Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP)


1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

EPA-740-D-23-002
December 2023
Office of Chemical Safety and
Pollution Prevention

United States

v/crM Environmental Protection Agency

Draft Risk Evaluation for
Tris(2-chloroethyl) Phosphate
(TCEP)

CASRN 115-96-8

December 2023
Page 1 of 572

-------
28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

TABLE OF CONTENTS

ACKNOWLEDGEMENTS	18

EXECUTIVE SUMMARY	19

1	INTRODUCTION	22

1.1	Scope of the Risk Evaluation	22

1.1.1	Life Cycle and Production Volume	22

1.1.2	Conditions of Use Included in the Draft Risk Evaluation	26

1.1.2.1 Conceptual Models	28

1.1.3	Populations Assessed	33

1.1.3.1 Potentially Exposed or Susceptible Subpopulations	33

1.2	Systematic Review	34

1.3	Organization of the Risk Evaluation	36

2	CHEMISTRY AND FATE AND TRANSPORT OF TCEP	37

2.1	Physical and Chemical Properties	37

2.2	Environmental Fate and Transport	39

2.2.1	Fate and Transport Approach and Methodology	39

2.2.2	Summary of Fate and Transport Assessment	42

2.2.3	Weight of the Scientific Evidence Conclusions for Fate and Transport	45

2.2.3.1 Strengths, Limitations, Assumptions, and Key Sources of Uncertainty for the Fate and

Transport Assessment	45

3	RELEASES AND CONCENTRATIONS OF TCEP IN THE ENVIRONMENT	46

3.1	Approach and Methodology	46

3.1.1 Industrial and Commercial	46

3.2	Environmental Releases	49

3.2.1	Industrial and Commercial	49

3.2.1.1 Summary of Daily Environmental Release Estimates	50

3.2.2	Consumer Releases	57

3.2.3	Weight of the Scientific Evidence Conclusions for Environmental Releases from
Industrial, Commercial, and Consumer Sources	57

3.2.3.1 Strengths, Limitations, Assumptions, and Key Sources of Uncertainty for the

Environmental Release Assessment	57

3.3	Concentrations of TCEP in the Environment	59

3.3.1	Ambient Air Pathway	60

3.3.1.1	Measured Concentrations in Ambient Air	61

3.3.1.2	EPA Modeled Concentrations in Ambient Air and Air Deposition
(IIOAC/AERMOD)	61

3.3.1.2.1 TCEP Partitioning between Gaseous Phase and Particulate Phase	63

3.3.2	W ater Pathway	64

3.3.2.1	Geospatial Analyses of Environmental Releases	64

3.3.2.1.1 Geospatial Analysis for Tribal Exposures	65

3.3.2.2	Measured Concentrations in Surface Water	67

3.3.2.3	Measured Concentrations in Precipitation	69

3.3.2.4	Measured Concentrations in Surface Water Databases	69

3.3.2.5	EPA Modeled Surface Water Concentrations (E-FAST, VVWM-PSC)	71

3.3.2.6	EPA Modeled Surface Water Concentrations via Air Deposition (AERMOD)	74

Page 2 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

73	3.3.2.7 Measured Concentrations in Wastewater	75

74	3.3.2.8 Measured Concentrations in Sediment	76

75	3.3.2.9 EPA Modeled Sediment Concentrations (VVWM-PSC)	77

76	3.3.2.10 EPA Modeled Sediment Concentrations via Air Deposition (AERMOD)	79

77	3,3,3 Land Pathway	79

78	3.3.3.1 Measured Concentrations in Soil	80

79	3.3.3.2 EPA Modeled Soil Concentrations via Air Deposition (AERMOD)	80

80	3.3.3.3 Measured Concentrations in Biosolids	81

81	3.3.3.4 EPA Calculated Soil Concentrations via Biosolids	81

82	3.3.3.5 Measured Concentrations in Groundwater	82

83	3.3.3.6 Measured Concentrations in Groundwater Databases	82

84	3.3.3.7 EPA Modeled Groundwater Concentrations via Leaching (DRAS)	84

85	3.4 Concentrations of TCEP in the Indoor Environment	86

86	3.4.1 Indoor Air Pathway	86

87	3.4.1.1 Measured Concentrations in Indoor Air	86

88	3.4.1.2 Measured Concentrations in Personal Air	88

89	3.4.1.3 EPA Modeled Indoor Concentrations as a Ratio of Ambient Air	89

90	3.4.1.4 Reported Modeled Concentrations in Indoor Air	89

91	3.4,2 Indoor Dust Pathway	90

92	3.4.2.1 Measured Concentrations in Indoor Dust	90

93	3.4.2.2 Reported Modeled Concentrations in Indoor Dust	92

94	4 ENVIRONMENTAL RISK ASSESSMENT	93

95	4.1 Environmental Exposures	94

96	4.1.1 Approach and Methodology	94

97	4.1,2 Exposures to Aquatic Species	95

98	4.1.2.1 Measured Concentrations in Aquatic Species	95

99	4.1.2.2 Calculated Concentrations in Aquatic Species	96

100	4.1.2.3 Modeled Concentrations in the Aquatic Environment	97

101	4.1.3 Exposures to Terrestrial Species	98

102	4.1.3.1 Measured Concentrations in Terrestrial Species	98

103	4.1.3.2 Modeled Concentration in the Terrestrial Environment	99

104	4.1.4 Trophic Transfer Exposure	100

105	4.1.5 Weight of the Scientific Evidence Conclusions for Environmental Exposures	102

106	4.1.5.1 Strengths, Limitations, Assumptions, and Key Sources of Uncertainty for the

107	Environmental Exposure Assessment	102

108	4.2 Environmental Hazards	104

109	4.2.1 Approach and Methodology	104

110	4.2,2 Aquatic Species Hazard	105

111	4.2.3 Terrestrial Species Hazard	107

112	4,2,4 Environmental Hazard Thresholds	110

113	4.2.4.1 Aquatic Species COCs Using Empirical and SSD Data	110

114	4.2.4.2 Aquatic Species COCs Using ECOSAR Modeled Data	Ill

115	4.2.4.3 Terrestrial Species Hazard Values	Ill

116	4,2.5 Summary of Environmental Hazard Assessment	112

117	4.2,6 Weight of the Scientific Evidence Conclusions for Environmental Hazards	114

118	4.2.6.1 Strengths, Limitations, Assumptions, and Key Sources of Uncertainty for the

119	Environmental Hazard Assessment	114

120	4.3 Environmental Risk Characterization	118

Page 3 of 572

-------
121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

4.3.1	Risk Characterization Approach	120

4.3.1.1 Risk Characterization Approach for Trophic Transfer	124

4.3.2	Risk Characterization for Aquatic Receptors	127

4.3.3	Risk Characterization for Terrestrial Receptors	134

4.3.4	Risk Characterization Based on Trophic Transfer in the Environment	135

4.3.5	Connections and Relevant Pathways from Exposure Media to Receptors	138

4.3.5.1	Aquati c Receptors	138

4.3.5.2	Terrestrial Receptors	138

4.3.6	Summary of Environmental Risk Characterization	140

4.3.6.1	COUs with Quantified Release Estimates	140

4.3.6.2	COUs without Quantified Release Estimates	146

4.3.7	Overall Confidence and Remaining Uncertainties Confidence in Environmental Risk
Characterization	150

4.3.7.1	Trophic Transfer Confidence	150

4.3.7.2	Risk Characterization Confidence	153

5 HUMAN HEALTH RISK ASSESSMENT	155

5.1 Human Exposures	156

5.1.1	Occupational Exposures	157

5.1.1.1	Approach and Methodology	157

5.1.1.2	Summary of Inhalation Exposure Assessment	162

5.1.1.3	Summary of Dermal Exposure Assessment	167

5.1.1.4	Weight of the Scientific Evidence Conclusions for Occupational Exposure	169

5.1.1.4.1 Strengths, Limitations, Assumptions, and Key Sources of Uncertainty for the

Occupational Exposure Assessment	169

5.1.2	Consumer Exposures	171

5.1.2.1	Approach and Methodol ogy	171

5.1.2.2	Consumer COUs and Exposure Scenarios	172

5.1.2.2.1	Consumer Exposure Routes Evaluated	176

5.1.2.2.2	Inhalation Exposure Assessment	180

5.1.2.2.3	Dermal Exposure Assessment	182

5.1.2.2.4	Oral Exposure Assessment	184

5.1.2.2.5	Qualitative Exposure Assessment	187

5.1.2.3	Summary of Consumer Exposure Assessment	189

5.1.2.4	Weight of the Scientific Evidence Confidence for Consumer Exposure	192

5.1.2.4.1 Strengths, Limitations, Assumptions, and Key Sources of Uncertainty for the

Consumer Exposure Assessment	195

5.1.3	General Population Exposures	200

5.1.3.1	Approach and Methodol ogy	201

5.1.3.1.1 General Population Exposure Scenarios	205

5.1.3.2	Summary of Inhalation Exposure Assessment	206

5.1.3.3	Summary of Dermal Exposure Assessment	208

5.1.3.3.1	Incidental Dermal from Swimming	208

5.1.3.3.2	Incidental Dermal Intake from Soil	209

5.1.3.4	Summary of Oral Exposures Assessment	210

5.1.3.4.1	Drinking Water Exposure	211

5.1.3.4.2	Fish Ingestion Exposure	215

5.1.3.4.3	Subsistence Fish Ingestion Exposure	219

5.1.3.4.4	Tribal Fish Ingestion Exposure	219

Page 4 of 572

-------
169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

5.1.3.4.5	Incidental Oral Ingestion from Soil	221

5.1.3.4.6	Incidental Oral Ingestion from Swimming	222

5.1.3.4.7	Human Milk Exposure	224

5.1.3.4.8	Dietary Exposure (non-TSCA)	226

5.1.3.5	Exposure Reconstruction Using Human Biomonitoring Data and Reverse Dosimetry 228

5.1.3.6	Summary of General Population Exposure Assessment	232

5.1.3.6.1 General Population Exposure Results	232

5.1.3.7	Weight of the Scientific Evidence Conclusions for General Population Exposure	237

5.1.3.7.1	Strengths, Limitations, Assumptions, and Key Sources of Uncertainty for the
General Population Exposure Assessment	238

5.1.3.7.2	Strengths, Limitations, and Key Sources of Uncertainty for the Human Milk
Pathway	241

5.1.4	Aggregate Exposure Scenarios	243

5.1.5	Sentinel Exposures	245

5.2	Human Health Hazard	246

5.2.1	Approach and Methodology	246

5.2.2	Toxicokinetics Summary	248

5.2.3	Non-cancer Hazard Identification and Evidence Integration	249

5.2.3.1	Critical Human Health Hazard Outcomes	250

5.2.3.1.1	Neurotoxicity	250

5.2.3.1.2	Reproductive Toxi city	253

5.2.3.1.3	Developmental Toxicity	257

5.2.3.1.4	Kidney Toxicity	260

5.2.3.2	Other Human Health Hazard Outcomes	261

5.2.4	Genotoxicity Hazard Identification and Evidence Integration	267

5.2.5	Cancer Hazard Identification, MOA Analysis, and Evidence Integration	268

5.2.5.1	Human Evidence	268

5.2.5.2	Animal Evidence	268

5.2.5.3	MOA Summary	270

5.2.5.4	Evidence Integration Summary	272

5.2.6	Dose-Response Assessment	273

5.2.6.1	Selection of Studies and Endpoints for Non-cancer Toxicity	274

5.2.6.1.1	Non-cancer Points of Departure for Acute Exposure	274

5.2.6.1.2	Non-cancer Points of Departure for Short-Term and Chronic Exposures	278

5.2.6.1.3	Uncertainty Factors Used for Non-cancer Endpoints	286

5.2.6.2	Selection of Studies and Endpoint Derivation for Carcinogenic Dose-Response
Assessment	286

5.2.7	Weight of the Scientific Evidence Conclusions for Human Health Hazard	287

5.2.7.1	Strengths, Limitations, Assumptions, and Key Sources of Uncertainty for the Hazard
Identification and Selection of PODs for Human Health Hazard Assessment	288

5.2.7.1.1	Acute Non-cancer	288

5.2.7.1.2	Short-Term and Chronic Non-cancer	289

5.2.7.1.3	Cancer	291

5.2.7.2	Human Health Hazard Confidence Summary	292

5.2.8	Toxicity Values Used to Estimate Risks from TCEP Exposure	292

5.2.9	Hazard Considerations for Aggregate Exposure	294

5.3	Human Health Risk Characterization	294

5,3.1 Risk Characterization Approach	294

5.3.1.1 Estimation of Non-cancer Risks	296

Page 5 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

218	5.3.1.2 Estimation of Cancer Risks	296

219	5.3.2 Summary of Human Health Risk Characterization	297

220	5.3.2.1 Summary of Risk Estimates for Workers	297

221	5.3.2.1.1 COUs/OESs with Quantitative Risk Estimates	297

222	5.3.2.1.2 COUs/OESs without Quantitative Risk Estimates	304

223	5.3.2.2 Summary of Risk Estimates for Consumers	305

224	5.3.2.2.1 COUs with Quantitative Risk Estimates	305

225	5.3.2.2.2 COUs without Quantitative Risk Estimates	311

226	5.3.2.3 Summary of Risk Estimates for the General Population	311

227	5.3.2.3.1 COUs with Quantitative Risk Estimates	311

228	5.3.2.3.2 COUs without Quantitative Risk Estimates	324

229	5.3.2.4 Summary of Risk Estimates for Infants from Human Milk	326

230	5.3.3 Risk Characterization for Potentially Exposed or Susceptible Subpopulations	329

231	5.3.4 Risk Characterization for Aggregate and Sentinel Exposures	338

232	5.3.5 Overall Confidence and Remaining Uncertainties in Human Health Risk

233	Characterization	341

234	5.3.5.1 Occupational Risk Estimates	341

235	5.3.5.2 Consumer Risk Estimates	343

236	5.3.5.3 General Population Risk Estimates	346

237	5.3.5.4 Hazard Values	349

238	6 UNREASONABLE RISK DETERMINATION	355

239	6.1 Unreasonable Risk to Human Health	357

240	6.1.1 Populations and Exposures EPA Assessed to Determine Unreasonable Risk to Human

241	Health	357

242	6,1,2 Summary of Unreasonable Risks to Human Health	357

243	6.1.3 Basis for EPA's Determination of Unreasonable Risk to Human Health	358

244	6.1.4 Unreasonable Risk in Occupational Settings	359

245	6.1.5 Unreasonable Risk to Consumers	359

246	6.1.6 Unreasonable Risk to the General Population	360

247	6.1.7 Unreasonable Risk to Infants from Human Milk	361

248	6.2 Unreasonable Risk to the Environment	361

249	6.2.1 Populations and Exposures EPA Assessed to Determine Unreasonable Risk to the

250	Environment	362

251	6.2,2 Summary of Unreasonable Risks to the Environment	362

252	6.2,3 Basis for EPA's Determination of Unreasonable Risk of Injury to the Environment	362

253	6.3 Additional Information Regarding the Basis for the Unreasonable Risk Determination	363

254	6.3.1 Additional Information about COUs Characterized Qualitatively	363

255	REFERENCES	371

256	APPENDICES	402

257	Appendix A ABBREVIATIONS, ACRONYMS, AND GLOSSARY OF SELECT TERMS	402

258	A.l Abbreviations and Acronyms	402

259	A.2 Glossary of Select Terms	405

260	Appendix B REGULATORY AND ASSESSMENT HISTORY	407

261	B.l Federal Laws and Regulations	407

262	B.2 State Laws and Regulations	408

Page 6 of 572

-------
263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

306

307

308

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

B,3 International Laws and Regulations	409

B.4 Assessment History	410

Appendix C LIST OF SUPPLEMENTAL DOCUMENTS	411

Appendix D PET ATT,ED EVALUATION OF POTENTIALLY EXPOSED OR

SUSCEPTIBLE SUBPOPULATIONS	415

D.l PESS Based on Greater Exposure	415

D,2	PESS Based on Greater Susceptibility	418

Appendix E PHYSICAL AND CHEMICAL PROPERTIES AND FATE AND TRANSPORT

DETAILS	425

E.l	Physical and Chemical Properties Evidence Integration	425

E. 1.1 Physical Form	425

E. 1.2 Vapor Density	425

E. 1.3 Octanol: Air Partition Coefficient (Log Koa)	425

E. 1.4 Henry's Law Constant (HLC)	425

E.l.5 Flash Point	426

E.l.6 Autoflammability	426

E.2	Fate and Transport	427

E.2.1 Approach and Methodology	427

E.2.1.1 EPI Suite™ Model Inputs	428

E.2.1.2 Fugacity Modeling	428

E.2.1.3 OECD Pov and LRTP Screening Tool	429

E.2.1.4 Evidence Integration	430

E.2.2 Air and Atmosphere	430

E.2.3 Aquatic Environments	432

E.2.3.1 Surface Water	432

E.2.3.2 Sediments	434

E.2.4 Terrestrial Environments	434

E.2.4.1 Soil 	434

E.2.4.2 Groundwater	435

E.2.4.3 Landfills	436

E.2.4.4 Biosolids	437

E.2.4.5 Key Sources of Uncertainty	437

E.2.5 Persistence Potential of TCEP	437

E.2.5.1 Destruction and Removal Efficiency	437

E.2.5.2 Removal in Wastewater	438

E.2.5.3 Removal in Drinking Water Treatment	439

E.2.6	Bioaccumulation Potential of TCEP	439

E.2.6.1	Key Sources of Uncertainty	442

Appendix F ENVIRONMENTAL HAZARD DETAILS	444

F.l	Approach and Methodology	444

F.2 Hazard Identification	444

F.2.1	Aquatic Hazard Data	444

F.2.1.1	Web-Based Interspecies Correlation Estimation (Web-ICE)	444

F.2.1.2 Species Sensitivity Distribution (SSD)	447

F.2.2 Terrestrial Hazard Data	452

F. 2.3 Evi dence Integrati on	453

Page 7 of 572

-------
309

310

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325

326

327

328

329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

344

345

346

347

348

349

350

351

352

353

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

F.2.3.1 Weight of the Scientific Evidence	454

Appendix G ENVIRONMENTAL RISK DETAILS	458

G, 1 Risk Estimation for Aquatic Organisms	458

G,2 Risk Estimation for Terrestrial Organisms	465

G.3	Trophic Transfer Analysis Results	466

Appendix H GENERAL POPULATION EXPOSURE DETAILS	469

H.	1 Exposure Factors	469

H,2 Water Pathway	470

H.2.1 Surface Water and Groundwater Monitoring Database Retrieval and Processing	470

H.2.1.1 Water Plots and Figures Generated in R	470

H.2.2 Methodology for Obtaining New Flow Data (2015 to 2020)	472

H.2.3 E-FAST: Predicted Flowing Surface Water Concentrations (First Tier Modeling)	472

H.2.3.1	E-FAST Exposure Activity Parameters	476

H.2.4 VVWM-PSC: Predicted Flowing Surface Water Concentrations (Second Tier Modeling) 477

H.3 Ambient Air Pathway	477

H.3.1 Modeling Approach for Estimating Concentrations in Ambient Air	478

H.3.2 Ambient Air: Screening Methodology	478

H.3.3 Ambient Air: AERMOD Methodology	481

H.4 Human Milk Pathway	490

H. 4.1 V erner Model	492

H.4.2 Milk Ingestion Rates by Age	495

H.4.3 Modeled Milk Concentrations	496

H.4.4 Infant Exposure Estimates	496

H.4.5 Infant Risk Estimates	505

H.4.6	Sensitivity Analysis	513

H.5	Landfill Analysis Using DRAS	514

Appendix I CONSUMER EXPOSURE DETAILS	517

I.1	Approach and Methodology	517

I.1.1	Consumer Exposure Model (CEM)	517

1.1.1 Inputs	519

I.1.1.1	Consumer Exposure Modeling and Sensitivity Analysis	519

1.1.1 Results	519

1.1.1.1 Navigating Supplemental Consumer Modeling Results	519

1.1.1.1 CEM 3.0 User Guide and Appendices	522

Appendix J HUMAN HEALTH HAZARD DETAILS	523

J. 1 Toxicokinetics and PBPK Models	523

J. 1.1 Absorption	523

J. 1.2 Distribution	524

J. 1.3 Metabolism	524

J. 1.4 Elimination	525

J. 1.5 PBPK Modeling Approach	525

J,2 Detailed Mode of Action Information	525

J.2.1 Mutagenicity	526

J.2.2 Other Modes of Action	527

J.2.3 Mode of Action Conclusions	528

Page 8 of 572

-------
354

355

356

357

358

359

360

361

362

363

364

365

366

367

368

369

370

371

372

373

374

375

376

377

378

379

380

381

382

383

384

385

386

387

388

389

390

391

392

393

394

395

396

397

398

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

J.3 Dose-Response Derivation 528

J.3.1 Adjustments for All PODs (Non-cancer and Cancer) 528

J.3.2 Non-cancer Dose-Response Modeling 529

J.3.2.1 Calculating Daily Oral Human Equivalent Doses (HEDs) 529

J.3.2.2 Use of Oral HED as Dermal HED 530

J.3.2.3 Extrapolating to Inhalation Human Equivalent Concentrations (HECs) 530

J.3.2.4 TCEP Non-cancer HED and HEC Calculations for Acute Exposures 531

J.3.2.5 TCEP Non-cancer HED and HEC Calculations for Short-Term and Chronic

Exposures 531

J.3.3 Cancer Dose-Response Modeling 532

J.3.3.1 Calculating Daily Oral Cancer Slope Factors (CSFs) 532

J.3.3.2 Use of Oral CSF as Dermal CSF 532

J.3.3.3 Extrapolating to Inhalation Unit Risks (IURs) 533

J.3.3.4 TCEP CSF and IUR Calculations for Lifetime Exposures 533

Appendix K EVIDENCE INTEGRATION FOR HUMAN HEALTH OUTCOMES 535

K. 1 Evidence Integration Tables for Major Human Health Hazard Outcomes 536

K.2 Evidence Integration Statements for Health Outcomes with Limited Data 557

Appendix L GENOTOXICITY DATA SUMMARY 559

L. 1.1 Chromosomal Aberrations 559

L. 1.1.1 In Vivo Data 559

L. 1.1.2 In Vitro Data 560

L.1.2 Gene Mutations 560

L. 1.2.1 In Vitro Studies 560

L. 1.3 Other Genotoxicity Assays 561

Appendix M EXPOSURE RESPONSE ARRAY FOR HUMAN HEALTH HAZARDS 567

Appendix N DRAFT EXISTING CHEMICAL EXPOSURE LIMIT (ECEL) DERIVATION. 568

N. 1 Draft Occupational Exposure Value Calculations 568

N.2 Summary of Air Sampling Analytical Methods Identified 571

LIST OF TABLES

Table 1-1. Conditions of Use in the Risk Evaluation for TCEP 26

Table 2-1. Physical and Chemical Properties of TCEP 37

Table 2-2. Environmental Fate Properties of TCEP 40

Table 3-1. Crosswalk of Conditions of Use (COUs) to Occupational Exposure Scenarios Assessed 46

Table 3-2. Summary of EPA's Daily Release Estimates for Each OES and EPA's Overall Confidence in

these Estimates for 2,500 lb Production Volume 51

Table 3-3. Summary of EPA's Release Estimates for Each COU/OES and EPA's Overall Confidence in

these Estimates 54

Table 3-4. Excerpt of Ambient Air Modeled Concentrations and Deposition for the Use of Paints and

Coatings - Spray Application OES, 2,500 lb Production Volume, 95th Percentile Release

Estimate, Suburban Forest Land Category Scenario 63

Table 3-5. Summary of Modeled Surface Water Concentrations for the 2,500 lb, High-End Release

Estimates 73

Table 3-6. Summary of Modeled Benthic Pore Water and Sediment Concentrations for the 2,500 lb

Production Volume, High Estimate Releases 78

Page 9 of 572

-------
399

400

401

402

403

404

405

406

407

408

409

410

411

412

413

414

415

416

417

418

419

420

421

422

423

424

425

426

427

428

429

430

431

432

433

434

435

436

437

438

439

440

441

442

443

444

445

446

447

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table 3-7. Potential Groundwater Concentrations (|ig/L) of TCEP Found in Wells within 1 Mile of a

Disposal Facility Determined Using the DRAS Model 85

Table 4-1. TCEP Fish Concentrations Calculated from VVWM-PSC Modeled Industrial and

Commercial TCEP Releases 97

Table 4-2. Aquatic Organisms Environmental Hazard Studies Used for TCEP 106

Table 4-3. Terrestrial Organisms Environmental Hazard Studies Used for TCEP 109

Table 4-4. Environmental Hazard Thresholds for Aquatic Environmental Toxicity 114

Table 4-5. Environmental Hazard Thresholds for Terrestrial Environmental Toxicity 114

Table 4-6. TCEP Evidence Table Summarizing the Overall Confidence Derived from Hazard

Thresholds 117

Table 4-7. Risk Characterization to Corresponding Aquatic and Terrestrial Receptors Assessed for the

Following COUs 121

Table 4-8. Terms and Values Used to Assess Potential Trophic Transfer of TCEP for Terrestrial Risk

Characterization 125

Table 4-9. Environmental Risk Quotients (RQs) by COU with Production Volumes of 2,500 lb/year for
Aquatic Organisms with TCEP Surface Water Concentration (ppb) Modeled by VVWM-

PSC 130

Table 4-10. Environmental Risk Quotients (RQs) by COU with Production Volumes of 2,500 lb/year for
Aquatic Organisms with TCEP Pore Water Concentration (ppb) Modeled by VVWM-

PSC 131

Table 4-11. Environmental Risk Quotients (RQs) by COU with Production Volumes of 2,500 lb/year for
Aquatic Organisms with TCEP Sediment Concentration (ppb) Modeled by VVWM-PSC

132

Table 4-12. Risk Quotients (RQs) Calculated Using Monitored Environmental Concentrations from

WQX/WQP 133

Table 4-13. Risk Quotients (RQs) Calculated Using TCEP in Surface Water from Monitored

Environmental Concentrations from Published Literature 133

Table 4-14. Risk Quotients (RQs) Calculated Using TCEP Concentrations in Sediment from Published

Literature 134

Table 4-15. Calculated Risk Quotients (RQs) Based on TCEP Soil Concentrations (mg/kg) as Calculated

Using Modeled Data 134

Table 4-16. Risk Quotients (RQs) Calculated Using TCEP Soil Concentrations from Published

Literature 135

Table 4-17. Risk Quotients (RQs) for Screening Level Trophic Transfer of TCEP in Terrestrial

Ecosystems Using EPA's Wildlife Risk Model for Eco-SSLs 136

Table 4-18. Risk Quotients (RQs) Calculated with Highest Mean TCEP Soil Concentration (5.89E-03
mg/kg) from Monitored Values in Published Literature for Screening Level Trophic
Transfer of TCEP in Terrestrial Ecosystems Using EPA's Wildlife Risk Model for Eco-

SSLs 137

Table 4-19. Selected Risk Quotients (RQs) (Highest Fish TCEP Concentrations) Based on Potential
Trophic Transfer of TCEP from Fish to American Mink (Mustela vison) as a Model

Aquatic Predator Using EPA's Wildlife Risk Model for Eco-SSLs 137

Table 4-20. Exposure Scenarios (Production Volume of 2,500 lb TCEP/year) and Corresponding

Environmental Risk for Aquatic Receptors with TCEP in Surface Water, Sediment, and

Pore Water 143

Table 4-21. Exposure Scenarios (Production Volume of 2,500 lb TCEP/year) and Corresponding
Environmental Risk for Terrestrial Receptors with TCEP in Soil (Invertebrates) and

Trophic Transfer 144

Table 4-22. TCEP Evidence Table Summarizing Overall Confidence Derived for Trophic Transfer... 152

Page 10 of 572

-------
448

449

450

451

452

453

454

455

456

457

458

459

460

461

462

463

464

465

466

467

468

469

470

471

472

473

474

475

476

477

478

479

480

481

482

483

484

485

486

487

488

489

490

491

492

493

494

495

496

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table 4-23. TCEP Evidence Table Summarizing Overall Confidence for Environmental Risk

Characterization 154

Table 5-1. Summary for Each Occupational Exposure Scenarios (OES) 160

Table 5-2. Summary of Total Number of Workers and ONUs Potentially Exposed to TCEP for Each

OES 162

Table 5-3. Summary of Inhalation Exposure Results for Workers Based on Monitoring Data for Each

OES 164

Table 5-4. Summary of Inhalation Exposure Results for Workers Based on Exposure Modeling for Each

OES 164

Table 5-5. Summary of Inhalation Exposure Results for ONUs Based on Monitoring Data and Exposure

Modeling for Each OES 166

Table 5-6. Summary of Dermal Retained Dose for Workers Based on Exposure Modeling for Each OES

168

Table 5-7. Summary of Consumer COUs, Exposure Scenarios, and Exposure Routes 173

Table 5-8. CEM 3.0 Model Codes and Descriptions 176

Table 5-9. Crosswalk of COU Subcategories, CEM 3.0 Scenarios, and Relevant CEM 3.0 Models Used

for Consumer Modeling 177

Table 5-10. Summary of Key Parameters for Article Modeling in CEM 3.0a 179

Table 5-11. Steady State Air Concentrations and Respirable Particle of TCEP from Consumer Modeling

(CEM 3.0) 180

Table 5-12. Chronic Dermal Average Daily Doses (mg/kg-day) of TCEP from Consumer Article

Modeling for Adults and Children 3 to 6 Years of Age (CEM 3.0) 184

Table 5-13. Chronic Ingestion Average Daily Doses (mg/kg-day) of TCEP from Consumer Article

Modeling for Adults and Infants 1 to 2 Years of Age (CEM 3.0) 186

Table 5-14. Summary of Commercial Paints and Coatings Concentrations and Density of TCEP 187

Table 5-15. Summary of Acute Daily Rate of Consumer Articles Modeled with CEM 3.0 190

Table 5-16. Summary of Chronic Average Daily Doses of Consumer Articles Modeled with CEM 3.0

191

Table 5-17. Summary of Lifetime Average Daily Doses of Consumer Articles Modeled with CEM 3.0

192

Table 5-18. Weight of the Scientific Evidence Confidence for Chronic Consumer Exposure Modeling

Scenarios 193

Table 5-19. Sensitivity Analysis for Chronic Consumer Exposure Modeling Scenarios 195

Table 5-20. Summary of Sampling Date for TCEP Weight Fraction Data 198

Table 5-21. Summary of Indoor Monitoring Data of TCEP from U.S. Studies 199

Table 5-22. Summary of Environmental Monitoring Data of TCEP from the Literature for U.S. Studies

202

Table 5-23. Excerpt of Ambient Air Modeled Concentrations for the 2,500 lb Production Volume, High-
End Release Estimate for all COUs at 100 m, Suburban Forest Land Category Scenario

208

Table 5-24. Modeled Incidental Dermal (Swimming) Doses for all COUs for Adults, Youths, and

Children, for the 2,500 lb High-End Release Estimate 209

Table 5-25. Modeled Soil Dermal Doses for the Commercial Use of Paints and Coatings COU, for

Children 210

Table 5-26. 50th Quantile Distances and 30Q5 and Harmonic Mean 50th Quantile Dilution Factors for

Relevant TCEP SIC 213

Table 5-27. Modeled Drinking Water Ingestion Estimates for Diluted Surface Water Concentrations for
Adults for All Industrial and Commercial COUs for the 2,500 lb High-End Release
Estimate 213

Page 11 of 572

-------
497

498

499

500

501

502

503

504

505

506

507

508

509

510

511

512

513

514

515

516

517

518

519

520

521

522

523

524

525

526

527

528

529

530

531

532

533

534

535

536

537

538

539

540

541

542

543

544

545

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table 5-28. Modeled Drinking Water Ingestion Estimates for Surface Water Concentrations for Adults
for All Industrial and Commercial COUs for the 2,500 lb High-End Release Estimate 214

Table 5-29. Landfill Releases of TCEP from Two Commercial and Industrial OESs 214

Table 5-30. Estimated Average Daily Doses, Lifetime Average Daily Doses, and Lifetime Average

Daily Concentrations for Adults from Groundwater Concentrations by DRAS 215

Table 5-31. Fish Tissue Concentrations Calculated from Modeled Surface Water Concentrations and

Monitoring Data 216

Table 5-32. Adult General Population Fish Ingestion Doses by Scenario Based on a Production Volume

of 2,500 lb/year and High-End Release Distribution 218

Table 5-33. Adult Subsistence Fisher Doses by Scenario Based on a Production Volume of 2,500 lb/year

and High-End Release Distribution 219

Table 5-34. Adult Tribal Fish Ingestion Doses by Scenario Based on a PV of 2,500 lb/year, High-End

Release Distribution, and Two Fish Ingestion Rates 220

Table 5-35. Modeled Soil Dermal Doses for the Commercial Use of Paints and Coatings OES for

Children for the 2,500 lb High-End Release Estimates 221

Table 5-36. Modeled Incidental Oral (Swimming) Doses for All COUs, for Adults, Youth and Children,

for the 2,500 lb High-End Release Estimate 223

Table 5-37. Concentrations of Foods Found in the Monitoring Literature in ng/g 228

Table 5-38. Human TCEP/BCEP U.S. Biomonitoring Datasets by Population, Type, and Number 231

Table 5-39. Reconstructed Daily Intakes from Creatinine Adjusted Urinary BCEP Concentrations from

NHANES (2013-2014) 232

Table 5-40. General Population Acute Oral Ingestion Estimates for Drinking Water Summary Table 234

Table 5-41. Summary of General Population Chronic Oral Exposures 235

Table 5-42. Summary Acute and Chronic General Population Dermal Exposures 236

Table 5-43. Summary of General Population Inhalation Exposures 236

Table 5-44. Overall Confidence for General Population Exposure Scenarios 237

Table 5-45. Qualitative Assessment of the Uncertainty and Variability Associated with General

Population Assessment 240

Table 5-46. Comparison among Studies with Sensitive Endpoints Considered for Acute Exposure

Scenarios 276

Table 5-47. Dose-Response Analysis of Selected Studies Considered for Acute Exposure Scenarios . 277
Table 5-48. Comparison among Studies with Sensitive Endpoints Considered for Short-Term Exposure

Scenarios 281

Table 5-49. Dose-Response Analysis of Selected Studies Considered for Short-Term Exposure

Scenarios 282

Table 5-50. Comparison among Studies with Sensitive Endpoints Considered for Chronic Exposure

Scenarios 284

Table 5-51. Dose-Response Analysis of Selected Studies Considered for Chronic Exposure Scenarios

285

Table 5-52. Dose-Response Analysis of Kidney Tumorsa for Lifetime Exposure Scenarios 287

Table 5-53. Confidence Summary for Human Health Hazard Assessment 292

Table 5-54. Non-cancer HECs and HEDs Used to Estimate Risks 293

Table 5-55. Cancer IUR and CSF Used to Estimate Risks 293

Table 5-56. Exposure Scenarios, Populations of Interest, and Hazard Values 294

Table 5-57. Occupational Risk Summary for 2,500-Pound Production Volume 299

Table 5-58. Acute and Chronic Non-cancer Consumer Risk Summary 308

Table 5-59. Lifetime Cancer Consumer Risk Summary 310

Table 5-60. General Population Acute Drinking Water (Oral Ingestion) Non-cancer Risk Summary ..316
Table 5-61. Acute Fish Ingestion Non-cancer Risk Summary 317

Page 12 of 572

-------
546

547

548

549

550

551

552

553

554

555

556

557

558

559

560

561

562

563

564

565

566

567

568

569

570

571

572

573

574

575

576

577

578

579

580

581

582

583

584

585

586

587

588

589

590

591

592

593

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table 5-62. General Population Chronic Water and Soil Ingestion Non-cancer Risk Summary 318

Table 5-63. Chronic Fish Ingestion Non-cancer Risk Summary 319

Table 5-64. General Population Lifetime Cancer Oral Ingestion Risk Summary Table 320

Table 5-65. Lifetime Cancer Risk Summary for General Population and Fish Consumption 321

Table 5-66. General Population Dermal Acute and Chronic Non-cancer Risk Summary 322

Table 5-67. Lifetime Cancer Risk Summary for General Population and Fish Consumption 323

Table 5-68. General Population Lifetime Cancer Inhalation Risk Summary Tablea 323

Table 5-69. Summary of PESS Considerations Incorporated into the Risk Evaluation 330

Table 5-70. Summary of Detection Frequencies and Sampling Dates for Relevant Consumer Products

Containing TCEP 336

Table 5-71. Suggested Consumer Population Sizes Based on Characterization of Consumer Article

Detection Frequencies 337

Table 5-72. Overall Confidence for Acute, Short-Term, and Chronic Human Health Non-cancer Risk

Characterization for COUs Resulting in Risks 350

Table 5-73. TCEP Evidence Table Summarizing Overall Confidence for Lifetime Human Health Cancer

Risk Characterization for COUs Resulting in Risks 352

Table 6-1. Supporting Basis for the Draft Unreasonable Risk Determination for Human Health

(Occupational COUs) 365

Table 6-2. Supporting Basis for the Draft Unreasonable Risk Determination for Human Health

(Consumer COUs) 367

Table 6-3. Supporting Basis for the Draft Unreasonable Risk Determination for Human Health (Infant

Risks from Human Milk Ingestion, Upper Milk Intake Rate) 368

Table 6-4. Supporting Basis for the Draft Unreasonable Risk Determination for the Environment 370

LIST OF FIGURES

Figure 1-1. TSCA Existing Chemicals Risk Evaluation Process 22

Figure 1-2. TCEP Life Cycle Diagram 23

Figure 1-3. Reported Aggregate TCEP Production Volume (lb) 2012-2020 25

Figure 1-4. TCEP Conceptual Model for Industrial and Commercial Activities and Uses: Potential

Exposure and Hazards 29

Figure 1-5. TCEP Conceptual Model for Consumer Activities and Uses: Potential Exposures and

Hazards 30

Figure 1-6. TCEP Conceptual Model for Environmental Releases and Wastes: General Population

Hazards 31

Figure 1-7. TCEP Conceptual Model for Environmental Releases and Wastes: Ecological Exposures and

Hazards 32

Figure 1-8. Populations Assessed in this Draft Risk Evaluation 33

Figure 1-9. Diagram of the Systematic Review Process 35

Figure 2-1. Transport, Partitioning, and Degradation of TCEP in the Environment 44

Figure 3-1. An Overview of How EPA Estimated Daily Releases for Each OES 48

Figure 3-2. Concentrations of TCEP (ng/m3) in Ambient Air from 2000 to 2019 61

Figure 3-3. Map of Nationwide Measured TCEP Water Concentrations Retrieved from the Water

Quality Portal, 1995 to 2022 65

Figure 3-4. Map Indicating Norman Landfill in Proximity to Tribal Lands 66

Figure 3-5. Groundwater Concentration of TCEP Reported near Twenty-Nine Palms Reservation near

Coachella, California 67

Figure 3-6. Concentrations of TCEP (ng/L) in Surface Water from 1980 to 2017 68

Figure 3-7. Concentrations of TCEP (ng/L) in Precipitation from 1994 to 2014 69

Page 13 of 572

-------
594

595

596

597

598

599

600

601

602

603

604

605

606

607

608

609

610

611

612

613

614

615

616

617

618

619

620

621

622

623

624

625

626

627

628

629

630

631

632

633

634

635

636

637

638

639

640

641

642

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Figure 3-8. Frequency of Nationwide Measured TCEP Surface Water Concentrations Retrieved from the

Water Quality Portal, 2003 to 2022 70

Figure 3-9. Time Series of Nationwide Measured TCEP Surface Water Concentrations Retrieved from

the Water Quality Portal, 2003 to 2022 71

Figure 3-10. Concentrations of TCEP (ng/L) in Wastewater from 2001 to 2018 76

Figure 3-11. Concentrations of TCEP (ng/g) in Sediment from 1980 to 2017 77

Figure 3-12. Concentrations of TCEP (ng/L) in the Not Specified Fraction of Groundwater from 1978 to

2017 82

Figure 3-13. Frequency of Nationwide Measured TCEP Groundwater Concentrations Retrieved from the

Water Quality Portal, 1995 to 2021 83

Figure 3-14. Time Series of Nationwide Measured TCEP Groundwater Concentrations Retrieved from

the Water Quality Portal, 1995 to 2021 84

Figure 3-15. Concentrations of TCEP (ng/m3) in Indoor Air from 2000 to 2016 88

Figure 3-16. Concentrations of TCEP (ng/m3) in Personal Inhalation in General Population

(Background) Locations from 2013 to 2016 88

Figure 3-17. Concentrations of TCEP (ng/g) in Indoor Dust from 2000 to 2019 91

Figure 4-1. Measured Concentrations of TCEP (ng/g) in Aquatic Species - Fish from 2003 to 2016 .... 96
Figure 4-2. Measured Concentrations of TCEP (ng/g) in Terrestrial Species - Bird from 2000 to 2016 99
Figure 4-3. Measured Concentrations of TCEP (ng/g) in the Wet Fraction of Terrestrial Species - Plant

in Remote (Not Near Source) Locations from 1993 to 1994 99

Figure 4-4. Trophic Transfer of TCEP in Aquatic and Terrestrial Ecosystems 101

Figure 5-1. Approaches Used for Each Component of the Occupational Assessment for Each OES ... 158

Figure 5-2. Consumer Pathways and Routes Evaluated in this Assessment 172

Figure 5-3. Photo of TCEP Label on Wooden Television Stand 176

Figure 5-4. Potential Human Exposure Pathways to TCEP for the General Population 201

Figure 5-5. Direct and Indirect Exposure Assessment Approaches Used to Estimate General Population

Exposure to TCEP 204

Figure 5-6. Modeled Exposure Points for Finite Distance Rings for Ambient Air Modeling (AERMOD)

205

Figure 5-7. General Population Inhalation Concentrations (ppm) by Distance (m) in Log Scale 207

Figure 5-8. Concentrations of TCEP (ng/L) in Drinking Water from 1982 to 2014 211

Figure 5-9. Concentrations of TCEP (ng/g) in the Wet Fraction of Dietary from 1982 to 2018 227

Figure 5-10. Concentrations of TCEP (ng/L) in the Unadjusted Urine from 2015 to 2019 229

Figure 5-11. Concentrations of BCEP (ng/L) in the Creatinine-Adjusted Urine from 2014 to 2019 229

Figure 5-12. Concentrations of BCEP from NHANES data for the U.S. Population from 2011 to 2014

230

Figure 5-13. Concentrations of TCEP (ng/wipe) in Surface Wipes from 2014 to 2018 230

Figure 5-14. Concentrations of TCEP (ng/wipe) in Silicone Wristbands from 2012 to 2015 230

Figure 5-15. Aggregate Chronic Average Daily Doses (CADDs) for Each Consumer COU, Lifestage244
Figure 5-16. EPA Approach to Hazard Identification, Data Integration, and Dose-Response Analysis for

TCEP 247

Figure 5-17. Exposure Response Array for Short-Term and Chronic Exposure Durations by Likely

Hazard Outcomes 278

Figure 5-18. Aggregate CADDs for Consumer Use of textiles in Outdoor Play Structures at Adult,

Youth2, and Youth 1 Life Stages 339

Figure 5-19. Aggregate Acute Average Daily Doses (ADRs) for Carpet Back Coating, Childl, and

Infant2 Life Stages 340

Figure 5-20. Consumer Modeling Time Series Results for Acoustic Ceilings 345

Figure 5-21. Consumer Modeling Time Series Results for Wood Flooring 345

Page 14 of 572

-------
643

644

645

646

647

648

649

650

651

652

653

654

655

656

657

658

659

660

661

662

663

664

665

666

667

668

669

670

671

672

673

674

675

676

677

678

679

680

681

682

683

684

685

686

687

688

689

690

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Figure 5-22. Consumer Modeling Time Series Results for Insulation 346

LIST OF APPENDIX TABLES

Table_Apx B-l. Federal Laws and Regulations 407

Table_Apx B-2. State Laws and Regulations 408

Table_Apx B-3. International Laws and Regulations 409

TableApx B-4. Assessment History of TCEP 410

TableApx D-l. PESS Evidence Crosswalk for Increased Exposure 416

Table Apx D-2. PESS Evidence Crosswalk for Biological Susceptibility Considerations 420

Table Apx F-l. Web-ICE Predicted Species that Met Model Selection Criteria 446

Table Apx F-2. Considerations that Inform Evaluations of the Strength of the Evidence within an

Evidence Stream {i.e., Apical Endpoints, Mechanistic, or Field Studies) 456

Table Apx G-l. Calculated Risk Quotients Based on TCEP Sediment Concentrations (ppb) as

Calculated Using Modeled Data for Air Deposition to Sediment 458

Table Apx G-2. Environmental Risk Quotients by Exposure Scenario with Production Volumes of
2,500 lb/year for Aquatic Organisms with TCEP Surface Water Concentration (ppb)

Modeled by VVWM-PSC 459

Table Apx G-3. Environmental Risk Quotients by Exposure Scenario with Production Volumes of
2,500 lb/year for Aquatic Organisms with TCEP Pore Water Concentration (ppb)

Modeled by VVWM-PSC 460

Table Apx G-4. Environmental RQs by Exposure Scenario with Production Volumes of 2,500 lb/year
for Aquatic Organisms with TCEP Sediment Concentration (ppb) Modeled by VVWM-

PSC 461

Table Apx G-5. Environmental RQs by Exposure Scenario with Production Volumes of 25,000 lb/year
for Aquatic Organisms with TCEP Surface Water Concentration (ppb) Modeled by

VVWM-PSC 462

Table Apx G-6. Environmental RQs by Exposure Scenario with Production Volumes of 25,000 lb/year
for Aquatic Organisms with TCEP Pore Water Concentration (ppb) Modeled by VVWM-

PSC 463

Table Apx G-l. Environmental RQs by Exposure Scenario with Production Volumes of 25,000 lb/year
for Aquatic Organisms with TCEP Sediment Concentration (ppb) Modeled by VVWM-

PSC 464

Table Apx G-8. Calculated RQs Based on TCEP Soils Concentrations (mg/kg) as Calculated Using

Modeled Data for Air Deposition to Soil 465

Table Apx G-9. RQs Based on Potential Trophic Transfer of TCEP in Terrestrial Ecosystems Using

EPA's Wildlife Risk Model for Eco-SSLs (Equation 4-1) 466

Table Apx G-10. RQs Based on Potential Trophic Transfer of TCEP from Fish to American Mink as a
Model Aquatic Predator Using EPA's Wildlife Risk Model for Eco-SSLs (Equation 4-1)

468

Table_Apx H-l. Body Weight by Age Group 469

Table_Apx H-2. Fish Ingestion Rates by Age Group 469

Table Apx H-3. Crosswalk of COU and OES, Abbreviations, and Relevant SIC Codes 474

Table Apx H-4. Harmonic Mean, 30Q5, 7Q10, and 1Q10 50th Percentile Flows for Relevant TCEP SIC

Codes 475

Table Apx H-5. Incidental Dermal (Swimming) Modeling Parameters 476

Table Apx H-6. Incidental Oral Ingestion (Swimming) Modeling Parameters 476

Table Apx H-7. Ambient Air Release Inputs Utilized for Ambient Air Modeling: IIOAC and AERMOD
Methodology for TCEP 479

Page 15 of 572

-------
691

692

693

694

695

696

697

698

699

700

701

702

703

704

705

706

707

708

709

710

711

712

713

714

715

716

717

718

719

720

721

722

723

724

725

726

727

728

729

730

731

732

733

734

735

736

737

738

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table_Apx H-8. Settings for Gaseous Deposition 485

Table_Apx H-9. Settings for Particle Deposition 485

TableApx H-10. Description of Daily or Period Average and Air Concentration Statistics 488

TableApx H-ll. Key Chemical Characteristics of TCEP 491

Table Apx H-12. Data Input Requirements for the Multi-compartment Model 493

Table_Apx H-13. Mean and Upper Milk Ingestion Rates by Age 496

Table Apx H-14. Range of Modeled Milk Concentrations by Maternal Group 496

Table Apx H-15. Average Infant Doses via Human Milk Exposure from Maternal Consumer Use

Scenarios 497

Table Apx H-16. Average Infant Doses from Maternal Workers Based on Mean Milk Intake Rate ... 499
Table Apx H-17. Average Infant Doses from Maternal Workers Based on Upper Milk Intake Rate... 500
Table Apx H-18. Average Infant Doses via Human Milk Exposure from Maternal General Population

Oral Exposures Based on Mean Milk Intake Rate 501

Table Apx H-19. Average Infant Doses via Human Milk Exposure from Maternal General Population

Oral Exposures Based on Upper Milk Intake Rate 502

Table Apx H-20. Average Infant Doses via Human Milk Exposure from Maternal Tribal Fish Ingestion

Based on Mean Milk Intake Rate 503

Table Apx H-21. Average Infant Doses via Human Milk Exposure from Maternal Tribal Fish Ingestion

Based on Upper Milk Intake Rate 504

Table Apx H-22. Infant Risks via Human Milk Exposure from Maternal Consumer Use Scenarios... 505
Table Apx H-23. Infant Risks via Human Milk Exposure from Maternal Occupational Use Scenarios

Based on Mean Milk Intake Rate 506

Table Apx H-24. Infant Risks via Human Milk Exposure from Maternal Occupational Use Scenarios

Based on Upper Milk Intake Rate 507

Table Apx H-25. Infant Risks via Human Milk Exposure from Maternal General Population Oral

Exposures Based on Mean Milk Intake Rate 508

Table Apx H-26. Infant Risks via Human Milk Exposure from Maternal General Population Oral

Exposures Based on Upper Milk Intake Rate 509

Table Apx H-27. Infant Risks via Human Milk Exposure from Tribal Maternal Fish Exposures Based

on Mean Milk Intake Rate 510

Table Apx H-28. Infant Risks via Human Milk Exposure from Tribal Maternal Fish Exposures Based

on Upper Milk Intake Rate 511

Table_Apx H-25. Variables and Values Used in Sensitivity Analysis 513

Table_Apx H-30. Input Variables for Chemical of Concern 515

Table_Apx H-31. Waste Management Unit (WMU) Properties 516

Table_Apx K-l. Evidence Integration for Neurotoxicity 536

Table_Apx K-2. Evidence Integration for Reproductive Effects 540

Table Apx K-3. Evidence Integration for Developmental Effects 543

Table_Apx K-4. Evidence Integration Table for Kidney Effects 546

Table_Apx K-5. Evidence Integration Table for Liver Effects 548

Table_Apx K-6. Evidence Integration Table for Cancer 550

Table_Apx L-l. Results of In Vivo Micronucleus Test 560

Table Apx L-2. Results of Bacterial Reverse Mutation Test in Salmonella typhimurium 561

Table_Apx L-3. TCEP Genotoxicity Studies 563

TableApx N-l. Limit of Detection (LOD) and Limit of Quantification (LOQ) Summary for Identified
Air Sampling Analytical Methods 572

Page 16 of 572

-------
739

740

741

742

743

744

745

746

747

748

749

750

751

752

753

754

755

756

757

758

759

760

761

762

763

764

765

766

767

768

769

770

771

772

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

LIST OF APPENDIX FIGURES

FigureApx E-l. Box and Whisker Plots of Reported Physical and Chemical Property Data Values .. 427
FigureApx E-2. Screen Capture of EPI Suite™ Parameters Used to Calculate Fate and Physical and

Chemical Properties for TCEP 428

Figure Apx E-3. EPI Suite™ Level III Fugacity Modeling Graphical Result for TCEP 429

Figure Apx E-4. Screen Capture of OECD Pov and LRTP Screening Tool Parameters Used to Calculate

TCEP's LRTP 430

Figure Apx F-l. SSD Toolbox Interface Showing HC05s and P-Values for Each Distribution Using
Maximum Likelihood Fitting Method Using TCEP's Acute Aquatic Hazard Data

(Etterson, 2020) 448

Figure Apx F-2. AICc for the Six Distribution Options in the SSD Toolbox for TCEP's Acute Aquatic

Hazard Data (Etterson, 2020) 449

Figure Apx F-3. Q-Q Plot of TCEP Acute Aquatic Hazard Data with the Weibull Distribution (Etterson,

2020) 450

Figure Apx F-4. SSD Distribution for TCEP's Acute Hazard Data (Etterson, 2020) 451

Figure_Apx F-5. TRV Flow Chart 453

Figure Apx H-l. Example Tooltips from Media Maps and Time Series Graphs 471

Figure_Apx H-2. Overview of EPA's Screening Level Ambient Air Pathway Methodology 478

Figure Apx H-3. Modeled Exposure Points Locations for Finite Distance Rings 482

Figure_Apx H-4. Modeled Exposure Points for Area Distance 483

Figure Apx H-5. Cuticular Resistance as a Function of Vapor Pressure 486

Figure Apx H-6. Compartments and Exposure Routes for Verner Model 492

Figure_Apx H-7. Sensitivity Analysis of Model Inputs Measured as Elasticity 513

Figure Apx 1-1. Screenshot of Lifetime Average Daily Doses (LADDs) Bar Chart Displaying Tool Tip

for Acoustic Ceiling, Inhalation Estimate 520

Figure Apx 1-2. Screenshot of Lifetime Average Daily Doses (LADDs) Bar Chart Displaying Function

to Compare Data on Hover, for Insulation Estimates 520

Figure Apx 1-3. Screenshot of Lifetime Average Daily Doses (LADDs) Bar Chart Displaying Bar Chart

that Deselects Inhalation Estimate and Selects Ingestion and Dermal Estimates 521

Figure Apx 1-4. Screenshots of Lifetime Average Daily Doses (LADDs) Bar Chart Displaying a

Cropped Subsection of the Figure 521

Figure Apx M-l. Exposure Response Array for Likely and Suggestive Human Health Hazard

Outcomes 567

Page 17 of 572

-------
773

774

775

776

777

778

779

780

781

782

783

784

785

786

787

788

789

790

791

792

793

794

795

796

797

798

799

800

801

802

803

804

805

806

807

808

809

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

ACKNOWLEDGEMENTS

This report was developed by the United States Environmental Protection Agency (U.S. EPA or the
Agency), Office of Chemical Safety and Pollution Prevention (OCSPP), Office of Pollution Prevention
and Toxics (OPPT).

Acknowledgements

The Assessment Team gratefully acknowledges the participation, input, and review comments from
OPPT and OCSPP senior managers and science advisors and assistance from EPA contractors Battelle
(Contract No. EP-W-16-017), GDIT (Contract No. HHSN316201200013W), SPS (Contract No.
68HERC20D0021), ERG (Contract No. 68HERD20A0002), ICF (Contract No. 68HERH22F0259),
Versar (Contract No. EP-W-17-006), SRC (Contract No. 68HERH19D0022). Special acknowledgement
is given for the contributions of technical experts from EPA's Office of Research and Development
(ORD), including Sandy Raimondo for her review of the Web-ICE methodology in Appendix F.2.1.1.

As part of an intra-agency review, the draft TCEP Risk Evaluation was provided to multiple EPA
Program Offices for review. Comments were submitted by Office of the Administrator/Office of
Children's Health Protection, Office of Air and Radiation, Office of General Council, Office of
Research and Development, and Office of Water.

Docket

Supporting information can be found in the public docket, Docket ID CEP A-HQ-QPPT-2023-0265Y
Disclaimer

Reference herein to any specific commercial products, process or service by trade name, trademark,
manufacturer, or otherwise does not constitute or imply its endorsement, recommendation, or favoring
by the United States Government.

Authors: James Bressette, Xiah Kragie, and Andrea Pfahles-Hutchens (Assessment Leads), Kara
Koehrn (Management Lead), Kesha Forrest (Branch Chief), Yousuf Ahmad, Andrea Amati, Edwin
Arauz, Amy Benson, Sarah Gallagher, Lauren Gates, Chris Green, Leigh Hazel, Keith Jacobs, Rachel
McAnallen, Claudia Menasche, Catherine Ngo, Chloe O'Haire, Joseph Rappold.

Contributors: Chris Brinkerhoff, Sandy Raimondo, Cecelia Tan.

Technical Support: Mark Gibson, Hillary Hollinger.

This draft risk evaluation was reviewed by OPPT and OCSPP leadership.

Page 18 of 572

-------
810

811

812

813

814

815

816

817

818

819

820

821

822

823

824

825

826

827

828

829

830

831

832

833

834

835

836

837

838

839

840

841

842

843

844

845

846

847

848

849

850

851

852

853

854

855

856

857

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

EXECUTIVE SUMMARY

The EPA has evaluated tris(2-chloroethyl) phosphate, or TCEP, under the Toxic Substances Control Act
(TSCA). In this draft risk evaluation, EPA preliminarily finds that TCEP presents an unreasonable
risk of injury to human health and the environment.

In December 2019, EPA designated TCEP as a high-priority substance for TSCA evaluation and in
August 2020 released the final scope of the risk evaluation. This draft risk evaluation assesses human
health risk to workers, consumers, and the general population, as well as risk to the environment.

Although U.S. production of TCEP has decreased by about 99 percent since 2014, it is still used
domestically to make some paints and coatings and as a flame retardant and plasticizer for specific
aerospace applications. In the past, TCEP was processed in many products made in the United States,
including fabrics and textiles, some types of foam, and construction materials—some of which may still
be in use today. TCEP may still be found in a wide range of goods that are imported into the United
States.

Because TCEP is mixed into but not chemically bonded to materials, it can leach out of products and
into the environment. TCEP that is released into the environment from manufacturing processes or
leaching from products primarily ends up in water, sediment, soil, or dust. TCEP may leach out of
materials dumped in landfills and reach groundwater or surface water. It can also be released into the air.
If TCEP enters the atmosphere, it can be deposited in lakes and rivers through rain and snowfall. TCEP
can be carried long distances via air and water and has been detected in the Arctic. TCEP concentrations
may be even higher indoors than outdoors, because TCEP can leach out of consumer products such as
carpets or wooden TV stands and attach to household dust. Although TCEP is persistent in the
environment {i.e., it does not easily degrade) and has been detected in organisms such as fish exposed to
TCEP in surface water, it does not appear to bioaccumulate because it is not found to accumulate in
people or animals at greater concentrations than exist in the environment.

Unreasonable Risk to Human Health

Data from laboratory animal testing shows that exposure to TCEP may increase the risk of adverse
effects in people such as kidney cancer and other cancers, as well as harm to neurological and
reproductive systems (Section 5.2.5.3). EPA evaluated the risks of people experiencing these cancers
and harmful neurological and reproductive effects from being exposed to TCEP at work, in the home, by
breastfeeding, and by eating fish taken from TCEP-contaminated water. When determining
unreasonable risk of TCEP to human health, EPA also accounted for potentially exposed and susceptible
populations—pregnant women, infants exposed through human milk, children and adolescents
(especially males), people who experience aggregated or sentinel exposures, fenceline communities who
live near facilities that emit TCEP, firefighters, and people and tribes whose diets include large amounts
of fish (Section 5.3.3).

Workers with the greatest potential for exposure to TCEP are those who spray TCEP-containing paints
or coatings, or workers who are involved in processing a 2-part resin used in paints, coatings, and
polyurethane resin castings for aerospace applications (Section 5.3.2.1). Outside the workplace, adults,
infants, and children may be most at risk if they breathe or ingest TCEP that comes out of fabrics,
textiles, foam, and wood products and that either attaches to dust or otherwise gets into indoor air
(Section 5.3.2.2). Infants and children may be at risk if they mouth products containing foam, textiles, or
wood that contain TCEP (Section 5.3.2.3) or are breastfed (Section 5.3.2.4). People who are subsistence
fishers may be at high risk if they eat TCEP-contaminated fish; tribal people for whom fish is important

Page 19 of 572

-------
858

859

860

861

862

863

864

865

866

867

868

869

870

871

872

873

874

875

876

877

878

879

880

881

882

883

884

885

886

887

888

889

890

891

892

893

894

895

896

897

898

899

900

901

902

903

904

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

dietarily and culturally have even higher risk than the general population and subsistence fishers
(Section 5.3.3).

EPA's assessment preliminarily shows unreasonable risks of cancer and noncancer health effects
from half of the TCEP conditions of use (COUs) to (1) breastfed infants, (2) people who handle
TCEP or handle products formulated with TCEP at work, (3) people who breathe or ingest dust
from TCEP that comes off of consumer products, and (4) people who eat large amounts of fish
contaminated with TCEP. For workers, there are certain activities where acute, short-term, chronic and
lifetime exposures to TCEP—especially from contact with skin—contribute to unreasonable risk.

Outside the work environment, TCEP presents unreasonable risk to adults, children, and infants with
acute, short-term/chronic, and lifetime exposure to TCEP, mainly from breathing or ingesting TCEP-
containing dust or eating TCEP-contaminated fish. TCEP presents unreasonable risk to children and
infants with acute and short-term/chronic exposure from mouthing consumer products that contain
TCEP. EPA also assessed whether breast-feeding infants were at higher risk than their mothers and
determined that they are not.

Unreasonable Risk to the Environment

Based on data for three fish species and predictive models for sediment-dwelling organisms, EPA
assessed TCEP exposures to the aquatic environment when TCEP leaches or is released into water
through the manufacturing, processing, or use of TCEP or TCEP-containing materials. EPA's
assessment preliminarily shows that chronic exposure to TCEP results in unreasonable risk to fish
from using TCEP as a laboratory chemical and to sediment-dwelling organisms for all uses that
were quantitatively evaluated. EPA preliminarily determined that acute exposure to TCEP does not
present unreasonable risk to aquatic organisms (vertebrate and invertebrate species). Data on soil
invertebrates and mammals indicate that acute and chronic exposure to TCEP does not present
unreasonable risks to land-dwelling animals.

Considerations and Next Steps

A total of 20 COUs were evaluated for TCEP (see Table 1-1). EPA preliminarily determined that the
following nine COUs contribute to the unreasonable risk that TCEP presents, considered singularly or in
combination with other TCEP exposures:

•	Manufacturing (import);

•	Processing - incorporation into formulation, mixture, or reaction product - paint and coating
manufacturing;

•	Processing - incorporation into formulation, mixture, or reaction product - polymers used in
aerospace equipment and products;

•	Processing - incorporation into article - aerospace equipment and products;

•	Commercial use - paints and coatings;

•	Commercial use - laboratory chemicals;

•	Consumer use - furnishing, cleaning, treatment/care products - fabric and textile products;

•	Consumer use - furnishing, cleaning, treatment/care products - foam seating and bedding
products; and

•	Consumer use - construction, paint, electrical, and metal products - building/construction
materials - wood and engineered wood products - wood resin composites.

The following five COUs were preliminary determined not to contribute to the unreasonable risk:

•	Processing - recycling;

•	Distribution in commerce;

Page 20 of 572

-------
905

906

907

908

909

910

911

912

913

914

915

916

917

918

919

920

921

922

923

924

925

926

927

928

929

930

931

932

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

•	Industrial use - aerospace equipment and products;

•	Commercial use - aerospace equipment and products; and

•	Consumer use - construction, paint, electrical, and metal products - building/construction
materials - insulation.

In addition, there are six COUs for which EPA does not have sufficient information to determine
whether they contribute to TCEP's unreasonable risks (see Section 5.3.2.3.2 and Section 6.3.1):

•	Commercial use - furnishing, cleaning, treatment/care products - fabric and textile products;

•	Commercial use - furnishing, cleaning, treatment/care products - foam seating and bedding
products;

•	Commercial use - construction, paint, electrical, and metal products - building/construction
materials - wood and engineered wood products - wood resin composites;

•	Commercial use - construction, paint, electrical, and metal products - building/construction
materials - insulation;

•	Consumer use - paints and coatings; and

•	Disposal.

It also is important to note that, in addition to the lack of information on six COUs, the estimates of risk
in the TCEP evaluation include assumptions and modeled predictions around which there are varying
levels of uncertainty. That being said, the totality of information and weight of the scientific evidence
give EPA confidence that under the known, intended, and reasonably foreseen COUs that are subject to
evaluation and regulation under TSCA, TCEP presents unreasonable risks to human health and the
environment.

This draft risk evaluation has been released for public comment and will undergo independent, expert
scientific peer review. EPA will issue a final TCEP risk evaluation in 2024 after considering input from
the public and peer reviewers. If in the final risk evaluation EPA determines that TCEP presents
unreasonable risk to human health or the environment, EPA will initiate regulatory action to mitigate
those risks.

Page 21 of 572

-------
933

934

935

936

937

938

939

940

941

942

943

944

945

946

947

948

949

950

951

952

953

954

955

956

957

958

959

960

961

962

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

1 INTRODUCTION

EPA has evaluated tris(2-chloroethyl) phosphate (TCEP) under the Toxic Substances Control Act
(TSCA). TCEP is primarily used as an additive flame retardant and plasticizer in polymers used in
aerospace equipment and products and as an additive flame retardant in paint and coating
manufacturing. In the past, TCEP was processed in many products made in the United States, including
fabrics and textiles, some types of foam, and construction materials—some of which may still be in use
today. TCEP may also be imported in articles intended for consumer use. Section 1.1 provides
production volume, life cycle diagram (LCD), conditions of use (COUs), and conceptual models used
for TCEP; Section 1.2 includes an overview of the systematic review process; and Section 1.3 presents
the organization of this draft risk evaluation. Figure 1-1 describes the major inputs, phases, and
outputs/components of the TSCA risk evaluation process, from scoping to releasing the final risk
evaluation.

Inputs

Existing Laws, Regulations,
and Assessments
Use Document
Public Comments

Public Comments on
Draft Scope Document

Analysis Plan
Testing Results
Data Evaluation Process
Data Integration

Public Comments on
Draft RE
• Peer Review Comments
on Draft RE

Phase

Outputs

Figure 1-1. TSCA Existing Chemicals Risk Evaluation Process

1.1 Scope of the Risk Evaluation

EPA evaluated risk to human and environmental populations for TCEP. Specifically for human
populations, the Agency evaluated risk to (1) workers and occupational non-users (ONUs) via inhalation
and oral routes; (2) workers via dermal routes; (3) consumers via inhalation, dermal, and oral routes; and
(4) the general population via oral, dermal, and inhalation routes. In this risk evaluation the general
population includes various subpopulations such as subsistence fishers and tribal populations. For
environmental populations, EPA evaluated risk to (1) aquatic species via water and sediment, and (2)
terrestrial species via air and soil leading to dietary exposure.

1.1.1 Life Cycle and Production Volume

The LCD shown below in Figure 1-2 depicts the COUs that are within the scope of the draft risk
evaluation during various life cycle stages, including manufacturing, processing, use (industrial,
commercial, consumer), distribution, and disposal. The LCD has been updated since it was included in
the TCEP final scope document (U.S. EPA. 2020b) to correspond with minor updates to the COUs. The
information in the LCD is grouped according to the Chemical Data Reporting (CDR) processing codes
and use categories, including functional use codes for industrial uses and product categories for

Page 22 of 572

-------
963

964

965

966

967

968

969

970

971

972

973

974

975

976

977

978

979

980

981

982

983

984

985

986

987

988

989

990

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

industrial, commercial, and consumer uses. The CDR Rule under TSCA requires U.S. manufacturers
(including importers) to provide EPA with information on the chemicals they manufacture or import into
the United States. EPA collects CDR data approximately every 4 years with the latest collections
occurring in 2006, 2012, 2016, and 2020.

Descriptions of the industrial, commercial, and consumer use categories identified from the CDR are
included in the LCD (Figure 1-2) (U.S. EPA. 2016d). The descriptions provide a brief overview of the
use category; the Supplemental Information on Environmental Release and Occupational Exposure
Assessment (U.S. EPA. 20231) contains more detailed descriptions (e.g., process descriptions, worker
activities, process flow diagrams, equipment illustrations) for each manufacture, processing, use, and
disposal category.

Because TCEP is also known to co-occur in formulation with other flame retardants, such as 2,2-
bis(chloromethyl)-propane-l,3-diyltetrakis(2-chloroethyl) bisphosphate (V6), this draft risk evaluation
evaluates TCEP when it co-occurs with other flame retardants in commercial and consumer products
(e.g., when it co-occurs with V6). However, it does not evaluate the other flame retardants.

TRIS(2-CHLOROETHYL) PHOSPHATE (TCEP) (CAS RN l-BS-8)
MFG/IMPORT PROCESSING INDUSTRIAL, COMMERCIAL, CONSUMER USES

Incorporation into
Formulation, Mixture, or
Reaction product

(Flame retardant in paint

and coating
manufacturing; Polymers

used in aerospace
equipment and products)

Incorporation into articlt

(Aerospace equipment and
products)

Recycling

Paints and CoatingV

Other use*

Aerospace equipment and
products; laboratory
chemicals

Furnishing, Cleaning,
Treatment/Care Product^2

Fabric and textile products;
foam setting and bedding
products

Construction, Paint,
Electrical, and Metal
Products*-2

Building/construction
materials not covered
elsewhere; wood resin
composites and insulation

RELEASES and WASTE
DISPOSAL

Disposal

See Conceptual Model fo
En vironmentaI Releases
and Wastes

~
~

Manufacture (Including
Import)

Processing

Uses:

1. Industrial/Commercial

2. Consumer

Figure 1-2. TCEP Life Cycle Diagram

1 Due to lack of reasonably available data, including current CDR data, EPA cannot differentiate between import
and processing sites.

2 See Table 1-1 for additional details on uses.

As evident in Figure 1-3, import, production volume, and uses of TCEP in the United States have
curtailed in recent years. Although CDR data show production volumes for TCEP in chemical form in
the tens of thousands of pounds from 2012 to 2015, the most recent updated 2020 CDR data showed that
no company reported the manufacture (including import) of TCEP in the United States from 2016 to
2020. However, the reporting threshold for TCEP in CDR is 25,000 lb and some manufacturing could be

Page 23 of 572

-------
991

992

993

994

995

996

997

998

999

1000

1001

1002

1003

1004

1005

1006

1007

1008

1009

1010

1011

1012

1013

1014

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

occurring below that threshold ( )20a).1 The production volumes for TCEP reported to CDR

for years 2012 to 2015 were all from one company, Aceto US LLC, a chemical manufacturer and
supplier importing TCEP in chemical form. Aceto US LLC indicated to EPA that TCEP was imported
and used as a flame retardant for unsaturated polyester resins and for aircraft furniture (

2020b). Note that prior to 2012, production volume in CDR was reported in ranges. From 1986 to 2002,
the production volume reported to CDR (previously known as the Inventory Update Rule, or IUR) was
between 1 and 10 million lb. In 2006, the production volume reported was between 500,000 and 1
million lb and in 2011 the production volume was withheld.

To supplement the CDR data, EPA also considered Datamyne import volume information that shows
593 lb of TCEP imported in 2020. Descartes Datamyne is a commercial searchable trade database that
covers the import-export data and global commerce of more than 50 countries (across 5 continents) and
includes cross-border commerce of the United States with over 230 trading partners (Descartes. 2020).
The trade data are gathered from the U.S. Customs Automated Manifest System. Since 2014, total
imports of TCEP in chemical form range in volume over the time from approximately 96,823 lb (in
2014) to 593 lb (in 2020) (Descartes. 2020). Note that for 2014, the Aceto US LLC data is included in
the total production volume for CDR and Datamyne. For 2020, Sigma Aldrich Corp reported the 593
lb.2

The 2016 CDR reporting data and Datamyne import volume data for TCEP in chemical form are
provided in Figure 1-3. TCEP imported in articles is not captured in these data. Note, EPA only recently
added TCEP to the Toxics Release Inventory (TRI) with the first year of reporting from facilities due
July 1, 2024.

1 Note that because CDR generally does not include information on impurities or manufacturing solely in small quantities for
research and development, and because small manufacturers are exempt from 2020 CDR reporting, some manufacturing
could be occuring at small manufacturers. However, EPA does not consider domestic manufacturing of TCEP to be
reasonably foreseeable. Lastly, TCEP imported in articles would not be captured in CDR.

2 Due to the nature of Datamyne data, some shipments containing TCEP may be excluded due to being categorized under
other names that were not included in the search terms. There also may be errors in the data that prevent shipment records
containing the chemical from being located. Datamyne does not include articles/products containing the chemical unless the
chemical name is included in the description; however, based on descriptions provided on the bills of lading, Figure 1-3
provides an estimate of the volume of TCEP imported as the chemical (not in an identified product) from 2012 to 2020.

Page 24 of 572

-------
1015

1016

1017

1018

1019

1020

1021

1022

1023

1024

1025

1026

1027

1028

1029

1030

1031

1032

1033

1034

1035

1036

1037

1038

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Reported Aggregate TCEP Production Volume (lbs.) by Year

i.L.J

2012 2013 2014 2015 2016 2017 2018 2019 2020
¦ CDR Production Volume ¦ Datamyne Import Volume

Figure 1-3. Reported Aggregate TCEP Production Volume (lb) 2012-2020

Note: CDR data for the 2016 reporting period is available via ChemView. Because of an ongoing CBI
substantiation process required by amended TSCA, the CDR data available in this draft risk evaluation is more
specific than currently provided in ChemView (U.S. EPA. 2019a). For 2014, Aceto US LLC's production volume
is included in both the CDR data and the Datamyne data.

Given the uncertainties in the current production volume for TCEP, EPA used two production volumes
in its analyses for this draft risk evaluation: 2,500 and 25,000 lb. The 2,500 lb production volume is used
as a more realistic estimate reflecting current production volumes, while 25,000 lb is used as an upper
bound estimate based on the 2020 CDR reporting threshold. There are several reasons why EPA
considers 2,500 lb to be a more realistic production volume. First, the decreasing aggregate TCEP
production volumes according to CDR and Datamyne, as shown in Figure 1-3, suggest that the
production volume is now somewhere below the 2020 CDR reporting threshold of 25,000 lb, with
Datamyne showing 593 lb of TCEP imported in 2020 and generally the most recent Datamyne
information (2017 to 2020) in the low thousands of pounds or lower. Additionally, EPA received public
comments (EPA-HQ-OPPT-2018-0476-0041) on the final scope document (U.S. EPA. 2020b)
confirming industry's transition away from the domestic use of TCEP.

Communication with industry further supported the declining use of TCEP as many companies have
since discontinued or reformulated products that contained TCEP, even though TCEP is still in use for
several commercial and consumer COUs (EPA-HQ-OPPT-2018-0476-0056). However, there is no
federal ban on the manufacture, process, or use of TCEP that would prevent production volumes from
increasing again (see Appendix B for the regulatory history of TCEP). Therefore, EPA used these two

160,000

140,000

120,000

100,000

(/)

§ 80,000
o

60,000

40,000

20,000

Page 25 of 572

-------
1039

1040

1041

1042

1043

1044

1045

1046

1047

1048

1049

1050

1051

1052

1053

1054

1055

1056

1057

1058

1059

1060

1061

1062

1063

1064

1065

1066

1067

1068

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

production volumes to characterize what is possible and what is realistic given reasonably available
information. Given EPA's research, the 25,000 lb upper bound production volume is believed to be an
overestimate of current production volumes in the United States. For these reasons, the 2,500 lb
production volume is used throughout this draft risk evaluation as EPA has more confidence that it is
reflective of current production volumes. Estimates using the upper bound of 25,000 lb are presented in
appendices and supplemental files.

1 »1,2 Conditions of Use Included in the Draft Risk Evaluation

The Final Scope of the Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) CASRN115-96-8
(I 1020b) identified and described the life cycle stages, categories and subcategories that

comprise COUs that EPA planned to consider in the risk evaluation. All COUs for TCEP included in
this draft risk evaluation are reflected in the LCD (Figure 1-2) and conceptual models (Section 1.1.2.1).
Table 1-1 below presents all COUs for TCEP.

In this draft risk evaluation, EPA made edits to the COUs listed in the final scope document. These edits
reflect EPA's improved understanding of the COUs based on further outreach and public comments
received, which have been added to the reference(s) column of Table 1-1. Changes include removing
"flame retardant" as the exclusive functional use in the processing conditions of use; editing industrial
and commercial use in "aircraft interiors and products" to "aerospace equipment and products"; and
improved the description of the COU to avoid using the "products not covered elsewhere" description
from the CDR reporting codes. EPA did not receive public comments on additional commercial uses
that fall into the "Other use" category aside from laboratory chemicals, the Agency removed "e.g.,"
from the COU, "Commercial use - other use - e.g., laboratory chemicals."

All COUs assessed in this Risk Evaluation are considered on-going uses. However, there are several
COUs for which part of the life cycle has ceased, such as manufacturing (including import) and
processing. However, other parts of the lifecycle including recycling, commercial or consumer use, and
disposal are on-going. These COUs are identified in Table 1-1 and include four COUs for commercial
use and five COUs for consumer use.

Table 1-1. Cont

itions of Use in the Risk Evaluation for TCEP

Life Cycle
Stage"

Category''

Subcategory'

Reference(s)

Manufacturing

Import

\(2016d)

Processing

Processing -
incorporation into
formulation, mixture, or
reaction product

Paint and coating
manufacturing

(U.S. EPA. 2019a: Duratec. 2018: U.S.
EPA. 2017b: PPG. 2016. 2010)

Flame Control Coatings_meeting
memo

Processing -
incorporation into
formulation, mixture, or
reaction product

Polymers used in
aerospace equipment
and products

EPA-HQ-OPPT-2018-0476-0015;
EPA-HQ-OPPT-2018-0476-0012: BJB
Enterprises ( . EPA-HO-OPPT-

2018-0476-0045; Summary of email
exchanges

Processing -
incorporation into article

Aerospace equipment
and products

EPA-HQ-OPPT-2018-0476-0006;
EPA-HQ-OPPT-2018-0476-0045;
Boeing meeting memo

Recycling

(U.S. EPA. 2019a)

Page 26 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Life Cycle
Stage"

Category''

Subcategory'

Reference(s)

Distribution in
Commerce

Distribution in
commerce

Distribution in
commerce



Industrial Use

Other use

Aerospace equipment
and products

EPA-HQ-OPPT-2018-0476-0006;
Boeing meeting memo



Other use

Aerospace equipment
and products

EPA-HQ-OPPT-2018-0476-0006



Paints and coatings

Paints and coatings

4 (2019a); Alliance for
Automotive Innovation



Laboratory chemicals

Laboratory chemical

America (2018)



Furnishing, cleaning,
treatment/care products

Fabric and textile
products^

EPA-HQ-OPPT-2018-0476-0015

Commercial
Use

Furnishing, cleaning,
treatment/care products

Foam seating and
bedding products^

Staoleton et al. ( ; Stapleton &
Hammel meeting memo



Construction, paint,
electrical, and metal
products

Building/construction
materials - insulation^

EPA-HQ-OPPT-2018-0476-0015;
EPA-HQ-OPPT-2018-0476-0041; EC

(2009). cites I ARC (1990)



Construction, paint,
electrical, and metal
products

Building/construction
materials - wood and
engineered wood
products - wood resin
composites^

EC (2009). cites I ARC (1990). OECD
(2006); IPCS (1998)



Paints and Coatings

Paints and coatings^

4 (2019a); Alliance for
Automotive Innovation



Furnishing, cleaning,
treatment/care products

Fabric and textile
products^

EPA-HQ-OPPT-2018-0476-0015



Furnishing, cleaning,
treatment/care products

Foam seating and
bedding products^

Staoleton et al. ( ; Stapleton &
Hammel meeting memo

Consumer Use

Construction, paint,
electrical, and metal
products

Building/construction
materials - insulation^

EPA-HQ-OPPT-2018-0476-0015;
EPA-HQ-OPPT-2018-0476-0041; EC

(2009). cites I ARC (1990)



Construction, paint,
electrical, and metal
products

Building/construction
materials -wood and
engineered wood
products - wood resin
composites^

EC (2009). cites I ARC (1990). OECD
(2006); IPCS (1998)

Disposal

Disposal

Disposal®



a Life Cycle Stage Use Definitions (40 CFR 711.3)

-	"Industrial Use" means use at a site at which one or more chemicals or mixtures are manufactured (including
imported) or processed.

-	"Commercial Use" means the use of a chemical or a mixture containing a chemical (including as part of an
article) in a commercial enterprise providing saleable goods or services.

-	"Consumer Use" means the use of a chemical or a mixture containing a chemical (including as part of an
article, such as furniture or clothing) when sold to or made available to consumers for their use.

Page 27 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Life Cycle
Stage"

Category''

Subcategory'

Reference(s)

- Although EPA has identified both industrial and commercial uses here for purposes of distinguishing scenarios
in this document, the Agency interprets the authority over "any manner or method of commercial use" under
TSCA section 6(a)(5) to reach both.
b These categories of COU appear in the LCD, reflect CDR codes, and broadly represent COUs of TCEP in
industrial and/or commercial settings and for consumer uses.
c These subcategories reflect more specific COUs of TCEP.
d Manufacturing (including import) and processing for these COUs has ceased.

'' This COU use includes associated disposal of those COUs for which manufacturing (including import) and
processing have ceased.

1069	1.1.2.1 Conceptual Models

1070	The conceptual model in Figure 1-4 presents the exposure pathways, exposure routes, and hazards to

1071	human populations from industrial and commercial activities and uses of TCEP. Figure 1-5 presents the

1072	conceptual model for consumer activities and uses, Figure 1-6 presents general population exposure

1073	pathways and hazards for environmental releases and wastes, and Figure 1-7 presents the conceptual

1074	model for ecological exposures and hazards from environmental releases and wastes.

Page 28 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

INDUSTRIAL AND COMMERCIAL
ACTIVITIES/USES

EXPOSURE PATHWAY

EXPOSURE ROUTE

POPULATIONS

HAZARDS

1075

1076

1077

1078

Manufacture (incl.
Import)

Processing:
Incorporation into
formulation, mixture,
or reaction product
Incorporation into
article

Recycling

Painting and Coatings

Other use

Furnishing, Cleaning,
Treatment/Care
Products

Construction, Paint,
Electrical,and Metal
Products

Waste Handling, Treatment,
and Disposal

	1

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

CONSUMER ACTIVITIES/USES	EXPOSURE PATHWAYS	EXPOSURE ROUTES	POPULATIONS	HAZARDS

Conceptual Models)

1079

1080	Figure 1-5. TCEP Conceptual Model for Consumer Activities and Uses: Potential Exposures and Hazards

1081	The conceptual model presents the exposure pathways, exposure routes, and hazards to human populations from consumer activities and uses of TCEP.

1082

Page 30 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

RELEASES AND WASTES FROM
INDUSTRIAL/COMMERCIAL/
CONSUMER USES

Industrial Pre1

	 Treatment or

Industrial

	wyvT

EXPOSURE PATHWAYS

EXPOSURE
ROUTES

POPULATIONS

HAZARDS

Wastewater or
Liquid Wastes

Solid Wastes
Liquid Wastes

Emissions to
Air

1083

1084

1085

1086

Indirect discharge
T

~ POTW

^ Underground
Injection

Hazardous and
-> Municipal
Waste Landfill

Fish Ingestion—

1

Hazardous and
Municipal

Waste
Incinerators

Off-site Waste
Transfer

RecycTing,

Other
Treatment

Hazards Potentially
Associated with
Acute and/or Chronic
Exposures

Figure 1-6. TCEP Conceptual Model for Environmental Releases and Wastes: General Population Hazards

The conceptual model presents the exposure pathways, exposure routes, and hazards to human populations from releases and wastes from industrial,
commercial, and/or consumer uses of TCEP.

Page 31 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

RELEASES AND WASTES FROM

INDUSTRIAL/COMMERCIAL/	EXPOSURE PATHWAYS	POPULATIONS	HAZARDS

CONSUMER USES

1087

1088	Figure 1-7. TCEP Conceptual Model for Environmental Releases and Wastes: Ecological Exposures and Hazards

1089	The conceptual model presents the exposure pathways, exposure routes, and hazards to environmental populations from releases and wastes from

1090	industrial, commercial, and/or consumer uses of TCEP.

Page 32 of 572

-------
1091

1092

1093

1094

1095

1096

1097

1098

1099

1100

1101

1102

1103

1104

1105

1106

1107

1108

1109

1110

1111

1112

1113

1114

1115

1116

1117

1118

1119

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

_M.3 Populations Assessed	

Based on the conceptual models presented in Section 1.1.2.1, Figure 1-8 presents the human and
ecological populations assessed in this draft risk evaluation. Specifically for humans, EPA evaluated risk
to workers and ONUs via inhalation routes and risk to workers via dermal routes; risk to consumers via
inhalation, dermal, and oral routes; risk to the general population via oral, dermal, and inhalation routes.
For environmental populations, EPA evaluated risk to aquatic species via water and sediment, and risk
to terrestrial species via air, soil, and water leading to dietary exposure. Human health risks were
evaluated for acute, short-term/subchronic, chronic, and lifetime exposure scenarios as appropriate, and
environmental risks were evaluated for acute and chronic exposure scenarios, as applicable based on
reasonably available exposure and hazard data as well as the relevant populations for each. All
consumers of products containing TCEP were considered users of those products, and bystanders were
not assessed separately because all the consumer COUs assessed were article scenarios. For the purposes
of article exposures, consumers and bystanders are considered the same.

Environmental

Aquatic

Terrestrial

Surface
Water

Sediment

Soil

Air

AquaticSpecies

\

. /

Terrestrial
Species

f	 	N.

Surface
water

v 	y

Human Health

(includes PESS*)

Occupational

(includes adolescents and
women of reproductive
age)

V	

Consumer

(includes children)

Workers

Inhalation
Dermal

ONUs

Inhalation

Users

Inhalation

Ingestion

Dermal

*PESS: Potentially exposed orsusceptible subpopulations

General
Population

(includes fenceline)

v	y

/"	N

All lifestages
\	

•	Inhalation

•	Ingestion

•	Dermal

Figure 1-8. Populations Assessed in this Draft Risk Evaluation

1.1.3.1 Potentially Exposed or Susceptible Subpopulations

TSCA Section 6(b)(4)(A) requires that risk evaluations "determine whether a chemical substance
presents an unreasonable risk of injury to health or the environment, without consideration of costs or
other non-risk factors, including an unreasonable risk to a potentially exposed or susceptible
subpopulation identified as relevant to the risk evaluation by the Administrator, under the conditions of
use." TSCA section 3(12) states that "the term 'potentially exposed or susceptible subpopulation'' means
a group of individuals within the general population identified by the Administrator who, due to either
greater susceptibility or greater exposure, may be at greater risk than the general population of adverse
health effects from exposure to a chemical substance or mixture, such as infants, children, pregnant
women, workers, or the elderly."

This draft risk evaluation considers potentially exposed or susceptible subpopulations (PESS)
throughout the human health risk assessment (Section 5). Considerations related to PESS can influence

Page 33 of 572

-------
1120

1121

1122

1123

1124

1125

1126

1127

1128

1129

1130

1131

1132

1133

1134

1135

1136

1137

1138

1139

1140

1141

1142

1143

1144

1145

1146

1147

1148

1149

1150

1151

1152

1153

1154

1155

1156

1157

1158

1159

1160

1161

1162

1163

1164

1165

1166

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

the selection of relevant exposure pathways, the sensitivity of derived hazard values, the inclusion of
particular human populations, and the discussion of uncertainties throughout the assessment.

Evaluation of the qualitative and quantitative evidence for PESS begins as part of the systematic review
process, where any available relevant published studies and other data are identified. If adequate and
complete, this evidence informs the derivation of exposure estimates and human health hazard
endpoints/values that are protective of PESS.

EPA has identified a list of specific PESS factors that may contribute to a group having increased
exposure or biological susceptibility, such as lifestage, occupational and certain consumer exposures,
nutrition, and lifestyle activities. For TCEP, the Agency identified how the risk evaluation addressed
these factors as well as any remaining uncertainties. For the TCEP draft risk evaluation, EPA accounted
for the following PESS groups: infants exposed through human milk from exposed individuals, children
and male adolescents who use consumer articles or among the exposed general population, subsistence
fishers, tribal populations, pregnant women, workers and consumers who experience aggregated or
sentinel exposures, fenceline communities who live near facilities that emit TCEP, and firefighters. See
Section 5.3.3 and Appendix D for details related to this analysis.

1.2 Systematic Review

The U.S. EPA's Office of Pollution Prevention and Toxics (EPA/OPPT) applies systematic review
principles in the development of risk evaluations under the amended TSCA. TSCA section 26(h)
requires EPA to use scientific information, technical procedures, measures, methods, protocols,
methodologies, and models consistent with the best available science and base decisions under section 6
on the weight of scientific evidence. Within the TSCA risk evaluation context, the weight of the
scientific evidence is defined as "a systematic review method, applied in a manner suited to the nature of
the evidence or decision, that uses a pre-established protocol to comprehensively, objectively,
transparently, and consistently identify and evaluate each stream of evidence, including strengths,
limitations, and relevance of each study and to integrate evidence as necessary and appropriate based
upon strengths, limitations, and relevance" (40 CFR 702.33).

Systematic review supports the risk evaluation in that data searching, screening, evaluation, extraction,
and evidence integration and is used to develop the exposure and hazard assessments based on
reasonably available information. EPA defines "reasonably available information" to mean information
that EPA possesses or can reasonably obtain and synthesize for use in risk evaluations, considering the
deadlines for completing the evaluation (40 CFR 702.33).

In response to comments received by the National Academies of Sciences, Engineering, and Medicine
(NASEM), TSCA Scientific Advisory Committee on Chemicals (SACC) and public, EPA developed the
Draft Systematic Review Protocol Supporting TSCA Risk Evaluations for Chemical Substances (U.S.
E 21) to describe systematic review approaches implemented in TSCA risk evaluations. In
response to recommendations for chemical specific systematic review protocols, the Draft Risk
Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Systematic Review Protocol (	?23n)

(also referred to as the "TCEP Systematic Review Protocol") describes clarifications and updates to
approaches outlined in the 2021 Draft Systematic Review Protocol that reflect NASEM, SACC and
public comments as well as chemical-specific risk evaluation needs. For example, EPA has updated the
data quality evaluation process and will not implement quantitative methodologies to determine both
metric and overall data or information source data quality determinations. Screening decision
terminology (e.g., "met screening criteria" as opposed to "include") was also updated for greater
consistency and transparency and to more appropriately describe when information within a given data

Page 34 of 572

-------
1167

1168

1169

1170

1171

1172

1173

1174

1175

1176

1177

1178

1179

1180

1181

1182

1183

1184

1185

1186

1187

1188

1189

1190

1191

1192

1193

1194

1195

1196

1197

1198

1199

1200

1201

1202

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

source met discipline-specific title and abstract or full-text screening criteria. Additional updates and
clarifications relevant for TCEP data sources are described in greater detail in the TCEP Systematic
Review Protocol (U.S. EPA. 2023n).

The systematic review process is briefly described in Figure 1-9 below. Additional details regarding
these steps are available in the 2021 Draft Systematic Review Protocol (U.S. EPA. 2021). Literature
inventory trees and evidence maps for each discipline (e.g., human health hazard) displaying results of
the literature search and screening, as well as sections summarizing data evaluation, extraction, and
evidence integration are included in the TCEP Systematic Review Protocol (U.S. EPA. 2023n).

• Based on the
approach
described in the
Literature
Search Strategy
documents.

• Title/abstractand
full-text screening
based on pre-
defined

inclusion/exclusion
criteria.

• Evaluateand
document the
quality of studies
based on pre-
defined criteria.

Data Search

Data Screen

'j§>

Data

Evaluation

~ —
~ —
~ —
~ —

Figure 1-9. Diagram of the Systematic Review Process

• Extract relevant

information based

on pre-defined



templates.



Data



Extraction

	 1



|

• Evaluate results
both within and
across evidence
streams to develop
weight of the
scientific evidence
conclusions.

Evidence
Integration



EPA used reasonably available information, defined in 40 CFR 702.33, in a fit-for-purpose approach, to
develop a risk evaluation that relies on the best available science and is based on the weight of the
scientific evidence in accordance with TSCA sections 6 and 26. EPA reviewed reasonably available
information and evaluated the quality of the methods and reporting of results of the individual studies
using the evaluation strategies described in the 2021 Draft Systematic Review Protocol (U.S. EPA.
2021) and the TCEP Systematic Review Protocol (U.S. EPA. 2023n).

EPA also identified key assessments conducted by other EPA programs and other U.S. and international
organizations. Depending on the source, these assessments may include information on COUs (or the
equivalent), hazards, exposures, and potentially exposed or susceptible subpopulations. Some of the
most pertinent assessments that were consulted for TCEP include the following:

•	U.S. EPA's 2009 Provisional Peer-Reviewed Toxicity Values (PPR1V) for Tris(2-
chloroethvDphosphate (TCEP) (CASRN115-96-8):

•	2009 European Union Risk Assessment Report: CAS: 115-96-8: Tris (2-chloroethyl) phosphate,
TCEP:

•	Environment Canada and Health Canada's 2009 Screening Assessment for the Challenge
Ethanol, 2-chloro- phosphate (3:1) (Tris(2-chlrorethyl) phosphate fTCEPl):

•	Australia's 2016 Ethanol, 2-chloro- phosphate (3:1): Human health tier II assessment:

•	Australia's 2017 Ethanol, 2-chloro- phosphate (3:1): Human health tier III assessment:

•	ATSDR's 2012 ToxicologicalProfile for Phosphate Ester Flame Retardants:

•	NTP's 1991 Technical Report on Toxicology and Carcinogenesis Studies of Tris(2-chloroethyl)
Phosphate (CASRN 115-96-8) in F344 N Rats and B6C3F1 Mice (Gavage Studies) : and

•	IARC's 1999 Monographs on the Evaluation of Carcinogenic Risks to Humans Volume 71.

Page 35 of 572

-------
1203

1204

1205

1206

1207

1208

1209

1210

1211

1212

1213

1214

1215

1216

1217

1218

1219

1220

1221

1222

1223

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

1.3 Organization of the Risk Evaluation

This draft risk evaluation for TCEP includes five additional major sections, a list of REFERENCES, and
several APPENDICES. Section 2 summarizes basic physical-chemical characteristics as well as the fate
and transport of TCEP. Section 3 includes an overview of releases and concentrations of TCEP in the
environment. Section 4 provides a discussion and analysis of the environmental risk assessment,
including the environmental exposure, hazard, and risk characterization based on the COUs for TCEP.
Section 5 presents the human health risk assessment, including the exposure, hazard, and risk
characterization based on the COUs. Section 5 also includes a discussion of PESS based on both greater
exposure and/or susceptibility, as well as a description of aggregate and sentinel exposures. Sections 4
and 5 both discuss any assumptions and uncertainties and how they impact the draft risk evaluation.
Finally, Section 6 presents EPA's proposed determination of whether the chemical presents an
unreasonable risk to human health or the environment as a whole chemical approach and under the
assessed COUs.

Appendix A provides a list of abbreviations and acronyms as well a glossary of select terms used
throughout this draft risk evaluation. Appendix B provides a brief summary of the federal, state, and
international regulatory history of TCEP. Appendix C lists all separate supplemental files associated
with this draft risk evaluation, which can be accessed through hyperlinks included in the references. All
subsequent appendices include more detailed analysis and discussion than are provided in the main body
of this draft risk evaluation for TCEP.

Page 36 of 572

-------
1224

1225

1226

1227

1228

1229

1230

1231

1232

1233

1234

1235

1236

1237

1238

1239

1240

1241

1242

1243

1244

1245

1246

1247

1248

1249

1250

1251

1252

1253

1254

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

2 CHEMISTRY AND FATE AND TRANSPORT OF TCEP

Physical and chemical properties determine the behavior and characteristics of a chemical that inform its
condition of use, environmental fate and transport, potential toxicity, exposure pathways, routes, and
hazards. Environmental fate and transport include environmental partitioning, accumulation,
degradation, and transformation processes. Environmental transport is the movement of the chemical
within and between environmental media, such as air, water, soil, and sediment. Transformation or
degradation occur through reaction of the chemical in the environment. Thus, understanding the
environmental fate of TCEP informs the determination of the specific exposure pathways, and potential
human and environmental populations that EPA considered in this draft risk evaluation.

2.1 Physical and Chemical Properties

EPA gathered and evaluated physical and chemical property data and information according to the
process described in the 2021 Draft Systematic Review Protocol (	2021). During the

evaluation of TCEP, EPA considered both measured and estimated physical and chemical property
data/information summarized in Table 2-1, as applicable. More details are given in Appendix E.l.
Information on the full, extracted dataset is available in the supplemental file Draft Risk Evaluation for
Tris (2-chloroethyl) Phosphate (TCEP) - Systematic Review of Data Quality Evaluation and Data
Extraction Information for Physical and Chemical Properties (U.S. EPA. 2023f).

TCEP is a clear, transparent liquid with a slight odor (DOE. 2016; U.S. EPA. 2015b; ECB. 2009; Lewis
and Hawlev. 2007; Weil. 2001) and low viscosity (\ Wli, r ^ \)). As a chlorinated phosphate ester,

TCEP is used as a flame-retardant additive and plasticizer that melts around -55 °C and begins to
decompose at 320 °C (MI , I * n \ b; Toscano and Coleman. 2012; ECB. 2009; IARC.
1990). TCEP is appreciably soluble in water with water solubility of 7,820 mg/L at 20 °C and a low log
K,)\v (1.78) ( * ii \ K 2015b; EC. 2009; ECB. 200"; \ cibmeeen et at.. 2005). With a vapor
pressure of 0.0613 nimHg at 25 °C 0 v >1 P \ l %; Dobry and Kelln l ) and a boiling point of
330 °C(! v n \ 2019b; DO ,i > H \	^ oes. 2014; Toscano and Coleman. 2012).

TCEP has low volatility and is categorized as a semi-volatile organic compound (SVOC) (ECHA. 2018;
TERA. 2015). However, TCEP will become more volatile when the temperature increases (0.5 nimHg at
145 °C) (Toscano a ;man. 2012; NTP. 1992).

Table 2-1. Physical and Chemical Properties of TCEP

Property

Selected Value"

Reference(s)

Overall Quality
Determination''

Molecular formula

C6H12CI3O4P





Molecular weight

285.49 g/mol





Physical form

Clear, transparent liquid
with slight odor

(DOE. 2016; U.S. EPA. 2015b; ECB.
2009; Lewis and Hawlev, 2007; Weil,
2001)

High

Melting point

-55 °C

(DOE, 2016; U.S. EPA, 2015a. b;
Toscano and Coleman. 2012)

High

Boiling point

330 °C

(U.S. EPA. 2019b; DOE. 2016; U.S.
EPA. 2015a; Havnes. 2014; Toscano and

Coleman, 2012)

High

Density

1.39 g/cm3 at 25 °C

(DOE. 2016; Havnes. 2014; Toscano and
Coleman. 2012)

High

Page 37 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Property

Selected Value"

Reference(s)

Overall Quality
Determination''

Vapor pressure

0.0613 mmHg at 25 °C

1957)

High

Vapor density

9.8 (air = 1)

( 19)

High

Water solubility

7,820 mg/L at 20 °C

(U.S. EPA. 2015b: EC. 2009: ECB.
2009; Verbraeeen et aL 2005)

High

Octanol: water partition
coefficient (log Kow)

1.78

ru.s. EPA. 2015b: EC. 2009: ECB.
2009: Verbreiaaen et aL 2005)

High

Octanol:air partition
coefficient (log Koa)

7.86 to 7.93

(Okeme et aL 2020)

High

Henry's Law constant

2.945E-06 atmm3/mol at
25 °C (calculated)

ru.s. EPA. 2012d)

High

Flash point

225 °C (closed cup)

ru.s. EPA. 2015a)

High

Autoflammability

480 °C

( 19: ECB. 2009)

Medium

Viscosity

45 cP at 20 °C

( C. 1990)

High

Refractive index

1.4721

(Havnes. 2014; Dobrv and Keller, 1957)

High

a Measured unless otherwise noted

b "Overall Quality Determinations" apply to all references listed in this table

Page 38 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

1255

2.2 Environmental Fate and Transport

TCEP - Environmental Fate and Transport (Section 2.2):

Key Points

EPA evaluated the reasonably available information to characterize the environmental fate and
transport of TCEP, the key points are summarized below:

•	TCEP exists in both gaseous and particle phases under environmentally relevant conditions
and partitions to organic carbon in the air. TCEP is not expected to undergo significant direct
photolysis, but TCEP in the gaseous phase will rapidly degrade in the atmosphere (ti/2 = 5.8
hours).

•	TCEP is not expected to undergo abiotic degradation processes such as photolysis and
hydrolysis in aquatic environments under environmentally relevant conditions. However,
TCEP's rate of hydrolysis is highly dependent on pH and temperature conditions.

•	TCEP does not biodegrade in water under aerobic conditions but will volatilize from surface
water despite its low Henry's Law constant (2.945xlCT6 atmm3/mol at 25 °C).

•	TCEP can be transported to sediment from overlying surface water through advection and
dispersion of dissolved TCEP and deposition of suspended solids containing TCEP.
However, TCEP may partition between surface water and sediments to varying degrees
because of its wide range of log Koc values (2.08 to 3.46) and high water solubility (7,820
mg/L), which could contribute to its mobility in the environment.

•	TCEP accumulation in soil is unlikely because of its log Koc values. Due to its high water
solubility and despite its low Henry's Law constant, TCEP in moist soil will both migrate to
groundwater and volatilize.

•	TCEP will be minimally removed via conventional drinking water and wastewater treatment
and will be retained in wastewater effluents with a low fraction being adsorbed onto sludge.

•	TCEP has been detected in surface water and groundwater samples; point sources include
wastewater effluents and landfill leachates.

•	TCEP has been detected in surface water, air, and snow in remote locations with no known
source of releases but is known to undergo long-range transport through atmospheric, plastic
debris, and other natural processes.

•	TCEP does not bioaccumulate in aquatic fish but may in benthic fish. When TCEP
concentrations are transferred to higher trophic levels in the food web, trophic dilution
occurs.

•	Overall, TCEP appears to be a persistent mobile organic compound (PMOC). PMOCs can
dissolve in water or bind to particles, resulting in longer environmental half4ives and greater
potential for long-range transport—especially in the air, water, and sediment compartments.

1256

1257

1258

1259

1260

1261

1262

2.2.1 Fate and Transport Approach and Methodology

Reasonably available environmental fate data—including biotic and abiotic biodegradation rates,
removal during wastewater treatment, volatilization from lakes and rivers, and organic carbon:water
partition coefficient (log Koc)—are the parameters used in the current draft risk evaluation. In assessing
the environmental fate and transport of TCEP, EPA considered the full range of results from data
sources that were rated high-quality. Information on the full extracted dataset is available in the
supplemental file Draft Risk Evaluation for Tris (2-chloroethyl) Phosphate (TCEP) - Systematic Review

Page 39 of 572

-------
1263

1264

1265

1266

1267

1268

1269

1270

1271

1272

1273

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

of Data Quality Evaluation and Data Extraction Information for Environmental Fate and Transport
0 v < < \ 3r). Other fate estimates were based on modeling results from EPI Suite™ (U.S. EPA.
2012d), a predictive tool for physical and chemical properties and environmental fate estimation.3
Information regarding the model inputs is available in Appendix E.

Table 2-2 provides selected environmental fate data that EPA considered while assessing the fate of
TCEP and were updated after publication of Final Scope of the Risk Evaluation for Tris(2-chloroethyl)
Phosphate (TCEP) CASRN115-96-8 (	20b) with additional information identified through

the systematic review process.

Table 2-2. Environmental Fate Properties of TCEP

Property or
Endpoint

Value"

Reference(s)

Overall Quality
Determination

Indirect

photodegradation

ti/2 =5.8 hours (based on -OH rate constant of
2.2E-11 cm3/mole-sec at 25 °C and 12-hour day
with 1.5E06 -OH/cm3; estimated)6

(U.S. EPA. 2012d)

High

Direct

photodegradation

Not expected to be susceptible to direct photolysis
by sunlight because the chemical structure of TCEP
does not contain chromophores that absorb at
wavelengths >290 nm

(HSDB. 2015)

High

Hydrolysis half-
life

ti/2 = 2 years at pH 8 and 25 °C (estimated)

(Saint-Hilaire et ah.

2011)

High

ti/2 = 0.083 days at pH 13; no significant degradation
observed over 35 days at pH 7, 9, and 11

(Su et al.. 2016)

Aerobic
biodegradation

Water: 13% and 4% /28 days (OECD 301B) at 10
and 20 mg/L test substance concentration in
activated domestic sludge, adaption not specified

(Life Sciences
Research Ltd.
1990b)

High

Soil: DT50 = 17.7 days; 78%/40 days based on test
substance concentration of 50 (ig/kg

(Hurtado et al..
2017)

Anaerobic
biodegradation

No data





Bioconcentration
factor (BCF)
(L/kg, unless
noted)

Whole body BCF = 0.31 ± 0.06, 0.16 ± 0.03, and
0.34 ± 0.04 attest substance concentrations of 0.04,
0.2, and 1.0 mg/L, respectively in the muscle of
juvenile Atlantic salmon (Salmo salar)

(Arokwe et al.,
2018)

High

BCF = 1.0 ± 0.1 (muscle), 4.3 ± 0.2 (liver), 2.6 ±
0.2 (brain), 1.6 ± 0.1 (gill), and 1.6 ± 0.1 (kidney) at
test substance concentration of 9.1 (ig/L for juvenile
common carp (Cyprinus carpi0) (OECD 305)

(Tans et al.. 2019)

BCF = 0.8 ± 0.1 (muscle), 2.4 ± 0.1 (liver), 2.2 ± 0.1
(brain), 1.9 ± 0.2 (gill) attest substance

(Wans et al..
2017a)

3 See EPI (Estimation Programs Interface) Suite™ for addit io mat i nfo rmatio n and supporting documents about this freely
available, online suite of programs, which was reviewed by the EPA Science Advisory Board (SAB. 2007').

Page 40 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Property or
Endpoint

Value"

Reference(s)

Overall Quality
Determination



concentration of 893 (ig/L, respectively for zebrafish
(Danio rerio) (OECD 305)





Bioaccumulation
factor (BAF)
(L/kg, unless
noted)

Mean BAF = 794 (muscle), 1,995 (liver), 1,995
(kidney), and 1,995 (gill)

(Bekele et aL

2021)

High

Mean BAF = 30.7 (muscle) and 70.7 (liver) for
crucian carp (Carassius auratus)

(Choo et aL, 2018)

Mean BAF = 2,198 at test substance concentration
of 0.464 ng/L for walleye (Sander vitreus)

(Guo et aL. 2017b)

Mean BAF = 1,248 for snakehead (Ophiocephalus
argus), 191 for catfish (Clarias batrachus), 109-202
for mud carp (Cirrhinus molitorella), 207 for
crucian carp (Carassius auratus), and 463 for
Oriental River prawn (Macrobrachium nipponense)

(Liu et aL. 2019a)

Mean BAF = 6,310 for benthic invertebrates (soft
tissue); 2,690 for pelagic fish (organ); 4,270 for
benthic fish (organ and whole body)

("Wans et aL.
2019b)

Organic
carbon: water
partition
coefficient
(log Koc)

2.08-2.52

(Cristate et aL.
2017)

High

3.23 ±0.23

(Wans et aL,
2018a)

3.32 (mean; range 2.5-4.06)

(Zhang et aL,

2018b)

3.46 ±0.48

(Zhang et aL,

2018b)

Removal in
wastewater
treatment

Approximately -5% removal after primary
treatment; -19.1% overall removal

(Kim et aL, 2017)

High

Trophic
magnification
factor (TMF)

Benthic food web: 2.6 (tentative due to small sample
size, n = 15)

(Brandsma et aL.

High

No significant relationship with pelagic food web
and total food web

2015)

Antarctic food chain: 5.2

(Fu et aL. 2020)

No significant relationship with trophic level

(Zhao et aL, 2018)

Biota-sediment
accumulation
factor (BSAF)

Mean BSAF (L/kg): 1.09 (muscle) and 2.49 (liver)
for Crucian carp (Carassius auratus)

(Choo et aL. 2018)

High

Mean BSAF: 0.015-0.171

(Liu et aL. 2019a)

Mean BSAF: 2.19E-03 for benthic invertebrates
and 1.48E-03 for benthic fishes

(Wans et aL,
2019b)

a Measured unless otherwise noted

b Information estimated usine EPI Suite™(U.S. EPA, 2012c)

Page 41 of 572

-------
1274

1275

1276

1211

1278

1279

1280

1281

1282

1283

1284

1285

1286

1287

1288

1289

1290

1291

1292

1293

1294

1295

1296

1297

1298

1299

1300

1301

1302

1303

1304

1305

1306

1307

1308

1309

1310

1311

1312

1313

1314

1315

1316

1317

1318

1319

1320

1321

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

2.2.2 Summary of Fate and Transport Assessment

Numerous studies have described TCEP as a "ubiquitous" contaminant because it is commonly found in
various environmental compartments such as indoor air and dust, outdoor air, surface water, drinking
water, groundwater, soil, sediment, biota, and even precipitation all over the world (Awonaike et ai,

2021; Ma et ai, JOJ I; ^icpp et al., 2021; Choo am! iHi J020; Li et ai, 2019b; Tan et ai, 2019; Arukwe
et ai, 2018; Kim and Kantian, 2018; Cao *'t i , Uurtado -n jI JO i ; \\ et ai, JO I j; Hi adman
et ai, 2014; Padhye et ai, i, 1 uMale et al i , < Madman et ai, 2012; Regnery and Puttmann,
2010b; Benotti et ai. 2009; Fries and Puttmann. , ). This is because TCEP is primarily used as
an additive plasticizer and flame retardant. When used as an additive, TCEP is added to manufactured
materials via physical mixing rather than chemical bonding and as a result, TCEP can easily leach or
diffuse into its surrounding environment (Qi et ai, 2019; Liu et ai, 2014; Wei et ai, 2014; AT SDR,
2012; van der Veen and de Boer. 2012; EC. 200(, i v ^ _i<09; NICNAS. 20011 TCEP's physical and
chemical properties suggests that its main mode of distribution in the environment is through water and
soil, depending on where it is being released (Appendix E.2.1.2) (TERA, 2015; U.S. EPA, 2012d;
Regnerv and Puttmann. 2010b; Zhang et ai. 2009).	

Multiple studies have identified urban sources as sources of TCEP in the environment through fugitive
emissions to air (Abdollahi et ai. 1 ^ I , < ,uo et ai. 2015; Moller et ai. 2011). The exact sources of
TCEP emissions from urban environment are unknown, however they are likely the articles that were
treated with or containing TCEP (Abdollahi et ai, 2017; Luo et ai, 2015; Wei et ai, JO I I; Moller et ai,
2011; Aston et a 5). Compared to outdoor air, TCEP concentrations are significantly higher in
indoor air, because TCEP has the potential to volatilize from treated products and diffuse into air, as
well as partition onto dust, due to its use as an additive (Qi et ai, 2*- m , h ^ \ , Liu et ai, 2014;
ATSDR. 2012; EC. 2009; NICNAS. 2001). Atmospheric deposition has been identified as an important
source of TCEP to surface water, especially in urban areas. Several studies showed that higher TCEP
concentrations in precipitation were generally seen in densely populated areas with high traffic volume
(Kim and Kannan, 2018; Regnerv and Puttmann, 2010b; Regnery and Puettmann, 2009; Marklund et ai,
2005b). In addition, storm water and urban runoff can contribute to additional emissions to surface
water.

TCEP can be transported to sediment from overlying surface water by advection and dispersion of
dissolved TCEP and by deposition of suspended solids containing TCEP. However, TCEP may partition
between surface water and sediments to varying degrees because of its wide range of log Koc values
(2.08 to 3.46) (Wang et ai, 2018a; Zhang et ai, 2018b; Cristale et ai, ) and high water solubility
(7,820 mg/L) (Lee et ai. 2018; Ma et ai. JO I ; i'undsma ei a I JO I ¦*; i ao et ai. 2012). which could
contribute to its mobility in the environment. Higher concentrations of TCEP in sediment are expected
to be found at potential source locations (e.g., near urban and industrialized areas) (Chokwe and
Okonkwo, 2019; Tan et ai, 2019; Lee et ai, 2018; Wang et ai, 2018a; Cao et ai, 2017; Maruya et ai,
2016; Cristale et ai. 2013). TCEP accumulation in soil is expected to be unlikely. Due to its high water
solubility (7,820 mg/L), dissolved TCEP was observed to be mobile and migrated to groundwater by
common soil transport processes such as advection and diffusion ( op et ai, 2021; Buszka et ai,
2009; Barnes et ai. 2004). TCEP in the soil was seen to be vertically transported to deeper soil horizons,
causing TCEP concentration in the surface soil to be lower (He et ai. 101 ; Hacaloni et ai. 2008).

Most flame retardants that have "High" or "Very High" persistence designations, such as TCEP, are
persistent because they are expected to be stable by design to maintain their flame-retardant properties
throughout its lifetime in products (	i). Based on multiple monitoring studies, TCEP

appears to be a persistent mobile organic compound (PMOC). PMOCs can dissolve in water or bind to
particles, resulting in longer environmental half-lives and greater potential for long-range transport

Page 42 of 572

-------
1322

1323

1324

1325

1326

1327

1328

1329

1330

1331

1332

1333

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

(Blum et at.. 2019; Rodeers et al.. 2018; Reemtsma et al. 2016). TCEP was detected in both lake and
marine waters of the Arctic, where TCEP was quantified in water and air far from human settlements
(>500 km). Atmospheric deposition and watershed runoff may be the primary sources of TCEP in these
remote waters where TCEP is unlikely to be rapidly transformed by hydrolysis, photolysis, or
biodegradation (Na et al.. 2020; McDonough et al.. 2018; Li et al.. 2017b). These findings indicate that
TCEP has the potential to undergo long-range transport in air and water. TCEP's long-range transport
potential (LRTP) was seen to be significantly underestimated when using its physical and chemical
properties in quantitative structure-activity relationship (QSAR) models because the behavior of TCEP
in the environment often does not align with its physical and chemical properties. A detailed summary
of physical and chemical properties and a fate and transport assessment of TCEP is available in
Appendix E.

Page 43 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE

December 2023

Emissions
from
Source

Dry Deposition

Dispersal

Wet Deposition

Vapor pressure	Indirect Atmosphenc

0 0613 mmHg	Photolysis

t,Q= 5.8 h

Wastewater facility / W"Constant
Indirect/Direct discharge / 2 945E 06 a'mm3/™'

Aerobic Biodegradation
Rate = moderate

Landfills

Surface Water

Aqueous	—-—sr

Photolysis WS = 7,820 mg/L

No degradation	#\

Runoff

Land applied biosolids

Aerobic
Biodegradation
Rate = low

Anaerobic
1	Biodegradation

Log Kqc = 2.08 - 3.46 Rate = Negl'9lble

Log Koc =
2 08-346

Bioconcentration
BCF = 0.16-4.3

Hydrolysis =
Negligible

Sediment/
Pore Water

Groundwater

Anaerobic
Biodegradation
Rate = Negligible

1334

1335

1336

1337

1338

1339

1340

Figure Legend

Negligible

» Low/Slow

Moderate

High/Fast/Strong
^Very High/Rapid/Strong

Partitioning/T ransportation

~ T ransformation/Degradation

Wastewater Facility

Figure 2-1. Transport, Partitioning, and Degradation of TCEP in the Environment"

a The diagram depicts the distribution (grey arrows), transport and partitioning (black arrows), and the
transformation and degradation (white arrows) of TCEP in the environment. The width of the arrow is a
qualitative indication of the likelihood that the indicated partitioning will occur or the rate at which the indicated
degradation will occur (i.e., wider arrows indicate more likely partitioning or more rapid degradation).

Page 44 of 572

-------
1341

1342

1343

1344

1345

1346

1347

1348

1349

1350

1351

1352

1353

1354

1355

1356

1357

1358

1359

1360

1361

1362

1363

1364

1365

1366

1367

1368

1369

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

2,2,3 Weight of the Scientific Evidence Conclusions for Fate and Transport

2.2.3.1 Strengths, Limitations, Assumptions, and Key Sources of Uncertainty for the
Fate and Transport Assessment

Given the consistent results from numerous high-quality studies, there is a robust confidence that TCEP

•	is not expected to undergo significant direct photolysis (Appendix E.2.2);

•	will partition to organic carbon in the air (Appendix E.2.2);

•	will exist in both the gas and particle phases (Appendix E.2.2);

•	showed no significant degradation after undergoing hydrolysis but hydrolysis rate was seen to
increase with increasing pH (Appendix E.2.3.1);

•	does not undergo biodegradation in water under aerobic conditions (Appendix E.2.3.1);

•	will volatilize from surface water and moist soil (Appendixes E.2.3.1 and E.2.4.1);

•	produces hazardous byproducts when undergoing thermal degradation (Appendix E.2.5.1);

•	will not be removed after undergoing wastewater treatment and will be retained in effluents with
low fraction being adsorbed onto sludge (Appendix E.2.5.2);

•	is minimally removed after undergoing conventional drinking water treatment (Appendix
E.2.5.3); and

•	has the ability to undergo long-range transport (Appendixes E.2.2 and E.2.3.1).

As a result of limited studies identified, there is a moderate confidence that TCEP

•	will partition to organic carbon in sediment and soil (Appendixes E.2.3.2 and E.2.4.1);

•	will enter surface water and groundwater from landfills (Appendix E.2.4.3);

•	will not bioaccumulate in fish residing in the water column (Appendix E.2.6);

•	may bioaccumulate in benthic fish (Appendix E.2.6); and

•	does not bioaccumulate when TCEP concentrations are transferred to higher trophic levels in the
food web (Appendix E.2.6).

Very limited evidence on anaerobic biodegradation of TCEP exists because only one medium-quality
study on anaerobic biodegradation in water was identified and no degradation was observed (Appendix
E.2.3.2). Additionally, no anaerobic biodegradation in sediment study was identified. A detailed
discussion of strengths, limitations, assumptions, and key sources of uncertainty for the fate and
transport assessment of TCEP is available in Appendix E.

Page 45 of 572

-------
1370

1371

1372

1373

1374

1375

1376

1377

1378

1379

1380

1381

1382

1383

1384

1385

1386

1387

1388

1389

1390

1391

1392

1393

1394

1395

1396

1397

1398

1399

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

3 RELEASES AND CONCENTRATIONS OF TCEP IN THE
ENVIRONMENT

EPA estimated environmental releases of TCEP. Section 3.1 describes the approach and methodology
for estimating releases. Estimates of environmental releases are presented in Section 3.2. Section 3.3
presents the approach, methodology, and estimates of environmental concentrations that result from
environmental releases of TCEP.

3.1 Approach and Methodology

3.1.1 Industrial and Commercial	

EPA categorized the COUs listed in Table 1-1 into occupational exposure scenarios (OESs) (see Table
3-1). EPA developed the OESs to group processes or applications with similar sources of release and
occupational exposures that occur at industrial and commercial workplaces within the scope of the risk
evaluation. For each OES, occupational exposure and environmental release results are provided and
expected to be representative of the entire population of workers and sites involved for the given OES in
the United States. Note that EPA may define only a single OES for multiple COUs, while in other cases
multiple OESs may be developed for a single COU. For example, the paint and coating manufacturing
COU has two associated OESs—a 1-part coatings scenario and a 2-part reactive coatings scenario. EPA
makes this determination by considering variability in release and use conditions and whether the
variability can be captured as a distribution of exposure or instead requires discrete scenarios.
Specifically, the 1-part coatings tend to be water-based formulations and could potentially have a greater
release to water whereas the 2-part reactive coatings could have greater release to incineration or
landfill. Further information on specific OESs is provided in Supplemental Information on
Environmental Release and Occupational Exposure Assessment (	10231).

All COUs assessed in this Risk Evaluation are considered on-going uses. However, there are several
COUs for which part of the life cycle has ceased, such as manufacturing (including import) and
processing. However, other parts of the lifecycle including recycling, commercial or consumer use, and
disposal are on-going. These COUs are identified in Table 3-1 and include four COUs for commercial
use and five COUs for consumer use.

Table 3-1. Crosswalk of Conditions of Use (COUs) to Occupational Exposure Scenarios Assessed

COU

OES

Life Cycle Stage"

Category''

Subcategory'

Manufacture

Import

Import

Repackaging

Processing

Incorporated into
formulation,
mixture, or
reaction product

Paint and coating
manufacturing

Incorporation into paints and
coatings - 1-part coatings

Incorporation into paints and
coatings - 2-part reactive coatings

Incorporated into
formulation,
mixture, or
reaction product

Polymers used in aerospace
equipment and products

Formulation of TCEP into 2-part
reactive resins

Incorporated into
article

Aerospace equipment and
products

Processing into 2-part resin article

Recycling

Recycling

Recycling e-waste

Page 46 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

cou

OES

Life Cycle Stage"

Category''

Subcategory'

Distribution

Distribution

Distribution in commerce

Distribution activities (e.g.,
loading) considered throughout
life cycle, rather than using a
single distribution scenario

Industrial Use

Other use

Aerospace equipment and
products

Installation of article

Commercial Use

Other use

Aerospace equipment and
products

Use and/or maintenance of
aerospace equipment and products

Paints and coatings

Paints and coatings

Use of paints and coatings - spray
application OES

Other use

Laboratory chemicals

Lab chemical - use of laboratory
chemicals

Furnishing,
cleaning,
treatment/care
products

Fabric and textile products^

End of service life disposal^
(releases and exposures not
quantified)

Foam Seating and Bedding
Products^

End of service life disposal^
(releases and exposures not
quantified)

Construction,
paint, electrical,
and metal products

Building/construction
materials - insulation^

End of service life disposal^
(releases and exposures not
quantified)

Building/construction
materials - wood and
engineered wood products -
wood resin composites^

End of service life disposal^
(releases and exposures not
quantified)

Disposal

Disposal

Disposal®

Waste disposal (landfill or
incineration, covered in each
COU/OES as opposed to a
separate COU)

a Life Cycle Stage Use Definitions (40 CFR 711.3)

-	"Industrial Use" means use at a site at which one or more chemicals or mixtures are manufactured (including
imported) or processed.

-	"Commercial Use" means the use of a chemical or a mixture containing a chemical (including as part of an
article) in a commercial enterprise providing saleable goods or services.

-	"Consumer Use" means the use of a chemical or a mixture containing a chemical (including as part of an article,
such as furniture or clothing) when sold to or made available to consumers for their use.

-	Although EPA has identified both industrial and commercial uses here for purposes of distinguishing scenarios in
this document, the Agency interprets the authority over "any manner or method of commercial use" under TSCA
section 6(a)(5) to reach both.

h These categories of COUs appear in the LCD, reflect CDR codes, and broadly represent COUs of TCEP in industrial
and/or commercial settings and for consumer uses.
c These subcategories reflect more specific COUs of TCEP.

d This COU includes associated disposal of those COUs for which manufacturing (including import) and processing
have ceased.

'' Section 3.2 provide details on these OESs.

Page 47 of 572

-------
1400

1401

1402

1403

1404

1405

1406

1407

1408

1409

1410

1411

1412

1413

1414

1415

1416

1417

1418

1419

1420

1421

1422

1423

1424

1425

1426

1427

1428

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

The 2016 CDR data (U.S. EPA. 2019a) included a single reporting site, Aceto Corporation in Port
Washington, New York, importing TCEP, with no downstream industry sectors identified. TCEP was
not reported in the 2020 CDR (U.S. EPA. 2020a). EPA did identify other data on current import
volumes and possible import sites from Datamyne, as presented in Figure 1-3, which showed some
TCEP imports below the CDR threshold of 25,000 lb/site-yr. Nevertheless, processors of TCEP may be
purchasing the chemical from importers (see Supplemental Information on Environmental Release and
Occupational Exposure Assessment (U.S. EPA. 20231) for details). Therefore, EPA assumed TCEP may
still be imported at volumes below the CDR reporting threshold and EPA assessed the following two
potential scenarios: (1) one site importing 25,000 lb, and (2) one site importing 2,500 lb. EPA modeled
environmental releases and occupational exposures for these hypothetical scenarios. For each OES,
where monitoring data were not available, daily releases were estimated per media of release based on
EPA Standard Models, Generic Scenarios (GSs), and/or Emission Scenario Documents (ESDs) to
generate annual releases and for the estimation of associated release days. TCEP is not listed on the
National Emissions Inventory (NEI) and was only recently added to TRI, with the first year of reporting
from facilities due July 1, 2024. EPA describes its approach and methodology for estimating daily
releases and for detailed facility level results in Supplemental Information on Environmental Release
and Occupational Exposure Assessment (U.S. EPA. 20231).

OES
Occupational
Assessment

Inhalation
Exposure

Dermal
Exposure

# of Workers,
ONUs Exposed

Monitoring
Data



Modeling



Modeling



# Workers or
ONUs per site



Number of
facilities





















HSIA, Reports,
NIOSH, OSHA



NF/FF, ESD



DEVL model



BLS, Census,
ESD



Census, NEI,
TRI, DMR, CDR

Figure 3-1. An Overview of How EPA Estimated Daily Releases for Each OES

BLS = Bureau of Labor Statistics; DEVL = Dermal Exposure to Volatile Liquids model; DMR = Discharge
Monitoring Report; ELG = Effluent Limitation Guidelines; HSIA = Halogenated Solvents Industry Alliance;
NF/FF = Near-Field/Far Field; NIOSH = National Institute of Occupational Safety and Health; OSHA =
Occupational Safety and Health Administration

The releases of TCEP were estimated for each media applicable to the OES. For TCEP, releases could
occur to water, air, or disposal to land. TCEP released could be in the form of liquid (neat or in
formulation), vapor, and/or solid waste.

Page 48 of 572

-------
1429

1430

1431

1432

1433

1434

1435

1436

1437

1438

1439

1440

1441

1442

1443

1444

1445

1446

1447

1448

1449

1450

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

3.2 Environmental Releases

TCEP - Environmental Releases (Section 3.2):

Key Points

EPA evaluated the reasonably available information for releases of TCEP to the environment. The
key points of the environmental releases are summarized below:

• EPA assessed environmental releases of TCEP from industrial and commercial sources as well
as consumer products.

o For industrial and commercial sources, EPA used data from literature, relevant ESDs, or
GSs to estimate environmental releases to air, surface water, and waste disposal from a
generic facility for each OES. Some OESs could not be quantified due to insufficient
data. Of the OESs that could be quantified, the highest release estimates were from

¦	Incorporation into paints and coatings - 1-part coatings

¦	Incorporation into paints and coatings - 2-part reactive coatings

¦	Formulation of TCEP-containing reactive resins (for use in 2-part systems)

¦	Use of paints and coatings - spray application OES.

o For consumer products, EPA did not have enough information to assess environmental
releases quantitatively. However, the Agency acknowledges that there may be TCEP
releases to the environment via the demolition and disposal of consumer articles, as well
as to wastewater via domestic laundry. These releases were assessed qualitatively. EPA
included anecdotal information from peer-reviewed literature on releases from consumer
articles in Section 5.1.2.2.5.

3.2.1 Industrial and Commercial	

EPA combined its estimates for each activity that is reasonably expected to occur during each OES.
These activities were based on using data from literature, relevant ESDs or GSs. Once these activities
were identified, existing EPA models and parameters (e.g., the EPA/OPPT Mass Transfer Coefficient
model, EPA/OPPT Penetration model, ChemSTEER User Guide, etc.) were used in a Monte Carlo
simulation to create a distribution of releases. From this distribution EPA provides a high-end (95th
percentile) and central tendency (50th percentile) release values as well as a range of potential release
days. The releases presented are assumed to be representative of what would be reasonably expected to
occur at an individual generic site. In some cases, where it was not reasonable to assume a single generic
site due to throughput constrictions presented in the relevant source (e.g., it is not reasonable to assume
that a single paint application site or laboratory would use the entire PV of 25,000 lb), a range of
potential number of sites is presented in Table 5-2. A summary of these ranges of releases across OESs
is presented in Table 3-2. See Supplemental Information on Environmental Release and Occupational
Exposure Assessment (U.S. EPA. 20231) for more details on deriving the overall confidence score for
each OES. For some OESs, EPA was not able to estimate or did not anticipate there to be releases; for
example:

•	EPA was not able to quantify disposal of articles that historically contained TCEP with
reasonably available information. This was assessed qualitatively.

•	Installation of articles are not expected to lead to significant releases because the articles are
expected to already be in final form (e.g., electronic potting) and not expected to undergo further
processing (i.e., shaping, sanding cutting, etc.).

Page 49 of 572

-------
1451

1452

1453

1454

1455

1456

1457

1458

1459

1460

1461

1462

1463

1464

1465

1466

1467

1468

1469

1470

1471

1472

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

•	EPA was not able to quantify releases of TCEP that could occur during the recycling of e-waste.
Sources used for this provided monitoring data from breathing zone measurements from various
locations within a facility that recycles e-waste that contained very small amounts of TCEP dust.
The source of TCEP was not identified and the source further stated that TCEP is rarely used in
electronics. EPA expects releases that could occur during this activity to be minimal and only
potentially occur at a small subset of facilities.

•	EPA lacks production volume data to assess TCEP exposure from distribution into commerce
due to the declining production and manufacturing in recent years. Although manufacturing,
processing, and distribution into commerce of TCEP is declining (see Section 1.1.1, Table 3-1);
distribution into commerce that has occurred, is ongoing, or is likely to occur during a COU
subject to evaluation; and exposure to human or ecological populations has occurred or is likely
to occur; will be included in the risk evaluation as an exposure associated with a COU.

3.2.1.1 Summary of Daily Environmental Release Estimates

Table 3-2 and Table 3-3 provide estimated releases that could occur during each OES, the expected
media of release if releases are expected to occur during that OES, and possible number of sites where
releases could occur. The estimated daily releases are based on a 2,500 lb production volume. For most
cases, the number of sites is based on a single generic site; however, in some cases, such as use of paints
and coatings and laboratory chemicals, a distribution of the number of sites was created. The
distributions for number of sites were created for these OESs to provide variability in the potential
number of sites and is further explained in the Supplemental Information on Environmental Release and
Occupational Exposure Assessment (	20231).

Page 50 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

1473	Table 3-2. Summary of EPA's Daily Release Estimates for Each OES and EPA's Overall Confidence in these Estimates for 2,500 lb

1474	Production Volume





Estimated Dailv



Estimated Release







cou

OES

Release Range across
Sites (kg/sitc-dav)

Type of Discharge,"
Air Emission,6 or

Frequency Range
across Sites (days)''

Number of
Facilities'

Overall
Confidence

Sou rccs





Central

High-End

Transfer for Disposal'

Central

High-End







Tendency



Tendency











6.35E00

9.89E00

Surface water

4

4





Peer-

Manufacture

Repackaging

3.18E-04

6.03E-04

Fugitive or stack air

4

4

1 generic site

Medium

reviewed

(Import)

N/A

N/A

Waste disposal (landfill
or incineration)

N/A

N/A

literature®
(GS/ESD)





1.02E01

3.52E01

Surface water

6

2





Peer-

Processing

Incorporation into paints and

1.56E-03

9.60E-03

Fugitive or stack air

6

4

1 generic site

High

reviewed

coatings - 1-part coatings

1.53E00

9.27E00

Waste disposal (landfill
or incineration)

7

2

literature®
(GS/ESD)





2.71E01

3.19E01

Surface water

1

1









Incorporation into paints and

3.65E-03

7.90E-03

Fugitive air

1

1





Peer-

reviewed

literature®

Processing

coatings - 2-part reactive

3.75E-03

1.99E-02

Stack air

1

1

1 generic site

High



coatings

3.40E01

340E01

Waste disposal (landfill
or incineration)

1

1





(GS/ESD)





2.52E01

3.15E01

Surface water

1

1









Formulation of TCEP-

3.25E-03

8.83E-03

Fugitive air

1

1





Peer-

reviewed

literature^

Processing

containing reactive resins (for

2.73E-03

2.07E-02

Stack air

1

1

1 generic site

High



use in 2-part systems)

340E01

340E01

Waste disposal (landfill
or incineration)

1

1





(GS/ESD)





N/A

N/A

Surface water

N/A

N/A





Peer-

Processing

Processing into 2-part resin

3.30E-04

9.90E-04

Fugitive or stack air

55

113

1 generic site

High

reviewed

article

3.98E-01

2.50E00

Waste disposal (landfill
or incineration)

92

17

literature®
(GS/ESD)

Processing

Recycling e-waste

EPA did not have sufficient data to estimate these releases

Distribution

Distribution in commerce

Distribution activities (e.c

g., loading) considered throughout life cycle, rather than using a single distribution scenario.

Industrial

Use

Installation of articles

Releases expected to be negligible

Page 51 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023





Estimated Dailv



Estimated Release







cou

OES

Release Range aeross
Sites (kg/sitc-dav)

Type of Discharge,"
Air Emission,6 or

Frequency Range
across Sites (days)''

Number of
Facilities'

Overall
Confidence

Sou rccs





Central

High-End

Transfer for Disposal'

Central

High-End







Tendenev



Tendency









Use and/or maintenance of

Releases expected to be negligible













aerospace equipment and
products





















2.37E00

2.32E01

Surface water

1

2

95th







Use of paints and coatings -
spray application8

1.25E01

1.14E02

Fugitive air

1

2

Percentile:

2,031

50th

Percentile:
281



Peer-
reviewed
literature®
(GS/ESD)



N/A

N/A

Waste disposal (landfill
or incineration)

N/A

N/A

Medium





3.96E-01^

8.83E-01^

Surface water

220

214

13 (1st
percentile) -



Peer-



Lab chemical - use of

6.47E-05^

7.99E-05^

Fugitive or stack air

220

214

High

reviewed

Commercial
Use

laboratory chemicals

N/A

N/A

Waste disposal (landfill
or incineration)

N/A

N/A

6 (5th
percentile)

literature®
(GS/ESD)

Furnishing, cleaning,
treatment/care products

•	Fabric and textile
products

•	Foam seating and
bedding products



















Construction, paint,
electrical, and metal products
• Building/construction
materials - insulation

Manufacturing and Processing of these COU's has ceased, EPA does not have sufficient data to estimate the releases that
may occur during disposal of already existing products



• Building/construction
materials - wood and



















engineered wood
products - wood resin



















composites

















1 )lspnsal

1 )ispiisal

Wasic 1 )is|x>

sal (l.andlill

or Incineration. ancrcd i

ii each ('()l

()l :s as opposed lo a separate (()l )



Dnccl disclia

gc lo surface ualcr. induce! discharge lo noii-PUTW, indirect discharge to POTW











Emissions via fugitive air; stack air; or treatment via incineration













c Transfer to surface impoundment, land application, or landfills

d Where available, EPA used peer reviewed literature (e.g., generic scenarios or emission scenario documents) to provide a basis to estimate the number of release days
of TCEP within a COU.

Page 52 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

cou

OES

Estimated Daily
Release Range across
Sites (kg/sitc-dav)

Type of Discharge,"
Air Emission,6 or
Transfer for Disposal'

Estimated Release
Frequency Range
across Sites (days)''

Number of
Facilities'

Overall
Confidence

Sou rccs

Central „ .
„ . High-End
Tendency

Central . r .
„ . High-End
Tendency

e Where available, EPA used peer reviewed literature (e.g., generic scenarios or emission scenario documents) data to provide a basis to estimate the number of sites
using TCEP within a condition of use.

^ "High-end" is the 5th percentile and "Central Tendency" is the 1st percentile. See Section 3.10 of Engineering Supplemental file for rationale of using the 1st and 5th
percentiles.

g Multiple throughput and site scenarios are presented in Table 5-1 of the Engineering Supplemental file.

1475

1476

Page 53 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

1477 Table 3-3. Summary of EPA's Release Estimates for Each COU/OES and EPA's Overall Confidence in these Estimates

Life Cycle
Stage







Surface
Water

Air

Waste Disposal

Overall
Confidence



Category

Subcategory

OES

Fugitive
Air

Stack Air

Landfill

Incineration

Sources

Manufacture
(Import)

Import

Import

Repackaging

0

0

0

m

m

Medium

Peer-reviewed

literature®

(GS/ESD)



Incorporated
into

formulation,

Paint and

coating

manufacturing

Incorporation
into paints and
coatings - 1-
part coatings









0

High

Peer-reviewed

literature®

(GS/ESD)



mixture, or

reaction

product

Incorporation
into paints and
coatings - 2-
part coatings







0



High

Peer-reviewed

literature®

(GS/ESD)

Processing

Incorporated
into

formulation,
mixture, or
reaction
product

Polymers used
in aerospace
equipment and
products

Formulation of
TCEP-
containing
reactive resins
(for use in 2-
part systems)







0



High

Peer-reviewed

literature^

(GS/ESD)



Incorporated
into article

Aerospace
equipment and
products

Processing into
2-part resin
article

0

0

0

0

0

High

Peer-reviewed

literature®

(GS/ESD)



Recycling

Recycling

Recycling e-
waste

~

~

~

~

~

Medium

NIOSH
HHE's used
for exposure
estimates;
insufficient
data to
estimate
releases

Distribution

Distribution

Distribution in
commerce

Distribution in
Commerce

Distribution activities (e.g., loading) considered throughout life cycle, rather than using a
single distribution scenario.

Industrial

Use

Other use

Aerospace
equipment and
products

Installation of
article



0

0

0

0

Medium

Releases not
expected to
occur during

Page 54 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Life Cycle
Stage

Category

Subcategory

OES

Surface
Water

Air

Waste Disposal

Overall
Confidence

Sources

Fugitive
Air

Stack Air

Landfill

Incineration





















handling of

aerospace

articles

Commercial
Use

Other use

Aerospace
equipment and
products

Use and/or
maintenance of
aerospace
equipment and
products

m

m

m

m

m

Medium

Releases not
expected to
occur during
handling of
aerospace
articles

Paints and
coatings

Paints and
coatings

Use of paints
and coatings -
spray

application oes
1,000 kg daily
throughput

a

0

0

0

0

Medium

Peer-reviewed

literature®

(GS/ESD)

Other use

Laboratory
chemicals

Lab chemical -
use of
laboratory
chemicals

a

0

0

0

0



Peer-reviewed

literature®

(GS/ESD)

Furnishing,
cleaning,
treatment/care
products

Fabric and
textile products



~

~

~

~

~

Medium

Peer-reviewed
literature®

Foam seating
and bedding
products



~

~

~

~

~

Medium

Peer-reviewed
literature®

Construction,
paint,

electrical, and
metal products

Building/
construction
materials -
insulation



~

~

~

~

~

Medium

Peer-reviewed
literature®

Page 55 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Life Cycle
Stage

Category

Subcategory

OES

Surface
Water

Air

Waste Disposal

Overall
Confidence

Sources

Fugitive
Air

Stack Air

Landfill

Incineration

Building/
construction
materials -
wood and
engineered
wood products -
wood resin
composites



III

III

III

III

III

Medium

Peer-reviewed
literature®

Disposal





Disposal

Evaluated as part of each OES as opposed to a standalone OES

0 Estimated releases HI- Estimated releases but not anticipated LJ- Releases not quantified, assessed qualitatively

1478

Page 56 of 572

-------
1479

1480

1481

1482

1483

1484

1485

1486

1487

1488

1489

1490

1491

1492

1493

1494

1495

1496

1497

1498

1499

1500

1501

1502

1503

1504

1505

1506

1507

1508

1509

1510

1511

1512

1513

1514

1515

1516

1517

1518

1519

1520

1521

1522

1523

1524

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

3.2.2	Consumer Releases	

Environmental releases to the environment may occur from consumer articles containing TCEP via the
end-of-life disposal and demolition of consumer articles in the built environment, as well as from the
associated down-the-drain release of TCEP from domestic laundry that removes TCEP containing dust
from clothing to wastewater. It is difficult for EPA to quantify these ends-of-life and down-the-drain
laundry exposures due to limited information on source attribution of the consumer COUs. In previous
assessments, EPA has considered down-the-drain analysis for consumer products scenarios where there
is reasonably foreseen exposure scenario where it can be assumed the consumer product (e.g., drain
cleaner, lubricant, oils) will be discarded directly down-the-drain. Although EPA acknowledges that
there may be TCEP releases to the environment via the demolition and disposal of consumer articles and
the release of TCEP to wastewater via domestic laundry, the Agency did not quantitatively assess these
scenarios due to lack of reasonably available information. EPA instead assessed them qualitatively.
Anecdotal information in the peer-reviewed literature helps qualitatively describe how TCEP may be
potentially released to the environment from consumer articles (Section 5.1.2.2.5).

3.2.3	Weight of the Scientific Evidence Conclusions for Environmental Releases from
Industrial, Commercial, and Consumer Sources

For each OES, EPA considered the assessment approach, the quality of the data and models, and
uncertainties in assessment results to determine a level of confidence as presented in Supplemental
Information on Environmental Release and Occupational Exposure Assessment (	20231). EPA

determined that the various GSs and ESDs had overall quality determinations of high or medium,
depending on the GS/ESD. The GSs and ESDs are documents developed by EPA or OECD that are
intended to provide an overview of an industry and identify potential release and exposure points for that
industry; they cover processes and are not specific to any chemical. This lack of chemical specificity
creates an uncertainty in the overall release estimate—the assessed parameter values may not always be
representative of applications specific to TCEP use in each OES. Another uncertainty is lack of
consideration for release controls. The GS/ESDs assume that all activities occur without any release
controls and in an open-system environment where vapor and particulates freely escape. Actual releases
may be less than estimated if facilities utilize pollution control methods. Although TCEP monitoring
data would be preferred to modeled estimates from generic scenarios, monitoring data were not
available for almost all the OESs included in the draft risk evaluation. EPA strengthened modeled
estimates by using Monte Carlo modeling to allow for variation in environmental release calculation
input parameters according to the GS/ESD and other literature sources. The Agency was unable to
quantitatively assess releases to the environment from consumer products containing TCEP.

3.2.3.1 Strengths, Limitations, Assumptions, and Key Sources of Uncertainty for the
Environmental Release Assessment
Use of Reporting Year-Release Trends Analysis

The 2016 CDR only had one reporter of TCEP while the 2020 CDR had no reporters; it is assumed that
TCEP has been largely phased out of products it was historically used in such as flexible and rigid foam
products. EPA expects that current users of TCEP do not surpass the CDR reporting threshold of 25,000
lb per site-year (i.e., less than 25,000 lb/year is used at any given site).

EPA searched the DMR database for TCEP monitoring data from 2010 to 2021. Monitoring data were
available for locations in California; however, TCEP was not detected in any of the effluents of the
POTWs that were monitored (	2022b). DMR data are submitted by NPDES permit holders to

states or directly to the EPA according to the monitoring requirements of the facility's permit. States are
required to load only major discharger data into DMR and may or may not load minor discharger data.

Page 57 of 572

-------
1525

1526

1527

1528

1529

1530

1531

1532

1533

1534

1535

1536

1537

1538

1539

1540

1541

1542

1543

1544

1545

1546

1547

1548

1549

1550

1551

1552

1553

1554

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

The definition of major vs. minor discharger is set by each state and could be based on discharge volume
or facility size. Due to these limitations, some sites that discharge may not be included in the DMR
dataset. It is uncertain the extent to which sites not captured in these databases release TCEP into the
environment or whether the releases are to water, air, or landfill. TCEP was officially added to TRI at
the end of 2022. However, companies will not have to report on their possible management and/or use
of TCEP until July 2024.

EPA also searched other databases including the Water Quality Portal (WQP), where monitoring trends
indicate a downward trend of TCEP concentrations in surface water (see Figure 3-9).

Use of Generic Scenario and Emission Scenario Documents for Number of Facilities

In some cases, the number of facilities for a given OES was estimated using GSs and ESDs, which are
peer-reviewed. These documents typically attempt to find and map applicable North American Industry
Classification System (NAICS) codes to an OES. This is done by identifying keywords relevant to that
OES and entering them into the search tool on the U.S. Census Bureau's website. The results are
reviewed for relevancy and the most applicable NAICS codes are selected. It is possible that the NAICS
codes selected may not fully represent all potential types of sites for a given OES.

Uncertainties Associated with Number of Release Days Estimate

EPA did not have site specific data for the number of release days for most OESs. Typically, in these
cases, the Agency assumed that an activity occurs once per day (e.g., a facility may process a single
batch per day). In the event that this assumption leads to a number of operating days that exceeds 365
days, it may be assumed that a site will be processing more than one batch per day. Given the relatively
small production volume of TCEP being assessed this situation was not encountered. However, it is
possible that this could lead to either an under or over estimation of the number of release days. In
certain circumstances, EPA chose 250 days a year as the upper bound of possible number of operating
days because that is considered the maximum number of days a worker would be exposed, for most
OESs the number of release days was well under this value.

Page 58 of 572

-------
1555

1556

1557

1558

1559

1560

1561

1562

1563

1564

1565

1566

1567

1568

1569

1570

1571

1572

1573

1574

1575

1576

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

3.3 Concentrations of TCEP in the Environment

TCEP - Concentrations in the Environment (Section 3.3):

Key Points

EPA evaluated the reasonably available information on concentrations of TCEP in the environment.

The key points on environmental concentrations are summarized below:

• EPA assessed environmental concentrations of TCEP in air, water, and land (soil, biosolids and
groundwater).

o For the air pathway, measured data from a variety of locations within and outside of the
United States provided TCEP concentrations near facilities and locations that would
represent general population exposure, as well as in remote locations. EPA also modeled
ambient air concentrations and deposition from facilities releasing TCEP to air. The
Agency expects dry and wet air deposition of TCEP from air to land and surface waters
may be an important source of TCEP to the ambient environment,
o For the water pathway, EPA found measured data on TCEP in surface water,

precipitation, groundwater, wastewater, and the sediment compartment. The Agency also
modeled TCEP concentrations in surface water and sediment, including air deposition
contributions to each, near facilities releasing TCEP. EPA expects surface water and
sediment to be the main environmental exposure pathways for aquatic organisms,
o For the land pathway, EPA found measured concentrations of TCEP in soil, biosolids, and
groundwater. The Agency modeled soil concentrations from air deposition and biosolids
as well as groundwater concentrations from landfill leachate. EPA does not expect TCEP
concentrations to accumulate in soil; rather, TCEP in soil is expected to migrate to
groundwater.

The environmental exposure characterization focuses on aquatic and terrestrial releases of TCEP from
hypothetical facilities that use, manufacture, or process TCEP under industrial and/or commercial COUs
subject to TSCA regulations. To characterize environmental exposure, EPA assessed point estimate
exposures derived from both measured and predicted concentrations of TCEP in ambient air, surface
water, and landfills in the United States.

A literature search was also conducted to identify peer-reviewed or gray sources of TCEP monitoring
and reported modeled data. The tornado plots in the subsequent sections are a summary of the
monitoring for the various environmental media. The plots provide the range of media concentrations in
monitoring various studies. The plots are split by U.S. and non-U.S. data, fraction (e.g., vapor, gas,
particle; see Figure 3-9) and the studies are ordered from top to bottom from newer to older data. The
plots are colored to indicate general population, remote, near facility, and unknown population
information.

For more information on TCEP monitoring data, please see the following documents:

•	Environmental Monitoring Concentrations Reported by Media Type (U.S. EPA. 2023 g).

•	Environmental Monitoring and Biomonitoring Concentrations Summary Table (U.S. EPA.
2023f).

•	Data Quality Evaluation Information for General Population, Consumer, and Environmental
Exposure. (U.S. EPA. 2023v)

Page 59 of 572

-------
1577

1578

1579

1580

1581

1582

1583

1584

1585

1586

1587

1588

1589

1590

1591

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

• Data Extraction Information for General Population, Consumer, and Environmental Exposure

(U.S. EPA. 20230")

3.3.1 Ambient Air Pathway

EPA searched peer-reviewed literature, gray literature, and databases to obtain concentrations of TCEP
in ambient air. Section 3.3.1.1 displays the aggregated results of reported monitoring concentrations for
ambient air found in the peer-reviewed and gray literature from the systematic review. Section 3.3.1.2
reports EPA modeled ambient air concentrations and deposition fluxes.

Ambient air concentrations of TCEP were measured in six studies in the United States (Figure 3-2).
Br adman et al. f: recorded a maximum concentration of 1.60 |ig/m3 at 14 early childhood education
facilities in California between May 2010 and May 201 1. Peverly et al. (2 sampled TCEP in
ambient air at 13 locations across Chicago, Illinois. They demonstrated that TCEP ambient air
concentrations (maximum of 0.335 |ig/m3) were slightly higher nearer to downtown Chicago than
suburban Chicago.

Page 60 of 572

-------
1592

1593

1594

1595

1596

1597

1598

1599

1600

1601

1602

1603

1604

1605

1606

1607

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE

December 2023

3.3.1.1 Measured Concentrations in Ambient Air

Mix Combined Vapor/Gas and Particulate

3985267 - Guo el ah, 2017 - CA.US

NonUS Combined Vapor/Gas and Particulate

6994279 - Bohlin-Ni/zcttoct al.. 2019 - NO

5386424 - Raucrt ct al.. 2018 - AR.BR.CL.MX
5386424 - Raucrt ct al., 2018 ¦ AR.BO.BR.CL,CO,CR,MX
632484 - Ohura ct al., 2006 - JP

US Particulate

NonUS Particulate

NonUS Vapor/Gas

2939998 - Peverly et al,. 2015 - US
5163441 - Salamova ct al.. 2016 - US
3864979 - Clark ct a].. 2017 - US
3027503 - Salamova ct aJ.. 2014 - US
3027503 - Salamova ct al.. 2014 - US
2539068 - Bradman et al.. 2014 - US

6816026 - Maccira ct al., 2020 - ES
5163827 - Wong ct al.. 2018 - SE
3862723 - Li ct al.. 2017 - AQ
5469544 - Suhring ct al., 2016 - CA
3466615 - AbdoIIahi et al.. 2017 - CA
5176506 - Marklunil ct al.. 2005 - Fl
1927779 - Saito ct al.. 2007 - JP

3862723 - Li et al.. 2017 - AQ
5017070 - Kun-Karakus et al.. 2018 - TR
5017070 - Kun-Karakus ct al.. 2018 - TR

IOA-5

a	General Population (Background)

¦	Remote (Not Near Source)

IBB Near Facility (Highly Exposed)

¦	Unknown/Not Specified

V	Lognormal Distribution (CT and 90th percentile)

A	Normal Distribution (CT and 90th percentile)

gj	Noil-Detect

V
lA

D v

C v

4 &

V

EJA

0.01	0.1

Concentration (ng/m3 )

Figure 3-2. Concentrations of TCEP (ng/m3) in Ambient Air from 2000 to 2019

3.3.1.2 EPA Modeled Concentrations in Ambient Air and Air Deposition
(HQAC/AERMOD)

EPA used the Integrated Indoor-Outdoor Air Calculator (IIOAC), and the American Meteorological
Society (AMS)/EPA Regulatory Model (AERMOD) to estimate ambient air concentrations and air
deposition of TCEP from facility releases. IIOAC uses pre-run results from a suite of AERMOD
dispersion scenarios at a variety of meteorological and land-use settings, as well as release emissions, to
estimate particle deposition at different distances from sources that release chemical substances to the
air. AERMOD, a higher tier model, was utilized to incorporate refined parameters for gaseous as well as
particle deposition. AERMOD is a steady-state plume model that incorporates air dispersion based on
planetary boundary layer turbulence structure and scaling concepts, including treatment of both surface
and elevated sources, and both simple and complex terrain.

Page 61 of 572

-------
1608

1609

1610

1611

1612

1613

1614

1615

1616

1617

1618

1619

1620

1621

1622

1623

1624

1625

1626

1627

1628

1629

1630

1631

1632

1633

1634

1635

1636

1637

1638

1639

1640

1641

1642

1643

1644

1645

1646

1647

1648

1649

1650

1651

1652

1653

1654

1655

1656

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Industrial and commercial release estimates are presented in Section 3.2. Table 3-3 provides the
following COUs/OESs that have ambient air releases (stack or fugitive). These facility releases were
utilized to model ambient air concentrations and deposition via AERMOD and IIOAC.

The full set of inputs and results of IIOAC and AERMOD are presented in Appendix H.3. For the initial
IIOAC runs, EPA modeled each of the fugitive air and stack air release scenarios for the seven relevant
OESs. In addition, due to initial uncertainties in the particle size, EPA ran IIOAC for both fine and
coarse particle settings for TCEP. In IIOAC, all calculated air concentrations of fine and coarse particles
are capped by an upper limit equal to the National Ambient Air Quality Standard	I) for

particulate matter (PM). These limits are 35 and 150 (.ig/m3 for fine and coarse particles (i.e., the
NAAQS for PM2.5 and PM10), respectively. These limits were met for all the OESs with stack
emissions. In addition, this limit was reached for the fine particle size, fugitive emissions run for the
commercial use of paints and coatings (Appendix H.3).

A further limitation of IIOAC is that it does not model gaseous deposition. Due to the inability to model
gaseous deposition, and due to the initial screening results meeting the NAAQS caps, EPA decided to
run a higher tier model (AERMOD) for the ambient air pathway.

AERMOD is a steady-state Gaussian plume dispersion model that incorporates air dispersion based on
planetary boundary layer turbulence structure and scaling concepts, including treatment of both surface
and elevated sources and both simple and complex terrain. AERMOD can incorporate a variety of
emission source characteristics, chemical deposition properties, complex terrain, and site-specific hourly
meteorology to estimate air concentrations and deposition amounts at user-specified population
distances and at a variety of averaging times. Readers can learn more about AERMOD, equations within
the model, detailed input and output parameters, and supporting documentation by reviewing the
AERMOD Users' Guide (	).

Additional parameters were required to run the higher tier model, AERMOD. EPA reviewed available
literature and referenced the fenceline methodology (Draft Screening Lev ~oach for Assessing
Ambient Air and Water Exposures to Fenceli nmunities Version 1.0) to select input parameters for
deposition, partitioning factors between the gaseous and particulate phases, particle sizes,
meteorological data, urban/rural designations, and physical source specifications. A full description of
the input parameters selected for AERMOD and details regarding post-processing of the results are
provided in Appendix H.3.3.

AERMOD was run under two land categories: suburban forested and bodies of water. A limited set of
AERMOD tests suggested suburban-forest was a reasonable and appropriately health-protective default
land-cover selection when land-cover analysis is not possible. Bodies of water typically led to the
highest deposition values. Ambient air concentrations for both land categories for each OES are
presented in Appendix H.3.3. Table 3-4 is an excerpt of the modeled annual air release data for the Use
of paints and coatings - spray application OES, 2,500 lb production volume, 95th percentile release
estimate, suburban forest land category scenario. The ambient air modeled concentrations and deposition
values are presented for two meteorology conditions (Sioux Falls, South Dakota, for central tendency
meteorology [MetCT]; and Lake Charles, Louisiana, for higher-end meteorology [MetHIGH]), 10
distances, and 3 percentiles (10th, 50th and 95th percentiles). These results indicate a maximum ambient
air concentration of 2.55 ng/m3 at 10 m from the facility and maximum deposition of 17.5 g/m2 at 30 m
from the facility for the Use of paints and coatings - spray application OES, 2,500 lb production
volume, 95th percentile release estimate, suburban forest land category scenario.

Page 62 of 572

-------
1657

1658

1659

1660

1661

1662

1663

1664

1665

1666

1667

1668

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table 3-4. Excerpt of Ambient Air Modeled Concentrations and Deposition for the Use of Paints
and Coatings - Spray Application OES, 2,500 lb Production Volume, 95th Percentile Release
Estimate, Suburban Forest Land Category Scenario		

Meteorology"

Distance (m)

Concentration (ng/mJ)by
Percentile

Deposition (g/m2) by Percentile

10th

50th

95th

10th

50th

95th

MetCT

10

4.98E-01

9.27E-01

1.11E00

3.29

7.00

8.14

MetCT

30

1.11E-01

2.84E-01

4.16E-01

2.80

5.90

7.67

MetCT

30-60

5.80E-02

1.34E-01

2.86E-01

1.22

2.67

5.78

MetCT

60

3.40E-02

9.42E-02

1.58E-01

8.46E-01

1.87

2.58

MetCT

100

1.15E-02

3.36E-02

6.45E-02

2.82E-01

6.68E-01

9.63E-01

MetCT

100-1,000

1.09E-04

5.21E-04

4.90E-03

2.21E-03

9.07E-03

8.13E-02

MetCT

1,000

5.92E-05

1.82E-04

7.95E-04

1.39E-03

3.43E-03

9.51E-03

MetCT

2,500

7.91E-06

2.39E-05

1.49E-04

1.86E-04

4.53E-04

1.78E-03

MetCT

5,000

2.29E-06

8.21E-06

4.83E-05

5.36E-05

1.71E-04

6.49E-04

MetCT

10,000

7.68E-07

2.56E-06

1.76E-05

1.85E-05

5.44E-05

2.68E-04

MetHIGH

10

5.90E-01

1.03E00

2.55E00

5.88

1.04

3.29

MetHIGH

30

1.12E-01

2.71E-01

7.05E-01

2.74

6.69

17.5

MetHIGH

30-60

4.87E-02

1.27E-01

4.32E-01

1.29

3.17

11

MetHIGH

60

2.88E-02

8.69E-02

2.23E-01

7.09E-01

2.06

5.33

MetHIGH

100

8.77E-03

3.08E-02

8.21E-02

2.13E-01

7.06E-01

1.93

MetHIGH

100-1,000

6.85E-05

4.23E-04

4.60E-03

1.61E-03

9.60E-03

1.06E-01

MetHIGH

1,000

3.25E-05

1.62E-04

6.08E-04

7.75E-04

3.68E-03

1.47E-02

MetHIGH

2,500

4.54E-06

2.52E-05

9.06E-05

1.06E-04

5.21E-04

2.19E-03

MetHIGH

5,000

1.30E-06

9.54E-06

2.87E-05

3.03E-05

1.97E-04

6.75E-04

MetHIGH

10,000

2.74E-07

4.19E-06

1.32E-05

7.09E-06

8.75E-05

2.99E-04

a MetCT refers to meteorological conditions from Sioux Falls, South Dakota, and MetHIGH refers to meteorological
conditions from Lake Charles, Louisiana. Since the scenarios are not at real locations, they were modeled twice
using two different meteorological stations. These central tendency and high-end estimates were determined during
the development of EPA's IIOAC.

3.3.1.2.1 TCEP Partitioning between Gaseous Phase and Particulate Phase

Dry and wet air deposition of TCEP to land and surface waters may be an important source of TCEP to
the ambient environment. Air deposition may be the result of particle deposition and/or gaseous
deposition.

There is conflicting information about the particle size of TCEP and whether TCEP is present in the gas
or particle phase. A study of offices in China suggests that the mass median aerodynamic diameters
(MMAD) of TCEP is coarse, between 4 and 5 |im, and that the contribution of TCEP is due to indoor
rather than outdoor air (Yame et at.. 2014). Another Chinese study suggests that only 22 percent of

Page 63 of 572

-------
1669

1670

1671

1672

1673

1674

1675

1676

1677

1678

1679

1680

1681

1682

1683

1684

1685

1686

1687

1688

1689

1690

1691

1692

1693

1694

1695

1696

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

TCEP is found among particle size fractions of dust samples less than 43 |im (He et ai. 2018c). A third
Chinese study indicates that the MM AD of TCEP is fine, between 1 and 2 |im (Cao et ai. 2019).
Schreder et al.	indicates that TCEP is not detected in respirable particulate fractions (<4 |im). A

team of Canadian scientists sought to make sense of these discrepancies by examining the gas-particle
partitioning of organophosphate esters. Okeme C evaluated gas-particle partitioning in indoor and
outdoor air by using a group of single-parameter and poly-parameter models. Their predictions suggest
that TCEP should be in the gas phase contrary to measurements. Okeme (2018) suggests that the
unexpectedly high particle fractions reported in many studies is due to sampling artifact. Okeme (2018)
argues that many of the studies with high particle fractions do not account for safe sampling volumes,
and that gas-phase sorption could be contributing substantially to the mass of TCEP captured on the
filters.

As described in the Appendix H.3.3, EPA selected a proportion of emissions in gaseous phase of 82
percent and the proportion in particle phase of 18 percent based on Wolschke et al. C

3.3.2 Water Pathway	

EPA searched peer-reviewed literature, gray literature, water databases to obtain concentrations of
TCEP in surface water, precipitation, and sediment. Sections 3.3.2.1, 3.3.2.3, 3.3.2.7, and 3.3.2.8 display
the aggregated results of reported monitoring and reported modeled concentrations for surface water,
precipitation, and sediment found in the peer-reviewed and gray literature as a result of systematic
review. Sections 3.3.2.4 provides surface water concentrations as a results of surface water databases.
Sections 3.3.2.5, 0, 3.3.2.9, and 3.3.2.10 report EPA modeled surface water and sediment
concentrations.

3.3.2.1 Geosjjatial Analyses of Environmental Releases

No location information is available for facilities that produce, manufacture, or use TCEP. The surface
water data from the Water Quality Portal (WQP) shows TCEP concentration distributed across the
United States. Figure 3-3 indicates the detected water concentrations from the WQP from 1995 to 2022.
Many additional sample sites recorded non-detects, which are not shown in this figure.

Page 64 of 572

-------
1697

1698

1699

1700

1701

1702

1703

1704

1705

1706

1707

1708

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE

December 2023



0>

W

Media
D Effluent

Finished Water

[Groundwater
Hyporheic zone
Landfill effluent
Leadiate
Surface Water
Missing

Figure 3-3. Map of Nationwide Measured TCEP Water Concentrations Retrieved from the Water
Quality Portal, 1995 to 2022

Source: EPA Accessible Link to Interactive Figure.

Size of the dots indicate magnitude of concentration; see Appendix H.2.1 for more details.

3.3.2.1.1 Geospatial Analysis for Tribal Exposures

Although EPA did not identify facilities that release TCEP on or near tribal lands, TCEP has been
detected in surface water and/or groundwater on or near tribal lands. Groundwater samples collected in
2000 downgradient of the Norman Landfill had TCEP concentrations between 0.22 to 0.74 pg/L. Figure
3-4 indicates that the Norman Landfill was also located within a few miles from the Chickasaw Tribal
Lands in Oklahoma. The landfill closed in 1985, was covered with a clay cap, and vegetated (lames et
al.. 2004).

Page 65 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE

December 2023

1711

1712	In 2018, concentrations in groundwater of up to 2.4 |ig/L were detected at the Twenty-Nine Palms Band

1713	of Missions Indians in Coachella, California (Figure 3-5). These concentration data were provided by

1714	EPA's STORage and RETrieval (STORET) Data Warehouse rather than collected as part of landfill

1715	monitoring efforts like the example above. This site was monitored again in 2019 (0.24 |ig/L) and twice

1716	in 2021 (0.79 to 0.84 ug/L) (STORET via ( WIS et al.. 2022V).

1717

Page 66 of 572

-------
1718

1719

1720

1721

1722

1723

1724

1725

1726

1727

1728

1729

1730

1731

1732

1733

1734

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE

December 2023

A summary of surface water monitoring studies is provided in Figure 3-6. Six U.S. studies were
identified (five in the "US Not Specified" section and one in the "Mix Not Specified"). Sengupta et al.
(20141 reported TCEP concentrations at 581 ng/L in October 2011 and 785 ng/L in July 2011 in the Los
Angeles and San Gabriel Rivers during low flow conditions. TCEP concentrations in the Santa Clara
River, California, were recorded up to 810 ng/L during low flow events in 2013 ( vlaruva et al.. 2016).

A Korean study found midstream concentrations of TCEP 9 times higher than upstream values (234 vs.
15.0 ng/L) (Choo et al.. 2018). This study suggested that a potential cause of the elevated TCEP
concentrations was due to an industrial complex involving fiber manufacture being located near the
midstream site.

Figure 3-5. Groundwater Concentration of TCEP Reported near Twenty-Nine Palms Reservation
near Coachella, California

Source: EPA Accessible Link to Interactive Figure.

See Appendix H.2.1 for more details.

3.3.2.2 Measured Concentrations in Surface Water

x

240
ugA

Surface wstw Site s 10
Rrvef/Sfeeam

X

TEPAM

Page 67 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE

December 2023

1735

US Not Specified

Mix Not Specified
NonUS Not Specified

4182703 - Maruya cl al., 2016 - US
4181598 - Sengupta ct al.. 2014 - US
4253347 - Padhye el al., 2014 - US
5469762 - Giorgino cl al., 2007 - US
3353787 - Kolpin ct al.. 2002 - US

4530235 - Scon ct al., 1996 - CA.US

5305891 - Gadclha ct al.. 2019 - FT
5428453 -GaoctaL 2019 -SE
5469295 - McDonough ct al.. 2018 ¦ CA.GL
4829919 - Blum ct al.. 2018 - SE
5428638 - Blum ct al,. 2018 - SE
5469301 - Chooet al.. 2018 - KR
3862723 - Li ct al.. 2017 - AQ
3860951 - Loos ct al.. 2017 - DE
5499542 - Gustavsson ct al.. 2018 - SE
5469274 - Scott ct al.. 2014 - AU
1788425 - Cristalc ct al.. 2013 - GB
4330586 - Malamoros ct al.. 2012 - DK
2588430 - Regnery and Piittmann, 2010 - DE
2919589 - Caldertin-Preciado et al.. 2011 - ES
5469263 - Regnery and Piittmann. 2010 - DE
5469315 - Gourmclon ct al.. 2010 - FR
1250860 - Rodilet al..20l2-ES
2593950 - Quednow and Piittmann. 2009 - DE
1619118 - Andrcscn ct al.. 2007 - DE
4832200 - Andrcscn ct al.. 2004 - DE
5469313 - Fries and Puitmann. 2003 - DE
5469312 - Fries and Puttmann. 2001 - DE

IOA-4

| General Population (Background)

Near Facility (Highly Exposed)

¦ Remote (Not Near Source)

& Normal Distribution (CT ami 90th percentile)
V Lognormal Distribution (CT and 90th percentile)
S3 Non-Detect

A A

VlV

w



¦a





~~









A A

W

t7V

^7

o

¦nri

& v

Cv

0.1	1	10

Concentration (ng/L) (pt I )

NonUS Not Specified

| General Population (Background)
y LognormaJ Disiribulion (CT and 90th percentile)

2919504 ¦ Ishikawa cl al., 1985 ¦ JP
2919504 - Ishikawa ct al., 1985 - JP

10M

w

0.001

0.01

1736

1737

1738

1739

0.1	1	10

Concentration (ng/L) (pt 2)

vv

100

1000

10*4

Figure 3-6. Concentrations of TCEP (ng/L) in Surface Water from 1980 to 2017

Page 68 of 572

-------
1740

1741

1742

1743

1744

1745

1746

1747

1748

1749

1750

1751

1752

1753

1754

1755

1756

1757

1758

1759

1760

1761

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

3.3.2.3 Measured Concentrations in Precipitation

Scott et al. (1996) recorded concentrations of TCEP in precipitation samples from 14.4 to 52.3 ng/L in
Ontario, Canada, collected in 1994 (Figure 3-7).

LIS

4530235 - Scott et a!., 19% - US



B Genera] Population (Background)
I Remote (Not Near Source)



NonUS

3862723-Li ct al..2017-AQ















1









2662833 - Mihajlovic and Fries. 2012 - DE
2662833 - Mihajlovic and Fries. 2012 - DE





¦m

ii







2588430 - Rcgncry and Piitimann. 2010 - DE













2588430 - Regncry and PUttmann. 2010 - DE
2598725 - Rcgncry and Puettmann. 2009 - DE
2598725 - Rcgncry and Puettmann. 2009 - DE
2598725 - Regncry and Puettmann. 2009 - DE





¦

¦







2598725 - Regnery and Puettmann. 2009 - DE













5469313 - Fries and Puttmann. 2003 - DE





HI







0.01 0

1 1 10 100
Concentration (ng/L)

1000

Figure 3-7. Concentrations of TCEP (ng/L) in Precipitation from 1994 to 2014

3.3.2.4 Measured Concentrations in Surface Water Databases

Measured surface water concentrations were obtained from EPA's Water Quality Exchange (WQX)
using the WQP tool, which is the nation's largest source of water quality monitoring data and includes
results from EPA's STORage and RETrieval (STORET) Data Warehouse, the U.S. Geological Survey
(USGS) National Water Information System (NWIS), and other federal, state, and tribal sources.

The complete record of national monitoring of surface water reported by the WQP were reviewed to
summarize the prevalence of TCEP in raw surface water (NWIS et al.. 2022). Data retrieved in January
2023 included sampling dates from 2001 to 2022 and resulted in 9,892 available sample results (Figure
3-8.). Full details of the retrieval and processing of ambient surface water monitoring data from the
WQP are presented in Appendix H.2. Figure 3-8. shows the range of TCEP concentrations detected in
surface water samples the lowest detected sample concentrations within the data set are 0.02 |ig/L. Most
(95 percent) of the sample records available had no level of TCEP detected above the reported detection
limit for the analysis (referred to as "non-detects"). The highest detection limit was 2,720 |ig/L. The 466
detected values ranged from 0.47 to 7.66 |ig/L, with a median of 0.23 |ig/L.

Page 69 of 572

-------
1762

1763

1764

1765

1766

1767

1768

1769

1770

1771

1772

1773

1774

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

0.1	0.2	0.5	1	2

Detected TCEP Surface Water Concentration (ug/L)

Figure 3-8. Frequency of Nationwide Measured TCEP Surface Water Concentrations Retrieved
from the Water Quality Portal, 2003 to 2022

The highest concentrations of TCEP detected in surface water in the United States is 7.66 |ig/L, detected
in August 2013 in Rochester, New York (NWIS via [WQP]). This monitoring location is on the Genesee
river at Ford Street bridge within 1,500 feet downstream of an abandoned Vacuum Oil plant on the west
bank of the Rochester's Plymouth-Exchange neighborhood. The Vacuum Oil plant is a brownfield site
that is being managed by the New York State Department of Environmental Conservation (DEC). EPA
lacks data to confirm whether Vacuum Oil is the source of TCEP. Concentrations of up to 2.55 |ig/L
have been detected in Oregon as recent as October 2020 (STORET via [WQP]). Figure 3-9 demonstrates
that surface water concentrations of TCEP have been decreasing over the last two decades.

Page 70 of 572

-------
1775

1776

Mil

1778

1779

1780

1781

1782

1783

1784

1785

1786

1787

1788

1789

1790

1791

1792

1793

1794

1795

1796

1797

1798

1799

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Water Monitoring in the US by Time (excluding non-detects)

1

0.5

—I

W3 0

c
o

4—'

03

4—>

£ -0.5

c
o
U
no
O

-1

-1.5

2005	2010	2015	2020

Time of Sampling

Figure 3-9. Time Series of Nationwide Measured TCEP Surface Water Concentrations
Retrieved from the Water Quality Portal, 2003 to 2022

Source: EPA Accessible Link to Interactive Figure
See Appendix H2.1 for more details.

3.3.2.5 EPA Modeled Surface Water Concentrations (E-FAST, VVWM-PSC)	

A tiered modeling approach was implemented for estimating surface water concentrations of TCEP.
EPA's Exposure and Fate Assessment Screening Tool, version 2014 (E-FAST 2014) (U.S. EPA. 2007b).
a simple dilution-based model, was first used to estimate total chemical surface water concentrations in
streams. As E-FAST 2014 does not consider chemical partitioning into various media due to physical
and chemical properties (Kow, Koc), it tends to overestimate total surface water concentrations and
underestimate the chemical concentration that is sorbed to soil. Because TCEP's physical and chemical
properties lends it to potentially partitioning into various media (Section 2.2.2), E-FAST 2014-derived
exposures that were greater than the most conservative environmental- or human health-relevant point of
departure (POD) were triaged for further modeling using the VVWM-PSC model which incorporates
partitioning and degradation. The VVWM-PSC model was also used to estimate settled sediment in the
benthic region of streams.

Predicted surface water concentrations were modeled for facility releases as detailed in Section 3.2. The
aquatic modeling was conducted with E-FAST 2014 using hypothetical annual release/loading amounts
(kg/yr) and estimates of the number of days per year that the annual load is released (see Section 3.2 for
more information). As appropriate, two scenarios were modeled per release: release of the annual load
over an estimated maximum number of operating days per year. Additionally, the Probabilistic Dilution
Model (PDM), a module of E-FAST 2014, was run to predict the number of days a stream concentration
will exceed the designated COC value.





•





• ••
_ ¦

•

*

1 * / *

•« /.

N * *
^ f ••

:A

•Ji7v

• w ••

• S

3

•



•

• •
*

• •

§Tl a •

V A ,«] •

t- • •

.%• n i
p •• v • • •• •

• ¦

.

*, • • a

•

•



.S ' '

rg : ..

s. i •• •
• ••

»

* •
: •







• •
•

• •
•• ••

•

• • • *

:t £ i

•* i\







Page 71 of 572

-------
1800

1801

1802

1803

1804

1805

1806

1807

1808

1809

1810

1811

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table 3-5 release estimates are presented based on a 2,500 lb per site-year, high-end estimate release
scenarios, the only deviation from this is the Use of paints and coatings and the Lab chemical OESs.
These deviations are due to single site throughput constraints within the models used, in these cases, the
PV of 2,500 lb/year was used to create a distribution of the possible number of sites. The 2,500 lb was
not divided by COU, rather the full 2,500 lb was considered for each COU. Since CDR reporting is done
on a per site-year basis, EPA estimated a 2,500 lb per site-year. Section 3.2 provides a summary of the
release estimates for each COU/OES. For the maximum days of release scenarios, surface water
concentrations under 7Q10 flow conditions for E-FAST 2014 ranged from 1.27><103 to l.ll><104for the
various exposure scenarios. Results for VVWM-PSC are overall slightly lower for all scenarios since
VVWM-PSC accounts for additional sink effects that are not accounted for in E-FAST 2014. For more
information on E-FAST 2014 and VVWM-PSC, including information on input parameters, see
Appendix H.2.

Page 72 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

1812 Table 3-5. Summary of Modeled Surface Water Concentrations for the 2,500 lb, High-End Release Estimates

Life Cycle
Stage

Category

Subcategory

OES

Inputs

E-FAST 2014

WWM-PSC

Days of
Release

Estimated 7Q10
Flow (mJ/day)

Daily Pollutant
Load (kg/day)

Dailv Concentration
— 7Q10 Oig/L)

Dailv Concentration
-7Q10 Qig/L)

Manufacture

Import

Import

Repackaging

4

4,130

9.88

2,392

2,390

Processing

Incorporated
into

formulation,
mixture, or
reaction
product

Paint and

coating

manufacturing

Incorporation into
paints and
coatings - 1-part
coatings

2

3,380

35.18

10,407

10,200

Incorporation into
paints and
coatings - 2-part
coatings

1

3,380

31.89

9,436

8,280

Polymers used
in aerospace
equipment and
products

Formulation of
TCEP into 2-part
reactive resins

1

2,850

31.54

11,066

9,190

Commercial
Use

Paints and
coatings

Paints and
coatings

Use of paints and
coatings - spray
application

2

4,130

23.26

5,631

5,590

Other use

Laboratory
chemicals

Lab chemical -
use of laboratory
chemicals

182

4,130

0.40

96

96

1813

Page 73 of 572

-------
1814

1815

1816

1817

1818

1819

1820

1821

1822

1823

1824

1825

1826

1827

1828

1829

1830

1831

1832

1833

1834

1835

1836

1837

1838

1839

1840

1841

1842

1843

1844

1845

1846

1847

1848

1849

1850

1851

1852

1853

1854

1855

1856

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

3.3.2.6 EPA Modeled Surface Water Concentrations via Air Deposition (AERMOD)

A study in the lower great lakes suggested that TCEP undergoes net gas phase deposition to lakes at a
flux of-3,980 ng/nr per day (Ma et al. 2021). Other studies in the open ocean have suggested that the
air-water gas exchanges were dominated by volatilization from seawater to air for TCEP 146 ± 239
ng/nr per day (Li et al.. 2017b).

EPA used IIOAC and AERMOD to estimate air deposition from facility releases and to calculate a
resulting pond water concentration near a hypothetical facility. Pond water concentrations from air
deposition were estimated for the COUs with air releases. Air deposition modeling was conducted using
IIO AC and AERMOD. Due to limitations of IIO AC in incorporating gaseous and particulate deposition,
deposition results from the AERMOD were utilized in calculating pond water concentrations. A
description of the ambient air modeling and the deposition results are provided in Section 3.3.1.2. Using
the modeled deposition rates, the TCEP concentration in pond water was calculated with the following
equations:

Equation 3-1

AnnDep = TotDep x Ar x CF

Where:

AnnDep
TotDep
Ar
CF

Total annual deposition to water body catchment (|ig)
Annual deposition flux to water body catchment (g/m2)
Area of water body catchment (m2)

Conversion of grams to micrograms

Equation 3-2

Where:

PondWaterConc =

AnnDep

PondWaterConc

AnnDep

Ar

Pond Depth

CF

Ar x Pond Depth

Annual-average concentration in water body (|ig/L)

Total annual deposition to water body (|ig)

Area of water body (m2); default = 10,000 m2 from EPA OPP

standard farm pond scenario

Depth of pond; default = 2 m from EPA OPP standard farm pond
scenario

Conversion of cubic meters to liters

Appendix H.3.3 presents the range of calculated pond water concentrations for the different emission
scenarios. The highest estimated 95th percentile pond water concentration, across all exposure scenarios,
for the 2,500 lb production volume, high-end estimate was for commercial use of paints and coatings
scenario:

•	1.07xl03 |ig/L or 1,070 |ig/L at 100 m from the source; and

•	8.10 |ig/L at 1,000 m from the source.

Page 74 of 572

-------
1857

1858

1859

1860

1861

1862

1863

1864

1865

1866

1867

1868

1869

1870

1871

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

3.3.2.7 Measured Concentrations in Wastewater

Laundry wastewater may be the primary source of TCEP to wastewater treatment plant influent and
subsequently to the aquatic environment. This theory suggests that the TCEP in the indoor environment
is transferred to indoor dust that is subsequently transferred to clothing. The dust is removed from the
clothing during laundry and this wastewater reaches the wastewater treatment plants. Not all wastewater
treatment plants are fully effective in removing TCEP, and the subsequent effluent may result in higher
concentrations in the aquatic environment (Schreder and La Guardia. 2014). Wastewater monitoring
data from multiple locations in Emeryville, California corroborates this theory, as the highest levels of
TCEP were shown to come from industrial laundry services at levels of 3.72 |ig/L in wastewater
(Jackson and Sutton. 2008). A study in Albany, New York between 2013 and 2015 indicated mean
influent concentrations of 1,430 ng/L and effluent concentrations of 1,100 ng/L of TCEP (Kim et at..
2017). The monitoring data suggests that U.S. values of TCEP in wastewater appear to be higher than
concentrations in other high-income countries as shown in Figure 3-10.

Page 75 of 572

-------
1872

1873

1874

1875

1876

1877

1878

1879

1880

1881

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE

December 2023



Raw Influent





| Treated Effluent





Untreated Effluent at Discharge Origin



, ¦ Untreated Combined Sewer Overflow



V Lognormal Distribution (CT and 90th percentile»



A Normal Distribution (CT and 90th percentile)

US



H Non-Detect





3862000 - Kim cl al.. 2017 - US





¦Ka





3862000 - Kim et al.. 2017 - US





M



2528320 - Schreder and La Guardia. 2014 - US









2528320 - Schreder and La Guardia. 2014 - US











5469289 - Laws cl al.. 2011 - US











1408465 - Jackson and Suuon. 2008 - US





1 A A



1408465 - Jackson and Sutlon. 2008 - US









5743010 - Lorainc and Pcttigrov, 2006 - US





4A













NonUS











7002475 - Norwegian Environment. 2019 - NO





•





5428453 - Gao ct al.. 2019 - SB





V





5428453 ¦ Gao ct al.. 2019 - SE











4457234 - Been et al.. 2017 - BE





m





5664394 - Launay ct al.. 2016 - DE





f w





5664394 - Launay ct al.. 2016 - DE











4143122- Blum ct al.. 2017-SE











3035438 - O'Brien ct al.. 2015 - AU





¦B





5469315 - Gourmclon et al.. 2010 - FR





za





1250860 - Rodil ct al.. 2012 - ES





wmsa





1250860 - Rodil ct al.. 2012 - ES





W





5162720 - Meyer and Bcslcr. 2004 - DE





EA





5162720 - Meyer and Besier. 2004 - DE











8683710 - Marklund ct al. 2005 - SE











8683710 - Marklund et al. 2005 - SE





av





8683710 - Marklund ct al. 2005 - SE











8683710 - Marklund et al. 2005 - SE





¦ES





8683710 - Marklund ct al. 2005 - SE





&





8683710 - Marklund cl al. 2005 - SE





^37





5469313 - Fries and Puttmann. 2003 - DE





¦





5469313 - Fries and Puttmann. 2003 - DE











0.01 0.1

10 100 1000 10*4 10*5



Concentration (ng/L)



Figure 3-10. Concentrations of TCEP (ng/L) in Wastewater from 2001 to 2018

3.3.2.8 Measured Concentrations in Sediment

Limited information was available on measured concentrations of TCEP in sediment in the United
States. Maruva et al. (2016 detected TCEP in coastal embayments at up to 6.98 ng/g dry weight in
Marina Del Ray, Los Angeles, California, in 2013. The mean sediment TCEP concentration was 2.2
ng/g with a 90th percentile value of 4.0 ng/g Maruva et al. (2016). Concentrations of TCEP were
reported at a maximum of 41 ng/g in sediment samples of the Elbe River at the mouths of five tributaries
after a flooding event in Europe in August 2002 (Stachel et al.. 2005).

Page 76 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

1882

1883

1884

1885

1886

1887

1888

1889

US Drv

4182703 - Maruya el al., 2016 - US



HHU General Population (Background)

Near Facility {Highly Exposed)
¦ Unknown/Not Specified
V Lognormal Distribution (CT and 90th percentile)
A Normal Distribution (CT and 90th percentile)



NonUS Drv

5305891 - Gaddha ct al.. 2019 - PT

















5470119 - Chokwe and Okonkwo. 2019 - ZA



V V





5469301 - Choo et al.. 2018 - KR



Km





5740077 - Siachcl ct al., 2005 - CZ.DE
2919504 - Ishikawa ct al., 1985 - JP







NonUS Wet

2935128 * Brandsma ct al.. 2015 - NL



v





0.001

0.01

0.1 1 10
Concentration (ng/g)

100

Figure 3-11. Concentrations of TCEP (ng/g) in Sediment from 1980 to 2017

3.3.2.9 EPA Modeled Sediment Concentrations (VVWM-PSC)	

A summary of the benthic pore water and sediment concentrations modeled using VVWM-PSC are
summarized by COU/OES in Table 3-6. Modeled estimates are presented for the 2,500 lb production
volume, high-end estimate release scenarios. Section 3.2.2 provides a summary of the release estimates
for each COU/OES. For the maximum day of release scenarios, sediment concentrations ranged from
8.94x 102 to 5.04x 103 |ig/kg for the 2,500 lb production volume, high-end estimate release scenarios.

Page 77 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

1890	Table 3-6. Summary of Modeled Benthic Pore Water and Sediment Concentrations for the 2,500 lb Production Volume, High

1891	Estimate Releases

Life Cycle
Stage

Category

Subcategory

OES

Inputs

VVWM-PSC

Days of
Release

Estimated
7Q10 Flow
(m3/day)

Daily
Pollutant

Load
(kg/day)

Benthic Pore Water
Concentration
(Hg/L)

Sediment
Concentration

(ng/g)

Manufacture

Import

Import

Repackaging

4

4,130

9.88

155

894

Processing

Incorporated
into

formulation,
mixture, or
reaction
product

Paint and coating
manufacturing

Incorporation into
paints and coatings
- 1-part coatings

2

3,380

35.18

339

1,960

Incorporation into
paints and coatings
- 2-part coatings

1

3,380

31.89

155

893

Polymers used in
aerospace
equipment and
products

Formulation of
TCEP into 2-part
reactive resins

1

2,850

31.54

185

1,070

Commercial
Use

Paints and
coatings

Paints and
coatings

Use of paints and
coatings - spray
application OES

2

4,130

23.26

180

1,040

Other use

Laboratory
chemicals

Lab chemical - use
of laboratory
chemicals

182

4,130

0.40

66

380

1892

Page 78 of 572

-------
1893

1894

1895

1896

1897

1898

1899

1900

1901

1902

1903

1904

1905

1906

1907

1908

1909

1910

1911

1912

1913

1914

1915

1916

1917

1918

1919

1920

1921

1922

1923

1924

1925

1926

1927

1928

1929

1930

1931

1932

1933

1934

1935

1936

1937

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

For more information on the VVWM-PSC methodology, including inputs used, please see Appendix
H.2.4.

3.3.2.10 EPA Modeled Sediment Concentrations via Air Deposition (AERMOD)

EPA used AERMOD to estimate air deposition from facility releases and calculate a resulting sediment
concentration near a hypothetical facility. Sediment concentrations from air deposition were estimated
for the condition of use scenarios with air releases. Air deposition modeling was conducted using IIOAC
and AERMOD. Due to limitations of IIOAC in incorporating gaseous and particulate deposition,
deposition results from the AERMOD were utilized in calculating sediment concentrations. A
description of the modeling and the deposition results is provided above in Section 3.3.1.2. Additional
details on IIOAC and AERMOD are presented in Appendix H.3.3. Using the modeled deposition rates,
the TCEP concentration in sediment was calculated with the following equations:

Equation 3-3

AnnDep = TotDep x Ar x CF

Where:

AnnDep
TotDep
Ar
CF

Total annual deposition to water body catchment (|ig)
Annual deposition flux to water body catchment (g/m2)
Area of water body catchment (m2)

Conversion of grams to micrograms

Equation 3-4

Sediment Concentration

Where:

Sediment Cone

AnnDep

Ar

Pond Depth

Mix
Dens

0=

AnnDep

Ar x Mix x Dens

Annual-average concentration in water body (|ig/kg)

Total annual deposition to water body (|ig)

Area of water body (m2); default = 10,000 m2 from EPA OPP
standard farm pond scenario

Depth of pond; default = 2 m from EPA OPP standard farm pond
Scenario

Mixing depth (m); default = 0.1 m

Density of sediment; default = 1,300 kg/m3 from the European
Commission Technical Guidance Document (	B).

Appendix H.3.3 presents the range of calculated sediment concentrations for the different emission
scenarios. Equation 3-4 is conservative as it does not include a water solubility parameter. The highest
estimated 95th percentile sediment concentration amongst all exposure scenarios was for the 2,500 lb
production volume, high end estimate release commercial use of paints and coatings scenario:

•	1.64xl04 |ig/kg or 16,400 |ig/kg at "fenceline" population (100 m from the source); and

•	1.25xl02 |ig/kg or 125 |ig/kg at "community" population (1,000 m from the source).

3.3.3 Land Pathway

EPA searched peer-reviewed literature, gray literature, water databases to obtain concentrations of
TCEP in soil, biosolids, and groundwater. Sections 3.3.3.1, 3.3.3.3, and 3.3.3.5 display the aggregated
results of reported monitoring and reported modeled concentrations for soil, sediment, and groundwater

Page 79 of 572

-------
1938

1939

1940

1941

1942

1943

1944

1945

1946

1947

1948

1949

1950

1951

1952

1953

1954

1955

1956

1957

1958

1959

1960

1961

1962

1963

1964

1965

1966

1967

1968

1969

1970

1971

1972

1973

1974

1975

1976

1977

1978

1979

1980

1981

1982

1983

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

found in the peer reviewed and gray literature as a result of systematic review. Section 3.3.3.6 provides
groundwater concentrations from water databases. Sections 3.3.3.2, 3.3.3.4, and 3.3.3.7 report EPA
modeled and estimated soil and groundwater concentrations.

3.3.3.1	Measured Concentrations in Soil

There are no reported soil concentrations of TCEP in the United States. A research team in Germany
observed concentrations of TCEP from 5.07 to 23.48 ng/g dry weight. Snow melt appears to be a
contributor to amplified soil concentrations. The highest soil concentrations were observed one day after
snow melt at 23.48 ng/g, whereas soil concentrations at the same location before snowfall were below 8
ng/g. The meltwater generated at the snow surface percolated downwards due to gravity picking up
chemicals present at the snow grain edge (Mihailovic ar	2012). These authors suggested that the

source of the TCEP may be due to its use in cars (Mihailovic et ai. 2011). TCEP levels ranged from
1.03 to 2.30 ng/g dry weight in Bursa, Turkey, a city known for its textile and automotive parts
manufacturing (Kurt-Karakus et al. 2018).

3.3.3.2	EPA Modeled Soil Concentrations via Air Deposition (AERMOD)

EPA used AERMOD to estimate air deposition from facility releases and calculate a resulting soil
concentration near a hypothetical facility.

Soil concentrations from air deposition were also estimated for the COUs with air releases (see Table
3-3 for a crosswalk of COU/OES with air releases). The air deposition modeling was conducted using
IIO AC and then AERMOD. A description of the modeling and the deposition results is provided above
in Section 3.3.1.2. Using the modeled deposition rates, the TCEP concentration in soil was calculated
with the following equations:

Equation 3-5

AnnDep = TotDep x Ar x CF

Where:

AnnDep
TotDep
Ar
CF

Equation 3-6

Total annual deposition to soil (|ig)
Annual deposition flux to soil (g/m2)
Area of soil (m2)

Conversion of grams to micrograms

Where:

SoilConc

AnnDep

Mix

Ar
Dens

SoilConc =

AnnDep

Ar x Mix x Dens

Annual-average concentration in soil (|ig/kg)

Total annual deposition to soil (|ig)

Mixing depth (m); default = 0.1 m from the European Commission
Technical Guidance Document (TGD) (ECB. 2003)

Area of soil (m2)

Density of soil; default = 1,700 kg/m3 from TGD (ECB. 2003)

The above equations assume instantaneous mixing with no degradation or other means of chemical
reduction in soil over time and that TCEP loading in soil is only from direct air-to-surface deposition
(i.e., no runoff).

Page 80 of 572

-------
1984

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017

2018

2019

2020

2021

2022

2023

2024

2025

2026

2027

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Appendix 481H.3.3 presents the range of calculated soil concentrations corresponding to the emission
scenarios considered. From the table, the highest estimated 95th percentile soil concentration amongst
all exposure scenarios was for the commercial use of paints and coatings scenario:

•	1.14x 104 |ig/kg at "fenceline" population (100 m from the source); and

•	8.65x 101 |ig/kg at "community" population (1,000 m from the source)

3.3.3.3	Measured Concentrations in Biosolids

Wastewater and liquid waste treatment can result in effluent discharge to water and land application of
biosolids. A study of a wastewater treatment plant in New York reported means of combined sludge
concentrations (40.1 ng/g dry weight), ash (47.7 ng/g dry weight), and sludge cake (78.9 ng/g dry
weight) (Kim et	). TCEP in concentrations up to 3 17 ng/g dry weight (mean of 10.6 ng/g) was

detected in sewage sludge collected from wastewater treatment plants located in the United States
(Wane et at.. 2019c). Due to its persistence, it is likely that dissolved TCEP will eventually reach
surface water and groundwater via runoff after the land application of biosolids. TCEP has been found at
concentrations of 4 ng/g in Canada in biosolids (Woudneh et at.. 2015).

3.3.3.4	EPA Calculated Soil Concentrations via Biosolids

Section 2.2.3.1 indicates that TCEP will not be removed after undergoing wastewater treatment and will
be retained in effluents with a low fraction being adsorbed onto sludge.

To assess soil concentrations resulting from biosolid applications, EPA relied upon modeling work
conducted in Canada (ECU	) that used Equation 60 from TGD (	33), as follows:

Equation 3-7

Dur _ ^sludge ^ ARsiUdge

soil ~ Dsoil x BDsoa

Where:

PECsoii	= Predicted environmental concentration (PEC) for soil (mg/kg)

Csiudge	= Concentration in sludge (mg/kg)

ARsiudg	= Application rate to sludge amended soils (kg/m2/yr); default = 0.5 from

Table A-11 of TGD

Dsoil	= Depth of soil tillage (m); default = 0.2 m in agricultural soil and 0.1 m in

pastureland from Table A-l 1 of TGD
BDsoii	= Bulk density of soil (kg/m3); default = 1,700 kg/m3 from Section 2.3.4 of

TGD

The concentration in sludge was assumed as 0.079 mg/kg dry weight based on Kim et at. (21 Using
these assumptions, the estimated soil concentrations after the first year of application were 0.116 |ig/kg
in tilled agricultural soil and 0.232 |ig/kg in pastureland.

A limitation of Equation 3-7 is that it assumes no losses from transformation, degradation, volatilization,
erosion, or leaching to lower soil layers. Section 3.3.3.7 describes the potential leaching of TCEP from
landfills. Additionally, it is assumed there is no input of TCEP from atmospheric deposition and there
are no background TCEP accumulations in the soil. EPA has also assumed that there is only one
application of biosolids per year.

Page 81 of 572

-------
2028

2029

2030

2031

2032

2033

2034

2035

2036

2037

2038

2039

2040

2041

2042

2043

2044

2045

2046

2047

2048

2049

2050

2051

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

3.3.3.5 Measured Concentrations in Groundwater

TCEP was detected in a groundwater plume downgradient (0.22 to 0.74 |ig/L) of the Norman Landfill,
Oklahoma. The Norman Landfill is a municipal unlined landfill (subtitle D) established in 1920 and
closed in 1985 (Barnes et al.. 2004). One domestic well in Elkhart, Indiana reported TCEP
concentrations of 0.65 to 0.74 |ig/L between 2000 and 2002. This domestic well was near Himco Dump,
a historical waste site, used for disposal until 1976 (Buszka et al.. 2009). A study from Fort Devens,
Massachusetts reported concentrations of 0.28 to 0.81 |ig/L at monitoring wells down-gradient of a land
application facility (Hutchins et al.. 1984). These studies suggest that there is potential for TCEP to
migrate to groundwater and domestic wells from nearby non-hazardous waste landfills (e.g., Norman
Landfill) or historical waste sites (e.g., Himco Dump, Indiana, Fort Devens, Massachusetts).



Near Facility (Highly Exposed)





¦ General Population (Background)





V Lognormal Distribution (CT and 90th percentile)



A Normal Distribution (CT and 90th percentile)

US

B1 Non-Detect



5469289 - Laws et al.. 2011 - US









1	S



3975066 - Hopple el al.. 2009 - US























4912133 - Buszka ct al., 2009 - US









/HI



4832201 - Barnes el al.. 2008 - US













5469339 - Barnes el al.. 2004 - US







M r T ¦



1316091 - Hutchins el al.. 1984 - US













NonUS













5428453 - Gao et al., 2019 - SE





v



2579610 - Regnery et al.. 2011 - DE











II

<
<]











2579610 - Regnery ct al.. 2011 - DE







5469313 - Fries and Puttmann. 2003 - DE













5469312 - Fries and Puttmann. 2001 - DE















KM





5469582 - Yasuhara. 1994 - JP







9





0.01 0,

i i

10 100 1000



Concentration (ng/L)

1

Figure 3-12. Concentrations of TCEP (ng/L) in the Not Specified Fraction of Groundwater from
1978 to 2017

3.3.3.6 Measured Concentrations in Groundwater Databases

Data were retrieved from the WQP to characterize observed concentrations of TCEP in groundwater.
These monitored values may or may not represent locations used as a source for drinking water and are
analyzed to characterize the observed ranges of TCEP concentrations in groundwater—irrespective of
the reasons for sample collection. Data retrieved in January 2023 included sampling dates from 1995 to
2021 and resulted in 51 detected results. Figure 3-13 shows most (98%, n = 3,325) of the sample records
available had no TCEP detected above the reported detection limit for the analysis (referred to as "non-
detects"). The 51 detects had a median value of 0.21 |ig/L. Full details of the retrieval and processing
groundwater monitoring data from the WQP are presented in Appendix H.2.

Page 82 of 572

-------
2052

2053

2054

2055

2056

2057

2058

2059

2060

2061

2062

2063

2064

2065

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

0.2 0.5 1 2	5 10 20	50

Detected TCEP Groundwater Concentration (ug/L)

500 1,001

Figure 3-13. Frequency of Nationwide Measured TCEP Groundwater Concentrations Retrieved
from the Water Quality Portal, 1995 to 2021

The highest concentrations of TCEP detected in groundwater in the United States is 610 |ig/L, detected
in April 2002 in Idaho. Other samples at similar locations in April 2004 were an order of magnitude
lower (2.8 to 94 |ig/L) (NWIS et al.. 2022). These estimates are from groundwater wells along the
Gooding Milner Canal in the Magic Valley. Also in 2002, TCEP was detected in groundwater in
Belleview, Florida, at a concentration of 3.5 |ig/L. A more recent value (May 2017) detected TCEP in
groundwater at a concentration of 2.4 |ig/L in New Mexico. The New Mexico monitoring location is a
well in the Four Hills Village in Albuquerque, New Mexico, which is about 1 to 2 miles from the
Kirtland AFB Landfill. Generally, based on the WQP data, concentrations of TCEP in groundwater have
been decreasing over the last two decades.

Page 83 of 572

-------
2066

2067

2068

2069

2070

2071

2072

2073

2074

2075

2076

2077

2078

2079

2080

2081

2082

2083

2084

2085

2086

2087

2088

2089

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

1995

Water Monitoring in the US by Time (excluding non-detects)

I .
• ••

• h'

2000

2005	2010

Time of Sampling

2015

2020

Figure 3-14. Time Series of Nationwide Measured TCEP Groundwater Concentrations Retrieved
from the Water Quality Portal, 1995 to 2021

Source: EPA Accessible Link to Interactive Figure.

See Appendix H.2.1 for more details.

3.3.3.7 EPA Modeled Groundwater Concentrations via Leaching (DRAS)

Landfills may have various levels of engineering controls to prevent groundwater contamination. These
can include industrial liners, leachate capturing systems, and routine integration of waste. However,
groundwater contamination from disposal of consumer, commercial, and industrial waste streams
continues to be a prominent issue for many landfills throughout the United States (Li et al.. 2015; Li et
al.. 2013). These contaminations may be attributed to perforations in the liners, failure of the leachate
capturing system, or improper management of the landfills. Groundwater contamination with TCEP may
occur when the chemical substance is released to landfills, underground injection wells, or surface
impoundments. Due to its physical and chemical properties (e.g., water solubility, Henry's law constant)
and fate characteristics (e.g., biodegradability, half-life in groundwater), TCEP is anticipated to persist
in groundwater for substantially longer than in other media.

Several sources of TCEP may contribute to groundwater concentrations including industrial facility
releases and disposal of consumer products in landfills. With many manufacturing and processing uses
phased out, EPA expects environmental releases of TCEP from industrial facilities to be declining. In
fact, EPA has seen concentrations in surface water and groundwater generally declining over time.
However, environmental releases from landfills may remain (or increase). EPA considered the potential
for groundwater contamination following disposal of waste containing TCEP to landfills.

Page 84 of 572

-------
2090

2091

2092

2093

2094

2095

2096

2097

2098

2099

2100

2101

2102

2103

2104

2105

2106

2107

2108

2109

2110

2111

2112

2113

2114

2115

2116

2117

2118

2119

2120

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

This assessment was completed using the Hazardous Waste Delisting Risk Assessment Software
(DRAS). DRAS was specifically designed to address the Criteria for Listing Hazardous Waste identified
in Title 40 Code of Federal Regulations (40 CFR) Section 261.11(a)(3), a requirement for evaluating
proposed hazardous waste delisting. In this assessment, DRAS is being utilized to determine potential
groundwater concentrations of TCEP after TCEP-containing consumer products have been disposed of
into a non-hazardous waste landfill. To understand possible exposure scenarios from these ongoing
practices, EPA modeled groundwater concentrations of TCEP leaching from landfills where TCEP or
consumer products containing TCEP have been disposed. The greatest potential for release of disposed
TCEP to groundwater is from landfills that do not have an adequate liner system.

Potential groundwater concentrations resulting from disposal of TCEP to landfills vary across landfill
loading rates and concentrations of TCEP in leachate. Estimated exposures presented here are therefore
based on varying landfill conditions. Production volumes of 2,500 lb (1,134 kg) and 25,000 lb (11,340
kg) are used as potential loading rates. This assumes that a combination of raw TCEP and TCEP in
commercial and consumer goods all goes to a single landfill each year.

Masoner et al. (2014a) analyzed leachate concentrations from various landfills across the United States
in 2011 and 2012. In 2011, the reported range of TCEP in leachate concentrations in these landfills
ranged from 8.0><10_1 to 3.2X101 |ig/L, with a median of l.OxlO1 |ig/L and a detection frequency of 35
percent. In 2012, the maximum leachate concentration was 9.1 x 10_1 |ig/L with a detection frequency of
27 percent (Masoner et al.. ). To account for the uncertainties in these estimates a range of leachate
concentrations were selected for the DRAS model. Because DRAS calculates a weight adjusted dilution
attenuation factor (DAF) rather than a groundwater concentration, a back of the envelop computation
was used to convert the DAF to a potential concentration that people living within one mile of a landfill
might be exposed if the release were not identified and remediated. For more information on the DRAS
model please see Appendix H.5.

Table 3-7. Potential Groundwater Concentrations (jig/L) of TCEP Found in Wells within

1 Mile of a Disposal Faci

ity Determined Using the DRAS Model

Leachate Concentration

Loading Rate (kg)

(jig/L)

1.00E03

1.00E04

1.00E-01

1.08E-03

1.01E-02

1.00E00

1.08E-02

1.01E-01

1.00E01

1.08E-01

1.01E00

1.00E02

1.08E00

1.01E01

Concentrations organized by potential loading rates (kg) and potential leachate concentrations (ng/L)

Page 85 of 572

-------
2121

2122

2123

2124

2125

2126

2127

2128

2129

2130

2131

2132

2133

2134

2135

2136

2137

2138

2139

2140

2141

2142

2143

2144

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

3.4 Concentrations of TCEP in the Indoor Environment

TCEP - Concentrations in the Indoor Environment (Section 3.4):

Key Points

EPA evaluated the reasonably available information for concentrations of TCEP in the indoor
environment. The key points are summarized below:

• The indoor environment exposure characterization focused on consumer uses, disposals, and
background exposures of TCEP.
o Indoor air monitoring data show TCEP in particulate or vapor/gas form with

concentrations primarily between 1 x 10~2 and 1 x 104 ng/m3.
o Indoor dust is an important exposure pathway for TCEP. EPA found monitoring data
showing a range of TCEP concentrations in indoor dust in residential spaces, public
spaces, and vehicles, with concentrations as high as 167,532 ng/g in homes.

The indoor environment exposure characterization focuses on consumer uses, disposals, and background
exposures of TCEP. In addition to the contribution from consumer uses, indoor environment TCEP
concentrations were estimated from ambient contributions for air.

Note that indoor air and dust concentrations from consumer uses are presented in Section 5.1.2,
Consumer Exposures.

For more information on TCEP indoor monitoring and reported indoor modeling data, please see:

•	Environmental Monitoring Concentrations Reported by Media Type (U.S. EPA. 2023 g).

•	Environmental Monitoring and Biomonitoring Concentrations Summary Table (U.S. EPA.
2023f).

•	Data Quality Evaluation Information for General Population, Consumer, and Environmental
Exposure. (U.S. EPA. 2023v)

•	Data Extraction Information for General Population, Consumer, and Environmental Exposure
(U.S. EPA. 2023p)

3.4.1 Indoor Air Pathway

3.4.1.1 Measured Concentrations in Indoor Air

The indoor air monitoring data indicates indoor air concentrations primarily between 1 x 10~2 and 1 x 104
ng/m3 ranges. One study indicated particulate concentrations of TCEP of up to l.lxlO7 ng/m3 max in
PM2.5 (Wallner et al.. 2012). This study may have had issues with sampling artifacts due to the use of
glass filters as described by Okeme (2018) (see Section 3.3.1.2 for more details). There was only one
study on vapor/gas in the United States. Dodson et al. (2017) has a 95th percentile concentration of 37
ng/m3 TCEP in vapor/gas.

Page 86 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE

December 2023

2145

2146

US Combined Vapor/Gas and Paniculate

5432871 - Dodson cl aU 2019 - US

NQftUS Oirn.fyne (pt 1)

100

I (XX)

10A4

Page 87 of 572

-------
2147

2148

2149

2150

2151

2152

2153

2154

2155

2156

2157

2158

2159

2160

2161

2162

2163

2164

2165

2166

2167

2168

2169

2170

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

(continued)

NonlJS Vapor/Gas

2537005 - Fromme ct al., 2014 - DE
788335 - Bcrgh ct al.. 2011 - SE
788335 - Bcrgh ct al.. 2011 ¦ SE
5469670 - Luongo and Ocsiman. 2016 ¦ SE
1249459 - Bcrgh ct al.. 2011 - SE
779503 - Haitmann ct al.. 2004 - CH
779503 - Hartmann ct al., 2004 - CH
1949033 - Yoshida ct al.. 2006 - JP
789515-Olake et al.. 2004-JP
1598712-Otakcctal . 2001 - JP

IOA-4

¦ Public Space
Residential
Vehicle

y Lognornial Distribution (CT and 90th percentile)

I V V

¦

0.1	I	10

Concentration (ng/m3) (pt 2)

100

1000

10*4

Figure 3-15. Concentrations of TCEP (ng/m3) in Indoor Air from 2000 to 2016

3.4.1.2 Measured Concentrations in Personal Air	

Two studies measured TCEP in personal air in the U.S. Personal air refers to the area within the
breathing zone. Schreder et al. (2016) conducted a study on white-collar workers in urban, suburban,
and rural areas of Washington State. Participants were instructed to wear an Institute of Occupational
Medicine (IOM) sampler affixed to a shirt collar within the breathing zone continually during a 24-hour
day during normal activities, including at home and at work, traveling to and from home and work,
shopping, and socializing, and to wear or hang the sampler at breathing zone level during sleep.

Schreder et al. (2016) reported mean and maximum inhalable (>4 |im) TCEP concentrations of 19.1
ng/m3 and 77.8 ng/m3 respectively, detected in 8/9 participants. La Guardia and Hale (2015) conducted a
study measuring flame retardants among the personal air of four gymnastic coaches at their workplace
and their homes. TCEP was not detected in the personal air of these coaches. Okeme et al. (2018)
reported a median personal air concentration of three Canadian office workers of 34 ng/m3.
Polydimethylsiloxane (silicone rubber) brooches were used for the sampling methodology, and the three
participants wore the samplers for 7 days.

US Particulate

3222316 - Schreder ct al, 2016 - US





¦¦¦ General Population (Background)



NonUS Particulate

5017615 - Okeme et al.. 2018 - CA

NonUS Vapor/Gas





















III











3357642 - Xu ct al., 2016 - NO

0.001 0.01 0.1 1 10 100

Concentration (ng/m3)

Figure 3-16. Concentrations of TCEP (ng/m3) in Personal Inhalation in General Population
(Background) Locations from 2013 to 2016

Page 88 of 572

-------
2171

2172

2173

2174

2175

2176

2177

2178

2179

2180

2181

2182

2183

2184

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

3.4.1.3	EPA Modeled Indoor Concentrations as a Ratio of Ambient Air

IIOAC calculates a mean and high-end indoor air concentration based on the outdoor/ambient air
concentration and the mean and high-end indoor-outdoor ratios. In IIOAC, indoor-outdoor ratios of 0.65
and 1 are used for the mean and high-end ratios, respectively. The indoor-outdoor ratio of 0.65 is used to
calculate indoor air concentrations corresponding to the mean outdoor air concentration for each
potentially exposed population. The indoor-outdoor ratio of 1 is used to calculate the indoor air
concentration corresponding to the 95th percentile of outdoor air concentration of each potentially
exposed population.

IIOAC was used as a tier 1 screening model before estimating ambient exposures via AERMOD.

Results of IIOAC are presented in Appendix H.3.

3.4.1.4	Reported Modeled Concentrations in Indoor Air

Shin i	reported TCEP emission rates in a whole house of 48.417 mg/day. Emission rate refers

to the amount of chemical emitted per unit time.

Page 89 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE

December 2023

2185

2186

3.4.2 Indoor Dust Pathway

3.4.2.1 Measured Concentrations in Indoor Dust

2187

2188

us

Mix

NonUS

4161719 - Hoffman el al.. 2017 - US
5163584 - Phillips el al.. 2018 - US
6968217 - Shin et al- 2019 - US
3012534 - La Guardia and Hale. 2015 - US
3012534 - La Guardia and Hale. 2015 - US
2343712 - Stapleton el al.. 2014 - US
2528320 • Schreder and La Guardia. 2014 - US
2539068 - Bradman el al., 2014 - US
2215665 - Shin el al.. 2014 - US
1676728 -Fang el al.. 2013-US
1676728 - Fang el al., 2013 - US
3864462 - Caslorina et al., 2017 - US

5184432 - Tan el al.. 2019 - CN.US

5043338 - Velazquez-G6mcz et al., 2019 - ES
5043338 - Veldzquez-GcSmez et al.. 2019 - ES
5043338 - Veldzquez-Gomez et al.. 2019 - ES
5163693 - Raniakokko et al.. 2019 - F1
5165944 - Liu and Mabury. 2019 - CA
5412073 - Giovanoulis et al.. 2019 - SE
3223090 - Longer el al.. 2016 - DK
3223090 - Longer et al., 2016 ¦ DK
4292121 - Christia et al.. 2018 - GR
4292129 - Deng et al.. 2018 - CN
4292133 - Persson et al.. 2018 - SE
3862555 - Zhou el al., 2017 - DE
3862555 - Zhou el al.. 2017 - DE
3862555 - Zhou et al.. 2017 - DE
4285929 - He et al.. 2018 - AU

0.01

| I Residential
¦ Public Space
Vehicle

V Lognnrnial Distribution (CT and 90th percentile)
A Normal Distribution (CT and 90th percentile)

E7V
M

I V V

I

I V V

1 V V

V
' V

AA

¦7 V

w

IV V

IV V

I V V
I VV
I V V

~ v

I V

0.1

K7V

I V V

10	100	1000	10*4

Concentration (ng/g) (pt 1)

10A5

I0A6

10^7

Page 90 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE

December 2023

2189 (continued)

NonUS

4285929- He ct al..20l8-AU
4285929-He et al..20l8-AU
4292136 - Larsson et al.. 2018 - SE
3005686 - Takeuchi ct al.. 2015 - JP
3357642 - Xu ct al., 2016 - NO
4178500 - Kim and Tanabe, 2017 - KR
4178500 - Kim and Tanabe. 2017 - KR
4433160 - Kadcmoglou ct al., 2017 - GB.NO
1313395 - Wallncr et al.. 2012 - AT
3604490 - Tbkumura ct al.. 2017 - JP
3975074 - Sugcng ci al. 2017 - NL
4433160 - Kadcmoglou et al.. 2017 - GB
4829235 - Ail Bamai ct al., 2018 - JP
1927602 - Ali ct al.. 2012 - NZ
2537005 • Fromme ci al.. 2014 - DE
2540527 - Brandsma et al.. 2014 - NL
2540527 - Brandsma ct al.. 2014 - NL
3350460 - Coclho ct al.. 2016 - PT
5164389 - Brommcr ct al.. 2012 - DE
788335 - Bcrgh et al.. 2011 - SE
788335 - Bergh et al.. 2011 - SE
1927614 - Van den Ecde et al.. 2012 - BE.ES.RO
2542290 - Tajima ct al.. 2014 - JP
2543095 - Fan ct al., 2014 - CA
3015040 - Mizouchi ct al.. 2015 - JP
5469392 - Bastiaensen ct al.. 2019 - JP
5469670 - Luongo and Ocstman. 2016 - SE
697390 - Kana/awa ct al.. 2010 - JP
2919501 - Marklund ct al.. 2003 - SE
2919501 - Marklund ct al.. 2003 - SE

0.01

¦	Residential

¦	Public Space
¦¦¦ Vehicle

BBI Other

y Lognonnal Distribution (CT and 90th percentile)
A Normal Distribution (CT ami 90th percentile)

^^7



¦IHD







^37

AA

I V V

I V V

K3 v

V7

ka v

¦7 V

^7

~V

o.i

10	100	1000 10" 4

Concentration (ng/g) (pi 2)

I0A5

10*6

10*7

NonUS

2919501 - Marklund ct al.. 2003 - SE
2919501 - Marklund et al.. 2003 - SE
4731349 - Ingerowski ct al.. 2001 - DE

0.01

j Residential
IB Vehicle
A Normal Distribution (CT ami 90th percentile)









1











I



I	10	100	1000	I0A4

Concentration (ng/g) (pi 3)

10*5 10*6

10*7

Figure 3-17. Concentrations of TCEP (ng/g) in Indoor Dust from 2000 to 2019

Page 91 of 572

-------
2195

2196

2197

2198

2199

2200

2201

2202

2203

2204

2205

2206

2207

2208

2209

2210

2211

2212

2213

2214

2215

2216

2217

2218

2219

2220

2221

2222

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Concentrations of TCEP in dust were significantly higher in facilities with napping equipment (e.g.,
foam beds and mats) made from foam (Bradman et alj ). Correlations between organophosphate
esters in dust and consumer products containing foams, furniture, and electronics strongly implicate
household items as sources of these chemicals (Abate and Martincigh. 2019). In the United States,
concentrations of TCEP in dust are reported at 50.2 ng/g in houses and up to 1,080 ng/g in cars (Fane et
ai. 2013). Phillips et al. (2018) reported maximum concentrations of TCEP of 167,532 ng/g and a
geometric mean of 864.1 ng/g in North Carolina homes from September 2014 to April 2016 as part of
the Toddler's Exposure to SVOCs in the Indoor Environment (TESIE) study. A study of the Center for
the Health Assessment of Mothers and Children of Salinas (CHAMACOS) cohort in California reported
similar concentrations of TCEP as the TESIE cohort. It found that TCEP levels in dust are significantly
associated with the presence of extremely worn carpets (Castorina et al.. 2017).

3.4.2.2 Reported Modeled Concentrations in Indoor Dust

Castorina et al. ( reported modeled oral doses of 0.064 |ig/kg-day for pregnant women via
residential indoor dust in Salinas Valley, California. Schreder et al. (2016) reported 50th percentile
modeled intakes for children (82.8 ng/day) and adults (41.4 ng/day). Ingerowski et al. (2001). a low-
quality study, reported a range of dust intakes of from 0.2 to 2 |ig/day.

Rantakokko et a	modeled inhalation, dermal, and oral intakes of TCEP in children from indoor

dust. Fiftieth percentile intakes were highest for dust ingestion (2.9 ng/kg-day) vs. dermal absorption
(1.3 ng/kg/day) and inhalation (0.023 ng/kg-day). This suggests that for children's exposure to dust, oral
routes may be the most important avenue of exposure. Kademoglou et al. (2017) modeled adult and
toddler daily dust intakes from European homes and offices. They reported mean toddler dust intakes of
14.195 ng/kg/day for the high intake rate and 3.549 ng/kg/day in houses located in the United Kingdom.
Adult intakes were higher in houses (0.624 ng/kg bw with high intake rate) vs. offices (0.0214 ng/kg bw
with high intake for 8 hours spent in offices). The highest observed modeled dust intakes (1.38 |ig/kg-
day) were reported for children at a kindergarten in Hong Kong (Deng et al.. 2018b).

Page 92 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

2223	4 ENVIRONMENTAL RISK ASSESSMENT

2224	EPA assessed environmental risks of TCEP exposure to aquatic and terrestrial species. Section 4.1

2225	describes the environmental exposures through surface water, sediment, soil, air, and diet via trophic

2226	transfer. Environmental hazards for aquatic and terrestrial species are described in Section 4.2, while

2227	environmental risk is described in Section 4.3.

Page 93 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

2228	4.1 Environmental Exposures

2229

TCEP - Environmental Exposures (Section 4.1):

Key Points

EPA evaluated the reasonably available information for environmental exposures of TCEP to aquatic
and terrestrial species. The key points of the environmental exposure assessment are summarized
below:

•	EPA expects the main environmental exposure pathways for TCEP to be surface water,
sediment, and soil. The ambient air exposure pathway was also assessed for its contribution via
deposition to these media.

•	TCEP exposure to aquatic species through surface water and sediment were modeled to
estimate concentrations near industrial and commercial uses. These results were compared to
measured concentrations of TCEP from databases (i.e., WQP) or published literature from a
variety of locations.

o Modeled data estimate surface water concentrations in the low thousands of ppb (Table
4-9) and sediment concentrations low thousands of ppb (Table 4-11) near industrial and
commercial uses.

o Monitoring data show TCEP surface water concentrations in the United States generally

decreasing over the last two decades,
o While EPA does not expect TCEP to bioaccumulate in higher trophic levels in the food
web, biomonitoring from the published literature show TCEP in the tissue of several
aquatic species including fish in the Great Lakes and harbor seals in San Francisco Bay.
o EPA also estimated fish tissue concentrations by COU using the modeled water releases
from industrial and commercial uses.

•	TCEP exposure to terrestrial species through soil, air, and surface water was also assessed using
modeling and monitoring data.

o TCEP exposure to terrestrial organisms occurs primarily through diet via the soil
pathway, with deposition from air to soil being a source. Exposure through diet was
assessed through a trophic transfer analysis, which estimated the transfer of TCEP from
soil through the terrestrial food web using representative species,
o TCEP exposure to terrestrial organisms from surface water ingestion is typically
ephemeral. Therefore, the trophic transfer analysis for terrestrial organisms assumed
TCEP exposure concentrations for wildlife water intake are equal to TCEP soil
concentrations for each corresponding exposure scenario,
o Direct exposure of TCEP to terrestrial receptors via air was not assessed quantitatively
because dietary exposure was determined to be the driver of exposure to wildlife. The
contribution of TCEP exposure from inhalation relative to the ingestion exposure route is
not expected to drive risk because of dilution associated environmental conditions.

2230	4.1.1 Approach and Methodology

2231	Soil and surface water are the major environmental compartments for TCEP (see Section 2.2.2). The

2232	environmental exposure assessment focuses on TCEP concentrations in surface water, sediment, and soil

2233	as these are the media used to determine risks to aquatic and terrestrial organisms (see Section 4.3).

2234	Ambient air was also assessed for its contribution via deposition to these media.

Page 94 of 572

-------
2235

2236

2237

2238

2239

2240

2241

2242

2243

2244

2245

2246

2247

2248

2249

2250

2251

2252

2253

2254

2255

2256

2257

2258

2259

2260

2261

2262

2263

2264

2265

2266

2267

2268

2269

2270

2271

2272

2273

2274

2275

2276

2277

2278

2279

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Monitoring information for aquatic and terrestrial species are presented in Sections 4.1.2 and 4.1.3
below. Reported monitoring information on environmental media (e.g., surface water, sediment, air) are
presented in Section 3.3. When available, measured TCEP concentrations from databases (i.e., WQP) or
published literature were as used as comparative exposure concentrations for risk quotient (RQ)
calculations and are presented in Section 4.3.

EPA utilized various models to assess the environmental concentrations resulting from the industrial and
commercial release estimates (Section 3.3). These models are E-FAST, VVWM-PSC, IIOAC, and
AERMOD. Additional information on these models is available in Section 3.3. TCEP surface water
concentrations (ppb) were modeled by E-FAST and VVWM-PSC. TCEP pore water and benthic
concentrations were modeled using VVWM-PSC as described in Section 3.3.2.9. TCEP concentrations
in soil and water via air deposition at the community level (1,000 m from the source) were modeled as
described in Sections 3.3.2.10 and 1.1.1, respectively. Reported and modeled surface water and sediment
concentrations were used to assess TCEP exposures to aquatic species.

Measured and modeled soil concentrations were utilized to assess risk to terrestrial species via trophic
transfer (see Section 4.1.4). Specifically, trophic transfer of TCEP and potential risk to terrestrial
animals was based on modeled soil data from AERMOD and concentrations reported within Mihailovic
and Fries (2012). Potential risk to aquatic dependent wildlife utilized surface water concentrations
modeled via VVWM-PSC for each COU in combination TCEP fish concentrations calculated using the
whole body BCF reported within (Arukwe et at.. 2018). Exposure factors for terrestrial organisms used
within the trophic transfer analyses are presented in Section 4.1.4. Application of exposure factors and
hazard values for organisms at different trophic levels is detailed within Section 4.3 and utilized
equations as described in the U.S. EPA Guidance for Developing Ecological Soil Screening Levels (U.S.
EPA. 2005a).

For more information on TCEP monitoring data in aquatic and terrestrial species, please see the
following supplemental documents:

•	Environmental Monitoring Concentrations Reported by Media Type (	23g).

•	Environmental Monitoring and Biomonitoring Concentrations Summary Table (U.S. EPA.
2023a

•	Data Quality Evaluation Information for General Population, Consumer, and Environmental
Exposure. (U.S. EPA. 2023v)

•	Data Extraction Information for General Population, Consumer, and Environmental Exposure

(	2023p)

4.1.2 Exposures to Aquatic Species

4.1.2.1 Measured Concentrations in Aquatic Species

A graphical survey of TCEP concentrations in fishes within reasonably available published literature
(seven studies) is presented in Figure 4-1. Guo e*	measured concentrations of TCEP in fish

samples in the Great Lakes Basin using the Great Lakes Fish Monitoring and Surveillance Program
(GLFMSP) sampling protocol. TCEP was found in more than 50 percent of the fish samples at a
geometric mean of 13.3 ng/g lipid, including lake trout (Salvelinus namaycush) or walleye (Sander
vitreus). The lipid-based concentrations of TCEP in Lake Erie fish were significantly higher than those
of the other four Great Lakes. These concentrations are in line with lipid-based concentrations from
Sundkvist et al. (2010). who measured TCEP in mussels (Mytilus edulis), herring (iClupeidae), eelpout

Page 95 of 572

-------
2280

2281

2282

2283

2284

2285

2286

2287

2288

2289

2290

2291

2292

2293

2294

2295

2296

2297

2298

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

(.Zoarces viviparus), salmon (Scilmo salar), and perch (Perca fluviatilis) in Swedish lakes and coastal
areas.

TCEP has been recorded in the blubber of harbor seal (Phoca vitalina) within the San Francisco Bay at a
median concentration of 3.4 ng/g (Sutton et al.. 2019). Sutton et al. (2019) indicated that blubber might
not be a good indicator of exposure to hydrophilic phosphate-based flame retardants due to degradation
and metabolism. Two European studies present lipid concentrations of TCEP in aquatic mammals at
similar levels to the lipid concentrations in fish shown above (Sala et al.. 2019; Hallanger et al.. 2015).

Mix Lipid

3985267 - Guo CI al.. 2017 - CA.US - Other

NonUS Lipid

5164.108 - Sanli'n et al.. 2016 - ES - Whole Organism
5162922 - Hallanger ci al., 2015 - NO - Other
2586188 - Sundkvist et al., 2010 - SE - Muscic/Filci
2586188 - Sundkvist el al.. 2010 - SE - Muscle/Filet
2586188 - Sundkvist et al., 2010 - SE - Muscte/Filet

NonUS Wet

5469301 - Choo et al., 2018 - KR - Liver
5469301 - Choo et al., 2018 - KR - Muscle/Filet
5469301 - Choo et al.. 2018 - KR - Other
5469297 - McGoldrick et al.. 2014 - CA - Other
2935128 - Brandsma et al.. 2015 - NL - Other
6992056 - Evenset et al., 2009 - NO - Liver
6992056 - Evenset et al.. 2009 - NO - Muscle/Fillet
6992056 - Evenset et al., 2009 - NO - Whole Organism

0.001

— General Population (Background)

| Remote ( Not Near Source)

Bl Near Facility (Highly Exposed)

A Normal Distribution (CT and 90th percentile)
y Lognormal Distribution (CT and 90th percentile)

/¦A'

A

W

tv

W





^2?









O v



Concentration (ng/g)

Figure 4-1. Measured Concentrations of TCEP (ng/g) in Aquatic Species - Fish from 2003 to 2016

4.1.2.2 Calculated Concentrations in Aquatic Species	

In addition to considering monitoring data from published literature, EPA modeled concentrations in
fish for each industrial and commercial release scenario (Table 4-1). Concentrations of TCEP in fish
were calculated by multiplying the VVWM-PSC modeled surface water concentrations for each
industrial and commercial releases scenario by the bioconcentration factor of 0.34 L/kg (Arukwe et al..
2018) (Table 2-2). These conservative whole fish TCEP concentrations were utilized within the
screening level assessment for trophic transfer as described in Section 4.1.4.

Page 96 of 572

-------
2299

2300

2301

2302

2303

2304

2305

2306

2307

2308

2309

2310

2311

2312

2313

2314

2315

2316

2317

2318

2319

2320

2321

2322

2323

2324

2325

2326

2327

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table 4-1. TCEP Fish Concentrations Calculated from VVWM-PSC Modeled Industrial and
Commercial TCEP Releases

Scenario Name

Production
Volume (lb/year)

Release
Distribution"

SWC
(jig/L)

Fish Concentration

(ng/g)

Import and repackaging

2,500

High-End

2,370

805

Incorporation into paints
and coatings - 1-part
coatings

2,500

High-End

10,300

3,502

Incorporation into paints
and coatings - 2-part
reactive coatings

2,500

High-End

9,340

3,175

Use in paints and coatings
at job sites

2,500

High-End

5,580

1,897

Formulation of TCEP
containing reactive resin

2,500

High-End

10,900

3,706

Laboratory chemicals

2,500

High-End

96

32

SWC = surface water concentration

a Production volume of 2,500 lb TCEP/year uses high-end estimates (95th percentile for all COUs except the laboratory
chemicals COU that uses the 1st percentile).

These calculated whole fish results are two to three orders of magnitude higher than the reported fish
concentrations in Guo et al. (201 M, who reported a geometric mean of 35.6 ng/g lipid in Lake Erie.
Guo et al. (2 also reported a geometric mean concentration of TCEP in Great Lakes water of
4.64 10 4 |.ig/L via Venier et al. (20141 while Arukwe et al. (2 used a water concentration of
7.75><102 |ig/L to derive the BCF within laboratory-controlled experiments. The current TCEP surface
water concentrations modeled via VVWM-PSC are one to two orders of magnitude greater that values
reported in \wkwe et at	however, it is important to consider that modeled concentration are

intended to represent COU-based source release concentrations.

4.1.2.3 Modeled Concentrations in the Aquatic Environment

E-FAST was used to estimate total TCEP surface water concentration within lotic (i.e., flowing) systems
and represents TCEP concentration within the water column. The days of exceedance modeled in E-
FAST are not necessarily consecutive and could occur throughout a year at different times. Days of
exceedance is calculated as the probability of exceedance multiplied by the total modeled days of
release. While both E-FAST and VVWM-PSC consider dilution and variability in flow, the VVWM-
PSC model can estimate a time-varying surface water concentration, partitioning to suspended and
settled sediment, and degradation within compartments of the water column. VVWM-PSC considers
model inputs of physical and chemical properties of TCEP (i.e., Kow, Koc, water column half-life,
photolysis half-life, hydrolysis half-life, and benthic half-life), allowing EPA to model predicted pore
water and sediment concentrations.

The VVVM-PSC model utilized relatively low stream orders (i.e., depth of 2 m) as a conservative
approach for modeling stream reach. Results within PSC are reported as the maximum concentration
value of the investigated chemical over the specified averaging periods (e.g., 1-day, 3-day, etc.) as well
as a time-series graph of surface water and benthic pore water concentrations (	). TCEP

surface water concentrations (ppb) were modeled by E-FAST and VVWM-PSC and are presented in
Table 4-9 for each COU at a production volume of 2,500 lb per year. TCEP pore water concentration

Page 97 of 572

-------
2328

2329

2330

2331

2332

2333

2334

2335

2336

2337

2338

2339

2340

2341

2342

2343

2344

2345

2346

2347

2348

2349

2350

2351

2352

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

and sediment concentration modeled by VVWM-PSC are presented within Table 4-10 and Table 4-11,
respectively.

EPA used IIOAC and AERMOD to estimate air deposition from facility releases and calculate a
resulting pond water concentration near a hypothetical facility. Pond water concentrations from air
deposition were estimated for the COUs with air releases (Table 4-7). AERMOD results indicate air
deposition to water are not drivers of risk and have significantly reduced TCEP concentrations when
compared to TCEP when modeled within the water column, pore water, and sediment modeling via E-
FAST and VVWM-PSC. For example, the highest estimated 95th percentile pond water concentration
from annual deposition from air to water, across all exposure scenarios, was 8.1 |ig/L for the
Commercial use of paints and coatings scenario at an annual production volume of 2,500 lb. This
highest modeled concentration (8.1 |ig/L) within a pond at 1,000 m from a point source was
approximately 150 times lower than the lowest surface water concentration modeled using VVWM-PSC
(1,270 |ig/L as a maximum 1-day average concentration for the laboratory chemicals scenario at an
annual production volume of 2,500 lb). Although the IIO AC and AERMOD were applied to a generic farm
pond setting to calculate concentrations of TCEP in pond surface water and pond sediment, these models do
not account for media exchange of the chemical of interest as WWM-PSC does.

4.1.3 Exposures to Terrestrial Species

4.1.3.1 Measured Concentrations in Terrestrial Species

Two studies (see Figure 4-2) have reported concentrations of TCEP and a TCEP metabolite bis(2-
chloroethyl) phosphate (BCEP) in bird eggs (Guo et at.. 2018; Stubbings et at.. 2018). From these two
studies the mean concentration of TCEP in birds by wet weight is 5.3 ng/g with a 90th percentile value
of 9.7 ng/g. BCEP was among the most abundant metabolites (0.38 to 26 ng/g ww) in bald eagle
(Haliaeetus leucocephalus) eggs. These values are results of the Michigan Bald Eagle Biosentinel
Program archive that sampled bald eagles in the Great Lakes Region from 2000 to 2012.

Page 98 of 572

-------
2353

2354

2355

2356

2357

2358

2359

2360

2361

2362

2363

2364

2365

2366

2367

2368

2369

2370

2371

2372

2373

2374

2375

2376

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

US Wet

NonUS Wet

5166846 - Gilo el al.. 2018 - US - Bluodjg
5166846 - Guoet al., 2018 - US - Egg (whole)

2823276 - Hubcr ct al.. 2015 ¦ NO ¦ Egg (whole)
4181327 - Chen el al.. 2012 - CA - Egg (whole)
4931691 - Greaves and Letcher. 2014 - CA - Blood
4931691 - Greaves and Letcher. 2014 - CA - Liver
4931691 - Greaves and Letcher, 2014 ¦ CA - Other
4931691 - Greaves and Lelcher. 2014 - CA - Adipose Tissue
4931691 - Greaves and Lelcher. 2014 - CA - Egg (yolk)
4931691 - Greaves and Lelcher, 2014 - CA - Muscle/Filet

NonUS Dry

5017003 - Monclus et al.. 2018 - ES - Feathers
2542346 - Eulaers el al.. 2014 - NO - Feathers
2935128 - Brandsma el al., 2015 - NL - Egg (whole)

IOA-6

¦ I General Population (Background)

| Remote (Not Near Source)

Near Facility (Highly Exposed)
ts Non-Detect

y Lognormal Distribution (CT and 90th percentile)

vv

E7

W

*
m

vv

0.01	1

Concentration (ng/g)

Figure 4-2. Measured Concentrations of TCEP (ng/g) in Terrestrial Species - Bird from 2000 to
2016

Aston et al. (1996) reported TCEP in pine needles (Pinusponderosa) at six out of nine collection sites in
the Sierra Nevada Foothills in the mid-1990s with a geometric mean TCEP concentration of 142 ng/g
and a range of 10 ng/g to 1,950 ng/g (Figure 4-3). Although the source of the TCEP is unknown, the
authors suspected that concentrations may have been due to aerial transport and deposition from nearby
point sources such as incinerators. Samples reported within Aston et al. (1996) were collected in 1993
and 1994 with concentrations from this study representing a period with significantly higher
concentrations of TCEP in production and use (see Section 1.1.1).







| Remote (Not Near Source)



US Wet

5469881 - Aston et al,. 1996 - US - Foliage



y Lot; normal Distribution (CT and 90th percentile)





0.01

0.1

1 10 100 1000
Concentration (ng/g)

10*4

Figure 4-3. Measured Concentrations of TCEP (ng/g) in the Wet Fraction of Terrestrial Species -
Plant in Remote (Not Near Source) Locations from 1993 to 1994

4.1.3.2 Modeled Concentration in the Terrestrial Environment

The contribution of exposure risk from inhalation relative to the ingestion exposure route is not expected
to drive risk because of dilution associated environmental conditions (U.S. EPA. 2003a. b). In addition,
TCEP is not persistent in air due to its short half-life in the atmosphere (ti/2 = 5.8 hours) and because
particle-bound TCEP is primarily removed from the atmosphere by wet or dry deposition (U.S. EPA.
2012d). Air deposition to soil modeling is described in Section 3.3.3.2. EPA determined the primary
exposure pathway for terrestrial organisms is through soil via dietary uptake via trophic transfer. As
described in Section 3.3.3.2, IIOAC and subsequently AERMOD were used to assess the estimated
release of TCEP via air deposition from specific exposure scenarios to soil. Estimated concentrations of

Page 99 of 572

-------
2377

2378

2379

2380

2381

2382

2383

2384

2385

2386

2387

2388

2389

2390

2391

2392

2393

2394

2395

2396

2397

2398

2399

2400

2401

2402

2403

2404

2405

2406

2407

2408

2409

2410

2411

2412

2413

2414

2415

2416

2417

2418

2419

2420

2421

2422

2423

2424

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

TCEP that could be in soil via air deposition at the community level (1,000 m from the source) exposure
scenarios have been calculated and are presented in Appendix G.2.

4.1,4 Trophic Transfer Exposure	

Trophic transfer is the process by which chemical contaminants can be taken up by organisms through
dietary and media exposures and transferred from one trophic level to another. EPA has assessed the
available studies collected in accordance with the 2021 Draft Systematic Review Protocol (
2021) relating to the biomonitoring of TCEP.

TCEP is released to the environment by various exposure pathways (see Figure 2-1). The exposure
pathway for terrestrial organisms is through soil; deposition of TCEP from air to soil is the primary
exposure pathway. A secondary source of TCEP contamination in soil is from the application of
biosolids. However, the concentration of TCEP in soil from biosolids is two orders of magnitude less
than the TCEP soil concentration from air deposition (see Section 3.3). Therefore, biosolid application is
not expected to drive risk within the terrestrial environment. The exposure pathway for water includes
runoff from soil (e.g., after a rain event), deposition from air, and direct releases from water treatment
plants. Sediment TCEP concentrations determined by VVMW-PSC modeling range from 2.6- to 108.8-
fold greater than surface water concentration across all COUs (see Section 3.3.2.9). Indicating that
sediment acts as a sink for TCEP and a source of elevated exposure to TCEP through the dietary
exposure pathway for higher trophic levels in the water column that feed on benthic organisms. Trophic
magnification is not expected in the water column or terrestrial environments but may occur where
TCEP concentrations are high (i.e., in the benthic zone) (Table 2-2).

Representative avian and mammal species are chosen to connect the TCEP transport exposure pathway
via terrestrial trophic transfer from earthworm (Eisenia fetida) uptake of TCEP from contaminated soil
through invertivore avian (American woodcock [Scolopax minor]) and mammal (short-tailed shrew
[Blarina brevicauda]) species, to the American kestrel (Falco sparverius) that feeds on invertebrates,
avian, and small terrestrial vertebrates.

American woodcocks primarily feed on invertebrates with a preference for earthworms. When
earthworms are not available, other soil invertebrates and a small proportion of vegetation may be
consumed. Depending on the location and season, earthworms may comprise 58 to 99 percent of
American woodcock diet (	). Short-tailed shrews primarily feed on invertebrates with

earthworms comprising approximately 31 percent (stomach volume) to 42 percent (frequency of
occurrence) of their diet. American kestrels have a varied diet that includes invertebrates and vertebrates
(mammal, avian, and reptile). The proportion of prey type will vary by habitat and prey availability. For
trophic transfer analysis, the American kestrel diet comprised equal proportions of the three
representative prey species (i.e., one-third earthworm, one-third American woodcock, and one-third
short-tailed shrew), which approximates the dietary composition of the American kestrel winter diet
reported in Meyer and Balgooven (1987). The calculations for assessing TCEP exposure from soil
uptake by earthworms and the transfer of TCEP through diet to higher trophic levels are presented in
Section 4.3.1.10. Because surface water sources for wildlife water ingestion are typically ephemeral, the
trophic transfer analysis for terrestrial organisms assumed TCEP exposure concentration for wildlife
water intake are equal to soil concentrations for each corresponding exposure scenario.

The representative semi-aquatic terrestrial species is the American mink (Mustela vison), whose diet is
highly variable depending on their habitat. In a riparian habitat, American mink derive 74 to 92 percent
of their diet from aquatic organisms, which includes fish, crustaceans, birds, mammals, and vegetation
(Alexand 7). Similar to soil concentrations used for terrestrial organisms, the highest modeled

Page 100 of 572

-------
2425

2426

2427

2428

2429

2430

2431

2432

2433

2434

2435

2436

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE

December 2023

surface water TCEP concentrations with a production volume of 25,000 lb/year was used as a surrogate
for the TCEP concentration found in the American mink's diet in the form of both water intake and a
diet of fish. For trophic transfer, fish concentrations shown in Table 4-1 are used in conjunction with
trophic transfer calculations in Section 4.3.1.1.

Terrestrial TMF

Surface Water

F-Arthwuna

Dapiuui

Ifenduc 7xat:

:und Water

Figure Legend







¦

Partitioriingrr ransportation



~

T ransformation/Degradation









¦

Wastewater Facility













Figure 4-4. Trophic Transfer of TCEP in Aquatic and Terrestrial Ecosystems

The diagram demonstrates uptake from media to biota and trophic transfer through the food web (blue
arrows). The width of the arrows shows relative chemical transport between biota or media. Within the
aquatic environment, the benthic zone is bounded by dashed black lines from the bottom of the water
column to sediment surface and subsurface layers. The depth that the benthic environment extends into
subsurface sediment is site specific. The conceptual model illustrates BCFs, BSAFs, and TMFs for

Page 101 of 572

-------
2437

2438

2439

2440

2441

2442

2443

2444

2445

2446

2447

2448

2449

2450

2451

2452

2453

2454

2455

2456

2457

2458

2459

2460

2461

2462

2463

2464

2465

2466

2467

2468

2469

2470

2471

2472

2473

2474

2475

2476

2477

2478

2479

2480

2481

2482

2483

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

aquatic organisms as shown in Appendix E.2.6. Food intake rates (FIRs) are shown for terrestrial
vertebrates.

4,1,5 Weight of the Scientific Evidence Conclusions for Environmental Exposures

4.1.5.1 Strengths, Limitations, Assumptions, and Key Sources of Uncertainty for the
Environmental Exposure Assessment

Concentrations of TCEP in environmental and biological media are expected to vary. Release from
industrial facilities, indoor sources, and long-range transport may all contribute to concentrations of
TCEP in the environment. Determining the source apportionment of TCEP from each is complex.
Proximity to facilities and other sources is likely to lead to elevated concentrations compared to
locations that are more remote. No manufacturing or processing facility locations were identified for
releases to TCEP. The inability to locate releases by these location contributes to a layer of uncertainty
when selecting model input parameters that are typically informed by location (e.g., meteorological data,
land cover parameters for air modeling, flow data for water modeling).

Limited monitoring data are available for aquatic and terrestrial species in the United States. In addition,
monitoring data collected in previous years when production volume and associated releases of TCEP
into the environment are expected to have been higher than they are currently and expected to be in the
future. When considering older monitoring data and monitoring data from international sources, there
are uncertainties associated with using these data because it is unknown whether those sampling sites are
representative of current sites within the United States. Recent and future estimated levels of TCEP in
the area may be lower than past levels due to reported reductions in releases over time. The predicted
concentrations may be lower than concentrations that consider more years of releases or releases
associated with higher production volumes.

In modeling environmental concentrations of TCEP, EPA acknowledges the conservative nature of the
E-FAST model and the additional refinement provided by the VVWM-PSC model. Water dilution
models can be used to determine the concentration of a chemical in the surface water after a source
emits the chemical into a water body. Because the E-FAST model default values encompass either a
combination of upper percentile and mean exposure parametric values, or all upper percentile parametric
values, the resulting model predictions represent high-end exposures estimates. A simple dilution model
such as E-FAST provides exposure estimates that are derived from a simple mass balance approach and
does not account for partitioning between compartments within a surface water body or degradation over
time in different media, parameters which are relevant to TCEP. For these reasons, EPA utilized a two-
tier approach by complementing the E-FAST modeling with more refined estimates from the PSC model
to describe further environmental exposures.

When modeling using E-FAST, EPA assumed that primary treatment removal at POTWs occurred with
0 percent removal efficiency. EPA recognizes that this is a conservative assumption that results in no
removal of TCEP prior to release to surface water. Section 2.2.1 and Appendix E.2.5.2 discusses the
recalcitrance of TCEP to wastewater treatment systems. This assumption reflects both the uncertainty of
the type of wastewater treatment that may be in use at a direct discharging facility and the TCEP
removal efficiency in that treatment.

EPA used a combination of chemical-specific parameters and generic default parameters when
estimating surface water, sediment, soil, and fish-tissue concentrations. For estimated soil concentrations
from air deposition, specifically, EPA recognizes that different default parameters for gaseous vs.
particle partitioning, may result in concentrations of a higher magnitude. However, EPA used central

Page 102 of 572

-------
2484

2485

2486

2487

2488

2489

2490

2491

2492

2493

2494

2495

2496

2497

2498

2499

2500

2501

2502

2503

2504

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

tendency, high production volume, and high-end, central tendency production volume values to
characterize the variability within and across scenarios. To estimate soil concentrations, EPA also used
central tendency and high-end meteorological inputs.

Comparison of model outputs with monitored values offers one way to ground truth the combination of
model inputs and outputs used. EPA compared monitoring and modeled surface water, sediment, soil,
and fish-tissue concentration estimates. Estimates of fish-tissue concentrations are further discussed in
Section 5.1.3.4.2. In summary, EPA compared monitored and modeled fish tissue concentrations and
found modeled fish concentrations were two to three orders of magnitude higher than those reported for
whole fish within published literature (Section 4.1.2.2). The conservative approach for calculated fish
tissue concentrations presented in Section 4.1.2.2 was utilized for trophic transfer analysis to semi-
aquatic mammals (Section 4.3.1.10). In comparison to measured values reported within published
literature, these calculated values should be viewed as organisms with direct proximity to source of
TCEP release as calculated using VVWM-PSC.

EPA conducted modeling of TCEP concentrations in surface water, pore water, and sediment based on
the assumption that releases entered lotic (flowing) aquatic systems. Although EPA did not consider the
potential impact of persistence and longer-term sinks in lake and estuary environments, localized
deposition of TCEP within 1,000 m from hypothetical release sites from air to soil, water, and sediment
were modeled for each applicable COU via IIOAC and AERMOD.

Page 103 of 572

-------
2505

2506

2507

2508

2509

2510

2511

2512

2513

2514

2515

2516

2517

2518

2519

2520

2521

2522

2523

2524

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

4.2 Environmental Hazards

TCEP - Environmental Hazards (Section 4.2):

Key Points

EPA evaluated the reasonably available information for environmental hazard endpoints associated
with TCEP exposure. The key points of the environmental hazard assessment are summarized below:

• Aquatic species hazard:

o Aquatic hazard data were available for TCEP for three species of fish; however, no
aquatic invertebrate or aquatic plant studies were reasonably available.

o To estimate hazards (mortality) from acute exposures, EPA supplemented the empirical
data with hazard predictions from an EPA predictive tool, Web-based Interspecies
Correlation Estimation. These data were used with the empirical fish data to create a
Species Sensitivity Distribution and calculate a TCEP concentration of concern (COC)
for acute exposures of aquatic species (85,000 ppb) representing the lower 95th percentile
of an HC05 (Table 4-4).

o EPA also calculated a COC for chronic exposures (growth and development of the
Japanese medaka) to aquatic species (55.9 ppb) using empirical fish data (Table 4-4).

• Terrestrial species hazard:

O Terrestrial hazard data for TCEP were available for soil invertebrates, mammals, and
avian species.

O Based on empirical toxicity data for nematodes and earthworms, the chronic hazard
threshold for terrestrial invertebrate is 612 mg/kg soil (Table 4-5).

O Empirical toxicity data for mice and rats were used to estimate a chronic toxicity
reference value (TRY) for terrestrial mammals of 44 mg/kg-bw/day (Table 4-5).

4.2.1 Approach and Methodology	

During scoping, EPA reviewed potential environmental hazards associated with TCEP and identified 14
sources of environmental hazard data shown in Figure 2-10 of Final Scope of the Risk Evaluation for
Tris(2-chloroethyl) Phosphate (TCEP) CASRN115-96-8 (U.S. EPA. 2020b).

EPA completed the review of environmental hazard data/information sources during risk evaluation
using the data quality evaluation metrics and the data quality criteria described in the 2021 Draft
Systematic Review Protocol (U.S. EPA. 2021). Studies were assigned an overall quality determination
of high, medium, low, or uninformative.

EPA assigned an overall quality determination of high or medium to 14 acceptable aquatic toxicity and
17 acceptable terrestrial toxicity studies. For the aquatic studies, two species had appropriate endpoint
concentrations (LC50) for assessing acute hazards. The modeling approach, Web-based Interspecies
Correlation Estimation (Web-ICE) (Version 3.3), can both predict toxicity values for environmental
species that are absent from a dataset and can provide a more robust dataset to estimate toxicity
thresholds. EPA used Web-ICE to supplement empirical data for TCEP for aquatic organisms. Details
outlining the method are included in Appendix F. For terrestrial species, all mammal studies were from
mice and rats used as human health model organisms. These studies were used to calculate a toxicity

Page 104 of 572

-------
2525

2526

2527

2528

2529

2530

2531

2532

2533

2534

2535

2536

2537

2538

2539

2540

2541

2542

2543

2544

2545

2546

2547

2548

2549

2550

2551

2552

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

reference value (TRV) for mammals, which is expressed as doses in units of mg/kg-bw/day. Although
the TRV for TCEP is derived from laboratory mice and rat studies, because body weight is normalized,
the TRV can be used with ecologically relevant wildlife species to evaluate chronic dietary exposure to
TCEP. Representative wildlife species chronic hazard thresholds are evaluated in the trophic transfer
assessments using the TRV.

4.2,2 Aquatic Species Hazard
Toxicity to Aquatic Organisms

EPA assigned an overall quality determination of high or medium to 14 acceptable aquatic toxicity
studies. These studies contained relevant aquatic toxicity data for Japanese medaka (Oryzias latipes),
rainbow trout (Oncorhynchus mykiss), and zebrafish (Danio rerio). EPA identified three aquatic toxicity
studies, displayed in Table 4-2, as the most relevant for quantitative assessment. The remaining 11
studies were represented by results at a sub-organ or mechanistic level, which were considered to be
separated from direct population level effects or did not demonstrate effect(s) at the test concentrations
employed within their study concentrations gradients. The Web-ICE application was used to predict
LC50 toxicity values for 18 additional aquatic organisms (16 fish, 1 amphibian, and 1 aquatic
invertebrate species) from the rainbow trout and zebrafish 96-hour LC50 data (Raimondo and Barron.
2010). The test species (n = 2) and predicted species (n = 18) toxicity data were subsequently used to
calculate the distribution of species sensitivity to acute TCEP exposure.

Aquatic Vertebrates

Fish: Relevant acute toxicity studies for fish that included LC50 data were assigned an overall quality
determination of high for two 96-hour static condition (Alzualde et at.. 2018; Life Sciences Research
L 3a) fish toxicity studies, which evaluated the median lethal concentrations (LC50) from
exposure to TCEP. The acute 96-hour LC50 values for fish were 249 mg/L for rainbow trout (Life
Sciences Research Ltd. 1990a) and 279 mg/L for zebrafish embryo (Alzualde et ai. 2018). The LC50
study for rainbow trout did not meet the assumptions of the Probit test. Therefore, a non-linear
interpolation was used to approximate the LC50 value. The zebrafish embryo study by talde et al.
(2 used a nonlinear regression test (sigmoidal dose-response curve) to calculate the LC50.

Page 105 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

2553 Table 4-2. Aquatic Organisms Environmental Hazard Studies Used for TCEP

Duration

Test Organism (Species)

Endpoint

Hazard Values
(mg/L)

Geometric
Mean" (mg/L)

Effect

Citation
(Data Evaluation
Rating)

AijiKilic \ cilcbi'alcs

Chronic

Fish: Japanese medaka

(Oryzias latipes)

14-day NOEC/LOEC

0.25/1.25

0.559

Developmental
Growth

( n et ah, 2016) (Hieh)

Acute

Fish: rainbow trout

(Oncorhynchus mykiss)

96-hour LC50
96-hour NOEC/LOEC

249
50/100

70.7

Mortality

(Life Sciences Research

L )a) (High)

Fish: zebrafish embryo
(Danio rerio)

96-hour LC50

279

-

Mortality

(Alzualde et aL 2018)

96-hour EC50
96-hour NOEC/LOEC

118

114/171

139.7

Developmental/
Growth

(High)

a Geometric mean of definitive values only.

2554

Page 106 of 572

-------
2555

2556

2557

2558

2559

2560

2561

2562

2563

2564

2565

2566

2567

2568

2569

2570

2571

2572

2573

2574

2575

2576

2577

2578

2579

2580

2581

2582

2583

2584

2585

2586

2587

2588

2589

2590

2591

2592

2593

2594

2595

2596

2597

2598

2599

2600

2601

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

The ChV is the geometric mean of the lowest-observed-effect concentration (LOEC) and no-observed-
effect concentration (NOEC). The overall quality determination for relevant studies with ChV values
were high for two 96-hour studies for rainbow trout and zebrafish (Alzualde et ai. 2018; Life Sciences
Research Ltd. 1990a) and one 14-day study for Japanese medaka (Sun et at..: ). The 96-hour
rainbow trout had a ChV of 70.7 mg/L for mortality (Life Sciences Research Ltd. 1990a). the 96-hour
zebrafish embryo had a ChV of 139.7 mg/L for development and growth (Alzualde et ai. 2018). and the
14-day Japanese medaka had a ChV of 0.559 mg/L for development and growth (Sun et at.. 2016).

No chronic exposure duration data for fish were available. The Sun et at. (2016) study encompassed 14-
day TCEP exposures across approximately 9 days of embryo development followed by approximately 5
days of larval development. The duration of this experimental exposure covering all of embryogenesis
and 5 days of larval development represents sensitive lifestages for fishes. As a result, the Japanese
medaka 14-day NOEC/LOEC for development and growth was the most sensitive endpoint within the
reasonably available data and will be considered a chronic hazard value. For the chronic toxicity
assessment of fish an assessment factor and/or acute-to-chronic ratio will be applied to the chronic
health value (ChV) and will be described within Section 4.2.4.1.

Amphibians

No amphibian studies were available to assess potential hazards from TCEP exposure. However,
modeled data from Web-ICE predicted a bullfrog (Lithobates catesbeianus) 96-hour LC50 of 333 mg/L.
Therefore, amphibians are accounted for within the Web-ICE and species sensitivity distribution (SSD)
results.

Aquatic Invertebrates

No aquatic invertebrate studies were available to assess potential hazards from TCEP exposure.
However, modeled data from Web-ICE predicted daphnia (Simocephalus vetulus) 48-hour EC50 of 337
mg/L. In addition, EPA's Ecological Structure Activity Relationships (ECOSAR) model predicted a
daphnia 48-hour LC50 of 170 mg/L and a ChV of 10 mg/L from TCEP exposure (	322c).

Aquatic Plants

No aquatic plant or algae studies were available to assess potential hazards from TCEP exposure.
However, the ECOSAR model predicted a green algae 96-hour EC50 of 210 mg/L and a ChV of 72
mg/L (	322c).

4,2,3 Terrestrial Species Hazard

EPA assigned an overall quality determination of high or medium to 17 acceptable terrestrial toxicity
studies. These studies contained relevant terrestrial toxicity data for two Norway rat (Rattus norvegicus)
strains (F334 and Sprague-Dawley), two mouse (Mas musculus) strains (CD-I IGS and B6C3F1), 1
earth worm (Eisenia fetida), and 1 nematode (round worms; Caenorhabditis elegans). EPA identified a
total of seven terrestrial toxicity studies, displayed in Table 4-3, as the most relevant for quantitative
assessment.

Terrestrial Vertebrates

Five relevant chronic toxicity studies for terrestrial vertebrates that included no-observed-effect level
(NOEL) and/or lowest-observed-effect level (LOEL) data were assigned an overall quality
determination of high or medium with reproduction, mortality, and/or neurotoxicity (e.g., lesions to
hippocampus) endpoints for rodents (n = 4) and thyroid effects for the single avian toxicity study. One
study with a medium overall quality determination was for the reproduction endpoints reported within

Page 107 of 572

-------
2602

2603

2604

2605

2606

2607

2608

2609

2610

2611

2612

2613

2614

2615

2616

2617

2618

2619

2620

2621

2622

2623

2624

2625

2626

2627

2628

2629

2630

2631

2632

2633

2634

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Matthew	30). Mortality endpoints within the same study received an overall quality

determination of high.

Similarities among mammalian studies with ecologically relevant, population-level effects were
observed. Of the three studies that included mice, two studies resulted in LOEL values. Reproductive
effects (NOEL = 175 mg/kg, LOEL = 700 mg/kg) due to reduced sperm count was shown in Matthews
et al. (1990). An initial dose gradient for a single dose reproduction study found that the lowest test dose
with mortality effects in mice was LOEL = 1,000 mg/kg (Hazletom Laboratories. 1983). Additionally,
ataxia and tremors were noted shortly after dosing of the mice, which may be related to neurotoxicity.
Male rats were more sensitive (NOEL = 88 mg/kg, LOEL = 175 mg/kg) to TCEP exposure through the
oral route for mortality endpoints than females (NOEL = 175 mg/kg, LOEL = 350 mg/kg) (Matthews et
al.. 1990). The 2-year studies for neurotoxicity (degenerative lesions of cerebrum and brain stem) and
mortality endpoints showed a NOEL of 44 mg/kg and a LOEL of 88 mg/kg (]	). A 60-day

Sprague-Dawley rat study also resulted in neurotoxicity with lesions in the hippocampus (Yang et al..
2018a). These studies indicate that neurotoxicity of the brain may be a mode of action (MOA) for TCEP
exposures in rodents.

For avian species, one high-quality study was available for the American kestrel (Fei iit^ ^ al. 2^ l ).
The study reported statistically significant increases in the plasma free thyroid hormones
triiodothyronine (T3) and thyroxine (T4) (LOEL = 0.0025 mg/kg-bw/day) with no effects on body
weight or food consumption from 21-day TCEP exposure through the diet.

Soil Invertebrates

Relevant chronic toxicity studies for soil invertebrates included two studies that were assigned an overall
quality determination of high. The earthworm had a NOEL of 0.1 mg/kg soil and a LOEL of 1.0 mg/kg
soil at 3, 7, and 14 days of exposure to TCEP that showed a significant dose response relationship with
degradation of the digestive tract and exfoliation of the typhi osole (Yame et al.. 2018b). The nematode
study results show a NOEL of 500 mg/kg soil and a LOEL of 750 mg/kg soil at 3 days exposure to
TCEP for reduced growth and shortened lifespan, and an LC50 of 1,381 mg/kg soil at 6 days exposure
to TCEP (Xu et al. ).

Terrestrial Plants

No terrestrial plants studies were available to assess potential hazards from TCEP exposure.

Page 108 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

2635 Table 4-3. Terrestrial Organisms Environmental Hazard Studies Used for TCEP

Duration

Test Organism

Endpoint

Hazard Values
(mg/kg)"

Geometric Mean''
(mg/kg)

Effect

Citation
(Data Evaluation Rating)

\ laninials

Chronic

F344/N rats
(Rattus norvegicus)

2-year NOEL/LOEL

44/88

62.2

Neurotoxicity/
mortality

fNTP. 1991b) (Hieh)

16-week
NOEL/LOEL

Female: 175/350
Male: 88/175

247.5
124.1

Mortality

(Matthews etaL 1990)
(High)

B6C3F1 mice (Mus

musculus)

16-week NOEL/
LOEL

175/700

495.0

Reproduction

(Matthews etaL. 1990)
(Medium)

Sprague-Dawley rat

(Rattus norvegicus)

60-day NOEL/LOEL

50/100

70.7

Neurotoxicity

( e et aL 2018a) (Hiizh)

Acute

CD-I IGS outbred mice

(Mus musculus)

8-day LOEL

1,000

NA

Mortality

(Hazleton Laboratories,
83)(High)

A\ Kill

Chronic

American kestrel (Falco
sparverius)

14-day LOEL

0.0025

NA

Thyroid

( 	Mlli^i)

Soil in\cilclii'iilcs

Chronic

Earth worm (Eisenia
fetida)

3, 7, 14-day,
NOEC/LOEC

0.1/1.0

0.3

Gastrointestinal

( e et aL, 2018b)(Hieh)

Acute

Nematode

(Caenorhabditis elegans)

3-day NOEC/LOEC
6-day LC50

500/750
1,381

612.4
NA

Growth/mortality

(Xii et aL, 2017) (Hiah)

a Hazard values for mammals and avian are in mg/kg-bw/day.

b Geometric means of definitive values only (i.e., >48 mg/kg was not used in the calculation).

2636

Page 109 of 572

-------
2637

2638

2639

2640

2641

2642

2643

2644

2645

2646

2647

2648

2649

2650

2651

2652

2653

2654

2655

2656

2657

2658

2659

2660

2661

2662

2663

2664

2665

2666

2667

2668

2669

2670

2671

2672

2673

2674

2675

2676

2677

2678

2679

2680

2681

2682

2683

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

4,2,4 Environmental Hazard Thresholds

EPA calculates hazard thresholds to identify potential concerns to aquatic and terrestrial species. For
aquatic species, the hazard threshold is called a concentration of concern (COC), and for terrestrial
species, the hazard threshold is called a hazard value or toxicity reference value (TRV). These terms
(COC, TRV, and hazard value) describe how the hazard thresholds are derived and can encompass
multiple taxa or ecologically relevant groups of taxa as the environmental risk characterization serves
populations of organisms within a wide diversity of environments. See Appendix F for more details
about how EPA weighed the scientific evidence. Hazard thresholds are then used to calculate RQs in the
risk characterization step of the environmental risk evaluation. After weighing the scientific evidence,
EPA selects the appropriate toxicity value from the integrated data to use as a hazard threshold for each
assessment type.

For aquatic species, EPA estimates hazard by calculating a COCs for a hazard threshold. COCs can be
calculated using a deterministic method by dividing a hazard value by an assessment factor (AF)
according to EPA methods (	;, 2014b. 2012b).

Equation 4-1

COC = toxicity value '¦ AF

COCs can also be calculated using probabilistic methods. For example, an SSD can be used to calculate
a hazardous concentration for 5 percent of species (HC05). The HC05 estimates the concentration of
TCEP that is expected to be protective for 95 percent of species. This HC05 can then be used to derive a
COC, and the lower bound of the 95 percent confidence interval (CI) of the HC05 can be used to
account for uncertainty instead of dividing by an AF. Aquatic hazard values within Section 4.2.2 are
presented in mg/L, while the subsequent section will demonstrate the calculation of acute and chronic
COC in |ig/L or ppb to conform with conform with modeled and monitored environmental media
concentrations presenting within Section 4.3 Environmental Risk Characterization.

4.2.4.1 Aquatic Species COCs Using Empirical and SSD Data

For the acute COC, EPA used the 96-hour LC50 toxicity data from rainbow trout and zebrafish studies
from Table 4-2 as surrogate species to predict LC50 toxicity values for 18 additional aquatic organisms
(16 fish, 1 amphibian, and 1 aquatic invertebrate species) using the Web-ICE application (Raimondo and
Barron. 2010). The test species (n = 2) and predicted species (n = 18) toxicity data were then used to
calculate the distribution of species sensitivity to TCEP exposure through the SSD toolbox as shown in
Appendix F.2.1.2 (Etterson. 2020). The calculated HC05 was 121.5 mg/L (95 percent CI = 85.0 to 170.6
mg/L). The lower 95 percent CI of the HC05 was then multiplied by 1,000 to convert mg/L to |ig/L (or
ppb) resulting in 85,000 |ig/L. The chronic COC was derived from the ChV of the 14-day LOEC/NOEC
of 0.559 mg/L for Japanese medaka with the application of an AF of 10. The ChV for Japanese medaka
represents effects of development and growth throughout the embryo and larval period for this species
(Sun et al. 2016).

Secondary acute and chronic COCs were derived from the previously described COCs for aquatic
organisms within the water column. Acute data from the use of Web-ICE and subsequent SSD includes
empirical data from fishes and modeled data from: fishes, an amphibian, and the freshwater daphnid
(Simocephalus vetulus). A secondary acute COC was calculated with an addition AF of 10 applied to the
acute COC and a secondary chronic COC was calculated with an AF of 100 applied previously
described fish ChV. This approach considers the data landscape for TCEP environmental hazards and
acknowledges the increased uncertainty associated with the limited number of hazard studies available

Page 110 of 572

-------
2684

2685

2686

2687

2688

2689

2690

2691

2692

2693

2694

2695

2696

2697

2698

2699

2700

2701

2702

2703

2704

2705

2706

2707

2708

2709

2710

2711

2712

2713

2714

2715

2716

2717

2718

2719

2720

2721

2722

2723

2724

2725

2726

2727

2728

2729

2730

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

for aquatic species that will be reflected in the overall confidence derived from hazard thresholds
detailed in Section 4.2.6.1.

The acute COC derived from the HC05 for TCEP is 85,000 |ig/L or ppb.

The secondary acute COC with the additional AF of 10 = 85.0 mg/L/(AF of 10) x 1,000 = 8,500 |ig/L or
ppb.

For the chronic COC, the ChV of the 14-day LOEC/NOEC of 0.559 mg/L for Japanese medaka, based
on development and growth was used. Therefore, the chronic COC = 0.559 mg/L/(AF of 10) = 0.0559
mg/L x 1,000 = 55.9 |ig/L or ppb.

The chronic COC for TCEP is 55.9 ppb.

A secondary chronic COC with the additional AF of 10 = 0.559 mg/L/([AF of 10] [AF of 10]) = 0.00559
mg/L x 1,000 = 5.59 ppb.

4.2.4.2	Aquatic Species COCs Using ECOSAR Modeled Data

ECOSAR modeling estimated potential TCEP hazard values for green algae and daphnia that are
currently not represented with empirical data. The potential extension of information from ECOSAR to
create COCs for aquatic plants and acute and chronic benthic COCs was considered as an alternative
approach to the previously detailed COCs using a combination of empirical and Web-ICE SSD results.
Specifically, predictions for green algae included a 96-hour EC50 of 210 mg/L and a ChV of 72 mg/L
(U.S. EPA. 2022c). Estimated daphnia hazard values were reported with a 48-hour LC50 of 170 mg/L
and ChV of 10 mg/L (U.S. EPA. 2022cY

A COC for aquatic plants was derived with an AF of 100 to account for uncertainties associated with
ECOSAR to empirical hazard values. Acute and chronic COCs are represented using ECOSAR values
from daphnid EC50 and ChV values. An acute COC was derived from the ECOSAR-predicted daphnid
48-hour LC50 of 170 mg/L with an AF of 50 applied. This AF for the acute COC is represented with the
application of an AF of 5 for acute invertebrate hazard value and an additional AF of 10 for uncertainties
associated with the use of an ECOSAR hazard value for a water column invertebrate. A chronic COC
from ECOSAR modeled data utilized the daphnid ChV of 10 mg/L with an AF of 100 applied. As a
result, the chronic COC is represented with the application of an AF (10) for chronic invertebrate hazard
and an additional AF (10) for uncertainties associated with the use of an ECOSAR hazard value for a
water column.

The algae COC derived from an ECOSAR 96-hr LC50 for TCEP with an additional AF of 100 = 210
mg/L/(AF of 100) x 1,000 = 2,100 |ig/L or ppb.

The acute COC derived from an ECOSAR daphnid 48-hr LC50 for TCEP with an additional AF of 50 =
170 mg/L/(AF of 50) x 1,000 = 3,400 |ig/L or ppb.

The chronic COC derived from an ECOSAR daphnid ChV for TCEP with an additional AF of 100 = 10
mg/L/(AF of 100) x 1,000 = 100 |ig/L or ppb.

4.2.4.3	Terrestrial Species Hazard Values

For terrestrial species, EPA estimates hazard by using a hazard value for soil invertebrates, a
deterministic approach, for calculating a TRV for mammals. The TRV is expressed as doses in units of

Page 111 of 572

-------
2731

2732

2733

2734

2735

2736

2737

2738

2739

2740

2741

2742

2743

2744

2745

2746

2747

2748

2749

2750

2751

2752

2753

2754

2755

2756

2757

2758

2759

2760

2761

2762

2763

2764

2765

2766

2767

2768

2769

2770

2771

2772

2773

2774

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

mg/kg-bw/day. Although the TRV for TCEP is derived from laboratory mice and rat studies, body
weight is normalized, therefore the TRV can be used with ecologically relevant wildlife species to
evaluate chronic dietary exposure to TCEP. Representative wildlife species chronic hazard threshold
will be evaluated in the trophic transfer assessments using the TRV. The following criteria were used to
select the data to calculate the TRV with NOEL and/or LOEL data (	)0?a). For more details

see Appendix F.2.2.

Step 1: At least three results and two species tested for reproduction, growth, or mortality general
end points.

•	The minimum dataset required to derive either a mammalian or avian TRV consists of three
results (NOEL or LOEL values) for reproduction, growth, or mortality for at least two
mammalian or avian species. If these minimum results are not available, then a TRV is not
derived.

Step 2: Are there three or more NOELs in reproduction or growth effect groups?

•	Calculation of a geometric mean requires at least three NOEL results from either the
reproduction or growth effect groups.

•	Because there was a single reproduction effect result and no growth effect results, then
proceed to Step 3.

Step 3: If there is at least one NOEL result for the reproduction or growth effect groups:

•	Then the TRV is equal to the lowest reported no-observed-adverse-effect level (NOAEL) for
any effect group (reproduction, growth, or mortality), except in cases where, the NOEL is
higher than the lowest bounded LOEL.

•	Then the TRV is equal to the highest bounded NOEL below the lowest bounded LOEL.

For TCEP, the NOEL for reproduction is 350 mg/kg-bw/day, and the lowest mortality LOEL is 88
mg/kg-bw/day with a NOEL of 44 mg/kg-bw/day.

Toxicity Reference Value (TRV) for Terrestrial Toxicity

The chronic TRV for mammals is 44 mg/kg-bw/day.

For soil invertebrates, EPA estimates hazard by calculating the ChV for a hazard threshold. The ChV is
the geometric mean of the NOEC and LOEC values. Although the most sensitive adverse outcome from
TCEP exposure is for earthworm gastrointestinal damage, the ecologically relevant effects for soil
invertebrates are for reproduction, population, and growth. The nematode NOEC (500 mg/kg soil) and
LOEC (750 mg/kg soil) for reduced growth and shortened lifespan are used to calculate the ChV.

The ChV for soil invertebrates is 612.4 mg/kg soil.

4,2,5 Summary of Environmental Hazard Assessment

For acute aquatic exposures to TCEP, the 96-hour LC50 toxicity values are 249.0 and 279.1 mg/L for
rainbow trout and zebrafish, respectively, from two high-quality studies ( iatde et at.. 2018; Life
Sciences Research I )0a). For chronic aquatic exposures, a ChV is 0.559 mg/L from the Japanese
medaka 14-hour NOEC7LOEC for development and growth (Sun et at.. 2016). No studies were available
for aquatic plants. However, the ECOSAR model estimated a green algae 96-hour EC50 of 210 mg/L
and a ChV of 72 mg/L (U ,S. EPA. 2022c). Although no amphibian or aquatic invertebrate studies were
available to assess potential hazards from TCEP exposure, modeled data from Web-ICE provided a
bullfrog LC50 of 333 mg/L and a daphnid LC50 of 337 mg/L. In addition, the ECOSAR model

Page 112 of 572

-------
2775

2776

2777

2778

2779

2780

2781

2782

2783

2784

2785

2786

2787

2788

2789

2790

2791

2792

2793

2794

2795

2796

2797

2798

2799

2800

2801

2802

2803

2804

2805

2806

2807

2808

2809

2810

2811

2812

2813

2814

2815

2816

2817

2818

2819

2820

2821

2822

2823

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

estimated a daphnid 48-hour LC50 of 170 mg/L and ChV of 10 mg/L from TCEP exposure (U.S. EPA.

2022c).

EPA utilizes COCs derived from aquatic species with empirical and SSD data addressing uncertainties
using additional assessment factors as described in Section 4.2.4.1. EPA also considered ECOSAR
predictions. The acute COC is represented by an SSD with Web-ICE representing fish, an amphibian,
and a daphnid species. The representation of an SSD and derived acute COC was chosen over the
potential extrapolation of a single existing daphnid ECOSAR value. Similarly, the chronic COC derived
from a high-quality study on embryo/larval development in medaka serves as a sensitive endpoint as
compared to the alternative application of an AF of 100 with single daphnid ChV from ECOSAR.

EPA calculated COCs for aquatic organisms inhabiting the water column, which are summarized in
Table 4-4. These COCs will be utilized to determine risk to aquatic organisms from modeled and
published concentrations of TCEP in surface water, benthic pore water, and sediment. EPA calculated
an acute COC from the HC05 of 85,000 ppb for aquatic organisms and a secondary acute COC of 8,500
ppb based on the LC50 toxicity values from 2 test species and 16 additional fish, 1 amphibian, and 1
aquatic invertebrate species using Web-ICE (Raimondo and Barron. 2010). The test species (n = 2) and
derived species (n = 18) toxicity data were then used to calculate the distribution of species sensitivity to
TCEP exposure through the SSD toolbox (Etterson. 2020). The calculated HC05 was 121,500 |ig/L. The
acute COC = lower 95 percent CI of the HC05 = 85,000 |ig/L ppb, and 8,500 ppb secondary acute COC
with the additional AF of 10. For the chronic COC, the ChV of the 14-day LOEC/NOEC of 0.559 mg/L
for Japanese medaka, based on development and growth, was used with the application of an AF of 10,
resulting in 55.9 ppb. EPA also calculated a secondary chronic COC from the chronic COC with an
additional AF of 10, resulting in 5.59 ppb.

For chronic terrestrial mammalian exposures to TCEP, the NOEL, and/or LOEL toxicity data ranged
from a rat NOEL of 50 mg/kg-bw/day to a mouse LOEL of 1,000 mg/kg-bw/day for reproduction,
mortality, and/or neurotoxicity endpoints, and were assigned an overall quality determination of high for
all five studies with the exception of one medium overall quality determination for a reproduction
endpoint (Yang et ai. 2018a; Matthews et al. I , x h I t k Matthews et ai. 1990; Hazleton
Laboratories. 1983). EPA calculated chronic toxicity to mammals from TCEP exposure using a TRV.
The TRV is equal to the highest NOAEL below the lowest LOAEL for mortality. The chronic TRV for
mammals is 44 mg/kg-bw/day (Table 4-5). The TRV is then used as the chronic hazard threshold for
representative species during the trophic transfer assessments.

For soil invertebrate exposure to TCEP, a NOEC of 500 mg/kg soil and a LOEC of 750 mg/kg soil at
three days exposure to TCEP was expressed for reduced growth and shortened lifespan of nematodes.
The ChV is 612 mg/kg soil for growth and reduced lifespan (Xii et al.. 2017) (Table 4-5).

Hazard threshold values for earthworms and American kestrels (Table 4-4) are represented by toxicity
endpoints, including degradation of the digestive track in earthworms and increases in plasma thyroid
hormones in kestrels. Although the most sensitive adverse outcome within soil invertebrates from TCEP
exposure is for earthworm, the ecologically relevant effects for soil invertebrates are for reduced growth
and shortened lifespan with a ChV of 612 mg/kg soil, from which an RQ value can be calculated.
Similarly, while the hazard value for the American kestrel within this analysis is based on elevated
plasma free thyroid concentrations at 7 days, the study did not detect any effects on free thyroid
concentrations, kestrel growth (i.e., body weight), nor food consumption at the conclusion of the 21-day
dietary exposure study with TCEP (Fernie et al.. ). Because the apical assessment endpoint of
growth was not affected, it is difficult to assess the ecological relevancy of the change.

Page 113 of 572

-------
2824

2825

2826

2827

2828

2829

2830

2831

2832

2833

2834

2835

2836

2837

2838

2839

2840

2841

2842

2843

2844

2845

2846

2847

2848

2849

2850

2851

2852

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table 4-4. Environmental Hazard Thresholds for Aquatic Environment

tal Toxicity

Environmental Aquatic Toxicity

Hazard Value

(^g/L)

Assessment
Factor (AF)

coc

(Hg/L)

Acute aquatic exposure:

Lower 95% CI of HC05 from SSD

85,000

N/Afl

85,000

Chronic aquatic exposure: based on fish ChV

559

10

55.9

Secondary acute aquatic exposure: based on
Lower 95% CI of HC05 from SSD

85,000

10

8,500

Secondary chronic aquatic exposure: based on fish ChV

559

100

5.59

a Used lower 95% CI of the HC05 to account for uncertainties rather than an AF

Table 4-5. Environmental Hazard Thresholds for Terrestrial Environmental Toxicity

Environmental Terrestrial Toxicity

Hazard Value or TRV

Mammal

44 mg/kg-bw/day

American Kestrel (Falco sparverius)

0.0025 mg/kg-bw/day

Nematode (Caenorhabditis elegans)

612 mg/kg soil

Earthworm (Eisenia fetida)

0.3 mg/kg soil

4.2.6 Weight of the Scientific Evidence Conclusions for Environmental Hazards	

EPA uses several considerations when weighing and weighting the scientific evidence to determine
confidence in the environmental hazard data. These considerations include the quality of the database,
consistency, strength and precision, biological gradient/dose response, and relevance (see Appendix
F.2.3.1) and are consistent with the 2021 Draft Systematic Review Protocol (	2021). Table 4-6

summarizes how these considerations were determined for each environmental hazard threshold.

Overall, EPA considers the evidence for chronic mammalian hazard thresholds robust, the evidence for
aquatic vertebrate and invertebrate and terrestrial invertebrates hazard thresholds moderate, and the
evidence for chronic avian hazard thresholds slight. Hazard confidence in COCs for secondary acute and
chronic assessments with additional assessment factors are ranked as slight. A more detailed explanation
of the weight of the scientific evidence, uncertainties, and overall confidence levels is presented in
Appendix F.2.3.1.

4.2.6.1 Strengths, Limitations, Assumptions, and Key Sources of Uncertainty for the
Environmental Hazard Assessment
Quality of the Database; and Strength (Effect Magnitude) and Precision

All the studies used to calculate COCs (aquatic fish), TRVs (terrestrial mammals), and hazard thresholds
(terrestrial invertebrates) received a high overall quality determination from the systematic review data
quality evaluation. Effect size was not reported for mammal studies. Effect size was reported for aquatic
fish and nematode studies using LC50s.

Model approaches such as Web-ICE have more uncertainty than empirical data and are not substitutes
for empirical data when determining the hazard or risk. For aquatic organisms, three fish species were
represented in the empirical data from systematic review, and two of these species had data appropriate
for the SSD model. EPA was able to supplement the dataset for aquatic organisms for TCEP with
predictions from Web-ICE, which included predictions for 16 fish species, 1 amphibian species, and 1
invertebrate species. The use of two species available as inputs for the Web-ICE application reduces the

Page 114 of 572

-------
2853

2854

2855

2856

2857

2858

2859

2860

2861

2862

2863

2864

2865

2866

2867

2868

2869

2870

2871

2872

2873

2874

2875

2876

2877

2878

2879

2880

2881

2882

2883

2884

2885

2886

2887

2888

2889

2890

2891

2892

2893

2894

2895

2896

2897

2898

2899

2900

2901

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

confidence in the Web-ICE and subsequent SSD output. However, the use of the probabilistic approach
within this risk evaluation increases confidence compared to a deterministic approach using the two
studies on fishes with acute hazard study endpoints. The use of the lower 95 percent CI instead of a
fixed AF of 5 also increases confidence as it is a more data-driven way of accounting for uncertainty.

A 14-day study with a ChV as an endpoint of growth and development was used to calculate the chronic
COC. The 14-day exposure was conducted throughout both sensitive embryo and larval developmental
periods within the Japanese medaka fish (Sun et al. ). The study duration, developmental periods
of TCEP exposure, and application of an AF 10 increase confidence that the chronic COC was not
underestimated. There were no reasonably available empirical toxicity data available for benthic
organisms. Using the acute and chronic COCs creates an additional uncertainty associated with
extrapolating water column organism sensitivity from TCEP exposure. With the addition of an AF of 10
for secondary chronic COC calculations, confidence decreased that toxicity to aquatic organisms was
represented by empirical data.

For terrestrial mammal species, no wildlife studies were available from systematic review; however,
four high-quality level studies with two species, mice and rats, represented were used from human
health animal model studies. A TRV derived from the mammal studies was used to calculate the hazard
threshold in mg/kg-bw.

For avian species, a single, high-quality level study was available for the American kestrel. The avian
study detected transient differences in thyroid hormone level with no apparent effects on body weight or
food consumption. Although the test did not detect any effects on apical assessment endpoints of
regulatory interest {i.e., impaired growth, survival, or reproduction) and the ecological relevancy of
change in thyroid hormone level is uncertain, the study is still useful for the trophic transfer assessment.
For example, if the results of the trophic transfer show that exposure from TCEP is lower than {i.e., is
protective for) the hazard threshold for effect on thyroid hormones, then a qualitative assertion can be
made that the exposure levels from TCEP do not indicate risk.

For soil invertebrates, two high-quality level soil invertebrate studies were available. The earthworm
study did not have an ecologically relevant endpoint effect, although the earthworm is still useful for
assessing trophic transfer hazards both because of its direct ingestion of soil and because the earthworm
is expected to be part of the diet of other trophic levels (short-tailed shrew, woodcock, and American
kestrel).

Consistency: For aquatic fish species, the behavior effect of hypoactivity under dark phase stimulation
and development/growth effects was similar in Japanese medaka and zebrafish. Activity under light and
dark phases, as well as development/growth effects, were not tested with rainbow trout. Mortality effects
for NOEC/LOEC and LC50s were similar for zebrafish and rainbow trout. The mortality endpoint was
not reported in the Japanese medaka study. However, there is still some uncertainty associated with the
small number of studies (n = 3) to assess consistency in outcomes.

For terrestrial mammal species, human health animal model studies (rats) are in agreement with respect
to neurotoxicity effects resulting from lesions to the brain. Confidence is robust on the MOA for rats on
exposure to TCEP via diet due to neurotoxic effects with lesions to the brain. Three studies included
mice; however only a single study resulted in a LOEL for mortality. The maximum dose in all the
studies that included both rats and mice were all below the single study for mice where the lowest test
concentration resulted in the LOEL.

Page 115 of 572

-------
2902

2903

2904

2905

2906

2907

2908

2909

2910

2911

2912

2913

2914

2915

2916

2917

2918

2919

2920

2921

2922

2923

2924

2925

2926

2927

2928

2929

2930

2931

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

The single avian, earthworm, and nematode studies were insufficient to characterize consistency in their
respective outcomes.

Biological Gradient/Dose-Response

A dose response was reported for all studies used for calculating hazard thresholds as well as the
earthworm study used in trophic transfer. However, because the American kestrel study only had one
dose concentration, no dose-response was reported.

Biological Relevance: Behavior and developmental/growth effects were in agreement between both
species tested, zebrafish and Japanese medaka (Alzualde et ai. 2018; Sun et al. ^ ). Mortality effects
were also in agreement between species tested (zebrafish and rainbow trout). All rat studies across
multiple strains exhibited brain lesions from TCEP exposure that was associated with the mortality
endpoint. Data were insufficient to observe correspondence of adverse outcomes across species within
taxa group for avian of terrestrial invertebrates.

Physical/Chemical Relevance: Empirical data were on the effects of the chemical of interest, which
increases confidence. TCEP was identified, including source, for all organisms. Purity was either not
reported or not analytically verified for rainbow trout, earthworm, one of the mouse/rat studies
(Matthews et al. 1990). and the American kestrel study (Fernie et ai. 2015).

Environmental Relevance: Additional uncertainty is associated with laboratory to field variation in
exposures to TCEP are likely to have some effect on hazard threshold; that is, gavage vs. natural forage
diet for mammals (rats and mice) and invertebrate substrate {i.e., nematodes maintained on nematode
growth medium and earth worms on artificial soil). Test conditions for fish species correspond well with
natural environmental conditions. The creation of secondary acute and chronic COCs considered the
data landscape for TCEP; however, these COCs have decreased environmental relevance when
compared to empirical and probabilistic methods employed when deriving acute and chronic COCs. The
application of addition AFs for these secondary COCs decreases confidence in relevance of these values
and potentially overestimates hazard.

Page 116 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table 4-6. TCEP Evidence Tab

e Summarizing the Overall Confidence Derived

'rom Hazard Thresholds

Types of Evidence

Quality of the
Database

Consistency

Strength and
Precision

Biological Gradient/
Dose-Response

Relevance"

Hazard
Confidence

Aquatic

Acute aquatic assessment

++

++

++

+++

+++

Moderate

Chronic aquatic assessment

++

++

++

+++

+++

Moderate

Secondary acute aquatic
assessment (+ AF)

+

++

++

+++

+

Slight

Secondary chronic aquatic
assessment (+ AF)

+

++

++

+++

+

Slight

1 cnvsliial

Chronic avian assessment

+

+

+

+

++

Slight

Chronic mammalian assessment

++

+++

+++

+++

+++

Robust

Terrestrial invertebrates

++

+

++

++

+++

Moderate

a Relevance includes biological, physical/chemical, and environmental relevance

+++ Robust confidence suggests thorough understanding of the scientific evidence and uncertainties. The supporting weight of the scientific evidence
outweighs the uncertainties to the point where it is unlikely that the uncertainties could have a significant effect on the hazard estimate.

++ Moderate confidence suggests some understanding of the scientific evidence and uncertainties. The supporting scientific evidence weighed against the
uncertainties is reasonably adequate to characterize hazard estimates.

+ Slight confidence is assigned when the weight of the scientific evidence may not be adequate to characterize the scenario, and when the assessor is
making the best scientific assessment possible in the absence of complete information. There are additional uncertainties that may need to be considered.

2933

Page 117 of 572

-------
2934

2935

2936

2937

2938

2939

2940

2941

2942

2943

2944

2945

2946

2947

2948

2949

2950

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

4.3 Environmental Risk Characterization

TCEP - Environmental Risk Characterization (Section 4.3):

Key Points

EPA evaluated the reasonably available information to support environmental risk characterization.

The key points of the environmental risk characterization are summarized below:

•	For aquatic species, chronic RQs are above 1 and have corresponding days of exceedance
greater than 14 days within the sediment compartment (sediment and benthic pore water) for 5
of 20 COUs (Table 4-20). Because of TCEP's affinity to bind to sediment and persistence in
the aquatic compartment, there could be a lasting effect on benthic biota and potential
community-level impacts from chronic TCEP exposure. EPA has moderate confidence in the
RQ inputs for the acute and chronic aquatic assessment.

•	For aquatic species, the laboratory chemicals COU resulted in a chronic RQ greater than 1
with over 14 days of exceedance within surface water (Table 4-20).

•	Monitoring data show RQs from TCEP surface water concentrations and sediment within the
WQP database or published literature were below 1 (Table 4-12). However, differences in
magnitude between modeled and measured concentrations may be due to measured
concentrations not being geographically or temporally close to releases of TCEP from a
facility.

•	For terrestrial species, EPA did not identify RQs greater than or equal to 1.

o RQs for soil invertebrates or terrestrial mammals were less than 1 using either modeled
soil concentrations or concentrations taken from the very limited monitoring data set
available (from an urban area of Germany) (Table 4-21). EPA has moderate confidence
in the RQ inputs for the terrestrial invertebrate assessment,
o RQs were below 1 for all representative species and corresponding trophic level using
TCEP soil concentrations from available published literature. RQs were below 1 for
semi-aquatic terrestrial receptors via trophic transfer from fish and using the highest
modeled TCEP surface water concentrations (Table 4-21). EPA has moderate confidence
in the RQ inputs for the screening level trophic transfer assessment.

EPA considered fate, exposure, and environmental hazard to characterize the environmental risk of
TCEP. For environmental receptors, EPA estimated: (1) risks to aquatic species via water and sediment,
and (2) to terrestrial species via exposure to soil by air deposition and through diet via trophic transfer.
Risk estimates to aquatic-dependent terrestrial species included exposures to TCEP through water and
diet. As described in Section 2.2.2, TCEP is described as a "ubiquitous" contaminant because it is
commonly found in various environmental compartments such as surface water, soil, sediment, and
biota. TCEP's physical and chemical properties suggests that its main mode of distribution in the
environment is water and soil, depending on the media of release (Figure 2-1; Appendix E.2.1.2). TCEP
has the potential to undergo long-range transport in air and water (LTRP) that could be significantly
underestimated when using its physical and chemical properties in QSAR models. Oftentimes TCEP's
behavior in the environment does not align with its physical and chemical properties. TCEP can be
transported to sediment from overlying surface water by advection and dispersion of dissolved TCEP
and by deposition of suspended solids containing TCEP. However, TCEP may partition between surface
water and sediments to varying degrees because of its wide range of Log Koc values (2.08 to 3.46)
(Zhang et al.. 2021: Wang et al.. 2018a: Zhang et al.. 2018b: Cristate et al.. 2013) and high water

Page 118 of 572

-------
2951

2952

2953

2954

2955

2956

2957

2958

2959

2960

2961

2962

2963

2964

2965

2966

2967

2968

2969

2970

2971

2972

2973

2974

2975

2976

2977

2978

2979

2980

2981

2982

2983

2984

2985

2986

2987

2988

2989

2990

2991

2992

2993

2994

2995

2996

2997

2998

2999

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

solubility (7,820 mg/L) ( v N \ jm b; EC. 2009; Hi' . '09), which could contribute to its
mobility in the environment. For example, TCEP in the soil was seen to be vertically transported to
deeper soil horizons, causing TCEP concentrations in the surface soil to be lower (He et al.. 2017;
Bacaloni et al. 2008). TCEP does not undergo hydrolysis under environmentally relevant conditions and
is considered persistent in water (Appendix E.2.3.1), sediment (Appendix E.2.3.2), and soil (Appendix
E.2.4.1).

Direct exposure of TCEP to terrestrial receptors via air was not assessed quantitatively because dietary
exposure was determined to be the driver of exposure to wildlife. The contribution of exposure risk from
inhalation relative to the ingestion exposure route is not expected to drive risk because of dilution-
associated environmental conditions (U.S. EPA. 2003a. b). The gaseous phase of TCEP is expected to
have a short half-life in the atmosphere (ti/2 = 5.8 hours) with a high Koa, suggesting this compound
would adsorb to organic carbon present in airborne particles (Okeme et al.. 2020; Ji et al.. 2019; Wane et
al JO I b; 1 c. i i1 \ . 11. •The resulting particle-bound TCEP would be expected to be removed
from the atmosphere through wet or dry deposition. Annual air deposition to water and soil was modeled
using AERMOD for applicable COUs (Table 4-7), and these modeled values are included as
components within the current environmental risk characterization.

EPA quantitatively assessed TCEP concentrations in surface water, pore water, sediment, and soil for
aquatic and terrestrial receptors via modeled concentrations (EFAST, VVWM-PSC, AERMOD)
representing COU-based releases of TCEP. As reported in Section 3.3.2.5, EPA estimated surface water
concentrations from COU based releases of TCEP and reported from 1,271 ppb (or |ig/L) to 11,066 ppb
with a production volume of 2,500 lb/year. Considered to be a minor component, annual air deposition
of TCEP to water was modeled using AERMOD indicating deposition to a lentic {i.e., relatively static)
system at 1,000 m from the source at 8.1 ppb, which was approximately 150 times less than the lowest
surface water concentration modeled using the model, VVWM-PSC. Mean (± SEM) TCEP surface
water concentrations in ambient water were 0.33 ± 0.02 ppb and ranged from 0.01 ppb to 7.66 ppb for
466 detected values in the WQP (2003 to 2022). TCEP water concentrations in published literature were
reported in Section 3.3.2 and represent ambient TCEP concentrations from surface waters and are not
associated with direct environmental releases of TCEP. Maximum TCEP concentrations in surface
waters were collected near urban environments recorded at 0.581, 0.785, and 0.810 ppb during low-flow
conditions in the Los Angeles, San Gabriel, and Santa Clara Rivers in California, respectively (Maruva
et al.. 2016; Sengupta et al.. 2014).

As reported in Section 3.3.2.9, modeled benthic pore water TCEP concentrations ranged from 138 to
873 ppb for the production volume of 2,500 lb/year, respectively. Modeled sediment concentrations
ranged from 893 ppb (or |ig/kg) to 5,040 ppb for the production volume of 2,500 lb/year. Air deposition
to sediment, as reported in Section 3.3.2.10, indicated the highest annual deposition at 1,000 m was 125
ppb, which is almost 7 times lower than the lowest sediment TCEP value modeled with VVWM-PSC
(Incorporation into paints and coatings - solvent borne at 893 ppb) and about 40 times lower than the
highest PSC value for laboratory chemicals (5,040 ppb). As reported in Section 3.3.3.2, calculated TCEP
soil concentrations resulting from modeled air deposition 1,000 m from the source with a production
volume of 2,500 lb/year ranged from 1.49xl0~6 to 0.0039 mg/kg and 1.92xl0~6 to 0.0055 mg/kg for
central tendency and high-end meteorology conditions.

Section 4.2 details available environmental hazard data and indicates that TCEP presents hazard to
aquatic and terrestrial organisms. For acute exposures, TCEP is a hazard to aquatic animals at 85,000
ppb based on the lower 95 percent CI of the HC05 resulting from an SSD utilizing EPA's Web-ICE
(Raimondo and Barron. 2010) and SSD toolbox applications (Etterson. 2020). For chronic exposures.

Page 119 of 572

-------
3000

3001

3002

3003

3004

3005

3006

3007

3008

3009

3010

3011

3012

3013

3014

3015

3016

3017

3018

3019

3020

3021

3022

3023

3024

3025

3026

3027

3028

3029

3030

3031

3032

3033

3034

3035

3036

3037

3038

3039

3040

3041

3042

3043

3044

3045

3046

3047

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

TCEP is a hazard to aquatic organisms with a ChV of 55.9 ppb for fish. For terrestrial exposures, TCEP
is a hazard to mammals at 44 mg/kg-bw/day and a hazard to soil invertebrates with a ChV of 612 mg/kg.
In addition, TCEP presented sub-organ level hazard values for birds at doses of 0.0025 mg/kg-bw/day
and for soil invertebrates at 0.3 mg/kg soil and will serve to supplement terrestrial receptors via a
conservative approach to estimate risk from trophic transfer.

EPA assigned an overall quality determination of high or medium to 14 acceptable aquatic toxicity
studies and 17 acceptable terrestrial toxicity studies (s qq Risk Evaluation for Tris(2-chloroethyl)
Phosphate - Systematic Review Supplemental File: Data Quality Evaluation of Environmental Hazard
Studies (	,3uV). The Risk Evaluation for / risf 2-chloroethyl) Phosphate - Systematic Review

Supplemental File: Data Quality Evaluation of Environmental Hazard Studies (	E023u)

presents details of the data evaluations for each study, including evaluations of each metric and overall
study quality level. As detailed in Section 4.2.6, EPA/OPPT considers the evidence for terrestrial
chronic mammalian robust, the evidence for aquatic hazard thresholds and terrestrial invertebrates
moderate, and the evidence for terrestrial chronic avian slight.

4.3.1 Risk Characterization Approach

EPA characterized the environmental risk of TCEP using RQs (U.S. EPA. 1998b; Bamthouse et at..
1982). which are defined as

Equation 4-2

RQ = Environmental Exposure Concentration/Hazard Threshold

Environmental exposure concentrations for each compartment {i.e., surface water, pore water, sediment,
and soil) were based on measured {i.e., monitored data and/or reasonably available literature) and/or
modeled {i.e., E-FAST, VVMW-PSC, AERMOD) concentrations of TCEP from Section 3.3
Concentrations of TCEP in the Environment. EPA calculates hazard thresholds to identify potential
concerns to aquatic and terrestrial species. These terms describe how the values are derived and can
encompass multiple taxa or ecologically relevant groups of taxa as the environmental risk
characterization serves populations of organisms within a wide diversity of environments. For hazard
thresholds, EPA used the COCs calculated for aquatic organisms, and the hazard values or TRVs
calculated for terrestrial organisms as detailed within Section 4.2.

RQs equal to 1 indicate that environmental exposures are the same as the hazard threshold. If the RQ is
above 1, the exposure is greater than the hazard threshold. If the RQ is below 1, the exposure is less than
the hazard threshold. RQs derived from modeled data for TCEP are shown in Table 4-9, Table 4-10, and
Table 4-11 for aquatic organisms, and Table 4-15 for terrestrial organisms. For aquatic species, acute
risk is indicated when the RQ is greater than or equal to 1 for acute exposures, or chronic risk is
indicated with a RQ greater than or equal to 1 with days of exceedance at or above 14 days for chronic
exposures. The chronic COC was derived from a 14-day exposure, therefore, the days of exceedance to
demonstrate risk reflects the exposure period for that hazard value. Secondary COCs were represented
from the acute COC and chronic COC with the application of an additional assessment factors (Table
4-4); however, confidence in these COCs are "slight." For terrestrial species, RQ values are calculated
from the hazard value for soil invertebrates (nematode) and TRV for mammals as detailed in Section
4.2.4, and risk is indicated when the RQ greater than or equal to 1.

EPA used modeled {e.g., E-FAST, VVWM/PSC, AERMOD) and measured {e.g., monitoring
information from peer-reviewed literature or relevant databases) data to characterize environmental
concentrations for TCEP and to calculate the RQ. Table 4-7 represents the COUs with relevant

Page 120 of 572

-------
3048

3049

3050

3051

3052

3053

3054

3055

3056

3057

3058

3059

3060

3061

3062

3063

3064

3065

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

environmental releases represented in the current risk characterization on aquatic and terrestrial
receptors. Exposure data are especially helpful to characterize exposures from facilities and/or COUs. In
the absence of facility-specific releases for TCEP, estimated releases were generated for a generic
facility for each COU with production volume scenarios set at 2,500 lb/year (Table 4-7). Exposure data
and corresponding RQ values produced with a production volume of 25,000 lb/year are presented within
Appendix G. Surface water monitoring data on TCEP from available databases such as the WQP and
published literature were used as additional approaches to characterize risk to aquatic receptors. The
purpose of using monitored data and published literature, when available, was to determine if
concentrations in the ambient environment exceeded the identified hazard benchmarks for aquatic and
terrestrial receptors while also providing support for or concurrence with modeled concentrations.

As described in Section 3.3.3.2, IIOAC and subsequently AERMOD were used to assess the estimated
release of TCEP via air deposition from specific exposure scenarios to soil (Table 4-7). Estimated
concentrations of TCEP that could be in soil via air deposition at the community level (1,000 m from the
source) exposure scenarios have been calculated.

Table 4-7. Risk Characterization to Corresponding Aquatic and Terrestrial Receptors Assessed

'or the Following COUs





RQ Values

RQ Values

COU (Life cycle stage/ Category/
Sub-category)

Occupational
Exposure Scenario

Calculated for
Aquatic
Receptors"

Calculated for
Terrestrial
Receptors''

Manufacture/ Import/ Import

Repackaging

Yes

Yes

Processing/ Incorporated into formulation,
mixture, or reaction product/ Paint and coating
manufacturing

Incorporation into
paints and coatings -
1-part coatings

Yes

Yes

Processing/ Incorporated into formulation,
mixture, or reaction product/ Paint and coating
manufacturing

Incorporation into
paints and coatings -
2-part reactive
coatings

Yes

Yes

Processing/ Incorporated into formulation,

Formulation of TCEP





mixture, or reaction product/ Polymers used in

into 2-part reactive

Yes

Yes

aerospace equipment and products

resins





Processing/ Incorporated into article/ Aerospace
equipment and products

Processing into 2-part
resin article

N/Ad

Yes

Processing/ Recycling/ Recycling

Recycling e-waste

EPA did not have sufficient data to
estimate these releases'7

Distribution in Commerce/ Distribution in

Distribution in

Distribution activities (e.g., loading)

commerce

commerce

considered throughout life cycle,
rather than using a single
distribution scenario

Industrial use/ Other use/ Aerospace equipment
and products

Installing article
(containing 2-part
resin) for aerospace
applications
(electronic potting)

Releases expected to be negligiblec

Commercial use/ Other use/ Aerospace equipment
and products

Installing article
(containing 2-part
resin) for aerospace
applications

Releases expected to be negligiblec

Page 121 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

COU (Life cycle stage/ Category/
Sub-category)

Occupational
Exposure Scenario

RQ Values
Calculated for
Aquatic
Receptors"

RQ Values
Calculated for
Terrestrial
Receptors''

Commercial use/ Paints and coatings/ Paints and
coatings

Use in paints and
coatings at job sites

Yes

Yes

Commercial use/ Laboratory chemicals/
Laboratory chemicals

Lab chemical - use of
laboratory chemicals

Yes

Yes

Commercial use/ Furnishing, cleaning, treatment
care products/ Fabric and textile products



End of service life disposal
(Releases and exposures not
quantified)c

Commercial use/ Furnishing, cleaning, treatment
care products/ Foam seating and bedding products



End of service life disposal
(Releases and exposures not
quantified)c

Commercial use/ construction, paint, electrical,
and metal products/ Building/construction
materials - insulation



End of service life disposal
(Releases and exposures not
quantified)c

Commercial use/ Construction, paint, electrical,
and metal products/ Building/construction
materials - wood and engineered wood products -
wood resin composites



End of service life disposal
(Releases and exposures not
quantified)c

Consumer use/Paints and coatings/ Paints and
coatings



No quantified environmental
releases from consumer uses''

Consumer use/Furnishing, cleaning, treatment
care products/ Fabric and textile products



No quantified environmental
releases from consumer uses''

Consumer use/ Furnishing, cleaning, treatment
care products/ Foam seating and bedding products



No quantified environmental
releases from consumer uses''

Consumer use/ Construction, paint, electrical, and
metal products/ Building/construction materials -
insulation



No quantified environmental
releases from consumer uses''

Consumer use/ Construction, paint, electrical, and
metal products/ Building/construction materials -
wood and engineered wood products - wood resin
composites



No quantified environmental
releases from consumer uses''

Disposal/ Disposal/ Disposal



Waste disposal (Landfill or
Incineration, covered in each
COU/OES as opposed to a separate
COU)c

a RQ values calculated for aquatic receptors based on TCEP releases from wastewater, WQP database, and published
literature

h RQ values calculated for terrestrial receptors based on TCEP releases as fugitive air and stack air deposition to soil,
trophic transfer, and published literature
c Section 3.2 provides details on these OESs

d Section 5.1.2.2.5 details the lack of information to characterize exposures for disposal of consumer wastes

3066

3067	EPA used IIOAC and AERMOD to estimate air deposition from hypothetical facility releases and

3068	calculate resulting sediment concentrations to a pond. Air deposition to sediment as reported in Section

3069	3.3.2.10 indicated the highest annual deposition at 1,000 m was 125 |ig/kg which is approximately 7

3070	times lower than the lowest sediment TCEP value modeled with VVWM-PSC (incorporation into paints

3071	and coatings - solvent borne at 893 |ig/kg) and approximately 40 times lower than the highest PSC

Page 122 of 572

-------
3072

3073

3074

3075

3076

3077

3078

3079

3080

3081

3082

3083

3084

3085

3086

3087

3088

3089

3090

3091

3092

3093

3094

3095

3096

3097

3098

3099

3100

3101

3102

3103

3104

3105

3106

3107

3108

3109

3110

3111

3112

3113

3114

3115

3116

3117

3118

3119

3120

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

value for laboratory chemicals (5,040 |ig/kg). RQs for each relevant COU listed in Table 4-7 were
calculated for air deposition to sediment at 1,000 m and are available are presented within Appendix G
for both production volumes and meteorological conditions. RQs were greater than 1 for TCEP use in
paints and coatings at job sites with both meteorological conditions for the 2,500 lb/year production
volume. All RQ values for the high production volume scenario of 25,000 lb/year were less than 1, with
the highest RQ at 0.13 for TCEP use in paints and coatings at job sites. The low production volume
scenario modeling used high-end estimates for at 95th percentile of the mean. RQs for the mean (50th
percentile) air to sediment deposition with the AERMOD for both meteorological models were below 1.
It is not anticipated that air deposition to water will significantly contribute as TCEP concentrations
within the water column, pore water, and sediment will utilize modeling via E-FAST and VVWM-PSC.

Frequency and duration of exposure can affect the potential for adverse effects in aquatic receptors.
Within the aquatic environment, a two-tiered modeling approach was employed to predict surface water,
pore water, and sediment TCEP concentrations. If the E-FAST predicted 7Q10 surface water
concentrations were greater than the chronic or acute COCs, the VVWM-PSC model was then used to
confirm whether the predicted surface water concentration days of exceedance as determined by the
acute COC and chronic COC. For TCEP, all six applicable OESs (Table 4-7) modeled in E-FAST
produced chronic RQ values greater or equal to 1, prompting the use of VVWM-PSC for greater
ecological resolution on TCEP concentrations and days of exceedance within the water column and
benthic compartments.

Environmental RQ values by exposure scenario with TCEP surface water concentrations (ppb) were
modeled by E-FAST and VVWM-PSC and are presented in Table 4-9. The max day average
concentrations produced by VVWM-PSC represent the maximum concentration (ppb) over a 1- or 14-
day average period corresponding with the acute or chronic COC used for the RQ estimate.
Environmental RQ values by exposure scenario for aquatic organisms with TCEP pore water
concentration and sediment concentration modeled by VVWM-PSC are presented within Table 4-10 and
Table 4-11, respectively. Scenarios and production volume allow for the calculation of RQs and days of
exceedance that for risk estimation to aquatic organisms (scenarios with an acute RQ greater than or
equal to 1, or a chronic RQ greater than or equal to 1 and 14 days or more of exceedance for the chronic
COC).

VVWM-PSC considers model inputs of physical and chemical properties of TCEP {i.e., Kow, Koc,
water column half-life, photolysis half-life, hydrolysis half-life, and benthic half-life) allowing EPA to
model predicted benthic pore water and sediment concentrations. The role of Koc within the VVWM-
PSC on sediment TCEP concentrations was investigated with a sensitivity analysis. Model inputs for
this physical and chemical property were represented as the mean and 5th percentile of the mean with
values of 2.82 and 2.13, respectively. Results of TCEP concentrations within surface water and benthic
pore water were not influenced by model inputs of Koc; however, sediment concentrations were highly
influenced by this model parameter. The use of the 5th percentile of the mean (2.13) produced TCEP
concentrations for sediment within one to two orders of magnitude of reported within published
literature (Maruva et at.. 2016; Stachel et at.. 2005). Results for VVWM-PSC model output presented
within Section 4.3.2 utilized a Koc value of 2.13, while results utilizing the mean of 2.82 are presented
within Appendix G in TableApx G-2, TableApx G-3, and TableApx G-4.

EPA considers the biological relevance of species that COCs or hazard values are based on when
integrating these values with the location of the surface water, pore water, and sediment concentration
data to produce RQs. Life-history and habitat of aquatic organisms influence the likelihood of exposure
above the hazard threshold in an aquatic environment. EPA has identified COC values associated with

Page 123 of 572

-------
3121

3122

3123

3124

3125

3126

3127

3128

3129

3130

3131

3132

3133

3134

3135

3136

3137

3138

3139

3140

3141

3142

3143

3144

3145

3146

3147

3148

3149

3150

3151

3152

3153

3154

3155

3156

3157

3158

3159

3160

3161

3162

3163

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

aquatic hazard values and include acute COC, chronic COC, secondary acute COC, and secondary
chronic COC. The acute COC for aquatic species is the lower 95 percent CI of the HCos of an SSD, a
modeled probability distribution of toxicity values from multiple taxa inhabiting the water column. The
chronic COC is represented by a growth and development endpoint from 14-day exposures to TCEP
within the water column. Calculated RQ values for pore water and sediment are represented with acute
and chronic COCs in addition to secondary COCs derived from acute and chronic COCs as detailed in
Section 4.2.4. The secondary acute COC and secondary chronic COC values have been applied to
environmental concentrations to demonstrate RQ values for pore water and sediment; however, the
confidence in these RQ inputs were described a "slight" within Table 4-6 as compared to the "moderate"
confidence determinations for the acute COC and chronic COC.

4.3.1.1 Risk Characterization Approach for Trophic Transfer

Trophic transfer is the process by which chemical contaminants can be taken up by organisms through
dietary and media exposures and transfer from one trophic level to another. Chemicals can be transferred
from contaminated media and diet to biological tissue and accumulate throughout an organisms' lifespan
(bioaccumulation) if they are not readily excreted or metabolized. Through dietary consumption of prey,
a chemical can subsequently be transferred from one trophic level to another. If biomagnification occurs,
higher trophic level predators will contain greater body burdens of a contaminant compared to lower
trophic level organisms.

EPA conducted screening level approaches for aquatic and terrestrial risk estimation based on exposure
via trophic transfer using conservative assumptions for factors such as: area use factor, TCEP absorption
from diet, soil, and water. Section E.2.5 details persistence as this compound is expected to persist
within aquatic and terrestrial environments. Under laboratory conditions, mean whole body BCF for
juvenile Atlantic Salmon (Salmo salaf) is reported as 0.34 L/kg wet weight for an experimental
exposure concentration of 1.0 mg/L (Arukwe et al.. 2018). TCEP is not considered bioaccumulative;
however, geometric mean concentrations within biota in Lake Erie have been reported at concentrations
of 35.6 ng/g lipid as reported by Guo et al. (7 in Section 4.1.2. Section 4.1 reports measured
concentrations of TCEP within biota with seven studies indicating TCEP concentrations within whole
fish and lipid (see Section 4.1.2.1), one study within a marine mammal (Section 4.1.2.1), and two studies
with terrestrial organisms (see Section 4.1.3.1). A screening level analysis was conducted for trophic
transfer and formulation of RQ values from aquatic and terrestrial hazard values. If RQ values were
greater than or equal to 1, risk estimation based on potential trophic transfer of TCEP is indicated from
this screening level approach and further refined analysis is warranted. If an RQ value is less than 1, risk
based on potential trophic transfer of TCEP is not indicated from screening level approach and no
further assessment is necessary. The screening level approach employs a combination of conservative
assumptions {i.e., conditions for several exposure factors included within Equation 4-3 below) and
utilization of the maximum values obtained from modeled and/or monitoring data from relevant
environmental compartments.

Following the basic equations as reported in Chapter 4 of the U.S. EPA Guidance for Developing
Ecological Soil Screening Levels (	35a). wildlife receptors may be exposed to contaminants

in soil by two main pathways: incidental ingestion of soil while feeding, and ingestion of food items that
have become contaminated due to uptake from soil. The general equation used to estimate the risk from
exposure via these two pathways is provided below:

Page 124 of 572

-------
3164

3165

3166

3167

3168

3169

3170

3171

3172

3173

3174

3175

3176

3177

3178

3179

3180

3181

3182

3183

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Equation 4-3

([Soilj *PS* FIR * AFsy] + [£f=1 Bij *Pt* [FIR + WIR] * AFiy]) * AUF

Where:

RQj =

Risk quotient for contaminant (j) (unitless)

Soilj =

Concentration of contaminant (j) in soil (mg/kg dry weight)

N

Number of different biota type (i) in diet

Bij =

Concentration of contaminant (j) in biota type (i) (mg/kg dry weight)

Pi

Proportion of biota type (i) in diet

FIR =

Food intake rate (kg of food [dry weight] per kg body weight per day)

WIR =

Water intake rate (kg of water per kg body weight per day)

AF a =

Absorbed fraction of contaminant (j) from biota type (i) (for screening



purposes set equal to 1)

AF sj =

Absorbed fraction of contaminant (j) from soil (s) (for screening purposes set



equal to 1)

HTj =

Hazard Threshold (mg/kg-BW[wet weight]/day)

Ps

Proportion of total food intake that is soil (kg soil/kg food)

AUF =

Area use factor (for screening purposes set equal to 1)

Table 4-8. Terms and Values Used to Assess Potential Trophic Transfer of TCEP for Terrestrial
Risk Characterization



Earthworm

(Eisenia fetida)

Short-Tailed Shrew

American

American Kestrel

(.Falco sparverius)

American

Term

(Blarina
brevicauda)

Woodcock

(Scolopax minor)

Mink

(Mustela visori)

Soilja

0.0055 mg/kgb
TCEP

0.0055 mg/kg b
TCEP

0.0055 mg/kg b
TCEP

0.0055 mg/kg b
TCEP

10.3 mg/Lc
TCEP

N

1

1

1

3

1









0.0055 mg/kg
TCEP (worm)



By

0.0055 mg/kg b
TCEP (soil)

0.0055 mg/kg TCEP
(worm)

0.0055 mg/kg
TCEP (worm)

0.0046 mg/kg
TCEP (short-tailed
shrew)

3.71 mg/kg d
TCEP (Fish)









0.0057 mg/kg
TCEP (woodcock)



P,

1

1

1

0.33

1

FIR

1

0.55 e

0.77e

0.30d

0.22e

WIR

1

0.223e

0.1e

Dietary hydration

0.104®

AFy-

1

1

1

1

1

AF.,

1

1

1

1

1

HTj

0.3 mg/kg -
soil/day

0.66 mg/kg-bw/day

N/Ar

0.0025 mg
TCEP/kg-bw/day

24.2 mg

TCEP/kg-

bw/day

Ps

1

0.03g

0.164g

0.057g

1

AUF

1

1

1

1

1

Page 125 of 572

-------
3184

3185

3186

3187

3188

3189

3190

3191

3192

3193

3194

3195

3196

3197

3198

3199

3200

3201

3202

3203

3204

3205

3206

3207

3208

3209

3210

3211

3212

3213

3214

3215

3216

3217

3218

3219

3220

3221

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Term

Earthworm

(Eisenia fetida)

Short-Tailed Shrew

(Blarina
brevicauda)

American
Woodcock

(Scolopax minor)

American Kestrel

(Falco sparverius)

American
Mink

(Mustela visori)

" TCEP concentration in surface water for Mink
h Highest soil concentration of TCEP obtained using AERMOD modeling (2,500 lb/year)
c Highest surface water concentration of TCEP obtained using WWM-PSC modeling (2,500 lb/year)
d Highest fish concentration (mg/kg) calculated from surface water concentration TCEP (WWM-PSC) and whole body
BCF of 0.34 (Arukwe et at.. 20.1.8')

' Exposure factors (FIR and WIR) sourced from EPA's Wildlife Exposure Factors Handbook (U.S. EPA, 1993b)
' No TCEP hazard threshold value for this representative species is available

g Soil ingestion as proportion of diet represented at the 90th percentile sourced from EPA's Guidance for Developing
Ecological Soil Screening Levels (U.S. EPA, 2005a)

Terrestrial hazard data are available for soil invertebrate and mammals using hazard values detailed in
Section 4.2.4. Representative avian and mammal species are chosen to connect the TCEP transport
exposure pathway via trophic transfer from earthworm uptake of TCEP from contaminated soil through
invertivore avian (American woodcock) and mammal (short-tailed shrew) species, to the American
kestrel that feeds on invertebrates as well as avian and small terrestrial vertebrates.

At the screening level, the conservative assumption is that the invertebrate diet for the American
woodcock and short-tailed shrew comprises 100 percent earthworms from contaminated soil. Similarly,
the dietary assumptions for the American kestrel are 100 percent of the invertebrate, avian, and mammal
diet are from the earthworm, American woodcock, and short-tailed shrew, respectively. Additionally,
the screening level analysis uses the highest modeled or monitored soil contaminate level to determine if
a more detailed assessment is required. Because surface water sources for wildlife water ingestion are
typically ephemeral, the trophic transfer analysis for terrestrial organism assumed TCEP exposure
concentration for wildlife water intake are equal to soil concentrations for each corresponding exposure
scenario.

Exposure factors for food intake rate (FIR) and water intake rate (WIR) were sourced from the EPA's
Wildlife Exposure Factors Handbook (U.S. EPA. 1993b). The proportion of total food intake that is soil
(Ps) is represented at the 90th percentile for representative taxa (short-tailed shrew, woodcock, and
hawk) and was sourced from calculations and modeling in EPA's Guidance for Developing Ecological
Soil Screening Levels (	2005a). Additional assumptions for this analysis have been considered

to represent conservative screening values (II	2005a). Within this model, incidental oral soil

exposure is added to the dietary exposure resulting in total oral exposure greater than 100 percent. In
addition, EPA assumes that 100 percent of the contaminant is absorbed from both the soil (AFSJ) and
biota representing prey (AFy). The proportional representation of time an animal spends occupying an
exposed environment is known the area use factor (AUF) and has been set at 1 for all biota within this
equation (Table 4-8).

The following hazard values were used for trophic transfer of TCEP from media (soil) through trophic
levels: earthworm ChV of 0.3 mg/kg soil, mammal TRV dose of 44 mg/kg-bw/day, and American
kestrel LOEL at doses of 0.0025 mg/kg-bw/day. Short-tailed shew and American mink hazard threshold
values were calculated from the mammal TRV (44 mg/kg-bw/day) to represent the mean short-tailed
shew and American mink body weight values of 0.015 kg and 0.55 kg, respectively, reported in EPA's
Wildlife Exposure Factors Handbook (U.S. EPA. 1993b). It is important to reiterate that hazard values
within this screening-level trophic transfer analysis for earthworm and American kestrel are represented
by endpoints of gastrointestinal damage and increaser plasma thyroid hormones, respectively. Although
the most sensitive adverse outcome within soil invertebrates from TCEP exposure is for earthworm, the

Page 126 of 572

-------
3222

3223

3224

3225

3226

3227

3228

3229

3230

3231

3232

3233

3234

3235

3236

3237

3238

3239

3240

3241

3242

3243

3244

3245

3246

3247

3248

3249

3250

3251

3252

3253

3254

3255

3256

3257

3258

3259

3260

3261

3262

3263

3264

3265

3266

3267

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

ecologically relevant effects for soil invertebrates are for reduced growth and shortened lifespan with a
ChV of 612 soil mg/kg from which an RQ value can also be calculated. The inclusion of earthworms
and kestrels from this screening-level analysis represent an additional conservative approach for
estimating risk to terrestrial organisms via trophic transfer.

For semi-aquatic terrestrial species, the TRV was used with the American mink for the screening level
assessment (Table 4-8). Similar to the above soil concentrations used as term Soil in Equation 4-1, the
highest surface water concentration modeled via VVWM-PSC was used as a surrogate for the TCEP
concentration found in the American mink's diet, which is highly variable depending on habitat. In a
riparian habitat, mink derive 74 to 92 percent of their diet from aquatic organisms, which includes fish,
crustaceans, birds, mammals, and vegetation (Alexander. 1977). The American mink was used as the
representative species for semi-aquatic mammals. As a conservative assumption, 100 percent of the
American mink's diet is predicted to come from fish. Fish concentration (mg/kg) was calculated using
surface water concentrations of TCEP from VVWM-PSC assuming a BCF of 0.34 as reported for whole
body values from 1 mg/L TCEP exposures under laboratory conditions ( :we et al. 2018).

4.3.2 Risk Characterization for Aquatic Receptors

The physical and chemical properties of TCEP and its persistence translate to removal from the water
column by particulate and sediment organic matter and persistence within sediment (see Section 2.2.2).
TCEP may partition between water and sediment due to its physical and chemical properties and, as a
result, exposure of TCEP and the duration of that exposure to organisms dwelling within the sediment
could be elevated. Many benthic invertebrates are detritivores, meaning they feed on dead plant and
animal material or contribute to the liberation of additional nutrient resources by further breaking down
these materials. Detritivorous benthic invertebrates often serve as an important food source for many
juvenile fishery and non-game resident species. In several cases, days of exceedance were greater in
pore water (Table 4-10) and sediment (Table 4-11) than the surface water (Table 4-9), further indicating
that TCEP would be a more persistent hazard to benthic dwelling organisms with increased durations of
exposure.

The VVWM-PSC model identified substantial deposition of TCEP to the sediment (Table 4-11) with a
production volume of 2,500 lb/year. Listed below are the 5 out of 20 COUs (Life cycle stage/ Category/
Sub-category with their respective OES) evaluated, RQs for chronic duration exposures were greater
than or equal to one with more than 14 days of exceedance within both pore water and sediment. A
major concern centered around the RQs within sediment and pore water is the lasting effects on benthic
biota and potential community-level impacts from chronic TCEP exposure within this aquatic
compartment.

Manufacture/ Import/ Import/ Import and Repackaging

Surface Water: Surface water acute RQ values for import and packaging TCEP was less than 1 via both
E-FAST and VVWM-PSC modeling. Both E-FAST and VVWM-PSC models demonstrated chronic
RQs greater than 1; however, no days of exceedance were greater than or equal to 14 days. Specifically,
E-FAST and VVWM-PCS days of exceedance were 2 and 5 days, respectively.

Pore Water: The pore water acute RQ for importing and repackaging TCEP was less than one the acute
COC. The chronic RQ for importing and repackaging TCEP was greater than one for the chronic COC
at 2.47. The corresponding days of exceedance for the chronic COC was 49 days.

Page 127 of 572

-------
3268

3269

3270

3271

3272

3273

3274

3275

3276

3277

3278

3279

3280

3281

3282

3283

3284

3285

3286

3287

3288

3289

3290

3291

3292

3293

3294

3295

3296

3297

3298

3299

3300

3301

3302

3303

3304

3305

3306

3307

3308

3309

3310

3311

3312

3313

3314

3315

3316

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Sediment: The sediment acute RQ for importing and repackaging TCEP was less than one for the acute
COC. The chronic RQ for importing and repackaging TCEP was greater than one for the chronic COC
at 14.29. The corresponding days of exceedance for the chronic COC was 119 days.

Processing/ Incorporated into Formulation, Mixture, or Reaction Product/ Paints and Coating
Manufacturing/ Incorporation into Paints and Coatings - 1-Part Coatings

Surface Water: Surface water acute RQ values for TCEP incorporation into paints and coatings - 1-part
coatings were less than 1 via both E-FAST and VVWM-PSC modeling. Both E-FAST and VVMW-PSC
models demonstrated chronic RQs greater than 1; however, no days of exceedance were greater than or
equal to 14 days. Specifically, E-FAST and VVWM-PCS days of exceedance were 0 and 4 days,
respectively.

Pore Water: The pore water acute RQ for TCEP incorporation into paints and coatings - 1-part coatings
was less than one for the acute COC. The chronic RQ for importing and repackaging TCEP was greater
than one for the chronic COC at 5.44. The corresponding days of exceedance for the chronic COC was
82 days.

Sediment: The sediment acute RQ for TCEP incorporation into paints and coatings - 1-part coatings was
less than one for the acute COC. Chronic RQs for importing and repackaging TCEP was greater than
one for the chronic COC at 31.31. The corresponding days of exceedance for the chronic COC was 145.

Processing/ Incorporated into Formulation, Mixture, or Reaction Product/ Paints and Coating
Manufacturing/ Incorporation into Paints and Coatings - 2-Part Coatings

Surface Water: Surface water acute RQ values for TCEP incorporation into paints and coatings -
resins/solvent-borne were less than 1 via both E-FAST and VVWM-PSC modeling. Both E-FAST and
VVMW-PSC models demonstrated chronic RQs greater than 1; however, no days of exceedance were
greater than or equal to 14 days. Specifically, E-FAST and VVWM-PCS days of exceedance were 0 and
3 days, respectively.

Pore Water: The pore water acute RQ for TCEP incorporation into paints and coatings - resins/solvent-
borne was less than one for the acute COC. The chronic RQ for importing and repackaging TCEP was
greater than one for the chronic COC at 2.49. The corresponding days of exceedance for the chronic
COC was 48 days.

Sediment: The sediment acute RQ for TCEP incorporation into paints and coatings - resins/solvent-
borne was less than one for the acute COC. The chronic RQs for importing and repackaging TCEP was
greater than one for the chronic COC at 14.29. The corresponding days of exceedance for the chronic
COC was 118 days.

Commercial use/ Paints and coatings/ Paints and coatings/ Use in Paints and Coatings at Job Sites

Surface Water: Surface water acute RQ values for TCEP use in paints and coatings at job sites were less
than 1 via both E-FAST and VVWM-PSC modeling. Both E-FAST and VVMW-PSC models
demonstrated chronic RQs greater than 1; however, no days of exceedance were greater than or equal to
14 days. Specifically, E-FAST and VVWM-PCS days of exceedance were 1 and 3 days, respectively.

Pore Water: The pore water acute RQ for TCEP use in paints and coatings at job sites was less than one
for the acute COC. The chronic RQs for paints and coatings at job sites was greater than one for the
chronic COC at 2.95. The corresponding days of exceedance for the chronic COC was 56 days.

Page 128 of 572

-------
3317

3318

3319

3320

3321

3322

3323

3324

3325

3326

3327

3328

3329

3330

3331

3332

3333

3334

3335

3336

3337

3338

3339

3340

3341

3342

3343

3344

3345

3346

3347

3348

3349

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Sediment: The sediment acute RQ for TCEP use in paints and coatings at job sites was less than one for
the acute COC. The chronic RQ for paints and coatings at job sites was greater than one for the chronic
COC at 17.01. The corresponding days of exceedance for the chronic COC was 125 days.

Processing/Incorporated into Formulation, Mixture, or Reaction Product/ Polymers Used in
Aerospace Equipment and Products/Formulation of TCEP into 2-Part Reactive Resins

Surface Water: Surface water acute RQ values for formulation of TCEP into 2-part reactive resins were
less than 1 via both E-FAST and VVWM-PSC modeling. Both E-FAST and VVMW-PSC models
demonstrated chronic RQs greater than 1, however, no days of exceedance were greater than or equal to
14 days. Specifically, E-FAST and VVWM-PCS days of exceedance were 1 and 3 days, respectively.

Pore Water: The pore water acute RQ for formulation of TCEP into 2-part reactive resins was less than
one for the acute COC. The chronic RQ for 2-part reactive resins was greater than one for the chronic
COC at 2.90. The corresponding days of exceedance for the chronic COC was 55 days.

Sediment: The sediment acute RQs for formulation of TCEP into 2-part reactive resins were less than
one for both the acute COC and secondary acute COC. Chronic RQs for 2-part reactive resins were both
greater than one for the chronic COC and secondary chronic COC at 16.74 and 167.44, respectively. The
corresponding days of exceedance for the chronic COC and secondary chronic COC were 124 and 190
days.

Commercial Use/Laboratory Chemicals/ Laboratory Chemicals/ Laboratory Chemicals

Surface Water: Within the water column, acute RQ values for laboratory chemicals were less than 1 via
both E-FAST and VVMM-PSC modeling. VVMW-PSC modeling demonstrated a chronic RQ of 1.74
with days of exceedance of 179.

Pore Water: The pore water acute RQs for laboratory chemicals was less than one for the acute COC.
The chronic RQ for laboratory chemicals was greater than one at 1.18. The corresponding days of
exceedance for the chronic COC was 84 days.

Sediment: The sediment acute RQ for laboratory chemicals was less than one for the acute COC. The
chronic RQ for laboratory chemicals was greater than one for the chronic COC at 6.80. The
corresponding days of exceedance for the chronic COC was 209 days.

Page 129 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

3350	Table 4-9. Environmental Risk Quotients (RQs) by COU with Production Volumes of 2,500 lb/year for Aquatic Organisms with

3351	TCEP Surface Water Concentration (ppb) Modeled by VVWM-PSC 		











Modeled Using VVWM-PSC

COU (Life Cycle
Stage/Category/Sub-category)

Occupational
Exposure Scenario

Production
Volume
(lb/year)"

Days of
Release

Release
(kg/day)

Max Day
Average
(ppb)"

COC
Type

COC
(ppb)

Days of
Exceedance
(days per
year)

RQ

Manufacture/ Import/ Import

Import and

2,500

4

9.88

2,390

Acute

85,000

N/A

0.03

repackaging

683

Chronic

55.9

5

12.22

Processing/ Incorporated into
formulation, mixture, or reaction
product/ Paint and coating
manufacturing

Incorporation into







10,200

Acute

85,000

N/A

0.12

paints and coatings -
1-part coatings

2,500

2

35.17

1,480

Chronic

55.9

4

26.48

Processing/ Incorporated into
formulation, mixture, or reaction

Incorporation into
paints and coatings -

2,500

1

31.89

8,280

Acute

85,000

N/A

0.10

product/ Paint and coating

2-part reactive

673

Chronic

55.9

3

12.04

manufacturing

coatings

















Commercial use/ Paints and

Use in paints and

2,500

2

23.25

5,590

Acute

85,000

NA

0.07

coatings/ Paints and coatings

coatings at job sites

804

Chronic

55.9

3

14.38

Processing/ Incorporated into
formulation, mixture, or reaction
product/ Polymers used in aerospace
equipment and products

Formulation of







9,190

Acute

85,000

N/A

0.11

TCEP into 2-part
reactive resins

2,500

1

31.53

789

Chronic

55.9

3

14.11

Commercial use/ Laboratory
chemicals/ Laboratory chemicals

Laboratory
chemicals

2,500

182

0.39

96

Acute

85,000

N/A

1.13E

-03







97

Chronic

55.9

179

1.74

a Production volume of 2,500 lb TCEP/year uses high-end estimates (95th percentile for all COUs except the laboratory chemicals COU uses the 1st percentile)
b Max day average represents the maximum concentration over a 1- or 14-day average period corresponding with the acute or chronic COC used for the RQ

estimate



















c VVWM-PSC model input parameter for KOC utilized the 5th percentile (2.13) of the mean (2.82)

N/A = Days of exceedance are modeled for the application of chronic COCs and do not apply for acute COCs and corresponding RQs





3352

3353

3354

Page 130 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

3355	Table 4-10. Environmental Risk Quotients (RQs) by COU with Production Volumes of 2,500 lb/year for Aquatic Organisms with

3356	TCEP Pore Water Concentration (ppb) Modeled by VVWM-PSC 			

COU (Life Cycle
Stage/Category/Sub-
category)

Occupational
Exposure Scenario

Production
Volume
(lb/year)"

Days of
Release

Release
(kg/day)

Benthic Pore

Water
Concentration
(ppb)"

Benthic Pore Water'

COC
Type

COC
(ppb)

Days of
Exceedance

RQ

Manufacture/ Import/ Import

Import and
repackaging

2,500

4

9.88

154

Acute

85,000

N/A

1.82E-03

138

Chronic

55.9

49

2.47

Processing/ Incorporated into
formulation, mixture, or
reaction product/ Paint and
coating manufacturing

Incorporation into
paints and coatings -
1-part coatings

2,500

2

35.17

339

Acute

85,000

N/A

3.99E-03

304

Chronic

55.9

82

5.44

Processing/ Incorporated into
formulation, mixture, or
reaction product/ Paint and
coating manufacturing

Incorporation into
paints and coatings -
2-part reactive
coatings

2,500

1

31.89

155

Acute

85,000

N/A

1.82E-03

139

Chronic

55.9

48

2.49

Commercial use/ Paints and
coatings/ Paints and coatings

Use in paints and
coatings at job sites

2,500

2

23.25

185

Acute

85,000

N/A

2.18E-03

165

Chronic

55.9

56

2.95

Processing/ Incorporated into
formulation, mixture, or
reaction product/ Polymers
used in aerospace equipment
and products

Formulation of
TCEP into 2-part
reactive resins

2,500

1

31.53

180

Acute

85,000

N/A

2.12E-03

162

Chronic

55.9

55

2.90

Commercial use/ Laboratory
chemicals/ Laboratory
chemicals

Laboratory
chemicals

2,500

182

0.39

66

Acute

85,000

N/A

7.76E-04

66

Chronic

55.9

84

1.18

a Production volume of 2,500 lb TCEP/year uses high-enc

estimates (95th percentile for all COUs except the laboratory chemicals COU uses the 1st percentile)

Max day average represents the maximum concentration over a 1- or 14-day average period corresponding with the acute or chronic COC used for the RQ

estimate



















c WWM-PSC model input parameter for Koc utilized the 5th percentile (2.13) of the mean (2.82)









N/A = Days of Exceedance are modeled for the application of chronic COCs and do not apply for acute COCs and corresponding RQs



3357

3358

Page 131 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

3359	Table 4-11. Environmental Risk Quotients (RQs) by COU with Production Volumes of 2,500 lb/year for Aquatic Organisms with

3360	TCEP Sediment Concentration (ppb) Modeled by VVWM-PSC				

COU (Life Cycle
Stage/Category/Sub-category)

Occupational
Exposure Scenario

Production
Volume
(lb/year)"

Days of
Release

Release
(kg/day)

Sediment
Concentration
(ppb)"

Sediment'

COC
Type

COC
(ppb)

Days of
Exceedance

RQ

Manufacture/ Import/ Import

Import and
repackaging

2,500

4

9.88

894

Acute

85,000

N/A

0.01

799

Chronic

55.9

119

14.29

Processing/ Incorporated into
formulation, mixture, or reaction
product/ Paint and coating
manufacturing

Incorporation into
paints and coatings -
1-part coatings

2,500

2

35.17

1,960

Acute

85,000

N/A

0.02

1,750

Chronic

55.9

145

31.31

Processing/ Incorporated into
formulation, mixture, or reaction
product/ Paint and coating
manufacturing

Incorporation into
paints and coatings -
2-part reactive
coatings

2,500

1

31.89

893

Acute

85,000

N/A

0.01

799

Chronic

55.9

118

14.29

Commercial use/ Paints and
coatings/ Paints and coatings

Use in paints and
coatings at job sites

2,500

2

23.25

1,070

Acute

85,000

N/A

0.01

951

Chronic

55.9

125

17.01

Processing/ Incorporated into
formulation, mixture, or reaction
product/ Polymers used in
aerospace equipment and
products

Formulation of TCEP
into 2-part reactive
resins

2,500

1

31.53

1,040

Acute

85,000

N/A

0.01

936

Chronic

55.9

124

16.74

Commercial use/ Laboratory
chemicals/ Laboratory chemicals

Laboratory chemicals

2,500

182

0.39

380

Acute

85,000

N/A

0.01

380

Chronic

55.9

209

6.80

a Production volume of 2,500 lb TCEP/year uses high-end estimates (95th percentile for all COUs except the laboratory chemicals COU uses the 1st
percentile)

h Max day average represents the maximum concentration over a 1- or 14-day average period corresponding with the acute or chronic COC used for the RQ
estimate

c WWM-PSC model input parameter for Koc utilized the 5th percentile (2.13) of the mean (2.82)

N/A = Days of exceedance are modeled for the application of chronic COCs and do not apply for acute COCs and corresponding RQs

3361

Page 132 of 572

-------
3362

3363

3364

3365

3366

3367

3368

3369

3370

3371

3372

3373

3374

3375

3376

3377

3378

3379

3380

3381

3382

3383

3384

3385

3386

3387

3388

3389

3390

3391

3392

3393

3394

3395

3396

3397

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

EPA used surface water monitoring data from the WQP and published literature to characterize the risk
of TCEP to aquatic organisms. These monitored surface water data reflect concentrations of TCEP in
ambient water. WQP data show an average (± SEM) concentration for TCEP of 0.33 ± 0.02 ppb in
surface water from 466 measurements taken throughout the United States between 2003 and 2022. The
highest concentration recorded during this period was 7.66 ppb, which was recorded in August 2013 in
Rochester, New York. Table 4-12 shows that RQ estimates were less than 1 for both acute and chronic
COCs. There are no sediment samples above the detection limit for TCEP in the WQP.

Table 4-12. Risk Quotients (RQs) Calculated Using Monitored Environmental Concentrations
from WQXAVQP	

Monitored Surface Water Concentrations
(ppb) from 2003-2022

RQ Using Acute COC of
85,000 ppb

RQ Using Chronic COC of
55.9 ppb

Mean (Standard Error of the Mean):
0.33 (0.02) ppb

3.88E-05

5.9E-03

Maximum: 7.66 ppb

9.01E-05

0.13

Five of the six studies from reasonably available published literature sampled waters within the United
States, while one included sample sites from both U.S. and Canadian waters (Scott et at.. 1996). All six
studies from published literature are represented by general population surface water sampling where
TCEP concentration are not associated with a specific facility. One study encompassed 85 sample sites
for TCEP with study design placing sampling directly downstream from "intense urbanization and
livestock production, detecting TCEP within 49 of the 85 samples and resulting in minimum and
maximum TCEP concentrations of 0.02 and 0.54 ppb, respectively" (Kolpin et at.. 2002). Across all
studies a total of 185 samples resulted in 141 samples with TCEP detected and 44 non-detected
samplings between 1994 and 2013. The mean (±SEM) for TCEP concentrations reported within surface
water in the reasonably available published literature is 0.16 (±0.05) ppb with minimum and maximum
concentrations of 0.0002 and 0.81 ppb, respectively.

Table 4-13 shows RQs estimates close to zero for both acute and chronic COCs.

Table 4-13. Risk Quotients (RQs) Calculated Using TCEP in Surface Water from Monitored
Environmental Concentrations from Published Literature

Monitored Surface Water Concentrations
(ppb) from Published Literature

RQ Using Acute COC of
85,000 ppb

RQ Using Chronic COC of
55.9 ppb

Mean (Standard Error of the Mean):
0.16 (0.05) ppb

1.8E-06

2.8E-03

Maximum: 0.81 ppb

9.5E-06

1.4E-02

Two studies representing TCEP sediment concentrations from the United States and another conducted
within Germany and the Czech Republic were presented within the reasonably available literature. The
study conducted in the United States sampled sediment within coastal embayments in southern
California and the Santa Clara River Watershed (Maruva et at.. 2016). The mean sediment TCEP
concentration was 2.2 |ig/kg and 90th percentile of the mean of 4.0 ppb with maximum TCEP
concentrations in sediment within coastal embayments and the Santa Clara Watershed at 6.98 ppb and
5.08 ppb, respectively (Maruva et at.. 2016). A survey of 37 sample sites along the Elbe River within
Germany and the Czech Republic following a flooding event in 2002 reported a range of TCEP in

Page 133 of 572

-------
3398

3399

3400

3401

3402

3403

3404

3405

3406

3407

3408

3409

3410

3411

3412

3413

3414

3415

3416

3417

3418

3419

3420

3421

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

sediment from less than 1 to 41 ppb and a median concentration of 7.4 ppb (Stachel et al. 2005). RQs
were less than 1 for acute COCs for all mean, median, and maximum TCEP concentrations (Table 4-14).
RQs for TCEP in sediment using the chronic COC were also less than one for all values within these
published studies.

Table 4-14. Risk Quotients (RQs) Calculated Using TCEP Concentrations in Sediment from
Published Literature

Monitored Sediment
Concentrations (ppb) from
Published Literature

RQ Using Acute
COC of 85,000 ppb

RQ Using Chronic
COC of 55.9 ppb

Reference
(Overall Quality
Determination)

Mean: 2.2 ppb

2.58E-05

0.03

(Maruva et al.. )

Maximum: 6.98 ppb

8.21E-05

0.12

(High)

Median: 7.4 ppb

8.70E-05

0.13

(Stachel et al.. 2005)

Maximum: 41 ppb

4.82E-04

0.73

(Medium)

4.3.3 Risk Characterization for Terrestrial Receptors

RQs were less than 1 for all relevant exposure scenarios when using the highest AERMOD predictions
for air deposition to soil at 1,000 m. Table 4-15 presents soil concentration and chronic RQ values from
the exposure scenario with the highest TCEP soil concentrations, indicating RQs below 1 for soil
organisms based on modeling data. The highest soil concentration recorded from AERMOD predictions
is 0.0055 mg/kg based on TCEP use in paints and coatings at job sites at 1,000 m. Soil concentrations
and RQ values for all scenarios, production volumes, and meteorology models are presented within
Table Apx G-8.

Table 4-15. Calculated Risk Quotients (RQs) Based on TCEP Soil Concentrations (mg/kg) as

Calculated Using Mot

eled Data

Occupational
Exposure Scenario

Production
Volume
(lb/year)"

Meteorological
Model6

Soil Concentration
(mg/kg) at 1,000 mc

Chronic RQ (Hazard
Value: 612 mg/kg)

Use in paints and
coatings at job sites

2,500

MetCT

3.97E-03

6.49E-06

MetHIGH

5.58E-03

9.11E-06

a Production volume of 2,500 lb TCEP/yr uses high-end estimates (95th percentile)

b The ambient air modeled concentrations and deposition values are presented for two meteorology conditions
(Sioux Falls, South Dakota, for central tendency meteorology; and Lake Charles, Louisiana, for higher-end
meteorology)

c Estimated concentrations of TCEP (90th percentile) that could be in soil via air deposition at a community (1,000
m from the source) exposure scenario

Risk characterization and trophic transfer for terrestrial receptors is based on modeled soil data from
AERMOD since there are no published literature or monitoring databases with TCEP soil concentrations
from U.S. sites and one comparative study from Germany (Mihailovic and Fries. ). Transient
increases in TCEP concentration have been observed with mean concentrations elevated from 0.008 to
0.023 mg/kg immediately following snowmelt conditions (Mihailovic and Fries. ). RQs to soil

Page 134 of 572

-------
3422

3423

3424

3425

3426

3427

3428

3429

3430

3431

3432

3433

3434

3435

3436

3437

3438

3439

3440

3441

3442

3443

3444

3445

3446

3447

3448

3449

3450

3451

3452

3453

3454

3455

3456

3457

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

invertebrates were below 1 for soil TCEP concentrations as reported for different sample periods from
Mihailovic a s (2012) (Table 4-16).

Table 4-16. Risk Quotients (RQs) Calculated Using TCEP Soil Concentrations from Published
Literature

Sample Collection
Conditions

Mean TCEP
Concentration in Soil
(mg/kg)

Chronic RQ (Hazard
Value: 612 mg/kg)

Reference
(Overall Quality
Determination)

Soil TCEP concentrations in

5.89E-03

9.62E-06



January





(Mihailovic and
11 t^s. 2012)
(High)

Soil TCEP concentration prior
to snowmelt

7.67E-03

1.25E-05

Soil TCEP concentration 24

2.34E10-02

3.76E-05

hours after snowmelt







4,3,4 Risk Characterization Based on Trophic Transfer in the Environment

Trophic transfer of TCEP and potential risk to terrestrial animals was evaluated using a screening level
approach conducted as described in the EPA's Guidance for Developing Ecological Soil Screening
Levels (	>a). TCEP concentrations within biota and resulting RQ values for all six relevant

COUs represented by seven OESs (Table 4-7), two production volume scenarios (2,500 and 25,000
lb/year), and two meteorological models for soil deposition are presented in
Table Apx G-9. Table 4-17 presents biota concentrations and RQ values for the highest soil
concentration via AERMOD (Paints and coatings at job sites) at the 2,500 production volumes. RQs
were below 1 for all soil concentrations and COUs based on the chronic hazard threshold for terrestrial
invertebrate identified within Section 4.2.4.3. The chronic TRV, calculated using empirical toxicity data
with mice and rats, also resulted in RQs less than 1 for all modeled soil concentrations. The overall
hazard confidence for the chronic mammalian assessment and terrestrial invertebrates reported within
Section 4.2.6 as robust and moderate, respectively, providing increased confidence in the application of
these ecologically relevant hazard thresholds.

Estimates of risk represented as RQ values were calculated using hazard thresholds with in vivo data
measuring ecologically relevant endpoints such as mortality, reproduction, or growth. These RQ values
are all below 1 for all species and corresponding trophic levels represented (Table 4-17). The earthworm
and American kestrel are important tools in this screening-level trophic transfer analysis as they
represent an animal with direct ingestion of soil {i.e., the earthworm) and as a top avian predator {i.e.,
the kestrel). Hazard values representing effects at the sub-organ level were identified for the earthworm
(alterations in gastrointestinal tract) and American kestrel (alterations in plasma thyroid hormone
levels). TCEP in biota calculated for the earthworm and American kestrel are at doses of 0.0055 and
0.0016 mg/kg/day, respectively, for the highest modeled soil TCEP concentration with a production
volume of 2,500 lb/year. They did not equal or exceed these species hazard thresholds described within
Section 4.2.4.3. The hazard value for the American kestrel (doses of 0.0025 mg/kg/day) did not result in
any detectable impacts to ecologically relevant endpoints of body weight or food consumption from this
21-day dietary exposure study with TCEP (Fernie et at.. 2015). One COU {i.e., Use in paints and
coatings at job sites) at the 25,000 lb/year production volume resulted in TCEP concentrations of 0.025
mg/kg/day; however, this production volume is believed to be an overestimate of current production
volumes in the United States (see Section 1.1.1). In addition, the screening-level analysis used equation

Page 135 of 572

-------
3458

3459

3460

3461

3462

3463

3464

3465

3466

3467

3468

3469

3470

3471

3472

3473

3474

3475

3476

3477

3478

3479

3480

3481

3482

3483

3484

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

terms (e.g., area use factor and the proportion of TCEP absorbed from prey and soil) all set to the most
conservative values further emphasizing a cautious approach to risk to TCEP via trophic transfer.

Table 4-17. Risk Quotients (RQs) for Screening Level Trophic Transfer of TCEP in Terrestrial
Ecosystems Using EPA's Wildlife Risk Model for Eco-SSLs a		

Organism

TCEP Concentration
in Biota

(mg/kg/day)''

Hazard
Threshold
(mg/kg-bw/day)

Reference for Hazard
Value or TRV
(Overall Quality
Determination)

RQ

Nematode

(Caenorhabditis
elegans)

0.0055

612

( et al.„ 2017) (Hiah)

9.0E-06

Mammal

0.004

44

N/Ac

9.8E-05

Short-tailed shrew
(Blarina
brevicauda)

0.004

0.66

N/Ac

0.007

Woodcock

(,Scolopax minor)

0.005

N/A

N/Ad

N/A

a Calculated using highest modeled soil TCEP concentrations with a production volume of 2,500 lb/year (0.0055
mg/kg); see also Equation 4-1.

b TCEP concentration represents the highest modeled soil concentration via AERMOD modeling with a production
volume of 2,500 lb/year.

c Mammal TCEP TRV value calculated using several studies as per (U.S. EPA, 2007a).

'' No TCEP hazard threshold value for this representative species is available.

There are no reported studies within the pool of reasonably available published literature that quantify
TCEP soil concentrations in the United States. A study with an overall quality determination of high
monitored TCEP soil concentrations in the summer (August) and winter (January and February) months
in Germany (Mihailovic and Fries. 2012). The soil collection site was characterized as being located
approximately 3 km from the city center of Osnabrueck and about 20 m from buildings constructed of
reinforced concrete with facades predominately comprised of glass. Biota concentrations and RQ values
were calculated using the same assumptions as described previously in Table 4-8, utilizing the highest
TCEP soil concentration reported in Mihailovic and Fries . Note that this study should be
considered to represent TCEP concentrations in soil from an ambient urban environment and is not
directly comparable to scenarios detailed within the current draft risk evaluation. In a related study at the
same site, the authors postulated that TCEP concentrations resulted from atmospheric deposition and
potentially from cars, and emphasizing the importance of considering atmospheric deposition of
chlorinated organophosphate esters (e.g., TCEP) in future risk assessments (Mihailovic et ai. 2011). The
RQs are below 1 for all species and corresponding trophic level represented (Table 4-18). TCEP
concentrations in biota calculated for the earthworm and American kestrel were 5.89x 10 3 and
1.70xl0~3 mg/kg/day, respectively, and do not equal or exceed these species hazard thresholds described
in Section 4.2.4.3.

Page 136 of 572

-------
3485

3486

3487

3488

3489

3490

3491

3492

3493

3494

3495

3496

3497

3498

3499

3500

3501

3502

3503

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table 4-18. Risk Quotients (RQs) Calculated with Highest Mean TCEP Soil Concentration
(5.89E-03 mg/kg) from Monitored Values in Published Literature for Screening Level Trophic

Transfer of TCEP in Terrestrial Ecosystems Using EPA's Wi

dlife Risk Model for Eco-SSLs a

Organism

TCEP
Concentration in
Biota (mg/kg/day)''

Hazard
Threshold
(mg/kg-bw/day)

Reference for Hazard
Value or TRV
(Overall Quality
Determination)

RQ

Nematode

(Caenorhabditis elegans)

5.89E-03

612

(Xu et ah. 2017) (Hiah)

9.6E-06

Mammal

4.60E-03

44

N/Ac

1.0E-04

Short-tailed shrew
(Blarina brevicauda)

4.60E-03

0.66

N/Ac

6.9E-03

Woodcock

(Scolopax minor)

5.70E-03

N/A

N/Ac

N/A

a As reported in (Mihailovic and Fries, 2012); see also Equation 4-1.

h TCEP concentration represents the highest mean recorded soil concentration (5.89E-03 mg/kg) as reported in
(Mihailovic and Fries, 2012).

c Mammal TCEP TRV value calculated using several studies as detailed in (U.S. EPA. 2007a).

J No TCEP hazard threshold value for this representative species is available.

RQs were below 1 for semi-aquatic terrestrial receptors via trophic transfer from fish and the highest
modeled TCEP surface water concentrations (Table 4-19). RQ and biota concentration values for all
COUs are presented within Table Apx G-10. The hazard confidence for the chronic mammalian
assessment was reported as robust within Section 4.2.6 and BCF values used to approximate TCEP
concentrations within fish were from a high-quality study (Anikwe et at.. 2018). The modeled TCEP
concentrations within this analysis are five orders of magnitude greater than surface water
concentrations identified from the WQP database and the published literature (Table 4-12 and Table
4-13). These results align with previous risk assessments that concluded that TCEP is not viewed as a
bioaccumulative compound ( v i i \ ,, U 2009; ECB. 20091

Table 4-19. Selected Risk Quotients (RQs) (Highest Fish TCEP Concentrations) Based on
Potential Trophic Transfer of TCEP from Fish to American Mink (Mustela vison) as a Model

Aquatic Predator

Jsing EPA's Wildlife Risk

Model for Eco-SSLs a

Occupational
Exposure
Scenario

Production
Volume
(lb/year)

Release
Distribution

SWCa
(ppb)

Fish
Concentration
(mg/kg)

American Mink

(Mustela vison)

TCEP in Biota
(mg/kg/day)

RQ

Formulation of
TCEP Containing
Reactive Resin

2,500

High-End

10,900

3.71

2.34

0.08

a See also Equation 4-1

b TCEP Surface Water Concentration (SWC) calculated using VVWM-PSC

Page 137 of 572

-------
3504

3505

3506

3507

3508

3509

3510

3511

3512

3513

3514

3515

3516

3517

3518

3519

3520

3521

3522

3523

3524

3525

3526

3527

3528

3529

3530

3531

3532

3533

3534

3535

3536

3537

3538

3539

3540

3541

3542

3543

3544

3545

3546

3547

3548

3549

3550

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

4,3,5 Connections and Relevant Pathways from Exposure Media to Receptors

4.3.5.1	Aquatic Receptors

Surface Water, Benthic Porewater, and Sediment

Within the aquatic environment, a two-tiered modeling approach was employed to predict surface water,
pore water, and sediment TCEP concentrations. If the E-FAST predicted 7Q10 surface water
concentrations were greater than the chronic or acute COCs, the VVWM-PSC model was then used to
confirm whether the predicted surface water concentration days of exceedance as determined by the
acute COC and chronic COC. For TCEP, all five applicable COUs (Table 4-7) modeled in E-FAST
produced chronic RQ values greater or equal to 1, prompting the use of VVWM-PSC for greater
ecological resolution on TCEP concentrations and days of exceedance within the water column and
benthic compartments (see Section 4.3.1).

Air Deposition to Water and Sediment

EPA used IIOAC and AERMOD to estimate air deposition from hypothetical facility releases and to
calculate pond water and sediment concentrations 1,000 m from the hypothetical facility. Pond water
concentrations from air deposition were estimated for the COUs with air releases (Table 4-7). The
highest estimated 95th percentile pond water concentration from annual deposition, across all exposure
scenarios, was 8.1 ppb for the Commercial use of paints and coatings scenario at an annual production
volume of 2,500 lb per year. This highest modeled concentration within a pond at 1,000 m from a point
source was approximately 150 times lower than the lowest surface water concentration modeled using
VVWM-PSC (1,270 ppb as a maximum 1-day average concentration for the Laboratory chemicals
scenario at an annual production volume of 2,500 lb per year). Air deposition to sediment as reported in
Section 3.3.2.10 indicated the highest annual deposition at 1,000 m was 125 ppb, which is about seven
times lower than the lowest sediment TCEP value modeled with VVWM-PSC (Incorporation into paints
and coatings - solvent borne at 893 ppb) and about 40 times lower than the highest PSC value for
laboratory chemicals (5,040 ppb). Using VVWM-PSC, sediment concentrations from aquatic releases of
TCEP ranged from 893 ppb to 5,040 ppb for the production volume of 2,500 lb/year, respectively, and
represent a significant driver of TCEP deposition to sediment within flowing water systems. Although
the IIO AC and AERMOD were applied to a generic farm pond setting to calculate concentrations of
TCEP in pond surface water and pond sediment, these models do not account for media exchange of the
chemical of interest as is the case for VVWM-PSC. In addition, it is not anticipated that air deposition to
water will significantly contribute as TCEP concentrations within the water column, pore water, and
sediment will utilize modeling via E-FAST and VVWM-PSC.

TCEP Runoff from Biosolids

Due to its persistence, it is likely that dissolved TCEP will eventually reach surface water via runoff
after the land application of biosolids. A review of reasonably available literature indicates that modeled
surface water, pore water, and sediment concentrations are approximately half the highest concentrations
and approximately 50 times greater than the mean values biosolid concentrations reported in Wane et al.
(2019c). Direct exposure of TCEP to aquatic receptors via biosolids was not assessed quantitatively (see
Section 3.3.3).

4.3.5.2	Terrestrial Receptors
Inhalation by Wildlife

Direct exposure of TCEP to terrestrial receptors via air was not assessed quantitatively because dietary
exposure was determined to be the driver of exposure to wildlife. The contribution of exposure risk from
inhalation relative to the ingestion exposure route is not expected to drive risk because of dilution
associated environmental conditions and the deposition of TCEP from air to soil (	s03a. b).

Page 138 of 572

-------
3551

3552

3553

3554

3555

3556

3557

3558

3559

3560

3561

3562

3563

3564

3565

3566

3567

3568

3569

3570

3571

3572

3573

3574

3575

3576

3577

3578

3579

3580

3581

3582

3583

3584

3585

3586

3587

3588

3589

3590

3591

3592

3593

3594

3595

3596

3597

3598

3599

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

The contribution of exposure risk from inhalation relative to the ingestion exposure route is not expected
to drive risk because of dilution associated environmental conditions and the deposition of TCEP from
air to soil (	003a. b). AERMOD results indicate a maximum ambient air concentration (95th

percentile, MetHIGH) of 6,08/10 7 [j,g/m3 at 1,000 m from a hypothetical facility for the Use of paints
and coatings - spray application OES under the 2,500 lb/year production volume using the Suburban
forest land category scenario (see Section 3.3.1.2). AERMOD results for the same conditions and COU
for air deposition to soil indicate a TCEP concentration of 5.58 (J,g/kg at 1,000 m from a hypothetical
facility (Table Apx G-8). In addition, TCEP is not persistent in air due to short half-life in the
atmosphere (ti/2 = 5.8 hours) (I v << \	) and because particle-bound TCEP is primarily removed

from the atmosphere by wet or dry deposition (see Section 4.1.3.2).

Biosolids

TCEP is released to the environment by various exposure pathways (Figure 2-1). The exposure pathway
for terrestrial organisms is through soil. Deposition of TCEP from air to soil is the primary exposure
pathway. A secondary source of TCEP contamination in soil is from the application of biosolids.
However, the maximum modeled concentration of TCEP in soil from biosolids (2.32><10~4 mg/kg for
pastureland) is two orders of magnitude less than the maximum modeled TCEP soil concentration from
air deposition 8.65><10~2 mg/kg (see Section 3.3). Therefore, biosolid application is not expected to have
an impact on the terrestrial risk assessment (see Section 4.1.4).

Air Deposition to Soil

As described in Section 3.3.3.2, EPA Modeled Soil Concentrations via Air Deposition (AERMOD),
IIOAC and subsequently AERMOD were used to assess the estimated release of TCEP via air
deposition from specific exposure scenarios to soil (Table 4-7). Estimated concentrations of TCEP that
could be deposited in soil via air deposition at the community level (1,000 m from the source) exposure
scenarios have been calculated (see Section 4.3.1).

Soil in Diet

Following the basic equations as reported within Chapter 4 of EPA's Guidance for Developing
Ecological Soil Screening Levels, wildlife receptors may be exposed to contaminants in soil by two main
pathways: incidental ingestion of soil while feeding, and ingestion of food items that have become
contaminated due to uptake from soil (	»05a). Within this model, incidental oral soil

exposure is added to the dietary exposure resulting in total oral exposure greater than 100 percent (see
Section 4.1.4).

Surface Water Ingestion in Wildlife

Because surface water sources for wildlife water ingestion are typically ephemeral, the trophic transfer
analysis for terrestrial organisms assumed TCEP exposure concentration for wildlife water intake are
equal to soil concentrations for each corresponding exposure scenario (see Section 4.1.4).

For semi-aquatic terrestrial species, the TRV was used with the American mink for the screening level
assessment (Table 4-8). Similar to the soil concentrations used as term Soil; in Equation 4-3, the highest
surface water concentration modeled via VVWM-PSC was used as a surrogate for the TCEP
concentration found in the American mink's diet (see Section 4.3.1.1).

Semi-aquatic Wildlife

The American mink was used as the representative species for semi-aquatic mammals. As a
conservative assumption, 100 percent of the American mink's diet is predicted to come from fish. Fish
concentration (mg/kg) was calculated using surface water concentrations of TCEP from VVWM-PSC

Page 139 of 572

-------
3600

3601

3602

3603

3604

3605

3606

3607

3608

3609

3610

3611

3612

3613

3614

3615

3616

3617

3618

3619

3620

3621

3622

3623

3624

3625

3626

3627

3628

3629

3630

3631

3632

3633

3634

3635

3636

3637

3638

3639

3640

3641

3642

3643

3644

3645

3646

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

assuming a BCF of 0.34 as reported for whole body values from 1 mg/L TCEP exposures under
laboratory conditions ( kwe et al. 2018). The conservative approach for calculated fish tissue
concentrations presented in Section 4.1.2.2 was utilized for trophic transfer analysis to semi-aquatic
mammals (see Section 4.3.1.10).

4,3,6 Summary of Environmental Risk Characterization

4.3.6.1 COUs with Quantified Release Estimates

EPA had uncertainty in the production volume and hazard value for sediment dwelling species;
however, even at the realistic production volume of 2,500 lb/year, EPA found chronic RQs above 1 with
more than 14 days of exceedance for aquatic receptors in the sediment compartment using both COCs
that help bound uncertainties in the hazard. Additionally, because of the physical-chemical and fate
properties, EPA expects TCEP to partition between water and sediment and be persistent within the
sediment compartment. Therefore, EPA has moderate confidence that there is risk to aquatic organisms
in the sediment compartment for 5 out of 20 COUs.

The current environmental risk characterization on TCEP utilizes two alternate production volume
assumptions for the calculation of RQ values. The 25,000 lb/year production volume is used as the high-
end estimation. It is based on the reporting threshold for TCEP in CDR; however, given EPA's research,
this is believed to be an overestimate of current production volumes in the United States. Therefore, the
2,500 lb production volume is reflective of estimated current production volumes. In the current section,
the analyses using 2,500 lb/year production volume are presented. Table 4-20 and Table 4-21 present
RQ values for exposure scenarios with a production volume of 2,500 lb/year and corresponding
environmental risk for aquatic and terrestrial receptors, respectively. Exposure data and corresponding
RQ values produced with a production volume of 25,000 lb/year are presented within the Appendix G.

Within the aquatic environment, chronic RQs for aquatic receptors from TCEP exposure are elevated
above one and have corresponding days of exceedance greater than 14 days within pore water and
sediment compartments of benthic environment based on the affinity and persistence of this compound.
EPA calculated risks to sediment organisms using two hazard thresholds (or COCs)—one representing a
more conservative threshold and the other a less conservative threshold that were referred to as
secondary acute COC and secondary chronic COC. Risk was consistently identified within sediment and
pore water using both COCs, which gives EPA more confidence the use of the COCs for RQ values
presented throughout Section 4.3.2. Secondary COCs represent the acute COC and chronic COC with
the application of additional assessment factors (Table 4-4); however, overall hazard confidence was
determined to be "slight." The overall hazard confidence for acute COC and chronic COC were both
rated as "moderate" (Table 4-6) with overall confidence in the RQ inputs also as "moderate" (Table
4-23). Acute and chronic COCs with "moderate" hazard confidence represent RQs within the current
summary section as the corresponding confidence in risk characterization RQ inputs were also rated as
"moderate" (Table 4-23).

Exposure concentrations were modeled based on COU related releases to the aquatic environment and
are represented by TCEP values within surface water, pore water, and sediment. Confidence in aquatic
exposure estimates is "moderate" with modeling parameters considering inputs from COUs and physical
and chemical and fate parameters specific to TCEP. Surface water monitoring data were available from
the WQP database and published literature, while monitoring data for TCEP in sediment was available
from published literature. Table 4-20 displays RQ estimates for all exposure scenarios with a production
volume of 2,500 lb/year in surface water TCEP concentrations modeled via VVWM-PSC modeling. For
TCEP modeled in surface water, one COU (Laboratory chemicals) had a chronic RQ greater than or

Page 140 of 572

-------
3647

3648

3649

3650

3651

3652

3653

3654

3655

3656

3657

3658

3659

3660

3661

3662

3663

3664

3665

3666

3667

3668

3669

3670

3671

3672

3673

3674

3675

3676

3677

3678

3679

3680

3681

3682

3683

3684

3685

3686

3687

3688

3689

3690

3691

3692

3693

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

equal to one and greater than 14 days of exceedance. The COU for laboratory chemicals resulted in
surface water concentrations 1.7 times above the chronic COC with 179 days of exceedance. The
Laboratory chemicals COU is characterized by greater days of released compared to other COUs with
quantified surface water releases, indicated by the exceedance of the chronic COC duration. For other
COUs with modeled TCEP concentrations for surface water, RQs using the chronic COC resulted in
values also greater than one; however, the days of exceedance were well below the days of exceedance
represented for chronic risk. All relevant TCEP exposure concentration values for both E-FAST and
VVWM-PSC results for modeled surface water concentrations are provided in Table 4-9. The overall
exposure confidence for acute and chronic aquatic assessment were both rated as "moderate" (Table
4-23) with the inclusion of physical and chemical parameters represented within models performed with
VVWM-PSC. No RQs over 1 were identified from TCEP surface water concentrations within the WQP
database or published literature (Table 4-12).

No acute RQs were greater than 1 for modeled surface water TCEP at 2,500 lb/year production volume
via both E-FAST and VVMW-PSC modeling.

Chronic RQs were not greater than 1 and days of exceedance were less than 14 days for surface water
TCEP modeled via VVWM-PSC at the 2,500 lb/year production volume for 4 of the 5 relevant COUs
(Life cycle stage/ Category/ Sub-category/ OES):

•	Manufacturer/ Import/ Import/ Repackaging

•	Processing/ Incorporated into formulation, mixture, or reaction product/ Paint and coating
manufacturing/ Incorporation into paints and coatings - 1-part coatings and 2-part reactive
coatings

•	Commercial use/ Paints and coatings/ Paints and coatings/ Use in paints and coatings at job sites

•	Processing/Incorporated into article/ Aerospace equipment and products/ Processing into 2-part
resin article

The VVWM-PSC model identified substantial deposition of TCEP to the benthic compartment, which
comprises sediment and benthic pore water. Physical and chemical properties including but not limited
to Koc, benthic half-life, and hydrolysis half-life within the VVWM-PSC model, aligns with the
partitioning to organic carbon in sediment (Appendix E.2.3.2) and persistence (Appendix E.2.3.1).

These parameters resulted in modeled data indicating TCEP concentrations residing within pore water
and sediment over longer durations of time (days of exceedance) when compared to results from surface
water concentrations for the chronic COC (55.9 ppb). For pore water, chronic RQs were greater than or
equal to 1 with over 14 days of exceedance for all five relevant COUs (Table 4-20). Days of exceedance
were greater in pore water (Table 4-10) than surface water (Table 4-9), indicating that TCEP will be a
more persistent hazard to benthic dwelling organisms with increased durations of exposure. All relevant
COCs and relevant flow data for VVWM-PSC results for modeled pore water concentrations are
available in Table 4-10. There are no pore water TCEP concentrations reported in the WQP database or
published literature.

No acute RQs were greater than or equal to 1 for modeled pore water TCEP at 2,500 lb/year production
volume via VVMW-PSC modeling.

Chronic RQs were greater than one with over 14 days of exceedance for pore water TCEP modeled via
VVWM-PSC at the 2,500 lb/year production volume for all five relevant COUs (Life cycle stage/
Category/ sub-category/ occupational exposure scenario):

•	Manufacturer/ import/ import/repackaging

Page 141 of 572

-------
3694

3695

3696

3697

3698

3699

3700

3701

3702

3703

3704

3705

3706

3707

3708

3709

3710

3711

3712

3713

3714

3715

3716

3717

3718

3719

3720

3721

3722

3723

3724

3725

3726

3727

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

•	Processing/ incorporated into formulation, mixture, or reaction product/ paint and coating
manufacturing/ incorporation into paints and coatings - 1-part coatings and 2-part reactive
coatings

•	Commercial use/ paints and coatings/ paints and coatings/ use in paints and coatings at job sites

•	processing/ incorporated into article/ aerospace equipment and products/ processing into 2-part
resin article

•	Commercial use/ laboratory chemicals/laboratory chemicals/ lab chemical - use of laboratory
chemicals

For sediment, chronic RQs were greater than 1 and greater than 14 days of exceedance within five
COUs (Table 4-20). As previously stated, concern for these RQs within sediment and pore water is the
lasting effects on benthic biota and potential community-level impacts from chronic TCEP exposure
within this aquatic compartment. Many benthic invertebrates are detritivores, meaning they feed on dead
plant and animal material or contribute to the liberation of additional nutrient resources by further
breaking down these materials. These detritivorous benthic invertebrates often serve as an important
food source for many juvenile fishery and non-game resident species. No RQs over 1 were identified
from TCEP sediment concentrations within published literature (Table 4-14).

No acute RQs were greater than or equal to 1 for modeled sediment TCEP at 2,500 lb/year production
volume via VVMW-PSC modeling.

Chronic RQs were greater than one with over 14 days of exceedance for sediment TCEP modeled via
VVWM-PSC at the 2,500 lb/year production volume for all five relevant COUs (Life cycle stage/
Category/ Sub-category/ Occupational exposure scenario):

•	Manufacturer/ import/ import/ repackaging

•	Processing/ incorporated into formulation, mixture, or reaction product/ paint and coating
manufacturing/ incorporation into paints and coatings - 1-part coatings and 2-part reactive
coatings

•	Commercial use/ paints and coatings/ paints and coatings/ use in paints and coatings at job sites

•	Processing/ incorporated into article/ aerospace equipment and products/ processing into 2-part
resin article

•	Commercial use/ laboratory chemicals/ laboratory chemicals/ lab chemical - use of laboratory
chemicals

Page 142 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

3728	Table 4-20. Exposure Scenarios (Production Volume of 2,500 lb TCEP/year) and Corresponding Environmental Risk for Aquatic Receptors

3729	with TCEP in Surface Water, Sediment, and Pore Water	

cou

Occupational
Exposure
Scenario"

Aquatic Receptors6

Surface Water

Sediment

Pore Water

Life Cycle
Stagc/Catcgo ry

Sub-category

Acute
RQ"

Conf in
Acute
RQ
Inputs''

Chronic
RQ'

DoE®

Conf in
Chronic

RQ
Inputs''

Acute
RQ"

Conf in
Acute
RQ
Inputs''

Chronic
RQf

DoE®

Conf in
Chronic

RQ
Inputs''

Acute
RQ"

Conf in
Acute
RQ
Inputs''

Chronic
RQ'

DoE®

Conf in
Chronic

RQ
Inputs''

Manufacture/
import

Import

Repackaging

0.03

Moderate

12.2

5

Moderate

0.01

Moderate

14.3

119

Moderate

1.8E-03

Moderate

2.5

49

Moderate

Processing/
incorporated
into formulation,
mixture, or
reaction product

Paint and

coating

manufacturing

Incorporation
into paints and
coatings - 1 -part
coatings

0.12

Moderate

26.5

4

Moderate

0.02

Moderate

31.3

145

Moderate

4.0E-03

Moderate

5.4

82

Moderate

Processing/
incorporated
into formulation,
mixture, or
reaction product

Paint and

coating

manufacturing

Incorporation
into paints and
coatings - 2-part
reactive coatings

0.10

Moderate

12.0

3

Moderate

0.01

Moderate

14.3

118

Moderate

1.8E-03

Moderate

2.5

48

Moderate

Processing/
incorporated
into formulation,
mixture, or
reaction product

Polymers
used in
aerospace
equipment
and products

Formulation of
TCEP into 2-
part reactive
resins

0.11

Moderate

14.1

3

Moderate

0.01

Moderate

16.7

124

Moderate

2.1E-03

Moderate

2.9

55

Moderate

Commercial
use/paints and
coatings

Paints and
coatings

Use in paints
and coatings at
job sites

0.07

Moderate

14.4

3

Moderate

0.01

Moderate

17.0

125

Moderate

2.2E-03

Moderate

3.0

56

Moderate

Commercial

use/laboratory

chemicals

Laboratory
chemicals

Lab chemical -
use of laboratory
chemicals

1.1E-03

Moderate

1.74

179

Moderate

0.01

Moderate

6.8

209

Moderate

7.8E-04

Moderate

1.1

84

Moderate

Modeled TCEP concentrations and RQ values for all relevant exposure scenarios are available in Table 4-9, Table 4-10, and Table 4-11.

" Production volume of 2,500 lb TCEP/yr uses high-end estimates (95th percentile for all COUs except the laboratory chemicals COU uses the 1 st percentile).
4 Risk assessed to aquatic receptors based on TCEP releases from wastewater, WQP database, and published literature.
c All exposure values and Days of Exceedance (DoE) modeled using VVWM-PSC.
d Acute Risk Quotient derived using a Concentration of Concern of 85,000 ppb.

e Conf = Confidence. Confidence in Acute Risk Quotient or Chronic Risk Quotient inputs is detailed in Section 4.3.7.2.

^Chronic Risk Quotient derived using a Primary Concentration of Concern of 55.9 ppb.

?Days of Exceedance (DoE) modeled using VVWM-PSC.

Page 143 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

3730

3731

Table 4-21. Exposure Scenarios (Production Volume of 2,500 lb TCEP/year) and Corresponding Environmental Risk for Terrestrial

cou





Terrestrial Receptors'

Life Cycle Stage/Category

Sub-category

Occupational
Exposure Scenario"

Meteroro-
logical Model6

Soil (invertebrates)''

Trophic Transfer (soil)''

Trophic Transfer

(water)2

RQ

Conf. in

Short-Tailed

Conf. in RQ

American

Conf. in









RQ Inputs^

Shrew RQ

Inputs^

Mink RQ

RQ Inputs^

Manufacture/import

Import

Repackaging

MetCT

2.4E-06

Moderate

1.8E-06

Robust

0.02

Robust

MetHI

3.1E-09

2.3E-06

Processing/incorporated into
formulation, mixture, or
reaction product

Paint and coating

Incorporation into
paints and coatings -
1-part coatings

MetCT

5.4E-08

Moderate

4.0E-05

Robust

0.08

Robust

manufacturing

MetHI

9.3E-08

6.8E-05

Processing/incorporated into
formulation, mixture, or
reaction product

Paint and coating

Incorporation into
paints and coatings -

MetCT

1.8E-08

Moderate

1.3E-05

Robust

0.07

Robust

manufacturing

2-part reactive
coatings

MetHI

3.9E-08

2.9E-05

Processing/incorporated into
formulation, mixture, or
reaction product

Polymers used in
aerospace equipment
and products

Formulation of TCEP
into 2-part reactive
resins

MetCT

2.0E-08

Moderate

4.7E-05

Robust

0.08

Robust

MetHI

4.2E-08

4.6E-05

Processing/incorporated into

Aerospace equipment

Processing into 2-part

MetCT

6.4E-08

Moderate

1.5E-05

Robust

NA

Robust

article

and products

resin article

MetHI

6.3E-08

3.1E-05

Commercial Use/paints and

Paints and coatings

Use in paints and

MetCT

6.5E-06

Moderate

0.005

Robust

0.04

Robust

coatings

coatings at job sites

MetHI

9.1E-06

0.007

Commercial Use/laboratory

Laboratory chemicals

Lab chemical - use of

MetCT

7.9E-08

Moderate

5.8E-05

Robust

7.0E-04

Robust

chemicals

laboratory chemicals

MetHI

7.6E-08

5.6E-05

" Production volume of 2,500 lb TCEP/year uses high-end estimates (95th percentile for all COUs except the laboratory chemicals COU uses the 1st percentile).



h The ambient air modeled concentrations and deposition values are presented for two meteorology conditions (MetCT: Sioux Falls, South Dakota, for central tendency
meteorology; and MetHI: Lake Charles, Louisiana, for higher-end meteorology).

c Risk assessed to terrestrial receptors based on TCEP releases as fugitive air and stack air deposition to soil, trophic transfer, and published literature.





d Estimated concentrations of TCEP (90th percentile) that could be in soil via air deposition at a community (1,000 m from the source) exposure scenario.



e Fish concentration (mg/kg) was calculated using surface water concentrations of TCEP from WWM-PSC assuming a BCF of 0.34 as reported for whole body values from 1
mg/L TCEP exposures under laboratory conditions (Arukwe et al. 2018).

f Conf = Confidence; Confidence in Risk Quotient (RQ) inputs are detailed in Section 4.3.7.2.













3732

Page 144 of 572

-------
3733

3734

3735

3736

3737

3738

3739

3740

3741

3742

3743

3744

3745

3746

3747

3748

3749

3750

3751

3752

3753

3754

3755

3756

3757

3758

3759

3760

3761

3762

3763

3764

3765

3766

3767

3768

3769

3770

3771

3772

3773

3774

3775

3776

3777

3778

3779

3780

3781

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

RQs were less than 1 for all relevant COUs for air deposition to soil at 1,000 m (Table 4-21). The
highest soil concentration from AERMOD predictions is 0.0055 mg/kg based on TCEP use in Paints and
coatings at job sites at 1,000 m with the 2,500 lb/year production volume and higher-end meteorology
condition. There are no published literature or monitoring databases with TCEP soil concentrations from
U.S. sites and one comparative study from Germany (Mihailovic a s. 2012). RQs for soil
invertebrates were less than 1 with soil TCEP concentrations as reported for different sample periods
from Mihailovic and Fries (2012) (Table 4-16). This study should be considered to represent TCEP
concentrations in soil from an ambient urban environment and is not directly comparable to scenarios
detailed within the current risk evaluation. Mihailovic et al. (2011) emphasized the importance of
atmospheric deposition of chlorinated organophosphate esters in risk assessments, which the current risk
evaluation has taken into consideration for environmental risk characterization.

Trophic transfer of TCEP and potential risk to terrestrial animals was based on modeled soil data from
AERMOD and concentrations reported within Mihailovic and Frie	A screening level approach

was conducted as described in EPA's Guidance for Developing Ecological Soil Screening Levels (U.S.

35a). The two analyses performed represented: (1) trophic transfer for animals from exposures
originating with TCEP soil concentrations and terrestrial prey items (Table 4-18), and (2) trophic
transfer based for animals from exposures with TCEP water concentrations and aquatic prey items
(Table 4-19). Table 4-21 demonstrates that RQs were less than 1 for any modeled soil concentrations
and COUs based on the chronic hazard threshold for terrestrial invertebrate identified in Appendix G.
The chronic TRV, calculated using empirical toxicity data with mice and rats, also demonstrated RQs
less than 1 for all modeled soil concentrations (Table 4-21). In addition, RQs were less than 1 for all
species represented within trophic levels using TCEP soil concentrations reported within Mihailovic and

(2012) (Table 4-18). For semi-aquatic animals, RQs were also less than 1 for semi-aquatic
terrestrial mammals via trophic transfer from fish and the highest modeled TCEP surface water
concentrations (Table 4-19). The results of these screening level trophic transfer analyses corroborate
previous risk assessments indicating TCEP is not a bioaccumulative compound (	2015a; EC.

2009; ECB. 2009).

In the current environmental risk characterization for aquatic and terrestrial organisms, EPA considered
aggregating exposure that a population would experience from multiple facilities in proximity releasing
TCEP to the environment. However, EPA did not aggregate across facilities for environmental
exposures or risk because location information was not available for facilities releasing TCEP to the
environment. Environmental media concentrations from monitoring data {i.e., not associated with a
specific exposure scenario or COU) were not aggregated with modeled environmental media
concentrations associated with a specific exposure scenario or COU. TCEP from monitored surface
water data reported within the WQP indicated a mean of 0.33 + 0.02 ppb (Section 4.3.2). Table 4-12
demonstrates that this mean surface water concentration for TCEP resulted in acute and chronic RQ
values of 3.8xl0~5 and 5.9x10 3, respectively. Similar database monitoring information were not
available for sediment TCEP concentrations; however, the model used to predict surface water,
sediment, and porewater TCEP concentrations was inclusive of physical and chemical properties {i.e.,
Kow, Koc, water column half-life, photolysis half-life, hydrolysis half-life, and benthic half-life) known
to contribute to TCEP's persistence within these media.

EPA also considered aggregating across pathways of exposure for aquatic and terrestrial organisms, but
did not, because releases of TCEP to surface water and sediment were found to significantly contribute
to these media when compared to deposition to water and/or sediment via air (see Section 4.3.5.1).
Similarly, the most significant pathway for exposure to terrestrial receptors is via soil, which was
modeled from air deposition (see Section 4.3.5.2). For aquatic organisms, surface water and sediment

Page 145 of 572

-------
3782

3783

3784

3785

3786

3787

3788

3789

3790

3791

3792

3793

3794

3795

3796

3797

3798

3799

3800

3801

3802

3803

3804

3805

3806

3807

3808

3809

3810

3811

3812

3813

3814

3815

3816

3817

3818

3819

3820

3821

3822

3823

3824

3825

3826

3827

3828

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

pathways involve primary exposure routes such as epithelial uptake (skin, gills) and oral uptake.
Aggregation of exposures via both surface water and dietary exposure was not conducted for aquatic
organisms because TCEP is not expected to bioaccumulate expect at very high concentrations that could
result in risk directly from surface water (see Appendix E.2.6). The screening level trophic transfer
analysis performed included TCEP within prey in addition to soil ingestion for terrestrial receptors and
water ingestion for semi-aquatic mammals (see Section 4.3.1.1).

4.3.6.2 COUs without Quantified Release Estimates

Table 4-7 represents the COUs for which quantitative risk characterization could be performed for
aquatic and terrestrial receptors. The following section represents a qualitative discussion of those
remaining COUs and subsequent OESs lacking quantitative risk estimates.

Recycling and Distribution and Commerce

EPA did not have sufficient data to estimate releases to the environment for the following COUs:

•	Processing - recycling

•	Distribution in commerce

EPA was not able to quantify releases of TCEP to the environment during the recycling of e-waste. E-
waste recycling activities include receiving e-waste at the facility, dismantling or shredding the e-waste,
and sorting the recycled articles and generated scrap materials (NIOSH. 2018; Yang et ai. 2013; Siodin
et ai. 2001). There are 1,455 recycling facilities in the United States (U.S. BL.S. 2016; U.S. Census
Bureau. , ) indicated via NAICS code 562920 - "Materials Recovery Facilities." However, only a
subset of electronic waste facilities is expected to handle TCEP-containing products. The exact number
of these facilities is unknown and data were not available on the volume or source of TCEP contained in
electronics processed at any of the facilities identified.

TCEP-containing materials from the recycling process are typically treated or disposed following the
initial processing and not reprocessed or reused (Yang et ai. 2013). EPA did not find reasonably
available data to quantify environmental releases of TCEP from e-waste facilities. The total releases are
expected to be low since TCEP is not typically used in electronics but is predominantly found in
polyurethane foam (Stapletom et ai. 2011). The NIOSH's Health Hazard Evaluation Program Report on
metals and flame retardants at an electronic recycling company categorized TCEP as "less commonly
used in electronics now and in the past" with a detection percentage 18 percent and range of "not
detectable" to 10 ng/m3 based on full-shift personal air sampling for 19 participants over 2 days (Grimes
et ai. 2019). A fraction of the products are recycled and recycling will likely be dispersed over many e-
waste sites. This qualitative analysis indicates that releases of TCEP to the environment are potentially
present from the recycling of e-waste. However, since TCEP releases are expected to be lower relative
to other quantified scenarios, the recycling COU would be expected to have lower risk than the
quantified scenarios described within Section 4.3.6.1.

Production volume data for TCEP is below reporting levels so the precise production volume is
unknown in order to fully assess TCEP exposure from distribution in commerce. Generally, TCEP
production volumes have declined and this decline would logically lead to decreased distribution into
commerce. Exposure to the environment during distribution in commerce is still possible from ongoing
manufacturing, processing, industrial, and commercial uses. EPA has assessed some risks related to
distribution in commerce (e.g., based on fugitive releases from loading operations) within other relevant
COUs (e.g., manufacturing/repackaging). However, EPA lacks data to assess all risks to the
environment from environmental releases and exposures related to distribution of TCEP in commerce.
Due to limited reasonably available data for the full set of possible exposures, EPA has not made any

Page 146 of 572

-------
3829

3830

3831

3832

3833

3834

3835

3836

3837

3838

3839

3840

3841

3842

3843

3844

3845

3846

3847

3848

3849

3850

3851

3852

3853

3854

3855

3856

3857

3858

3859

3860

3861

3862

3863

3864

3865

3866

3867

3868

3869

3870

3871

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

conclusions regarding risk for this COU separately from the risks already estimated for other relevant
COUs.

Aerospace Equipment and Products

EPA does not expect significant releases to the environment for the following COUs/OESs:

•	Industrial use - other use - aerospace equipment and products

o OES: Installing article (containing 2-part resin) for aerospace applications (electronic
potting)

•	Commercial use - other use - aerospace equipment and products

o OES: Installing article (containing 2-part resin) for aerospace applications

Specifically, EPA does not expect significant releases to occur during the installation of TCEP -
containing aircraft and aerospace articles into or onto the relevant transportation equipment. After
TCEP-containing resins have cured, EPA expects TCEP release will be limited by the hardened polymer
matrix. Releases may occur via the mechanism of "blooming" or volatilization from the cured resin
surface during the service life of the aircraft or aerospace article, but EPA expects that releases via this
mechanism during installation activities will be negligible (OE 09; NICNAS. 2001). The Agency
was not able to quantify environmental releases from blooming in addition to a lack of information on
the end use and service life of the product. EPA considered risk to the environment from installation of
TCEP-containing aircraft and aerospace articles into or onto the relevant transportation equipment. Risk
to the environment from releases of TCEP to the air via blooming from these COUs are expected to have
lower risk compared to quantified scenarios described within Section 4.3.6.1.

Commercial Uses (COUs) That Have Been Phased Out

The COUs listed below are only linked to end of service life disposal as manufacturing and processing is
not ongoing:

•	Commercial use - furnishing, cleaning, treatment/care products - fabric and textile products;

•	Commercial use - furnishing, cleaning, treatment/care products - foam seating and bedding
products;

•	Commercial use - construction, paint, electrical, and metal products - building/construction
materials - insulation; and

•	Commercial use - construction, paint, electrical, and metal products - building/construction
materials - wood and engineered wood products - wood resin composites

EPA has confirmed from literature sources that TCEP was used for these purposes in past decades.
However, these commercial uses were phased out beginning in the late 1980s or early 1990s and
replaced by other flame retardants or flame-retardant formulations. EPA did not locate data to estimate
the TCEP throughput used for these products, the amounts of these products that have already reached
the end of their service life, or amounts that have already been disposed. The Agency assumes that
products with TCEP that are still in use represents a fraction of the overall amount of TCEP previously
used for these purposes and these types of products (e.g., insulation and furniture) will result in a final
deposition to landfills for disposal. However, since TCEP releases are expected to be lower relative to
other quantified scenarios, these commercial COUs would be expected to have lower risk than the
quantified scenarios described within Section 4.3.6.1.

Page 147 of 572

-------
3872

3873

3874

3875

3876

3877

3878

3879

3880

3881

3882

3883

3884

3885

3886

3887

3888

3889

3890

3891

3892

3893

3894

3895

3896

3897

3898

3899

3900

3901

3902

3903

3904

3905

3906

3907

3908

3909

3910

3911

3912

3913

3914

3915

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Processing/Incorporated into Formulation, Mixture, or Reaction Product Processing/Incorporated
into Article

EPA identified the following environmental releases via waste disposal; however, the Agency was
unable to perform quantitative risk characterization of environmental releases related to waste disposal
for the following COUs:

•	Processing/incorporated into formulation, mixture, or reaction product/ paint and coating
manufacturing;

•	Processing/incorporated into formulation, mixture, or reaction product/ paint and coating
manufacturing;

•	Processing/incorporated into formulation, mixture, or reaction product/ polymers used in
aerospace equipment and products; and

•	Processing/incorporated into article/aerospace equipment and products

EPA was able to perform quantitative risk characterization (Table 4-7) on the COUs listed above based
on environmental releases to either fugitive or stack air and/or wastewater to onsite treatment or
discharge to POTW, where applicable (Table 3-2). Waste disposal refers to either landfill or incineration
and relies on inputs provided by the ESD or GSs. The proportion of the throughput that goes to either
landfills or incinerators was not detailed within the ESD or GS. Although details pertaining to the fate of
disposal to these waste streams were unknown, a qualitative analysis of the disposal COU is presented
below.

Consumer Uses

Although there is the possibility of environmental releases from consumer articles containing TCEP via
offgassing of consumer articles, down the drain release of TCEP from domestic laundry, the end-of-life
disposal and demolitions of consumer articles, EPA was unable to quantify the environmental releases
for the following COUs:

•	Consumer use - paints and coatings;

•	Consumer use - furnishing, cleaning, treatment/care products - fabric and textile products;

•	Consumer use - furnishing, cleaning, treatment/care products - foam seating and bedding
products;

•	Consumer use - construction, paint, electrical, and metal products - building/construction
materials - insulation; and

•	Consumer use - construction, paint, electrical, and metal products - building/construction
materials - wood and engineered wood products - wood resin composites

EPA was unable to quantify environmental exposures from consumer releases and disposal due to
limited information on source attribution of the consumer COUs. In previous assessments, EPA has
considered down the drain analysis for consumer products for which a reasonably foreseen direct
discharge exposure scenario can be assumed (e.g., drain cleaner, lubricant, oils). TCEP containing dust
present on consumer clothing may be released to the environment via domestic laundry; however, due to
uncertainties in the source attribution of consumer COUs to dust, and the subsequent loading of dust on
to clothing, EPA did not quantify environmental exposures for this scenario. Consumer releases to the
environment are anticipated to be less than occupational releases, and wastewater concentrations from
manufacturing, commercial and processing COUs were shown to be significantly lower than acute and
chronic COCs identified in Section 4.2.

Page 148 of 572

-------
3916

3917

3918

3919

3920

3921

3922

3923

3924

3925

3926

3927

3928

3929

3930

3931

3932

3933

3934

3935

3936

3937

3938

3939

3940

3941

3942

3943

3944

3945

3946

3947

3948

3949

3950

3951

3952

3953

3954

3955

3956

3957

3958

3959

3960

3961

3962

3963

3964

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Disposal

TCEP was among the 10 most frequently found compounds in a study that collected wastewater from
multiple sites in the Research Triangle Park area of North Carolina between 2002 and 2005 (Giorgino et
ai. 2007). The study detected TCEP in 61.9 percent of wastewater samples, with a maximum
concentration of 0.7 ppb. The maximum concentration from the USGS study (0.7 ppb) is similar to the
maximum surface water TCEP concentration reported within published literature (0.81 ppb) used to
calculate risks (see Section 4.3.2) and resulted in RQ values of less than one for both acute and chronic
COCs (Table 4-13). The researchers indicated that flame retardants were measured primarily at sites
downstream from municipal wastewater discharges and elevated concentrations were due to surface
waters collected at a site downstream from an industrial fire.

Incineration of articles containing TCEP may create localized environmental releases. Aston et al.
(1996) reported TCEP concentrations of up to 1.95 mg/kg in pine needles (Pinus ponderosa) in the
Sierra Nevada foothills in the mid-1990s (Table 4-3). The source of the TCEP is unknown; however,
authors suspected that these levels may have been due to aerial transport and deposition from nearby
point sources such as incinerators.

The demolition and removal of commercial and consumer articles may result in environmental
exposures to TCEP. Construction waste and old consumer products can be disposed of in municipal
solid waste landfills and construction and demolition landfills. Section 3.3.3.7 models the resulting
groundwater concentration that may occur from TCEP that leaches from landfills. Section 3.3.3.5
highlights suspected leaching of TCEP from nearby landfills (Norman Landfill, Himco Dump and Fort
Devens, MA) (Buszka et al.. 2009; Barnes et al.. 2004; Hutchins et al.. 1984). The Himco Dump is a
closed, formerly unlicensed landfill that included a 4-acre construction debris area. EPA issued a notice
in the Federal Register finalizing the deletion of part of the Himco Dump Superfund site from the
National Priorities List (NPL). The Indiana Department of Environmental Management (IDEM)
formally concurred with EPA's proposal on January 26, 2022, and EPA, proposed the Site for partial
deletion in March 2022 Groundwater from one well in Elkhart, Indiana, near the Himco Dump reported
TCEP concentrations of 0.65 ppb to 0.74 ppb (Buszka et al.. 2009). Fort Devens is also an EPA,
superfund site, a former army installation established in 1917 and closed in 1996. Monitoring wells
down-gradient of a land application facility near Fort Devens, Massachusetts, indicated TCEP
concentrations from 0.28 ppb to 0.81 ppb (Hutchins et a |). TCEP was detected throughout the
entire length of a leach ate plume near a municipal landfill (subtitle D) near Norman, Oklahoma (Barnes
et al.. 2004). TCEP concentration detected within the groundwater plume down-gradient of the Landfill
in Norman, Oklahoma, ranged from 0.22 ppb to 0.74 ppb (Barnes et al.. 2004). Leachate samples from
landfill sites in Japan detected TCEP at ranges from 4.1 to 5430 mg/mL with authors indicating that
plastic wastes may serve as the origin (Yasuhara. 1995).

Without a full characterization of non-hazardous landfill (e.g., Norman Landfill) conditions and
historical wastes (e.g., Himco Dump and Fort Devens) around the country, the data needed to produce
quantitative risk estimates for disposal is not reasonably available. EPA does not have data representing
municipal and managed landfills and is uncertain how often contaminant migration occurs given modern
practices of non-hazardous landfill and historical site management. Source attribution of the consumer
uses to the leaching concentration exhibited within Sections 3.3.3.6 and 3.3.3.7 are not available;
therefore, it is unknown if these concentrations are the result of consumer and/or commercial disposal.
The possibility of environmental exposure to TCEP after the release from disposal of consumer wastes
exists. The maximum TCEP concentrations recorded within groundwater at the Norman Landfill, Himco
Dump, and Ft. Devens are 0.74 ppb, 0.81 ppb, and 0.74 ppb, respectively—which are similar to the to
the maximum surface water concentrations reported within published literature (0.81 ppb) used to

Page 149 of 572

-------
3965

3966

3967

3968

3969

3970

3971

3972

3973

3974

3975

3976

3977

3978

3979

3980

3981

3982

3983

3984

3985

3986

3987

3988

3989

3990

3991

3992

3993

3994

3995

3996

3997

3998

3999

4000

4001

4002

4003

4004

4005

4006

4007

4008

4009

4010

4011

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

calculate risks (see Section 4.3.2) resulting in RQ values less than one for both acute and chronic COCs
(Table 4-13). TCEP releases from disposal of consumer and commercial articles are expected to be
lower relative to other quantified scenarios, the disposal COU would be expected to have lower risk than
the quantified scenarios described within Section 4.3.6.1.

4,3.7 Overall Confidence and Remaining Uncertainties Confidence in Environmental
Risk Characterization

The overall confidence in the risk characterization combines the confidence from the environmental
exposure, hazard threshold, and trophic transfer sections. This approach aligns with the 2021 Draft
Systematic Review Protocol (	2021) and Systematic Review Protocol for the Draft Risk

Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) (	2023rt). The confidence from the

trophic transfer section was completed in the same manner as the confidence in hazard threshold
presented in Section 4.2.6 and Appendix F.2.3.1. For trophic transfer, EPA considers the evidence for
chronic mammalian robust, the evidence for invertebrates moderate, and the evidence for chronic avian
slight (Table 4-22). Synthesis of confidence for exposure, hazard, and trophic transfer (when applicable)
resulted in the following confidence determinations for risk characterization RQ inputs: (1) robust for
chronic mammalian evidence, (2) moderate for acute and chronic aquatic evidence, and (3) slight for
secondary acute and secondary chronic aquatic assessments with additional assessment factors and
chronic avian evidence (Table 4-23).

4.3.7.1 Trophic Transfer Confidence

Quality of the Database; and Strength (Effect Magnitude) and Precision

Several conservative assumptions were applied across different representative organisms within trophic
groups to represent a screening level approach. For example, modeled TCEP concentrations within
water (VVWM-PSC) and soil (via AERMOD) were applied to all COUs. TCEP concentrations obtained
from these models were specific to each COU and production volume scenarios. Examination of
potential risk from TCEP using this hazard value should be viewed as a conservative approach
employed using both AERMOD modeled data and soil concentrations within published literature
(Mihailovic and Fries.! ).

Trophic transfer analysis utilized American woodcock and American kestrel within the soil-based
pathway to determine potential risk from TCEP. The hazard value for the raptor species is limited to a
single study observing increased thyroid hormone production with no effects on body weight or food
consumption from a 21-day feeding study (Fernie et at.. 2015). No representative hazard data were
available for the woodcock as an avian insectivore. RQ values were not calculated for the woodcock,
which served as a prey item to the kestrel, representing uptake and transfer from a soil invertebrate to
insectivore to carnivore.

Short-tailed shrew and American mink were employed as representative species using a mammalian
TRV adjusted to their respective body weights. Mammalian hazard values for trophic transfer utilized
ecologically relevant endpoints from high-quality studies originating from human health animal model
investigations. The resulting TRV (Table 4-5) derived from mammal studies was used to calculate the
hazard threshold in mg/kg-bw. Because the TRV is scaled by body weight, smaller representative
species will have greater body burden from TCEP exposure than larger species.

For soil invertebrates, two high-quality soil invertebrate studies were available. Trophic transfer analysis
used an ecologically relevant ChV from a nematode with endpoints related to reduced growth and
shortened lifespan. The earthworm hazard value was also demonstrated in this analysis, although the
earthworm did not have an ecologically relevant endpoint effect. The earthworm is still useful for

Page 150 of 572

-------
4012

4013

4014

4015

4016

4017

4018

4019

4020

4021

4022

4023

4024

4025

4026

4027

4028

4029

4030

4031

4032

4033

4034

4035

4036

4037

4038

4039

4040

4041

4042

4043

4044

4045

4046

4047

4048

4049

4050

4051

4052

4053

4054

4055

4056

4057

4058

4059

4060

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

assessing trophic transfer hazards because of its direct ingestion of soil. The earthworm also serves as a
relevant prey item for all trophic levels {i.e., short-tailed shrew, woodcock, and American kestrel).

Consistency

Inputs for soil and water TCEP concentrations displayed similarities among modeled and monitored
concentrations. The highest soil concentrations modeled via AERMOD (Table 4-15) were within one
order of magnitude to the highest soil concentrations reported within published literature (Table 4-16)
(Mihailovic and Fries. ^ ). Concentrations of TCEP in whole fish reported within published literature
(Guo et al. 2017b) represent concentrations two to three orders of magnitude lower than calculated fish
TCEP concentrations (see Section 4.1.2). Any comparison to measured values reported within published
literature should be viewed conservatively as organisms with direct proximity to source of TCEP release
and resulting surface water concentrations as calculated using VVWM-PSC.

Biological Relevance

The use of hazard values derived from singular studies for American kestrel, earthworm, and nematode
are limiting in biological relevance; however, the application of conservative assumptions at each
trophic level ensures a cautious approach to determining potential risk. For example, if the results of the
trophic transfer show that exposure from TCEP is lower than the hazard threshold for thyroid effects,
than a qualitative assertion can be made that the exposure levels from TCEP do not indicate risk. For
avian species, only a single high-quality level study was available for the American kestrel with no
hazard value for the avian insectivore within this analysis. The short-tailed shrew and American mink
were selected as appropriate representative mammals for the soil- and aquatic-based trophic transfer
analysis, respectively (U.S. EPA. 1993b). Overall, the use of exposure factors {i.e., feed intake rate,
water intake rate, the proportion of soil within the diet) from a consistent resource assisted in addressing
species specific differences within the RQ equation (	93b).

Physical and Chemical Relevance

The highest modeled TCEP concentrations for water and soil were used to investigate potential risk
from trophic transfer. Assumptions within the trophic transfer equation (Equation 4-3) for this analysis
have been considered to represent conservative screening values (	2005a) and those

assumptions were applied similarly for each trophic level and representative species. Applications across
representative species included assuming 100 percent TCEP bioavailability from both the soil (AF*,) and
biota representing prey (AFy). It is likely these considerations overrepresent TCEP's ability to transfer
among trophic levels; however, it is a precaution built into the screening level approach (U.S. EPA.
2005a).

Environmental Relevance

Although several aspects of the RQ equation were conservative and represented various species, there
are still uncertainties associated with overall relevance of this model to fit all wildlife scenarios for
potential TCEP risk. The current trophic transfer analysis investigated potential risk resulting from
TCEP exposure in media such as soil and water. This analysis was extended to represent uptake from
those media to soil invertebrates and fishes as a basis of trophic transfer from these prey to other higher
trophic levels. Analysis included TCEP soil concentrations from published literature but ultimately
relied on modeled TCEP water concentrations as the monitored TCEP values from WQP are three to
five orders of magnitude less than modeled concentrations. The area use factor is the home range size
relative to the contaminated area {i.e., site/home range = AUF with the AUF within this screening level
analysis designated as 1 for all organisms). Application of this value in the RQ equation increases the
conservative approach to trophic transfer analysis for larger animals such as mammals and birds
assuming longer residence within an exposed area or a large exposure area.

Page 151 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table 4-22. TCEP Evidence Table Summarizing Overall Con

Idence Derivet

for Trophic Transfer

Types of Evidence

Quality of the
Database

Consistency

Strength and
Precision

Biological Gradient/
Dose-Response

Relevance"

Trophic Transfer
Confidence

Aquatic

Acute Aquatic Assessment

N/A

N/A

N/A

N/A

N/A

N/A

Chronic Aquatic Assessment

N/A

N/A

N/A

N/A

N/A

N/A

Aqualic pkinls (\ascular and alyac

\ A

\ A

\ A

1 cnvslnal

\ A

\ A

\ A

Chronic Avian Assessment

+

++

+

N/A

+

Slight

Chronic Mammalian Assessment

+++

++

++

N/A

++

Moderate

Terrestrial invertebrates

++

++

++

N/A

++

Moderate

a Relevance includes biological, physical/chemical, and environmental relevance.

+ + + Robust confidence suggests thorough understanding of the scientific evidence and uncertainties. The supporting weight of the scientific evidence
outweighs the uncertainties to the point where it is unlikely that the uncertainties could have a significant effect on the hazard estimate.

+ + Moderate confidence suggests some understanding of the scientific evidence and uncertainties. The supporting scientific evidence weighed against
the uncertainties is reasonably adequate to characterize hazard estimates.

+ Slight confidence is assigned when the weight of the scientific evidence may not be adequate to characterize the scenario, and when the assessor is
making the best scientific assessment possible in the absence of complete information. There are additional uncertainties that may need to be considered.

4062

Page 152 of 572

-------
4063

4064

4065

4066

4067

4068

4069

4070

4071

4072

4073

4074

4075

4076

4077

4078

4079

4080

4081

4082

4083

4084

4085

4086

4087

4088

4089

4090

4091

4092

4093

4094

4095

4096

4097

4098

4099

4100

4101

4102

4103

4104

4105

4106

4107

4108

4109

4110

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

4.3.7.2 Risk Characterization Confidence

Environmental risk characterization evaluated confidence from environmental exposures and
environmental hazards. Hazard confidence was represented by evidence type as reported previously in
Section 4.2.6. Trophic transfer confidence was represented by evidence type as reported in the preceding
Section 4.3.7.1. Exposure confidence has been synthesized from Section 4.1.5.1 and is further detailed
in the current section. The following confidence determinations for risk characterization RQ inputs are:
(1) robust for chronic mammalian evidence, (2) moderate for acute and chronic aquatic evidence, and
(3) slight for secondary acute and secondary chronic aquatic assessments and chronic avian evidence
(Table 4-23).

Surface water concentration of TCEP were modeled initially using E-FAST and further refined using
VVWM-PSC. Refined modeling with VVWM-PSC allowed estimates of TCEP pore water and sediment
concentrations in addition to providing modeled days of exceedance for each compartment. Uncertainty
associated with location-specific model inputs (e.g., flow parameters and meteorological data) is present
as no facility locations were identified for TCEP releases.

The modeled data represent estimated concentrations near hypothetical facilities that are actively
releasing TCEP to surface water, while the reported measured concentrations represent sampled ambient
water concentrations of TCEP. Differences in magnitude between modeled and measured concentrations
may be due to measured concentrations not being geographically or temporally close to known releasers
of TCEP. VVWM-PSC allowed for the application of a standard, conservative set of parameters and
adjust for physical-chemical properties of TCEP. For example, stream reach was set to represent a
waterway with a width of 8 m and depth of 2 m.

Physical and chemical properties including, but not limited to Koc, benthic half-life and hydrolysis half-
life appear to accurately represent TCEP's persistence; however, sensitivity analysis indicated that Koc
input parameters heavily influenced the role of sediment deposition to sediment. As a result, Koc was
represented as both the mean (2.82) and the 5th percentile of the mean (2.13), as detailed within Section
4.3.1. Maruva et al. (2016) represents an ambient environmental monitoring study within the published
literature that made both surface water and sediment collections at the same sites and similar time
periods within a watershed. Surface water collected in August and October 2013 and sediment samples
collected in September 2013 were taken at 6 sites downstream of urban areas along the Santa Clara
River in Southern California. TCEP sediment concentrations were consistently one order of magnitude
higher than TCEP surface water concentrations across all sample sites. Specifically, mean (+ SE) TCEP
concentrations for surface water and sediment were 0.32 + 0.04 ppb and 2.59 + 0.75 ppb, respectively.
Although a single study, Maruva c	illustrates how TCEP within the water column of a

flowing system can sorb to sediment to produce elevated concentrations. The WQP data and published
literature on surface water TCEP concentrations is three to four orders of magnitude lower than modeled
surface water concentrations. Confidence in the exposure components of the RQ inputs for benthic
assessment is supported as studies within published literature are one to three orders of magnitude lower
than results obtained from VVMW-PSC modeling. Confidence in exposure parameters for surface water
have been rated "moderate" as the results are modeled from directly downstream from a hypothetical
facility releasing TCEP.

Similar to aquatic exposures for TCEP, environmental exposures to soil invertebrates, mammals, and
avian species relied on modeling air deposition to soil via AERMOD with supporting information from
published literature. The AERMOD model included two meteorological conditions (Sioux Falls, South
Dakota, for central tendency meteorology; and Lake Charles, Louisiana, for higher-end meteorology) in

Page 153 of 572

-------
4111

4112

4113

4114

4115

4116

4117

4118

4119

4120

4121

4122

4123

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

addition to different production volumes (2,500 and 25,000 lb/year) to characterize potential amounts of
annual TCEP deposition to soil from air. One high-quality comparative study on TCEP soil
concentrations was identified within the published literature. TCEP fish tissue concentrations within the
Great Lakes (Guo et al. 2017b) are two to three orders of magnitude lower than the TCEP tissue
concentrations calculated using a whole organism BCF value from another high-quality study (Arukwe
et al.. 2018). Modeled soil concentrations were within one order of magnitude of a single study from
published literature (Mihailovic and Fries. 2012); however, it is important to note that similarity with a
single study is not enough to build confidence in the relevance or accuracy of modeled results.

Table 4-23. TCEP Evidence Table Summarizing Overall Confidence for Environmental
Risk Characterization

Types of Evidence

Exposure

Hazard

Trophic
Transfer

Risk
Characterization
RQ Inputs

Ac

uatic

Acute aquatic assessment

++

++

N/A

Moderate

Chronic aquatic assessment

++

++

N/A

Moderate

Secondary acute aquatic
assessment (+ AF)

++

+

N/A

Slight

Secondary chronic aquatic
assessment (+ AF)

++

+

N/A

Slight

1 cnvslnal

Chronic avian assessment

++

+

+

Slight

Chronic mammalian assessment

++

+++

++

Robust

Terrestrial invertebrates

++

++

++

Moderate

+ + + Robust confidence suggests thorough understanding of the scientific evidence and uncertainties. The supporting
weight of the scientific evidence outweighs the uncertainties to the point where it is unlikely that the uncertainties
could have a significant effect on the hazard estimate.

+ + Moderate confidence suggests some understanding of the scientific evidence and uncertainties. The supporting
scientific evidence weighed against the uncertainties is reasonably adequate to characterize hazard estimates.

+ Slight confidence is assigned when the weight of the scientific evidence may not be adequate to characterize the
scenario, and when the assessor is making the best scientific assessment possible in the absence of complete
information. There are additional uncertainties that may need to be considered.

Page 154 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

4124	5 HUMAN HEALTH RISK ASSESSMENT

4125	EPA assessed human health risks of TCEP exposure to workers and ONUs, consumers, and the general

4126	population. Section 5.1 describes exposures to workers and ONUs via inhalation and oral routes;

4127	workers via dermal routes; consumers via inhalation, dermal, and oral routes; and the general population

4128	via oral, dermal, and inhalation routes. Human health hazards, including cancer and non-cancer endpoint

4129	identification and dose-response, are described in Section 5.2. Human health risk characterization is

4130	described in Section 5.3.

Page 155 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

4131 5.1 Human Exposures

TCEP - Human Exposures (Section 5.1):

Key Points

EPA evaluated all reasonably available information for occupational, consumer, and general
population exposure to TCEP, including consideration of the potential for increased susceptibility
across PESS considerations (see also Section 5.3.3 and Appendix D). The following bullets
summarize the key points of this section of the draft risk evaluation:

• Workers and ONUs can be exposed to TCEP via inhalation by dust or vapor.

o However, large amounts of dust are not expected to be generated based on the types of

activities that occur during the processing or use of TCEP-containing products or articles,
o Workers can also be exposed to mists generated during the spray application of TCEP-

containing paint products, but ONUs are not expected to be present during this use.
o Workers will be exposed to TCEP via dermal exposure when processing liquid TCEP.
however, once TCEP has been incorporated into an article the ability for appreciable
amounts of TCEP to be absorbed through the skin will decrease significantly as there is
little need for further processing of an article during installation.

•	Chronic TCEP exposures from consumer articles to infants and children are the most relevant
duration and populations of interest. Children's mouthing activity is an important factor when
estimating exposure to TCEP in consumer products.

o For consumer exposures, the inhalation route dominates exposure for building and
construction materials such as roofing insulation, acoustic ceilings, and wood flooring.
Exposures to infants and children for fabric and textiles, foam seating and bedding
products, and wooden TV stands is dominated by the oral route,
o Inhalation exposures are highest for building and construction products due to emission

of vapors from consumer articles,
o Dermal exposures are highest for wood resin products to children,
o Ingestion exposures are highest for foam seating and bedding products for children.

•	Fish ingestion is the most important exposure scenario for TCEP exposure to the general
population. BAF and fish ingestion rate are sensitive parameters that influence these exposure
estimates. Tribal populations for whom fish is important dietarily and culturally may have even
higher exposure than the general population and subsistence fishers.

•	Fenceline communities may have elevated exposure from facilities that release TCEP. No site-
specific information was available for TCEP, so EPA varied several inputs to show a range of
possible exposures from a hypothetical facility.

•	EPA identified several PESS groups: Infant exposure to TCEP via human milk was estimated
by considering a maternal dose due to occupational, consumer, and general population
exposures. Firefighters were identified as a PESS group through occupational exposure
(Section 5.3.3). Children and infants were identified as PESS through consumer exposure.
Subsistence fishers, children, infants, and fenceline communities were identified as PESS
through general population exposures.

4132

4133

Page 156 of 572

-------
4134

4135

4136

4137

4138

4139

4140

4141

4142

4143

4144

4145

4146

4147

4148

4149

4150

4151

4152

4153

4154

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

5.1.1 Occupational Exposures

TCEP - Occupational Exposures (Section 5.1.1):

Key Points

EPA evaluated the reasonably available information for occupational exposures. The key points of
the occupational exposure assessment are summarized below:

•	Occupational exposure data available for TCEP:

o EPA only identified monitoring data for dust occurring within an electronic waste
recycling facility; monitoring data for the remaining COUs/OESs was not found, most
likely because TCEP does not have an assigned OSHA PEL and is therefore not typically
tested for in the workplace,
o For OESs that do not have data, EPA used relevant generic scenario and/or emission
scenario documents to identify worker activities and exposure routes that are reasonably
expected to occur. Exposure distributions were then created using Monte Carlo
simulation with 100,000 iterations and the Latin Hypercube sampling method.

•	The OES, use of paints and coatings - spray application, had the highest occupational
exposure for inhalation and dermal exposure; this is due to mist being generated during
application as well as a higher dermal loading value:

o Inhalation exposure for use of paints and coatings - spray application ranges from 5.500
mg/m3 (95th percentile, 8-hr TWA, resin-based paints) to 1.7/10 1 mg/m3 (50th
percentile, 8-hr TWA, water-based paints). EPA identified mist generation as the main
driver of exposure but is not expected to occur during other COUs/OESs.
o Dermal acute retained dose (mg/kg-day) ranges from 8.02 (95th percentile) to 1.48 (50th
percentile).

The following subsections briefly describe EPA's approach to assessing occupational exposures and
results for each condition of use assessed. For additional details on development of approaches and
results refer to the Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Supplemental
Information File: Supplemental Information on Environmental Release and Occupational Exposure
Assessment (U.S. EPA. 20231). As discussed in Section 3.1.1, EPA has mapped the industrial and
commercial COUs to OESs in Table 3-1.

5.1.1.1 Approach and Methodology	

As described in the Final Scope of the Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP)
CASRN115-96-8 (U.S. EPA. 2020b). for each COU, EPA distinguishes exposures for workers and
ONUs. Normally, a primary difference between workers and ONUs is that workers may handle TCEP
and have direct contact with the chemical, while ONUs are working in the general vicinity of workers
but do not handle TCEP and do not have direct contact with it. Where possible, for each COU, EPA
identified job types and categories for workers and ONUs.

As discussed in Section 3.1.1, EPA established OESs to assess the exposure scenarios more specifically
within each COU. Table 3-1 provides a crosswalk between COUs and OESs. Figure 5-1 provides the
approaches used by EPA to estimate exposures for the OESs included in this draft risk evaluation of
TCEP. EPA did not identify any relevant inhalation exposure monitoring data to TCEP vapor for any of
the OESs, because TCEP does not have an Occupational Safety and Health Act (OSHA) permissible

Page 157 of 572

-------
4155

4156

4157

4158

4159

4160

4161

4162

4163

4164

4165

4166

4167

4168

4169

4170

4171

4172

4173

4174

4175

4176

4177

4178

4179

4180

4181

4182

4183

4184

4185

4186

4187

4188

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

exposure limit (PEL). For two OESs monitoring data was available for TCEP in dust. The quality of the
monitoring data was evaluated using the data quality review evaluation metrics and the categorical
ranking criteria described in the Draft Systematic Review Protocol Supporting TSCA Risk Evaluations
for Chemical Substances (U.S. EPA. 2021). Relevant data were assigned an overall quality
determination of high, medium, low, or uninformative. In addition, EPA established an overall
confidence for the data when integrated into the occupational exposure assessment. The Agency
considered the assessment approach, the quality of the data and models, as well as uncertainties in
assessment results to assign an overall confidence level of robust, moderate, or slight.

Where monitoring data were reasonably available, EPA used this data to characterize central tendency
and high-end inhalation exposures. Where no inhalation monitoring data were identified, but inhalation
exposure models were reasonably available, EPA estimated central tendency and high-end exposures
using only modeling approaches. If both inhalation monitoring data and exposure models were
reasonably available, where applicable, EPA presented central tendency and high-end exposures using
both. EPA only identified measured dermal exposure estimates for dust generated at e-waste facilities.
Monitoring data were not reasonably available for any other COUs. EPA standard models, such as the
EPA Mass Balance Inhalation Model and Fractional Absorption Model, were used to estimate high-end
and central tendency inhalation and dermal exposures for workers in each OES.

For many cases, EPA did not have monitoring data to estimate inhalation exposure for ONUs. In some
cases, this was addressed with the use of exposure models, when available. However, most OESs do not
contain inhalation exposure estimates for ONUs. In general, EPA expects ONU exposures to be less
than worker exposures. Dermal exposure for ONUs was not evaluated because these employees are not
expected to be in direct contact with TCEP.

Figure 5-1. Approaches Used for Each Component of the Occupational Assessment for Each OES

CDR = Chemical Data Reporting; GS = Generic Scenario; ESD = Emission Scenario Document; BLS = Bureau
of Labor Statistics; NIOSH (HHE) = National Institute of Occupational Safety and Health (Health Hazard
Evaluations); Fab = Fractional Absorption Model

In Table 5-1, EPA provides a summary for each OES by indicating whether monitoring data were
reasonably available; how many data points were identified, the quality of the data; EPA's overall
confidence in the data; whether the data were used to estimate inhalation exposures for workers and

Page 158 of 572

-------
4189

4190

4191

4192

4193

4194

4195

4196

4197

4198

4199

4200

4201

4202

4203

4204

4205

4206

4207

4208

4209

4210

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

ONUs; and whether EPA used modeling to estimate inhalation exposure to dust, vapors, or mist and
dermal exposures for workers and ONUs.

Table 5-2 provides a summary of EPA estimates for the total number of potentially exposed workers and
ONUs for each OES. To prepare these estimates, EPA first attempted to identify NAICS codes
associated with each OES. For these NAICS codes, EPA then reviewed Standard Occupational
Classification (SOC) codes from the Bureau of Labor Statistics (BLS) and classified relevant SOC codes
as workers or ONUs. All other SOC codes were assumed to represent occupations where exposure is
unlikely. EPA also estimated the total number of facilities associated with the NAICS codes previously
identified based on data from the U.S. Census Bureau.

EPA then estimated the average number of workers and ONUs potentially exposed per generic site by
dividing the total number of workers and ONUs by the total number of facilities. Finally, using EPA's
estimates for the number of facilities using TCEP, the Agency was able to estimate the total number of
workers and ONUs potentially exposed to TCEP for each OES. Additional details on EPA's approach
and methodology for estimating the number of facilities using TCEP and the number of workers and
ONUs potentially exposed to TCEP can be found in th q Draft Risk Evaluation for Tris(2-chloroethyl)
Phosphate (TCEP) - Supplemental Information File: Supplemental Information on Environmental
Release and Occupational Exposure Assessment (	)231).

Page 159 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

4211 Table 5-1. Summary for Each Occupational Exposure Scenarios (PES)



Inhalation Exposure

Dermal Exposure

OES



Monitorin

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

OES

Inhalation Exposure

Dermal Exposure

Monitoring

Modeling

Inhalation Exposure
Confidence"

Monitoring

Modeling

Dermal Exposure
Confidence"

Worker

# Data
Points

ONU

# Data
Points

Overall
Quality
Determ-
ination

Worker

ONU

Worker

ONU

Worker

Overall
Quality
Determ-
ination

Worker

Worker

ONU

Commercial use - lab
chemical - use of
laboratory chemicals

X

N/A

X

N/A

N/A

V

X

Robust

Moderate

X

N/A

V

Moderate

N/A

commercial uses:
furnishing, cleaning,
treatment/care products
fabric and textile
products

•	Foam seating and
bedding products

Construction, paint,
electrical, and metal
products

•	Building/construction
materials - insulation

•	Building/construction
materials - wood and
engineered wood
products - wood
resin composites

X

N/A

X

N/A

N/A

X

X

N/A

N/A

X

N/A

X

N/A

N/A

Disposal

Evaluated as part of each OES as opposed to a standalone OES





Where EPA was not able to estimate ONU inhalation exposure from monitoring data or models, this was assumed equivalent to the central tendency experienced by workers for
the corresponding OES; dermal exposure for ONUs was not evaluated because they are not expected to be in direct contact with TCEP.

" Robust confidence suggests thorough understanding of the scientific evidence and uncertainties. The supporting weight of the scientific evidence outweighs the uncertainties to
the point where it is unlikely that the uncertainties could have a significant effect on the hazard estimate.

Moderate confidence suggests some understanding of the scientific evidence and uncertainties. The supporting scientific evidence weighed against the uncertainties is reasonably
adequate to characterize hazard estimates.

Slight confidence is assigned when the weight of the scientific evidence may not be adequate to characterize the scenario, and when the assessor is making the best scientific
assessment possible in the absence of complete information. There are additional uncertainties that may need to be considered.

4212

4213

Page 161 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

4214	5.1.1.2 Summary of Inhalation Exposure Assessment

4215	Table 5-2 summarizes the number of facilities and total number of exposed workers for all OESs.

4216

4217	Table 5-2. Summary of Total Number of Workers and ONUs Potentially Exposed to TCEP for

4218	EachOES"

OES

Total Exposed
Workers /
Site

T otal
Exposed

ONUs/
Site

Total Exposed / Site
(Exposure davs/vr

High-End-
Central Tendency)

Number of
Facilities"

Notes

Manufacture
(import) -
repackaging

1

0

1

(7-4)

1 generic site

424690 - Other
Chemical and Allied
Products Merchant
Wholesalers

Processing -
incorporation into
paints and coatings
- 1-part coatings

14

5

19

(38-6)

1 generic site

325510 - Paint and
Coating Manufacturing

Processing -
incorporation into
paints and coatings
- 2-part reactive
coatings

14

5

19

(2-1)

1 generic site

325510 - Paint and
Coating Manufacturing

Processing -
formulation of
T CEP-containing
reactive resins (for
use in 2-part
systems)

27

12

39

(6-1)

1 generic site

325211 - Plastics
Material and Resin
Manufacturing

Processing -
processing into 2-
part resin article

75

64

139

(250 - 72)

1 generic site

326400 - Aerospace
Product and Parts
Manufacturing

Processing -
recycling e-waste

2

2

4

(250 - 250)

Unknown

562920 - Materials
Recovery Facilities

Distribution - distribution in commerce

Distribution activities (e.g., loading) considered throughout life
cycle, rather than using a single distribution scenario

Industrial use -
installation of
article

75

64

139

(250 - 250)

1 generic site

326400 - Aerospace
Product and Parts
Manufacturing

Commercial use -
Use and/or
maintenance of
aerospace
equipment and
products

75

64

139

(250 - 250)

1 generic site

326400 - Aerospace
Product and Parts
Manufacturing

Commercial use -
use of paints and
coatings - spray
application

3

0

3

Sites vary based
on multiple
throughput
scenarios; see
Table 3-2

811121 - Automotive
Body, Paint, and
Interior Repair and
Maintenance

Page 162 of 572

-------
4219

4220

4221

4222

4223

4224

4225

4226

4227

4228

4229

4230

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

OES

Total Exposed
Workers /
Site

T otal
Exposed
ONUs/
Site

Total Exposed / Site
(Exposure davs/vr

High-End-
Central Tendency)

Number of
Facilities"

Notes



4

0

4

(Exposure days based
on 1-, 2-, or 250-day
scenarios)



238320 - Painting and
Wall Covering
Contractors

Commercial Use -
lab chemical - use
of laboratory
chemicals

3

3

6

(220 - 214)

13 sites (1st
percentile)

6 sites (5th
percentile)

541380 - Testing
laboratories

541713	- Research and
development in
nanotechnology

541714	- Research and
development in
biotechnology (except
nanobiotechnology)

541715	- Research and
development in the
physical, engineering,
and life sciences
(except nanotechnology
and biotechnology)
621511 - Medical
Laboratories

Commercial Uses -

•	Furnishing, cleaning,
treatment/care products

o Fabric and textile products
o Foam seating and bedding
products

•	Building/construction materials

o Insulation

o Wood resin composites

Manufacturing and processing for
these COU's has ceased

N/A

Disposal

Evaluated as part of each OES as opposed to a standalone OES

a EPA's approach and methodology for estimating the number of facilities using TCEP and the number of workers
and ONUs potentially exposed to TCEP can be found in the Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate
(TCEP) - Supplemental Information File: Supplemental Information on Environmental Release and Occupational
Exposure Assessment ( 231).

A summary of inhalation exposure results based on monitoring data and exposure modeling for each
OES is presented for workers in Table 5-3 and Table 5-4, respectively. ONUs are presented in Table
5-5. These tables provide a summary of time-weighted average (TWA) inhalation exposure estimates as
well as acute exposure concentrations (AC), average daily concentrations (ADC), lifetime average daily
concentrations (LADC), and subchronic average daily concentration (SCADC). The ADC is used to
characterize risks for chronic non-cancer health effects whereas the LADC is used for chronic cancer
health effects. The SCADC represents repeated exposure for approximately 30 days. Additional details
regarding AC, ADC, LADC, and SCADC calculations along with EPA's approach and methodology for
modeling inhalation exposure can be found in Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate
(TCEP) - Supplemental Information File: Supplemental Information on Environmental Release and
Occupational Exposure Assessment (U.S. EPA. 20231).

Page 163 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

4231 Table 5-3. Summary of Inhalation Exposure Results for Workers Based on Monitoring Data for Each PES

OES

Inhalation Monitoring (Worker, ppm)

TWA

AC

ADC

LADC

SADC

High-End

Central
Tendency

High-End

Central
Tendency

High-End

Central
Tendency

High-End

Central
Tendency

High-End

Central
Tendency

Processing - recycling e-waste

9.68E-04

1.00E-07

6.6E-04

6.80E-08

4.51E-04

4.66E-08

2.31E-04

1.85E-08

4.83E-04

4.99E-08

Industrial use - installation of
article

1.3E-05

1.3E-05

8.8E-06

8.8E-06

6.5E-06

6.5E-06

3.1E-06

2.4E-06

6.5E-06

6.5E-06

Commercial use - use and/or
maintenance of aerospace
equipment and products

1.3E-05

1.3E-05

8.8E-06

8.8E-06

6.5E-06

6.5E-06

3.1E-06

2.4E-06

6.5E-06

6.5E-06

4232

4233

4234	Table 5-4. Summary of Inhalation Exposure Results for Workers Based on Exposure Modeling for Each PES

OES

Inhalation Modeling (Worker, mg/m3)

TWA (8-hr)

AC

ADC

LADC

SADC

High-End

Central
Tendency

High-End

Central
Tendency

High-End

Central
Tendency

High-End

Central
Tendency

High-End

Central
Tendency

Manufacture (import) -
repackaging

4.1E-02

1.1E-02

2.8E-02

7.5E-03

3.1E-03

8.9E-05

1.2E-04

3.4E-05

3.7E-03

1.1E-03

Processing - incorporation into
paints and coatings - 1-part
coatings

1.0E-01

1.7E-02

7.1E-02

1.1E-02

8.0E-04

1.9E-04

3.2E-04

7.3E-05

9.2E-03

2.2E-03

Processing - incorporation into
paints and coatings - 2-part
reactive

4.0E-01

9.6E-02

2.7E-01

6.5E-02

7.9E-04

1.9E-04

3.1E-04

7.1E-05

9.6E-03

2.3E-03

Processing - formulation of
TCEP-containing reactive
resins (for use in 2-part
systems)

4.1E-01

7.4E-02

2.8E-01

5.1E-02

8.4E-04

1.8E-04

3.3E-04

6.9E-05

1.0E-02

2.2E-03

Page 164 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023



Inhalation Modeling (Worker, mg/mJ)

OES

TWA (8-hr)

AC

ADC

LA DC

SADC

High-End

Central
Tendency

High-End

Central
Tendency

High-End

Central
Tendency

High-End

Central
Tendency

High-End

Central
Tendency

Processing - processing into 2-
part resin article

1.8E-02

3.4E-03

1.2E-02

2.3E-03

2.3E-03

3.9E-04

9.2E-04

1.5E-04

8.1E-03

1.6E-03

Distribution - distribution in

Distribution activities (e.g., loading) considered throughout life cycle, rather than using a single distribution scenario

commerce





















Commercial use - paints &

1.1E00

1.7E-01

7.5E-01

1. IE—01

2.1E-03

3.1E-04

1.1E-03

1.3E-04

2.5E-02

3.8E-03

coatings - spray (1-part
coatings, 1-day application)
(OES #7)





















Commercial use - paints &

1.1E00

1.7E-01

7.5E-01

1. IE—01

4.1E-03

6.3E-04

2.1E-03

1.37E-04

5.0E-02

7.7E-03

coatings - spray (1-part
coatings, 2-day application)





















Commercial use - paints &

1.1E00

1.7E-01

7.5E-01

1. IE—01

5.1E-01

7.9E-02

2.6E-01

3.1E-02

5.5E-01

8.4E-02

coatings - spray (1-part
coatings, 250-day application)





















Commercial use - paints &

5.5E00

8.5E-01

3.8E00

5.7E-01

1.0E-02

1.6E-03

5.3E-03

6.3E-04

1.3E-01

1.9E-02

coatings - spray (2-part
coatings, 1-day application)





















Commercial use - paints &

5.5E00

8.5E-01

3.8E00

5.7E-01

2.1E-02

3.1E-03

1.1E-02

1.3E-03

2.5E-01

3.8E-02

coatings - spray (2-part
coatings, 2-day application)





















Commercial use - paints &

5.5E00

8.5E-01

3.8E00

5.7E-01

2.6E00

3.9E-01

1.3E00

1.6E-01

2.8E00

4.2E-01

coatings - spray (2-part
coatings, 250-day application)





















Commercial use - lab chemical

9.3E-04

5.8E-04

7.9E-04

5.1E-04

4.3E-04

2.7E-04

1.5E-04

8.8E-05

4.6E-04

2.9E-04

- use of laboratory chemicals





















Disposal

Assessed as part of each OES and not as a stand-alone OES

4235

Page 165 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

4236 Table 5-5. Summary of Inhalation Exposure Results for ONUs Based on Monitoring Data and Exposure Modeling for Each PES

OES

Inhalation Monitoring (ONU, mg/mJ)

TWA

AC

ADC

LA DC

SADC

High-End

Central
Tendency

High-End

Central
Tendency

High-End

Central
Tendency

High-End

Central
Tendency

High-End

Central
Tendency

Recycling of e-waste

1.9E-04

1.0E-07

1.3E-04

6.8E-08

8.9E-05

4.7E-08

4.5E-05

1.9E-08

9.5E-05

5.0E-08

Note that for many cases, EPA was not able to estimate inhalation exposure for ONUs, but EPA expects these to be lower than inhalation exposure for workers.

4237

Page 166 of 572

-------
4238

4239

4240

4241

4242

4243

4244

4245

4246

4247

4248

4249

4250

4251

4252

4253

4254

4255

4256

4257

4258

4259

4260

4261

4262

4263

4264

4265

4266

4267

4268

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

5.1.1.3 Summary of Dermal Exposure Assessment

Table 5-6 presents the estimated dermal acute retained dose for workers in various exposure scenarios.
The exposure estimates are provided for each OES based on the maximum possible exposure
concentration (Yderm), which is the highest concentration level of TCEP that a worker handles
throughout the process. The exposure concentration is determined based either on EPA's review of
currently available products and formulations containing TCEP or the assumption that neat TCEP is
handled to formulate these products.

The occupational dermal dose estimates assume one exposure event (applied dose) per workday and that
absorption through and into the skin may occur for up to 8 hours as representative of a typical workday.
Also, it is assumed that workers will thoroughly wash their hands with soap and water at the end of their
shifts. Regarding material remaining in the skin post-washing, EPA considers the quantity of material
remaining in the skin as potentially absorbable in accordance with OECD Guidance Document 156
(OECD. 2022). Therefore, overall occupational dermal exposure consists of the amount absorbed during
the 8-hour workday plus the amount remaining in the skin after washing the hands at the end of the 8-
hour workday.

In order to estimate occupational dermal exposures to TCEP, EPA relied on fractional absorption data
from Abdallah et al. (2016). This study used a low concentration (-0.005 wt % in acetone) of TCEP for
in vitro dermal absorption testing of a finite dose {i.e., 500 ng/cm2) over a 24-hour period. As mentioned
above, the occupational exposure estimates are based on a typical 8-hour workday. Cumulative
absorption data from Abdallah et al. C show 82.69 ng/cnr absorbed after 8 hours of exposure and
the fraction remaining in the skin is 0.068 after 24 hours of exposure. Because there were no data for the
quantity remaining in the skin after 8 hours of exposure, EPA conservatively assumed that the quantity
in the skin after 24 hours of exposure is representative of the amount remaining in the skin after 8 hours
of exposure. EPA used the cumulative absorption data to determine the fraction absorbed after an 8-hour
exposure period (0.165), and then conservatively added the fraction remaining in the skin at 24 hours
(0.068). Therefore, the overall fractional absorption from an 8-hour exposure was calculated for a dilute
solution containing TCEP as fabs = 0.165 + 0.068 = 0.233.

Page 167 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

4269	Table 5-6. Summary of Dermal Retained Dose for Workers Based on Exposure Modeling for Each

4270	OES



Max TCEP
Weight Fraction

Non-occluded Worker Dermal Retained Dose

OES

Dose (mg/day)



(Max Yderm)

High-End

Central
Tendency

Manufacture (import) - repackaging

1.0E00

6.54E00

2.18E00

Processing - incorporation into paints

1.0E00

6.54E00

2.18E00

and coatings - 1-part coatings







Processing - incorporation into paints

1.0E00

6.54E00

2.18E00

and coatings - 2-part reactive coatings







Processing - formulation of TCEP-

1.0E00

6.54E00

2.18E00

containing reactive resins (for use in 2-
part systems)







Processing - processing into 2-part
resin article

4.0E-01

2.62E00

8.73E-01

Processing - recycling e-waste

1.40E-05

4.4E-05

1.8E-05

Distribution - distribution in commerce

Distribution activities (e.g., loading) considered throughout life cycle,
rather than using a single distribution scenario

Industrial use - installation of article

N/A

N/A

N/A

Commercial use - use and/or

N/A

N/A

N/A

maintenance of aerospace equipment
and products







Commercial use - use of paints and

0.25

8.02E00

1.48E00

coatings - spray application OES







Commercial use - lab chemical - use of

1.0

6.54E00

2.18E00

laboratory chemicals







Commercial uses:

N/A

N/A

N/A

• Furnishing, cleaning,
treatment/care products
o Fabric and textile products







o Foam seating and bedding
products

• Construction, paint, electrical, and







metal products
o Building/construction
materials - insulation







o Building/construction
materials - wood and







engineered wood products -







wood resin composites







Disposal

Evaluated as part of each OES as opposed to a standalone OES

All dermal exposure scenarios are considered to be to a finite dose; therefore, no scenario is considered occluded.

4271

Page 168 of 572

-------
4272

4273

4274

4275

4276

4277

4278

4279

4280

4281

4282

4283

4284

4285

4286

4287

4288

4289

4290

4291

4292

4293

4294

4295

4296

4297

4298

4299

4300

4301

4302

4303

4304

4305

4306

4307

4308

4309

4310

4311

4312

4313

4314

4315

4316

4317

4318

4319

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

5.1.1.4 Weight of the Scientific Evidence Conclusions for Occupational Exposure

Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Supplemental Information File:
Supplemental Information on Environmental Release and Occupational Exposure Assessment (U.S.
E 231) provides a summary of EPA's overall confidence in its inhalation exposure estimates for
each of the OESs assessed.

5.1.1.4.1 Strengths, Limitations, Assumptions, and Key Sources of Uncertainty for
the Occupational Exposure Assessment

Number of Workers

Several uncertainties surround the estimated number of workers potentially exposed to TCEP. Current
CDR data reported in 2020 do not show production volumes that exceed the threshold of 25,000 pounds
and therefore, information was not available to estimate the number of workers associated with
manufacturing, processing, or use of TCEP.

There are inherent limitations to the use of CDR data as reported by manufacturers and importers of
TCEP. Manufacturers and importers are only required to report if they manufactured or imported more
than 25,000 lb of TCEP at a single site during any calendar year; as such, CDR may not capture all sites
and workers associated with any given chemical because it is possible for entities to use less than the
CDR threshold. Therefore, EPA assumes that any ongoing manufacturing, import, processing, or use of
TCEP occurs using volumes below the CDR threshold.

There are also uncertainties with BLS data, which are used to estimate the number of workers for the
remaining COUs. First, BLS' OES employment data for each industry/occupation combination are only
available at the 3-, 4-, or 5-digit NAICS level, rather than the full 6-digit NAICS level. This lack of
granularity could result in an overestimate of the number of exposed workers if some 6-digit NAICS are
included in the less granular BLS estimates but are not likely to use TCEP for the assessed applications.
EPA addressed this issue by refining the OES estimates using total employment data from the U.S.
Census' Statistics of U.S. Businesses (SUSB). However, this approach assumes that the distribution of
occupation types (SOC codes) in each 6-digit NAICS is equal to the distribution of occupation types at
the parent 5-digit NAICS level. If the distribution of workers in occupations with TCEP exposure differs
from the overall distribution of workers in each NAICS, then this approach will result in inaccuracy but
would be unlikely to systematically either overestimate or underestimate the count of exposed workers.

Second, EPA's judgments about which industries (represented by NAICS codes) and occupations
(represented by SOC codes) are associated with the uses assessed in this report are based on EPA's
understanding of how TCEP is used in each industry. Designations of which industries and occupations
have potential exposures is nevertheless subjective, and some industries/occupations with few exposures
might erroneously be included, or some industries/occupations with exposures might erroneously be
excluded. This would result in inaccuracy but would be unlikely to systematically either overestimate or
underestimate the count of exposed workers.

Analysis of Exposure Monitoring Data

This risk evaluation uses existing worker exposure monitoring data to assess exposure to TCEP during
some COUs, depending on availability of data. To analyze the exposure data, EPA categorized each data
point as either "worker" or "occupational non-user." The categorizations are based on descriptions of
worker job activity as provided in literature and EPA's judgment. In general, samples for employees that
are expected to have the highest exposure from direct handling of TCEP are categorized as "worker" and
samples for employees that are expected to have the lower exposure and do not directly handle TCEP
are categorized as "occupational non-user."

Page 169 of 572

-------
4320

4321

4322

4323

4324

4325

4326

4327

4328

4329

4330

4331

4332

4333

4334

4335

4336

4337

4338

4339

4340

4341

4342

4343

4344

4345

4346

4347

4348

4349

4350

4351

4352

4353

4354

4355

4356

4357

4358

4359

4360

4361

4362

4363

4364

4365

4366

4367

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Exposures for ONUs can vary substantially. Most data sources do not sufficiently describe the proximity
of these employees to the TCEP exposure source. As such, exposure levels for the "occupational non-
user" category will have high variability depending on the specific work activity performed. It is
possible that some employees categorized as "occupational non-user" have exposures similar to those in
the "worker" category depending on their specific work activity pattern.

Some scenarios have limited exposure monitoring data in literature, if any. Where there are few data
points available, it is unlikely the results will be representative of worker exposure across the industry.
In cases where there was no exposure monitoring data, EPA used monitoring data from similar COUs as
a surrogate. For example, EPA did not identify inhalation monitoring data for installation of aircraft and
aerospace articles based on the systematic review of literature sources. However, EPA estimated
inhalation exposures for this OES using monitoring data for TCEP exposures during furniture
manufacturing (Makinen et at.. 2009). EPA expects that inhalation exposures during furniture
manufacturing occur from handling or contacting TCEP-containing products, which is comparable to
inhalation exposures expected during installation of TCEP-containing products for aircraft or aerospace
applications. While these COUs have similar worker activities contributing to exposures, it is unknown
that the results will be fully representative of worker exposure across different COUs.

Where sufficient data were reasonably available, the 95th and 50th percentile exposure concentrations
were calculated using reasonably available data. The 95th percentile exposure concentration is intended
to represent a high-end exposure level, while the 50th percentile exposure concentration represents a
typical exposure level. The underlying distribution of the data, and the representativeness of the
reasonably available data, are not known. Where discrete data were not reasonably available, EPA used
reported statistics {i.e., 50th and 95th percentile). Since EPA could not verify these values, there is an
added level of uncertainty.

EPA calculated ADC and LADC values assuming workers and ONUs are regularly exposed during their
entire working lifetime, which likely results in an overestimate. Individuals may change jobs during
their career such that they are no longer exposed to TCEP, and actual ADC and LADC values would be
lower than the estimates presented.

The following describe additional uncertainties and simplifying assumptions associated with use of this
modeling approach for TCEP:

•	No OSHA PEL (Very Little Monitoring Data): While EPA has confidence in the models used, it
is possible that they may not account for variability of exact monitoring processes and practices
at an individual site.

•	No 2020 CDR Reporters and Only One 2016 CDR Reporter (with No Downstream Details
Provided): Assumptions of an ongoing production volume of 2,500 and 25,000 lb per site-year
could overestimate actual amount of TCEP handled at a given site, thus overestimating actual
exposures and releases. Release and exposure information using the 25,000 lb per site-year is
provided in the Engineering Supplemental file.

Modeled Dermal Exposures

The Fractional Absorption Model is used to estimate dermal exposure to TCEP in occupational settings.
The model assumes a fixed fractional absorption of the applied dose; however, fractional absorption
may be dependent on skin loading conditions. The model also assumes a single exposure event per day
based on existing framework of the EPA/OPPT 2-Hand Dermal Exposure to Liquids Model and does
not address variability in exposure duration and frequency. Additionally, the studies used to obtain the
underlying values of the quantity remaining on the skin (Qu) did not take into consideration the fact that

Page 170 of 572

-------
4368

4369

4370

4371

4372

4373

4374

4375

4376

4377

4378

4379

4380

4381

4382

4383

4384

4385

4386

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

liquid retention on the skin may vary with individuals and techniques of application on and removal
from the hands. Also, the data used were developed from three kinds of oils; therefore, the data may not
be applicable to other liquids. Based on the uncertainties described above, EPA has a moderate level of
confidence in the assessed baseline exposure (see Table 5-1).

5.1.2 Consumer Exposures	

TCEP - Consumer Exposures (Section 5.1.2):

Key Points

EPA evaluated the reasonably available information for the following consumer exposures, the key
points of which are summarized below:

•	Limited information is available on TCEP in consumer products.

0 There are no current safety data sheets.

0 Weight fraction estimates in some cases were derived from literature values that were over
20 years old and from maximum values reported in Washington State databases.

•	The highest exposure estimates were from inhalation of the roofing insulation scenario (1.42
mg/kg/d) and the wood flooring scenario (1.24 mg/kg/day). However, EPA's confidence in
these estimates is low. Of the scenarios with moderate or robust confidence, the highest
inhalation and oral exposure estimates were from the textile for children's outdoor play
structures scenario (0.0604 mg/kg/day, 0.185 mg/kg/day, respectively).

•	Inhalation is the driver for exposure to building and construction materials (e.g., roofing
insulation, acoustic ceiling) and wood flooring for adults.

•	Oral ingestion is the driver for exposure for fabric and textile products, foam seating and
bedding products, and wooden tv stands for children and infants.

5.1.2.1 Approach and Methodology	

The migration of additive flame retardants from indoor sources such as building materials, fabrics,
textiles, and wood articles (from either ongoing COUs or in service products/articles at the end of their
life cycle) appear to be a likely source of flame retardants found in indoor dust, suspended particles, and
indoor air (Dodson et al.. 2012; Weschler and Nazaroff 2010). However, the relative contribution of
different sources of TCEP in these matrices is not well characterized. For example, building insulation,
textiles, and paints and coatings that contain TCEP have differing magnitudes of emissions that depend
on a variety of differing conditions.

Modeling was conducted to estimate exposure from the identified consumer COUs. Exposures via
inhalation, oral, and dermal routes to TCEP-containing consumer products were estimated using EPA's
Consumer Exposure Model (CEM) Version 3.0 (U.S. EPA. 2019d). Figure 5-2 below displays the
embedded models within CEM 3.0.

Page 171 of 572

-------
4387

4388

4389

4390

4391

4392

4393

4394

4395

4396

4397

4398

4399

4400

4401

4402

4403

4404

4405

4406

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

i— Partition to Air

AJNHl:
Inhalation of
Airborne
Emissions from
Articles (Gas &
Particulate)

ARTICLE SOURCE

r



AJNGl:





Ingestion of





Airborne





Emissions from





Articles





(Particulates)





A_DER1: Dermal
Absorption of

Airborne
Emissions from
Articles

Aggregate

Total
Indoor Dose
by Pathway
and
Receptor

HUMAN RECEPTOR

Figure 5-2. Consumer Pathways and Routes Evaluated in this Assessment

CEM 3.0 estimates acute dose rates and chronic average daily doses for inhalation, ingestion, and
dermal exposures of consumer products and articles. CEM 3.0 gives exposure estimates for various
lifestages, including the following:

Adult

(>21 years)

Youth 2

(16-20 years)

Youth 1

(11-15 years)

Child 2

(6-10 years)

Child 1

(3-5 years)

Infant 2

(1-2 years)

Infant 1

(<1 year)

Lifetime LADD/LADC (lifetime average daily dose/lifetime average daily concentration)

Exposure inputs for these various lifestages are provided in the EPA's CEM Version 3.0 Appendices
(U.S. EPA. 2019e). CEM 3.0 acute exposures are for an exposure duration of 1 day, and chronic
exposures are for an exposure duration of 1 year. For more information on specific use patterns, and
exposure inputs for populations, please see H.4.6 (Consumer Exposure). A summary of key parameters
used for the various consumer exposures scenarios are provided in Table 5-10.

5.1.2.2 Consumer COUs and Exposure Scenarios

Page 172 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

4407 Table 5-7. Summary of Consumer CPUs, Exposure

Scenarios, and Exposure Routes

Life Cycle
Stage

Category

Subcategory

Consumer Use and Exposure
Scenario

Form(s)

Routes Evaluated

Consumer User

Oral

Inhalation

Dermal

Consumer
Use

Paints and
coatings

Paints and coatings

N/A

Liquid





Q

Vapor



Q



Mist





Q

Consumer
Use

Furnishing,
cleaning,
treatment/care
products

Fabric and textile products

Direct contact through use of
products/articles containing
TCEP

Air/Particulate



~



Dust

~



~

Article/Product
Contact/Mouthing

~



~

Consumer
Use

Furnishing,
cleaning,
treatment/care
products

Foam seating and bedding
products

Direct contact through use of
products/articles containing
TCEP

Air/Particulate



~



Dust

~



~

Article/Product
Contact/Mouthing

~



~

Consumer
Use

Construction,
paint,

electrical, and
metal products

Building/construction
materials - insulation

Direct contact through use of
building/construction materials
made containing TCEP

Air/Particulate



~



Dust

~



~

Article/Product
Contact"







Building/construction
materials - wood and
engineered wood products -
wood resin composites

Direct contact through use of
wood and wood products made
containing TCEP

Air/Particulate



~



Dust

~



~

Article/Product
Contact/Mouthing

~



~

Disposal

Wastewater,
liquid wastes,
and solid
wastes

Wastewater, liquid wastes,
and solid wastes

Direct contact through use of
products/articles containing
TCEP

Article/Product Contact





Q

Dust





Q

Air/Particulate



Q



Long-term emission/mass-
transfer through use of products
containing TCEP

Dust





Q

Air/Particulate



Q



Quantitatively assessed; Q = Qualitatively assessed
a Contact with the product is not expected (see Section 5.1.2.2.1).

4408

Page 173 of 572

-------
4409

4410

4411

4412

4413

4414

4415

4416

4417

4418

4419

4420

4421

4422

4423

4424

4425

4426

4427

4428

4429

4430

4431

4432

4433

4434

4435

4436

4437

4438

4439

4440

4441

4442

4443

4444

4445

4446

4447

4448

4449

4450

4451

4452

4453

4454

4455

4456

4457

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Paints and Coatings

Consumers are no longer able to purchase paints and coatings containing TCEP because their domestic
retail production and manufacturing has ceased. It is possible that old paint cannisters stored in
basements, crawlspaces, and/or garages may result in exposure to TCEP from off-gassing or during use
by consumers. Furthermore, the exposure to paints and coatings containing TCEP may occur via an
article scenario in which the paint and coating has already been applied. There is a higher likelihood that
older buildings may have used TCEP-containing paints and coatings when the use of TCEP in consumer
paints and coatings was more common. This dried scenario is like the acoustic ceilings/drywall scenario
that was assessed for the building/construction materials COU. The exposure scenario of dried paints
and coatings present in the indoor environment is qualitatively assessed.

Due to limited information regarding the use of paints and coatings and the uncertainties surrounding the
weight fraction, activity and use patterns, and duration of use, EPA did not quantitatively assess the use
of paints and coatings containing TCEP.

Fabric and Textile Products

In a study of the CHAMACOS cohort in California, Castorina et al. (2 indicates that TCEP levels in
dust are significantly associated with the presence of extremely worn carpets. Crowding, poor housing
quality, and lack of maintenance by landlords can result in "extremely worn" carpets, warranting
replacement. This suggests that individuals who are lower socioeconomic status may have increased
exposure to TCEP due to the inability to replace extremely worn carpets.

Ion as et al. C measured TCEP concentrations in different types (e.g., hard plastic, soft plastic and
rubber, wood and foam and textile) of childrens toys in Antwerp, Belgium. This study reported a median
TCEP concentration of 3 jug/g, mean of 10 jug/g, and maximum of 45 |ig/g of TCEP in 36 percent in 25
foam and textile products sampled. For soft plastics and rubber products, a detection frequency of 42
percent in 31 toys with a median of 5 jug/g, mean of 10 jug/g, and maximum of 65 |ig/g was reported.
For hard plastic toys, the study author reported a detection frequency of 14 percent in 50 toys with a
median of 2 jug/g, mean of 10 jug/g, and maximum of 25 jug/g. These mean concentrations correspond to
a weight fraction of 0.001 percent.

EPA searched the Ecology Washington database (WSDE. 2023) in August 2022 and retrieved various
information for fabric and textile products containing TCEP. The Ecology Washington database
sampled for fabric and textile products that are likely to be mouthed or used by children under the age of
three. The database had 67 products classified as textiles (synthetic fibers and blends), there were 2
detects at 0.01 percent and 1.3 percent. The 1.3 percent weight fraction was detected in the surface
textile of a children's mini chair. The database indicated four detects of TCEP in carpet padding and rug
mats. The weight fractions for these carpet products ranged from 0.01 to 0.02 percent.

Little additional information was found in the literature search on the percentages of TCEP in carpet
back coating. A European patent has suggested that flame retardants may be generally used in carpet
back coating at between 5 to 30 percent (Herrlich et al.. 2013).

Two scenarios were modeled for the fabric, textile, and leather products not covered elsewhere—one for
an outdoor children's play structure and one for carpet back coating. The CEM 3.0 scenario used for
both scenarios were Fabrics: curtains, rugs, wall coverings (see Table 5-9). Values of 1.3 percent for
fabric in children's play structure and 0.02 percent for the carpet back coating were selected for weight
fractions for consumer modeling as these values are believed to be more representative of products
readily available in the United States.

Page 174 of 572

-------
4458

4459

4460

4461

4462

4463

4464

4465

4466

4467

4468

4469

4470

4471

4472

4473

4474

4475

4476

4477

4478

4479

4480

4481

4482

4483

4484

4485

4486

4487

4488

4489

4490

4491

4492

4493

4494

4495

4496

4497

4498

4499

4500

4501

4502

4503

4504

4505

4506

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Foam Seating and Bedding Products

Various studies have reported the use of TCEP in furniture, automotive, and bedding foams (Maddela et
ai. 2020). In the early 2000s, Ineerowski et al. (2001) recorded TCEP in mattresses at 890 mg/kg (0.09
percent) in Germany. All et al. (2012) reported much lower concentrations of TCEP on mattresses
surfaces (0.11 jug/g) in New Zealand. Two different case reports reported the acute death of dogs (a pit
bull, a German shepherd, and a rottweiler) after chewing old automobile foams. The case studies found
significant amounts (>2 ppm) of TCEP in their stomach contents (Lehner et al.. 2010).

Fane et al. (1 has measured another flame retardant (V6) at levels of 3.63 percent in couch foam and
5.3 percent in auto foams. TCEP has been reported to be an impurity in V6 of up to 14 percent. V6 is the
dimer of TCEP, and it would be expected that TCEP would be an impurity of a V6 mixture. Hence, the
product of these two values suggests TCEP is available in couch foams at 0.51 percent and in auto
foams at 0.74 percent (Fane et al.. 2013). Although Ineerowski et al. (2001) recorded TCEP in
polyurethane soft foam at 19,800 mg/kg (1.98 percent), values from Fane et al. (1 were selected for
this furniture foam and auto foam scenarios as they were thought to be more current and representative
of the U.S. population.

For the foam toy block scenario, a weight fraction of 0.64 percent was calculated using information from
Fane et al. (1 . This was based on the knowledge of 4.6 percent of V6 in polyurethane foam with an
understanding that TCEP has been reported to be an impurity in V6 of up to 14 percent. lonas et al.
(2 reports a lower weight fraction (0.001 percent) of TCEP in 25 foam and textile toys.

Building/Construction Materials - Insulation

TCEP has been reportedly used in building materials, including wood preservations coatings, glass fiber
wallpapers, and acoustic ceilings (Maddela et al.. 2020). High TCEP concentrations in dust (94 mg/kg)
at a Swedish library were suggested to have been due the use of TCEP in the acoustic ceiling (Marklund
et al.. 2003).

Ineerowski et al. (2001) reported TCEP in polyurethane soft foam at 19,800 mg/kg (1.98 percent), and
68,000 mg/kg (6.8 percent) in acoustic ceilings. Kaiiwara et al. (2011) recorded concentrations of TCEP
in insulation boards of up to 10 ng/g in products purchased in Japan.

To assess the building/construction materials scenario, two exposure scenarios were run in CEM 3.0:
roofing insulation (under the Plastic articles - foam insulation scenario) and acoustic ceiling (under the
Drywall scenario). The weight fractions used for this modeling were 1.98 and 6.8 percent, respectively.
These exposures scenarios measured the chronic release of TCEP from the roofing insulation and
acoustic ceiling to the indoor air and indoor dust. They did not consider do-it-yourself (DIY) scenarios
of a consumer installing these articles because they are no longer commercially available.

Wood and Engineered Wood Products

A case study reported neurotoxic signs (muscular weakness) experienced by a 5-year-old child after
exposure to TCEP. It was postulated that the exposure was due to wood paneling that had been treated
with a wood preserver coating containing 3 percent TCEP. However, TCEP in dust was not quantified.
The study reported 600 mg/kg (0.06 percent) of TCEP in wood as cited in (SCHI	). lonas et al.

(7 reported a detection frequency of 25 percent in 8 wooden toys with a median of 4 |ig/g, mean of
4 jug/g, and maximum of 5 jug/g, which corresponds to a mean weight fraction of 0.0004 percent. The
products sampled in lonas et al. (2 were around 2007, with around half of the products coming from
China.

Page 175 of 572

-------
4507

4508

4509

4510

4511

4512

4513

4514

4515

4516

4517

4518

4519

4520

4521

4522

4523

4524

4525

4526

4527

4528

4529

4530

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE

December 2023

Anecdotally, TCEP concentrations have been reported to be present in imported wooden TV stands. The
photo below lists TCEP on a California Proposition 65 label on a wooden TV stand product imported to
the United States from Malaysia (Figure 5-3).

Importer: NOBLE HOUSE HOME FURNISHINGS LLC.
Production date: October 2020
Compliance level: TSCA Title VI

A WARNING: This product can expose you to chemicals
°r°e,MI Ph°SPh-' ^ are known to

8° .o ^ssZzz::™'-For more inf°r™tion

Figure 5-3. Photo of TCEP Label on Wooden Television Stand

Source: Photo by Yousuf Ahmad, U.S. EPA.

To assess the wood and engineered wood products scenario, two exposure scenarios for wood products
(exposure from wood flooring and wooden TV stand) was run in CEM 3.0 utilizing the wood articles:
hardwood floors, furniture predefined scenario with a weight fraction of 3 percent.

Wastewater, Liquid Wastes, and Solid Wastes

Consumers may be exposed to articles containing TCEP during the handling of disposal and waste. The
removal of articles in DIY renovation scenarios may lead to direct contact with articles and the dust
generated from the articles leading to consumer exposure. Due to the difficulties in quantifying
consumer disposal of products containing TCEP, consumer disposal of TCEP was not quantitatively
assessed for this risk evaluation. Section 5.1.2.2.5 discusses the qualitative assessment for consumer
disposals including the landfilling of building products and articles that contain TCEP.

5.1.2.2.1 Consumer Exposure Routes Evaluated

The COUs that were evaluated for TCEP were all articles. As such, the relevant underlying models
utilized for TCEP included those listed in Table 5-8 below.

Table 5-8. CEM

3.0 Model Codes and Descriptions

Model Code

Description

E6

Emission from article placed in environment

A INIi l

Inhalation from article placed in environment

AJNG1

Ingestion after inhalation

AJNG2

Ingestion of article mouthed

AJNG3

Incidental ingestion of dust

A DERI

Direct transfer from vapor phase to skin

A DER2

Dermal dose from article where skin contact occurs

Page 176 of 572

-------
4531

4532

4533

4534

4535

4536

4537

4538

4539

4540

4541

4542

4543

4544

4545

4546

4547

4548

4549

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Model Code

Description

ADER3

Dermal dose from skin contact with dust

CEM 3.0 contains 73 specific product and article categories and several generic categories that can be
user-defined for any product and article. Table 5-9 presents a crosswalk between the COU subcategories
with these predefined scenarios. In some cases, one COU mapped to multiple scenarios, and in other
cases one scenario mapped to multiple COUs.

Table 5-9. Crosswalk of COU Subcategories, CEM 3.0 Scenarios, and Relevant CEM 3.0 Models
Used for Consumer Modeling	

TCEP COU Subcategory

Exposure Scenario

CEM 3.0 Scenario

o\

z
X

z
O

z
O

z
O

o

O

m

Carpet back coating

Fabrics: curtains, rugs,
wall coverings

Fabric and textile products

Textile for outdoor
children's outdoor
play structures

Fabrics: curtains, rugs,
wall coverings

Foam seating and bedding
product

Foam used in
automobiles, foam
used in living room
furniture

Plastic articles: furniture
(sofa, chairs, tables)

Mattress

Plastic articles: mattresses

Other foam objects
(toy blocks)

Plastic articles: other
objects with potential for
routine contact (toys,
foam blocks, tents)

Building/construction
materials - insulation

Insulation

Plastic articles: foam
insulation

Acoustic ceiling

Drywall (acoustic ceiling)

Building/construction
materials - wood and
engineered wood products
- wood resin composites

Wood flooring

Wood articles: hardwood
floors, furniture

Wooden TV stand

Wood articles: hardwood
floors, furniture

In total, the four COUs for TCEP were mapped to nine CEM 3.0 scenarios. Relevant consumer
behavioral pattern data (i.e., use patterns) and product-specific characteristics were applied to each of
the scenarios. For more information on specific use patterns and product-specific characteristics please
see Appendix H.4.6 (Consumer Exposure).

Inhalation, oral and dermal routes were evaluated for each of the article COUs. The article model
Ingestion of article mouthed (AING2) was only evaluated for the COUs where it was anticipated that
mouthing of the product would occur. For example, it is unlikely that a child will mouth roofing
insulation or an acoustic ceiling, hence the A ING2 Model was deemed inappropriate for estimating
exposure for these COUs. The A DER2 Model (dermal dose from article where skin contact occurs)

Page 177 of 572

-------
4550

4551

4552

4553

4554

4555

4556

4557

4558

4559

4560

4561

4562

4563

4564

4565

4566

4567

4568

4569

4570

4571

4572

4573

4574

4575

4576

4577

4578

4579

4580

4581

4582

4583

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

was not used for estimating dermal exposure to roofing insulation and acoustic ceilings because dermal
contact is not expected to occur for these articles.

The chronic and lifetime exposure estimates are the most relevant durations for consumer articles.
Furnishings, building materials, and foam seating and bedding products are typically used over a longer
time frame than other types of consumer products with direct applications (e.g., household cleaners,
solvents). The exposure scenario of relevance for consumers for building and construction materials,
fabric and textile products, and foam seating and bedding products is that of a repeated exposure over a
chronic duration. As such, the exposure estimates presented in the successive sections focus on the
chronic average daily doses rather than the acute estimates. A summary of the acute, chronic, and
lifetime exposure estimates are presented in Section 5.1.2.3 and further discussed in Appendix H.4.6
(Consumer Exposure).

The CEM Version 3.0 was selected for the consumer exposure modeling as the most appropriate model
to use based on the type of input data available for TCEP-containing consumer products. The advantages
of using CEM to assess exposures to consumers and bystanders are as follows:

•	CEM model has been peer-reviewed;

•	CEM accommodates the distinct inputs available for the products containing TCEP; and

•	CEM uses the same calculation engine to compute indoor air concentrations from a source as the
higher-tier Multi-Chamber Concentration and Exposure Model (MCCEM) but does not require
measured chamber emission values (which are not available for TCEP).

Consumer modeled exposure estimates were compared to the reported monitoring and reported modeled
estimates for indoor air and indoor dust. Residential indoor air, indoor dust, and personal breathing zone
data were identified and evaluated during systematic review (	023p. v). Sections 3.4.1 and

3.4.2 provide a summary of the reported monitoring and reported modeled data in indoor air and indoor
dust. A challenge in comparing EPA modeled exposures estimates with the reported monitoring and
modeled data in the literature is that EPA's modeled exposure estimates are by COU, whereas reported
information in the literature are not typically specified by COU. For a characterization of model
sensitivity and full exposure results, see Appendix H.4.6 (Consumer Exposure).

Page 178 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

4584 Table 5-10. Summary of Key Parameters for Article Modeling in CEM 3.0"

Consumer
Exposure
Scenarios

Initial
Concentration of
SVOC in Article
(mg/cmJ)

Weight
Fraction of
Chemical
(%)

Density
Product/Article
(g/cmJ)

Duration of
Article
Contact (min)

Frequency of
Article Contact
(Events/Day)

Surface
A rea of
Article (m2)

Thickness
of Article
Surface
Layer (m)

Inter/one
Ventilation
Rate (m3/h)

Use
Environment
Volume (mJ)

Textile-
outdoor
play

structures

4.03E00

1.30

0.31

180

1

17.8608

0.055

1E-30

492

Carpet back
coating

4.00E-02

0.02

0.2

1,140

5

1.6

0.5

1E-30

492

Foam living
room

2.22E01

0.74

0.03

600

10

0.4225

0.01

88.6092

50

Foam auto

2.22E01

0.74

0.03

600

1

0.4225

0.01

9.4872

2.4

Mattresses

2.67E-02

0.09

0.03

600

1

3.097

0.5

107.01

36

Other foam
objects

1.92E-01

0.64

0.03

3.8

40

0.6606

0.01

108.978

50

Roofing
insulation

5.94E-01

1.98

0.03

0

1

158

0.5

1E-30

492

Wood
flooring

3.00E01

3.00

1

1,140

10

211

0.1

88.6092

50

Wood TV
stand

3.00E01

3.00

1

120

10

1.38

0.1

88.6092

50

Acoustic
ceiling

1.12E01

6.80

0.165

0

1

12.6

0.5

107.01

36

"For detailed information on selection of parameters refer to Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Supplemental Information File:
Consumer Exposure Modeling Inputs (U.S. EPA, 2023c).

4585

Page 179 of 572

-------
4586

4587

4588

4589

4590

4591

4592

4593

4594

4595

4596

4597

4598

4599

4600

4601

4602

4603

4604

4605

4606

4607

4608

4609

4610

4611

4612

4613

4614

4615

4616

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

5.1.2.2.2 Inhalation Exposure Assessment

Due to its vapor pressure of 0.0613 mm Hg at 25 °C, it is expected that under non-heated conditions
TCEP concentrations in air would be negligible. However, research has indicated that inhalation
exposure of TCEP can be higher than dermal exposure (Ortiz Carrizales. 2018). In addition,
concentrations of TCEP in the indoor air have been shown to be higher than ambient air concentrations
(Wong et ai. 2018). In general, concentrations of organophosphate flame retardants increase both
indoors and outdoors during warmer seasons (Wane et ai. 2019a).

Generally, TCEP release is higher at higher temperatures. However, the material to air coefficient (Kma)
values for TCEP have been shown to be similar at 35 and 55 °C. This implies that after reaching a
certain temperature, TCEP emission rates increase in a KMA-independent manner with further increase in
temperature. The Kma value at 23 °C for polyisocyanurate (PIR) foam was 7.76x 106 and for
polyurethane foam (PUF) was 3.87 106 (Maddeta et ai. 2020).

Due to its presence in particulates both less than and greater than 2.5 |im, and its presence in the gaseous
phase, EPA expects both inhalation pathways (<2.5 |im deposits in lung and <0.1 |im deposits in
alveolar region) and ingestion pathways (>2.5 |im deposits in mouth) to be contributors to TCEP
exposure. See Section 3.3.1.2.1 for more details regarding the particle vs. gas phase distribution of
TCEP. Consumer inhalation exposure to TCEP is expected through the direct inhalation of indoor air
and dust. Table 5-11 below illustrates the steady state SVOC concentrations and respirable particle (RP)
concentrations resulting from consumer exposure to articles containing TCEP.

Table 5-11. Steady State Air Concentrations and Respirable Particle of TCEP from Consumer

Modeling (CEM 3.0)

COU Subcategory

Consumer Scenario

Air SVOC
(mg/m3)

Respirable
Particles
(ng/mg)

Fabric and textile products

Carpet back coating

3.06E-02

3.79E-02

Textile-outdoor play
structures

3.96E00

4.80E00

Foam seating and bedding product

Foam auto

1.04E-04

2.43E-05

Foam living room

9.33E-06

3.33E-06

Mattresses

4.45E-04

1.33E-04

Other foam objects

1.26E-05

4.50E-06

Building/construction materials - insulation

Roofing insulation

9.32E00

1.13E01

Acoustic ceiling

7.52E-01

2.25E-01

Building/construction materials - wood and
engineered wood products - wood resin composites

Wood flooring

8.11E00

3.30E00

Wood TV stand

5.31E-02

2.16E-02

The insulation scenario followed by the wood-resin scenario had the highest TCEP air concentrations
(9.32 and 8.11 mg/m3 respectively).

Exposures doses (chronic average daily inhalation doses [CADDs]) for all of the COU subcategories
were estimated for the inhalation pathway via the following formulae) (A INHl):

Page 180 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

4617

Equation 5-1



4618





4619



Caas x FracTime x InhalAfter x CF1

CAD D Air = ava	

Air BW x CF2

4620

Equation 5-2



4621





4622

CAD Dparticuiate

SVOCRPairavg x RPairavg x (1 — IFRP)FracTime x InhalAfter x CFX

BW x CF2

4623





4624

Equation 5-3



4625



CADDtotal — CADDAir + CAD DpartiCUiate

4626





4627

Where:



4628

CADDAir

= Potential Chronic Average Daily Dose, air (mg/kg-day)

4629

CADDParticuiate = Potential Chronic Average Daily Dose, particulate (mg/kg-day)

4630

CADDtotai

= Potential Chronic Average Daily Dose, total (mg/kg-day)

4631

r

°gas_avg

= Average gas phase concentration (|ig/m3)

4632

SVOCRPairavg

= Average SVOC in RP concentration, air (|ig/mg)

4633

RPair avg

= Average RP concentration, air (mg/m3)

4634

IFrp

= RP ingestion fraction (unitless)

4635

FracTime

= Fraction of time in environment (unitless)

4636

InhalAfter

= Inhalation rate after use (m3/hr)

4637

CFi

= Conversion factor (24 hr/day)

4638

BW

= Body weight (kg)

4639

cf2

= Conversion factor (1,000 |ig/mg)

4640





4641

Exposures doses (Acute Dose rate ADRs) for all of the COU subcategories were estimated for the

4642

inhalation pathway via

the following formulae (A INH1):

4643





4644

Equation 5-4



4645





4646



Cgas max x FracTime x InhalAfter x CFX
ADRait ~ BW x CF2

4647





4648

Equation 5-5



4649





4650

SVOCRPair max x RPair avg x FracTime x InhalAfter x CFt
ADRPartiCUiate — BW x CF

4651





4652

Equation 5-6



4653





4654



ADRfotal ~ ADRAir + ADRparticulate

4655





4656

Where:



4657

ADRAir

= Potential Acute Dose Rate, air (mg/kg-day)

4658

AD Rp articulate

= Potential Acute Dose Rate, particulate (mg/kg-day)

4659

ADRt0tai

= Potential Acute Dose Rate, total (mg/kg-day)

Page 181 of 572

-------
4660

4661

4662

4663

4664

4665

4666

4667

4668

4669

4670

4671

4672

4673

4674

4675

4676

4677

4678

4679

4680

4681

4682

4683

4684

4685

4686

4687

4688

4689

4690

4691

4692

4693

4694

4695

4696

4697

4698

4699

4700

4701

4702

4703

4704

4705

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Cgas max	=	Maximum gas phase concentration (|ig/m3)

SVOCRPair max	=	Maximum SVOC in RP concentration, air (|ig/mg)

RPair_max	=	Maximum RP concentration, air (mg/m3)

FracTime	=	Fraction of time in environment (unitless)

InhalAfter	=	Inhalation rate after use (m3/hr)

CF1	=	Conversion factor (24 hr/day)

BW	=	Body weight (kg)

CF2	=	Conversion factor (1,000 |ig/mg)

The ADR and CADD equations (Equation 5-1, Equation 5-2, Equation 5-3, Equation 5-4,

Equation 5-5, and

Equation 5-6) for A INHl consider both contributions from air and particulates. The average gas phase
concentration is considered for CADDair, and the maximum gas phase concentration is considered for
ADRair. The average SVOC in the RP concentration is considered for CADDparticulate, and the
maximum SVOC in the RP concentration is considered for ADRparticulate. CADDair and
CADDparticulate are summed to obtain CADDtotal. Similarly, ADRair and ADRparticulate are
summed to get ADRtotal. The SVOC in the RP concentration is given in |ig/mg and is multiplied by an
average RP concentration (in mg/m3).

Although the inhalation exposures to consumer articles containing TCEP are dominated by gas phase air
concentrations versus the SVOC in RP concentrations, EPA decided to include both in the inhalation
exposure estimates. Therefore, EPA presented consumer inhalation values as doses (mg/kg-day), rather
than concentrations (mg/m3), because the dose values expressed as mg/kg-day included contributions
from both the gas and particulate phases.

CEM 3.0 outputs include inhalation doses for all lifestages. Inhalation doses are calculated for lifestages
by varying the BW and inhalation rate for the various population groups. These inhalation dose
calculations are simplified and do not take into consideration lifestages differences in ventilation,
anatomy, and metabolism. This risk evaluation presents one inhalation value (the adult value) by COU
(see Table 5-15 and Table 5-16). Appendix 1.1.1 presents the reported CEM inhalation doses with
breathing weight and body weight adjustments for all lifestages.

A summary of the acute, chronic, and lifetime inhalation doses are presented in Section 5.1.2.3. Table
5-10 presents a summary of the key parameters used for consumer modeling with CEM 3.0. For more
information on CEM 3.0, input parameters, sensitivity analysis, and assumptions used for consumer
modeling please see Appendix I.

5.1.2.2.3 Dermal Exposure Assessment

Consumers may be dermally exposed to TCEP via skin contact with consumer articles, skin contact with
dust generated from consumer articles, or the deposition of vapor generated from articles onto the skin.
CEM 3.0 contains dermal modeling components that estimate absorbed dermal doses resulting from
dermal contact with chemicals found in consumer products: Direct transfer from vapor phase to skin
(A DERl), dermal dose from article where skin contact occurs (A DER2), and dermal dose from skin
contact with dust (A DER3). All three models were used to estimate exposure to articles containing
TCEP, except for A DER2, which was not used for the Building/construction materials - insulation
COU because direct article contact with skin was not expected.

Page 182 of 572

-------
4706

4707

4708

4709

4710

4711

4712

4713

4714

4715

4716

4717

4718

4719

4720

4721

4722

4723

4724

4725

4726

4727

4728

4729

4730

4731

4732

4733

4734

4735

4736

4737

4738

4739

4740

4741

4742

4743

4744

4745

4746

4747

4748

4749

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Contact of skin with articles drives the dermal exposure estimate in cases where contact is expected.
Otherwise, skin contact with dust is the driver of dermal exposure. The following equation was used to
calculate CADD for A DER2:

Equation 5-7

Where:

CADD

Cart	=

SA/BW
FRabsart =

EDcr	=

ATcr	=

L

Many of these parameters are calculated within CEM. The parameter / is a function of duration of article
contact (min/day). A DER3 has a similar formula:

Equation 5-8

Where:

Dustcrwg

AF
FA
EvD

CFi

Compared to ADER2, this formula substitutes a chronic weighted dust concentration for the chemical
concentration and replaces the / term with an adherence factor {AF) and frequency of article contact
(EvD).

A key parameter in estimating results for A DER2 and A DER3 is fraction absorbed (Fabs). While the
duration of interaction with materials that contain TCEP may be shorter than the duration that was tested
in the dermal absorption study (i.e., a 24-hour exposure), EPA cannot assume that consumers would
immediately wash their hands following contact with treated objects (e.g., carpets). Therefore, the dose
that is deposited on the skin during an activity would be expected to remain on the skin until the skin is
eventually washed. As a result, EPA applied a 24-hour value for fraction absorbed (35.1 percent) from
Abdallah et al. (2016) to all consumer dermal exposures scenarios.

Table 5-12 provides the chronic dermal doses from each of the underlying models in CEM 3.0 and for
adults and children 3-6 years of age. All life-stages were analyzed. For more information on the
consumer dermal exposure inputs, equations, results (for all life-stages) and sensitivity analysis please
see Appendix I and EPA's CEM 3.0 Appendices (	g).

SA

Cart * rTTT * ^ * F^abs art * EDcr

CADD =	—	—	-	

ATcr

Potential Chronic Average Daily Dose (mg/kg-day)

Chemical concentration in article (mg/cm3)

Surface area to body weight ratio (cm2/kg)

Fraction absorbed (unitless)

Exposure duration, chronic (years)

Averaging time, chronic (years)

Average distance a diffusing molecule travels per contact (cm/day)

Si4

Dustcr WQt x -pTTT x AF x FAX EvD x EDcr

CADD =	-	——	—	

CFX x ATcr

Chronic weighted dust concentration (p,g/mg)
Adherence factor of dust to hand (mg/cm2-event)
Fraction absorbed (unitless)

Frequency of article contact per day (events/day)
Conversion factor (insert value)

Page 183 of 572

-------
4750

4751

4752

4753

4754

4755

4756

4757

4758

4759

4760

4761

4762

4763

4764

4765

4766

4767

4768

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table 5-12. Chronic Dermal Average Daily Doses (mg/kg-day) of TCEP from Consumer Article
Modeling for Adults and Children 3 to 6 Years of Age (CEM 3.0) 		

COU Subcategory

Consumer
Scenario

Life
Stage

A DERI
Vapor to
Skin

ADER2

Skin
Contact

ADER3
Skin Contact
with Dust

Total
Chronic
Dermal ADD

Fabric and textile
products

Carpet back
coating

Adult

2.29E-07

3.16E-04

8.60E-06

3.25E-04

Child

3.68E-07

5.07E-04

5.53E-05

5.63E-04

Textile-outdoor
play structures

Adult

2.97E-06

1.26E-02

2.10E-04

1.29E-02

Child

4.77E-06

2.03E-02

1.35E-03

2.17E-02

Foam seating and
bedding product

Foam auto

Adult

3.87E-10

5.65E-03

4.44E-09

5.65E-03

Child

6.43E-10

9.38E-03

2.95E-08

9.38E-03

Foam living
room

Adult

6.95E-10

1.26E-02

5.40E-09

1.26E-02

Child

1.15E-09

2.10E-02

3.59E-08

2.10E-02

Mattresses

Adult

1.33E-07

6.14E-03

3.99E-07

6.14E-03

Child

2.20E-07

1.02E-02

2.65E-06

1.02E-02

Other foam
objects

Adult

2.41E-10

2.23E-04

7.40E-09

2.23E-04

Child

4.19E-10

3.87E-04

5.15E-08

3.88E-04

Building/construction
materials - insulation

Roofing
insulation

Adult

3.49E-05

0

2.50E-04

2.84E-04

Child

5.61E-05

0

1.61E-03

1.66E-03

Acoustic ceiling

Adult

2.81E-06

0

8.48E-06

1.13E-05

Child

4.53E-06

0

5.45E-05

5.91E-05

Building/construction
materials - wood and
engineered wood
products - wood
resin composites

Wood flooring

Adult

6.08E-05

2.37E-01

1.33E-03

2.38E-01

Child

9.76E-05

3.80E-01

8.55E-03

3.89E-01

Wood TV stand

Adult

3.98E-07

7.68E-02

8.71E-06

7.68E-02

Child

6.38E-07

1.23E-01

5.59E-05

1.23E-01

5.1.2.2.4 Oral Exposure Assessment

Consumers may be exposed to TCEP via transfer from hand to mouth, ingestion after inhalation, mouthing of
articles, and the incidental ingestion of dust generated from consumer articles. CEM 3.0 contains an
ingestion modeling component that estimates ingestion doses resulting from consumer products:
ingestion after inhalation (AING1), ingestion of article mouthed (AING2), and incidental ingestion
from dust (A ING3). All three models were used to estimate exposure to articles containing TCEP,
except for A ING2, which was not used for the building/construction materials COU as mouthing of the
article was not expected.

Mouthing is a particular important route for estimating exposure to children and infants who may have
higher exposures to toys and children's products. CEM 3.0 has four choices for mouthing scenarios: 0, 1
(low), 10 (medium), and 50 (high) cm2. The high mouthing input was selected for outdoor play
structures and other foams (toy blocks) because these are children's products. The medium values were
selected for carpet back coating, wood flooring, wooden TV stand, foam furniture in the living room,
foam seat in an automobile, and the mattress scenarios.

The following equation was used to calculate CADD for A ING2:

Page 184 of 572

-------
4769

4770

4771

4772

4773

4774

4775

4776

4777

4778

4779

4780

4781

4782

4783

4784

4785

4786

4787

4788

4789

4790

4791

4792

4793

4794

4795

4796

4797

4798

4799

4800

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Equation 5-9

Where:

CADD

MR

CA

Dm

EDcr

CFl

ATcr

BW

CADD =

MR x CAx Dmx EDcr x CF1
BW x ATcr x CF2

Potential Chronic Average Daily Dose (mg/kg-day)

Migration rate of chemical from article to saliva (mg/cm2/hr)

SA/BW= Surface area to body weight ratio (cm2/kg)

Duration of mouthing (min/hr)

Exposure duration, chronic (years)

Conversion factor (24 hr/day)

Averaging time, chronic (years)

Body weight (kg) = Conversion factor (60 min/hr)

The following equation was used to calculate CADD for A ING3:

Equation 5-10

Where:

CADD

Dustcrwgt

FracTime

Dusting

BW

CF

CADD =

Dustcr wgt x FracTime x Dusting

BW x CF

Potential Chronic Average Daily Dose (mg/kg-day)

Chronic weighted dust concentration ([j,g/mg)

Fraction of time in environment (unitless)

Dust ingestion rate (mg/day)

Body weight (kg)

Conversion factor (1,000 (J,g/mg)

The chronic weighted dust concentration was weighted between the dust available from the respirable
portion, floor dust, and abraded particles.

Table 5-13 presents the chronic ingestion doses from each of the underlying models in CEM 3.0 and for
adults and infants 1 to 2 years of age. All life-stages were analyzed. For more information on the
consumer dermal exposure inputs, equations, results (for all life-stages) and sensitivity analysis please
see Appendix I and EPA's CEM 3.0 Appendices (	2).

Page 185 of 572

-------
4801

4802

4803

4804

4805

4806

4807

4808

4809

4810

4811

4812

4813

4814

4815

4816

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table 5-13. Chronic Ingestion Average Daily Doses (mg/kg-day) of TCEP from Consumer Article

Modeling for Adults and Infants 1 to 2 Years of Age (CEM 3.0



COU Subcategory

Consumer Scenario

Life
Stage

AING1
Ingestion

after
Inhalation

AING2
Mouthing

A ING3
Ingestion
of Dust

Total Chronic
Ingestion
ADD

Fabric and textile
products

Carpet back coating

Adult

3.44E-08

0

2.47E-05

2.47E-05

Infant

1.25E-07

2.22E-01

3.14E-04

2.22E-01

Textile-outdoor play
structures

Adult

4.13E-06

0

3.02E-04

3.06E-04

Infant

1.50E-05

2.22E-01

3.83E-03

2.26E-01

Foam seating and
bedding product

Foam auto

Adult

6.66E-10

0

3.22E-10

9.88E-10

Infant

2.43E-09

2.22E-01

4.09E-09

2.22E-01

Foam living room

Adult

7.55E-12

0

7.83E-10

7.91E-10

Infant

2.75E-11

2.22E-01

9.94E-09

2.22E-01

Mattresses

Adult

6.70E-10

0

1.45E-07

1.46E-07

Infant

2.44E-09

2.70E-01

1.84E-06

2.70E-01

Other foam objects

Adult

9.69E-12

0

1.05E-09

1.06E-09

Infant

3.53E-11

1.11E00

1.33E-08

1.11E00

Building/construction
materials - insulation

Roofing insulation

Adult

9.82E-06

0

7.19E-03

7.20E-03

Infant

3.58E-05

0

9.13E-02

9.13E-02

Acoustic ceiling

Adult

1.12E-06

0

2.44E-04

2.45E-04

Infant

4.07E-06

0

3.10E-03

3.10E-03

Building/construction
materials - wood and
engineered wood
products - wood resin
composites

Wood flooring

Adult

9.21E-06

0

1.91E-03

1.92E-03

Infant

3.36E-05

2.22E-01

2.43E-02

2.46E-01

Wood TV stand

Adult

6.03E-08

0

1.25E-05

1.26E-05

Infant

2.20E-07

2.22E-01

1.59E-04

2.22E-01

For children and infants, mouthing was the dominant route of exposure. For teenagers and adults,
ingestion of dust was the dominant route of exposure as no mouthing of the consumer articles are
expected.

Sensitivity analyses indicated that "Area of article mouthed" was the driver for the mouthing estimates.
The area of article mouthed was more important for the ingestion estimate compared to the initial
concentration of the SVOC in the article, the density of the article, the surface area of the article, and the
duration of article contact.

For more information on the consumer ingestion exposure inputs, equations, results (for all life-stages)
and sensitivity analysis please see Appendix I and EPA's CEM Version 3.0 User Guide and Appendices
(I	1022a).

Page 186 of 572

-------
4817

4818

4819

4820

4821

4822

4823

4824

4825

4826

4827

4828

4829

4830

4831

4832

4833

4834

4835

4836

4837

4838

4839

4840

4841

4842

4843

4844

4845

4846

4847

4848

4849

4850

4851

4852

4853

4854

4855

4856

4857

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

5.1.2.2.5 Qualitative Exposure Assessment
Paints and Coatings

A review of literature reporting TCEP used outside the US from the early 2000s provides some evidence
of the use of TCEP in paints and coatings. Ineerowski et al. (2001) detected TCEP in 85 percent of 983
household products in Germany and reported TCEP in wood preservation coatings at a concentration of
10,000 mg/kg (1.0%). Haumann and Thumulla (2002) detected TCEP in paints at a maximum of 840
mg/kg (0.084 percent) in Germany prior to 2002 (TERA. 2013).

Table 5-14 is a summary of the information gathered for the commercial use of paints and coatings
COU. This data indicate TCEP is used at a high-end of 25 percent in commercial paints and coatings.

Table 5-14. Summary of Commercial Paints and Coatings Concentrations and Density of TCEP

Paint Products

TCEP Concentration
(Mass Fraction)

Product Density
(kg/m3)

Low-End

High-End

Low-End

High-End

7 Industrial and commercial paints and
coatings

0.1%

25%

1,000

1,490

Consumer exposures to articles that have been coated with TCEP-containing paints and coatings will
mimic consumer exposures from the article scenarios (e.g., acoustic ceilings, wood resin products). The
paints and coatings scenario within CEM 3.0 is for the active application of paints and coatings in a
product scenario. Thus, for this risk evaluation, the dried paints and coatings scenario can be considered
a part of the quantitatively assessed articles scenarios.

The maximum weight fractions (25 percent) presented in Table 5-14. are up to 4 times higher than the
weight fractions available for consumer articles (6.8 percent). This suggests that commercial and
industrial products contain higher levels of TCEP than products and articles available for the consumer
market. With the increasing availability of commercial and industrial products sold on the internet and
the increase in DIY trends, consumers potentially could obtain paints and coatings that contain TCEP at
concentrations applicable to commercial uses.

The dermal route is the most important route to consider for exposures to paints and coatings containing
TCEP. The occupational dermal exposure estimates for workers using TCEP-containing paints and
coatings are presented in Section 5.1.1.3. The commercial use of paints and coatings results in a high-
end exposure estimate of 8.02 mg/day and a central tendency estimate of 1.48 mg/day (see Table 5-6).
This scenario is based on a spray application scenario under working conditions for non-occluded
scenarios.

Differences in the occupational and consumer exposure scenarios of paints and coatings provide context
to this qualitative assessment. Products available for the industrial and commercial market are
formulated differently than for consumers. Moreover, workers work with industrial grade formulations
that have higher concentrations of TCEP and may be exposed to paints and coatings containing TCEP
under exposures scenarios that result in higher exposures (e.g., spray application vs. typical domestic
painting).

Page 187 of 572

-------
4858

4859

4860

4861

4862

4863

4864

4865

4866

4867

4868

4869

4870

4871

4872

4873

4874

4875

4876

4877

4878

4879

4880

4881

4882

4883

4884

4885

4886

4887

4888

4889

4890

4891

4892

4893

4894

4895

4896

4897

4898

4899

4900

4901

4902

4903

4904

4905

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Wastewater, Liquid Wastes, and Solid Wastes

At the end of their life cycles, consumer articles may be disposed of in municipal solid waste landfills,
construction, and demolition landfills, or undergo incineration. Groundwater monitoring data in Section
3.3.3.5 suggests that TCEP can migrate from municipal unlined landfills to groundwater via landfill
leachate. Water discharges from laundered clothing that picks up TCEP may also be a potential source
of TCEP to surface waters. The successive sections attempt to describe TCEP exposures that may be a
result of the disposal, demolition and removal of household articles and dust containing TCEP. Due to
the difficulties in source attribution, EPA was unable to relate consumer COUs to these TCEP
exposures. However, they are qualitatively discussed to capture additional ways individuals may be
exposed to TCEP via consumer articles.

Wastewater: Section 3.3.2.7 states that laundry wastewater may contribute to elevated environmental
surface water concentrations of TCEP. Clothing has been hypothesized to act as a sink for TCEP to
transfer organophosphate esters from the indoor environment to surface waters via wastewater from
domestic and commercial laundry sources (Schreder and La Guardia. 2014). A study investigating the
relationship between the fate of phthalates and flame retardants transferring from clothing to laundry
wastewater found that chemicals with a log Kow less than 4 showed a greater than 80 percent release to
laundry water, whereas chemicals with a log Kow greater than 6 only showed less than 10 percent
release to laundry wastewater (Saini et ai. 2016). Furthermore, these findings also suggest that dermal
exposure to TCEP may be enhanced from clothing to sweat (Saini et ai. 2016).

TCEP was among the 10 most frequently found compounds, detected at 61.9 percent in wastewater
samples (maximum of 0.7 |ig/L), in a study that collected wastewater from multiple sites in Research
Triangle Park area of North Carolina between 2002 and 2005 (Giorgino et ai. 2007). Flame retardants
were measured primarily at sites downstream from municipal wastewater discharges and at a site
downstream from an industrial fire. TCEP samples were detected in four of eight sites, and at three of
three sites that had major upstream wastewater discharges. A possible explanation for TCEP detection at
the one other site (without an upstream wastewater discharge) was that a fire at an industrial cleaning-
supply warehouse occurred upstream a few months before the sampling event. It is believed that water
applied to control the fire had entered the nearby tributary. In addition, two of these sites near
wastewater discharges are also located near state recreation areas where public facilities, campgrounds,
dump stations, swimming beaches and boating access are available (Gioreino et ai. 2007).

Solid Wastes: A CDC NIOSH report evaluated the occupational exposure to flame retardants at four
gymnastics studios in the mid-2010s. The researchers sampled old foam blocks, mats, padded equipment
and employees via hand wipe samples before and after work. TCEP was detected at 343 ng/ft2 at one of
the gymnastics studios in June 2014, but was not detected in April 2015 after the replacement of new
foam blocks (Broadwater et ai. 2017). A similar study measured 1.6 to 1.9 ju.g/g dry weight of TCEP in
polyurethane foam blocks in a Seattle gym. TCEP was detected at a mean concentration of 1.18 |ig/g dry
weight in gym dust concentrations across four gyms. Dust samples were collected from the homes of
four gym instructors. TCEP was found at a mean concentration of 2.5 |ig/g dry weight at the instructors'
residences (La Guardia and Hale.: ).

A study from the Sierra Nevada foothills suggests that the presence of TCEP on the surfaces of
ponderosa pine needles can be explained by the aerial transport and deposition from nearby point
sources where chemicals were released during the incineration of plastic waste articles (Aston et ai.

1996).

Page 188 of 572

-------
4906

4907

4908

4909

4910

4911

4912

4913

4914

4915

4916

4917

4918

4919

4920

4921

4922

4923

4924

4925

4926

4927

4928

4929

4930

4931

4932

4933

4934

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Recycling: TCEP is not typically used in electronics but is predominantly found in polyurethane foam
(PUF) (Stapleton et al. 2011). A CDC NIOSH report assessed employee exposure to flame retardants at
an electronics recycler in November 2016 and February 2017. TCEP was detected in surface wipe
samples at the disassembly workstation at 154 ng/100 cm2. The report indicated the workers were
incorrectly wearing N95 respirators and were dry sweeping. To prevent exposure to airborne TCEP dust
particles, the report recommends prohibiting dry sweeping to clean work areas (Grimes et al.. 2019).

Landfills: The demolition and removal of consumer articles may result in exposures to TCEP.
Construction waste and old consumer products can be disposed of in municipal solid waste landfills and
construction and demolition landfills. Section 3.3.3.7 models the resulting groundwater concentration
that may occur from leaching of TCEP from landfills. Section 3.3.3.5 highlights suspected leaching of
TCEP from nearby landfills (Norman Landfill, Himco Dump, and Fort Devens) (Buszka et al.. 2009;
Barnes et al.. 2004; Hutchins et al.. 1984). The Himco Dump is a closed unlicensed landfill that included
a 4-acre construction debris area. EPA issued a notice in the Federal Register finalizing the deletion of
part of the Himco Dump Superfund site from the National Priorities List (NPL). The Indiana
Department of Environmental Management (IDEM) formally concurred with EPA's proposal on
January 26, 2022, and	jposed the site for partial deletion in March, 2022. Fort Devens is also an

EPA, superfund site, a former army instillation site that was established in 1917 and closed in 1996, is
also a closed superfund sites. TCEP was detected throughout the entire length of a leachate plume near a
municipal landfill (subtitle D) near Norman, Oklahoma (Barnes et al.. 2004). Leachate samples from
landfill sites in Japan detected TCEP at ranges from 4.1 to 5430 mg/mL This study suggested that the
origin may be due to plastic wastes (Yasuhara. 1995).

Without a full characterization of non-hazardous landfill (e.g., Norman Landfill) conditions and
historical wastes (e.g., Himco dump and Ft. Devens) around the country, EPA is uncertain how often
contaminant migration occurs given modern practices of non-hazardous landfill and historical site
management. However, the possibility of exposure to TCEP after the release from disposal of consumer
wastes exists.

5.1.2.3 Summary of Consumer Exposure Assessment

Page 189 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table 5-15. Summary of Acute Daily Rate of Consumer Artie

es Modeled with CEM 3.0

COU Sub-category

Consumer Exposure
Scenario

Life-Stage

Exposure Dose (mg/kg/day)

Oral

Inhalation

Dermal

Fabric and textile products

Carpet back coating

Adult

2.43E-04

5.11E-02

4.03E-04

Children

1.84E-01

N/A

1.05E-03

Textile for children's
outdoor play structures

Adult

3.84E-03

1.06E00

1.53E-02

Children

2.35E-01

N/A

3.73E-02

Foam seating and bedding product

Foam automobile

Adult

3.01E-07

2.89E-04

5.65E-03

Children

1.81E-01

N/A

9.39E-03

Foam living room

Adult

1.86E-07

5.19E-04

1.26E-02

Children

1.81E-01

N/A

2.10E-02

Mattress

Adult

3.50E-06

3.15E-03

6.16E-03

Children

4.95E-02

N/A

1.03E-02

Foam - other (toy block)

Adult

2.47E-07

7.02E-04

2.24E-04

Children

9.03E-01

N/A

4.00E-04

Building/construction materials -
insulation

Roofing insulation

Adult

8.87E-02

2.32E01

3.64E-03

Children

1.27E00

N/A

2.07E-02

Acoustic ceiling

Adult

5.92E-03

5.31E00

3.35E-04

Children

8.45E-02

N/A

1.52E-03

Building/construction materials -
wood and engineered wood products
wood resin composites

Wood flooring

Adult

1.42E-01

2.21E02

3.46E-01

Children

2.21E00

N/A

1.03E00

Wooden TV stand

Adult

9.32E-04

1.45E00

7.75E-02

Children

1.94E-01

N/A

1.28E-01

4936

4937

Page 190 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table 5-16. Summary of Chronic Average Daily Doses of Consumer Articles IV

odeled with CEM 3.0

COU Sub-category

Consumer Exposure
Scenario

Life-Stage

Exposure Dose (mg/kg/day)

Oral

Inhalation

Dermal

Fabric and textile products

Carpet back coating

Adult

2.48E-05

4.66E-03

3.25E-04

Children

1.8 IE—01

N/A

5.63E-04

Textile for outdoor
children s outdoor play
structures

Adult

3.06E-04

6.04E-02

1.29E-02

Children

1.85E-01

N/A

2.17E-02

Foam Seating and Bedding Product

Foam automobile

Adult

9.88E-10

7.94E-07

5.65E-03

Children

1.81E-01

N/A

9.38E-03

Foam living room

Adult

7.90E-10

1.42E-06

1.26E-02

Children

1.81E-01

N/A

2.10E-02

Mattress

Adult

1.45E-07

6.79E-05

6.14E-03

Children

4.95E-02

N/A

1.02E-02

Foam-other (toy block)

Adult

1.05E-09

1.92E-06

2.23E-04

Children

9.03E-01

N/A

3.88E-04

Building/construction materials -
insulation

Roofing insulation

Adult

7.20E-03

1.42E00

2.84E-04

Children

1.03E-01

N/A

1.66E-03

Acoustic ceiling

Adult

2.45E-04

1.15E-01

1.13E-05

Children

3.50E-03

N/A

5.91E-05

Building/construction materials -
wood and engineered wood products
wood resin composites

Wood flooring

Adult

1.92E-03

1.24E00

2.38E-01

Children

2.08E-01

N/A

3.89E-01

Wooden TV stand

Adult

1.26E-05

8.09E-03

7.68E-02

Children

1.81E-01

N/A

1.23E-01

4939

Page 191 of 572

-------
4940

4941

4942

4943

4944

4945

4946

4947

4948

4949

4950

4951

4952

4953

4954

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table 5-17. Summary of Lifetime Average Daily Doses of Consumer Articles Modeled with CEM

3.0

COU Sub-category

Consumer Exposure
Scenario

Exposure Dose (mg/

ig/day)

Oral

Inhalation

Dermal

Fabric and textile products

Carpet back coating

2.02E-02

6.03E-03

1.56E-05

Textile for outdoor
children's outdoor play
structures

0

0

0

Foam seating and bedding
product

Foam automobile

2.01E-02

1.03E-06

7.62E-05

Foam living room

2.01E-02

1.84E-06

1.70E-04

Mattress

1.73E-02

8.78E-05

8.34E-05

Foam - other (toy block)

0

0

0

Building/construction materials -
insulation

Roofing insulation

1.72E-02

1.84E00

3.31E-04

Acoustic ceiling

5.84E-04

1.48E-01

1.13E-05

Building/construction materials -
wood and engineered wood
products - wood resin composites

Wood flooring

2.47E-02

1.60E00

4.90E-03

Wooden TV stand

2.01E-02

1.05E-02

1.03E-03

5.1.2.4 Weight of the Scientific Evidence Confidence for Consumer Exposure

The overall exposure confidence for the various consumer scenarios ranged from slight to moderate.
Low confidence in the exposure estimates were mainly due to data uncertainties. Information on article
weight fraction was sparse, and it was unclear whether many of the literature values were still relevant
for articles used today. EPA considered a worst-case approach to consumer weight fraction and varied
this parameter in the sensitivity analysis as reported in Appendix H.4.6 (Consumer Exposure).
Information on exposure scenarios (e.g., mouthing durations, use durations, frequency of contacts per
day) were also limited. Furthermore, limited monitoring data were available to corroborate the modeled
consumer exposure estimates and validate current use of TCEP in consumer articles. In addition, there
are uncertainties related to CEM 3.0 modeling approaches (e.g., deterministic vs. stochastic approaches,
background concentrations, assumptions for dermal absorption parameters).

Page 192 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

4955 Table 5-18. Weight of the Scientific Evidence Confidence for Chronic Consumer Exposure Modeling Scenarios

Consumer
Condition of Use

Confidence

Confidence
in Model
Default

Values''

Confidence in User-Selected Varied Inputs'

Monitoring
Data

Overall

Category

Subcategory

Form

in Model

Used8

Density
Usedrf

Use
Duration1'

Weight
Fraction'

Room of

Use®

Dermal
Kp, Fabs,
Mouthing'1

Exposure
Confidence'



Carpet back

Article

++

+++

++

+++

++

+++

+

Limited

Moderate



coating





















Fabric and
textile

Textile for
outdoor

Article

+++

+

++

++

++

++

++

Limited

Moderate

products

children's
outdoor play
structures





















Building/
construction

Roofing
insulation

Article

++

++

+

N/A

+

+++

+

None

Slight

materials -
insulation

Acoustic
ceiling

Article

+

++

+

N/A

+

++

+

Limited

Slight



Foam

Article

+++

+++

++

++

++

+++

+

Limited

Moderate



automobile





















Foam
seating and
bedding
product

Foam living
room

Article

+++

+++

++

+++

++

+++

++

Limited

Moderate

Mattress

Article

+++

+++

++

+++

+

+++

+

None

Slight

Foam-other
(toy block)

Article

+++

+++

++

++

+

+++

++

None

Slight

Building/

Wood flooring

Article

+++

+++

++

+++

+

+++

+

None

Slight

construction
materials -
wood and

Wooden TV
stand

Article

+++

+++

++

++

+

+++

+

Limited

Moderate

engineered
wood























products -
wood resin























composites























a Confidence in Model Used considers whether model has been peer reviewed, as well as whether it is being applied in a manner appropriate to its design and

objective. The model used (CEM 3.0) has been peer reviewed, is publicly available, and has been applied in a manner intended, to exposures associated with

Page 193 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Consumer
Condition of Use

Category

Subcategory

Form

Confidence
in Model

Used"

Confidence
in Model
Default

Values''

Confidence in User-Selected Varied Inputs'

Density
Used'

Use
Duration1'

Weight
Fraction'

Room of

Use®

Dermal
Kp, Fabs,
Mouthing'1

Monitoring
Data

Overall
Exposure
Confidence'

4956

4957

uses of household products and/or articles. Medium was selected for the carpet-back coating scenario and a roofing insulation scenario because of uncertainties
surrounding the barrier layers. Low was selected for acoustic ceiling because the related CEM scenario was Drywall, and these products have different product
characteristics.

b Confidence in Model Default Values considers default value data source(s) such as building and room volumes, interzonal ventilation rates, and air exchange
rates. These default values are all central tendency values (i.e., mean or median values) sourced from EPA's Exposure Factors Handbook(\3.S. EPA.: )
(U.S. EPA. 20170). Low was selected for outdoor play structures, as there were uncertainties on the area volumes related to this scenario.
c Confidence in User-Selected Varied Inputs considers the quality of their data sources, as well as relevance of the inputs for the selected consumer condition of
use.

¦' Density Used was primarily based on gray literature values available for product descriptions. (1987)

e Use Duration is primarily sourced from the EPA's Exposure Factors Flandbook and by the judgment of the exposure assessor.

^Weight fraction of TCEP in articles was sourced from the available literature and database values.

•v Room of use (zone 1 in modeling) is informed by professional judgment of the exposure assessor based on the article scenario. The reasonableness of these
judgments is considered in the reported confidence ratings.

h The dermal permeability coefficient (Kp) used (0.022 cm/hr) and fraction absorbed (Fabs) used (35.1%) was derived from a study of TCEP tested on human ex
vivo skin (Abdattah et at. 2016). Frequency of mouthing (Low, Medium. High) was estimated using the assessors judgment when considering the exposure
scenario. Literature values override (2000) CEM 3.0 default values for fraction absorbed.

1 + + + Robust confidence suggests thorough understanding of the scientific evidence and uncertainties. The supporting weight of the scientific evidence
outweighs the uncertainties to the point where it is unlikely that the uncertainties could have a significant effect on the exposure estimate.

+ + Moderate confidence suggests some understanding of the scientific evidence and uncertainties. The supporting scientific evidence weighed against the
uncertainties is reasonably adequate to characterize exposure estimates.

+ Slight confidence is assigned when the weight of the scientific evidence may not be adequate to characterize the scenario, and when the assessor is making
the best scientific assessment possible in the absence of complete information. There are additional uncertainties that may need to be considered.

Page 194 of 572

-------
4958

4959

4960

4961

4962

4963

4964

4965

4966

4967

4968

4969

4970

4971

4972

4973

4974

4975

4976

4977

4978

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

5.1.2.4.1 Strengths, Limitations, Assumptions, and Key Sources of Uncertainty for
the Consumer Exposure Assessment

EPA recognizes the need to include an uncertainty analysis. One important distinction for such an
analysis is variability vs. uncertainty—both aspects need to be addressed. Variability refers to the
inherent heterogeneity or diversity of data in an assessment. It is a quantitative description of the range
or spread of a set of values and is often expressed through statistical metrics, such as variance or
standard deviation, which reflect the underlying variability of the data. Uncertainty refers to a lack of
data or an incomplete understanding of the context of the risk evaluation decision.

Variability cannot be reduced but can be better characterized. Uncertainty can be reduced by collecting
more or better data. Quantitative methods to address uncertainty include non-probabilistic approaches
such as sensitivity analysis and probabilistic or stochastic methods. Uncertainty can also be addressed
qualitatively by including a discussion of factors such as data gaps and subjective decisions or instances
where professional judgment was used.

Uncertainties associated with approaches and data used in the evaluation of consumer exposures are
described below. A sensitivity analysis was conducted for the following COUs to understand the drivers
for the inhalation, ingestion, and dermal estimates (Table 5-19).

Table 5-19. Sensitivity Analysis for Chronic Consumer Exposure Modeling Scenarios

Consumer Conditions of Use

User-Selected Varied Inputs"

Results

Subcategory

Consumer
Exposure
Scenario

Initial SVOC
Concentration in
Article (mg/cm3) h

Mouthing
Duration
(min)'

Surface
Area of
Article (m2)

Events

per
day (n)

Fabric and
textile products

Textile for
outdoor
children's play
structures

4.03
0.93
0.30

High

(8.4/7/10)

Low

(2.3/3.65/5)





Mouthing duration is
a driver of ingestion
exposures.

Building/
construction
materials -
insulation

Roofing
insulation

0.594
0.180
0.06







SVOC concentration
is a driver of
inhalation
exposures.

Building/
construction
materials -
wood and
engineered
wood products
- wood resin
composites

Wood flooring

30
12



211
105

10
5

SVOC concentration
is a driver of dermal
exposures.

Surface area of the
article and Events
per day (n) influence
the dermal exposure
estimates

"User selected inputs were varied for each of the listed consumer exposure scenarios.
b Initial SVOC concentration in article is a function of the product weight fraction and article density.
c The high mouthing duration defaults in CEM 3.0 were 10 min/event for an infant (<1 year of age), 7 min/event for
an infant aged 1-2 years, and 8.4 min/event for a child 3-5 years. EPA modified the mouthing durations to 5
min/event for infants <1 years, 3.65 min/event for 1-2 years, and 2.3 min/event for children 3-5 years to test the
sensitivity of this parameter.

Page 195 of 572

-------
4979

4980

4981

4982

4983

4984

4985

4986

4987

4988

4989

4990

4991

4992

4993

4994

4995

4996

4997

4998

4999

5000

5001

5002

5003

5004

5005

5006

5007

5008

5009

5010

5011

5012

5013

5014

5015

5016

5017

5018

5019

5020

5021

5022

5023

5024

5025

5026

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

A clear finding of the sensitivity analysis indicated that the initial SVOC concentration (a product of the
density and weight fraction) was a significant driver in the inhalation and dermal exposure estimates for
all scenarios. The initial SVOC concentration was also relevant for the ingestion estimate for the
inhalation scenario, likely because there was no estimate for direct mouthing of this COU. Mouthing
duration is an important driver of ingestion exposures for children's play structures. For full results on
the sensitivity analysis please refer to Appendix I (Consumer Exposures).

In the absence of parameter information from the literature, EPA used scientific judgement to select
parameters for consumer modeling. There are uncertainties associated with any scientific judgment. The
Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Supplemental Information File:
Consumer Exposure Modeling Inputs (	)23c) provides a full list of parameters and

description of rationale as to why certain parameter values were selected.

Weight Fraction

The key uncertainty in the consumer exposures assessment was the availability of relevant article weight
fractions data. The Ecology Washington database was the main source of weight fraction information
for the fabric, textile, and leather products scenarios. The 1.3 percent weight fraction for Textiles in
outdoor play structures was based on a value from the Washington State Database where the maximum
weight fraction of 67 articles was 1.3 percent (WSDE. 2023). Of the 67 articles, there were only 2 that
contained TCEP. The other article had a level of TCEP of 0.5 percent. Additionally, the database
indicated four detects of TCEP in carpet padding and rug mats (ranged from 0.01 to 0.02 percent). This
illustrates the limited data availability of weight fraction information for the fabric and textile products
scenario.

The building and construction products scenario (e.g., insulation, acoustic ceiling, wood resin products)
relied on old, foreign literature values from Ingerowski et al. (2 as cited in SCHER (2012).
Anecdotal information from the literature suggested TCEP is present in these products but did not have
specific information on weight fraction and article concentrations.

Values from Fang et al. (2013) were used to estimate weight fractions for foam seating and bedding
products. There are uncertainties in these estimates because concentrations of V6 (a dimer of TCEP)
were utilized in determining a TCEP weight fraction. This study measured TCEP at 14 percent as an
impurity in V6, and hence this proportion was used to estimate weight fractions of foam seating and
bedding products (Fane et al.. 2013). There are uncertainties associated with how much TCEP is present
as an impurity in V6.

TCEP in articles are not captured in CDR or Datamyne databases, as Datamyne does not include
articles/products containing the chemical unless the chemical name is included in the description. Based
on descriptions provided on the bills of lading, Figure 1-3 provides an estimate of the volume of TCEP
imported as the chemical (not in an identified product or article) from 2012 to 2020. This limitation
further illustrates the difficulty in obtaining current concentrations and weight fractions of TCEP in
consumer products.

Duration and Frequency of Contact and Mouthing

For the carpet back coating scenario and wood flooring scenario, a literature value indicated that
children under 12 years old spend 19 hours per day indoors (EFH 2011). It was assumed that the
frequency of contact per day was 5 events for carpet and 10 events for flooring, and that the area
mouthed was 10 cm2. It should be noted that these values are conservative assumptions for duration and

Page 196 of 572

-------
5027

5028

5029

5030

5031

5032

5033

5034

5035

5036

5037

5038

5039

5040

5041

5042

5043

5044

5045

5046

5047

5048

5049

5050

5051

5052

5053

5054

5055

5056

5057

5058

5059

5060

5061

5062

5063

5064

5065

5066

5067

5068

5069

5070

5071

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

frequency of contact {i.e., typical frequency may be less than these estimates). The dermal exposure
estimates are sensitive to the frequency of events per day parameter.

A further limitation for the carpet back coating and insulation scenario is the presence of a boundary
layer {e.g., top of the carpet, drywall in between insulation and living space) between the TCEP
containing material and the potentially exposed human {e.g., infant, child, adult). CEM 3.0 uses an
overall mass transfer coefficient that is empirically estimated from an equation based on the AMEM
guidance (the complexity of individual phase mass transfer is subsumed into an overall mass transfer
coefficient that is either measured or estimated from a regression equation based on assorted chemical
measurements). Although CEM 3.0 does not explicitly consider a boundary layer in its modeling, this
does not mean that the model does not attempt to capture this complexity. Nevertheless, it is an
uncertainty associated with the consumer modeling for the scenarios where a boundary layer would be
expected. The modeling as conducted suggests that the TCEP would migrate to the surface of the carpet
from the back coating components, or the dust particles would migrate from the insulation behind the
drywall to the living area.

Oral ingestion estimates are driven by mouthing of articles for infants and children. A sensitive
parameter driving these estimates is the duration of mouthing parameters. The recommended estimates
from CEM 3.0 are 8.4 min/hr, 7 min/hr, and 10 min/hr for young children (aged 3-5 years), infants (1-2
years), and infants (<1 year), respectively.

Trends and Monitoring Data

The paucity of monitoring information related to the consumer COUs makes it difficult for EPA to have
confidence in whether the consumer articles are nationally representative. Moreover, the decreasing
trend of TCEP use, seen in the production volume data and environmental monitoring data, coupled with
the understanding that many manufactures have replaced TCEP with alternatives in their products, build
more uncertainty about the relevance of the consumer modeling to current consumers.

A systematic review of the peer-reviewed and gray literature revealed that there is limited information
related to weight fractions of TCEP in consumer articles. No SDS were available for TCEP in consumer
products. For the limited monitoring and experimental literature that was available, it is unclear how
relevant the concentrations of TCEP at the time of sampling is related to consumer articles that are
produced today.

In 2013, the State of California amended Technical Bulletin 117, a residential upholstered furniture
flammability standard that was first implemented in 1975. The original TB 117 required interior filling
materials of upholstered furniture to withstand exposure to a 12 second small open flame (the small
flame impingement test, a one second flame, and the open flame test). This was replaced with a smolder
resistance test, which tests a lighted cigarette on the fabric outside of the foam in 2013. TB 117-2013 is
of significance to consumer articles, particularly fabric and textiles, and foam seating and bedding
products, as article manufacturers no longer are required to meet the stringent flame standards of TB
117. Flame retardant concentrations in these articles are expected to decrease following this change. The
available monitoring and experimental data on TCEP used in this consumer assessment was gathered
pre-2013 (Table 5-20).

Page 197 of 572

-------
5072

5073

5074

5075

5076

5077

5078

5079

5080

5081

5082

5083

5084

5085

5086

5087

5088

5089

5090

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table 5-20. Summary of Sampling Date for TCEP Weight Fraction Data

COU Subcategory

Weight Fraction Selected

Source

Sampling Date

Fabric and textile
products

•	0.02% carpet back coating

•	1.3% fabric in children's play
structures

Ecology Washington
database ("WSDE. 2023)

2012

Foam seating and
bedding products

•	0.51% furniture foam

•	0.74% auto foam

•	0.64% toy foam blocks

Fang et al. (2013)

2009-2011

Building/construction
materials - insulation

•	1.98% insulation

•	6.8% acoustic ceiling

Ingerowski et al. (2001)

<2001

Building/construction
materials - wood and
engineered wood
products - wood
resin composites

• 3% hardwood floors, wooden
TV stand

(SCHER. 2012)

1997°

a Ion as et al. (2 did provide more recent (2007) data on TCEP in wood tovs at 0.0004%. However, due to the
recent evidence suggesting TCEP use in wooden TV stands, and because TB 117-2013 is relevant for upholstered
foam and furniture materials, EPA selected a weight fraction of 3% for consumer modeling.

Due to the limited information available on article weight fractions, EPA was unable to select a range of
weight fraction for each of the COUs, and rather proceeded to assess consumer exposures to TCEP
containing articles with a single discrete weight fraction value per article scenario. Additional sensitivity
analysis varying the initial SVOC concentration in the article was conducted to help characterize the
results (Table 5-19).

Ion as et al. f: stratified their data on TCEP in toys by time of manufacture (before and after 2007
when the REACH regulation went into force). Pre-2007, TCEP was detected in 32 percent of 63
childrens toys whereas post-2007 TCEP was detected in 22 percent of 51 childrens toys. Nevertheless,
consumer modeling was conducted with possible weight fractions to understand the potential exposure
of such products in furnishings and the built consumer environment.

Table 5-21 summarizes the indoor air and indoor dust monitoring data that was available in the United
States. For a description of statistical methods, methodology of data integration, and treatment of non-
detects and outliers used to generate these estimates, please see the Supplemental Information File:
Environmental Monitoring Concentrations Reported by Media Type (	2023 e).

Page 198 of 572

-------
5091

5092

5093

5094

5095

5096

5097

5098

5099

5100

5101

5102

5103

5104

5105

5106

5107

5108

5109

5110

5111

5112

5113

5114

5115

5116

5117

5118

5119

5120

5121

5122

5123

5124

5125

5126

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table 5-21. Summary of Indoor Monitoring Data of TCEP from U.S. Studies

Matrices

Location
Type

Count of Estimates

from Studies
Containing U.S. Data

Unit

Fraction

Average of
Arithmetic
Estimates

Average of
90th Percentile
Estimates

Indoor Air

Public spaces

1

ng/m3

Particulate

2.0E00

4.6E00

Residential

1

ng/m3

Vapor/gas

9.5E00

2.1E01

Indoor Dust

Public spaces

1

ng/g

Dry

8.2E02

1.9E03

Residential

9

ng/g

Dry

1.1E03

2.2E03

Vehicles

1

ng/g

Dry

4.2E03

8.9E03

The maximum SVOC air concentration of 9.32 mg/m3 for the insulation condition of use is five orders
of magnitude higher than the 90th percentile estimate of indoor residential air concentrations found in
one U.S. study (2.1><10~5 mg/m3) (Dodson et ai. 2017). The maximum respirable portion dust
concentration of 11.13 |ig/mg (1.1 x 107 ng/g) is four orders of magnitude higher than the 90th percentile
estimate of residential indoor dust concentrations among nine U.S. studies (2.2><103 ng/g).

Modeling Approach Uncertainties

CEM 3.0 is a deterministic model where the outputs are fully determined by the choices of parameter
values and initial conditions. Stochastic approaches feature inherent randomness, such that a given set of
parameter values and initial conditions can lead to an ensemble of different model outputs. The overall
approach to the CEM modeling is intended to capture a range of low- to high-intensity user exposure
estimates by varying only a limited number of key parameters that represent the range of consumer
product and use patterns for each scenario. A limited set of parameters were varied in the sensitivity
analysis described in Table 5-19. Since not all parameters were varied, there is uncertainty regarding the
full range of possible exposure estimates. Although these estimates are thought to reflect the range of
exposure estimates for the suite of possible exposures based on the varied parameters, the scenarios
presented are not considered bounding or "worst-case," as there are unvaried parameters that are also
identified as sensitive inputs held constant at a central tendency value. Because EPA's largely
deterministic approach involves choices regarding highly influential factors such as weight fraction and
mouthing duration, it likely captures the range of potential exposure levels although it does not
necessarily enable characterization of the full probabilistic distribution of all possible outcomes.

CEM 3.0 has a set of predefined consumer exposure scenarios that do not always line up with the
conditions of use. For example, the CEM scenario utilized for consumer exposure to carpet back coating
was Fabrics: curtains, rugs, wall coverings. There are uncertainties on how TCEP migrates from carpet
back coatings to the surface of carpets and rugs. The literature describes that triphosphate esters such as
TCEP have 'blooming potential' which refers to the ability for the chemical to diffuse from a rubber or
plastic material to the outer surface after curing ("SCHER. 2012). Furthermore, the study from Castorina
et al. (2017) has indicated that TCEP levels in dust are significantly associated with the presence of
extremely worn carpets, suggesting that TCEP can be sampled in the dust from carpets and make it to
the surface.

Background levels of TCEP in indoor air and indoor dust are not considered or aggregated in this
assessment; therefore, there is potential for underestimating consumer exposures. Furthermore,

Page 199 of 572

-------
5127

5128

5129

5130

5131

5132

5133

5134

5135

5136

5137

5138

5139

5140

5141

5142

5143

5144

5145

5146

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

consumer exposures were evaluated on a COU specific basis and are based on the use of a single
consumer article, not multiple articles in the indoor environment.

There are uncertainties regarding the use of the 35.1 percent dermal fraction absorption (Fabs) parameter
for the consumer dermal exposure estimates. This is the 24-hour value for fraction absorbed from
Abdallah et al. (2016). EPA cannot assume that consumers would immediately wash their hands
following contact with consumer articles. Therefore, it was assumed that the dose that deposited on the
skin during exposure to a consumer article would remain on the skin until the skin was eventually
washed. While the duration of interaction with materials that contain TCEP may be shorter than the
duration that was tested in the dermal absorption study (i.e., a 24-hour exposure), EPA decided to use
the 35.1 percent fraction absorption value from Abdallah et al. (2016). due to uncertainties related to
consumer hand-washing behaviors.

5.1.3 General Population Exposures	

TCEP- General Population Exposures (Section 5.1.3):

Key Points

EPA evaluated the reasonably available information for the following general population exposures,
the key points of which are summarized below:

•	Oral ingestion for subsistence fishers had the highest exposure estimates (2.17 to 75.5 mg/kg-
day) among all routes. The highest subsistence fishing exposure estimates were for the
incorporation into paints and coatings - resins/solvent-borne OES.

•	The hypothetical scenario of a child playing in mud near a facility releasing TCEP to the
ambient air resulted in the highest dermal exposures at a maximum of 7.97 mg/kg-day for use
of paints and coatings at job sites OES. Estimates for a child conducting activities with soil
(2.12xl0~3 mg/kg-day) and incidental soil ingestion (1.08xl0_1 mg/kg-day) were calculated.
Paints and coatings was the only OES for the children playing in mud scenario with MOEs
below the benchmark for non-cancer as described in Section 5.3.2.3.

•	The highest inhalation exposure concentrations were for the use of paints and coatings at job
sites OES at a central tendency estimate of 3.36x 10~5 and a 95th percentile of 8.21 x 10~5
|ig/m3.

•	Exposure estimates for drinking water non-dilute from surface water (1,46x 10~4 mg/kg-day)
were highest for the formulation of TCEP containing reactive resins OES.

•	Children in fenceline communities and subsistence fishers are PESS who may have elevated
exposure to TCEP compared to rest of general population due to industrial and commercial
environmental releases.

General population exposures occur when TCEP is released into the environment and the environmental
media is then a pathway for exposure. Section 3.3 provides a summary of the monitoring, database, and
modeled data on concentrations of TCEP in the environment. Figure 5-4 below provides a graphic
representation of where and in which media TCEP is estimated to be found and the corresponding route
of exposure.

Page 200 of 572

-------
5147

5148

5149

5150

5151

5152

5153

5154

5155

5156

5157

5158

5159

5160

5161

5162

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE

December 2023

Ambient Air
Inhalation

Landfills
(Industrial or
Muncipal)

| Sediment |	

Figure Legend

*¦ Negligible

~ Low/Slow

Moderate

High/Fast/Strong

Very High/Rapid/Strong

Partitioning/T ransportation
~ T ransformation/Degradation
Wastewater Facility

-ys-

Water
Recreation

Oral. Derma!

Surface Water

Wastewater
Facility

Drinking

Water
T reatment

Bathing
_ . ..	Water

Dermal.
Inhalation

Water

Oral

Figure 5-4. Potential Human Exposure Pathways to TCEP for the General Population"

11 The diagram presents the media (white text boxes) and routes of exposure (italics for oral, inhalation, or dermal)
for the general population. Sources of drinking water from surface or water pipes is depicted with grey arrows.

This diagram pairs with Figure 2-1 depicting the fate and transport of the subject chemical in the
environment.

5.1.3.1 Approach and Methodology

TCEP is used primarily as an additive flame retardant in a variety of materials. TCEP has been detected
in the indoor and outdoor environment and in human biomonitoring indicating that some amount of
exposure is occurring in some individuals, although exposures likely vary across the general population.
See Section 3.3 and Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Systematic
Review Supplemental File: Data Extraction Information for General Population, Consumer, and
Environmental Exposure (U.S. EPA. 2023p) for a summary of environmental and biomonitoring studies
where TCEP has been detected.

Page 201 of 572

-------
5163

5164

5165

5166

5167

5168

5169

5170

5171

5172

5173

5174

5175

5176

5177

5178

5179

5180

5181

5182

5183

5184

5185

5186

5187

5188

5189

5190

5191

5192

5193

5194

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Releases of TCEP are likely to occur through the following mechanisms: diffusion from sources, gas-
phase, and particle-phase mass-transfer, abrasion of materials to form small particulates through routine
use, and direct transfer from articles to dust adhered to the article surface. Releases of flame retardants
to the outdoor environment may occur through direct releases to water, land, and air as well as indirect
releases from the indoor environment.

For a more detailed discussion about indoor SVOC exposure, fate, and transport in the indoor
environment, please see Section 2.2.2.

Exposure to the general population was estimated for the industrial and commercial releases per OES.
Table 3-3 illustrates how the industrial and commercial releases to the environmental media varies by
OES.

Modeled air concentrations (Section 3.3.1.2) were utilized to estimate inhalation exposures (5.1.3.2) to
the general population at various distances from a hypothetical facility. Modeled surface water
concentrations (Section 3.3.2.5) were utilized to estimate oral drinking water exposures, oral fish
ingestions exposures, incidental oral exposures (Section 5.1.3.4), and incidental dermal exposures
(Section 5.1.3.3) for the general population. Modeled groundwater concentrations (Section 3.3.3.7),
were also used to estimate oral drinking water exposures (Section 5.1.3.4) to the general population.
Modeled soil concentrations (Section 3.3.3.2) via deposition were used to estimate dermal and oral
exposures (Sections 5.1.3.3 and 5.1.3.4) to children who play in mud and other activities with soil.

Exposures estimates from industrial and commercial releases of TCEP were compared to exposure
estimates from non-scenario specific monitoring data to ground truth the results (e.g., indoor dust
exposures). Table 5-22 summarizes the environmental media monitoring data that was available in the
United States. For a description of statistical methods, methodology of data integration and treatment of
non-detects and outliers used to generate these estimates please see th q Draft Risk Evaluation for Tris(2-
chloroethyl) Phosphate (TCEP) - Supplemental Information File: Environmental Monitoring
Concentrations Reported by Media Type (	23g).

Table 5-22. Summary of Environmental Monitoring Data of TCEP from the Literature for U.S.
Studies

Matrices

Location Type

Count of Estimates

from Studies
Containing U.S. Data

Unit

Fraction

Average of
Arithmetic
Estimates

Average of
90th Percentile
Estimates

1 \ ironnvnlul media

Ambient Air

General Population

6

ng/m3

Any

1.3E-01

2.5E-01

Drinking Water

General Population

1

ng/L

Any

4.9E00

9.5E00

Sediment

General Population

1

ng/g

Dry

2.3E00

4.1E00

Surface Water

General Population

5

ng/L

Any

1.3E02

2.5E02

Wastewater

Treated Effluent

2

ng/g

Wet

2.1E01

4.3E01

Treated Effluent

4

ng/L

Wet

8.1E02

1.2E03

Lcoliujiail media

Aquatic 1'isli

General Population

1

11*1 n

O' o

Lipid

l.uLul

:.5LU1

Page 202 of 572

-------
5195

5196

5197

5198

5199

5200

5201

5202

5203

5204

5205

5206

5207

5208

5209

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Matrices

Location Type

Count of Estimates

from Studies
Containing U.S. Data

Unit

Fraction

Average of
Arithmetic
Estimates

Average of
90th Percentile
Estimates

Terrestrial
Birds

General Population

2

ng/g

Wet

5.3E00

9.7E00

Terrestrial
Plants

Remote

1

ng/g

Wet

1.3E02

2.2E02

1 liinuin [iiomomloniuj

Human Hair

General Population

2

ng/g

Dry

2.7E02

4.2E02

Human Nails

General Population

1

ng/g

Dry

6.3E02

1.4E03

Figure 5-5 depicts the direct and indirect methods EPA used to estimate general population exposures.
The direct assessment used environmental release estimates that were related to the industrial and
commercial OES (see Section 3.2). Release estimates were used to model ambient air concentrations
(see Section 3.3.1.2), surface water concentrations (see Section 3.3.2.5), soil concentrations (see Section
3.3.3.2), and groundwater concentrations as a result of landfill leachate (see Section 3.3.3.7). EPA
modeled estimates for the environmental media were used to estimate inhalation, dermal and ingestion
doses for various anticipated scenarios (e.g., childrens dermal exposure to soil, fish ingestion for the
general population, drinking water ingestion exposure). Further information on the assessed exposure
scenarios is presented in the individual sections below. In addition, EPA estimated exposure doses using
an indirect estimation method via reverse dosimetry (see Section 5.1.3.5). Furthermore, to help "ground
truth" the results, the reported environmental monitoring and reported modeled data (i.e., TCEP
concentration and doses in dietary sources, dust, soil, ambient air, indoor air, and surface water) were
compared against the exposure estimates calculated from the direct assessment patterns.

Page 203 of 572

-------
5210

5211

5212

5213

5214

5215

5216

5217

5218

5219

5220

5221

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE

December 2023

Direct
Assessment

Indirect
Assessment

Figure 5-5. Direct and Indirect Exposure Assessment Approaches Used to Estimate General
Population Exposure to TCEP

For each exposure pathway, central tendency and high-end exposures were estimated. EPA's Guidelines
for Human Exposure Assessment defined central tendency exposures as "an estimate of individuals in
the middle of the distribution." It is anticipated that these estimates apply to most individuals in the
United States. Fligh-end exposure estimates are defined as "plausible estimate of indivi dual exposure for
those individuals at the upper end of an exposure distribution, the intent of which is to convey an
estimate of exposure in the upper range of the distribution while avoiding estimates that are beyond the
true distribution." It is anticipated that these estimates apply to some individuals, particularly those who
may live near facilities with elevated concentrations.

Page 204 of 572

-------
5222

5223

5224

5225

5226

5227

5228

5229

5230

5231

5232

5233

5234

5235

5236

5237

5238

5239

5240

5241

5242

5243

5244

5245

5246

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE

December 2023

5.1.3.1.1 General Population Exposure Scenarios

Figure 5-4 provides an illustration of the exposure scenarios considered for general population exposure.

Ambient Air Exposure Scenarios

The Ambient Air Methodology utilizing AERMOD evaluated exposures to human populations at eight
finite distances (10, 30, 60, 100, 1,000, 2,500, 5,000, and 10,000 m) and one area distance (100 to 1,000
m) from a hypothetical releasing facility for each OES. Human populations for each of the eight finite
distances were placed in a polar grid every 22.5 degrees around the respective distance ring. This results
in a total of 16 modeled exposure points around each finite di stance ring for which exposures are
modeled. Figure 5-6 provides a visual depiction of the placement of exposure points around a finite
distance ring. Although the visual depiction only shows exposure point locations around a single finite
distance ring, the same placement occurred for all eight finite distance rings.

1000 m

Exposure Points around each Finite Distance Ring

Releasing Facility

60 m

100 m

10,000 m

30-60 m

100-1,000 m

10 m

Location of
fx) Exposed
Individual

2,500 m

Figure 5-6. Modeled Exposure Points for Finite Distance Rings for Ambient Air Modeling
(AERMOD)

Modeled exposure points for the area distance evaluated were placed in a cartesian grid at equal
distances between 200 and 900 m around each releasing facility (or generic facility for alternative
release estimates). Exposure points were placed at 100-meter increments. This results in a total of 456
points for which exposures are modeled. Figure 5-6 provides a visual depiction of the placement of these
exposure points (each dot) around the area distance ring.

Although the ambient air is a minor pathway for TCEP, the general population may be exposed to
ambient air concentrations and air deposition because of TCEP releases. Relevant exposures scenarios

Page 205 of 572

-------
5247

5248

5249

5250

5251

5252

5253

5254

5255

5256

5257

5258

5259

5260

5261

5262

5263

5264

5265

5266

5267

5268

5269

5270

5271

5272

5273

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

considered in this draft risk evaluation include ambient air inhalation for populations living nearby
releasing facilities, and ingestion and dermal exposure of soil to children result of ambient air deposition
from a nearby facility.

Soil Exposure Scenarios

Air deposition fluxes from AERMOD were used to estimate soil concentrations at various distances
from the hypothetical facility for each OES (see Section 3.3.3.2). Oral ingestion and dermal absorption
exposure estimates of soil were calculated for children aged 3 to 6 years. Ingestion estimates were
calculated for a central tendency and high intake rate. Dermal absorption estimates were calculated for
two exposure scenarios: a child playing in mud, and a child performing activities with soil.

Water Exposure Scenarios

TCEP is expected to be found predominantly in water or soil. Section 3.3.2.5 provides modeled
estimates of TCEP in surface water due to release of TCEP to water. Section 1.1.1 provides model
estimates of TCEP in surface water due to air deposition to surface waters. Section 3.3.3.7 provides
modeled estimates of TCEP in groundwater due to estimated migration from landfill leachate. Each of
these estimates were used to calculate an exposure dose from drinking water for the general population.
Additionally, modeled surface water concentrations (see Section 3.3.2.5) were used to calculate a dermal
exposure estimate from swimming, incidental ingestion estimates from swimming, fish ingestion
exposure.

5.1.3.2 Summary of Inhalation Exposure Assessment

Modeled ambient air concentrations for various distances from a hypothetical facility for each COU are
presented in Section 3.3.1.2. Figure 5-7 below is a graph of the inhalation concentration by distances for
the low production volume (2,500 lb/year) low-end and high-end estimates by the central tendency and
high meteorology data. The x-axis is in log scale of distances in meters and the y-axis is in log scale of
the 50th percentile concentrations in ppm.

Page 206 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE

December 2023

5274

5275

5276

5277

5278

5279

Low Release (2500 lbs)

Low Release (2500 lbs)

Central Tendency Estimate

High-End Estimate

-5.04

E
a.

B -7.5 •)
c

0

flj

£ -100

1

u
c

8

§
I

JC

I

o>
o

-5.0 i

-10.0-

1.0 1.5 2.0 25 3.0 1.0

Log [Distance (m)]

20

2.5

3.0

Scenario
COM3
IND
MFG
-4— PROC-article
*- PROC -reactive
+~ PROC-resin

PROC-waterbom«

COM3 refers to Use m paints and coatmgs at pb sites.

IND refers to Use of Lab Chemicals

MFG refers to Repackaging of Import Containers

PROC-article refers to Processing into 2-part resin article.

PROCresm refers to Incorporation into points and coatings ¦ resins/solvent-borne

PROC-waterborne refers to Incorporation into paints and coatings - waterbome coatings

PROC reactive refers to Formulation of TCEP containing reactive resin

Figure 5-7. General Population Inhalation Concentrations (ppin) by Distance (m) in Log Scale

Table 5-23 below indicates the ambient air concentrations at one distance (100 m) for each of the OES.
For a full set data for all distances please see Appendix H.

Page 207 of 572

-------
5280

5281

5282

5283

5284

5285

5286

5287

5288

5289

5290

5291

5292

5293

5294

5295

5296

5297

5298

5299

5300

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table 5-23. Excerpt of Ambient Air Modeled Concentrations for the 2,500 lb Production Volume,

OESa

Meteorology

Source

Concentration (ppm)by Percentile

10th

50th

95th

Use in paints and coatings at job
sites

MetCT

FUGU

1.15E-05

3.36E-05

6.45E-05

MetHIGH

FUGU

8.77E-06

3.08E-05

8.21E-05

Use of laboratory chemicals

MetCT

ALL

1.51E-08

2.04E-08

3.33E-08

MetHIGH

ALL

1.16E-08

2.24E-08

3.32E-08

Repackaging of import containers

MetCT

ALL

1.50E-10

3.88E-10

9.12E-10

MetHIGH

ALL

2.34E-10

4.39E-10

1.12E-09

Processing into 2-part resin article

MetCT

ALL

1.48E-08

1.93E-08

2.70E-08

MetHIGH

ALL

9.46E-09

1.96E-08

2.72E-08

Incorporation into paints and
coatings - 2-part reactive coatings

MetCT

ALL

2.60E-11

1.60E-09

1.14E-08

MetHIGH

ALL

3.46E-10

2.29E-09

1.11E-08

Incorporation into paints and
coatings - 1-part coatings

MetCT

ALL

4.80E-09

1.31E-08

2.87E-08

MetHIGH

ALL

4.00E-09

1.35E-08

3.51E-08

Formulation of TCEP containing
reactive resin

MetCT

ALL

2.72E-11

1.78E-09

1.26E-08

MetHIGH

ALL

3.73E-10

2.52E-09

1.21E-08

a Table 3-3 provides a crosswalk of industrial and commercial COUs to OESs

5.1.3.3 Summary of Dermal Exposure Assessment

5.1.3.3.1 Incidental Dermal from Swimming

The general population may swim in affected surface waters (streams and lakes) that are affected by
TCEP contamination. Modeled surface water concentrations from EFAST 2014 were used to estimate
acute doses and average daily doses because of dermal exposure while swimming.

The following equations were used to calculate incidental dermal (swimming) doses for all COUs, for
adults, youth, and children:

Equation 5-11

ADR =

SWC x Kp XSA xETx CF1 x CF2
BW

Equation 5-12

ADD =

SWC x Kp x SA x ET x RD x ET x CF 1 x CF2
BW x AT x CF3

Where:

ADR = Acute Dose Rate (mg/kg-day)
ADD = Average Daily Dose (mg/kg-day)
SWC = Chemical concentration in water (|ig/L)

Page 208 of 572

-------
5301

5302

5303

5304

5305

5306

5307

5308

5309

5310

5311

5312

5313

5314

5315

5316

5317

5318

5319

5320

5321

5322

5323

5324

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Kp =

Permeability coefficient (cm/h)

SA

Skin surface area exposed (cm2)

ET

Exposure time (h/day)

RD

Release days (days/year)

ED

Exposure duration (years)

BW =

Body weight (kg)

AT =

Averaging time (years)

CF1 =

Conversion factor (1.0/ 10 3 mg/|ig)

CF2 =

Conversion factor (1,0x 10 3 L/cm3)

CF3 =

Conversion factor (365 days/year)

A summary of inputs utilized for these exposure estimates are provided in Appendix H.

EPA used the dermal permeability coefficient (Kp) (0.022 cm/h) derived by Abdallah et al. (2016) from
their in vitro study that measured TCEP absorption through excised human skin.

Table 5-24. Modeled Incidental Dermal (Swimming) Doses for all COUs for Adults, Youths, and
Children, for the 2,500 lb High-End Release Estimate			

OES"

Surface Water
Concentration

Adult (>21 years)

Youth (11-15 years)

Child (6-10 years)

30Q5
Cone.
(Hg/L)

Harmonic

Mean
Cone.
(Hg/L)

ADRpot
(mg/kg-
day)

ADD

(mg/kg-
day)

ADRpot
(mg/kg-
day)

ADD

(mg/kg-
day)

ADRpot
(mg/kg-
day)

ADD

(mg/kg-
day)

Repackaging of import
containers

862.129

1,366.528

1.39E-03

6.02E-06

1.06E-03

4.61E-06

6.44E-04

2.80E-06

Incorporation into
paints and coatings -
1-part coatings

3,819.444

5,912.114

6.14E-03

2.61E-05

4.70E-03

2.00E-05

2.85E-03

1.21E-05

Incorporation into
paints and coatings -
2-part reactive
coatings

3,462.800

5,360.066

5.57E-03

2.36E-05

4.27E-03

1.81E-05

2.59E-03

1.10E-05

Use in paints and
coatings at job sites

2,029.305

3,216.574

3.26E-03

1.42E-05

2.50E-03

1.09E-05

1.52E-03

6.58E-06

Formulation of TCEP
containing reactive
resin

4,844.722

6,245.374

7.79E-03

2.75E-05

5.97E-03

2.11E-05

3.62E-03

1.28E-05

Use of laboratory
chemicals

34.555

54.722

5.59E-05

2.41E-07

4.26E-05

1.85E-07

2.58E-05

1.12E-07

" Table 3-3 provides a crosswalk of industrial and commercial COUs to OES.

5.1.3.3.2 Incidental Dermal Intake from Soil

Dermal absorbed doses (DAD) were calculated for TCEP using the following formula:
Equation 5-13

_ Cson x CF x AF x ABSd x SAsou x EV
~	BW x AT

Where:

Page 209 of 572

-------
5325

5326

5327

5328

5329

5330

5331

5332

5333

5334

5335

5336

5337

5338

5339

5340

5341

5342

5343

5344

5345

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

AF = Adherence factor of soil to skin (mg/cm2-event)

ABSd = Dermal absorption fraction

SA = Skin surface area

EV = Events per day

BW = Body weight

AT = Averaging time

Modeled soil concentrations were calculated from 95th percentile air deposition (Section 3.3.3.2) for
100 and 1,000 m. These calculations were conducted for the COM-paints-use scenario (LOW PV -
2,500 lb, HE-95th percentile release). The dermal absorption fraction (ABSd) used was 35.1 percent
(Abdallah et at.. 2016). The skin surface area for the arms (0.106 m2), hands (0.037 nr), legs (0.195 nr)
and feet (0.049 m2), and body weight (18.6 kg) of a 3- to 6-year-old was used from the Exposure
Factors Handbook (U.S. EPA. 2017c). EPA used two different scenarios for the adherence factor of soil
to skin: 96 mg/cm2for a child playing in mud and 0.467 mg/cm2for children's activity with soil. With an
assumption of one event per day and an averaging time of 2 days, the dermal exposure estimates for the
different scenarios were as follows:

Table 5-25. Modeled Soil Dermal Doses for the Commercial Use of Paints and Coatings COU, for
Children

OES

Exposure
Scenario

Distance
(m)

95th Percentile Soil
Concentration

Dermal Absorbed Dose
(mg/kg-day)

Use in
paints and
coatings at
job sites

Activities
with soil

100

1.14E04

3.88E-02

1,000

8.65E01

2.12E-03

Playing in
mud

100

1.14E04

7.97E00

1,000

8.65E01

4.36E-01

5.1.3.4 Summary of Oral Exposures Assessment

Page 210 of 572

-------
5346

5347

5348

5349

5350

5351

5352

5353

5354

5355

5356

5357

5358

5359

5360

5361

5362

5363

5364

5365

5366

5367

5368

5369

5370

5371

5372

5373

5374

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

5.1.3.4.1 Drinking Water Exposure

us

Mix

NonUS

4253347 - Padhye cl al.. 2014 - US
3975066 - Hopple et al.. 2009 - US
3364193 - Kingsbury el al.. 2008 - US

3559503 - Focazio cl al.. 2008 - PR.US
1487184 - Lebel el al., 1987 - CA.US

3455908 - Lcc ct al.. 2016 - KR
5469210 - Valcarcel ct al., 2018 - ES
1250860 - Rodil ct al.. 2012 - ES
5469582 - Yasuhara. 1994 - JP

IOA-6

B	I General Population (Background)

I	Unknown/Not Specified

V Lognormal Distribution (CT and 90th percentile)

H	Non-Dctect

~v













































¦ w











v







|7V

*



l0A-4

0.01	1

Concentration (ng/L)

100

10*4

Figure 5-8. Concentrations of TCEP (ng/L) in Drinking Water from 1982 to 2014

A study of drinking water systems in the United States indicated a maximum of 470 ng/L and a median
of 120 ng/L of TCEP in finished water, and a maximum of 200 ng/L and a median of 140 ng/L in
distributed waters in 6 out of 19 drinking water systems. The drinking water systems collected samples
from 19 drinking water treatment plants (DWTPs) across the United States, representing drinking water
for more than 28 million Americans (Benotti et al.. 2009).

TCEP has been detected in tap water in Korea at a mean of 39.5 and a maximum of 87.4 ng/L as
recently as 2017 (Park et al.. 2018). Because the OPFR concentrations were correlated with the distance
of the pipes (both from the water intake source to the drinking water treatment facility and the drinking
water treatment facility to the sampling site), this study has suggested that a possible source of OPFRs in
tap water were pipes. Pipe materials are known to promote the formation of disinfection by products or
biofilms (Park et al.. 2018).

Drinking Water Intake Estimates via Modeled Surface Water Concentrations
Modeled surface water concentrations (see Sections 1.1.1 and 3.3.2.5) were used to estimate drinking
water exposures. A 0 percent drinking water treatment removal efficiency was used for the purposes of
this exposure estimation.

Drinking water intakes were calculated using the following formulae:

Equation 5-14

( DWT\

SWC X (1 - ¦^TYrf) x IRdw XRD X CF1

ADRpor = 				

P0T	BW X AT

Page 211 of 572

-------
5375

5376

5377

5378

5379

5380

5381

5382

5383

5384

5385

5386

5387

5388

5389

5390

5391

5392

5393

5394

5395

5396

5397

5398

5399

5400

5401

5402

5403

5404

5405

5406

5407

5408

5409

5410

5411

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Equation 5-15

( DWT\

SWC x (1 - ) x IRdw xEDxRD x CF1

ADDpnr = 				

P0T	BW x AT x CF2

Equation 5-16

f DWT\

SWC x (1 - ^rr) x IRdw x ED x RD x CF1

LADDpor = 				

P0T	BW x AT x CF2

Equation 5-17

/ DWT\

SWC x (1 - ) x ED x RD x CF1

LAD CP0T = 	-	 ^	

P0T	AT X CF2

Where:

ADRpot

Potential Acute Dose Rate (mg/kg/day)

ADDpor

Potential Average Daily Dose (mg/kg/day)

LADDpor

Potential Lifetime Average Daily Dose (mg/kg/day)

LADCpot

Potential Lifetime Average Daily Concentration in drinking water



(mg/L)

SWC

Surface water concentration (ppb or |ig/L; 30Q5 cone for ADR,



harmonic mean for ADD, LADD, LADC)

DWT

Removal during drinking water treatment (%)

IRdw

Drinking water intake rate (L/day)

RD

Release days (days/yr for ADD, LADD and LADC; 1 day for



ADR)

ED

Exposure duration (years for ADD, LADD and LADC; 1 day for



ADR)

BW

Body weight (kg)

AT

Exposure duration (years for ADD, LADD and LADC; 1 day for



ADR)

CF1

Conversion factor (1.0xl0~3 mg/|ig)

CF2

Conversion factor (365 days/year)

A method was derived to incorporate a dilution factor to estimate TCEP concentrations at drinking water
locations downstream from surface water release points. Since no location information was available for
facilities releasing TCEP, a dilution factor and distances to drinking water intake was estimated for each
relevant SIC code. Table 5-26 provides the 50th quantile distances and 50th quantile harmonic mean and
for the relevant SIC codes.

Page 212 of 572

-------
5412

5413

5414

5415

5416

5417

5418

5419

5420

5421

5422

5423

5424

5425

5426

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table 5-26. 50th Quantile Distances and 30Q5 and Harmonic Mean 50th Quantile Dilution

Factors for Relevant TC

EP SIC

SIC Codes

n

50th Quantile
Distance
(km)

50th Quantile

Dilution
Factor (30Q5)

50th Quantile Dilution
Factor
(Harmonic Mean)

Adhesives, Sealants,
Plastics, Resins,
Rubber Manufacturing

516

113.82

432.36

528.47

Paint Formulation

374

107.03

1,603.6

1,854.89

POTWs - All facilities

567

129.57

1,233.87

1,557.91

30Q5 = The lowest 30-day average flow that occurs (on average) once every 5 years

To calculate the diluted water concentrations the surface water concentrations from E-FAST modeling
were divided by the dilution factor. Table 5-27 presents the diluted drinking water concentrations for
adults for all industrial and commercial COUs.

Table 5-27. Modeled Drinking Water Ingestion Estimates for Diluted Surface Water
Concentrations for Adults for All Industrial and Commercial COUs for the 2,500 lb High-End
Release Estimate

OES"

Diluted Water Concentration

Adult (> 21 years)

Harmonic Mean
Concentration
(*ig/L)

30Q5
Concentration
(*ig/L)

ADRpot
(mg/kg-
day)

ADD

(mg/kg-
day)

LADDpot

(mg/kg-
day)

LADCpot
(mg/L)

Repackaging of import
containers

0.553

1.108

4.46E-05

1.67E-08

7.05E-09

6.41E-07

Incorporation into
paints and coatings -
1-part coatings

2.059

3.687

1.48E-04

6.20E-08

2.62E-08

2.39E-06

Incorporation into
paints and coatings -
2-part reactive
coatings

1.867

3.343

1.35E-04

5.62E-08

2.38E-08

2.16E-06

Use in paints and
coatings at job sites

1.303

2.607

1.05E-04

3.92E-08

1.66E-08

1.51E-06

Formulation of TCEP
containing reactive
resin

9.167

14.445

5.81E-04

2.76E-07

1.17E-07

1.06E-05

Use of laboratory
chemicals

0.022

0.044

1.79E-06

6.68E-10

2.83E-10

2.57E-08

a See Table 3-3 for a crosswalk of industrial and commercial COUs to OESs.

Table 5-28 provides the non-diluted drinking water intake estimates. In this case, it is assumed that the
surface water outfall is located very close (within a few km) to the population. The dilution factor
reduces the acute, chronic and lifetime exposure estimates by a factor of three.

Page 213 of 572

-------
5427

5428

5429

5430

5431

5432

5433

5434

5435

5436

5437

5438

5439

5440

5441

5442

5443

5444

5445

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table 5-28. Modeled Drinking Water Ingestion Estimates for Surface Water Concentrations for

OES"

Water Concentration

Adult (> 21 years)

Harmonic
Mean
Concentration
(Hg/L)

30Q5
Concentration
(jig/L)

ADRpot
(mg/kg-
day)

ADD

(mg/kg-
day)

LADDpot
(mg/kg-
day)

LADCpot
(mg/L)

Repackaging of
import containers

862.129

1,366.528

5.4992E-02

2.60E-05

1.10E-05

9.99E-04

Incorporation into
paints and coatings -
1-part coatings

3,819.444

5,912.114

2.3792E-01

1.15E-04

4.87E-05

4.43E-03

Incorporation into
paints and coatings -
2-part reactive
coatings

3,462.800

5,360.066

2.1570E-01

1.04E-04

4.41E-05

4.01E-03

Use in paints and
coatings at job sites

2,029.305

3,216.574

1.2944E-01

6.11E-05

2.59E-05

2.35E-03

Formulation of TCEP
containing reactive
resin

4,844.722

6,245.374

2.5133E-01

1.46E-04

6.17E-05

5.62E-03

Use of laboratory
chemicals

34.555

54.772

2.20E-03

1.04E-06

4.40E-07

4.01E-05

a Table 3-3 provides a crosswalk of industrial and commercial COUs to OES.

A summary of inputs utilized for these exposure estimates is presented in Appendix H.

Drinking Water via Leaching of Landfills to Groundwater

Groundwater concentrations from leaching from landfills was estimated for the 2,500 and 25,000 lb
production volume scenarios (see Table 3-7. in Section 3.3.3.7). The relevant COU/OES that may be
relevant for groundwater migration from landfill leachate are the incorporation into paints and coatings -
1-part coatings, and processing into formulation of TCEP containing reactive resin. These OESs result in
the following releases to landfill presented in Table 5-29. In addition, consumer articles could be
disposed to municipal solid waste landfills and construction and demolition landfills.

Table 5-29. Landfill Releases of TCEP from Two Commercial and Industrial OESs

OES

Number of Release
Days

Annual Release Per Site
(kg-site-yr)

Daily Release
(kg/site-day)

Incorporation into paints and
coatings - 1-part coatings

2

2.15E01

9.27E00

Formulation of TCEP
containing reactive resin

17

4.29E01

2.49E00

Section 3.3.3.7 estimates a range of groundwater concentrations because of industrial and commercial
releases. The range of concentrations varies due to leachate concentrations to be between 1.08x10 3 and
1.08X101 |ig/L. Using the same formulae for drinking water ingestion above, adult drinking water
estimates because of landfill leachate contamination are presented in Table 5-30.

Page 214 of 572

-------
5446

5447

5448

5449

5450

5451

5452

5453

5454

5455

5456

5457

5458

5459

5460

5461

5462

5463

5464

5465

5466

5467

5468

5469

5470

5471

5472

5473

5474

5475

5476

5477

5478

5479

5480

5481

5482

5483

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table 5-30. Estimated Average Daily Doses, Lifetime Average Daily Doses, and Lifetime Average
Daily Concentrations for Adults from Groundwater Concentrations by DRAS	

DRAS

Groundwater
Concentration

Adult (> 21 years)

ADD

(mg/kg-day)

LADDpot
(mg/kg-day)

LADCpot
(mg/L)

Low Estimate: Low Leachate Concentration -
2,500 lb Production Volume

1.08E-03

3.3E-11

1.4E-11

1.3E-09

High Estimate: High Leachate Concentration -
2,500 lb Production Volume

1.08E01

3.3E-07

1.4E-07

1.3E-05

These results would be further lowered if dilution was incorporated to these drinking water estimates.
Due to uncertainties in distance from drinking water intake location to the groundwater contamination
site the dilution was not estimated.

The complete set of exposure estimates for adults and infants relying on groundwater as a primary
drinking water source are presented in Appendix H.5.

5.1.3.4.2 Fish Ingestion Exposure

Surface water concentrations for TCEP associated with a particular COU were modeled using E-FAST
as described in Section 3.3.2.5. Surface water concentrations based on harmonic mean surface water
flows, which represents long-term average flow conditions, were used to estimate the concentration of
TCEP in fish tissue. As it takes time for chemical concentrations to accumulate in fish, a harmonic mean
flow is more appropriate than a low streamflow value (e.g., 7Q10) that occurs infrequently.

Furthermore, dilutions of surface water concentrations of TCEP further downstream of a facility's
outfall was not considered, as fish presumably reside within stream reaches receiving direct releases
from a facility. This approach takes into account that people often harvest fishes originating from
various locations regardless of known or unknown releases to the environment at that location; thus, it is
more conservative because it estimates higher concentrations of TCEP in fish.

EPA estimated exposure from fish consumption using an adult ingestion rate for individuals aged 16 to
<70 years, which is lower than all age groups per kilogram of body weight (thus more protective) except
for 6 to <11 and 11 to <16 years (I v < < \ 2014a). See Table Apx H-2 in Appendix H for more
information. The 50th percentile (central tendency) and 90th percentile ingestion rate (IR) for adults is
5.04 g/day and 22.2 g/day, respectively. The ADRs were calculated using the 90th percentile IR. EPA
typically uses the central tendency for chronic exposure estimates. However, EPA considers both the
central tendency and 90th percentile IRs to be reasonable for the general population. The 90th percentile
IR can also capture individuals within the general population that may have higher chronic exposures
but not as high as the subsistence fisher. As a result, EPA used both fish ingestion rates to estimate an
ADD and LADD. Exposure estimates via fish ingestion were calculated according to the following
equation:

Equation 5-18

SWC x BAF x IR x CF1 x CF2 x ED
ADR or ADD =	

AT XBW
Where:

ADR = Acute Dose Rate (mg/kg/day)

Page 215 of 572

-------
5484

5485

5486

5487

5488

5489

5490

5491

5492

5493

5494

5495

5496

5497

5498

5499

5500

5501

5502

5503

5504

5505

5506

5507

5508

5509

5510

5511

5512

5513

5514

5515

5516

5517

5518

5519

5520

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

ADD	=	Average Daily Dose (mg/kg/day)

SWC	=	Surface water (dissolved) concentration (|ig/L)

BAF	=	Bioaccumulation factor (L/kg wet weight)

IR	=	Fish ingestion rate (g/day)

CF1	=	Conversion factor (0.001 mg/|ig)

CF2	=	Conversion factor for kg/g (0.001 kg/g)

ED	=	Exposure duration (year)

AT	=	Averaging time (year)

BW	=	Body weight (80 kg)

The years within an age group (i.e., 54 years for adults) was used for the exposure duration and
averaging time to characterize non-cancer risks. For cancer, the years within an age group was also used
for the exposure duration while the averaging time is 78 years (i.e., lifetime).

A BAF is preferred in estimating exposure because it considers the animal's uptake of a chemical from
both diet and the water column. For TCEP, there are multiple wet weight BAF values reported for whole
fish collected from water bodies that contained TCEP (Table 2-2). The modeled surface water
concentrations were converted to fish tissue concentrations using the upper and lower bound of the
BAFs reported in literature: 2,198 L/kg wet weight for walleye (Sander vitreus) collected from the U.S.
Great Lakes (Guo et al. 2017b) and 109 L/kg wet weight for mud carp collected from an e-waste
polluted pond in China (Liu et al.. 2019a). While Guo e*	is the only U.S. study that measured

TCEP concentrations in fish samples and is presumably more representative of subsistence fisher in the
United States, EPA considered BAF values from non-U. S. studies because of uncertainties with
walleye's BAF and subsistence fishers consume more than just one fish species. As a result, BAF from
non-U.S. studies were considered.

Table 5-31 compares the fish tissue concentration calculated from the scenario-specific modeled surface
water concentrations using the two BAFs with measured fish tissue concentrations obtained from
literature. For comparison, Table 5-31 also includes fish tissue concentrations presented in Table 4-1
that were derived from a BCF. The overall range for scenario-specific fish concentrations based on
modeled concentrations is for wet weight, and monitoring studies reported both wet and lipid weight.
While the lipid content was not available to convert from lipid to wet weight, measured fish tissue
concentrations are still several orders of magnitude lower than that derived from modeled surface water
concentrations and BAF or BCF.

Table 5-31. Fish Tissue Concentrations Calculated from Modeled Surface Water Concentrations
and Monitoring Data			

Data
Approach

Data Description

Surface Water
Concentration (jig/L)

Fish Tissue
Concentration
(jig/kg)

Modeled

Surface

Water

Concentration

BAF (2,198) and the maximum
1-day average dissolved water
concentrations from PSC under
harmonic mean flow conditions

Overall range
3.4E01 to 4.8E03

Overall range
7.6E04 to 1.06E07, ww

BAF (109) and the maximum 1-
day average dissolved water
concentrations from PSC under
harmonic mean flow conditions

Overall range
3.4E01 to 4.8E03

Overall range
3.8E03 to 5.3E05, ww

Page 216 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Data
Approach

Data Description

Surface Water
Concentration (jig/L)

Fish Tissue
Concentration
(Jig/kg)



BCF and the maximum 1-day
average dissolved water
concentrations from PSC under
7Q10 flow conditions

Overall range
9.6E01 to 1.09E04

Overall range
3.2E01 to 3.71E03, ww

Fish Tissue
Monitoring
Data (Wild-
Caught)

7 studies with over 200 fish
tissue samples collected from 7
countries, including one U.S.
studv bv Guo et al. (2017b)

Only one non-U.S. study
collected water samples
from the same waterbody
and at the same time as the
fish tissue samples. Surface
water concentrations for
that study ranged from
1.5E-02to 2.34E-01

Central tendency range for

U.S. study
6.55E00 to 3.56E01, lw
Overall range among non-

U.S. studies
ND to 2.96, ww
ND to 1.87E02, lw

5521

5522	The exposures calculated using the modeled scenario-specific surface water concentrations and two

5523	BAFs are presented in Table 5-32.

5524

Page 217 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

5525	Table 5-32. Adult General Population Fish Ingestion Doses by Scenario Based on a Production Volume of 2,500 lb/year and High-End

5526	Release Distribution



SWCfl
(Hg/L)

ADR6
(mg/kg-dav)

ADD'' (mg/kg-dav)

LADD'' (mg/kg-dav)

Scenario Name

BAF

BAF

BAF

BAF

BAF

BAF



2,198

109

2,198

109

2,198

109





CT

HE

CT

HE

CT

HE

CT

HE

CT

HE

Import and Repackaging

8.62E02

5.25E-01

2.60E-02

1.19E-01

5.25E-01

5.92E-03

2.60E-02

8.26E-02

3.63E-01

4.10E-03

1.80E-02

Incorporation into Paints

3.82E03

2.33E00

1.15E-01

5.29E-01

2.33E00

2.62E-02

1.15E-01

3.66E-01

1.61E00

1.82E-02

7.98E-02

and Coatings - 1-Part























Coatings























Incorporation into Paints

3.46E03

2.11E00

1.05E-01

4.80E-01

2.11E00

2.38E-02

1.05E-01

3.32E-01

1.46E00

1.65E-02

7.24E-02

and Coatings - 2-Part























Reactive Coatings























Use in Paints and

2.03E03

1.24E00

6.13E-02

2.81E-01

1.24E00

1.39E-02

6.13E-02

1.95E-01

8.55E-01

9.65E-03

4.24E-02

Coatings at Job Sites























Formulation of TCEP

4.84E03

2.95E00

1.46E-01

6.71E-01

2.95E00

3.33E-02

1.46E-01

4.64E-01

2.04E00

2.30E-02

1.01E-01

Containing Reactive
Resin























Laboratory Chemicals

3.46E01

2.10E-02

1.04E-03

4.78E-03

2.10E-02

2.37E-04

1.04E-03

3.31E-03

1.46E-02

1.64E-04

7.22E-04

" Surface water concentrations based on harmonic mean flow conditions.















h ADR calculated using the 90th percentile fish ingestion rate (22.2 g/day). ADD and LADD were calculated using both the mean and 90th percentile fish ingestion rates,
5.04 g/day and 22.2 g/day respectively. An ADD based on the 90th percentile ingestion rate is the same as an ADR.

5527

Page 218 of 572

-------
5528

5529

5530

5531

5532

5533

5534

5535

5536

5537

5538

5539

5540

5541

5542

5543

5544

5545

5546

5547

5548

5549

5550

5551

5552

5553

5554

5555

5556

5557

5558

5559

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

5.1.3.4.3 Subsistence Fish Ingestion Exposure

Subsistence fishers represent a PESS group for TCEP due to their greatly increased exposure via fish
ingestion (142.4 g/day compared to a 90th percentile of 22.2 g/day for the general population) (U.S.

30b). The ingestion rate for subsistence fishers apply to only adults aged 16 to < 70 years. EPA
calculated exposure for subsistence fishers using Equation 5-18 and the same inputs as the non-
subsistence fisher except for the ingestion rate. Furthermore, unlike the general population fish ingestion
rates, there is no central tendency or 90th percentile IR for the subsistence fisher. The same value was
used to estimate both the ADD and ADR.

EPA is unable to determine subsistence fisher exposure estimates specific to younger lifestages based on
reasonably available information. The exposure estimates for an adult subsistence fisher in Table 5-33
were calculated using the array of modeled scenario-specific surface water concentrations and BAF.

Table 5-33. Adult Subsistence Fisher Doses by Scenario Based on a Production Volume of 2,500

b/year and High-End Release Distribution

Scenario Name

SWCa
(ug/L)

ADD, ADR

(mg/kg-day)
BAF 2,198

ADD, ADR

(mg/kg-day)
BAF 109

LADD

(mg/kg-day)
BAF 2,198

LADD

(mg/kg-day)
BAF 109

Import and repackaging

8.62E02

3.37E00

1.67E-01

2.34E00

1.16E-01

Incorporation into paints and coatings -
1-part reactive coatings

3.82E03

1.49E01

7.41E-01

1.03E01

5.13E-01

Incorporation into paints and coatings -
2-part reactive coatings

3.46E03

1.35E01

6.72E-01

9.38E00

4.65E-01

Use in paints and coatings at job sites

2.03E03

7.94E00

3.94E-01

5.50E00

2.73E-01

Formulation of TCEP containing reactive
resin

4.84E03

1.90E01

9.40E-01

1.31E01

6.51E-01

Laboratory chemicals

3.46E01

1.35E-01

6.70E-03

9.36E-02

4.64E-03

" Surface water concentrations based on harmonic mean flow conditions.

5.1.3.4.4 Tribal Fish Ingestion Exposure

Tribal populations represent another PESS group. In the United States there are a total of 574 federally
recognized American Indian Tribes and Alaska Native Villages and 63 state recognized tribes. Tribal
cultures are inextricably linked to their lands, which provide all their needs from hunting, fishing, food
gathering, and grazing horses to commerce, art, education, health care, and social systems. These
services flow among natural resources in continuous interlocking cycles, creating a multi-dimensional
relationship with the natural environment and forming the basis of Tamamvit (natural law) (Harper et al..
2012). Such an intricate connection to the land and the distinctive life ways and cultures between
individual tribes create many unique exposure scenarios that can expose tribal members to higher doses
of contaminants in the environment. However, EPA quantitatively evaluated only the tribal fish
ingestion pathway for TCEP because of data limitations and recognizes that this overlooks many other
unique exposure scenarios.

01 la) (Chapter 10, Table 10-6) summarizes relevant studies on tribal-specific fish IRs that
covered 11 tribes and 94 Alaskan communities. The highest mean IR per kilogram of body weight was
reported in a 1997 survey of adult members (16 years and older) of the Suquamish Tribe in Washington.
Adults reported a mean IR of 2.7 g/kg-day, or 216 g/day assuming an adult body weight of 80 kg. In

Page 219 of 572

-------
5560

5561

5562

5563

5564

5565

5566

5567

5568

5569

5570

5571

5572

5573

5574

5575

5576

5577

5578

5579

5580

5581

5582

5583

5584

5585

5586

5587

5588

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

comparison, the IRs for the adult subsistence fisher and general population are 142.2 and 22.2 g/day,
respectively. A total of 92 adults responded to the survey funded by ATSDR through a grant to the
Washington State Department of Health, of which 44 percent reported consuming less fish/seafood
today compared to 20 years ago. One reason for the decline is restricted harvesting caused by increased
pollution and habitat degradation (Duncan. 2000).

Because current fish consumption rates are suppressed by contamination, degradation, or loss of access,
EPA reviewed existing literature for IRs that reflect heritage rates. Heritage rates refer to those that
existed prior to non-indigenous settlement on tribal fisheries resources, as well as changes in culture and
lifeways (	•). Heritage IRs were identified for four tribes, all located in the Pacific

Northwest region, among available literature. The highest heritage IR was reported for the Kootenai
Tribe in Idaho at 1,646 g/day (Ridolfi. 2016) (that study was funded through an EPA contract). The
authors conducted a comprehensive review and evaluation of ethnographic literature, historical
accounts, harvest records, archaeological and ecological information, as well as other studies of heritage
consumption. The heritage IR is estimated for Kootenai members living in the vicinity of Kootenay
Lake in British Columbia, Canada; the Kootenai Tribe once occupied territories in parts of Montana,
Idaho, and British Columbia. It is based on a 2,500 calorie per day diet, assuming 75 percent of the total
caloric intake comes from fish and using the average caloric value for fish. Notably, the authors
acknowledged that assuming 75 percent of caloric intake comes from fish may overestimate fish intake.

EPA calculated exposure via fish consumption for tribes using Equation 5-18 and the same inputs as the
general population except for the IR. Two IRs were used: 216 g/day for current consumption and 1,646
g/day for heritage consumption. Similar to the subsistence fisher, EPA used the same IR to estimate both
the ADD and ADR. Limited information does report IRs specific to younger lifestages, but do indicate
that adults consume higher amounts of fish per kilogram of body weight. As a result, exposure estimates
are only provided for adults (Table 5-34).

Table 5-34. Adult Tribal Fish Ingestion Doses by Scenario Based on a PV of 2,500 lb/year, High-
End Release Distribution, and Two Fish Ingestion Rates 			

Scenario Name

swca

(ug/L)

ADD, ADR

(mjj/kjj-da.y)
BAF 2,198

ADD, ADR
(m«/k«-day)
BAF 109

LADD

(mjj/kjj-day)
BAF 2,198

LADD
(m«/k«-day)
BAF 109

Ciii'iviil 1110:111 fisli iimcsliim rule rcpurlcd b\ llie Sut|uaiiiisli Tribe (216 u'das)

Import and repackaging

S.oJ'LUJ'

rlJ'Luu

:^4L u i

v54Jjiu

L~oL Ul

Incorporation into paints and coatings -
1-part reactive coatings

3.82E03

2.27E01

1.12E00

1.57E01

7.78E-01

Incorporation into paints and coatings -
2-part reactive coatings

3.46E03

1.18E02

1.02E00

8.19E01

7.06E-01

Use in paints and coatings at job sites

2.03E03

6.94E01

5.97E-01

4.80E01

4.13E-01

Formulation of TCEP containing reactive
resin

4.84E03

1.66E02

1.43E00

1.15E02

9.87E-01

Laboratory chemicals

3.46E01

1.18E00

1.02E-02

8.18E-01

7.04E-03

1 Icriiauc lish mucslion rale < u da> i

Import and repackaging

8.62E02

2.95E01

1.46E00

2.04E01

1.01E00

Incorporation into paints and coatings -
1-part reactive coatings

3.82E03

1.31E02

6.47E00

9.04E01

4.48E00

Page 220 of 572

-------
5589

5590

5591

5592

5593

5594

5595

5596

5597

5598

5599

5600

5601

5602

5603

5604

5605

5606

5607

5608

5609

5610

5611

5612

5613

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Scenario Name

SWCa
(ug/L)

ADD, ADR

(mg/kg-day)
BAF 2,198

ADD, ADR

(mg/kg-day)
BAF 109

LADD

(mg/kg-day)
BAF 2,198

LADD

(mg/kg-day)
BAF 109

Incorporation into paints and coatings -
2-part reactive coatings

3.46E03

1.18E02

5.87E00

8.19E01

4.06E00

Use in paints and coatings at job sites

2.03E03

6.94E01

3.44E00

4.80E01

2.38E00

Formulation of TCEP containing reactive
resin

4.84E03

1.66E02

8.21E00

1.15E02

5.68E00

Laboratory chemicals

3.46E01

1.18E00

5.86E-02

8.18E-01

4.05E-02

" Surface water concentrations based on harmonic mean flow conditions.

5.1.3.4.5 Incidental Oral Ingestion from Soil

Average Daily Doses (ADD) were calculated for TCEP ingestion using the following formula:

Equation 5-19

ADD =

C x IR x EF x ED x CF

BW x AT

Where:

ADD

C

IR

EF

CF

BW

AT

Average Daily Dose (mg/kg/d)

Soil Concentration (mg/kg)

Intake Rate of contaminated soil (mg/d)

Exposure Frequency (d)

Conversion Factor (10x 10~6 kg/mg)

Body Weight (kg)

Averaging time (non-cancer: ED x EF, cancer: 78 years x EF)

Modeled soil concentrations were calculated from 95th percentile air deposition (see Section 3.3.3.2)
concentrations for 100 m and 1,000 m from a hypothetical facility. These calculations were conducted
for the COM-Paints-USE scenario (LOW PV - 2,500 lb, HE-95th percentile release).

The mean intake rate for children aged 3 to 6 years varies; 41 mg/d was selected for the mean intake rate
and 175.6 was selected for the 95th percentile intake rate (	). Body weight (18.6 kg) of a

3- to 6-year-old was estimated from the Exposure Factors Handbook (	).

Table 5-35. Modeled Soil Dermal Doses for the Commercial Use of Paints and Coatings OES for

Children for the 2,500 lb

ligh-End J

telease Estimates

OES

Distance
(m)

95th
Percentile Soil
Concentration

(ng/g)

Average Daily Dose
(Mean Intake)
(mg/kg-day)

Average Daily Dose
(95th Intake)
(mg/kg-day)

Use in paints and
coatings at job sites

100

1.14E04

2.51E-02

1.08E-01

1,000

8.65E01

1.91E-04

8.16E-04

Page 221 of 572

-------
5614

5615

5616

5617

5618

5619

5620

5621

5622

5623

5624

5625

5626

5627

5628

5629

5630

5631

5632

5633

5634

5635

5636

5637

5638

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

5.1.3.4.6 Incidental Oral Ingestion from Swimming

The general population may swim in affected surfaces waters (streams and lakes) that are affected by
TCEP contamination. Modeled Surface water concentrations from EFAST 2014 were used to estimate
acute doses and average daily doses due to ingestion exposure while swimming.

The following equations were used to calculate incidental oral (swimming) doses for all COUs, for
adults, youth, and children:

Equation 5-20

ADR =

SWC xIRx CF1

BW

Equation 5-21

ADD =

SWC xIRxEDxRD x CF 1
BW x AT x CF2

Where:

ADR

ADD

SWC

IR

RD

ED

BW

AT

CF1

CF2

Acute Dose Rate (mg/kg/day)

Average Daily Dose (mg/kg/day)

Surface water concentration (ppb or |ig/L)
Daily ingestion rate (L/day)

Release days (days/yr)

Exposure duration (years)

Body weight (kg)

Averaging time (years)

Conversion factor (l.OxlO-3 mg/|ig)
Conversion factor (365 days/year)

Page 222 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

5639	A summary of inputs utilized for these estimates are present in Appendix H.

5640

5641	Table 5-36. Modeled Incidental Oral (Swimming) Doses for All COUs, for Adults, Youth and Children, for the 2,500 lb High-End

5642	Release Estimate

OES"

Surface Water Concentration

Adult (>21 vrs)

Youth (11-15 vrs)

Child (6-10 vrs)

30Q5
Concentration
(HS/L)

Harmonic Mean
Concentration
(HS/L)

ADRpot
(m»/k«-
-------
5644

5645

5646

5647

5648

5649

5650

5651

5652

5653

5654

5655

5656

5657

5658

5659

5660

5661

5662

5663

5664

5665

5666

5667

5668

5669

5670

5671

5672

5673

5674

5675

5676

5677

5678

5679

5680

5681

5682

5683

5684

5685

5686

5687

5688

5689

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

5.1.3.4.7 Human Milk Exposure

Infants are a potentially susceptible population because of their higher exposure per body weight,
immature metabolic systems, and the potential for chemical toxicants to disrupt sensitive developmental
processes, among other reasons. To determine whether a quantitative analysis of infant exposure to
TCEP via human milk could be informative, EPA considered available exposure and hazard information
for TCEP. Based on its slight lipophilicity and small mass, TCEP has the potential to accumulate in
milk. In fact, available biomonitoring studies demonstrated the presence of TCEP in human milk. The
highest concentrations were observed by Kim et al. (20141 in which TCEP was measured in 89 milk
samples collected in three Asian countries (Philippines, Japan, Vietnam), ranging from non-detect to 512
ng/g lipid weight, with an average of 0.14 to 42 ng/g. Another study by Sundkvist et al. (2010) collected
milk samples from 286 mothers in Sweden, where concentrations ranged from 2.1 to 8.2 ng/g lipid
weight, with a median of 4.9 ng/g. One study by (He et al.. 2018a) collected three milk samples in
Australia, and concentrations ranged from non-detect to 0.47 ng/mL wet weight. No U.S. biomonitoring
studies on TCEP in human milk were identified.

The hazard endpoints identified for TCEP (neurotoxicity for acute scenarios; reproductive toxicity for
short-term/chronic scenarios as well as carcinogenicity) are relevant for the milk pathway and are
protective of effects that may occur in infants as described in Section 5.2. Because TCEP can transfer to
human milk and infants may be particularly susceptible to its health effects, EPA further evaluated
infant exposures through the milk pathway for specific COUs.

EPA considered all maternal groups—occupational, consumer, and general population—when modeling
milk concentrations. Maternal doses are presented in Section 5.1 for occupational, Section 5.1.2.3 for
consumer, and Section 5.1.3 for general population.

Milk concentrations were estimated based on the maternal doses using a multi-compartment
physiologically based pharmacokinetic (PBPK) model identified by EPA as the best available model
(Verner et al.. 2009; Verner et al.. 2008). hereafter referred to as the Verner model. Only chronic, and
not acute, maternal doses were considered because the model is designed to estimate only continuous
maternal exposure. For more information on the Verner model, including modeled compartments, data
input requirements, and its system of differential equations, refer to Appendix H.

The Verner Model requires all maternal doses to be entered as oral doses. For consumers, CEM provides
inhalation estimates as an internal oral dose; therefore, no route-to-route extrapolation was necessary.
The only adjustment for maternal consumer doses was to account for body weight differences. CEM
assumes a body weight of 80 kg, which is less representative of women of reproductive age because it
combines males and females. To derive a dose representative of women of reproductive age, EPA
applied an adjustment factor of 1.21 based on a body weight of 65.9 kg (80 kg/65.9 kg) (U.S. EPA.
201 la). The body weight of 65.9 kg is for women 16 to 21 years of age. Body weight increases with age
for women of childbearing age, thus reducing overall exposure estimates. As a result, 65.9 kg is the most
health protective. Furthermore, only chronic maternal doses from consumer scenarios were considered
because TCEP is primarily found in consumer articles that are typically used over a long-time frame.

For occupational exposure scenarios, high-end inhalation concentrations were converted to oral
equivalent doses using the following equation:

Page 224 of 572

-------
5690

5691

5692

5693

5694

5695

5696

5697

5698

5699

5700

5701

5702

5703

5704

5705

5706

5707

5708

5709

5710

5711

5712

5713

5714

5715

5716

5717

5718

5719

5720

5721

5722

5723

5724

5725

5726

5727

5728

5729

5730

5731

5732

5733

5734

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Equation 5-22

Inhalation Cone x ED x IR

Oral Equivalent Dose =

BW

Where:

Oral Equivalen t Dose
Inhalation Cone

In mg/kg-day

Inhalation concentration (mg/m3)
8-hour TWA (high-end) for workers
Inhalation rate 1.25 m3/hr for workers
Body weight (65.9 kg)

ED

IR
BW

For workers, maternal dermal doses include both chronic (ADD) and subchronic (SCADD). The
SCADC represents repeated exposure for 30 days or more. Dermal ADD and SCADD from high-end
exposure levels for workers without personal protective equipment (PPE) {i.e., gloves) were used to
estimate infant exposure. These values are presented in Section 5.1 and adjusted by body weight.
Inhalation ADD and SCADD were calculated using Equation 5-23.

Equation 5-23

For consumers and workers, maternal doses were combined across all exposure routes for each COU:
inhalation (using the oral equivalent dose calculated with Equation 5-22 and Equation 5-23), dermal,
and/or oral routes. For general population, maternal doses were not combined because certain exposure
pathways {i.e., fish ingestion and undiluted drinking water) demonstrated significantly higher doses than
others and will likely be the main driver of risk. EPA focused on these sentinel exposure pathways.

EPA used 30 years as the age of pregnancy throughout the human milk pathway. This parameter is
applicable to chemicals that accumulate over time. TCEP, being only slightly lipophilic and having a
half-life of less than 24 hours, is not expected to accumulate. Initial model simulations that varied the
age of pregnancy confirmed this expectation. A sensitivity analysis also showed that maternal age had a
negligible effect (see Appendix H).

Infant doses are calculated using the modeled milk concentrations and milk intake rates described in the
Agency's Exposure Factors Handbook (	£01 la) for multiple age groups within the first year

of life. The handbook presents a mean and upper (95th percentile) milk intake rate for each age group,
and infant doses were calculated using both ingestion rates. The model estimated an average dose for
each age group and each milk ingestion rate.

Appendix H.4.4 presents the average infant doses via the human milk pathway for all COUs within each
maternal group, as well as the range of modeled milk concentrations.

D xEFxEY

ADD or SCADC =

Where:

D

EF
EY
ATef
ATey

Oral-equivalent inhalation dose from Equation 5-22 (mg/kg-day)

Exposure frequency (days/yr) (22 days/year for SCADD, 250 days/year for ADD)

Working years (1 year for SCADD, 40 years for ADD)

Averaging time for exposure frequency (30 days for SCADD, 365 days for ADD)
Averaging time for exposure years (1 year for SCADD, 40 years for ADD)

Page 225 of 572

-------
5735

5736

5737

5738

5739

5740

5741

5742

5743

5744

5745

5746

5747

5748

5749

5750

5751

5752

5753

5754

5755

5756

5757

5758

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

5.1.3.4.8 Dietary Exposure (non-TSCA)

For general population exposure, literature values indicate dietary exposure from all food groups based
on monitoring data (Table 5-37). The exposure dose associated with ingesting food can be derived by
multiplying the concentration of chemical in food by the ingestion rate for that food and dividing by
body weight (U.S. EPA. 1992). Within this overall framework, exposures could be estimated by
grouping all foods and liquids together and using a generic overall exposure factor, disaggregating
discrete food groups, and using food group specific exposure factors, or estimating exposures for unique
food items.

Other EPA programs such as the Office of Pesticides (OPP) estimates exposure from food from using
two distinct pieces of information: the amount of a pesticide residue that is present in and on food {i.e.,
residue level), and the types and amounts of foods that people eat {i.e., food consumption). Residue
levels are primarily developed via crop field trials, monitoring programs, use information including the
percent of crop treated, and commercial and consumer practices such as washing, cooking, and peeling
practices. Various sources provide food consumption data, including the USDA's continuing survey of
Food Intake by Individuals (CSFII), the National Health and Nutrition Examination Survey (NHANES),
What We Eat in America (WWEIA). OPP uses the Dietary Exposure Evaluation Model - Food
Commodity Intake Database (DEEM-FCID) model to estimate dietary exposures. CEPA-HQ-OPP-2007-
0780-0001: DEEM-FCIDY

For this risk evaluation, EPA used available monitoring data to estimate central tendency and high-end
concentrations of TCEP in specific food groups. Figure 5-9 provides the monitoring concentrations of
TCEP in various food groups.

Page 226 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE

December 2023

US Wet

NonUS Wet

659041 - Fda, 1995 - US

5423396 - He et al..
5423396 - He et al„
5423396 - He el al.,
54233% • He et al..
5423396 - He et al..
5423396 - He et al..
5423396 - He et al..
4292130-Pomaet al.,
4292130-Pomaet al.,
4292130 - Poma et al.,
4292130 -Poma el al.,
4292130-Poma el al.,
4292130-Pomaet al.,
4292130-Poma el al.,
4292130- Poma el al,
5166285-Poma et al
5166285-Pomaetal.,
5166285 - Poma et al.,
5166285 - Pomaet al,
5166285 - Poma et al.
5166285 - Poma et al.,
5166285 - Poma et al.,
5166285 - Poma et al„
5166285 - Poma et al.

2018 -AU
2018-AU
2018-AU
2018-AU
2018-AU
2018-AU
2018-AU
2018 - BE
2018 - BE
2018-BE
2018 - BE
2018- BE
2018 - BE
2018 - BE
2018-BE
.2017 - SE
.2017 - SE
2017-SE
.2017 - SE
. 2017-SE
2017-SE
.2017 - SE
.2017 - SE
2017-SE

lOM

¦	fruit

¦	dairy

¦	fish ami shell fish

Srain

=iH— meat

non-dairy beverages
¦ Jib i other
¦¦¦ vegetables

baby food-infant formula

¦	fats and oils

y Lognormal Distribution (CT and ^Oth percentile)
A Normal Distribution (CT ami 90th percentile)
8 Non-Dctect



K^A

^7

*

10**4

0.001

0.01	0.1	I

Concentration (ng/g)

100

Figure 5-9. Concentrations of TCEP (ng/g) in the Wet Fraction of Dietary from 1982 to 2018

Page 227 of 572

-------
5762

5763

5764

5765

5766

5767

5768

5769

5770

5771

5772

5773

5774

5775

5776

5777

5778

5779

5780

5781

5782

5783

5784

5785

5786

5787

5788

5789

5790

5791

5792

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table 5-37. Concentrations of Foods Found in the Monitoring Literature in ng/g

Food Type

Count of Estimates from
All Studies (n)

Average of Arithmetic Mean
Estimates for All Data

Average of 90th Percentile
Estimates for All Data

Baby food/formula

1(17)

4.0E-01

6.2E-01

Dairy

3(45)

8.7E-02

1.3E-01

Fats and oils

1(10)

2.6E00

4.0E00

Fish and shellfish

1(53)

1.4E-01

3.2E-01

Fruit

1(5)

7.5E-02

9.8E-02

Grain

2(19)

2.3E-01

4.9E-01

Meat

2(50)

3.0E-02

4.7E-02

Vegetables

2(24)

1.4E-01

4.8E-01

Other

2(14)

1.9E-01

2.9E-01

Equations

The equation used to calculate the chronic dose for each age group due to dietary exposure of fruits,
grains, vegetables, meat, dairy, fats, and seafood is presented in Equation 5-24 below.

Equation 5-24

FCxIRx ED

Where:

ADD =

Average daily dose used for chronic non-cancer risk calculations due to ingestion



food group (mg/kg-day)

FC

TCEP concentration in food group (mg/g)

IR

Food group ingestion rate by age group (g/kg bw-day)

ED

Exposure duration

AT =

Averaging time

An Australian study indicated that more than 75 percent of the estimated daily intake of TCEP came
from dietary ingestion (4.1 out of 4.9 ng/kg bw/day). This study reported that grains (oatmeal, pasta,
bread) contributed 39 percent and nonalcoholic beverages contributed 32 percent of total TCEP intake
(He et at.. 2018b). Pom a et al. (2018) measured TCEP in different food groups in Belgium. In total they
found food intake of TCEP to be 207 ng/d and 2.8 ng/kg/day. TCEP was most concentrated in fats (49
ng/d) and grains (49 ng/d), followed by milk (3 1 ng/d), meat (23 ng/d), and cheese (23 ng/d). Pom a et al.
(7 suggests that the dietary intake was dominated by fats food group because of the inclusion of the
fish oil supplement fat food group, for which a total of 19 g/d was estimated.

5.1.3.5 Exposure Reconstruction Using Human Biomonitoring Data and Reverse
Dosimetry

EPA describes the approach used to estimate doses based on biomonitoring below. TCEP has been
quantified in human samples in hair, nails (Liu et al.. 2016; Liu et al.. 2015). blood serum, plasma (Zhao
et al.. 2017). urine (Figure 5-10), and human milk (Section 5.1.3.4.7).

Page 228 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

5793

5794

5795

5796

5797

5798

5799

NonUS Unadjusted

5469782 - He ct al, 2018 - AU
5562397 - Basiiaensen el al.. 2019 - BE
3020426 - Van Den Eede et al. 2015 - AU







| General Population (Background)





0

11 10 100 1000 10*4 10*5

Concentration (ng/L)

Figure 5-10. Concentrations of TCEP (ng/L) in the Unadjusted Urine from 2015 to 2019

BCEP, a metabolite of TCEP, has been reported in the 2011 to 2014 NHANES data (CDC. 20131 as
well as the peer-reviewed literature (Wang et al.. 2019d; He et al.. 2018a; Dodson et al.. 2014) (Figure
5-11, Figure 5-12).

US Creatinine Adjusted
US Unadjusted
NonUS Unadjusted

¦ General Population (Background)
y Lognornial Distribution (CT and 90ih percentile)

5164613 - Wang et al. 2019 - US

2533847 - Dodson ct al. 2014 - US

C7 V

5469782 - He et al. 2018 - AU
2537005 - Frommc ct al. 2014 - DE

0.1

5800

5801

5802

100	1000

Concentration (ng/L)

10*4

10*5

Figure 5-11. Concentrations of BCEP (ng/L) in the Creatinine-Adjusted Urine from 2014 to 2019

Page 229 of 572

-------
5803

5804

5805

5806

5807

5808

5809

5810

5811

5812

5813

5814

5815

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE

December 2023

Urinary Bis(2-chloroethyl) phosphate (BCEtP) (creatinine corrected) (2011 - 2014)

CAS Number 3040-56-0

Metabolite of Tris(2-chioroethy!) phosphate (TCEtP)

Geometric mean and selected percentiles of urine concentrations (in mqj'S of creatinine) for the U.S. population from the
National Health and Nutrition Examination Survey.

EJemo graphic

Survey

Geometric Mean

50th Percentile

75th Percentile

90th Percentile

95th Percentile

Sample

Categories

{Years}

(95% CI)

(95% CI)

(951% CI)

(95% CI)

(95% CI)

Size

Total population

11-12

0.481 (.443-.545)

.433 (.441-558)

.960 (-311-1.11)

2.11 (1.02-2.35)

3.30 (2.86-3.79)

2400

Total population

13-14

0.447 (.396-.505)

.853 (,337-.444)

.650 ( 743-361)

2.03(1.72-2.38)

3.04 (2.74-5.13)

2649

Age 6-11 yea's

11-12

0.968 (.B06-1.16)

.855 (.724-1.13)

1.88(1.51-2.14)

4.22 (2.S3-5.44)

6.77 (4.22-15.6)

394

Age 8-11 years

13-14

D.855 (.720-1.02)

.833 (.076-.931)

1.00(1.18-2.12)

4.25(3.39-5.43)

6.83 (4.97-8-99)

418

Age 12-19 yea's

11-12

0.574 (.433-.780)

.537 (.404-.000)

1.23 (.758-1.90)

3.11 (1.90-5.15)

5.15 (2.74-9.05.)

386

Age 12-10 years

13-14

D.510 (.429-020)

.442 (.350-.568)

1.D6 (763-1.38)

2.33 (1.7D-3.Q3)

4.48 (2.42-0.77)

423

Age 20+ years

11-12

O.+io (.306- 501)

.457 (.30B-.524)

.655(748-1.01)

1.87 (1.60-2.09)

2.80 (2.20-3.49)

1620

Age 20+ years

13-14

0.408 (.302-.46O)

.348 (.313-.383)

742 (.632-.S75)

1.87(1.42-2.31)

3.12 (2.38-4.69)

1508

Maes

11-12

0.449 (.413-.489)

.449 (.4D0-.506)

.805 (.779-1.02)

2.07(1.77-2.43)

3.2B (2.58-4.15)

1217

Maes

13-14

0.42 (.370-.476)

.373 (.322-.4Q0)

.820 (T25-.954)

2.01 (t .50-2.43)

3.70 (2.44-5.50)

1336

FeTiat^es

11-12

0.534 (.466-.612)

.534 (.404-.021)

1.04(879-122)

2.14(1.02-2.46)

3.41 (2.70-4.48)

1192

Feinaes

13-14

0.476 (.417-.543)

.407 (.35O-.407)

.909 (742-1.04)

2.00(1.75-2.41)

3.99 (2.61-526)

1313

Mexican Amercans

11-12

D.482 1.347-.669)

.509 (.381-.000)

1.05 (.673-1.61)

2.18 0.40-3.12)

3.12(1.97-0 71)

286

Mexican Americans

13-14

0.515 (.304-.672)

.477 (.343-.037J

1.01 (.666-1.47)

2.35(1.57-3.03)

3.10(2.43-6.34)

426

Non-Hispanic Backs

11-12

D.537 (.480-.590)

.517 (.460-.595J

1.10 (.527-1-29)

2.43(1.07-2.08)

3.70 (3.08-023)

666

Non-Hispanic Blacks

13-14

0.374 (.321-.435)

.328 (.267-.450)

732 (.630-.867)

1.50 (1.1B-1.80)

2.41 (1.88-3.17)

578

NoM-lispanic Whites

11-12

0.406 I.4D7-.535)

.451 (.399-.503)

.900 (767-1.09)

1.92(1.01-2.34)

2.00(2.41-3.72)

776

Non-Hispanic Whites

13-14

0.446 (.393-.506)

.379 (.333-.43T)

.857 (.731-1,00)

2.03(1.64-2.44)

4.0B (2.51-5.58)

1012

Ail Hispanics

11-12

0.529 (.446-.626)

.523 (.450- 013)

1.09(819-1 41)

2.45(1.97-2.94)

3.43(2.52-521)

552

All Hispanics

13-14

0.495 (.406-.604)

.472 (.371-.585)

.980(736-1.36)

2.27(1.69-275)

3.14 (253-3.94)

666

Asians

11-12

o.eoe (.512-.716)

.567 (.473-732)

1.20 (1.07-1 56)

2.77(2.11-3.62)

478 (2.77-7.50)

327

Asians

13-14

0.477 (.412-.553)

.442 (.371-500)

.702 (-606-128)

2.33(1.51-3.46)

4.1B (2.76-0.34)

281

Figure 5-12. Concentrations of BCEP from NHANES data for the U.S. Population from 2011 to
2014

TCEP has also been detected in personal hand wipes and wristbands (Figure 5-13, Figure 5-14). Xu et
al. (2016) calculated dermal absorption daily doses at a mean of 0.088 ng/kg/day.

m

5163584 - Phillips ci al. 2018 - US



1 General Population (Background}



2343712 - Staplclon ct al.. 2014 - US

NonUS

4292136 - Larsson el al.. 2018 - SE
3357642 - Xu el al., 2016 • NO













IC





0.1

10 100 1000 !0A4
Concentration (ng/wipe)

Figure 5-13. Concentrations of TCEP (ng/wipe) in Surface Wipes from 2014 to 2018

US

5165046 - Gibson et al., 2019 - US



B General Population (Background)
gj Non-Detect







^¦1





3361031 - Kilcet al. 2016-US

*







0.1

1

10 100
Concentration (ng/g)

1000

Figure 5-14. Concentrations of TCEP (ng/wipe) in Silicone Wristbands from 2012 to 2015

Page 230 of 572

-------
5816

5817

5818

5819

5820

5821

5822

5823

5824

5825

5826

5827

5828

5829

5830

5831

5832

5833

5834

5835

5836

5837

5838

5839

5840

5841

5842

5843

5844

5845

5846

5847

5848

5849

5850

5851

5852

5853

5854

5855

5856

5857

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

TCEP human biomonitoring data were previously extracted from peer-reviewed studies and curated to
produce one set of summary statistics per study. A total of two peer-reviewed studies, resulting in 6
datasets with sampling years from 2014 to 2018, reported TCEP data in human hair, human nails, and
human urine for the U.S. general population. Additional data are available for occupational workers and
highly exposed populations (Mayer et al. 2021; Shen et ai. 2018; Javatilaka et ai. 2017). Researchers
from the CDC measured urine samples for BCEP in 76 members of the general population and 146
firefighters who performed structure firefighting while wearing full protective clothing and respirators.
BCEP was detected in 10 percent of the general population, but the median concentration was too low to
quantify with acceptable repeatability and accuracy. For firefighters, BCEP was detected in 90 percent
of firefighters at a median of 0.86 ng/mL (Javatilaka et al..: ). Table 5-38 provides the number of
datasets for the general population and media type in the United States.

Table 5-38. Human TCEP/BCEP U.S. Biomonitoring Datasets by Population,
Type, and Number			

Population

Media Type

No. of Datasets

General Population

Human Hair

2

General Population

Human Nails

1

General Population (BCEP)

Human Urine

3

Urinary BCEP was selected as a biomarker of exposure for TCEP. Urinary BCEP is a recommended
target for biomonitoring of TCEP (Dodsom et al.. 2014). Furthermore, the robust dataset provided by the
NHANES survey that varies results across demographics, age groups, and time and allows for more
confidence in the values calculated by the exposure reconstruction.

Urinary volume and flow can vary between individuals due to differences in hydration status. One
approach to account for this variability is by taking creatinine-adjusted values for urinary concentration.
The NHANES data already provides creatinine adjusted values and more information on this adjustment
can be referenced in their fourth report (CDC. 2013).

Equation 5-25

Crr * Ctp

_ —cr	e

BW * Fue

Where:

DI	=	Daily intake of the parent compound (mg/kg-day)

Cc	=	Creatinine adjusted concentration of analyte in urine (mg biomarker/g creatinine)

Cre	=	Creatinine excretion rate (g creatinine/day)

BW	=	Body weight (kg)

Fue	=	Urinary excretion fraction (mg biomarker excreted/mg parent compound intake)

Kinetic data on the metabolism of TCEP is limited. Literature values have suggested a Fue of 0.07 based
on in vitro human liver microsomes (HLM) experiment, and a value of 0.13 based on in vitro human
liver S9 fraction experiment (Van den Eede et al.. 2013).

The creatinine excretion rate was normalized by body weight (in units of mg creatinine per kg
bodyweight per day). Cre can be estimated from the urinary creatinine values reported in biomonitoring
studies {i.e., NHANES) using the equations of Mage et al. (2008). Assessments from Health Canada and

Page 231 of 572

-------
5858

5859

5860

5861

5862

5863

5864

5865

5866

5867

5868

5869

5870

5871

5872

5873

5874

5875

5876

5877

5878

5879

5880

5881

5882

5883

5884

5885

5886

5887

5888

5889

5890

5891

5892

5893

5894

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

U.S. Consumer Product Safety Commission (CPSC) have used similar approaches to quantifying
creatinine excretion rate (Health Canada. 2020; CHAP. 2014).

To simplify this analysis, a few excretion rates were selected for various age groups (250 mg/day at 3
years and 1,750 mg/day for a 20-year-old adult male) from the literature (Mage et ai. 2008). The 2013-
2014 urinary BCEP concentrations were selected as the most recent and representative concentrations
for the U.S. population. Using the geometric mean and the 95th percentile concentrations from the 2013
to 2014 NHANES data, the daily intakes are estimated in Table 5-39.

Table 5-39. Reconstructed Daily Intakes from Creatinine Adjusted Urinary BCEP Concentrations
from NHANES (2013-2014).		

Statistic

Fue

3-year-old Intake (mg/kg-day)"

20-year-old Intake (mg/kg-day)''

Geomean

0.13

0.119

0.069

95th Percentile

0.13

0.952

0.525

Geomean

0.07

0.221

0.128

95th Percentile

0.07

1.768

0.975

a 3-year-old has a BW of 13.8 kg, and Cre of 250 mg/d. Used 6-11 year data for NHANES value (0.855 jxg/g
geomean and 6.83 jxg/g 95th percentile) since no data for younger lifestages available.

b 20-year-old has a BW of 80 kg, and Cre of 1,750 mg/d. Used Adult data for NHANES value (0.408 jxg/g geomean
and 3.12 jxg/g 95th percentile).

Wane et al. ( i similarly calculated exposure doses of 19 volunteers from Albany, NY of the parent
TCEP using creatinine adjusted urinary concentrations of BCEP. Wane et al. (2019d) found TCEP doses
to range 1 1.9 (50th percentile) to 38.6 ng/kg-tnv/day. Parameters used by Wane et al. (2019d) included a
0.63 value for Fue based on literature values for BDCIPP, and daily urine excretion values of 20 mL/kg-
bw/day and 22.2 mL/kg-tnv/day for children. Nevertheless, Wane et al. (2019d) stratified TCEP
exposure doses by gender, ethnicity and age, and indicated that females (7.82 ng/kg-bw/day) had higher
doses than males (4.35 ng/kg-bw/day), Caucasians (8.52 ng/kg-bw/day) had higher doses than Asians
(4.59 ng/kg-bw/day), and individuals aged 40 and above (9.61 ng/kg-bw/day) had higher doses than
lower age groups.

5.1.3.6 Summary of General Population Exposure Assessment

The general population can be exposed to TCEP from inhalation of air; dermal absorption of soils and
surface waters; and oral ingestion of TCEP in drinking water, fish, and soils. Infants can also be exposed
to TCEP via mother's milk. The sentinel exposure scenario for general population exposures was fish
consumption. Oral ingestion estimates of fish consumption are provided for the general population and
subsistence fishing populations, as well as tribal populations, with high end and central tendency BAF in
Table 5-41 Table 5-41.

5.1.3.6.1 General Population Exposure Results

Table 5-40 provides a summary of the acute oral exposure estimates for non-diluted and diluted drinking
water. Table 5-41 provides a summary of the chronic oral exposure estimates for non-diluted and diluted
drinking water; drinking water estimates based on landfill leaching to groundwater; incidental ingestion
of ambient waters during swimming general population and subsistence fisherman fish ingestion
estimates; and 50th and 95th percentile soil intakes at 100 and 1,000 m from hypothetical facilities.

Table 5-42 provides a summary of acute and chronic dermal exposures estimates of dermal exposure to
surface water when swimming and exposure estimates of dermal exposure to chronic concentration of

Page 232 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

5895	TCEP in soils. Table 5-43 below provide a summary of the relevant acute, chronic, and lifetime

5896	exposures. These summary tables present oral, dermal, and inhalation exposures as a result

5897	environmental releases (air, water, and disposal releases) for the applicable OES.

Page 233 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

5898 Table 5-40. General Population Acute Oral Ingestion Estimates for Drinking Water Summary Table

Acute Oral Exposure Estimates (mjj/kjj day)

OES"

Drinking Water

Drinking Water (diluted)

Adult

(>21
Years)

Infant
(Birth to
<1 Year)

Youth

(16-20
Years)

Youth
(11—15

Years)

Child

(6-10
Years)

Toddler

(1-5
Years)

Adult

(>21
Years)

Infant
(Birth to
<1 Year)

Youth

(16-20
Years)

Youth
(11-15

Years)

Child

(6-10
Years)

Toddler

(1-5 Years)

Import

5.5E-02

1.9E-01

4.2E-02

4.2E-02

5.4E-02

6.9E-02

4.5E-05

1.6E-04

3.4E-05

3.4E-05

4.4E-05

5.6E-05

Incorporation
into paints and
coatings - 1-part
coatings

2.4E-01

8.3E-01

1.8E-01

1.8E-01

2.3E-01

3.0E-01

1.5E-04

5.2E-04

1.1E-04

1.1E-04

1.5E-04

1.9E-04

Incorporation
into paints and
coatings - 2-part
reactive
coatings

2.2E-01

7.6E-01

1.7E-01

1.7E-01

2.1E-01

2.7E-01

1.3E-04

4.7E-04

1.0E-04

1.0E-04

1.3E-04

1.7E-04

Use in paints
and coatings at
job sites

1.3E-01

4.5E-01

9.9E-02

1.0E-01

1.3E-01

1.6E-01

1.0E-04

3.7E-04

8.1E-05

8.1E-05

1.0E-04

1.3E-04

Formulation of
TCEP
containing
reactive resin

2.5E-01

8.8E-01

1.9E-01

1.9E-01

2.5E-01

3.1E-01

5.8E-04

2.0E-03

4.5E-04

4.5E-04

5.7E-04

7.3E-04

Use of

laboratory

chemicals

2.2E-03

7.7E-03

1.7E-03

1.7E-03

2.2E-03

2.8E-03

1.8E-06

6.3E-06

1.4E-06

1.4E-06

1.8E-06

2.2E-06

" Table 3-3 provides a crosswalk of industrial and commercial COUs to OES.

5899

5900

Page 234 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

5901 Table 5-41. Summary of General Population Chronic Oral Exposures

Oral (nig/kg/day)

OESa

Drinking

Water
(Diluted)

Drinking

Water

Drinking
Water (via
Leaehing to
Groundwater)

Ambient
Water
(ineidcntal
ingestion)

Soil Intake
(50th) at 100
m

Soil Intake
(95th) at 100
m

Soil Intake
(50th) at 1,000
m

Soil Intake
(95th) at 1,000
m

Repackaging of import containers

1.67E-08

2.60E-05

N/A

1.29E-05

1.24E-10

5.30E-10

1.58E-12

6.78E-12

Incorporation into paints and coatings - 1-
part coatings

6.20E-08

1.15E-04

1.29E-06

5.59E-05

3.89E-09

1.67E-08

3.44E-11

1.47E-10

Incorporation into paints and coatings - 2-
part reactive coatings

5.62E-08

1.04E-04

N/A

5.07E-05

5.63E-10

2.41E-09

7.42E-12

3.18E-11

Use in paints and coatings at job sites

3.92E-08

6.11E-05

N/A

3.04E-05

9.15E-06

3.92E-05

4.77E-08

2.04E-07

Formulation of TCEP containing reactive
resin

2.76E-07

1.46E-04

N/A

5.90E-05

6.19E-10

2.65E-09

7.90E-12

3.38E-11

Processing into 2-part resin article

N/A

N/A

1.29E-06

N/A

5.30E-09

2.27E-08

5.41E-11

2.32E-10

Use of laboratory chemicals

6.68E-10

1.04E-06

N/A

5.20E-07

5.94E-09

2.54E-08

6.50E-11

2.78E-10

OES

General Population (GP)

Subsistenee Fisher (SF)

Tribes (Currentb)

Tribes (Heritage1)



BAF 2198

BAF 109

BAF 2198

BAF 109

BAF 2198

BAF 109

BAF 2198

BAF 109

Import

5.25E-01

2.60E-02

3.37E00

1.67E-01

1.89E01

9.40E-01

2.95E01

1.46E00

Incorporation into paints and coatings - 1-
part coatings

2.33E00

1.15E-01

1.49E01

7.41E-01

8.40E01

4.16E00

1.31E02

6.47E00

Incorporation into paints and coatings - 2-
part reactive coatings

2.11E00

1.05E-01

1.35E01

6.72E-01

1.18E02

3.77E00

1.18E02

5.87E00

Use in paints and coatings at job sites

1.24E00

6.13E-02

7.94E00

3.94E-01

6.94E01

2.21E00

6.94E01

3.44E00

Formulation of TCEP containing reactive
resin

2.95E00

1.46E-01

1.90E01

9.40E-01

1.66E02

5.28E00

1.66E02

8.21E00

Processing into 2-part resin article

N/A

N/A

N/A

N/A

N/A

N/A

N/A

N/A

Use of laboratory chemicals

2.10E-02

1.04E-03

1.35E-01

6.70E-03

1.18E00

3.77E-02

1.18E00

5.86E-02

" Table 3-3 provides a crosswalk of industrial and commercial COUs to OES.

b Current fish consumption rate at 216 g/day based on survey of Suquamish Indian Tribe in Washington (Section 5.1.3.4.4).
c Heritage fish consumption rate at 1,646 g/day based on study of Kootenai Tribe in Idaho (Section 5.1.3.4.4).

5902

5903

Page 235 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

5904 Table 5-42. Summary Acute and Chronic General Population Dermal Exposures	

Dermal (mg/kg/day)

OESfl

Surface Water
(Swimming)

Soil Mud at 100 m

Soil Activity at 100 m

Soil Mud at 1,000 m

Soil Activity at 1,000 m

Repackaging of import containers

6.00E-06

3.93E-07

1.91E-09

5.02E-09

2.44E-11

Incorporation into paints and
coatings - 1-part coatings

2.60E-05

1.23E-05

6.00E-08

1.09E-07

5.30E-10

Incorporation into paints and
coatings - 2-part reactive coatings

2.40E-05

1.78E-06

8.68E-09

2.35E-08

1.14E-10

Use in paints and coatings at job
sites

1.40E-05

2.90E-02

1.41E-04

1.51E-04

7.36E-07

Formulation of TCEP containing
reactive resin

2.80E-05

1.96E-06

9.54E-09

2.50E-08

1.22E-10

Processing into 2-part resin article

N/A

1.68E-05

8.18E-08

1.71E-07

8.34E-10

Use of laboratory chemicals

2.41E-07

1.88E-05

9.16E-08

2.06E-07

1.00E-09

"Table 3-3 provides a crosswalk of industrial and commercial COUs to OES.

5905

5906	Table 5-43. Summary of General Population Inhalation Exposures

Inhalation (jig/m3)

OES"

Ambient Air 50th

Ambient Air 95th

Repackaging of import containers

4.39E-10

1.12E-09

Incorporation into paints and coatings - 1-part coatings

1.35E-08

3.51E-08

Incorporation into paints and coatings - 2-part reactive coatings

2.29E-09

1.1IE—08

Use in paints and coatings at job sites

3.36E-05

8.21E-05

Formulation of TCEP containing reactive resin

2.52E-09

1.21E-08

Processing into 2-part resin article

1.96E-08

2.72E-08

Use of laboratory chemicals

2.24E-08

3.33E-08

a Table 3-3 provides a crosswalk of industrial and commercial COUs to OES.

5907

Page 236 of 572

-------
5908

5909

5910

5911

5912

5913

5914

5915

5916

5917

5918

5919

5920

5921

5922

5923

5924

5925

5926

5927

5928

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

5.1.3.7 Weight of the Scientific Evidence Conclusions for General Population
Exposure

Sections 5.1.3.2, 5.1.3.3, 5.1.3.4, and 5.1.3.5 summarize the direct and indirect exposure assessment
approaches taken to estimate general population exposures. A judgment on the weight of the scientific
evidence supporting the exposure estimate is decided based on the strengths, limitations, and
uncertainties associated with the exposure estimates. The judgment is summarized using confidence
descriptors: robust, moderate, slight, or indeterminate confidence descriptors.

EPA used general considerations (i.e., relevance, data quality, representativeness, consistency,
variability, uncertainties) as well as chemical-specific considerations for its weight of the scientific
evidence conclusions.

EPA modeled three routes of exposure: (1) inhalation from ambient air; (2) oral ingestion from drinking
water, fish ingestion, soil intake, and human milk intake; and (3) dermal exposures from surface water
and soil. Within each of these modeled pathways, EPA considered multiple variations in its analyses
{i.e., multiple distances for inhalation exposures, diluted vs non-diluted conditions for drinking water
exposures, high vs low BAF for fish ingestion) to help characterize the general population exposure
estimates and to explore potential variability. The resulting exposure estimates were a combination of
central tendency and high-end inputs for the various exposure scenarios. Modeled estimates were
compared with monitoring data to evaluate overlap, magnitude, and trends. Table 5-44 indicates the
confidence EPA has in their general population exposure estimates for each scenario.

Table 5-44. Overall Confidence for General Population Exposure Scenarios

Route

General Population Exposure Scenario

Confidence
(+ Slight, ++ Moderate, +++ Robust)

Oral

Drinking Water (diluted)

+++

Oral

Drinking Water

++

Oral

Drinking Water (via Leaching to Groundwater)

++

Oral

Surface Water (incidental ingestion)

++

Oral

Fish Ingestion (SF-HighBAF)

+

Oral

Fish Ingestion (GP-HighBAF)

+

Oral

Fish Ingestion (Tribal-HighBAF, Current or Heritage
Ingestion Rate)

+

Oral

Fish Ingestion (SF-LowBAF)

++

Oral

Fish Ingestion (GP-LowBAF)

++

Oral

Fish Ingestion (Tribal-LowBAF, Current or Heritage
Ingestion Rate)

++

Oral

Children's Soil Intake (50th) at 100 m

+

Oral

Children's Soil Intake (95th) at 100 m

+

Oral

Children's Soil Intake (50th) at 1,000 m

++

Oral

Children's Soil Intake (95th) at 1,000 m

++

Oral

Human Milk Intake

++

Dermal

Surface Water (swimming)

++

Dermal

Children playing in Mud at 100 m

+

Dermal

Children activities with Soil at 100 m

+

Page 237 of 572

-------
5929

5930

5931

5932

5933

5934

5935

5936

5937

5938

5939

5940

5941

5942

5943

5944

5945

5946

5947

5948

5949

5950

5951

5952

5953

5954

5955

5956

5957

5958

5959

5960

5961

5962

5963

5964

5965

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Route

General Population Exposure Scenario

Confidence
(+ Slight, ++ Moderate, +++ Robust)

Dermal

Children playing in Mud at 1,000 m

++

Dermal

Children activities with Soil at 1,000 m

++

Inhalation

Inhalation 100 m - MetCT

++

Inhalation

Inhalation 1,000 m - MetCT

+++

Inhalation

Inhalation 100 m - MetHIGH

++

Inhalation

Inhalation 1,000 m - MetHIGH

+++

5.1.3.7.1 Strengths, Limitations, Assumptions, and Key Sources of Uncertainty for
the General Population Exposure Assessment

No site-specific information was reasonably available when estimating release of TCEP to the
environment. Release estimates were provided for hypothetical sites. As such, there is considerable
uncertainty in the production volume estimate (2,500 lbs), and the resulting environmental release
estimates. In addition, there is uncertainty in the relevancy of the monitoring data to the modeled
estimates presented in this evaluation. Manufacturers have begun to phase out the use of TCEP as
demonstrated by the declining production volumes and the introduction of new regulations (e.g.,
California TB 117-2013) that have shifted the use away from TCEP and other organophosphate flame
retardants. For each release scenario, due to the lack of information on the distribution of TCEP across
industry sectors, it was assumed that the full production volume of 2,500 lbs was released for each COU.
This conservative assumption further contributes to the uncertainty when characterizing the resulting
modeled exposure estimates.

Drinking Water Estimates

Exposure estimates for the diluted drinking water estimates ranged from 0.022 to 9.167 ug/L which is 1-
2 orders of magnitude greater than the estimates found in the monitoring literature in the US: average of
4.9 ng/L and 90th percentile of 9.5 ng/L. The modeled estimates are more in line with a study of
drinking water systems from 19 drinking water systems across the US, where the median measured
concentrations of TCEP in finished water was 0.12 ug/L (Benotti et at.. 2009). There is uncertainty
surrounding the distance between release sites and drinking water intake locations. Nevertheless, the
assessment conducted analyses for diluted and undiluted drinking water estimates to account for this
uncertainty. Only 5 percent of surface water samples detected TCEP in the Water Quality Portal (see
Section 3.3.2.4).

The systematic review resulted in only a few cases demonstrating migration of TCEP to groundwater
from suspected landfill leachate (Buszka et at.. 2009; Barnes et at.. 2004; Hutchins et at.. 1984).
Furthermore, there are inherent uncertainties associated with estimating exposures from the transport of
chemicals through various media (e.g., landfill disposal to groundwater to drinking water). In addition,
TCEP was detected in only 2 percent of groundwater samples in the Water Quality Portal (see Section
3.3.3.6).

EPA has robust confidence in the diluted drinking water estimate, whereas EPA has moderate
confidence in the non-diluted drinking water estimates. EPA has slight confidence in the drinking water
estimates as a result of leaching from landfills to groundwater and subsequent migration to drinking
water wells.

Page 238 of 572

-------
5966

5967

5968

5969

5970

5971

5972

5973

5974

5975

5976

5977

5978

5979

5980

5981

5982

5983

5984

5985

5986

5987

5988

5989

5990

5991

5992

5993

5994

5995

5996

5997

5998

5999

6000

6001

6002

6003

6004

6005

6006

6007

6008

6009

6010

6011

6012

6013

6014

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Fish Ingestion Estimates

To account for the variability in fish consumption across the United States, fish intake estimates were
considered for both subsistence fishing populations and the general population. In estimating fish
concentrations, diluted surface water concentrations were not considered. It is unclear what level of
dilution may occur between the surface water at the facility outfall and habitats where fish reside. A
considerable source of uncertainty in the fish ingestion estimates was the selection of a bioaccumulation
factor (BAF). Two BAFs were considered (109 and 2198 L/kg wet weight) due to uncertainties with the
high end BAF value and to account for various fish species. No monitoring data were available
indicating the consumption of fish containing TCEP. EPA did find very limited monitoring data
indicating TCEP concentrations in fish tissue. The reported wet weight fish tissue concentrations in the
monitoring data are several magnitudes lower than the modeled estimates with either the low or high
BAF.

Soil and Swimming Ingest ion/Dermal Estimates

Two scenarios (children playing in mud and children conducting activities with soil) captured a wider
range of potential exposures to TCEP containing soils. EPA's Exposure Factors Handbook provided
detailed information on the child skin surface areas and event per day of the various scenarios (U.S.
E	). It is unclear how relevant dermal and ingestion estimates from soil exposure are as TCEP

is expected to migrate from surface soils to groundwater. Furthermore, there are inherent uncertainties
associated with estimating exposures from the transport of chemicals through various media (e.g., air to
land and subsequent soil ingestion and dermal absorption).

There are no recorded values of TCEP in soils in the US. A study in Germany reported highest
concentrations of TCEP in soil, 1 day after snow melt at 23.48 ng/g (Mihailovic and Frit	). The

95th percentile estimated modeled concentrations of soil because of air deposition for the use of paints
and coatings at job sites scenario was 1.14><104 ng/g at 100 m and 8.65X101 ng/g at 1000 m. The foreign
monitoring data is within range of the modeled soil estimates via air deposition. The child playing in
mud scenario assumes that the child will be exposed all over the arms, hands, legs, and feet.

Furthermore, there are uncertainties regarding the relevance of the selected dermal absorption fraction of
35.1 percent as discussed in the Section 5.1.2.4.1.

Non-diluted surface water concentrations were used when estimating dermal exposures to adults and
youth swimming in streams and lakes. TCEP concentrations will dilute when released to surface waters,
but it is unclear what level of dilution will occur when the general population swims in waters with
TCEP releases.

Inhalation

Modeled inhalation estimates are provided for a range of general population scenarios: various distances
from the emitting facility (10, 30, 60, 100, 1,000, 2,500, 10,000 m), two meteorology conditions (Sioux
Falls, South Dakota, for central tendency meteorology and Lake Charles, Louisiana, for higher-end
meteorology), central tendency and high-end release estimates for the low production volume (2,500
lbs), and 10th, 50th and 95th percentile exposure concentrations. Because no site-specific information
for TCEP release is available, EPA was unable to identify specific meteorological conditions that were
relevant to the air release.

Furthermore, EPA did not consider indoor to outdoor transfer of TCEP for general population inhalation
exposures. As discussed in Section 3.3.1.2.1, there are uncertainties surrounding the particle vs. gas
phase distribution of TCEP. It is unclear how sensitive this parameter is to the final inhalation and
deposition results. Use of paints and coatings at jobs sites was the OES with the highest modeled

Page 239 of 572

-------
6015

6016

6017

6018

6019

6020

6021

6022

6023

6024

6025

6026

6027

6028

6029

6030

6031

6032

6033

6034

6035

6036

6037

6038

6039

6040

6041

6042

6043

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

exposure estimates (8.21 x 10-5 ppm or 960 ng/m3) which is four orders of magnitude higher than the
average 90th percentile estimates for US data (3,1 x 10 1 ng/m3). Where information was unavailable,
EPA relied on AERMOD defaults when estimating inhalation exposures.

Reverse Dosimetry

Exposure estimates via reverse dosimetry provide an estimate of exposure based on biomonitoring
concentrations. Although NHANES provides nationally representative biomonitoring estimates, there is
no way to attribute the sources of TCEP to these biomonitoring estimates. NHANES only provided
urinary BCEP concentrations for the years 2011-2014. It is anticipated that these concentrations have
likely decreased due to the decrease in production volume and phase-out of TCEP to other alternatives.
In addition, there are modeling uncertainties associated with the reverse dosimetry calculation of
estimating internal TCEP doses from BCEP metabolite concentrations. Uncertainties include creatinine
adjustment and the accuracy of urinary excretion fraction. NHANES biomonitoring estimates do not
differentiate between TSCA and non-TSCA exposures. Hence, the reverse dosimetry estimates will be
an overestimate of the actual exposure levels due to TSCA COUs. The 95th percentile estimate for
TCEP intakes from reverse dosimetry is 1.8 mg/kg/day for children three years of age and 0.98 mg/kg/d
for adults 20 years of age. These reverse dosimetry estimates of TCEP were within an order of
magnitude of the highest general population, low BAF, oral fish intake estimates (0.33 mg/kg/day for
formulation of TCEP containing reactive resins OES). This corroboration builds confidence in the
plausibility of the general population fishing exposure estimates.

Key Variables, Parameters for General Population Assessment

Table 5-45 provides a list of key variables and parameters that influence the general population exposure
assessment. This table presents the sources of uncertainties and variabilities of key parameters for the
different exposure scenarios. For more detail on a comprehensive set of parameters used in the general
population exposure assessment, please see Appendix H.

Table 5-45. Qualitative Assessment of the Uncertainty and Variability Associated with General
Population Assessment 			

Variable Name

Relevant Section(s) in
Draft Risk Evaluation

Data Source(s)

Confidence
(Robust,
Moderate, Slight)

(icncral popnlalinn exposure assessmenl

Environmental release
estimates

0

EPA Modeled

+

Environmental monitoring
data

0

Extracted and evaluated data (all)
plus key studies

++

Fish intake rate

5.1.3.4.2, 0

(U.S. EPA. 2014a).
( a)

(Ridolfi. )

++

Exposure factors and activity
|\illeins

Appendix H

Exposure Factors Handbook
( )

+++

l\c\ iwamclci's lor modeling cn\ironmcnlal conccnlralions

Water modeling defaults:
river flow, dimensions,
characteristics

3.3.2.5, Appendix H

EFAST/VVWM -PSC defaults

++

Page 240 of 572

-------
6044

6045

6046

6047

6048

6049

6050

6051

6052

6053

6054

6055

6056

6057

6058

6059

6060

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Variable Name

Relevant Section(s) in
Draft Risk Evaluation

Data Source(s)

Confidence
(Robust,
ModernCe, Sliglil)

(icncml |n>|nikilmn exposure asscssmcnl

Air modeling defaults:
meteorological data,
indoor/outdoor transfer,

3.3.1.2, Appendix H

IIOAC/AERMOD defaults

++

Landfill leachate
concentrations and landfill
loading rates

3.3.3.7

DRAS defaults. (Masoner et aL

2016; Masoner et aL 2014b)

+

Drinking water treatment and
wastewater treatment removal

E.2.5.2, E.2.5.3, 2.2.2

("Life Sciences Research Ltd.
1990b. c)

(Padhve et aL, 2014; Benotti et

aL, 2009; Snvderet aL, 2006;
Westerhoff et aL. 2005;

Stackelbers et aL, 2004).

++

BAF

2.2, 5.1.3.4.2, 0

(Guo et aL, 2017b) and (Liu et aL,
2019a).

+ (high BAF)
++ (low BAF)

Gas phase vs. particulate
phase distribution, particle
size

3.3.1.2.1, Appendix H

(Okeme, 2018). fWolsclxke et aL,
2016).

++

1 luiiKin biomonilonng and iv\crsc dosimcli\ |\ii'amcleis

Biomonitoring data

5.1.3.5

Extracted and evaluated data (all)
plus key studies

++

Fraction of urinary excretion

5.1.3.5

( den Eede et aL, 2013).

++

Half-life in the body

Appendix H

litti3s://comDtox.eDa.aov/dasliboar

d/chemjM/adm^^

subtab/DTXSID5021411

++

Finally, EPA did not consider all possible exposure pathways, but rather focused on pathways that were
within the scope of its conceptual model. This may result in a potential underestimation of exposure in
some cases. Examples of exposure pathways that were not considered include incidental ingestion of
suspended sediment and surface water during recreational swimming and ingestion of non-fish seafood
such as aquatic invertebrates or marine mammals. However, EPA expects these exposures to be less
than those that were included in the overall assessment for the general population. As such, their impact
will likely be minimal and would be unlikely to influence the overall magnitude of the results.

5.1.3.7.2 Strengths, Limitations, and Key Sources of Uncertainty for the Human
Milk Pathway
Strengths of the Milk Model and Overall Approach

The Verner model integrates critical physiological parameters that includes pre- and postpartum changes
in maternal physiology, lactation, and infant growth. In addition, EPA implemented the Verner Model in
"R" to readily enable adjustments tailored to risk evaluation needs. For example, risk assessors can tailor
model inputs such as maternal doses to be more representative of women of reproductive age, thus
reducing the potential for underestimating infant doses. The overall approach to analyze infant exposure
through human milk also considers a wide range of data sources. It incorporates (1) available

Page 241 of 572

-------
6061

6062

6063

6064

6065

6066

6067

6068

6069

6070

6071

6072

6073

6074

6075

6076

6077

6078

6079

6080

6081

6082

6083

6084

6085

6086

6087

6088

6089

6090

6091

6092

6093

6094

6095

6096

6097

6098

6099

6100

6101

6102

6103

6104

6105

6106

6107

6108

6109

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

biomonitoring data (Section 5.1.3.4.7) on TCEP's potential transfer to human milk and its effects on
infants or development, (2) chemical properties influencing TCEP excretion in human milk, and (3) the
best available quantitative approaches for exposure. The half-life for TCEP was estimated using high-
throughput toxicokinetics, which predicts in vivo behavior based on in vitro measures from human
hepatocytes and plasma using simple toxicokinetics model (Wambaugh et at.. 2019). These
considerations were integrated into EPA's decision to proceed with a quantitative exposure analysis.

Uncertainty Associated with Predicting Accumulation in Milk

Well established criteria exist for predicting passive transport of chemicals across cell membranes,
including size, lipophilicity, water solubility, acid/base properties, and ionization. Nevertheless,
predictions of chemical accumulation via passive transport may be confounded by the pH gradient
between plasma and milk. The pH of human milk (7.08) is lower than plasma (7.42). Chemicals that are
weak acids or bases may accumulate to higher levels in milk than predicted based on passive diffusion
due to the pH gradient. For chemicals, the pH change can modify the molecular structure in a manner
that retards diffusion into the plasma medium that is more basic ( \l< ¦>» so-Am dot 2018; Wane and
Needham. 2007). It is not known if TCEP is subjected to ionization trapping because of the pH gradient.
Furthermore, it is not known whether TCEP is a substrate for active transporters in mammary epithelial
cells. These gaps in could introduce uncertainties in how much TCEP accumulates in milk, and thus an
infant's level of exposure.

Uncertainty in the Multi-compartment PBPK Model Inputs and Outputs

The multi-compartment PBPK model requires oral maternal doses. However, exposure can occur
through oral, dermal, and inhalation pathways for workers, consumers, and the general population.

While an inhalation-to-oral extrapolation of exposures was performed for TCEP to run the model,
differences in absorption potential and/or surface area between the lungs and gastrointestinal tract can
introduce uncertainties into the modeled milk concentrations. Also, enzymes involved in xenobiotic
metabolism are variably expressed across many organs and tissues, including sites of absorption such as
the gastrointestinal tract, lung, and skin (Bomifas and Blomeke. 2015; Lipworth. 1996). However, the
liver has the highest detoxification capacity in mammals (Schenk et ai. 2017). After oral administration,
xenobiotic chemicals absorbed from the gastrointestinal tract first pass through the liver before reaching
the systemic circulation. This "first-pass effect" may result in lower systemic bioavailability for
chemicals absorbed via the oral route compared to dermal and inhalation routes (Mehvar. 2018).
Therefore, route-to-route extrapolations may result in underestimating milk concentrations. For TCEP,
however, the effect on milk concentrations is expected to be small given its relatively slow clearance
rate {i.e., TCEP can partition to other parts of the body because it is not rapidly metabolized by the
liver).

Finally, a TCEP-specific source of uncertainty may derive from calculated rather than measured half-life
values and partition coefficients. See Table Apx H-12 in Appendix H for more information. The
calculated partition coefficients derive from Kow values, lipid and water fractions of blood and tissue,
and previously reported tissue compositions ("Verner et ai. 200N.	ai. 2003). The lack of

quantifiable uncertainty in these calculated values precludes a robust analysis of their contribution to
overall model uncertainty. However, a sensitivity analysis was conducted for TCEP to evaluate certain
chemical parameters' effects on model estimates. Overall, the model is sensitive to half-life where an
increase or decrease leads to a near equivalent change in the infant milk dose. Kow, which is used to
calculate partition coefficients, has a modest effect on the predicted infant dose. Infant doses are also
insensitive to alterations in milk lipid fraction. Appendix H.4.1 describes the results of the sensitivity
analysis in greater details.

Page 242 of 572

-------
6110

6111

6112

6113

6114

6115

6116

6117

6118

6119

6120

6121

6122

6123

6124

6125

6126

6127

6128

6129

6130

6131

6132

6133

6134

6135

6136

6137

6138

6139

6140

6141

6142

6143

6144

6145

6146

6147

6148

6149

6150

6151

6152

6153

6154

6155

6156

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Uncertainty and Variability Associated with Infant Exposure Dose: The Verner Model assumes
exclusive milk intake for the infant until the end of lactation for up to 12 months. It does not include a
weaning period where formula and/or solid foods are gradually introduced. Therefore, the model may
overestimate infant intake during periods of transition between human milk and formula or solid food
intake.

Weight of the Scientific Evidence for Human Milk Pathway

The weight of the scientific evidence judgement integrates various considerations to determine
confidence in the evaluation of infant's exposure to TCEP via human milk. The strengths of the Verner
PBPK Model are that it is peer-reviewed and well-documented (Verner et al. 2009; Verner et ai. 2008).
However, the model was not validated for TCEP because data were unavailable. It was validated using
data on persistent organic pollutants, which are more lipophilic and have much longer half-lives than
TCEP {i.e., 6 to 27 years vs. <24 hours) measured in mothers and infants from a Northern Quebec Inuit
population. Furthermore, it is unclear how uncertainties in model inputs like partition coefficients affect
modeled milk concentrations. The paucity of monitoring data also precludes EPA from ground truthing
modeled concentrations against measured data. As previously discussed, only one Australian study
measured TCEP concentrations by wet weight and in only three samples (He et al.. 2018a). Due to the
low number of data points, it is difficult for EPA to have confidence in the available monitoring data and
to use them to substantiate modeled concentrations. While there are uncertainties in the modeled milk
concentrations, the Verner PBPK model does reflect best available data identified by EPA, and as such,
EPA relied on it to evaluate the human milk pathway. The infant risk estimates based on the modeled
concentrations are always lower than the mothers; in fact, they are sometimes up to several magnitudes
lower. Therefore, EPA has moderate confidence that the evaluation approach is protective of infants
exposed through the human milk pathway.

5.1.4 Aggregate Exposure Scenarios	

EPA has defined aggregate exposure as "the combined exposures to an individual from a single chemical
substance across multiple routes and across multiple pathways (40 CFR 702.33)." The fenceline
methodology, (Draft Screening Level Approach for Assessing Ambient Air and Water Exposures to
Fenceline Communities Version 1.0), aggregated inhalation estimates and drinking water estimates from
co-located facilities. Due to the lack of site-specific data for TCEP, EPA was unable to employ this
approach.

Source attribution is a key challenge when attempting to characterize an aggregate exposure scenario.
When considering pathway specific estimates and aggregate exposures, there is uncertainty associated
with which pathways co-occur in each population group. Further, there is variability within a given
exposure pathway. For the same exposure scenarios, central tendency estimates are more likely to occur
than high-end estimates.

Aggregate Exposure across Routes

EPA presents total acute and chronic exposure estimates in the consumer assessment (Section 5.1.2.3
and Appendix 1.1.1). Generally, exposure estimates to consumer articles are dominated by a single route
(i.e., mouthing by infants and children). However, there are cases where aggregate exposures across
routes are important to consider when inhalation, dermal and ingestion estimates are within similar
ranges, and estimating risks from one route of exposure may underestimate the risk to a consumer COU.
The Supplemental TCEP Consumer Modeling Results includes a figure that aggregates the consumer
exposure estimates by route (inhalation, dermal, ingestion) for each COU, life stage combination:

Page 243 of 572

-------
6157

6158

6159

6160

6161

6162

6163

6164

6165

6166

6167

6168

6169

6170

6171

6172

6173

6174

6175

6176

6177

6178

6179

6180

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Aggregate Chronic Average Daily Doses (CADDs)

TCEPCOUs

bf;

£
aj"

LTl

o
Q
u
C

O

u

4

3
2

Dermal

Ingestion

Inhalation

¦¦I—-ill



£

£

£



£

m

z

£

u

u

u

O

u

u

U

0

O

0

O

0

0

O

CL

CL

CL

CL

o_

CL

CL







?

?

?

?

1/1

U1

LI

LI

l/l

l/l

LI

zr

zr

3

QJ

QJ

QJ

QJ

.

n

zr



3

3

3

Zl

0

D

CL

CL

CL

CL

i_

3"
NJ

IS 1

Qj
3

£

•<
O

cz

n

3"

3







zr

CL

3

TJ

ro

r-r

-<
O

c

ro ro o	a

n n °	°

rj rj

TJ	TJ
TJ TJ — —

qj	qj
¦<

ro ro

r-r —

n

LI LI

E E

n n

o o Ej

01	&	in

LI	LI	r-r

—	"	zr

d	E	nj

n	n

Q ~ 3* 3^ 3* ~ 3 3 ET O O O O

QIQIQIQIQIQiaiQIQIQIQlQIQJ

333333BBB3333

[j Qj L) Hi

_ _ _ c c c c

^	^	<	<	<	<	^ m m

t	I-	i.	i.	i.	i.	1/1 LH 1/1

"IT	£7	3	3	3	3	LI LI LT

„	_	Cra,	Cra,	Onq	Cra,	-< n ^

5 E,	>	^	Q	3	°

^ zr	±.

w ft	rt	J	^	w

CL O _

J= — EU

CL ~_ 3

CD CD CD I'D

ISJ

1= — QJ
— CL 3

W M

_ -< n

3" 3"

m ro
3

COU Life stage

=r I^G,	OTG,	era.	oro,

£	3=	5	=

C	C	—	Ed

^	=3.	CL	ZT

>
CL
C

= QJ

cl zr

Figure 5-15. Aggregate Chronic Average Daily Doses (CADDs) for Each Consumer COU,
Lifestage

Figure 5-15 demonstrates that for certain consumer products (outdoor play structures, wood resin and
wooden TV stand), exposure is not dominated by a single route and that it is important to consider
multiple routes of exposure. Section 5.3.4 further discusses the aggregate risk characterization of these
COUs and the relevant lifestages.

Aggregate Exposure across COUs

A worker may be involved in multiple activities that use TCEP that have varying multiple occupational
exposure scenarios. Consumers may have multiple articles at home that contain TCEP. For example, a
consumer could hypothetically have insulation with TCEP and have wooden articles containing TCEP in
the home. No evidence was found suggesting that a single consumer is exposed through multiple
consumer COUs. Due to lack of reasonably available data indicating co-exposures of multiple TCEP
containing activities or products in the occupational and indoor environment, EPA did not assess
aggregate exposure across consumer, commercial, or industrial COUs.

Aggregate Exposure across Exposure Scenarios

A child in the general population may be exposed TCEP via soil ingestion and drinking water. In the
case of the general population exposure estimates, a production volume of 2,500 lb used to estimate
releases for each individual occupational exposure scenario. EPA did not aggregate exposure estimates
to the general population because exposure estimates were based on release estimates assuming a
production volume of 2,500 lb per OES, and an aggregation would double count the production volume.

Page 244 of 572

-------
6181

6182

6183

6184

6185

6186

6187

6188

6189

6190

6191

6192

6193

6194

6195

6196

6197

6198

6199

6200

6201

6202

6203

6204

6205

6206

6207

6208

6209

6210

6211

6212

6213

6214

6215

6216

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Thus, in the example above the soil ingestion estimates were based on 2,500 lb per OES, and the
drinking water estimate was based on 2,500 lb per OES. Thus, it could be misleading to aggregate these
exposure estimates.

Furthermore, a child may be exposed to TCEP via mouthing of consumer articles as well as via drinking
water, fish ingestion, or inhalation of ambient air. The source of consumer exposure is via the consumer
purchase of finished articles containing TCEP, whereas the source of environmental exposure from soil
is due to the environmental release from a nearby hypothetical facility. EPA did not quantitively assess
aggregate exposure across exposure scenarios because no data was available indicating the co-exposure
of TCEP from multiple exposure scenarios.

5.1.5 Sentinel Exposures	

EPA defines sentinel exposure as "the exposure to a single chemical substance that represents the
plausible upper bound of exposure relative to all other exposures within a broad category of similar or
related exposures (40 CFR 702.33)." In terms of this draft risk evaluation, EPA considered sentinel
exposures by considering risks to populations who may have upper bound exposures; for example,
workers and ONUs who perform activities with higher exposure potential, or consumers who have
higher exposure potential or certain physical factors like body weight or skin surface area exposed. EPA
characterized high-end exposures in evaluating exposure using both monitoring data and modeling
approaches. Where statistical data are available, EPA typically uses the 95th percentile value of the
available dataset to characterize high-end exposure for a given condition of use. For general population
and consumer exposures, EPA occasionally characterized sentinel exposure through a "high-intensity
use" category based on elevated consumption rates, breathing rates, or user-specific factors.

EPA varied the general population exposure scenarios to help characterize the risk estimates. Risk
estimates were calculated for diluted and non-diluted drinking water conditions, soil intakes for
children's activities with soil and playing in mud scenario, and inhalation estimates at various distances
from a hypothetical facility. Furthermore, fish ingestion intakes were estimated using a high and low
BAF value for both subsistence fisherman and the general population. The sentinel exposure for these
general population exposure scenarios was fish ingestion for subsistence fisherman and fishers who are
members of tribes.

The sentinel exposure for the consumer assessments by route were inhalation from building and
construction materials (roofing insulation) for consumers, oral ingestion of TCEP from children's
mouthing of foam seating and bedding products (foam toy blocks), and children's dermal absorption of
TCEP from wood resin products (wood flooring).

Page 245 of 572

-------
6217

6218

6219

6220

6221

6222

6223

6224

6225

6226

6227

6228

6229

6230

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

TCEP - Human Health Hazards (Section 5.2):

Key Points

EPA evaluated the reasonably available information for human health hazards, including consideration
of the potential for increased susceptibility across PESS factors and acute, short-term, and chronic
exposures to TCEP (see also Section 5.3.3 and Appendix D). The key points of the human health
hazard assessment are summarized below:

•	Based on laboratory animal studies possible susceptible sex/lifestages are: (1) males for
reproductive toxicity with adolescents as potentially most susceptible, (2) females for
neurotoxicity, with potential greater sensitivity during pregnancy, and (3)
reproductive/developmental targets resulting in decreased fertility and viability of offspring

•	The acute non-cancer endpoint for TCEP was derived from tremors in pregnant female rats in a
developmental neurotoxicity study with aNOAEL of 40 mg/kg-day.

o Human equivalent dose (HED) (daily) = 9.46 mg/kg-day
o Human equivalent concentration (HEC) (continuous) = 51.5 mg/m3 (4.41 ppm),

extrapolated from oral data
o Benchmark margin of exposure (MOE) = 30, based on 10x intraspecies uncertainty factor
(UF) and 3x interspecies UFs

•	The short-term/chronic endpoint for TCEP was derived from reproductive organ effects
(decreases in seminiferous tubule numbers in adolescent male mice) in a 3 5-day oral feeding
study with a BMDL of 21 mg/kg-day.

o HED (daily) = 2.73 mg/kg-day

o HEC (continuous) = 14.9 mg/m3 (1.27 ppm), extrapolated from oral data
o Benchmark MOE = 30, based on 10* intraspecies and 3x interspecies UFs

•	The cancer endpoint for TCEP is based on the observation of kidney adenomas or carcinomas
in male rats from a 2-year oral gavage study.

o Oral/dermal cancer slope factor (CSF) (daily) = 2.45x 10~2 per mg/kg-day
o Inhalation unit risk (IUR) (continuous) = 4.51xl0~3 per mg/m3 (5.26xl0~2 per ppm),
extrapolated from oral data

5.2 Human Health Hazard

5.2.1 Approach and Methodology

EPA used the approach described in Figure 5-16 to evaluate, extract, and integrate evidence for TCEP
human health hazard and conduct dose-response modeling. This approach is based on the 2021 Draft
Systematic Review Protocol (U.S. EPA. 2021). updates to the systematic review processes presented in
the TCEP Systematic Review Protocol (U.S. EPA. 2023n). and the Framework for Raman Health Risk
Assessment to Inform Decision Making (U.S. EPA. 2014b).

Page 246 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE

December 2023

6232	Figure 5-16. EPA Approach to Hazard Identification, Data Integration, and Dose-Response Analysis

6233	for TCEP

6234

6235	For the human health hazard assessment, EPA systematically reviewed data sources identified in the

6236	literature search conducted in 2019. EPA first screened titles and abstracts and then full texts for

6237	relevancy using population, exposure, comparator, and outcome (PECO) screening criteria. Studies that

6238	met the PECO criteria were then evaluated for data quality using pre-established quality criteria and

6239	metrics. Although EPA used data quality criteria for many studies, EPA has not developed such criteria

6240	for toxicokinetics data other than dermal absorption studies. EPA also did not formally evaluate

6241	mechanistic studies for data qualtiy but did consider whether selected genotoxicity studies followed

6242	existing guidelines. Following data quality evaluation, EPA extracted the toxicological information from

6243	each evaluated study, including studies with uninformative quality determinations. The results of data

6244	quality evaluation and extraction of key study information for dermal absorption studies as well as

6245	human and animal phenotypic toxicity studies are presented in supplemental files (U.S. EPA. 2023o. q,

6246	w, x).

6247

6248	EPA considered studies that received low, medium, or high overall quality determinations for hazard

6249	identification, evidence integration, and dose-response analysis; only one part of the dermal absorption

6250	study was low quality. Information from studies of uninformative quality were only discussed on a case-

6251	by-case basis for hazard identification and evidence integration and were not considered for dose-

6252	response analysis. For example, if an uninformative study identified a significantly different outcome

6253	compared with high- or medium-quality studies and the uninformative rating was not expected to

6254	influence the specific results being discussed, EPA considered the uninformative study for the hazard

6255	outcome being considered.

6256

6257	After evaluating individual studies for data quality, EPA summarized hazard information by hazard

6258	outcome and considered the strengths and limitations of individual evidence streams (i.e., human studies

6259	of apical (phenotypic) endpoints if available, animal toxicity studies with phenotypic endpoints, and

6260	supplemental mechanistic information). The Agency integrated data from these evidence streams to

6261	arrive at an overall evidence integration conclusion for each health outcome category (e.g., reproductive

6262	toxicity). When weighing and integrating evidence to estimate the potential that TCEP may cause a

Page 247 of 572

-------
6263

6264

6265

6266

6267

6268

6269

6270

6271

6272

6273

6274

6275

6276

6277

6278

6279

6280

6281

6282

6283

6284

6285

6286

6287

6288

6289

6290

6291

6292

6293

6294

6295

6296

6297

6298

6299

6300

6301

6302

6303

6304

6305

6306

6307

6308

6309

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

given human health hazard outcome, EPA uses several factors adapted from Sir Bradford Hill (Hill.
1965). These elements include consistency, dose-response relationship, strength of the association,
temporal relationship, biological plausibility, and coherence, among other considerations. Sections 5.2.3,
5.2.4, and 5.2.5 discuss hazard identification and evidence integration conclusions for non-cancer hazard
outcomes, genotoxicity information, and cancer, respectively. Section 5.2.5 also presents an MO A
analysis for cancer.

EPA conducted dose-response analysis for the health outcome categories that received a judgment of
likely ("evidence indicates that TCEP exposure likely causes [health effect]") during evidence
integration. The Agency also conducted dose-response analysis for health outcomes that resulted in
suggestive evidence and compared the PODs {i.e., human equivalent concentrations [HECs] or human
equivalent doses [HEDs] divided by UFs for non-cancer effects; IURs or CSFs for cancer effects) for
both likely and suggestive evidence integration conclusions (	?23i). However, EPA only

considered the health outcomes and associated specific health effects from the likely evidence
integration judgments to use as toxicity values when estimating risks from exposure to TCEP.

If supported by statistically and/or biologically significant results and if the dose-response data could be
reasonably modeled, EPA conducted benchmark dose (BMD) modeling. The dose-response assessment,
including selection of studies and chosen PODs, is discussed in Section 5.2.6.

Finally, EPA assigns confidence ratings for each human health hazard outcome chosen for acute, short-
term, and chronic exposure scenarios. These ratings consider the evidence integration conclusions as
well as additional factors such as relevance of the health outcome (and associated health effect [s]) to the
exposure scenario (acute, short-term, or chronic) and PESS sensitivity. This overall weight of the
scientific evidence analysis is presented in Section 5.2.7.

Throughout each of these human health hazard analysis steps, EPA considered results of previous
analyses, including EPA's Provisional Peer-Reviewed Toxicity Values for Tris(2-chloroethyl)phosphate
(U.S. EPA. 2009) and the 2009 European Union Risk Assessment Report (E€B. 2009).

5.2.2 Toxicokinetics Summary

This section describes the absorption, distribution, metabolism, and elimination (ADME) data available
for TCEP. For full details on toxicokinetics see Appendix J.l. The PBPK model used to estimate doses
to infants ingesting human milk is described in Section 5.1.3.4.7 {Human Milk Exposure), with details
presented in Appendix H.4.

In Vivo ADME Information

EPA did not identify in vivo human studies that evaluated ADME information for TCEP by any route of
exposure. However, in vivo ADME studies in rats and mice found that radiolabeled TCEP is rapidly and
extensively absorbed following oral dosing (Burka et at.. 1991; Herretai. 1991). TCEP is primarily
eliminated in the urine, with more than 75 percent of a dose of 175 mg/kg eliminated within 24 hours for
both rats and mice (Burka et at.. 1991). TCEP distributes widely throughout the body. Herr
found radioactivity in blood, liver, and brain (including cerebellum, brainstem, caudate, hypothalamus,
cortex, hippocampus, and midbrain) in male and female rats. There was no significant difference in the
amount of TCEP present in blood and all brain regions after 24 hours of exposure (Heir et at.. 1991).

TCEP is predominantly metabolized in the liver in both rats and mice. Metabolites reported by Burka et
at. (1991) were bis(2-chloroethyl) hydrogen phosphate (BCHP, also identified as bis(2-chloroethyl)

Page 248 of 572

-------
6310

6311

6312

6313

6314

6315

6316

6317

6318

6319

6320

6321

6322

6323

6324

6325

6326

6327

6328

6329

6330

6331

6332

6333

6334

6335

6336

6337

6338

6339

6340

6341

6342

6343

6344

6345

6346

6347

6348

6349

6350

6351

6352

6353

6354

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

phosphate, or BCEP); bis(2-chloroethyl) 2-hydroxyethyl phosphate (BCGP); and bis(2-chloroethyl)
carboxymethyl phosphate (BCCP).

In Vitro Dermal Absorption

Although no dermal in vivo toxicokinetic studies are available, EPA identified Abdallah et al.
which measured dermal absorption using excised human skin in multiple in vitro experiments conducted
according to OECD TG 428, Skin Absorption: In Vitro Method. The experiments used exposures of
either 24 or 6 hours; acetone or 20 percent Tvveen 80 (polyoxyethylenesorbitan monooleate) in water as
the vehicle; 500 or 1,000 ng/cm2 application to skin; and finite (depletable) or infinite dose. EPA gave
each of the finite dose experiments overall quality determinations of medium. For the experiment that
claimed to investigate an infinite dose, EPA assigned a low overall quality determination scenario,
because conditions for infinite dosing (use of neat or large body of material) were not met and the results
did not reflect steady-state flux throughout the experiment (e.g., applied dose was depletable).

EPA used the 500 ng/cm2 24-hour finite dose application in acetone (0.005 percent solution) to estimate
absorption for workers because this was the only experiment for which the authors reported absorption
at multiple time points. Because EPA assumes workers wash their hands after an 8-hour shift, EPA used
the value of 16.5 percent, which is the amount of TCEP absorbed at 8 hours. In accordance with OECD
Guidance Document 156 (OECD. 2022). EPA also added the quantity of material remaining in the skin
(6.8 percent) at the end of the experiment as potentially absorbable.4 Therefore, EPA assumes workers
absorb 23.3 percent TCEP through skin and used this value to calculate risks for workers (see Section
5.1.1.3).

For consumer exposures and exposure to soil scenarios that assume hand washing does not occur for 24
hours, EPA used the value at 24 hours (28.3 percent) plus the amount remaining in skin (6.8 percent)
from the same experiment used for workers (500 ng/cm2 24-hour finite dose application in acetone);
total absorption was 35.1 percent absorption and was used to calculate risks (see Sections 5.1.2.2.3 and
5.1.3.3.2).

The estimates identified above apply to finite exposure scenarios for which the TCEP dose is depleted
over time. For exposure scenarios such as swimming in which a maximum absorption rate is expected to
be maintained (i.e., the dose is not depletable during the exposure duration), EPA used the dermal
permeability coefficient (Kp) of 2.2 10 2 cm/h derived by Abdallah et al. (2016) from the experiment
that used the 24-hour 1,000 ng/cm2 TCEP skin application to calculate risks (see Section 5.1.3.3.1).

023q) presents quality determinations for individual experiments conducted by Abdallah et
al. (2016). with EPA comments for each of the data quality metrics. Data extraction tables with details
on methods and results of the experiments are also presented in	)23q).

5.2.3 Non-cancer Hazard Identification and Evidence Integration

The sections below describe adverse outcome and mechanistic data available as well as evidence
integration conclusions for each human health hazard outcome (e.g., reproductive toxicity) that has been
examined and/or observed in TCEP toxicity studies. EPA identified only one epidemiological study
relevant to non-cancer endpoints. Therefore, evidence is primarily based on available laboratory animal
toxicity studies—almost exclusively via the oral route.

4 EPA used 6.8 percent (the total amount remaining in skin after washing) because the authors did not conduct tape stripping.

Page 249 of 572

-------
6355

6356

6357

6358

6359

6360

6361

6362

6363

6364

6365

6366

6367

6368

6369

6370

6371

6372

6373

6374

6375

6376

6377

6378

6379

6380

6381

6382

6383

6384

6385

6386

6387

6388

6389

6390

6391

6392

6393

6394

6395

6396

6397

6398

6399

6400

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Section 5.2.3.1 describes the critical adverse outcomes with the most robust laboratory animal findings
for TCEP that EPA considered for POD development {i.e., those with likely evidence integration
conclusions). Section 5.2.3.2 presents hazard identification and evidence integration for adverse
outcome with weaker evidence.

Appendix K provides more information on the evidence integration conclusions for the TCEP hazard
outcomes. The 2021 Draft Systematic Review Protocol (	21) describes the general process

of evidence evaluation and integration, with relevant updates to the process presented in the TCEP
Systematic Review Protocol (I v < < \ 2023n).

5.2.3.1 Critical Human Health Hazard Outcomes

The sections below focus on hazard identification and evidence integration of neurotoxicity,
reproductive toxicity, developmental toxicity, and kidney toxicity, which are the most sensitive critical
human health hazard outcomes associated with TCEP. These hazard outcome categories received likely
evidence integration conclusions, and sensitive health effects were identified for these hazard outcomes.

In the risk evaluation, neurotoxicity forms the basis of the POD used for acute exposure scenarios and
reproductive toxicity is the basis of the POD used for short-term and chronic exposure scenarios.

5.2.3.1.1 Neurotoxicity

Humans

EPA did not identify epidemiological studies that evaluated any potential neurological hazards.
Laboratory Animals

A review of high-quality acute, subchronic, and chronic studies in both rats and mice demonstrated
neurotoxic effects in both sexes following TCEP exposure.

Effects in Adults: Dosing from one to a few days in multiple studies resulted in several signs of
neurotoxicity. Female Fisher-344 rats administered 275 mg/kg of TCEP via oral gavage in a 1-day
toxicity study exhibited increased brain lesions, seizures, and behavior effects (Tilson et at.. 1990). NTP
reported that B6C3F1 mice administered the two highest doses (350 or 700 mg/kg-day) in a 16-
day study exhibited ataxia and convulsive movements during the first three days of dosing. (Moser et at..
2015) identified very slight to moderate tremors within five days of dosing at 125 mg/kg-day in 13
pregnant rats. Finally, pregnant mice administered 940 mg/kg-day TCEP via oral gavage were languid,
prostrate, and exhibited jerking movements during GDs 7 through 14 (Hazleton Laboratories. 1983).

Longer-term studies also resulted in multiple neurotoxic effects.	administered 0, 22, 44,

88, 175, or 350 mg/kg-day TCEP to rats for 16 weeks. Females exhibited greater sensitivity than males.
During week four, the highest two doses were accidentally doubled, and female rats showed ataxia,
excessive salivation, gasping, convulsions, as well as occasional hyperactivity. Rats exhibited necrosis
of hippocampal neurons with increased dose-response (8 of 10 females at 175 mg/kg-day; 10 of 10
females at 175 and 350 mg/kg-day; and 2 of 10 ales at 350 mg/kg-day); females also showed changes in
the thalamus. Mice did not exhibit neurotoxicity up to 700 mg/kg-day after 16 weeks exposure to TCEP

(I	).

Female SD rats were administered 0, 50, 100, or 250 mg/kg-day TCEP via oral gavage for 60 days
(Yang et at.. 2018a) and exhibited occasional periods of hyperactivity and periodic convulsions at the
highest dose, as well as learning impairment in the acquisition of the water maze tasks at particularly at

Page 250 of 572

-------
6401

6402

6403

6404

6405

6406

6407

6408

6409

6410

6411

6412

6413

6414

6415

6416

6417

6418

6419

6420

6421

6422

6423

6424

6425

6426

6427

6428

6429

6430

6431

6432

6433

6434

6435

6436

6437

6438

6439

6440

6441

6442

6443

6444

6445

6446

6447

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

100 and 250 mg/kg-day. Histopathological changes in the hippocampus were observed at the two
highest doses that included apoptosis and necrosis as well as invading inflammatory cells
and calcified or ossified foci in the brain cortex at the highest dose (Yame et at.. 2018a).

In a 2-year high-quality study in which rats were administered 0, 44, or 88 mg/kg-day TCEP via oral
gavage, more than 40 percent of 88 mg/kg-day females exhibited histopathological changes such as
focal gliosis, hemorrhage, mineralization, pigmentation, and hemosiderin in the brain stem and
cerebellum (	). Similar effects were not seen in male rats (only a six percent incidence of

hemorrhage in the pons vs. none in controls). Male mice exhibited some increase in mineralization of
the thalamus (56 and 52 percent at 175 and 350 mg/kg-day compared with 34 percent in controls) with
no T3nges in brain histology in F0 adult CD-I mice dosed with 700 mg/kg-day TCEP via gavage for
several weeks during a cross-over mating study.

Developmental Neurotoxicity: Moser et al. (1 assessed neurobehavioral effects and related
hormonal responses in a non-guideline study after dosing pregnant Long-Evans rats from GD 10 through
PND 22 via oral gavage of 0, 12, 40, and 90 mg/kg-day.5 The authors measured brain
acetylcholinesterase (AChE) activity, T3 and T4 levels, as well as brain and liver weights in offspring at
PND 6 and 22. Serum AChE was measured in pups at PND22 (after inhibiting butyl cholinesterase
activity). Liver weight, serum AChE, T3, and T4 of dams were measured when they were sacrificed at
PND22. No changes were observed for these measures except an increase in liver weight relative to
body weight of less than 10 percent in dams.

Multiple neurobehavioral tests were conducted. Using an elevated zero maze to measure anxiety-like
behavior, no variables attained statistical significance for offspring of exposed dams when evaluated at
PNDs 35 to 36 or PND 70 to 71. However, the data were highly variable, which could have precluded
detection of effects (Moser et aU ).

In the functional observational battery (FOB) of the offspring, hindlimb grip strength (PND 29 to 30)
and habituation (PND 29 to 30 and 78 to 79) did not differ from controls. The only significant FOB
domain in rats treated with TCEP was activity (sex by-dose-by-day) (p < 0.03), with only the vertical
activity counts in PND 29 to 30 males showing a dose effect (p < 0.01); post-hoc analysis showed no
differences (Moser et al.. 2015).

Offspring were then evaluated as adults (PND 83-101) and were tested for multiple outcomes in the
Morris water maze. In the spatial training portion, TCEP did not result in changes in learning the
platform position (latency, path length, path ratio); swim speed; or working memory (match-to-place).
However, during the memory test, TCEP showed statistically significant dose-response effects for time
in the correct quadrant and proximity score (p < 0.05), although rats in the 40 and 90 mg/kg-day groups
had a greater preference for the target compared to controls. Testing with a visual platform revealed no
differences in swim speed or latency. The authors observed a few differences in tests of spatial search
pattern, although these apparently did not influence the direct learning and memory measurements.

During the righting reflex evaluated from PND 2-4, offspring of high-dose TCEP-treated rats showed a
statistically significant sex-by-day interaction on PND 4 (p < 0.05), but there was no statistically
significant overall sex-by-day-by dose interaction. TCEP exposure was not associated with changes in
locomotion using a motor activity ontogeny (on PNDs 13, 17, and 21) or tests that included a light
transition component (PNDs 27 to 28 and 76 to 77) (Moser et al.. 2015). Overall, Moser et al. ('.
notes that the behavioral changes do not suggest biologically relevant adverse outcomes or

5 The highest dose was decreased from 125 to 90 mg/kg-day after 5 days.

Page 251 of 572

-------
6448

6449

6450

6451

6452

6453

6454

6455

6456

6457

6458

6459

6460

6461

6462

6463

6464

6465

6466

6467

6468

6469

6470

6471

6472

6473

6474

6475

6476

6477

6478

6479

6480

6481

6482

6483

6484

6485

6486

6487

6488

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

developmental toxicity.6 Other than tremors in dams early in the study, no TCEP-related adverse effects
were observed in this study.

Mechanistic Information

In a 1-day toxicity study, ICR male mice were administered via intraperitoneal injection a single dose at
concentrations of 0, 50, 100, and 200 mg/kg for 2 hours to evaluate the pharmacological effects of
TCEP. Combined administration of TCEP with psychoactive drugs; stimulants and depressants were
used to analyze the neurochemical mechanism involved in the increased activity ambulatory activity.
Data revealed that significantly high ambulatory activity was seen after the beginning of the
measurement and decrease gradually after the administration of 200 mg/kg of TCEP. The authors note
that these results suggest TCEP acts as a g-amino butyric acid (GAB A) antagonist and not as a
cholinergic agonist, and that TCEP increases ambulatory activity in ICR mice through a GABAergic
mechanism (Umezu et at.. 1998). The IJmezu et al. (1998) study was not considered for dose-response
analysis because it is not a relevant route of exposure, but it adds support to the potential neurotoxic
nature of TCEP.

(Yame et al.. 2018a) also conducted an analysis to identify possible biochemical processes and metabolic
pathways affected after chronic exposure to TCEP but found low levels of GABA in TCEP-treated
groups.

The metabolic pathway corresponding to GABA and other compounds provide a hypothesis to explore
the possible neurotoxicity mechanisms. These findings have not been further elucidated by additional
studies and thus are not conclusive regarding a mechanism for neurotoxicity.

Serum cholinesterase activity in female rats was 75 and 59 percent of controls (p < 0.01) at 175 or 350
mg/kg-day, respectively after 16-weeks repeated exposure.7 Serum cholinesterase activity was not
reduced in male rats or in either sex of mice after 16 weeks (V • l lb) \ loser et al. (Om > did not
identify changes in brain or serum AChE of offspring after developmental exposure. Although serum
cholinesterase activity may be associated with brain activity, U.S. EPA's Office of Pesticides science
policy (	lOOOd) concluded that the overall weight-of-evidence for serum cholinesterase

activity is the weakest link for brain cholinesterase.

Evidence Integration Summary

There were no human epidemiological studies available for TCEP and therefore, there is indeterminate
human evidence.

The evidence in animals is robust based on the magnitude and severity of histological changes in the
hippocampus and other regions of the brain, clinical signs of toxicity, and behavioral changes in female
rats. Results across available animal toxicological studies showed changes at the highest dose or
increases in a dose-response manner. Effects in offspring did not show greater effects than adults.

6	In a prenatal study, Kawashima et al. (1983) evaluated effects of TCEP exposure on neurodevelopment in Wistar rats.
The study is not in English, and the abstract identifies no adverse effects. EPA is translating this study and will evaluate this
for the final risk evaluation.

7	After 16 days, serum cholinesterase activities in female rats receiving 175 or 350 mg/kg-day were 79.7 and 81.8 percent of
controls, respectively; however, this study received an overall uninformative quality determination due to a viral infection.

Page 252 of 572

-------
6489

6490

6491

6492

6493

6494

6495

6496

6497

6498

6499

6500

6501

6502

6503

6504

6505

6506

6507

6508

6509

6510

6511

6512

6513

6514

6515

6516

6517

6518

6519

6520

6521

6522

6523

6524

6525

6526

6527

6528

6529

6530

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

The mechanistic data qualitatively support the evidence of hazard for TCEP however the data are
indeterminate for the specific mechanism of TCEP hazard and are not able to be used for dose response.
EPA considers the mechanistic evidence to be indeterminate.

Overall, EPA concluded that evidence indicates that TCEP likely causes neurotoxicity in humans under
relevant exposure circumstances. This conclusion is based on effects from oral studies in rats and mice
with dose levels between 22 and 700 mg/kg-day. Compared with exposure in adults, neurotoxicity is not
expected to be increased after developmental exposure based on a lack of effects in a prenatal/postnatal
study with doses up to 90 mg/kg-day (Table_Apx K-l).

5.2.3.1.2 Reproductive Toxicity

EPA guidance defines reproductive toxicity as a range of possible hazard outcomes that may occur after
treatment periods of adequate duration to detect such effects on reproductive systems (IJ..S J.

Although reproductive toxicity is often associated with developmental toxicity and cannot be easily
separated, this section describes male and female reproductive system toxicity (e.g., effects on sperm,
hormones) as well as changes in mating and fertility in a mouse continuous breeding study. Other
offspring effects from the continuous breeding study (e.g., decreases in live pups per litter) are described
in Section 5.2.3.1.3 (Developmental Toxicity).

Humans

EPA did not identify epidemiological or human dosing studies that evaluated potential reproductive
effects from TCEP exposure in the literature search conducted in 2019.

Laboratory Animals

Animal toxicity studies that evaluated reproductive effects after TCEP exposure consist of one
reproductive assessment by continuous breeding (RACB) in mice (NT	) and several repeated-

dose studies that evaluated reproductive organs and hormones in adult and adolescent mice and in adult
rats (Chen et al. 2015a: m » I lb, Matthews et al. 1990V

The high-quality RACB study (1ST	) dosed F0 male and female CD-I mice with 0, 175, 350, or

700 mg/kg-day TCEP for 1 week prior to cohabitation, 14 weeks cohabitation, and 3 weeks in a holding
period; F0 mice were allowed to produce up to 5 litters per breeding pair. After weaning of final litters,
the F0 male and female 700 mg/kg-day groups were crossbred with controls of the opposite sex to
determine influence of sex on reproductive outcomes. F1 animals in the final litters of the continuous
breeding phase received TCEP at the same doses as their parents for approximately 14 weeks (from
weaning through 74 days of age, during a one-week cohabitation phase, and during gestation and
lactation). The F1 animals were then evaluated for reproductive outcomes.8 Because F0 breeding pairs
produced no litters at 700 mg/kg-day, F1 dose groups were limited to 0, 175, and 350 mg/kg-day. F0
control and high dose (700 mg/kg-day) and F1 adult mice were examined for changes in reproductive
organs, sperm parameters, and estrous cyclicity.

Reproductive organs9 of F344 rats and B6C3Fi mice were evaluated in NTP 16-day, 16-18 week,10 and
2-year studies (I	) that received overall high-quality determinations, except the 16-day rat

8	The exposure duration was not clearly stated in NTP (1991a) for the F1 generation but Heindel et al (.1.989) states that the
continuous breeding protocol specifies that dosing of the F1 generation begins just after weaning.

9	Gross necropsy and histopathology: Males - epididymis, preputial gland, prostate, seminal vesicles, testis; Females - clitoral
gland, mammary glands, ovaries, uterus.

10	NTP (1.991b) stated that male rats were dosed for 18 weeks but Matthews et a I. (1.990) identified the studies as 16-week
studies (vs. an 18-week study for male rats), even though they are the same studies described in NTP (1991b).

Page 253 of 572

-------
6531

6532

6533

6534

6535

6536

6537

6538

6539

6540

6541

6542

6543

6544

6545

6546

6547

6548

6549

6550

6551

6552

6553

6554

6555

6556

6557

6558

6559

6560

6561

6562

6563

6564

6565

6566

6567

6568

6569

6570

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

study, which was uninformative due to a viral infection. Matthews et; 0) reported results of
additional reproductive measurements (e.g., sperm counts) from the 16 to 18 week NTP studies and
received a medium quality determination for the reported endpoints. Chen et al. ( , a high-quality
study, evaluated the male reproductive system at 0, 100, and 300 mg/kg-day TCEP for 35 days in an oral
feeding study of five-week-old adolescent male ICR mice	Oo) presents details extracted

from these studies.

Reproductive Outcomes from RACB: The F0 continuous breeding phase of	J, resulted in

decreased fertility;11 values of 72 percent fertility in the fifth litter per breeding pair at 350 mg/kg-day
and 67 to 0 percent in the second through fifth litters at 700 mg/kg-day (p < 0.05) contrasted with F0
control fertility of 97 percent. The 700 mg/kg-day dose also resulted in 25 or more cumulative days to
litter 12 vs. controls beginning in the second litter (p < 0.05).

During crossbreeding of F0 mice, the 700 mg/kg-day male x control female group resulted in lower
pregnancy13 and fertility indices (p < 0.05) but not when treated females were bred with untreated
males.1415 F1 breeding (both sexes dosed) resulted in decreased fertility at 350 mg/kg-day (highest dose;
p < 0.05).

Decreased fertility appeared earlier in the second generation (i.e., in the single litters produced according
to protocol) than in the first generation in which only in the second or subsequent litters from each of the
breeding F0 pairs were affected.

Male Reproductive Toxicity: In males, effects on reproductive organs and hormone levels were
identified but differed by study and dose. In adolescent mice, Chen et	i found 22 and 41

percent decreases in seminiferous tubule numbers at 100 and 300 mg/kg-day, respectively (p < 0.05) as
well as decreases in Leydig, Sertoli, and spermatogenic cells. The 300 mg/kg-day group also resulted in
a testis weight decrease of 13.6 percent and testicular testosterone decrease of 18 percent (p < 0.05) as
well as "absolute" disintegration of seminiferous tubules.

The RACB study (NTP. 1991a) identified a 34 percent decrease in epididymal sperm density, more than
3.4-fold increase in abnormal sperm, 45 percent fewer motile sperm, and a 30 percent decrease in testis
weight (p < 0.001) for the only tested dose (700 mg/kg-day) in the F0 adult CD-I mice. The treated F0
mice also exhibited minimal to mild testes hyperplasia (3/10 vs. 0/10 in controls). F1 male mice did not
exhibit effects on sperm or reproductive organs at either 175 or 350 mg/kg-day (NTP. 1991a).

In the 16-week repeated dose study B6C3Fi mice at 700 mg/kg-day exhibited decreases in absolute and
relative testes weights (p < 0.01) (	). Matthews et al. (1990) reported that the 700 mg/kg-day

mice in this study had slightly reduced sperm counts (p = 0.05). Neither effect was observed at 175
mg/kg-day or lower. No changes in testes weights were observed in male rats up to 175 mg/kg-day after
16 weeks (NTP. 1991b). and sperm morphology could not be conducted on the F344 rats in the 16-week

11	The percent of mated females with copulatory plugs that got pregnant.

12	This appears to be a measure of the number of days from start of cohabitation of the breeding pairs to the day when pups
were born.

13	Number of fertile pairs of the total number of cohabiting pairs.

14	The number of breeding pairs examined ranged from 18 to 20 among dose groups.

15	NTP (1991a) cited an inhalation study (Shepel'skaia andDvshginevich. .1.98.1.1 that administered TCEP at 0, 0.5, and 1.5
mg/m3 to male rats continuously for four months and then mated with unexposed females. Similar to the RACB results, dams
had significantly decreased litter si/e and also exhibited increased pre- and post-implantation loss at 1.5 mg/m3. Shepel'skaia
and Dvshginevich (.1.98.1.) appears to be an abstract in Russian; EPA could not obtain this study or evaluate its quality.

Page 254 of 572

-------
6571

6572

6573

6574

6575

6576

6577

6578

6579

6580

6581

6582

6583

6584

6585

6586

6587

6588

6589

6590

6591

6592

6593

6594

6595

6596

6597

6598

6599

6600

6601

6602

6603

6604

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

study due to technical difficulties (Matthews et a 3).1617 There were no changes in gross necropsy
or histopathology in the 16-day or 16-week NTP studies as identified in the text, or in the 2-year NTP
study as identified in incidence tables QS[	).

The crossbreeding results described earlier suggest offspring effects are greater from treated males vs.
treated females.

Female Reproductive Organ and Hormone-Related Effects: Adult F0 females administered 700 mg/kg-
day TCEP in the RACB study exhibited decreased postnatal dam weights but no changes in estrous
cyclicity. Lower doses were not examined, but the treated F1 female adults (175 or 350 mg/kg-day) also
exhibited no estrous cycle changes. Two of ten F1 females at 350 mg/kg-day had ovarian cysts, whereas
none of the ten controls exhibited cysts, although the authors did not suggest this to be a TCEP related
effect.18; lower doses were not evaluated. As noted earlier, even though the RACB identified effects
from treated female mice bred with untreated males, effects were less pronounced than those resulting
from treated males crossbred with untreated females (N	).

There were no changes in gross necropsy or histopathology in females in the 16-day or 16-week NTP
studies as noted in the text. No statistically or biologically noteworthy non-cancer effects were seen in
the 2-year study. Although adenocarcinomas occurred in three mice at 350 mg/kg-day (p < 0.05 in the
trend test), a fibroadenoma occurred in control mice; the trend for the combined tumor types was not
statistically significant, and the incidence of adenocarcinoma was within the range of historical controls

(I	).

Mechanistic and Supporting Information

In vitro studies provide some supporting mechanistic evidence of reproductive effects. Chen et al.
(2015b) identified several effects when mouse Leydig (TM3) cells were exposed to TCEP. At 100
|ig/mL TCEP, which did not result in significant cytotoxicity, effects included large decreases in one
gene associated with testosterone synthesis after all timepoints (6, 12, and 24 hours) and a second gene
at 24 hours. After stimulation of testosterone synthesis genes with human chorionic gonadotropin
(hCG), 100 |ig/mL TCEP still significantly decreased mRNA levels compared with controls or hCG.
Also at 100 |ig/mL and 24 hours exposure, testosterone secretion was decreased by about 50 percent
with TCEP alone and by about 39.9 percent (vs. hCG) after stimulation with hCG. TCEP exposure was
also associated with increased transcription of genes for antioxidant proteins.

16	NTP (1991a) provided more details of the sperm morphology and vaginal cytology examinations (SMVCE) from the 16-
week NTP study, citing an unpublished report (Gutati and Russell 1985") and partly described by (Matthews et al. 1990):
The doses evaluated for mice were 0, 44, 175, and 700 mg/kg-day. The 700 mg/kg-day B6C3Fi mice exhibited a 28 percent
decrease in epididymal sperm density; more than a doubling of abnormal sperm; a 22 percent decrease in testicular weight;
and decreased epididymis weights. Rats were evaluated at 0, 22, 88, and 175 mg/kg-day and Giilatl and Russell (.1.985) stated
that rats did not exhibit changes in epididymis and cauda epididymis weights or in percent abnormal epididymal sperm.
Sperm density was reported as being increased and motility was decreased in rats at 175 mg/kg-day even though Matthews et
al. (1.990) did not report the results due to technical difficulties. Gutati and Russell (.1.985) was not readily available and
therefore EPA did not evaluate it for data quality.

17	In (Shepel'skaia and Dyshginevich. 1981). cited by NTP (1991a). male rats exposed continuously to air
concentrations of TCEP for four months exhibited effects on meiosis, post meiotic growth, and maturity of spermatozoids
upon histopathological examination of males. Shepel'skaia and Dyshginevich (1981) appears to be an abstract in
Russian; EPA could not obtain this study or evaluate its quality.

18	In the F0 700 mg/kg-day dose group, two of 13 females also had ovarian cysts (one minimal, one mild) compared with
none among 12 controls. However, one instance of lymphoma associated with the ovary and one instance of oophoritis was
seen in the controls.

Page 255 of 572

-------
6605

6606

6607

6608

6609

6610

6611

6612

6613

6614

6615

6616

6617

6618

6619

6620

6621

6622

6623

6624

6625

6626

6627

6628

6629

6630

6631

6632

6633

6634

6635

6636

6637

6638

6639

6640

6641

6642

6643

6644

6645

6646

6647

6648

6649

6650

6651

6652

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Exposure to 300 |ag/mL TCEP (mostly after 24 hours) yielded generally greater changes in
transcriptional levels of genes associated with testosterone synthesis (mostly decreased); increased
transcription of genes encoding antioxidant proteins; increased activities of antioxidants; and decreased
secretion of testosterone. This concentration resulted in 31.4 percent lower viability of cells than
controls; thus, effects at this concentration may be at least partly secondary to cytotoxicity (Chen et al.
2015b). Overall, although some effects may have been due to general cytotoxicity, others are specific to
male reproductive toxicity (Chen et al.. 2015b).

TCEP exposure was not associated with estrogenic or anti-estrogenic effects using either a recombinant
yeast reporter gene assay or by inducing alkaline phosphatase in human endometrial cancer Ishikawa
cells (Follmaim and Wober. 2006). Reers et al. C also found no TCEP-related changes in
endogenous androgen receptor (AR) mediated gene expression in metastatic prostate cancer cells
(LNCaP) or in estrogen receptor a (ERa) and the aryl hydrocarbon receptor (AhR) target gene activation
using ECC-1 cells (endometrial carcinoma cells). Krivoshiev et al. (2016) reported that 1000 |iM TCEP
did not exhibit estrogenic activity in a cell proliferation assay using the breast adenocarcinoma cell line
(MCF-7) but did show anti-estrogenic activity when co-treated with 17P-estradiol (E2), yielding a 32
percent relative inhibitory effect. Viability of TCEP to MCF-7 cells was 93 percent of viability in
controls, and results are not expected to be overly influenced by cytotoxicity.

Evidence Integration Summary

There were no human epidemiological studies available for TCEP through the 2019 literature search, and
the human evidence is indeterminate for reproductive effects.

For the animal studies, which primarily received high or medium overall quality determinations,
biological gradients were seen for fertility index, number of litters per pair, and number of live pups per
litter, which were decreased in a dose-related manner the F0 generation CNTP. 1991a) and for testes
histopathology in mice (Chen et al.. 2015a). which exhibited increased magnitude and severity with
increasing dose.

Consistent findings included decreased numbers of live pups per litter observed at the same dose in F0
and F1 mice in the RACB, with increasing severity in the second generation (NTP. 1991a). and decreased
testes weights in mice at 300 mg/kg-day and higher (Chen et al.. 2015a; NIT l,),) Li, h). Decreases in
testosterone and related effects were observed in vivo and in vitro (Chen et al.. 2015a; Chen et al..
2015b). with related decreases in gene expression in vitro (Chen et al.. )

Within and among animal studies, coherent changes were seen between related types of effects.
Decreased testosterone in Chen et al. (2015a) and Chen et al. (2015b) support observed effects on testes
and sperm in other studies. Also, in the first generation of the RACB study (NTP. 1991a). male
reproductive effects were observed along with effects on fertility and live pups per litter.

Some effects differed among studies. Histopathological changes in the testes were also not routinely
identified. Chen et al. (2015a) observed changes in seminiferous tubules in adolescent ICR mice that
were not identified in other studies, including the F1 males in the RACB study that were dosed
beginning at weaning (NTP. 1991a). These differences lend uncertainty regarding the association of this
specific effect with TCEP exposure. However, studies differed in use of species or mouse strains and in
use of gavage vs. feeding. Chen et al. (2015a) was also conducted more than 20 years after the other
studies and differences in assessment methods could possibly explain the differences in results.

Page 256 of 572

-------
6653

6654

6655

6656

6657

6658

6659

6660

6661

6662

6663

6664

6665

6666

6667

6668

6669

6670

6671

6672

6673

6674

6675

6676

6677

6678

6679

6680

6681

6682

6683

6684

6685

6686

6687

6688

6689

6690

6691

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Effects on sperm were not identified in the F1 animals even though effects on live pups/litter and
fertility were observed in the RACB study (NTP. 1991a). However, in vitro studies suggest other
mechanisms (e.g., oxidative stress, as suggested by Chen et al. (2015b)) might be operating and could
contribute to the observed reproductive effects.

Overall, evidence in humans is indeterminate based on the lack of available studies. Evidence in animals
is moderate based on studies with decreased testes weight, sperm effects, and/or reduced fertility, and
some support from histopathological changes in testes. EPA considers the mechanistic evidence
(decreases in testosterone and genes expression but no direct estrogenic or androgenic agonism or
antagonism) to be slight. Overall, EPA concluded that evidence indicates that TCEP likely causes
reproductive toxicity in humans under relevant exposure circumstances. This conclusion is based on
effects primarily related to fertility in the RACB study and male reproductive toxicity and is based on
oral studies in rats and mice with dose levels between 22 and 700 mg/kg-day (Table_Apx K-2). EPA
guidelines for reproductive toxicity risk assessment (	96) state that findings in animals are

considered relevant to humans in the absence of evidence to the contrary.

5.2.3.1.3 Developmental Toxicity

identifies death, structural abnormalities, altered growth, and functional deficits as the
four major manifestations of developmental toxicity. This section describes relevant measurements
related to these outcomes and any identified effects (e.g., viability of offspring among fertile pairs) in
prenatal/postnatal studies in mice and rats and the continuous breeding study in mice. This section also
describes effects in animals measured during adolescence, a relevant developmental life stage (U.S.
E	). Mating and fertility outcomes resulting from the continuous breeding study are described in

Section 5.2.3.1.2 (Reproductive Toxicity).

Humans

EPA did not identify epidemiological or human dosing studies that evaluated potential developmental
effects from TCEP exposure in the literature search conducted in 2019.

Laboratory Animals

EPA identified two prenatal/postnatal animal studies, and both received high overall quality
determinations. Hazleton Laboratories (1983) administered 940 mg/kg-day TCEP via oral gavage to
female CD-I mice from GD 7 to 14. Dams exhibited clinical signs of neurotoxicity but no differences in
measures of live or dead pups per litter. In addition, there were no changes in fetal or pup weights.

Similarly, Long-Evans rat dams were dosed from GD 10 to PND 22 via oral gavage at 0, 12, 40, and 90
mg/kg-day (decreased from 125 mg/kg-day after 5 days) in the developmental neurotoxicity study
described in Section 5.2.3.1.1. There were no differences in litter size on PND 2 or changes in offspring
weight (Moser et al.. 2015).19 20 21

19	Kawashima et al. (.1.983). a foreign language study, evaluated viability of offspring; the study is being translated and EPA
will evaluate this for the final risk evaluation.

20	Limited information from the unavailable Russia inhalation study in rats (Shepet'skaia a	ihginevich. 1981")
identified decreased body weight and crown rump length in rat offspring at 0.5 mg/m3.

21	identified no effects on sex ratio in the first generation, and although significant differences in sex ratio
from controls were observed in the second generation, there is uncertainty in the change due to a discrepancy in reporting of
proportion of male offspring born alive at the highest dose (0.41 vs. 0.45).21 Two prenatal/postnatal studies did not identify
effects on sex ratio (Moser et al.. 1 ^ I ) tiazleton .Laboratories (1983). another prenatal study, did not describe
whether sex ratio was measured.

Page 257 of 572

-------
6692

6693

6694

6695

6696

6697

6698

6699

6700

6701

6702

6703

6704

6705

6706

6707

6708

6709

6710

6711

6712

6713

6714

6715

6716

6717

6718

6719

6720

6721

6722

6723

6724

6725

6726

6727

6728

6729

6730

6731

6732

6733

6734

6735

6736

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

In the RACB protocol I	, the 350 and 700 mg/kg-day mice exhibited decreases in average

number of litters per pair and live pups per litter (p < 0.001).

During crossbreeding of F0 mice, the 700 mg/kg-day male x control female group yielded decreased
live F1 pups per litter (statistical analysis not possible because only one litter was delivered). Results of
700 mg/kg-day females crossed with control males also led to decreases in live F1 pups per litter (p <
0.01 males; p < 0.05 both sexes). Outcomes from treated males x control females were more
pronounced, with production of just 1 litter with 3 live pups vs. 12 litters and 7.2 live pups per litter
from treated females x untreated males. The control x control group resulted inl2 litters and 10.3 live
pups per litter compared with either 700 mg/kg-day males or females crossbred with controls (NTP.
1991a).22 23

After F1 breeding, there were decreased numbers of live F2 pups per litter at the highest dose of 350
mg/kg-day (p < 0.05). Although live male F2 pups per litter were also reported as being significantly
decreased at 175 mg/kg-day (NTP. 1991a). EPA identified a discrepancy in NTP's Table 4-4 in the
proportion of males.

Effects were more pronounced across generations. The same dose (e.g., 350 mg/kg-day) resulted in
fewer live F2 pups per litter (7.6) than live F1 pups per litter (10.1) (NTP. 1991a).

Mechanistic and Supporting Information

Yonemoto et al. (1997) identified an IP50 (inhibitory concentration for cell proliferation) 3,600 |iM of
TCEP using rat embryo limb bud cells. The ID50 (inhibitory concentration for differentiation) was
identified as 1,570 |iM. The authors concluded that the high proliferation to differentiation ratio
suggested that TCEP should be investigated more fully for developmental toxicity.

In vivo and in vitro studies found TCEP to affect male reproductive hormones as noted in Section
5.2.3.1.2 including decreases in both testosterone secretion and decreases in a gene associated with
testosterone synthesis in mouse Leydig (TM3) cells (Chen et al.. 2015a; 2015b). These reproductive
studies may support observed developmental effects based on effects on offspring viability observed
after crossbreeding treated males with control females.

In other in vitro studies, TCEP was not associated with estrogenic or anti-estrogenic effects or changes
in AR-mediated gene expression or ER« and AhR target gene activation (Reefs et al.. 2016; Follmann
and Wober. 2006). TCEP did not exhibit estrogenic activity in in MCF-7 cells but did yield anti-
estrogenic activity when co-treated with E2 (Krivoshiev et al.. 2016).

Evidence Integration

There were no human epidemiological studies that investigated developmental outcomes from TCEP
through the 2019 literature search, and the human evidence is indeterminate for developmental effects.

Animal studies show moderate evidence for developmental effects. The prenatal and prenatal/postnatal
studies did not result in developmental outcomes. However, developmental outcomes such as decreased
live pups per litter were observed in the NTP RACB study (described in Section 5.2.3.1.2) with
increased severity in the second generation. Differences in study protocols between the RACB and

22	The number of breeding pairs examined ranged from 18 to 20 among dose groups.

23	Shepel'skaia	hginevich (.1.98.1.) cited in (NTP. 1991a") (unobtainable Russian abstract) resulted in dams with
significantly decreased litter size and increased pre- and post-implantation loss at 1.5 mg/m3.

Page 258 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

6737	prenatal studies may explain differences in outcomes. The developmental effects are supported by male

6738	reproductive toxicity from animal studies (Section 5.2.3.1.2).

6739

6740	The limited mechanistic evidence of reproductive toxicity can be relevant as considerations for

6741	developmental toxicity. EPA considers the supporting mechanistic data to be slight.

6742

6743	Overall, EPA concluded that evidence indicates that TCEP likely causes developmental toxicity in

6744	humans under relevant exposure circumstances. This conclusion is based on effects primarily related to

6745	fertility in the RACB study and is based on oral studies in mice and rats that evaluated doses of 12 to

6746	700 mg/kg-day (

Page 259 of 572

-------
6747

6748

6749

6750

6751

6752

6753

6754

6755

6756

6757

6758

6759

6760

6761

6762

6763

6764

6765

6766

6767

6768

6769

6770

6771

6772

6773

6774

6775

6776

6777

6778

6779

6780

6781

6782

6783

6784

6785

6786

6787

6788

6789

6790

6791

6792

6793

6794

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table Apx K-3). EPA guidelines for developmental toxicity risk assessment (	) state that

findings in animals are considered relevant to humans in the absence of evidence to the contrary.

5.2.3.1.4 Kidney Toxicity

Human

No human studies or other epidemiological studies for TCEP exposure were identified for potential
kidney effects.

Laboratory Animals

A review of the available animal toxicity studies for rats and mice identified the kidney as the target
organ in both sexes following TCEP exposure. In a short-term (28-day) repeated oral toxicity study,
male Fisher-344 rats were given a daily TCEP dose level of 350 mg/kg-day. Results showed signs of
scattered proximal tubular regeneration in the cortex and outer stripe of the outer medulla (Taniai et at..
2012a). Other findings after short-term exposure included increased absolute and relative kidney
weights in male rats at 175 and 350 mg/kg-day after 16-day oral repeated exposures.

Some effects were also observed after longer-term dosing. After 16 weeks of oral dosing, male rats had
increased absolute and relative kidney weights at high-dose only (350 mg/kg-day) and female rats
exhibited increased absolute and relative weights from 44 to 350 mg/kg-day (	). Both F0

males and female mice exhibited cytomegaly of renal tubule cells decreased kidney weights and after
dosing of 700 mg/kg-day TCEP for several weeks in a continuous breeding study (NTP. 1991a). In the
16-week study, male mice receiving 700 mg/kg-day had significantly reduced absolute kidney weights,
decreased by 19.4 percent compared to the controls. Relative-to-body kidney weights were decreased at
175, 350, and 700 mg/kg-day by 13.3 percent, 16.0 percent, and 14.1 percent compared to controls.
Tubule epithelial cells with enlarged nuclei (cytomegaly and karyomegaly) were observed in the kidneys
of high-dose (700 mg/kg) male and female mice. These lesions were mostly observed in the proximal
convoluted tubules of the inner cortex and outer stripe of the outer medulla.

In the 2-year bioassay, both sexes of rats and mice exhibited histopathological lesions in the kidney,
including renal tubule hyperplasia and in male and female rats and epithelial cytomegaly and
karyomegaly in both male and female mice (	).

In the 2-year study, karyomegaly was observed in 32 percent and 78 percent of male mice dosed at 175
and 350 mg/kg-day, respectively, compared to 4 percent of control animals. Karyomegaly was also
observed in 10 percent and 88 percent of female mice dosed at 175 and 350 mg/kg/day, respectively.
Hyperplasia of the renal tubule epithelium was observed in 6 percent and 4 percent of male and female
mice, respectively at 350 mg/kg-day compared to 2 percent and 0 percent of control male and female
mice (NTP. 1991b). High-dose male rats (88 mg/kg-day) exhibited 48 percent incidences of hyperplasia
of the renal tubule epithelium versus 0 percent in controls. High dose female rats also exhibited
increased incidence of focal hyperplasia of the renal tubule epithelium, by a 32 percent vs. 0 percent in
controls (	). The authors reported no changes blood urea nitrogen or creatinine in rats or

mice.

As noted in section 5.2.5.2, male rats after two years also exhibited dose-related increased incidence of
renal tubule adenomas vs. control rats (48 vs. 2 percent); one control and one high dose male developed
renal tubule carcinoma. High-dose female rats exhibited an increased incidence of renal tubule
adenomas, but to a lesser extent than male rats (10 vs. 0 in controls). Eight percent of high-dose male
mice had either renal tubule adenomas or adenocarcinomas compared with two percent in controls.

Page 260 of 572

-------
6795

6796

6797

6798

6799

6800

6801

6802

6803

6804

6805

6806

6807

6808

6809

6810

6811

6812

6813

6814

6815

6816

6817

6818

6819

6820

6821

6822

6823

6824

6825

6826

6827

6828

6829

6830

6831

6832

6833

6834

6835

6836

6837

6838

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Mechanistic Information

Mechanistic data also supported the conclusion that TCEP targets the kidney. In a 28-day gavage study,
markers for cell proliferation and apoptosis were increased in the kidneys (OSOM and cortex) of rats
(Taniai et al. 2012b). In vitro exposure of primary rabbit renal proximal tubule cells (PTCs) resulted in
reduced DNA synthesis, altered expression of cell cycle regulatory proteins, cytotoxicity, inhibition of
ion- and non-ion-transport functions, and there was increased expression of pro-apoptotic regulatory
proteins and decreased expression of proteins that inhibit apoptosis were also observed (Ren et al.. 2012;
Ren et al.. 2009. 2008).

Evidence Integration Summary

There were no human epidemiological studies available for TCEP and therefore, there is indeterminate
human evidence.

The evidence in laboratory animals is moderate based on incidences of kidney histopathology findings
that increased with dose in rats and mice of both sexes. Increased incidences of kidney histopathological
lesions were observed in rats and mice of both sexes following chronic exposures. Although less
consistent, changes in kidney weights were also observed in multiple species. EPA considers the
mechanistic evidence to be slight based on markers of cell proliferation and apoptosis in kidneys of rats
after 28-day gavage treatment and supporting in vitro evidence.

Overall, evidence indicates that TCEP exposure likely causes non-cancer kidney effects in humans
under relevant exposure circumstances based on oral studies with doses ranging from 22 to 700 mg/kg-
day in rats and mice (Table Apx K-4).

5.2.3.2 Other Human Health Hazard Outcomes

This section describes hazard identification and evidence integration for additional non-cancer health
outcome categories not considered to be critical to this risk evaluation based on the results of evidence
integration that identified evidence for these outcomes as suggestive or inadequate to assess effects.
These hazard outcomes are as follows: Skin and eye irritation, mortality, hepatic,
immune/hematological, thyroid, endocrine (other effects), lung/respiratory, and body weight.

Skin and Eye Irritation

Laboratory Animals: In a medium-quality study (Confidential. 1973). rabbits derm ally exposed to 0.5
mL (approximately 279 mg/kg24) TCEP for four hours did not show irritation through 48 hours at either
the intact or abraded skin sites. However, 0.4 mL/kg TCEP (equivalent to 556 mg/kg) was administered
to shaved dorsal skin of rabbits and repeated for four days, resulting in corrosivity and fissuring (FDRL.
1972). This study received an uninformative overall quality determination based on lack of information
on statistical analysis, and it is not clear how long TCEP was in contact with skin each day or when
corrosivity and fissuring first appeared.

TCEP was not irritating to eyes of rabbits when administered at 0.1 mL and observed for 72 hours
(Confidential. 1973) in a medium-quality study.

Evidence Integration Summary: The human evidence is indeterminate for skin and eye irritation. The
two readily available dermal irritation studies in animals showed inconsistent results and the single eye

24 According to the accompanying protocol, the dose was 0.5 mL TCEP (equivalent to 695 mg) and some sites were abraded.
Assuming 2.5 kg body weight of rabbits (2 to 3 kg was identified in the accompanying protocol), the dose was approximately
279 mg/kg-bw.

Page 261 of 572

-------
6839

6840

6841

6842

6843

6844

6845

6846

6847

6848

6849

6850

6851

6852

6853

6854

6855

6856

6857

6858

6859

6860

6861

6862

6863

6864

6865

6866

6867

6868

6869

6870

6871

6872

6873

6874

6875

6876

6877

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

irritation study of medium quality showed that TCEP is not irritating; these studies are indeterminate.
Although one study was uninformative, EPA considered that these results are not affected by the lack of
statistical analysis. Overall, the currently available evidence is inadequate to assess whether TCEP
causes irritation in humans (Appendix K.2).

Mortality

Laboratory Animals: EPA identified multiple oral studies and two dermal studies. In short-term oral
mouse studies, no female CD-I mice died at 940 mg/kg-day after dosing from GD 7 to 14 (Hazleton
Laboratories. 1983 V25 In a 16-day repeated-dose study, no mice died at doses up to 350 mg/kg-day
(I '•V * U>) 26 At higher doses, 13 to 20 percent female mice died at 1,000 mg/kg-day and all mice
died at 3,000 mg/kg-day after eight to fourteen days of exposure (NTP. 1991a; Hazleton Laboratories.
1983V

In longer-term studies, adult mortality was observed at lower doses in rats compared with mice. In 16 to
18 week subchronic studies that received medium-quality determinations for mortality, male and female
rats exhibited decreased survival as low as 175 and 350 mg/kg-day, respectively, but both groups
accidentally received double doses during week four; no mice died at doses up to 700 mg/kg-day after
16 weeks (Matthews et ai. 1990V27 No deaths occurred in rats or mice at lower doses (250 to 300
mg/kg-day) for 35 or 60 days (Yang et ai. 2018a; Chen et ai. 2015a); both studies received overall
high-quality determinations. In a high-quality 2-year study, rats exhibited decreased survival (by 27 to
29 percent) at 88 mg/kg-day, but mice did not exhibit differences in survival up to 350 mg/kg-day CNTP.
i< ).

In a medium-quality dermal irritation study, four of six rabbits died after a four-hour exposure to
approximately 279 mg/kg TCEP (Confidenti. 5).28 These rabbits exhibited narcosis and paralysis
before death. However,	did not report any deaths in rabbits dermally exposed to

approximately 556 mg/kg for 4 days. This study received an uninformative overall quality determination
based on lack of information on statistical analysis.

Decreases in numbers of live born animals after parental exposure are described in Section 5.2.3.1.2.
Evidence Integration Summary: Human evidence is indeterminate for mortality because there are no
human epidemiological studies. There is modest evidence in animal studies that shows higher mortality
in rats than mice on oral studies and uncertain potential for mortality via the dermal route given
conflicting results. Overall, evidence suggests but is not sufficient to conclude that TCEP exposure
causes mortality in humans under relevant exposure circumstances. This conclusion is based on oral
studies in rats and mice that assessed dose levels between 12 and 700 mg/kg-day and dermal studies in
rabbits at approximately 279 and 556 mg/kg-day (Appendix K.2).

Liver

25	Death occurred in pregnant female Wistar rats (Kawashima et at. 1983): this study is being translated and will be
evaluated]

26	No rats died in a short-term study at doses up to 700 mg/kg-day (NTP. 1991b") that received an uninformative overall data
quality determination due to a viral infection.

27	NTP (1991b) reported that 9 of 10 male rats survived at 175 mg/kg-day in the 16-week study compared with 4 of 10
reported by Matthews et at. (1.990). which is a report of the same study.

28	The 2009 European Union Risk Assessment Report (EC.B, 2009) reported results of an acute dermal study not readily
available to EPA in which four rabbits were each exposed dermally to 2,150 mg/kg for 24 hours, using occlusive patches. No
deaths, apparent signs of toxicity, or cholinesterase depression were observed in any of the rabbits 72 hours after treatment.

Page 262 of 572

-------
6878

6879

6880

6881

6882

6883

6884

6885

6886

6887

6888

6889

6890

6891

6892

6893

6894

6895

6896

6897

6898

6899

6900

6901

6902

6903

6904

6905

6906

6907

6908

6909

6910

6911

6912

6913

6914

6915

6916

6917

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Laboratory Animals: EPA identified multiple high-quality animal studies that reported liver weight,
histopathological changes, and one study measured enzyme changes. Liver weights were statistically
increased in multiple oral gavage rodent studies. In 16- or 18-week studies, rats and mice exhibited
absolute increases ranging from 10 to 84 percent and relative-to-body weight increases ranging from less
than 10 to 51 percent, with the largest increases in female rats at the highest dose of 350 mg/kg-day
(I	).29 At the 66-week sacrifice in the chronic bioassay, male rat absolute and relative liver

weights were increased by 20 and 19 percent, respectively at 88 mg/kg-day (the highest dose) but female
rats did not exhibit similar changes. Liver weight was not reported for mice in the chronic bioassay
(J	).30 F0 male mice (but not females) given 700 mg/kg-day TCEP for 18 weeks in a

continuous breeding study via oral gavage exhibited increases in relative and absolute liver weight of 20
and 15 percent, respectively, with no accompanying body weight changes (NTP. 1991a). No liver
weight changes were seen after 350 mg/kg-day in the F0 or F1 generation in the same study. Only the
16-day mouse study reported a decrease in (relative) liver weight in males (by 18 percent), but the
change was seen only at 44 mg/kg-day without a dose-response (NTP. 1991b).31

In the 2-year oral gavage bioassay, male mice had 6 and 16 percent incidence of eosinophilic liver foci
at 175 and 350 mg/kg-day compared with 0 incidence in controls. EPA conducted a Fischer's exact test
and identified the incidence at the highest dose to be statistically significant (p < 0.01). The foci are
believed to be precursors to hepatocellular neoplasms Q	). Because these foci were not

accompanied by increased basophilic and clear cell foci, which are considered part of the continuum
with hepatocellular adenomas, NT	states that it is uncertain whether eosinophilic foci were

associated with TCEP exposure. Adenomas and carcinomas are discussed in Section 5.2.5.2. At 700
mg/kg-day in the continuous breeding study, F0 male mice exhibited cytomegaly (10/12) and hepatitis
(4/12) vs. 0/10 per effect in controls; no other doses were evaluated in the F0 generation. F1 mice
exhibited minimal or mild changes in liver histology at 350 mg/kg-day (NTP. 1991a).

Liver enzyme activity was measured only at the 66-week sacrifice in the 2-year bioassay (NTP. 1991b).
Female rats at 88 mg/kg-day exhibited significantly decreased mean serum alkaline phosphatase (ALP)
and alanine transferase (ALT) values with no change in aspartate transaminase (AST). No information
was provided on the magnitude of change, and no differences were reported for male rats or mice of
either sex (NTP. 1991b). Although increases in liver enzyme activity are typically associated with liver
injury, decreases are harder to interpret. Decreases in serum ALT could occur after initial increases
resulting from liver injury and has been associated with decreased levels of vitamin B«, (Giannini et al.
2005). ALP is also present in bone and intestines and decreases have been associated with chronic
myelogenous leukemia, anemias, severe enteritis, and other conditions (Sharma et al l I, latroimi et
al. 2005).

Due to uncertainty and lack of information, EPA has not determined the decreased enzyme activities to
be adverse. Furthermore, except for the liver weight changes identified in the reproductive and
continuous breeding protocol in male mice at 700 mg/kg-day that were accompanied by

29	The 350 mg/kg-day female rats also had increased body weight (by 20 percent) compared with controls (NTP. 1991b").

30	In the 16-day rat study, females exhibited statistically significant increases in absolute and relative liver weights (by 17 and
14 percent, respectively) at 350 mg/kg-day but the study was uninformative due to a viral infection.

31	Chen et al. (2015a) found that male mice had decreases of 17.3 and 18.1 percent in absolute liver weight at 100 and 300
mg/kg-day, respectively after 35 days of dosing in an oral feeding study. Body weights were also decreased by 13.5 and 14.8
percent at 100 and 300 mg/kg-day respectively (estimated from graphs using Grablt!™ Copyright Datatrend Software, 1998-
2001. https://download.cnet.com/Grab~It~XP/3000~2053 4-4.1.084.litnit). EPA calculated decreased liver weights relative to
body weights for male mice of 3.5 and 3.6 percent at 100 and 300 mg/kg-day, respectively (Chen et al.. 2015a); therefore,
the changes were within 10 percent and not considered adverse.

Page 263 of 572

-------
6918

6919

6920

6921

6922

6923

6924

6925

6926

6927

6928

6929

6930

6931

6932

6933

6934

6935

6936

6937

6938

6939

6940

6941

6942

6943

6944

6945

6946

6947

6948

6949

6950

6951

6952

6953

6954

6955

6956

6957

6958

6959

6960

6961

6962

6963

6964

6965

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

histopathological changes, the increased liver weights in other studies are not clearly adverse due to the
lack of histopathological changes and lack of increased enzyme activity.

Mechanistic Information: EPA identified mechanistic studies in liver and liver cells from both in vivo
and in vitro studies. Limited mechanistic data indicate that TCEP may increase oxidative stress (based
on increased hepatic antioxidant enzyme activities and accompanying gene expression) in the livers of
male ICR mice after 35 days of dietary TCEP exposure (Chen et ai. 2015a). In vitro studies show that
TCEP induced oxidative stress, altered cellular energetics, and influenced cell signaling related to
proliferation, growth, and cell survival in the liver (Mennillo et ai. 2019; JO I h; JO I J016c; Zhang et
ai.2016b).

Evidence Integration Summary: There are no epidemiology studies that investigated liver effects, and
therefore human evidence is indeterminate.

Male mice exhibited a dose-related increase in eosinophilic foci after two years (as well as an increase in
hepatocellular adenoma) in a high-quality study (	). Increases in liver weights in male and

female rats occurred at lower doses as duration increased, and liver weights increased dose-dependently
in female rats and female mice at 16 weeks and in male rats at 66 weeks (|	). Only at a higher

dose (700 mg/kg-day) was concordance observed between increased liver weight and histopathological
changes (	).

However, "N	suggests an uncertain association between TCEP exposure and eosinophilic foci.

Also, there were no histopathology findings in rats or female mice, including no hypertrophy associated
with liver weight increases. Liver weight increases were seen in female rats after 16 days and 16 weeks,
but not 66 weeks of exposure. Increased liver weight was not seen in the 35-day study (Chen et ai.
2015a). No biologically relevant changes in serum enzymes were seen in the 2-year bioassay and were
not measured in shorter studies. Therefore, EPA determined that the animal evidence for adverse effects
on the liver based on these data are slight for the association between TCEP and adverse liver effects.

Mechanistic information shows biological gradients for the induction of hepatic oxidative stress
occurring earlier than apical endpoints. Also, across the in vitro studies, dose-related changes in
viability, oxidative stress, and impaired mitochondrial functioning were observed. Oxidative stress is a
plausible mechanism for eosinophilic foci (and tumor formation) that is relevant to humans. However,
few potential mechanisms were investigated in available studies and oxidative stress was demonstrated
in vivo at higher doses than those associated with liver lesions in the chronic study. This information
suggests mechanistic evidence for liver effects is slight.

Based on the indeterminate human evidence, slight animal evidence showing increased liver weights in
in the absence of relevant clinical chemistry findings or statistically significant histopathology changes,
EPA concluded that evidence suggests but is not sufficient to conclude that TCEP exposure causes
hepatic toxicity in humans under relevant exposure circumstances. This conclusion is based on studies
of mice and rats that assessed dose levels between 44 and 700 mg/kg-day (see Table_Apx K-5).

Immune/Hematological

Humans: Canbaz et ai (2015) did not identify an association between TCEP levels from mattress dust in
Swedish homes where 2-month-old children lived and the subsequent development of asthma when the
children reached ages 4 or 8 years in a medium-quality study.

Page 264 of 572

-------
6966

6967

6968

6969

6970

6971

6972

6973

6974

6975

6976

6911

6978

6979

6980

6981

6982

6983

6984

6985

6986

6987

6988

6989

6990

6991

6992

6993

6994

6995

6996

6997

6998

6999

7000

7001

7002

7003

7004

7005

7006

7007

7008

7009

7010

7011

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Laboratory Animals:	reported no chemical-related changes in hematological parameters in

rats or mice after 66 weeks of exposure and no histopathological changes in bone marrow, lymph nodes,
spleen, or thymus; rats did show a statistically significant increased trend in mononuclear cell leukemia
with increasing dose. No other in vivo animal toxicity studies were identified that studied specific
immune system changes.

Mechanistic: Three in vitro studies examined immune effects. Zhang et al. (2017a) found that TCEP
was associated with a decrease in the production of IL-6, an inflammatory cytokine, in the supernatant
of human hepatocytes (L02 cells). The authors stated that this result indicated that the IL-6/IL6R
pathway was not activated. Using the human hepatocellular carcinoma cell line HepG2, Krivoshiev et al.
(2018) found that TCEP altered gene expression of effector and regulatory proteins in the inflammatory
process and concluded that TCEP may influence inflammation and alter immune function. (Zhang et al..
2017b) found that liver cells co-exposed to both TCEP and benzo-a-pyrene activated pathways
associated with inflammation and increased expression of pro-inflammatory cytokines, whereas
exposure to TCEP alone did not yield similar changes.

Evidence Integration Summary: Evidence from an epidemiological study did not identify an association
between TCEP and childhood asthma and was indeterminate for immune and hematological effects; the
study evaluated only a single type of immune effect. Animal studies did not identify histopathological
changes in immune-related organs or in hematological parameters. A statistically significant increased
trend in mononuclear cell leukemia with increasing dose was seen in rats. In mechanistic studies, TCEP
was associated with decreases in an inflammatory cytokine and altered gene expression of inflammatory
proteins in two studies, but a third study identified inflammatory changes only after co-exposure with
benzo-a-pyrene.

Available evidence is indeterminate and therefore, is inadequate to assess whether TCEP may cause
immunological or hematological effects in humans under relevant exposure circumstances.

Thyroid

Humans: EPA did not identify any epidemiological studies that evaluated TCEP's association with non-
cancer effects on the thyroid. Hoffman et al. (2017). identified a statistically significant association
between TCEP exposure and thyroid cancer in a high-quality epidemiology study.

Animals: Moser et al. C found no changes in serum levels of total thyroxine (T4) and
triiodothyronine (T3) in Long-Evans dams or offspring at PNDs 6 and 22 when dosed up to 90 mg/kg-
day. J	evaluated histopathological changes in the thyroid and parathyroid in the 16-day, 16-

week, and 2-year rat and mouse studies. In the 2-year study, 12 percent of male mice (6 of 50) exhibited
follicular cell hyperplasia at 350 mg/kg-day vs. 6 percent of controls (3 of 60).	identified

increased incidences of thyroid neoplasms in rats in a 2-year cancer bioassay; the authors concluded that
there is uncertainty regarding an association with TCEP exposure.

Evidence Integration Summary: Based on these data, both human and animal evidence for non-cancer
thyroid effects is indeterminate. EPA also did not identify any mechanistic information specific to the
thyroid. Overall, the currently available evidence is inadequate to assess whether TCEP may cause non-
cancer thyroid changes in humans under relevant exposure circumstances.

Page 265 of 572

-------
7012

7013

7014

7015

7016

7017

7018

7019

7020

7021

7022

7023

7024

7025

7026

7027

7028

7029

7030

7031

7032

7033

7034

7035

7036

7037

7038

7039

7040

7041

7042

7043

7044

7045

7046

7047

7048

7049

7050

7051

7052

7053

7054

7055

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Endocrine (Other)

Animals: F0 male and female mice exhibited decreased adrenal weights after administration of 700
mg/kg-day TCEP for 18 weeks (NTP. 1991a).32 Similar effects were not observed in other studies.

Evidence Integration Summary: Based on indeterminate human and animal evidence and lack of
mechanistic support, the currently available evidence is inadequate to assess whether TCEP may cause
endocrine changes other than thyroid and reproductive hormones in humans.

Evidence related to reproductive hormones is assessed under discussed in Section 5.2.3.1.2 on
reproductive and developmental toxicity endpoints.

Lung/Respiratory

Animals: Lung weight changes were identified after 16 weeks (an increase of 17.5 percent in absolute
weight in 350 mg/kg-day female rats and decreases of 9 percent in absolute weight at 700 mg/kg-day in
female mice with relative-to-body lung weight decreases of 11.7 and 8.4 percent at 350 and 700
mg/kg/day, respectively).33 No changes were identified at the 66-week interim sacrifice in the 2-year
bioassay, and no non-cancer changes in histopathology were seen in rats or mice after two years other
than increased hemorrhage with dose in female rats presumed to be associated with cardiovascular
collapse in dying animals (	). All studies received high overall quality determinations.

Evidence Integration Summary: Based on a lack of epidemiological studies, human evidence is
indeterminate. In addition, animal data are indeterminate (no relevant histopathological effects, lung
weight changes in studies with high and uninformative overall quality determinations) based on high-
quality studies. Therefore, the currently available evidence is inadequate to assess whether TCEP may
cause lung or respiratory effects in humans under relevant exposure circumstances (Appendix K.2).

Body Weight

Animals: Changes in body weight are of concern and can suggest an underlying toxicity. For TCEP,
most studies ranging from 14 days at doses up to 1,000 mg/kg-day to two years at doses up to 88 and
350 mg/kg-day in rats and mice, respectively showed no body weight changes greater than 10 percent
(Yame et ai. 2018a; NTP. 1991a. b). Likewise, dams, fetuses, and pups exhibited no significant body
weight changes when dams were dosed up to 940 mg/kg-day during gestation or gestation and lactation
(Moser et al.. 2015; Hazleton Laboratories. 1983). Changes were also not observed in adjusted pup
weights, F0 or F1 dams at delivery, or in adult males in the continuous breeding study (NTP. 1991a).

Differences in body weights compared with controls were observed in only a few studies. Body weights
of male ICR mice decreased as much as 14.8 percent at 300 mg/kg-day TCEP after 35 days (Chen et al..
2015a). Another study identified a 20 percent increase among female rats after 16 weeks exposure to
350 mg/kg-day TCEP ( M « l lb).

In the continuous breeding study, F0 dam weights were decreased at 350 and 700 mg/kg-day from PNDs
7 through 21 (statistically significant trend, with up to 30 percent decrease for the single dam evaluated
at 700 mg/kg-day). In contrast, females in the 350 mg/kg-day group exhibited a 17 percent increase in body
weight at weaning but not during weeks 28 through 30 (N K V"'s I j). Overall, TCEP effects on body weight
were not consistent across studies and when observed, were not consistently increased, or decreased.

32	Kawashima et al. (.1.983) measured changes in pituitary weights; this study is being translated and will be evaluated for the
final risk evaluation.

33	A decrease was also seen in female rats after 16 days, but the study is uninformative due to a viral infection in the lungs
and salivary glands (NTP. 1991b').

Page 266 of 572

-------
7056

7057

7058

7059

7060

7061

7062

7063

7064

7065

7066

7067

7068

7069

7070

7071

7072

7073

7074

7075

7076

7077

7078

7079

7080

7081

7082

7083

7084

7085

7086

7087

7088

7089

7090

7091

7092

7093

7094

7095

7096

7097

7098

7099

7100

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Evidence Integration Summary: EPA identified no human studies that had information on body weight
changes and therefore, human evidence is indeterminate. In animal toxicity studies, TCEP effects on
body weight were not consistent across multiple studies. When body weight changes were observed,
they were not consistently increased or decreased. Therefore, the animal data are indeterminate. Overall,
the currently available evidence is inadequate to assess whether TCEP may cause changes in body
weight in humans under relevant exposure circumstances (Appendix K.2).

5.2.4 Genotoxicity Hazard Identification and Evidence Integration

For TCEP, several studies evaluated tests of clastogenicity (three in vivo micronucleus assays and one in
vitro chromosomal aberrations assay in mammalian cells), gene mutations (one forward mutation assay
in mammalian cells and six bacterial reverse mutation assays), and other genotoxicity and related
endpoints (two sister chromatid exchange assays, three comet assays, two cell transformation assays,
and one DNA binding assay) specific to TCEP. Although EPA did not evaluate these studies using
formal data quality criteria, selected studies were reviewed by comparing against current OECD test
guidelines and important deviations are noted below. EPA did not review the multiple studies that were
negative for gene mutations. When interpreting the results of these studies, EPA also consulted OECD
(2017).

Tests of clastogenicity and gene mutations can identify the potential for a chemical to induce permanent,
transmissible changes in the amount, chemical properties, or structure of DNA. One of three in vivo
micronucleus assays was readily available. Sala et al. (1982) administered TCEP via i.p. injection to
Chinese hamsters up to 250 mg/kg-day. Study methods deviated from OECD Test Guideline 474 (2016)
in several ways. Fewer erythrocytes (2,000 vs. 4,000) were scored than recommended, and the authors
did not verify that TCEP reached the bone marrow, although statistically significant results suggest this
was likely. Sala	2) used two hamsters per sex versus five per sex recommended by OECD TG

474 and used an exposure route that was not recommended. A firm conclusion is not possible given
several deviations from OECD TG 474. Also, the authors state that differences in the response between
sexes with variations among doses make interpretation difficult, resulting in an equivocal conclusion.
However, EPA combined results across sexes, based on a comparison of means test that indicated
similar results across sex and dose. This allowed greater statistical power (OECD. 2017). These combined
results showed statistically significant increases in micronuclei that showed a dose-response trend. No
information was provided to allow comparison with historical controls.

Two negative in vivo micronucleus studies using mice cited in the 2009 European Union Risk
Assessment Report (ECB. 2009) and a review article (Beth-Hubner. .1.999) were not available for review/4

TCEP also did not induce chromosomal aberrations in an in vitro assay using Chinese hamster ovary
cells (Galloway et a 7) that was mostly consistent with OECD Test Guideline 473 (2016a). except
that the authors scored only 100 cells per concentration compared with the recommended 300 per
concentration needed to conclude that a test is clearly negative.

A forward gene mutation assay using Chinese hamster lung fibroblasts (Sala et al.. 1982) and multiple
bacterial reverse gene mutation assays (Follmann and Wober. 2006; Haworth et al.. 198. ,

Prival et al.. 1977; Simmon et al.. 1977) were all negative for the induction of gene mutations. Most in
vitro gene mutation assays were conducted both with and without metabolic activation. In a study by
Nakamura et al. (1979). TCEP induced gene mutations in two Salmonella typhimurium strains. In strain

34 According to ECB (2009). the mouse i.p. study used doses from 175 to 700 mg/kg-day, and the oral study used a dose of
1,000 mg/kg. The original reports were not readily available for review.

Page 267 of 572

-------
7101

7102

7103

7104

7105

7106

7107

7108

7109

7110

7111

7112

7113

7114

7115

7116

7117

7118

7119

7120

7121

7122

7123

7124

7125

7126

7127

7128

7129

7130

7131

7132

7133

7134

7135

7136

7137

7138

7139

7140

7141

7142

7143

7144

7145

7146

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

TA1535, increases of four to seven times the control response were observed only with metabolic
activation and in TA100, increases were observed both with and without metabolic activation. The
reason for the inconsistency in results between Nakamura et al. (1979) and the other studies is unclear
because concentrations were comparable. One difference, however, is that Nakamura et	used a

mixture of PCBs (Kanechlor 500) for metabolic activation, whereas other studies used Aroclor 1254 or
did not appear to induce enzymes in the S9 fractions.

In addition to clastogenicity and gene mutation tests, other genotoxicity tests that measured DNA
damage or DNA binding been conducted using TCEP. Two sister chromatid exchange (SCE) assays
identified (1) equivocal results in Chinese hamster ovary cells (Galloway et al.. 1987). and (2)
statistically significant differences from controls in Chinese hamster lung fibroblasts but no clear dose
response (Sala et al.. 1982). In vitro comet assays in peripheral mononuclear blood cells (PMBCs)
identified DNA damage at the highest concentration, although it is not known whether this result was in
the presence of cytotoxicity (Bukowski et al.. 2019). Another comet assay did not identify DNA damage
in Chinese hamster fibroblasts either with or without metabolic activation (Follmann and Wober. 2006).
TCEP was also negative in a DNA binding assay (Lown et al.. 1980).

Sala et al. (1982) identified a high level of cell transformation in Syrian hamster embryo (SHE) cells but
a lower level using C3H10T1/2 cells with metabolic activation. These cell transformation results may
reflect direct or indirect genetic interactions or non-genotoxic mechanisms (OECD. 2007).

Overall, direct mutagenicity is not expected to be a predominant mode of action. Appendix L provides
additional details regarding TCEP genotoxicity studies as well as considerations regarding the quality of
the studies.

U.S. EPA's PPRTV (	309) concluded that the overall weight-of-evidence for the

mutagenicity of TCEP is negative. The PPRTV also acknowledged the weak positive result in the Ames
assay by Nakamura et a	and characterized the in vivo micronucleus assay in Chinese hamsters

(Sala et a 1) as equivocal.

5.2.5 Cancer Hazard Identification, MOA Analysis, and Evidence Integration

The sections below outline human (Section 5.2.5.1) and animal evidence (Section 5.2.5.2) for
carcinogenicity as well as and an MOA summary (Section 5.2.5.3) and a summary of evidence
integration conclusions (see Section 5.2.5.4).

5.2.5.1	Human Evidence

One high-quality case-control cancer study examined the association between TCEP/other flame-
retardant exposure and papillary thyroid cancer in adults (Hoffman et al.. 2017). TCEP concentrations in
dust were measured in 70 age- and gender-matched cases and controls in 2014 to 2016; no biological
measurements were collected for TCEP. The authors identified a median TCEP concentration of 400
ng/g in dust. Diagnosis of papillary thyroid cancer was positively associated with TCEP concentrations
above the median. The odds ratio is 2.42 (CI 1.10 to 5.33) (p < 0.05).

5.2.5.2	Animal Evidence

EPA identified one oral NTP cancer bioassay in which F344/N rats B6C3Fi mice (50 per sex per dose of
each species) were administered TCEP in corn oil via oral gavage for 5 days per week for 104 weeks.
Rats received 0, 44, or 88 mg/kg and mice received 0, 175, or 350 mg/kg (NT	). The study

received high overall quality determinations for the tumor incidence data.

Page 268 of 572

-------
7147

7148

7149

7150

7151

7152

7153

7154

7155

7156

7157

7158

7159

7160

7161

7162

7163

7164

7165

7166

7167

7168

7169

7170

7171

7172

7173

7174

7175

7176

7177

7178

7179

7180

7181

7182

7183

7184

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

N	identified multiple tumors and concluded that there is clear evidence of carcinogenic

activity of renal tubule adenomas in male and female rats. The authors also concluded that thyroid
follicular cell neoplasms and mononuclear cell leukemia in rats may have been related to TCEP
administration but acknowledge uncertainty related to this association. There was equivocal
carcinogenic evidence based on marginally increased incidence of renal tubule cell neoplasms in for
male mice and marginally increased incidence of harderian gland adenomas in female mice.35

Kidney Tumors

Rats: At the 66-week sacrifice, one high-dose male had a renal tubule adenoma. At the end of the study,
high-dose male rats exhibited increased incidences of renal tubule adenomas (48 percent) vs. control rats
(2 percent) (p < 0.001) and a dose-response trend was evident (p < 0.001). Male rats also exhibited
hyperplasia of the renal tubule epithelium, with 48 percent incidence at the high dose (vs. 0 percent in
controls). One control and one high dose male developed a renal tubule carcinoma. High-dose females
had a lower incidence of renal tubule adenomas (10 percent) but incidence was higher than controls (0
percent) (p < 0.05) with a statistically significant dose-response trend (p < 0.001). High dose females
also exhibited a 32 percent incidence of focal hyperplasia of the renal tubule epithelium vs. 0 percent in
controls.

Rats exhibited lower survival rates at 88 mg/kg-day after dosing with TCEP: 51 vs. 78 percent in
controls in males and 37 vs. 66 percent in controls for females. Female survival started to decrease at
week 70 and many rats exhibited brain lesions, whereas males' decreased survival was limited to the
final month of the study.

Mice: Mice exhibited no decreases in survival. At the end of the study, eight percent of high-dose male
mice had either renal tubule adenomas or adenocarcinomas compared with two percent in controls. Only
one low dose female exhibited a renal tubule adenoma. Six percent of mice exhibited renal tubule cell
hyperplasia. All treated mice had statistically significant increases in enlarged nuclei in renal tubule
epithelial cells (	). No kidney-related lesions were observed at the 66-week interim

sacrifice.36

Other Tumors

Hematopoietic system: Mononuclear cell leukemia (MNCL) was increased in male rats at both doses (28
and 26 percent, respectively) vs. 10 percent in controls. Because these are fatal neoplasms, life table
analyses are considered important and showed statistical significance for the low and high doses vs.
controls (p < 0.05) and for a dose-response trend (p = 0.01). Female rats exhibited a slight increase at the
high dose (40 percent) compared with controls (28 percent) and exhibited a dose-response trend (p
<0.01). Although MNCL may relate to TCEP exposure, the increase in male rats was not clearly dose-
related and was partly due to incidence that was lower than expected in the controls. In addition,

35	Takada et	19) dosed ddY mice at 0, 0.012, 0.06, 0.3, or 1.5 percent TCEP to ddY mice in the diet for 18 months
and identified increased incidence of tumors in multiple target organs; this study is not in English and was not translated or
evaluated for data quality. Takada et al. (1989) was, however, described in the 2009 PPRTV for TCEP (

2009) I v « « \ * _ 009) presented estimated doses for this study as 0, 9.3, 46.6, 232.8, and 1687.5 for males and 0, 10.7,
53.3, 266.7, and 1875 for females using measured data for body weight and food consumption from the bioassay in the
following equation: % diet x 10000 x estimated food consumption)/estimated body weight.

36	Takada et a I. (.1.989) identified an incidence of 82 percent renal cell adenomas and carcinomas in male mice at the highest
concentration vs. 4 percent in controls (p < 0.01).

Page 269 of 572

-------
7185

7186

7187

7188

7189

7190

7191

7192

7193

7194

7195

7196

7197

7198

7199

7200

7201

7202

7203

7204

7205

7206

7207

7208

7209

7210

7211

7212

7213

7214

7215

7216

7217

7218

7219

7220

7221

7222

7223

7224

7225

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

historical control values for these neoplasms are variable and all incidences in the current study were
within historical controls (J	).37

Thyroid: Other notable tumors in rats identified in the	bioassay included slightly increased

incidences of thyroid combined follicular cell adenomas and carcinomas observed in high-dose males
(10 vs. 2 percent control males) and in high-dose females (8 vs. 0 percent in controls). The incidence in
females exhibited a statistically significant dose-response trend and pairwise comparison at the highest
dose (p < 0.05). NTP concluded that these tumors may be related to TCEP exposure. However, the
increases were considered marginal. In addition, female rats did not exhibit thyroid follicular
hyperplasia, and	states that most thyroid carcinogens also cause hyperplasia.

Harderian Gland: At the 66-week sacrifice in x I V"'s I h 1, two high-dose female mice had adenomas
of the harderian gland and a third had a harderian gland carcinoma. In female mice, combined incidence
of harderian gland adenomas and carcinomas from both the 66-week and terminal sacrifices were
increased (5, 13, and 17 percent for controls, low, and high doses). Both the high-dose incidence vs.
controls and dose-response trend were statistically significant (p < 0.05).38

Liver: Male mice exhibited a significant positive trend for hepatocellular adenoma (p < 0.05) with 40,
36, and 56 percent incidence in controls, 175, and 350 mg/kg-day, respectively. However, the increase at
the high dose compared with controls was not statistically significant and there was no increase in
hepatocellular carcinomas compared with controls. Male mice also exhibited increased eosinophilic foci
(16 vs. 0 percent at the high dose compared with controls) but no increase in basophilic or clear cell foci,
which constitutes a morphological continuum with hepatocellular adenoma (I	).39

Uterine: Three female rats had uterine stromal sarcomas at the high dose but none in controls or the low-
dose group. Although the trend test was significant (p < 0.05), the incidence in the high dose group was
not significantly greater than in concurrent or historical controls and thus, NI	concluded that

the uterine tumors were not related to TCEP administration.

Mammary Gland: Three high-dose female mice had adenocarcinomas of the mammary gland with a
positive trend (p < 0.05). However, a fibroadenoma occurred in a female control; there was no
significant trend for fibroadenoma, or adenocarcinoma combined; and the incidence of adenocarcinomas
is within female historical vehicle controls. Therefore, I	concluded that the mammary gland

adenocarcinomas were not related to TCEP treatment.

5.2.5.3 MOA Summary

The U.S. EPA (2005b) Guidelines for Carcinogen Risk Assessment defines mode of action as "a
sequence of key events and processes, starting with the interaction of an agent with a cell, proceeding
through operational and anatomical changes and resulting in cancer formation." Hard (2 has
identified modes of action for renal tubule carcinogens that include direct DNA reactivity, indirect DNA
reactivity resulting from formation of free radicals, bioactivation involving glutathione conjugation,
mitotic disruption, sustained cell proliferation resulting from direct cytotoxicity, sustained cell

37	Takada et a I. (.1.989) found increased incidence of leukemia (type not specified) in female ddY mice (18 percent at -266.7
and 1,875 mg/kg-day) compared with two percent in controls (p < 0.05).

38	There were no increases in harderian gland tumors in male or female ddY mice (Takada et at. .1.989).

39	Takada et a I. (.1.989) identified increased hepatocellular adenomas or carcinomas in male ddY mice of 26 and 38 percent at
232.8 and 1688 mg/kg-day in the diet compared with 8 percent in controls (p < 0.01).

Page 270 of 572

-------
7226

7227

7228

7229

7230

7231

7232

7233

7234

7235

7236

7237

7238

7239

7240

7241

7242

7243

7244

7245

7246

7247

7248

7249

7250

7251

7252

7253

7254

7255

7256

7257

7258

7259

7260

7261

7262

7263

7264

7265

7266

7267

7268

7269

7270

7271

7272

7273

7274

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

proliferation after disruption of a physiologic process (such as alpha 2u-globulin nephropathy), chemical
exacerbation of chronic progressive nephropathy among others.

The target organ with the most robust evidence of carcinogenicity for TCEP is the kidney. In addition to
genotoxicity information on multiple cell types, EPA summarizes other biochemical and cellular effects
primarily in renal cells and kidneys. EPA did not conduct a formal analysis using concordance tables to
separately evaluate postulated MO As according to the International Programme on Chemical Safety
(IPCS) Conceptual Framework for Evaluating a Mode of Action for Chemical Carcinogenesis (Sonich-
Mullin et at.. 2001). Available data in vitro studies identified effects associated with TCEP and that
identify a variety of biochemical changes that might be relevant to induction of kidney tumors resulting
from TCEP exposure. However, only sparse in vivo evidence was available to understand the
temporality of precursor events associated with inducing kidney tumors.

Based on extensive data on tests of mutagenicity, EPA concludes that a mutagenic mode of action is not
a likely MOA for TCEP, as noted in Section 5.2.4 and Appendix L.

TCEP was associated with effects in 28-day studies in kidneys (OSOM and cortex) at 350 mg/kg-day
that included cell cycle deregulation, apoptosis, increases in regenerating tubules, and increased markers
of cell proliferation (but no accompanying proliferative lesions) (Taniai et at.. 2012b; Taniai et at..
2012a). The authors surmise that cell proliferation along with aberrant regulation of the cell cycle (e.g.,
from the G2 phase during which macromolecules are produced to prepare for cell division and through
the M phase of mitosis) may lead to chromosome instability linked to cancer. The accompanying
apoptosis may reflect aberrant cell cycle regulation (Taniai et at.. 2012b). It is also possible that DNA
damage may have been a precipitating factor in the increase of one of the markers (topoisomerase Ila)
(Taniai et at.. 2012a).

In vitro studies showed that primary rabbit renal proximal tubule cells (PTCs) exposed to TCEP
exhibited altered expression of cell cycle regulatory proteins, reduced DNA synthesis, inhibition of ion-
and non-ion-transport functions (e.g., decreased uptake of sodium, calcium, etc.), and induced
cytotoxicity. Increased expression of pro-apoptotic regulatory proteins and decreased expression of
proteins that inhibit apoptosis were also observed (Ren et at.. 2012; Ren et at.. 200". J008).

Studies of other tissues and cell types exposed to TCEP identified cell cycle changes, perturbation of
cell signaling pathways, markers of oxidative stress, impaired mitochondrial function, inhibition of
glutathione, and other effects (see Table Apx K-6).

In ^ K ! I I b >, the authors reported no hyperplasia in rats at the 66-week interim sacrifice in the
narrative (data tables not included). Although focal hyperplasia was observed and can be expected to be
a precursor to tumors, the only related finding regarding kidney tumors at the 66-week sacrifice was a
single renal tubule adenoma seen in female rats. Therefore, evidence of temporal progression from
hyperplasia to adenoma and then carcinoma is not available. At 2 years, hyperplasia was observed in
male rats, but incidence was slightly lower (0, 2, and 24) than adenomas (1, 5, and 24) compared with
hyperplasia at 0, 44, and 88 mg/kg-day. The lack of temporality and limited information on precursor
lesions and their relationship with tumors leads to uncertainty regarding dose-response progression from
hyperplasia to adenomas and carcinomas in males. Female rats did have higher rates of hyperplasia (0,
3, 16) than adenomas (0, 2, 5), at 0, 44, and 88 mg/kg-day, respectively.

Conclusion

Several studies have investigated biochemical and cellular changes in kidneys or renal cells that may be
associated with steps in an MOA for kidney cancer. EPA has not performed a formal analysis on

Page 271 of 572

-------
7275

7276

7277

7278

7279

7280

7281

7282

7283

7284

7285

7286

7287

7288

7289

7290

7291

7292

7293

7294

7295

7296

7297

7298

7299

7300

7301

7302

7303

7304

7305

7306

7307

7308

7309

7310

7311

7312

7313

7314

7315

7316

7317

7318

7319

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

postulated MO As (e.g., as in Sonich-Mullin et al. (21 ). However, available in vitro studies and a few
in vivo studies that identify multiple biochemical changes that might be relevant to induction of kidney
tumors There is sparse information on temporality and dose-response of potential pre-cursor events
within the in vivo studies and no clear NOAEL regarding tumor response to be able to model tumor
incidence with a nonlinear/threshold dose response analysis.

U.S. EPA's PPRTV (	309) concluded that the overall weight of evidence for mutagenicity is

negative and that no mechanistic data identify specific potential key events in an MOA for kidney or
other tumors induced by TCEP exposure other than a general association with known proliferative and
preneoplastic lesions.

5.2.5.4 Evidence Integration Summary

EPA concludes that TCEP is likely to be carcinogenic to humans using guidance from the Agency's
Guidelines for Carcinogen Risk Assessment (U.S. EPA. 2005b). This conclusion is based on clear
evidence of renal tubule adenomas and carcinomas in rats, equivocal evidence of kidney tumors in mice,
the rarity of the kidney tumors in rodents, and equivocal evidence of several other tumors in rats or
mice. Tumor incidence data are based an oral chronic bioassay in rats and mice that assessed dose levels
between 44 and 350 mg/kg-day. Table_Apx K-6 provides details regarding EPA's evidence integration
conclusion for cancer.40

There is indeterminate evidence in humans from a single high-quality case-control study that identified
an association between TCEP and papillary thyroid cancer (Hoffman et al.. 2017).

In laboratory animal studies, there is evidence of carcinogenicity in multiple two species and both sexes
in a single high-quality study. Evidence for kidney tumors is robust based on increased incidence of
renal tubule adenomas in male and female F344/N rats and marginal increases in these tumors in male
B6C3F1 mice CNTP. 1991b). The rarity of these tumors in F344/N rats and B6C3F1 mice strengthens the
evidence.

Lesions observed in kidneys include focal hyperplasia, renal tubular cell enlargement (karyomegaly),
and adenomas and carcinoma in rats and/or mice (N	) This continuum of has been observed

with renal tubular cell cancer in humans (Beckwith. 1999). Two-year cancer bioassay for a similar
chemical, tris (2,3-dibromopropyl) phosphate (CASRN 126-72-7), also resulted in kidney tumors in
male and female rats and male mice and karyomegaly in mice (NT	).

For MNCL, evidence is slight. NTP (1991b) observed significant pairwise increases and dose-response
trends of MNCL in male and female F344/N rats. However, MNCL is common in F344 rats, its
spontaneous incidence varies widely, and incidences in male rats exposed to TCEP were within
historical controls. Occurrence of these tumors is rare in mice and other strains of rats (Thomas et al..
2007). Further, there is uncertainty regarding similarity to tumors in humans. MNCL may be similar to
large granular lymphocytic leukemia (LGLL) in humans (Caldwell et	9; Caldwell. 1999;

Reynolds and Foon. 1984). particularly an aggressive form of CD3- LGL leukemia known as aggressive
natural killer cell leukemia (ANKCL) (Thomas et al.. 2007). However, Maronpot et al. ( ) note that
ANKCL is extremely rare with less than 98 cases reported worldwide, and the authors contend that
ANKCL has an etiology related to infection with Epstein-Barr virus, not chemical exposure.

40 Using the 2021 Draft Systematic Review Protocol (U.S. EPA. 202.1.'). the equivalent conclusion is that TCEP likely causes
cancer in humans under relevant exposure circumstances.

Page 272 of 572

-------
7320

7321

7322

7323

7324

7325

7326

7327

7328

7329

7330

7331

7332

7333

7334

7335

7336

7337

7338

7339

7340

7341

7342

7343

7344

7345

7346

7347

7348

7349

7350

7351

7352

7353

7354

7355

7356

7357

7358

7359

7360

7361

7362

7363

7364

7365

7366

7367

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Animal evidence for thyroid follicular cell tumors was slight based on increases seen in significant
pairwise increases of adenomas or carcinomas in female F344/N rats with a significant dose-response
trend but only marginal increases in male rats and no increase in B6C3F1 mice (1	). Although

a) notes that thyroid tumors in animal studies cannot be completely dismissed as a
hazard for humans, it appears that that rodents are more sensitive than humans to thyroid follicular cell
tumors induced by thyroid-pituitary disruption and thyroid stimulating hormone hyperstimulation
(Dvbing and Sannet l , I v < < \ l 98a). There is also slight evidence in animals for harderian
gland adenoma or carcinoma based on increased incidence in female B6C3F1 mice at the highest dose
only, but no increased incidence in rats or male B6C3F1 mice (	). Finally, slight evidence in

animals exists for hepatocellular tumors based on a dose-related trend in tumor incidence in only in one
sex of one species (male B6C3F1. mice) CNTP. 1 ).

The mechanistic evidence for carcinogenesis is slight. Available data indicates that TCEP has little if
any genotoxic potential. Limited additional data indicate that TCEP may influence cell signaling related
to proliferation, apoptosis, and ion transport, induce oxidative stress, alter cellular energetics in kidney
tissues and cells and in other cell types.

U.S. EPA's PPRTV (	309) also concluded that TCEP is likely to be carcinogenic to humans

based on information from oral animal bioassays that included clear evidence of renal tubule cell
adenomas in F344/N rats in K) ^! — lb), renal tubule adenomas and carcinomas in ddY mice in
Takada et al. (1989) as well as the rarity of these tumors. The PPRTV also describes evidence for other
tumors identified in these two bioassays as suggestive or equivocal.

The 2009 European Union Risk Assessment Report (ECB. 2009) concluded that TCEP has
carcinogenicity potential and cites the EU classification category 3 and R40—limited evidence of
carcinogenic effect. In contrast, the International Agency for Research on Cancer (IARC) designated
TCEP as not classifiable as to its carcinogenicity to humans in 1990 and again in 1999 (IARC.: ).

5.2.6 Dose-Response Assessment

According to U.S. EPA's 2021 Draft Systematic Review Protocol (U.S. EPA. 2021). hazard endpoints
that receive evidence integration judgments of demonstrates and likely would generally be considered
for dose-response analysis. Endpoints with suggestive evidence can be considered on a case-by-case
basis. Studies that received high or medium overall quality determinations (or low-quality studies if no
other data are available) with adequate quantitative information and sufficient sensitivity can be
compared.

There were no hazard outcome categories for which evidence demonstrates that TCEP causes the effect
in humans. Therefore, hazard outcomes that received likely judgements are the most robust evidence
integration decisions. The health effect with the most robust and sensitive POD among these likely
outcomes was used for risk characterization for each exposure scenario to be protective of other adverse
effects as described in the sections below.

Data for the dose-response assessment were selected from oral toxicity studies in animals. No acceptable
toxicological data were available by the inhalation route, and no PBPK models are available to
extrapolate between animal and human doses or between routes of exposure using TCEP-specific
information.

The PODs estimated based on effects in animals were converted to HEDs or CSFs for the oral and
dermal routes and HECs or IURs for the inhalation route. For this conversion, EPA used guidance from
to allometrically scale oral data between animals and humans. Although the guidance

Page 273 of 572

-------
7368

7369

7370

7371

7372

7373

7374

7375

7376

7377

7378

7379

7380

7381

7382

7383

7384

7385

7386

7387

7388

7389

7390

7391

7392

7393

7394

7395

7396

7397

7398

7399

7400

7401

7402

7403

7404

7405

7406

7407

7408

7409

7410

7411

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

is specific for the oral route, EPA used the same HEDs and CSFs for the dermal route of exposure as the
oral route because the extrapolation from oral to dermal routes is done using the human oral doses,
which do not need to be scaled across species. EPA accounts for dermal absorption in the dermal
exposure estimates, which can then be directly compared to the dermal HEDs.

For the inhalation route, EPA extrapolated the daily oral HEDs and CSFs to HECs and IURs using
human body weight and breathing rate relevant to a continuous exposure of an individual at rest. Based
on existing data (Heir et ai. 1991). absorption via the oral route may be greater than 95 percent.
Therefore, EPA assumed that absorption for the oral routes is 100 percent; there is no information
regarding absorption via the inhalation route, and therefore, EPA assumed 100 percent absorption via
this route. Therefore, no adjustment specific to absorption is needed for the oral and inhalation routes.
For consistency, all HEDs and the CSF are expressed as daily doses and all HECs are based on daily,
continuous concentrations (24 hours per day) using a breathing rate for individuals at rest. Adjustments
to exposure durations, exposure frequencies, and breathing rates are made in the exposure estimates used
to calculate risks for individual exposure scenarios.

Appendix J. 3 presents information on dose derivation, calculations for each of the PODs, and route-to-
route extrapolations. Considerations regarding the BMD modeling process as well as modeling results
for likely as well as suggestive TCEP outcomes are presented in the supplemental file Benchmark Dose
Modeling Results for TCEP (	:023b). A comparison of the PODs for likely and suggestive

health outcomes is presented visually in exposure response arrays within Appendix M, with calculations
for these PODs in an Excel spreadsheet in the supplemental file Human Health Hazard Points of
Departure Comparison Tables (U.S. EPA. 2023i).

5.2.6.1 Selection of Studies and Endpoints for Non-cancer Toxicity

EPA considered the suite of oral animal toxicity studies and likely individual adverse health effects
outcomes when considering non-cancer PODs for estimating risks for acute and short-term/chronic
exposure scenarios, as described in Section 5.2.6.1.1 and 5.2.6.1.2, respectively. EPA selected studies
and relevant health effects based on the following considerations:

•	Overall quality determinations;

•	Exposure duration;

•	Dose range;

•	Relevance (e.g., what species was the effect in, was the study directly assessing the effect, is the
endpoint the best marker for the tox outcome?);

•	Uncertainties not captured by the overall quality determination;

•	Endpoint/POD sensitivity;

•	Total UF; and

•	Uncertainty and sensitivity of BMR selection from BMD modeling.

The following sections provide comparisons of the above attributes for studies and hazard outcomes for
each of these exposure durations and details related to the studies considered for each exposure duration
scenario.

5.2.6.1.1 Non-cancer Points of Departure for Acute Exposure

To calculate risks for the acute exposure duration in the risk evaluation, EPA used a daily HED of 9.46
mg/kg (NOAEL of 40 mg/kg) from a prenatal/postnatal neurodevelopmental toxicity study (Moser et al.
2015) based on very slight to moderate tremors within five days of dosing at 125 mg/kg-day in 13 dams.

Page 274 of 572

-------
7412

7413

7414

7415

7416

7417

7418

7419

7420

7421

7422

7423

7424

7425

7426

7427

7428

7429

7430

7431

7432

7433

7434

7435

7436

7437

7438

7439

7440

7441

7442

7443

7444

7445

7446

7447

7448

7449

7450

7451

7452

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

EPA gave this study a high overall quality determination, and a UF of 30 was used for the benchmark
MOE during risk characterization.

Mice exhibited signs of neurotoxicity in other acute or short-term high-quality studies. In the NTP
(1 16-day study, mice exhibited ataxia and convulsive movements within three days at the two
highest doses with a daily HED of 16.6 mg/kg; data were only qualitatively described. Pregnant mice
administered 940 mg/kg-day TCEP via oral gavage were languid, prostrate, and exhibited jerking
movements during GDs 7 through 14 with an HED of 125 mg/kg-day (Hazleton Laboratory 5).
The HED from Moser et al. (2015) is more sensitive.

Tilson et 90) found that in addition to convulsions, female Fischer 344 rats exhibited
histopathological changes in the hippocampus and memory impairment in the Morris water maze after a
single oral gavage administration of 275 mg/kg and an HED of 65.0 mg/kg. Although EPA gave Tilson
et al. (1990) a high overall quality determination, the authors tested only a single dose level, which did
not allow a full understanding of the dose-response for TCEP. The POD is associated with greater
uncertainty because only a LOAEL was identified and a UF of 300 would be required for a benchmark
MOE analysis.

The high-quality intraperitoneal injection study by Umezu et al. (1998) provides qualitative support for
neurotoxicity; mice exhibited increased ambulatory activity at 100 and 200 mg/kg and 'light'
convulsions at 200 mg/kg after single administration of these doses. EPA did not consider this study to
be a candidate for the POD based on the exposure route.

EPA did not identify other studies of health outcomes with likely evidence integration judgments that
could be used for the acute exposure scenario.41 42 The continuous breeding protocol study (NTP. 1991a)
was not considered for acute exposure. The effects are more difficult to characterize as having occurred
following acute exposure or during a critical window in development than effects observed in prenatal
studies because the exposure paradigm includes exposure in male and female adults before and during
mating and in dams during gestation and lactation. Thus, offspring effects may be due to toxicity to
gametes prior to and during mating. Also,	identified reproductive and developmental

outcomes in litter two and subsequent litters, not the first litter from each dam. Finally, even though
some offspring toxicity may be mediated by the dam (as observed in the crossbreeding portion of NTP
(1991a)) prenatal studies (Moser et al < s , U izleton Laboratories. 1983) did not identify decreased
viability or other effects in offspring. Therefore, EPA considered decreased fertility and live pups as
most likely to occur after repeated exposure.

Table 5-46 presents a comparison of the attributes of studies and hazard endpoints considered for the
short-term exposure scenario and Table 5-47 summarizes the study PODs and pertinent information,
including HEDs and HECs. The bolded row represents the study and POD values used to calculate risks
for acute scenarios in the risk evaluation.

41	(Kawashima et al. 1983") is in a foreign language; EPA is translating the study and will evaluate it for the final risk
evaluation.

42	The 2009 European Union Risk Assessment Report (ECB, 2009) and other assessments identified acute lethality studies via
the oral, inhalation, and dermal routes that are not readily available to EPA. had extremely limited details (Smyth et at.

.1.95.1.1. or was a secondary source (Ulsanieret al. 1980"). Reported effects were LD5oS or LC50S that occurred at higher doses
or exposures, respectively; some studies reported results for a TCEP product (Fyrol CEF) of unknown purity.

Page 275 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

7453	Overall, the tremors observed in Moser	represent a sensitive endpoint that could occur in

7454	humans. The clinical signs of neurotoxicity (e.g., convulsions) were consistently observed across

7455	acute/short-term studies.

7456

7457	Table 5-46. Comparison among Studies with Sensitive Endpoints Considered for Acute Exposure

7458	Scenarios



Neurotoxicity

(Moser et ai, 2015)

Neurotoxicity
(NTP, 1991b)

Neurotoxicity
(Tilson et ai,

1990)

Neurotoxicity
(Hazleton

1983)

Overall Data

Quality

Determination

High

High

High

High

Exposure
Duration

Within 5 days

Within 3 days

1 day

8 days

Dose Range

12, 40, 125 mg/kg-
day (high dose
changed to 90 mg/kg-
day at 5 days)

0, 44, 88, 175, 350,
700 mg/kg-day

275 mg/kg

940 mg/kg-day

Relevance

Assumed to be
relevant to humans;
clearly adverse

Assumed to be
relevant to humans
(similar effect as
chosen POD);
clearly adverse

Assumed to be
relevant to humans
(similar effect as
chosen POD);
clearly adverse

Assumed to be
relevant to humans
(similar effect as
chosen POD);
clearly adverse

Uncertainties Not

Captured

Elsewhere

Effects observed only
at the highest dose

BMD modeling not
possible; only
qualitative outcome
information
available

Precision of POD is
limited because no
NOAEL was
identified

Precision of POD is
limited because no
NOAEL was
identified

Sensitivity of
POD for
exposure
scenario

Sensitive endpoint
with an identified
NOAEL

Less sensitive

Most sensitive when
considering
comparison with 300
benchmark MOE

Least sensitive

Total UF

30

30

300

300

7459

Page 276 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table 5-47. E

»ose-Response Analysis of Selected Studies Considered for Acul

te Exposure Scenarios

Target
Organ/
System

Species

Duration

Study POD/
Type (mg/kg)"

Effect

HEC
(mg/m3)
|ppm|

HED

(mg/kg)

UFs

Reference

Overall
Quality
Determination

Neurotoxicity

Long

Evans rats
(dams)

5 days

NOAEL = 40

Tremors

51.5
[4.41]

9.46

UFA=3
UFH=10
Total UF=30

Moser et al.

(2015)

High

Neurotoxicity

B6C3Fi
mice

16 days

NOAEL = 125

Convulsions,
ataxia within 3
days

90.4
[7.75]

16.6

UFA=3
UFH=10

Total UF=30

NIP (1991b)

High

Neurotoxicity

Fischer 344
rats

(females)

1 day

LOAEL = 275

Convulsions
brain lesions,
behavior
changes

354
[30.3]

65.0

UFA=3
UFH=10
UFl = 10
Total UF=300

Tilson et al.

a

High

Neurotoxicity

CD-I mice
(dams)

GD 7-14

LOAEL = 940

Jerking
movements,
languidity,
prostration

680
[58.3]

125

UFA=3
UFH=10
UFl =10

Total UF=300

Hazleton

Laboratories
I

High

" The PODs are duration adjusted to 7 days per week; therefore, any PODs from studies that dosed for 5 days per week were multiplied by 5/7.

7461

Page 277 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE

December 2023

7462	5.2.6.1.2 Non-cancer Points of Departure for Short-Term and Chronic Exposures

7463	Figure 5-17 presents exposure response arrays of the FIEDs for the likely hazard outcomes from the

7464	studies considered for the short-term and chronic HEDs. The FIEDs are presented within the hazard

7465	outcomes of reproductive, developmental, kidney toxicity, and neurotoxicity and ordered from lowest to

7466	highest to view relative sensitivities more easily.



















1 NOAELs and BMDLs: UF = 30; LOAELs: UF = 300 | | A NOAEL HED BBIVIDL

HED



~ LOAEL HED



























Relative kidney wt; 16 wk; rat (F); NTP 1991b

¦







Absolute kidney wt; 16 wk; rat(F); NTP 1991b

a""







Hyperplasia; 2 yr; rat (F); NTP 1991b









Karyomegaly; 2 yr; mouse (M); NTP 1991b

ijj 	 	







Hyperplasia; 2 yr; rat (M); NTP 1991b







c=

Absolute and relative kidney wt; 66 wk; rat (M); NTP 1991b

~~"a





"O

Relative kidney wt; 16 wk; mouse (M); NTP 1991b



A







Karyomegaly; 2 yr; mouse (F); NTP 1991b











Absolute and relative kidney wt; 16 wk; rat (M); NTP 1991b



~	







Absolute kidney wt; 16 wk; mouse (M); NTP 1991b



~	







Relative kidney wt; 16 d; mouse (F); NTP 1991b



~







Regenerating tubules, other histopathological changes; 28 d; rat (M); Taniai et al. 2012













No. of seminiferous tubules; 35 d; mouse (M); Chen et al. 2015

¦ '











Testiscular testosterone; 35 d; mouse (M); Chen et al. 2015



"a









Absolute and relative testes wt; 16 wk; mouse (M); NTP 1991b



... ^







>
"tj

Sperm count; 16 wk; mouse (M); Matthews et al. 1990



.... ^ _







-a
o

Task 4: Fertility and pregnancy index in Fl; 14 wk; mouse (M,F); NTP 1991a



H







Q_

a
cm

Testes wt; 35 d; mouse (M); Chen et al. 2015



¦









Task 2: Fertility, litter 5 in FO; Up to 18 wk; mouse (M,F); NTP 1991a



¦	









Task 2: Days to litter 2 and days to litter 3 in FO; Up to 18 wk; mouse (M,F); NTP 1991a



"a"









Task 3: Organ wt changes & histopathology; Sperm parameters; Pregnancy & fertility indices;

18 wk; mouse (M,F); NTP 1991a [1]



































Brain lesions; 2 yr; rat (F); NTP 1991b













Hippocampal lesions; 60 d; rat (F); Yang et al. 2018

~









Brain (hippocampal) necrosis; 16 wk; rat (F); NTP 1991b; Matthews et al. 1990

	







.t;
u
X

Changes in path length, Morris water maze; 60 d; rat (F); Yang et al. 2018









o
2

Ataxia, convulsions; 16 d; mouse (NS); NTP 1991b

	 A """"







=J
CD

Brain lesions; 2 yr; mouse (M); NTP 1991b

	~ "









Serum cholinesterase activity; 16 wk; rat (F); NTP 1991b; Matthews et al. 1990











Clinical observations; 60 d; rat (F); Yang et al. 2018

A~"









Prostration, jerking movements, languidity; 8 d; mouse (F); Hazleton Labs 1983









15

Task 2: Live male Fl pups/litter; Up to 18 wk; mouse (M); NTP 1991a

j£ 	







c
o
£

Task 4: Live F2 pups/litter; 14 wk; mouse (M,F); NTP 1991a









o_
_o

Task 2: FO mean litters/pair; Live total Fl pups/litter; Live female Fl pups/litter;

Up to 18 wk; mouse (M,F); NTP 1991a

A~	







o
>
CD

o

Task 3: Uve female Fl pups/litter (treated FO males); Uve male & total Fl pups/litter
(treated FO males or females); 18 wk; mouse (M,F); NTP 1991a

~











io

lOO

IO

oo











Dose (mg/kg-day)



[1] Task 3: Abs. epididymis wt in FO; Fluid & degenerated cells in epididymis in FO; Abs. & rel. testes wt

in FO; Interstitial

cell

hyperplasia in testes in FO; Ovarian cysts in FO; Sperm concentration, % motile, & 9o abnormal sperm

n FO; Pregnancy index

& fertility index (treated FO males)













7468	Figure 5-17. Exposure Response Array for Short-Term and Chronic Exposure Durations by Likely

7469	Hazard Outcomes

7470

Page 278 of 572

-------
7471

7472

7473

7474

7475

7476

7477

7478

7479

7480

7481

7482

7483

7484

7485

7486

7487

7488

7489

7490

7491

7492

7493

7494

7495

7496

7497

7498

7499

7500

7501

7502

7503

7504

7505

7506

7507

7508

7509

7510

7511

7512

7513

7514

7515

7516

7517

7518

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

EPA is using Chen et al. (2015a). the 35-day study in adolescent mice, to estimate non-cancer risks for
both the short-term and chronic exposure scenarios. The study received a high overall quality
determination, and the sensitive effect is a decrease in the numbers of seminiferous tubules (by 22 and
41 percent at 100 and 300 mg/kg-day, respectively) that is accompanied by absolute disintegration of
tubules and decreased testosterone levels and testes weights at 300 mg/kg-day.

EPA conducted BMD modeling, and several continuous BMD models adequately fit the seminiferous
tubule numbers, resulting in similar BMDL5s. The exponential 2 model fit resulted in the lowest Akaike
information criterion (AIC) and a good fit upon visual inspection. (	23b) presents additional

details, including the fits for all seven continuous models that were run and BMDL values for BMRs of
five percent RD and one SD.

For continuous data, EPA's BMD Technical Guidance recommends modeling the data using a BMR of
one standard deviation (SD) (	Eb) but lower response rates should be used when effects

are severe (e.g., frank). Thus, EPA used a BMR of 5 percent based on biological severity and identified
a BMDL5 of 21 mg/kg-day. The BMDLs for 1 SD and 10 percent were 61 and 43 mg/kg-day,
respectively. BMRs of 5 percent were also used for other severe or frank effects in the TCEP risk
evaluation, including decreased live pups per litter and brain necrosis. When evaluating male phthalate
syndrome, Blessinger et al. (2020) similarly used a BMR of 5 percent for all endpoints associated with
zero to moderate impacts on fertility. These endpoints included germ cell degeneration or depletion in
seminiferous tubules ranging from 5 to 75 percent (Blessinger et al.. 2020; Lanning et al.. 2002).

EPA calculated a daily HED of 2.79 mg/kg-day for Chen et al. (2015a) that accounts for allometric
scaling between mice and humans and is compared with a benchmark MOE of 30. HEDs for other
reproductive effects ranged from 9.51 to 93.1 mg/kg-day. Many are within an order of magnitude of
Chen et al. (2015a). The HEDs of 93.1 mg/kg-day are based on LOAELs that are 33 times greater (NTP.
1991a) and are used with a benchmark MOE of 300 instead of 30.

As noted in Section 5.2.3.1.2, hazard outcomes identified by Chen et al. (2015a) are supported by effects
on sperm, reproductive organ weight changes, and testes hyperplasia (NTP. 1991a. b; Matthews et al..
1990). Other reproductive and developmental outcomes were observed, including decreases in fertility
and live pups per litter in the continuous breeding toxicity study (NTP. 1991a).

There are uncertainties associated with using Chen et al. (2015a) for the POD. Other than minimal to
mild hyperplasia, histopathological changes in the testes were not routinely identified in other studies
(NTP. 1991a. b). However, Chen et al. (2015a) was conducted more than 20 years after the NTP studies
and some methods differed from older studies (e.g., preparation of tissues). Also, differences may reflect
use of different species or mouse strains, and in such cases,	96) recommends using the

most sensitive species in the absence of information to suggest otherwise.

There are limitations of (Chen et al.. 2015a. pp. author-year)' s study design and the BMD modeling
analysis. Doses for this feeding study may be imprecise because information on body weight and food
consumption were not reported. In addition, the sample size is small and as sample size decreases,
uncertainty in the true response rate increases. Finally, although EPA considered BMD modeling as
appropriate for this data set, in part because the lowest dose tested was a LOAEL, the BMR of 5 percent
is lower than the biologically and statistically adverse responses observed in the study (22.2 and 40.7
percent).

Page 279 of 572

-------
7519

7520

7521

7522

7523

7524

7525

7526

7527

7528

7529

7530

7531

7532

7533

7534

7535

7536

7537

7538

7539

7540

7541

7542

7543

7544

7545

7546

7547

7548

7549

7550

7551

7552

7553

7554

7555

7556

7557

7558

7559

7560

7561

7562

7563

7564

7565

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

As stated in EPA's Guidelines for Reproductive Toxicity Risk Assessment (	), human

males are particularly susceptible to chemicals that reduce numbers or quality of sperm. Chen et al.
(2015a) did not directly evaluate sperm numbers or quality but due to potential for the endpoint to affect
fertility, the magnitude of effects, and the potential for human males to be more susceptible than rodents,
EPA considers the significant effect on seminiferous tubules (which help produce, maintain, and store
sperm) to be of concern for human male reproduction and represents a relevant endpoint for the risk
evaluation.

Comparison of Studies Usedfor the Short-Term Exposure Scenario. In addition to Chen et al. (2015 a).
EPA considered sensitive effects from other studies ranging from a few days to 60 days for the short-
term POD that would be associated with a 30-day exposure scenario. Table 5-48 presents a comparison
of the attributes of multiple studies and hazard endpoints considered for the short-term exposure
scenario. Table 5-49 provides details of the studies, including PODs from the study or from dose-
response modeling, HECs, and HEDs. The bolded row represents the study and POD values used to
calculate risks for short-term and chronic scenarios in the risk evaluation.

HEDs for both Moser et al. (2015) and Yame et al. (2018a) are based on neurotoxicity, which are
relevant hazard outcomes observed across multiple studies and are within an order of magnitude of the
sensitive HED (2.79 mg/kg-day) from Chen et al. (2015a). In addition, they are oral gavage studies and
thus, dose levels are expected to be more precise compared with Chen et al. (2015a). a dietary study.
However, exposure durations (5 and 60 days) for these studies introduce some uncertainty regarding
applicability to the target 30-day exposure scenario compared with Chen et al. (2015a). a 35-day study.

Even though the HED from Chen et al. (2015a) is based on using a BMR below the observed data, other
short-term study and endpoint candidates also have limitations related to dose-response relationships.
Moser et al. (2015) observed effects only at the highest dose, and therefore, the HED is based on a
NOAEL, not a BMDL that considers the full dose-response curve. Similarly, the lowest HED (11.8
mg/kg-day) from Yang et al. (2018a) is based on a NOAEL; a similar HED from Yang et al. (2018a) (13
mg/kg-day, based on a BMDL20 of 55.0 mg/kg-day) also results in some uncertainty given typical
variability in the modeled neurobehavioral endpoint.

Taniai et al. (2012a). a 28-day study resulting in kidney proximal tubule regeneration, has a relevant
hazard outcome and an exposure duration closer to the short-term scenario. However, even less is
known about the dose-response relationship because the study used only a single dose level resulting in
a LOAEL and a benchmark MOE of 300 rather than 30 used with Chen et al. (2015a).

EPA considered developmental effects (decreased live pups per litter) and other outcomes from NTP
(1991a) to be relevant to humans and considered that these could occur following short-term exposures.
However, the POD for possible related reproductive effects observed by Chen et	1 is more

sensitive.

Overall, using Chen et al. (2015a) for the short-term exposure scenario in which adolescent male rats
were evaluated during a potentially sensitive life stage results in a sensitive POD for a relevant endpoint
for the risk evaluation. EPA considers this POD to be protective of other adverse effects identified in
TCEP toxicity studies, including developmental effects that may results from effects on male
reproductive organs.

Page 280 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

7566	Table 5-48. Comparison among Studies with Sensitive Endpoints Considered for Short-Term

7567	Exposure Scenarios					



Neurotoxicity

Neurotoxicity

Reproductive

Toxicity
(Chen et al,

2015a)

Developmental

Kidney Toxicity



(Moser et al.,

2015)

(Yang et al.,

2018a)

Toxicity
(NTF, 1991a)

(Taniai et at,
2012a)

Overall Data

Quality

Determination

High

High

High

High

Medium

Exposure
Duration

Within 5 days; less
applicable to short-
term exposure

60 days; less
applicable to short-
term exposure

35 days

Up to 18 weeks;
short-

term/chronic

28 days

Dose Range

12, 40, 125 mg/kg-
day (high dose
changed to 90
mg/kg-day at 5
days)

50, 100, 250 mg/kg-
day

100,300
mg/kg-day

F0: 175, 350,
700 mg/kg-day

350 mg/kg-day

Relevance

Endpoint assumed
to be relevant to

Endpoint assumed to
be relevant to

Endpoint
asssumed to be

Endpoint
assumed to be

Endpoint assumed
to be relevant to



humans

humans

relevant to
human male
reproduction
(

1996)

relevant to
humans

humans

Uncertainties Not

Dose-response less

Dose-response less

Dose precision

Some of the

Lack of

Captured
Elsewhere

precise: Use of
NOAEL

precise: Use of
NOAEL);
Neurobehavioral
outcomes (BMR of

unclear: dietary
study and no
information on
food

outcomes
uncertain (e.g.,
sensitivity of
decreased F2

understanding of
dose response and
greater uncertainty
due to use of





20%) had a similar
HED (13 mg/kg-day)
but effect is typically

consumption or
body weight

male pups per
litter) due to
errors in study

single dose level
resulting in a
LOAEL





variable



report



Sensitivity of

Within an order of

Within an order of

Most sensitive

Within an order

Less sensitive

Endpoint and

magnitude of the

magnitude of the

endpoint for the

of magnitude of

endpoint but is

POD

most sensitive

most sensitive

short-term

most sensitive

used with a larger



endpoint

endpoint

scenario

endpoint

benchmark MOE

Total UF/

30

30

30

30

300

Benchmark MOE











Uncertainty/

N/A

N/A

BMR of 5% is

BMR of 5% is

N/A

Sensitivity of





lower than

lower than



BMR Selection





responses in
study

responses in
study



7568

Page 281 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

7569 Table 5-49. Dose-Response Analysis of Selected Studies Considered for Short-Term Exposure Scenarios

Target Organ/
System

Species

Duration

Study POD/
Type (mg/kg-
day)

Effect

HEC
(mg/m3)
|ppm|

HED

(mg/kg-day)

UFs

Reference

Overall
Quality
Determination

Reproductive
Toxicity

ICR mice
(males)

35 days

BMDL5 =
21"

Decreased
numbers of
seminiferous
tubules

14.9
[1.27]

2.73

UFA=3
UFH=10
Total UF=30

Chen et al.
(2015a):
(Johnson et
al. 2003)

High

Neurotoxicity

Sprague-

Dawley

rats

(females)

60 days

NOAEL =
50

Hippocampal
lesions

64.3
[5.51]

11.8

UFA=3
UFH=10

Total UF=30

Yang et al.
(2018a);
(Sel grade and
Gilmour.
2010)

High

Developmental
Toxicity

CD-I

mice
(both)

Up to 18

weeks

BMDL5 =
71.5

Decreased live
male F1 pups per
litter

51.7
[4.43]

9.51

UFA=3
UFH=10

Total UF=30

NTP (1991a)

High

Kidney
Toxicity

F344 rats
(males)

28 days

LOAEL = 350

Regenerating
tubules in
kidneys

450
[38.6]

82.8

UFA=3

UFH=10

UFL=10

Total UF=300

Taniai et al.

Medium

a The BMDL based on 1SD is 61.2 mg/kg-day.

7570

Page 282 of 572

-------
7571

7572

7573

7574

7575

7576

7577

7578

7579

7580

7581

7582

7583

7584

7585

7586

7587

7588

7589

7590

7591

7592

7593

7594

7595

7596

7597

7598

7599

7600

7601

7602

7603

7604

7605

7606

7607

7608

7609

7610

7611

7612

7613

7614

7615

7616

7617

7618

7619

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Comparison of Studies and Hazard Outcomes for the Chronic Exposure Scenario: EPA generally
considers chronic studies to be those with exposure durations of > 10 percent of a lifetime. For TCEP,
these studies include the 16- and 18-week and 2-year NTP studies in rats and mice (I	). Also,

many of the endpoints in the RACB study (NTP. 1991a) (especially the crossbreeding and second-
generation effects) were measured after chronic exposure. Table 5-50 presents a comparison of the
attributes of sensitive endpoints from studies considered for the chronic exposure scenario, and Table
5-51 provides study details including PODs from the study or BMD modeling results, HECs, and HEDs.

Although it is a study with a shorter exposure duration, EPA chose Chen et al. (2015a) for the chronic
exposure scenarios because it resulted in an HED that is more sensitive (2.79 mg/kg-day) than most
longer-term results and covers a potentially sensitive life stage (adolescence).

Use of the shorter duration study by Chen et al. (2015 a). however, does lend uncertainty to the risk
evaluation because other longer-term studies are not as sensitive and because it is uncertain whether the
POD would be lower if Chen et al. (2015a) extended the exposure duration.

For the endpoints that resulted in likely evidence integration conclusions, most chronic studies received
high overall qualtiy determinations. There were a few exceptions. EPA gave medium overall quality
determinations to the sperm morphology and vaginal cytology results reported in the 16- and 18-week
NTP studies (Matthews et; 0) primarily based on limited information regarding methods and
results. Clinical observations described by N	for the 16- and 18-week studies in mice and rats

received uninformative overall quality determinations due to the lack of quantitative information for
these effects.

The single chronic endpoint more sensitive than Chen et al. (2015a) was increased relative kidney
weights for female rats from the 16-week NTP study, with an HED of 1.75 mg/kg-day (	).

However, EPA considered the changes in kidney weights for TCEP less relevant for predicting kidney
toxicity than other endpoints {i.e., kidney histopathology) because they were not consistently observed;
female rats had increased relative kidney weights after 16 weeks but not after 66 weeks, and female
mice had increased weights at 16 days but not at 16 weeks or the 66-week sacrifice. In addition, kidney
weight changes did not correspond to histopathology changes (	).

Histopathology is a more reliable endpoint for kidney effects and was observed in the 2-year studies
(NTP. 1991b); daily HEDs associated with hyperplasia and karyomegaly ranged from 5.49 to 14.2
mg/kg-day; most are within a factor of three of Chen et al. ( i and 14.2 mg/kg-day is roughly five
times higher.

Neurotoxicity was consistently observed across chronic studies with HEDs ranging from 7.43 to 22.8
mg/kg-day. These HEDs are all within an order of magnitude of Chen et al. (2015a).

The comparison of HEDs with reproductive endpoints described earlier and the comparisons with
kidney and neurotoxicity endpoints observed in the chronic studies demonstrates some consistency
across endpoints with respect to potency. These co-critical endpoints lend strength to using the sensitive
endpoint from Chen et al. (2015a) for the chronic duration.

Similar to Chen et al. (2015a). only two dose groups (44 and 88 mg/kg-day) were used in
2-year studies associated with the most sensitive of the kidney and neurotoxic effects, which somewhat
limits the understanding of the dose response relationship for these endpoints.

Page 283 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

7620	Overall, the HED from Chen et al. (2015a) associated with a relevant hazard outcome is protective of

7621	other observed adverse effects from chronic exposure to TCEP that include decreased fertility and live

7622	pups per litter as well as neurotoxicity and kidney histopathological effects.

7623

7624	Table 5-50. Comparison among Studies with Sensitive Endpoints Considered for Chronic

7625	Exposure Scenarios 				



Neurotoxicity
(MP, 1991b)

Reproductive
Toxicity (Chen et

al-, 2015a)

Developmental
Toxicity (

1991a)

Kidney (

1991b)

Overall Data Quality
Determination

High

High

High

High

Exposure Duration

2-year; chronic

3 5-day; short-term (<
chronic)

Up to 18 weeks;
short-term/chronic

2-year; chronic

Dose Range

44, 88 mg/kg-day

100, 300 mg/kg-day

F0: 175, 350, 700
mg/kg-day

44, 88 mg/kg-day

Relevance

Endpoint assumed
to be relevant to
humans

Endpoint assumed
relevance to human
male reproduction
( 4. 1996):
severity identified

Endpoint assumed
to be relevant to
humans

Endpoint assumed
to be relevant to
humans

Uncertainties Not
Captured Elsewhere

Dose-response less
precise (use of
NOAEL)

Dose precision
unclear based on
dietary study with no
information on food
consumption or body
weight changes

Decreases in live
pups per litter for
2nd generation less
clear due to error in
data.

Some

inconsistencies
between kidney
weight changes and
histopathology

Sensitivity of Endpoint
and POD

Most sensitive
among chronic
neurotoxic effects

Most sensitive across
hazard outcomes
(except increased
kidney weight in 16-
week study)

Less sensitive than
male reproductive
toxicity in Chen

Most sensitive
among chronic
histopathological
kidney effects; 16-
week kidney weight
change more
sensitive

Total UF

30

30

30

30

Uncertainty/Sensitivity of
BMR Selection

N/A

BMR of 5 percent,
predicted BMD and
BMDL values are
lower than doses
associated with
responses observed in
the study

BMR of 5 percent,
predicted BMD and
BMDL values are
lower than doses
associated with
responses in the
study

BMR of 10 percent

7626

Page 284 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table 5-51. E

•ose-Response Analysis of Selected Stut

ies Consideret

for Chronic Exposure Scenarios

Target Organ
System

Species/Sex
Exposed

Duration

Study
POD/Type
(mg/kg-day)

Effect

HEC
(mg/m3)
|ppm|

HED

(mg/
kg-day)

UFs

Reference

Overall
Quality
Determination

Reproductive
Toxicity

ICR mice
(male)

35 days

BMDL5 =
21"

Decreased
numbers of
seminiferous
tubules

14.9
[1.27]

2.73

UFA=3
UFH=10
Total UF=30

Chen et al.
(2015a):
(Johnson et al..
2003)

High

Neurotoxicity

F344 rats
(female)

Two years

NOAEL = 31.4

Brain lesions

40.4
[3.46]

7.43

UFA=3
UFH=10

Total UF=30

]

High

Developmenta
1 Toxicity

CD-I mice

Up to 18

weeks

BMDL5 = 71.5

Decreased live
F1 male pups
per litter

51.7
[4.43]

9.51

UFA=3
UFH=10

Total UF=30

NIP (1991a)

High

Kidney
Toxicity

F344 rats
(female)

Two years

BMDLio= 23.2

Renal tubule
hyperplasia

30
[2.6]

5.49

UFA=3
UFH=10

Total UF=30

]

High

a The BMDL based on 1SD is 61.2 mg/kg-day.

7628

Page 285 of 572

-------
7629

7630

7631

7632

7633

7634

7635

7636

7637

7638

7639

7640

7641

7642

7643

7644

7645

7646

7647

7648

7649

7650

7651

7652

7653

7654

7655

7656

7657

7658

7659

7660

7661

7662

7663

7664

7665

7666

7667

7668

7669

7670

7671

7672

7673

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

5.2.6.1.3 Uncertainty Factors Used for Non-cancer Endpoints

For the non-cancer health effects, EPA used a total UF of 30 for the benchmark MOEs for acute, short-
term, and chronic exposure durations for all exposure routes among studies that are used to estimate
risks. Other endpoints that used LOAELs for which EPA used a LOEAL-to-NOAEL UF of 10 and a
total benchmark MOE of 300.

1)	Interspecies Uncertainty Factor (UFa) of 3

EPA uses data from oral toxicity studies in animals to derive relevant HEDs, and (
201 la) recommends allometric scaling (using the 3/4 power of body weight) to account for
interspecies toxicokinetics differences for oral data. When applying allometric scaling, EPA
guidance recommends reducing the UFa from 10 to 3. The remaining uncertainty is associated
with interspecies differences in toxicodynamics. EPA also uses a UFa of 3 for the inhalation
HEC and dermal HED values because these values are derived from the oral HED.

2)	Intraspecies Uncertainty Factor (UFh) of 10

EPA uses a default UFh of 10 to account for variation in sensitivity within human populations
due to limited information regarding the degree to which human variability may impact the
disposition of or response to, TCEP.

3)	LOAEL-to-NOAEL Uncertainty Factor (UFl) of 1 or 10

The PODs chosen to calculate risks were either NOAELs or BMDL values and therefore, EPA
used a UFl of 1. EPA compared these values with other endpoints based on LOAELs, which
used a UFl of 10 to account for the uncertainty inherent in extrapolating from the LOAEL to the
NOAEL.

and U.S. EPA. (2002b) further discuss use of UFs in human health hazard dose-
response assessment.

5.2.6.2 Selection of Studies and Endpoint Derivation for Carcinogenic Dose-Response
Assessment

EPA considered the kidney tumors for derivation of toxicity values for the risk calculations based on the
evidence integration conclusion that the tumors are sensitive and robust, and that cancer is likely to be
caused by TCEP. The selection of representative cancer studies and tumors for dose-response analysis is
described below based on the following considerations:

•	Overall quality determination;

•	Sufficiency of dose-response information;

•	Strength of the evidence supporting the associated tumor type;

•	MO A conclusions;

•	Relevance (e.g., what species was the effect in, was the study directly assessing the effect, is the
endpoint the best marker for the tox outcome?);

•	Uncertainties not captured by the overall quality determination; and

•	Endpoint sensitivity.

Rodent bioassays identify increased incidences of kidney tumors in male F344/N rats, with a lower
increase in female rats (NT	). Treatment-related kidney tumors were also observed after two

years in male B6C3Fi mice (NTP. 1991b). EPA gave N	a high overall quality determination.

Page 286 of 572

-------
7674

7675

7676

7677

7678

7679

7680

7681

7682

7683

7684

7685

7686

7687

7688

7689

7690

7691

7692

7693

7694

7695

7696

7697

7698

7699

7700

7701

7702

7703

7704

7705

7706

7707

7708

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Based on a lack of adequate information on mechanisms or temporality and dose-response data for
precursor lesions to model the tumors using a threshold analysis, EPA used linear low-dose
extrapolation to estimate risks. U.S. EPA's PPRTV also used linear low-dose extrapolation in the
absence of specific mechanistic information.

EPA used the multistage models available in the BMD software and adjusted the data for mortality by
using animals still alive on the first day of cancer incidence. Therefore, animals dying from other causes
were not included in the analysis. For both male and female rats, kidney tumor incidence data
adequately fit one or both multistage models and tumors in males (adenomas and carcinomas) resulted
in the more sensitive CSF (0.0058 per mg/kg-day). The MR is based on daily, continuous
concentrations (24 hours per day) using a breathing rate for individuals at rest. Adjustments to exposure
durations, exposure frequencies, and breathing rates are made in the exposure estimates used to calculate
risks for individual exposure scenarios.

Table 5-52 presents the cancer PODs for modeled renal tumors. Because EPA has not concluded that
TCEP acts via a mutagenic mode of action, an age-dependent adjustment factor (ADAF) (

2005c) was not applied when estimating cancer risk for kidney tumors from TCEP exposure. EPA did
not use CSFs for combined tumors (across multiple target organs) for the risk evaluation but focused on
the tumors with the most robust evidence from the animal data.

See Appendix J.3 for dose-response derivation, including details on route-to-route extrapolation.
Considerations regarding the BMD modeling process for cancer and results are presented in Benchmark
Dose Modeling Results for TCEP (U.S. EPA. 2023b).

EPA did not use CSFs for combined tumors (across multiple target organs) for the risk evaluation but
focused on the tumors with the most robust evidence from the animal data.

Table 5-52. Dose-Response Analysis of Kidney Tumors" for Lifetime Exposure Scenarios

Tumors

Species (sex)

Oral/Dermal

CSF"b

IUR"

Extra Cancer Risk
Benchmark

Renal tubule
adenomas or
carcinomas

F344 rats (male)

0.0245 per mg/kg-
day

0.00451 permg/m3
(0.0526 per ppm)

1/10 4 (occupational)
1 x 10~4 to 1 x 10~6 (consumer,
general population)

Renal tubule
adenomas

F344 rats (female)

0.0220 per mg/kg-
day

0.00404 permg/m3
(0.0472 per ppm)

a CSFs and IURs were derived based on continuous exposure scenarios; CSFs from BMD modeling prior to allometric scaling
were 0.0058 and 0.0052 per mg/kg-day for male and female rats, respectively.

h U.S. EPA's PPRTV (U.S. EPA. 2009) calculated an oral CSF of 0.02 per me/ke-dav. also based on increased renal tubule
adenomas or carcinomas in male rats from NTP (1991b).

5.2,7 Weight of the Scientific Evidence Conclusions for Human Health Hazard

EPA considered evidence integration conclusions from Sections 5.2.3 and 5.2.5.4 and additional factors
listed below when choosing studies for dose-response modeling and for each exposure scenario (acute,
short-term/intermediate, and chronic), as described in Section 5.2.6. Additional considerations pertinent
to the overall hazard confidence levels that are not addressed in previous sections are described below
(see Section 5.2.7.1):

• Evidence integration conclusion (from Appendix K)

Page 287 of 572

-------
7709

7710

7711

7712

7713

7714

7715

7716

7717

7718

7719

7720

7721

7722

7723

7724

7725

7726

7727

7728

7729

7730

7731

7732

7733

7734

7735

7736

7737

7738

7739

7740

7741

7742

7743

7744

7745

7746

7747

7748

7749

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

o Demonstrates is rated as +++
o Likely is rated as ++
o Suggests is rated as +

•	Selection of most critical endpoint and study

•	Relevance to exposure scenario

•	Dose-response considerations

•	PESS sensitivity

Section 5.2.7.2 presents a summary table of confidence for each hazard endpoint and exposure duration.

5.2.7.1 Strengths, Limitations, Assumptions, and Key Sources of Uncertainty for the
Hazard Identification and Selection of PODs for Human Health Hazard
Assessment

5.2.7.1.1 Acute Non-cancer
Evidence Integration Conclusions

Clinical signs of neurotoxicity, histopathological changes in the brain, and neurobehavioral changes
measured in multiple studies were considered for the acute exposure scenario. EPA concluded that
TCEP likely causes neurotoxicity in humans under relevant exposure circumstances and assigned high
overall quality determinations to all acute studies considered.

Selection of Most Critical Endpoint and Study

EPA did not locate human studies that evaluated neurotoxicity. However, the tremors observed in Moser
et al. (2015) and similar neurotoxic effects in other studies are critical because they are adverse, and
neurotoxicity is consistently observed among acute and longer-term studies.

Offspring do not appear to be more sensitive for developmental neurotoxicity up to 90 mg/kg-day43
after exposure of pregnant rats during gestation and the early postnatal period based on results from
Moser et al. (2015). Viability and growth of offspring were also not affected after pregnant mice were
dosed with 940 mg/kg-day (Hazletom Laboratories. 1983).44

Relevance to Exposure Scenario

The candidate studies and endpoints for acute exposure identified neurotoxicity after one to eight days,
and EPA considered these durations relevant for the acute exposure scenario. Moser et al. (2015). the
study chosen to calculate risks, identified tremors within five days of exposure. There is some
uncertainty for this human exposure scenario given the lack of TCEP-specific information or models
(e.g., PBPK models) to extrapolate from animals to humans. EPA also extrapolated from oral HEDs to
inhalation HECs and dermal HEDs, which lends uncertainty for these routes. It is not known whether
these assumptions for the chosen POD would lead to over- or underprediction of risk from acute
exposure.

Dose-Response Considerations

None of the studies considered for acute exposure could be modeled using BMD models due to limited
dose-response information. EPA identified a NOAEL from Moser	but effects were seen

43	The study began with a dose of 125 mg/kg-day, which was lower to 90 mg/kg-day after 5 days due to toxicity in dams at
the highest dose.

44	A prenatal study in Wistar rats (Kawashima et al. 1983") in a foreign language will be translated it into English and
evaluated for the final risk evaluation.

Page 288 of 572

-------
7750

7751

7752

7753

7754

7755

7756

7757

7758

7759

7760

7761

7762

7763

7764

7765

7766

7767

7768

7769

7770

7771

7772

7773

7774

7775

7776

7777

7778

7779

7780

7781

7782

7783

7784

7785

7786

7787

7788

7789

7790

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

only at the highest dose. The other acute studies also identified only a NOAEL or LOAEL with effects
observed only at the highest dose or the only dose in the study.

Susceptible Subpopulations

Moser et al. (2015) evaluated effects in pregnant female rats. Given the lower HED for this study
compared with other acute studies, pregnant dams may be a susceptible subpopulation. However,
uncertainties exist because of limited dose response information for other studies. Non-pregnant female
rats are also shown to be a sensitive species and sex for neurotoxicity in longer-term studies as identified
in N]	. Offspring, as noted earlier, were not identified as more sensitive to neurotoxicity or

other effects from gestational and postnatal exposure of the dams.

5.2.7.1.2 Short-Term and Chronic Non-cancer
Evidence Integration Conclusions

EPA considered multiple animal toxicity studies and multiple hazard outcomes - reproductive toxicity,
neurotoxicity, developmental toxicity, and kidney toxicity - for the short-term and chronic exposure
scenarios. EPA concluded that TCEP likely causes all these outcomes in humans under relevant
exposure circumstances. EPA assigned the studies and endpoints high quality determinations except
Taniai et al. (2012a). which EPA gave a medium quality determination.

Selection of Most Critical Endpoint and Study

The nature of the effect chosen for calculating risks—differences in numbers and degeneration of
seminiferous tubules identified by Chen et al. (2015a)—is considered adverse, and the fertility of human
males is known to be sensitive to changes in sperm numbers and quality (	n5).

Neurotoxicity and kidney toxicity were also observed consistently among studies and HEDs were often
within an order of magnitude of each other.

The effects of Chen et al. (2015a) were the most sensitive after short-term exposure. Increased relative
kidney weight was most sensitive after chronic exposure, but EPA considered these weight changes less
predictive of kidney toxicity due to inconsistencies between short-term and longer-term studies and lack
of correlation with histopathology and clinical chemistry results in many cases.

Using Chen et al. (2015a) does lead to uncertainty because other studies did not report decreased
numbers or disintegration of seminiferous tubules; furthermore, related male reproductive effects were
only seen at higher doses in other studies. However, male reproduction was consistently affected in
several studies along with fertility and offspring viability. Thus, EPA considers the sensitive effects in
Chen et al. (2015a) to be relevant and differences might be due to species, test methods, or life stage.

There are several considerations that lend uncertainty as to whether risks could be underpredicted using
this POD. These include lack of human data; the known sensitivity of human males to reproductive
insults; and uncertainty about certain sensitive effects that could not be considered for a POD due to an
error in the results presented in the continuous breeding study (NTP. 1991a) or lack of full reports (see
Section 5.2.3.1.2).45

45 Data from Sfaepet'skaia and Dvshginevich (.1.98.1.) (cited in (NTP. 1991a")") suggests that reproductive effects by inhalation
(decreased fetal size) at 0.5 mg/m3 could be a LOAEC. Dividing this possible LOAEC by a total MOE of 300 yields
1.7 x 10-3 mg/m3, which is 300 times more sensitive than dividing the HEC of 14.9 mg/m3 based on Chen et a I. (20.1.5a) by the
total MOE of 30 (which results in 0.5 mg/m3). Even if the value of 0.5 mg/m3 from Shepel'skaia and Dvshginevich (1981) is
a NOAEC. the POD/MOE is still 30 times more sensitive than using the POD from Chen et a I. (2015a). Shepel'skaia and
Dvshginevich (.1.98.1.) was not readily available to EPA and appears to be only an abstract. Thus. EPA cannot consider
Shepel'skaia and Dvshginevich (.1.98.1.) for use in this risk evaluation.

Page 289 of 572

-------
7791

7792

7793

7794

7795

7796

7797

7798

7799

7800

7801

7802

7803

7804

7805

7806

7807

7808

7809

7810

7811

7812

7813

7814

7815

7816

7817

7818

7819

7820

7821

7822

7823

7824

7825

7826

7827

7828

7829

7830

7831

7832

7833

7834

7835

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

There is some uncertainty as to whether this POD is protective of a full range of effects. For example,
chronic studies did not evaluate neurobehavioral batteries. In addition, EPA did not locate any studies
that investigated TCEP's association with acoustic startle responses or social behaviors.

Relevance to Exposure Scenarios

The 35-day exposure used by Chen et al. (2015a) is more relevant than the shorter and longer studies of
5 or 60 days (e.g., Moser et al. (2015) and Yang et al. (2018a)) for the short-term exposure scenario,
which EPA defines as a 30-day exposure for this risk evaluation. Although the 28-day Taniai et al.
(2012a) study is well-suited for short-term exposures, other study aspects limit its suitability, including
testing at only 350 mg/kg-day.

There is inherent uncertainty in assuming that a 35-day toxicity study in rodents during male
adolescence is applicable to a similar exposure duration in human adolescent males for the endpoint of
decreased numbers of seminiferous tubules.

Using Chen et al. (2015a) to represent chronic exposure durations adds uncertainty to the risk
evaluation. If the specific effect identified by Chen et al. (2015a) were measured in a chronic study in
the same species starting in adolescence, the POD could be more sensitive. Therefore, it is possible that
risks might be under-predicted. Yet, among the available chronic studies, HEDs were less sensitive than

Chen et al. (2015a).

For all studies and endpoints, no TCEP-specific information was available for extrapolation to humans
and EPA relied on allometric scaling based on BW3 4 Route-to-route extrapolation to inhalation HECs
and dermal HEDs results in additional uncertainty. EPA cannot predict whether the assumptions
regarding route extrapolation for the chosen POD would lead to over- or underprediction of risk from
short-term exposure for the dermal route.46

Dose-Response Considerations

Chen et al. (2015a) fed TCEP to rats in a dietary study and do not report information on food
consumption. Thus, EPA does not know the precise doses received by the rats. However, the data
adequately fit several BMD models based on statistics and visual inspection and resulted in similar
BMDLs among the fit models. Also, use of the BMDL allowed EPA to use a relatively low total UF of
30. Given the severity of the effect (large percent decrease in numbers of tubules and significant
degeneration), EPA chose a BMR of 5 percent.

Although other short-term studies with relevant sensitive effects used three treatment levels (vs. two for
Chen et al. (2015a)). EPA identified limitations for these other studies that included the inability to
conduct BMD modeling, use of only one dose (with LOAEL only) or an effect seen only at the highest
dose. Sensitive chronic neurotoxic and kidney effects are from studies with two treatment levels;
neurotoxicity could not be modeled (and only a NOAEL is available) but kidney hyperplasia could be
modeled and yielded an appropriate BMDL.

Susceptible Subpopulations

Chen et al. (2015a) evaluated a sensitive sex life stage (male adolescent mice) and identified a sensitive
POD among critical endpoints. Other studies and endpoints considered for short-term and chronic

46 Limited data from Shepel'skaia and Dyshginevich (1981) (cited in	and likely only an abstract)

suggests a possible greater sensitivity to TCEP via inhalation.

Page 290 of 572

-------
7836

7837

7838

7839

7840

7841

7842

7843

7844

7845

7846

7847

7848

7849

7850

7851

7852

7853

7854

7855

7856

7857

7858

7859

7860

7861

7862

7863

7864

7865

7866

7867

7868

7869

7870

7871

7872

7873

7874

7875

7876

7877

7878

7879

7880

7881

7882

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

exposure identified sexes that might be more sensitive to certain effects. For example, female rats were
more sensitive for neurotoxicity.

5.2.7.1.3 Cancer
Evidence Integration Conclusions

EPA concludes that TCEP is likely to be carcinogenic to humans using guidance from U.S. EPA's
Guidelines for Carcinogen Risk Assessment (	)5b) based on information from a high-

quality study (N	).

Selection of Most Critical Endpoint and Study

Of the organs that exhibited tumors in	, EPA used the tumor type with the most robust

evidence - kidney adenomas and carcinomas - and used a CSF that was the most sensitive among
modeled kidney tumor incidence.

EPA considers increased incidence of renal tubule adenomas and carcinomas to be adverse, relevant to
humans, and representative of a continuum of benign to malignant tumors and was the only target organ
with robust evidence of increased tumors. There is some support for TCEP's association with thyroid
tumors in humans based on a case control study (Hoffman et at.. 2017).

Of the kidney tumors,	identified primarily adenomas and only one carcinoma. Thus, the

risk of malignant tumors is less certain; if humans are like rodents, use of the CSF from
could result in an over prediction of malignant cancer. However, if humans are more sensitive and
develop malignancies sooner, risks may be underpredicted.

Relevance to Exposure Scenarios

N	is a 2-year bioassay and is relevant for chronic exposures in humans. However, like non-

cancer endpoints, use of allometric scaling among species and route-to-route extrapolation to inhalation
HECs and dermal HEDs leads to some uncertainties and the impacts on risks are unknown.

Dose-Response Considerations

There is no complete understanding regarding mechanism(s) of cancer and there is also a lack of
appropriate precursors to cancer in the available in vivo studies with respect to temporality and dose
response (e.g., the single dose used by Taniai et al. (2012a) is higher than doses associated with tumors).
Therefore, EPA used linear low dose extrapolation a BMDLio. Because direct mutagenicity is not likely
to be the predominant MO A, using linear low dose extrapolation is a health conservative analysis that
would overpredict risks assuming that TCEP acts via a threshold MOA.

Use of tumor data for only one target organ (i.e., not combining incidence with other target organ
tumors) may result in some underestimation of risk, however. Therefore, the net effect of the dose-
response modeling, considering the benchmark risk levels used in the risk evaluation (1 in 10,000 to 1 in
1,000,000) is not known.

Susceptible Subpopulations

The single human study identified regarding TCEP exposure and thyroid cancer did not identify a
specific susceptible subpopulation (Hoffman et al.. 2017). Availability of a high-quality animal study
using two species and both sexes suggests possible sensitivities by sex (e.g., higher incidence of kidney
tumors in male rats).

Page 291 of 572

-------
7883

7884

7885

7886

7887

7888

7889

7890

7891

7892

7893

7894

7895

7896

7897

7898

7899

7900

7901

7902

7903

7904

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

The dose-response model applied to animal tumor data employed low-dose linear extrapolation, and this
assumes any TCEP exposure is associated with some positive risk of getting cancer. However, EPA did
not identify specific human groups that are expected to be more susceptible to cancer following TCEP
exposure even though there is likely to be variability in susceptibility across the human population.
Other than relying on animal tumor data for the more sensitive sex, the available evidence does not
allow EPA to evaluate or quantify the potential for increased cancer risk in specific subpopulations.
Given that a mutagenic mode of action is unlikely, EPA does not anticipate greater cancer risks from
early life exposure to TCEP.

5.2.7.2 Human Health Hazard Confidence Summary

Table 5-53 summarizes the confidence ratings for each factor for critical human health hazards
considered for acute, short-term, chronic, and lifetime exposure scenarios. The bolded rows are the
health endpoints for each exposure scenario used to calculate risks. Alternate PODs for health outcomes
are not bolded in the table.

Table 5-53. Confidence Summary for Human Health Hazard Assessment

Hazard
Domain

Evidence
Integration
Conclusion

Selection of Most
Critical Endpoint
and Studv

Relevance to
Exposure
Scenario

Dose-Response
Considerations

PESS
Sensitivity

Overall
Hazard
Confidence

Acute non-cancer

Neurotoxicity

+ +

+ + +

+ +

+ +

+ +

Moderate

Shorl-lcrm non-cancei'

Reproductive

+ +

+ +

+ + +

+

+ +

Moderate

Neurotoxicity

+ +

+

+ +

+ +

+ +

Moderate

Developmental

+ +

+

+ + +

+ +

+ +

Moderate

Kidnc>

+ +

+

+ + +

+

+

Moderate

( lironic 11011-canccr

Reproductive

+ +

+ +

+

+

+ +

Moderate

Neurotoxicity

+ +

+

+ + +

+ +

+ +

Moderate

Developmental

+ +

+

+ + +

+ +

+ +

Moderate

Kidnc\

+ +

+

+ + +

+ +

+

Moderate

( anccr

Kidney Cancer

+ +

+ +

+ + +

+ +

+ +

Moderate

+ + + Robust confidence suggests thorough understanding of the scientific evidence and uncertainties. The supporting
weight of the scientific evidence outweighs the uncertainties to the point where it is unlikely that the uncertainties could
have a significant effect on the hazard estimate.

+ + Moderate confidence suggests some understanding of the scientific evidence and uncertainties. The supporting
scientific evidence weighed against the uncertainties is reasonably adequate to characterize hazard estimates.
+ Slight confidence is assigned when the weight of the scientific evidence may not be adequate to characterize the
scenario, and when the assessor is making the best scientific assessment possible in the absence of complete information.
There are additional uncertainties that may need to be considered.

5.2.8 Toxicity Values Used to Estimate Risks from TCEP Exposure	

After considering hazard identification and evidence integration, dose-response evaluation, and weight
of the scientific evidence of POD candidates, EPA chose two non-cancer endpoints for the risk
evaluation—one for acute exposure scenarios and a second one for short-term and chronic scenarios
(Table 5-54). Cancer risks were estimated using increased kidney tumors in male rats (Table 5-55).
HECs and IURs are based on daily continuous (24-hour) exposure and HEDs and CSFs are daily values.
All studies received high overall quality determinations.

Page 292 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table 5-54.

Von-cancer HECs and HE

)s Used to Estimate Risks

Exposure
Scenario

Target Organ
System

Species
(Sex)

Duration

POD

(mg/kg-day)

Effect

HEC
(mg/m3)
|ppm|

HED

(mg/
kg-day)

Benchmark
MOE

Reference(s)

Acute

Neurotoxicity

Long Evans
rats (dams)

5 days

NOAEL =
40

Tremors

51.5
[4.41]

9.46

UFA=3
UFH=10

Total UF=30

Moser et al.

(2015)

Short-term
and Chronic

Reproductive
Toxicity

ICR mice
(male)

35 days

BMDL5 =
21

Decreased

seminiferous

tubules

14.9
[1.27]

2.73

UFA=3
UFH=10

Total UF=30

Chen et al.
(2015a):
("Johnson et al..

2003)

Table 5-55.

Cancer IUR and CSF Uset

to Estimate Risks

Exposure
Scenario

Target Organ
System

Species
(Sex)

Duration

POD

(mg/kg-day)

Effect

IUR
(per mg/m3)
[per ppm]

CSF
(per mg/
kg-day)

Benchmark
Risk Levels

Reference

Chronic/
Lifetime

Kidney tumors

Fischer
344/N rats
(male)

2 years

CSF from
BMD model
= 0.0058 per
mg/kg-day

Increased
renal tubule
adenomas or
carcinomas

0.00451
[0.0526]

0.0245

1E10-4

(occupational)

1E-4 to 1E-6

(consumer,

general

population)

]

7909

7910

Page 293 of 572

-------
7911

7912

7913

7914

7915

7916

7917

7918

7919

7920

7921

7922

7923

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

5.2.9 Hazard Considerations for Aggregate Exposure	

For use in the risk evaluation and assessing risks from other exposure routes, EPA conducted route-to-
route extrapolation of the toxicity values from the oral studies for use in the dermal and inhalation
exposure routes and scenarios. Because the health outcomes are systemic and are based on the oral
studies, EPA considers it is possible to aggregate risks across exposure routes for all exposure durations
and endpoints for the selected PODs identified in Sections 5.2.6.1 and 5.2.6.2.

5.3 Human Health Risk Characterization

TCEP - Human Health Risk Characterization (Section 5.3):

Key Points

EPA evaluated all reasonably available information to support human health risk characterization.

The key points of the human health risk characterization are summarized below:

•	Dermal exposures drive risks to workers in occupational settings and both cancer risks and
non-cancer MOEs that met benchmarks were observed for most COUs, whereas risks and
MOEs from inhalation exposure met benchmarks for multiple commercial paints and coatings
use scenarios within a single COU.

•	Fish ingestion is the primary exposure route driving risks to the general population. People who
are subsistence fishers may be at high risk if they eat TCEP-contaminated fish; tribal people for
whom fish is important dietarily and culturally have even higher risk than the general
population and subsistence fishers.

•	Mouthing by infants and children is the primary exposure route driving risks to consumers for
articles expected to be mouthed.

•	Infants exposed through human milk ingestion are not more sensitive than the mothers. The
COUs that present infant risks also result in maternal risks. There are no COUs that show
infant risks but not maternal risks. Therefore, protecting the mother will also protect the infant
from exposure via human milk.

5.3.1 Risk Characterization Approach

The exposure scenarios, populations of interest, and toxicological endpoints used for evaluating risks
from acute, short-term/intermediate, and chronic/lifetime exposures are summarized in Table 5-56.

Table 5-56. Exposure Scenarios, Populations of Interest, and Hazard Values



Workers



Male and female adolescents and adults (>16 years old) directly working with TCEP



under light activity (breathing rate of 1.25 m3/hr)



Exposure durations



• Acute - 8 hours for a single workday (most OESs)



• Short-term - 8 hours per workday for 22 working days



• Chronic - 8 hours per workday for 250 days per year for 31 or 40 working years



Exposure routes - Inhalation and dermal

Page 294 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Populations of Interest
and Exposure Scenarios

Occupational Non-users

Male and female adolescents and adults (>16 years old) indirectly exposed to TCEP
within the same work area as workers (breathing rate of 1.25 m3/hr)

Exposure durations

• Acute, Short-term, and Chronic - same as workers
Exposure route - Inhalation



Consumers

Male and female infants, children and adults using articles that contains TCEP
Exposure durations

•	Acute - 1 day exposure

•	Chronic - 365 days per year
Exposure routes

•	Adults - Inhalation and dermal

•	Infants and Children - Inhalation, dermal, and oral

Populations of Interest
and Exposure Scenarios

General Population

Male and female infants, children and adults exposed to TCEP through drinking water,
ambient water, ambient air, soil, and diet
Exposure durations

•	Acute - Exposed to TCEP continuously for a 24-hour period

•	Chronic - Exposed to TCEP continuously up to 33 years

Exposure routes - Inhalation, dermal, and oral (depending on exposure scenario)



Infants (Human Milk Pathway)

Infants exposed to TCEP through human milk ingestion
Exposure durations

•	Short term - Exposed to TCEP continuously for 30 days

•	Chronic - Exposed to TCEP continuously for one year
Exposure routes - Oral

Health Effects, Hazard
Values, and Benchmarks

Non-cancer Acute Hazard Values b

Sensitive health effect: Neurotoxicity

HEC Daily, continuous = 51.5 mg/m3 (4.41 ppm)

HED Daily = 9.46 mg/kg; dermal and oral

Total acute UF (benchmark MOE)

= 30 (UFA = 3; UFh= 10)c

Non-cancer Short-Term/Chronic Values b

Sensitive health effect: Male reproductive effects

HEC Daily, continuous = 14.9 mg/m3 (1.27 ppm)

HED Daily = 2.73 mg/kg; dermal and oral

Total short-term/chronic UFs (benchmark MOE)

= 30 (UFA = 3; UFh= 10)c

Cancer Hazard Values b

Both values based on renal tumors

IUR Daily, continuous = 0.00451 per mg/m3 (0.0526 per ppm)

CSFdo^ = 0.0245 per mg/kg-day

" The chronic duration is the most relevant exposure scenario for the consumer COUs and is used to assess chronic non-
cancer and lifetime cancer risks. Acute exposure duration non-cancer risks are presented to help characterize risk.
b The inhalation HEC and IUR are extrapolated from the oral HED or CSF, which are estimated using allometric scaling
(BW3/4) and are associated with continuous or daily exposures. The HEC and IUR values assume a resting breathing rate
(0.6125 m3/hr). The dermal HED is assumed to equal the oral HED. See Appendix J.3 and Benchmark Dose Modeling
Results for TCEP in 023b) for dose derivation.
c Total UFs in the benchmark MOE.

UFa = interspecies (animal to human); UFH = intraspecies (human variability)

Page 295 of 572

-------
7924

7925

7926

7927

7928

7929

7930

7931

7932

7933

7934

7935

7936

7937

7938

7939

7940

7941

7942

7943

7944

7945

7946

7947

7948

7949

7950

7951

7952

7953

7954

7955

7956

7957

7958

7959

7960

7961

7962

7963

7964

7965

7966

7967

7968

7969

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

5.3.1.1 Estimation of Non-cancer Risks

EPA used a margin of exposure (MOE) approach to identify potential non-cancer risks. The MOE is the
ratio of the non-cancer POD divided by a human exposure dose. Acute, short-term, and chronic MOEs
for non-cancer inhalation and dermal risks were calculated using the following equation:

Equation 5-26.

Non — cancer Hazard Value (POD)

M0E= 	Human Exposure

Where:

MOE

Non-cancer Hazard Value (POD)
Human Exposure

Margin of exposure for acute, short-term, or chronic
risk comparison (unitless)

HEC (mg/m3) or HED (mg/kg-day)

Exposure estimate (mg/m3 or mg/kg-day)

MOE risk estimates may be interpreted in relation to benchmark MOEs. Benchmark MOEs are typically
the total UF for each non-cancer POD. The MOE estimate is interpreted as a human health risk of
concern if the MOE estimate is less than the benchmark MOE (i.e., the total UF). On the other hand, if
the MOE estimate is equal to or exceeds the benchmark MOE, the risk is not considered to be of concern
and mitigation is not needed. Typically, the larger the MOE, the more unlikely it is that a non-cancer
adverse effect occurs relative to the benchmark. When determining whether a chemical substance
presents unreasonable risk to human health or the environment, calculated risk estimates are not "bright-
line" indicators of unreasonable risk, and EPA has the discretion to consider other risk-related factors in
addition to risks identified in the risk characterization.

5.3.1.2 Estimation of Cancer Risks

Extra cancer risks for repeated exposures to a chemical were estimated using the following equations:

Equation 5-27

Inhalation Cancer Risk = Human Exposure x IUR

or

Dermal or Oral Cancer Risk = Human Exposure x CSF

Where:

Risk	=	Extra cancer risk (unitless)

Human Exposure	=	Exposure estimate (LADC in ppm)

IUR	=	Inhalation unit risk (risk per mg/m3)

CSF	=	Cancer slope factor (risk per mg/kg-day)

Estimates of extra cancer risks are interpreted as the incremental probability of an individual developing
cancer over a lifetime following exposure (i.e., incremental or extra individual lifetime cancer risk).

EPA considers a range of extra cancer risk from 1 x 10~4 to 1 x 10 6 to be relevant benchmarks for risk
assessment (U.S. EPA. ). Consistent with NIOSH guidance (Whittaker et at.. 2016). under TSCA
EPA typically applies a 1 x 10~4 benchmark for occupational scenarios in industrial and commercial work
environments subject to OSHA requirements. EPA typically considers the general population and
consumer benchmark for cancer risk to be within the range of 1x 10~6 and 1 x 10 4. Again, it is important

Page 296 of 572

-------
7970

7971

7972

7973

7974

7975

7976

7977

7978

7979

7980

7981

7982

7983

7984

7985

7986

7987

7988

7989

7990

7991

7992

7993

7994

7995

7996

7997

7998

7999

8000

8001

8002

8003

8004

8005

8006

8007

8008

8009

8010

8011

8012

8013

8014

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

to note that these benchmarks are not bright lines and EPA has discretion to find unreasonable risks
based on other risk-related considerations based on analysis. Exposure-related considerations (e.g.,
duration, magnitude, population exposed) can affect EPA's estimates of the excess lifetime cancer risk.

5.3.2 Summary of Human Health Risk Characterization

5.3.2.1 Summary of Risk Estimates for Workers

EPA estimated cancer risks and non-cancer MOEs for workers exposed to TCEP for multiple COUs
based on the occupational exposure estimates described in Section 5.3.2.1.1. Complete risk calculations
and results for the occupational OES/COUs are available in Draft Risk Evaluation for Tris(2-
chloroethyl) Phosphate (TCEP) - Supplemental Information File: Risk Calculator for Occupational
Exposures (U.S. EPA. 2023k).

5.3.2.1.1 COUs/OESs with Quantitative Risk Estimates

Table 5-57 summarizes cancer and non-cancer risk estimates for the inhalation and dermal exposures for
all OESs assessed. These risk estimates are based on exposures estimated for workers who do not use
PPE such as gloves or respirators. When both monitoring and modeling data were available for
inhalation exposures, EPA only presented the risk estimates for the most reliable data source in the
summary table. Estimates for inhalation and dermal exposures that have PPE factored in are contained in
the Draft Risk Evaluation for TCEP - Supplemental Information File: Risk Calculator for Occupational
Exposures (U.S. EPA. 2023k).

Exposure data for ONUs were not available for most COUs except for recycling (with recycling e-waste
as the relevant OES). For the COUs and OESs without ONU-specific exposure data, EPA assumed risks
would be equal to or less than risks to workers who handle materials containing TCEP as part of their
job. The inhalation risk values used for workers are also presented for ONUs in Table 5-57. EPA
assumed that ONUs are not exposed dermally.

Within the commercial use of paints and coatings COU, EPA did not calculate short-term or chronic
non-cancer risks or lifetime cancer risks for the 1-day spray application for commercial paint and
coating scenarios (OES #7 and #10) because risks were most appropriately assessed using only the
inhalation HEC and dermal HED values for acute exposures. Likewise, EPA did not calculate chronic
non-cancer or lifetime cancer risks for the 2-day commercial paint and coating spray application (OES
#8 and #11) given the very limited number of days per year of exposure. However, for OESs exposures
longer than one day per year, EPA also compared exposure with the acute hazard PODs.

Risks from Inhalation Exposure

Cancer inhalation risk estimates were above 1 in 10,000 for the commercial use of paints and coatings
COU for both central tendency and high-end exposures. These risks were associated with two OESs:
250-day applications of either 1- or 2-part sprays. Risk estimates were less than 1 in 10,000 for the
remaining six occupational COU subcategories.

In addition, inhalation non-cancer MOEs were less than benchmark MOEs for the commercial use of
paints and coatings COU for high-end exposures. Within this COU, high-end acute exposure for all
three OESs associated with 2-part spray applications resulted in MOEs less than the benchmark MOE of
30. For high-end short-term/chronic exposures, MOEs were less than the benchmark MOE of 30 for the
250-day applications of either 1- or 2-part sprays. No other COU/OES combinations resulted in MOEs
less than the non-cancer benchmark MOEs; this includes the commercial and industrial uses for the

Page 297 of 572

-------
8015

8016

8017

8018

8019

8020

8021

8022

8023

8024

8025

8026

8027

8028

8029

8030

8031

8032

8033

8034

8035

8036

8037

8038

8039

8040

8041

8042

8043

8044

8045

8046

8047

8048

8049

8050

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

installation of aerospace articles, which used surrogate monitoring data to estimate inhalation exposures
that could occur during these activities.

Risks from Dermal Exposure

More COU categories were associated with worker dermal risks above 1 in 10,000. Cancer dermal risk
estimates were above 1 in 10,000 for both central tendency and high-end exposures for certain
subcategories and OESs within the following five COU categories: import; incorporation into
formulation, mixture, or reaction products; processing - incorporation into an article; commercial use of
paints and coatings; and other commercial use - laboratory chemicals.

Additional dermal cancer risks above 1 in 10,000 were observed for only high-end exposures within a
single COU category (Processing - incorporation into formulations, mixtures, or reaction products) and
two associated OESs (Incorporation into 2-part paints and coatings and Formulation of 2-part reactive
resins).

Three COU categories had chronic non-cancer dermal MOEs less than the benchmark value of 30 for
both high-end and central tendency exposures. These were Processing - incorporation into articles,
Commercial use of paints and coatings, and Other commercial use - laboratory chemicals. Two
additional COUs were associated with MOEs lower than 30 for only high-end exposures; these were
Import and processing - incorporation into formulation, mixture, or reaction products.

For the short-term exposure scenario, MOEs were less than 30 for five COUs for at least some OESs.
Within two of these COUs, certain OESs had MOEs less than 30 for only high-end exposures—

Flame retardant in paints and coatings manufacture (2-part coatings and polymers in aerospace
equipment) and Commercial use of paints and coatings (2-day application for 1-part coatings).

For the acute exposure scenario, five COUs had dermal MOEs of less than 30 for both central tendency
and high-end exposures. One of these five COUs (commercial use of paints and coatings) also had some
OESs (1-part sprays) for which MOEs were less than 30 for only high-end exposures.

Processing/recycling was the single COU with cancer dermal risks less than 1 in 10,000 and all non-
cancer MOEs greater than benchmark values. Dermal risk estimates were not calculated for industrial
and commercial use of aerospace equipment products because EPA does not expect dermal exposure for
this COU because TCEP will be entrained in the polymer matrix.

Page 298 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

8051 Table 5-57. Occupational Risk Summary for 2,500-Pound Production Volume

cou





Exposure
Route and
Duration



Estimates for No PPE

Overall

Life Cycle

Stage/

Subcategory

OES

Population

Exposure
Level

Acute Non-
cancer MOE

Short-Term Non-
cancer MOE

Chronic Non-
cancer MOE

Lifetime
Cancer

Confidence
in Risk

Category









UFs = 30

UFs = 30

UFs = 30

Risk

Estimates









Inhalation
8-hr TWA

Central

6.8E03

1.4E04

1.7E05

1.5E-07









Worker

Tendency









Moderate









High-End

1.9E03

4.0E03

4.9E04

5.5E-07



Manufacturing/
import

Import

Repackaging

ONU"

Inhalation
8-hr TWA

Central
Tendency

6.8E03

1.4E04

1.7E05

1.5E-07

Slight



High-End

1.9E03

4.0E03

4.9E04

5.5E-07











Central

4.3E00

9.4E00

1.14E02

2.3E-04









Worker

Dermal

Tendency









Moderate











High-End

1.4E00

1.8E00

2.2E01

1.6E-03











Inhalation
8-hr TWA

Central

4.6E03

6.7E03

7.7E04

3.3E-07









Worker

Tendency









Moderate





Incorporation
into paints and
coatings - 1-
part coatings



High-End

7.3E02

1.6E03

1.9E04

1.4E-06







ONU"

Inhalation
8-hr TWA

Central
Tendency

4.6E03

6.7E03

7.7E04

3.3E-07

Slight







High-End

7.3E02

1.6E03

1.9E04

1.4E-06









Central

4.3E00

6.3E00

7.6E01

3.5E-04





Flame



Worker

Dermal

Tendency









Moderate

Processing/
processing -
incorporation
into

retardant in:







High-End

1.4E00

5.7E-01

4.0E00

8.6E-03



paint and

coating

manufacturing



Worker

Inhalation
8-hr TWA

Central
Tendency

7.9E02

6.5E03

7.9E04

3.2E-07

Moderate





High-End

1.9E02

1.6E03

1.9E04

1.4E-06



formulation,



Incorporation



Inhalation
8-hr TWA

Central

7.9E02

6.5E03

7.9E04

3.2E-07



mixture, or



into paints and

ONU"

Tendency









Slight

reaction product



coatings - 2-
part coatings



High-End

1.9E02

1.6E03

1.9E04

1.4E-06









Central

4.3E00

3.8E01

4.6E02

5.8E-05









Worker

Dermal

Tendency









Moderate







High-End

1.4E00

6.3E00

7.6E01

4.5E-04



Polymers used

Formulation
of TCEP into
2-part reactive
resin

Worker

Inhalation
8-hr TWA

Central
Tendency

1.0E04

6.7E03

8.1E04

3. IE—07

Moderate



in aerospace
equipment
and products



High-End

1.9E02

1.5E03

1.8E04

1.5E-06





ONU"

Inhalation
8-hr TWA

Central
Tendency

1.0E04

6.7E03

8.1E04

3. IE—07

Slight

Page 299 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

cou

OES

Population

Exposure
Route and
Duration

Exposure
Level

Estimates for No PPE

Overall
Confidence

in Risk
Estimates

Life Cycle

Stajjc/
Category

Subcategory

Acute Non-
cancer MOE

UFs = 30

Short-Term Non-
cancer MOE

UFs =30

Chronic Non-
cancer MOE

UFs = 30

Lifetime
Cancer
Risk





High-End

1.9E02

1.5E03

1.8E04

1.5E-06

Worker

Dermal

Central
Tendency

4.3E00

3.8E01

4.6E02

5.8E-05

Moderate

High-End

1.4E00

2.1E00

2.5E01

1.4E-03

Processing/
processing -
incorporation
into article

Aerospace
equipment
and products

Processing
into 2-part
resin article

Worker

Inhalation
8-hr TWA

Central
Tendency

2.2E04

9.0E03

3.8E04

6.6E-07

Moderate

High-End

4.2E03

1.8E03

6.3E03

4.1E-06

ONU"

Inhalation
8-hr TWA

Central
Tendency

2.2E04

9.0E03

3.8E04

6.6E-07

Slight

High-End

4.2E03

1.8E03

6.3E03

4.1E-06

Worker

Dermal

Central
Tendency

1.1E01

4.3E00

1.6E01

1.7E-03

Moderate

High-End

3.6E00

1.4E00

1.5E00

2.3E-02

Processing/
recycling

Recycling

Processing -
recycling e-
waste

Worker

Inhalation
8-hr TWA

Central
Tendency

7.6E08

3.0E08

3.2E08

8.4E-11

Moderate

High-End

7.8E04

3.1E04

3.3E04

1.0E-06

ONU

Inhalation
8-hr TWA

Central
Tendency

7.6E08

3.0E08

3.2E08

8.4E-11

Moderate

High-End

4.0E05

1.6E05

1.7E05

2.0E-07

Worker

Dermal

Central
Tendency

5.2E05

2.0E05

2.2E05

1.2E-07

Moderate

High-End

2.2E05

8.5E4

9.1E04

3.8E-07

Commercial
use/paints and
coatings

Paints and
coatings

Commercial
use - paints &
coatings -
spray (1-part
coatings, 1-
day

application)

Worker

Inhalation
8-hr TWA

Central
Tendency

4.5E02

N/A

N/A

N/A

Moderate

High-End

6.9E01

N/A

N/A

N/A

ONU"

Inhalation
8-hr TWA

Central
Tendency

4.5E02

N/A

N/A

N/A

Slight

High-End

6.9E01

N/A

N/A

N/A

Worker

Dermal

Central
Tendency

3.2E01

N/A

N/A

N/A

Moderate

High-End

5.9E00

N/A

N/A

N/A

Commercial
use - paints &
coatings -

Worker

Inhalation
8-hr TWA

Central
Tendency

4.5E02

1.9E03

N/A

N/A

Moderate

High-End

6.9E01

3.0E02

N/A

N/A

Page 300 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

cou





Exposure
Route and
Duration



Estimates for No PPE

Overall

Life Cycle
Stajjc/

Subcategory

OES

Population

Exposure
Level

Acute Non-
cancer MOE

Short-Term Non-
cancer MOE

Chronic Non-
cancer MOE

Lifetime
Cancer

Confidence
in Risk

Category









UFs = 30

UFs =30

UFs = 30

Risk

Estimates





spray (1-part
coatings, 2-

ONU"

Inhalation
8-hr TWA

Central
Tendency

4.5E02

1.9E03

N/A

N/A

Slight





day



High-End

6.9E01

3.0E02

N/A

N/A





application)

Worker

Dermal

Central
Tendency

3.2E01

1.4E02

N/A

N/A

Moderate











High-End

5.9E00

2.6E01

N/A

N/A











Inhalation
8-hr TWA

Central

4.5E02

1.8E02

1.9E02

1.4E-04







Commercial
use - paints &
coatings -
spray (1-part

Worker

Tendency









Moderate







High-End

6.9E01

2.7E01

2.9E01

1.2E-03







ONU"

Inhalation
8-hr TWA

Central
Tendency

4.5E02

1.8E02

1.9E02

1.4E-04

Slight





coatings, 250-



High-End

6.9E01

2.7E01

2.9E01

1.2E-03





day

application)

Worker

Dermal

Central
Tendency

3.2E01

1.3E01

1.3E01

2.0E-03

Moderate

Commercial

Paints and







High-End

5.9E00

2.3E00

2.5E00

1.4E-02



use/paints and

coatings





Inhalation
8-hr TWA

Central

9.0E01

N/A

N/A

N/A



coatings



Commercial
use - paints &
coatings -
spray (2-part

Worker

Tendency









Moderate







High-End

1.4E01

N/A

N/A

N/A







ONU"

Inhalation
8-hr TWA

Central
Tendency

9.0E01

N/A

N/A

N/A

Slight





coatings, 1-



High-End

1.4E01

N/A

N/A

N/A





day

application)

Worker

Dermal

Central
Tendency

6.4E00

N/A

N/A

N/A

Moderate











High-End

1.2E00

N/A

N/A

N/A







Commercial
use - paints &
coatings -
spray (2-part

Worker

Inhalation
8-hr TWA

Central
Tendency

9.0E01

3.9E02

N/A

N/A

Moderate







High-End

1.4E01

5.9E01

N/A

N/A







ONU"

Inhalation
8-hr TWA

Central
Tendency

9.0E01

3.9E02

N/A

N/A

Slight





coatings, 2-



High-End

1.4E01

5.9E01

N/A

N/A





day

application)

Worker

Dermal

Central
Tendency

6.4E00

2.8E01

N/A

N/A

Moderate











High-End

1.2E00

5.1E00

N/A

N/A



Page 301 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

cou





Exposure
Route and
Duration



Estimates for No PPE

Overall

Life Cycle
Stajjc/

Subcategory

OES

Population

Exposure
Level

Acute Non-
cancer MOE

Short-Term Non-
cancer MOE

Chronic Non-
cancer MOE

Lifetime
Cancer

Confidence
in Risk

Category









UFs = 30

UFs = 30

UFs = 30

Risk

Estimates









Inhalation
8-hr TWA

Central

9.0E01

3.8E01

3.8E01

7.1E-04







Commercial
use - paints &
coatings -
spray (2-part

Worker

Tendency









Moderate







High-End

1.4E01

5.4E00

5.8E00

6.0E-03







ONU"

Inhalation
8-hr TWA

Central
Tendency

9.0E01

3.8E01

3.8E01

7.1E-04

Slight





coatings, 250-



High-End

1.4E01

5.4E00

5.8E00

6.0E-03





day

application)

Worker

Dermal

Central
Tendency

6.4E00

2.5E00

2.7E00

9.9E-03

Moderate











High-End

1.2E00

4.6E-01

5.0E-01

6.9E-02











Inhalation
8-hr TWA

Central

5.8E06

2.3E06

2.5E06

1.1E-08









Worker

Tendency









Slight









High-End

5.8E06

2.3E06

2.5E06

1.1E-08

Industrial
Use/Other Use

Aerospace
equipment

Installation of
articles

ONU"

Inhalation
8-hr TWA

Central
Tendency

5.8E06

2.3E06

2.5E06

1.1E-08

Slight

products



High-End

5.8E06

2.3E06

2.5E06

1.1E-08













Central

N/A

N/A

N/A

N/A









Worker

Dermal

Tendency









N/A











High-End

N/A

N/A

N/A

N/A











Inhalation
8-hr TWA

Central

5.8E06

2.3E06

2.5E06

1.1E-08









Worker

Tendency









Slight





Use and/or
maintenance
of aerospace
equipment
and products



High-End

5.8E06

2.3E06

2.5E06

1.1E-08



Commercial
Use/Other Use

Aerospace
equipment

ONU"

Inhalation
8-hr TWA

Central
Tendency

5.8E06

2.3E06

2.5E06

1.1E-08

Slight

products



High-End

5.8E06

2.3E06

2.5E06

1.1E-08





Worker

Dermal

Central
Tendency

N/A

N/A

N/A

N/A

N/A











High-End

N/A

N/A

N/A

N/A











Inhalation
8-hr TWA

Central

1.0E05

5.1E04

5.5E04

4.0E-07









Worker

Tendency









Moderate

Commercial

Laboratory

Laboratory



High-End

6.5E04

3.2E04

3.5E04

6.8E-07



Use/ Other Use

chemicals

chemicals



Inhalation
8-hr TWA

Central

1.0E05

5.1E04

5.5E04

4.0E-07









ONU"

Tendency









Slight









High-End

6.5E04

3.2E04

3.5E04

6.8E-07

Page 302 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

cou

OES

Population

Exposure
Route and
Duration

Exposure
Level

Estimates for No PPE

Overall
Confidence

in Risk
Estimates

Life Cycle

Stajjc/
Category

Subcategory

Acute Non-
cancer MOE

UFs = 30

Short-Term Non-
cancer MOE

UFs =30

Chronic Non-
cancer MOE

UFs = 30

Lifetime
Cancer
Risk





Worker

Dermal

Central
Tendency

4.3E00

1.7E00

2.7E00

9.7E-03

Moderate

High-End

1.4E00

5.7E-01

7.6E-01

4.5E-02

Disposal/
Disposal

Disposal

Disposal

Evaluated as part of each OES as opposed to a standalone OES

8052

Page 303 of 572

-------
8053

8054

8055

8056

8057

8058

8059

8060

8061

8062

8063

8064

8065

8066

8067

8068

8069

8070

8071

8072

8073

8074

8075

8076

8077

8078

8079

8080

8081

8082

8083

8084

8085

8086

8087

8088

8089

8090

8091

8092

8093

8094

8095

8096

8097

8098

8099

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

5.3.2.1.2 COUs/OESs without Quantitative Risk Estimates
Distribution in Commerce

Distribution in commerce includes transporting TCEP or TCEP-containing products between work sites
or to final use sites as well as loading and unloading from transport vehicles. Individuals in occupations
that transport TCEP-containing products (e.g., truck drivers) or workers who load and unload transport
trucks may encounter TCEP or TCEP-containing products.

Because TCEP production volumes have declined, and no companies reported manufacture or import of
TCEP on the 2020 CDR, this decline would logically lead to decreased distribution into commerce.
Therefore, exposure and risk would also likely have declined with time. Exposure is possible from
ongoing manufacturing, processing, industrial, and commercial uses, and EPA estimated exposure and
risk to workers from relevant activities (e.g., loading articles), where relevant, as part of these other
COUs (e.g., during manufacturing/repackaging). These exposures were generally combined with
exposures from other activities, and EPA assessed risks based on these combined exposures as part of
these other COUs. Due to limited data for the full set of possible exposures, EPA's confidence in this
exposure is indeterminate. Therefore, EPA cannot characterize risk to workers for this COU separately
from the risks already estimated for other relevant COUs.

Commercial Uses that Have Been Phased Out

EPA determined that some commercial use COUs for TCEP are not ongoing uses. These COUs are

•	Commercial use - furnishing, cleaning, treatment/care products - fabric and textile products;

•	Commercial use - furnishing, cleaning, treatment/care products - foam seating and bedding
products;

•	Commercial use - construction, paint, electrical, and metal products - building/construction
materials - insulation; and

•	Commercial use - construction, paint, electrical, and metal products - building/construction
materials - wood and engineered wood products - wood resin composites.

TCEP was used for these purposes in the past, but the COUs were phased out beginning in the late
1980s or early 1990s and replaced by other flame retardants or flame-retardant formulations. EPA did
not locate data to estimate (1) the amount of TCEP used in these products, (2) the amounts of these
products that have already reached the end of their service life, or (3) the amounts that have already been
disposed. Based on the years that the phase-out occurred, many of these products are likely to no longer
be in use because the end of their service life was already reached (e.g., commercial roofing has an
estimated lifespan of 17 to 20 years). EPA assumes that any of these products still used commercially
represent a fraction of the overall amount of TCEP previously used for these purposes.

For these reasons, EPA has not quantified these risks, and EPA's confidence in this exposure is
indeterminate. Therefore, EPA cannot characterize risk for these COUs, but included a qualitative
description of what is known from the reasonably available information.

Disposal

Waste handling, disposal, and/or treatment includes waste disposal (landfilling or incineration) as well
as water (e.g., releases to wastewater treatment and POTWs) and air releases (e.g., fugitive and stack air
emissions). Workers engaged in these activities at the facilities where TCEP is processed and used, as
well as workers at off-site waste treatment and disposal facilities (e.g., landfills, incinerators, POTWs)
could be exposed to TCEP.

Page 304 of 572

-------
8100

8101

8102

8103

8104

8105

8106

8107

8108

8109

8110

8111

8112

8113

8114

8115

8116

8117

8118

8119

8120

8121

8122

8123

8124

8125

8126

8127

8128

8129

8130

8131

8132

8133

8134

8135

8136

8137

8138

8139

8140

8141

8142

8143

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

EPA estimated releases to landfills for the following two COU/OES combinations:

•	Processing - incorporation into formulation, mixture, or reaction product - paint/coating
manufacture - 1-part coating OES; and

•	Processing - incorporation into articles - aerospace equipment and products - processing in two-
part resin article OES.

EPA estimated releases to incinerators for the following two COU/OES combinations:

•	Processing - incorporation into formulation, mixture, or reaction product - paint/coating
manufacture - 2-part coating OES; and

•	Processing - incorporation into formulation, mixture, or reaction product - polymers in
aerospace equipment and products - formulation of reactive resins OES.

Both releases to landfills and incinerators rely on inputs provided by ESDs or GSs. However, the ESDs
and GSs do not specify the proportion of the throughput that goes to either of these two disposal
practices. Therefore, EPA was unable to further quantify environmental releases related to these two
disposal processes.

For three of the COUs/OESs listed above, EPA was able to perform quantitative risk characterization
that included releases to onsite wastewater treatment or discharge to POTWs, where applicable (see
Table 3-2). Any worker exposures associated with on-site waste treatment were combined with other
exposures as relevant for the above COUs.

Waste treatment or disposal is expected to be negligible for industrial and commercial uses related to
installing articles for aerospace applications. For the COUs of manufacturing/repackaging, commercial
use of paints and coatings, commercial use of laboratory chemicals, and disposal to landfills or
incinerators are not expected but EPA estimated surface water releases that could include release to
wastewater treatment or POTWs.

For the commercial uses that have been phased out, any currently used products that contain TCEP are
expected to be disposed in landfills but will represent just a fraction of previous amounts from when
TCEP was used more widely. Data are lacking with which to estimate exposure and risk from disposal
or waste treatment activities for these COUs and EPA has not quantified such risks. For e-waste
recycling, there is also too little information to estimate exposure from disposal and only a small portion
of e-waste is expected to contain TCEP. Therefore, EPA's confidence in these exposures is
indeterminate and cannot characterize risk for the disposal or waste treatment activities for these COUs.

5.3.2.2 Summary of Risk Estimates for Consumers

5.3.2.2.1 COUs with Quantitative Risk Estimates

Table 5-58 summarizes the dermal, inhalation, and ingestion MOEs used to characterize non-cancer risk
for acute, short term, and chronic exposure and presents these values for all life stages for each COU.
Table 5-59 summarizes the dermal, inhalation, and ingestion lifetime cancer risk estimates for each
consumer COU. Risk estimates in Table 5-58 and Table 5-59 are only presented for COUs, routes, and
age groups that are below the non-cancer risk benchmarks or above the lifetime cancer benchmarks. For
cancer, EPA uses a range of cancer benchmarks from 1 in 10,000 to 1 in 1,000,000 to consider and
characterize lifetime cancer risks from consumer exposure. Table 5-59 presents the risk estimates that
were above the lifetime cancer benchmark of 1 in 1,000,000.

Page 305 of 572

-------
8144

8145

8146

8147

8148

8149

8150

8151

8152

8153

8154

8155

8156

8157

8158

8159

8160

8161

8162

8163

8164

8165

8166

8167

8168

8169

8170

8171

8172

8173

8174

8175

8176

8177

8178

8179

8180

8181

8182

8183

8184

8185

8186

8187

8188

8189

8190

8191

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Although CEM 3.0 provides inhalation exposure doses for each age group, inhalation exposure risk
estimates were calculated for the adult exposure scenario. Inhalation risk estimates for other lifestages
are presented in Appendix I. These adjusted inhalation exposure doses are estimated using breathing rate
and body weight considerations for each age group. Body weight- and inhalation rate-adjusted inhalation
risk estimates for younger life stages should be interpreted with caution. Despite accounting for
breathing rate and body weight, adjusted inhalation exposures for younger age groups may be inaccurate
because there are other considerations (e.g., elimination kinetics) that may differ among age groups
(	2012a). Information on the inputs used for consumer modeling using CEM 3.0 are presented

in Section 5.1.2 and Appendix I.

Acute and Chronic Risks

Children and infants have acute oral MOEs less than the benchmark of 30 for foam toy blocks, roofing
insulation, and wood flooring. Infants have acute oral MOEs less than the benchmark of 30 for all of the
COUs except acoustic ceilings. Chronic oral MOEs for children and infants are below the benchmark of
30 for fabric and textiles, foam seating and bedding products, wood flooring and wooden TV stands.
Infants and children have a greater susceptibility to TCEP exposure due to mouthing behaviors
associated with toys (e.g., outdoor play structures, foam blocks). As discussed in Section 5.1.2.2.4, EPA
selected a high mouthing parameter (50 cm2) for the COUs that were designed for children. For other
products that had the potential for mouthing, EPA selected medium mouthing parameters (10 cm2).
Mouthing duration had a pronounced impact on the oral exposures for children and infants (see
Appendix I).

Section 5.1.2.2.3 describes the parameters selection and assumptions considered for the dermal exposure
assessment. Acute and chronic dermal MOEs for all lifestages are below the benchmark of 30 for wood
flooring. Chronic dermal MOEs for children and infants are below the benchmark of 30 for wooden TV
stands. Sensitivity analyses indicated that the initial SVOC concentration in the article (a product of the
article density and the weight fraction) is a driver of dermal exposures. The consumer modeling suggests
direct contact with wooden articles (e.g., wood flooring, wooden TV stands) results in greater exposure
than dermal doses mediated from dust generated from consumer articles.

Chronic inhalation MOEs for acoustic ceilings, wood flooring, and insulation are below the benchmark
of 30. Acute inhalation MOEs for textiles in outdoor play structures, acoustic ceilings, wood flooring,
wooden TV stands, and insulation are below the benchmark of 30. Sensitivity analyses indicated that the
initial SVOC concentration in the article (a product of the article density and the weight fraction) is a
driver of inhalation exposures for insulation. For more information on the inhalation exposure estimates,
see Section 5.1.2.2.2.

Lifetime Cancer Risks

Inhalation from insulation presents the highest lifetime cancer risk (4,50/ 10 2), followed by inhalation
exposure from wood floorings (3,92/ 10 2) (Table 5-59). In comparing inhalation risks from wood floors
to a wooden TV stand, wood flooring has a larger cancer inhalation risk estimate by two orders of
magnitude. This suggests that the space (surface area) a wood article occupies in the home environment
has a relationship to the magnitude of inhalation risk. Lifetime cancers risks for wood flooring is
dominated by inhalation route whereas lifetime cancer risks for wooden TV stand is dominated by the
ingestion route. This may be explained by the relatively large surface area for wood flooring versus
wooden TV stands. Wood articles (e.g., wood flooring, wooden TV stands) have a higher lifetime cancer
risk for oral exposures (6,05/10 4 and 4,93/10 4) compared to dermal exposure (1.20/ 10 4 and
2,52/ 10 5), Carpet and foam products (e.g., mattresses, foam furniture, automobile foams) are

Page 306 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

8192	dominated by oral cancer risks relative to other routes. The contribution of mouthing exposure from

8193	these articles at younger lifestages may be contributing to the overall cancer risk.

Page 307 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

8194 Table 5-58. Acute and Chronic Non-cancer Consumer Risk Summary

COIJ

Consumer
Use Scenario

Exposure
Route

Age

Non-cancer MOEs"

Overall Confidence
Non-cancer MOEs

Life Cycle
Stage/Categorv

Subcategory

Group
(years)

Acute MOE

UFs = 30

Chronic MOE

UFs = 30





Carpet back
coating

Oral

Child: 3-5

51

15







Oral

Infant: 1-2

42

12

Moderate



Fabric and textile
products

Oral

Infant: <1

18

5





Textile for
children's

Oral

Child: 3-5

40

15





Oral

Infant: 1-2

35

12

Moderate





outdoor play
structures

Oral

Infant: <1

17

5





Inhalation

Adult: >21

9

45



Consumer use/





Oral

Child: 3-5

52

15



furnishing,
cleaning,
treatment, and



Foam auto

Oral

Infant: 1-2

43

12

Moderate





Oral

Infant: <1

18

5



care products



Foam living
room

Oral

Child: 3-5

52

15





Foam seating and
bedding products

Oral

Infant: 1-2

43

12

Slight



Oral

Infant: <1

18

5





Mattress

Oral

Infant: 1-2

35

10

Slight





Oral

Infant: <1

18

5





Foam-other (toy
block)

Oral

Child: 3-5

11

3







Oral

Infant: 1-2

9

2

Slight





Oral

Infant: <1

4

1



Consumer use/

Building/
construction

Roofing
insulation

Inhalation

Adult: >21

0.4

2



construction,
paints, electrical,
and metal

Oral

Child: 3-5

7

27

Slight

materials -
insulation

Oral

Infant: 1-2

8

30

products

Acoustic ceiling

Inhalation

Adult: >21

2

24



Page 308 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

cou

Consumer
Use Scenario

Exposure
Route

Age
Group
(years)

Non-cancer MOEs"

Overall Confidence
Non-cancer MOEs

Life Cycle
Stagc/Categorv

Subcategory

Acute MOE

UFs = 30

Chronic MOE

UFs = 30

Consumer use/
construction,
paints, electrical,
and metal
products

Building/
construction
materials - wood
and engineered
wood products -
wood resin
composites

Wood flooring

Dermal

Adult: >21

27

12

Slight

Dermal

Youth: 16-20

29

12

Dermal

Youth: 11-15

27

11

Dermal

Child: 6-10

21

9

Dermal

Child: 3-5

9

7

Dermal

Infant: 1-2

8

6

Dermal

Infant: <1

7

5

Inhalation

Adult: >21

0.4

2

Oral

Child: 3-5

4

13

Oral

Infant: 1-2

5

11

Oral

Infant: <1

5

5

Wooden TV
stand

Dermal

Child: 6-10

95

28

Moderate

Dermal

Child: 3-5

74

22

Dermal

Infant: 1-2

64

19

Dermal

Infant: <1

55

16

Inhalation

Adult: >21

7

337

Oral

Child: 3-5

49

15

Oral

Infant: 1-2

40

12

Oral

Infant: <1

18

5

8195

8196

8197

8198

Page 309 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

8199 Table 5-59. Lifetime Cancer Consumer Risk Summary

cou

Consumer Use Scenario

Exposure Route

Lifetime Cancer Risk
Estimates"

Overall Confidence in
Cancer Risk Estimate

Life Cycle
Stage/Categorv

Subcategory

Consumer use/
furnishing, cleaning,
treatment, and care
products

Fabric and textile
products

Carpet back coating

Oral

4.94E-04

Moderate

Inhalation

1.48E-04

Dermal

3.82E-07

Foam seating and
bedding products

Foam automobile

Oral

4.93E-04

Moderate

Inhalation

2.51E-08

Dermal

1.87E-06

Foam living room

Oral

4.93E-04

Moderate

Inhalation

4.51E-08

Dermal

4.17E-06

Mattress

Oral

4.23E-04

Slight

Inhalation

2.15E-06

Dermal

2.04E-06

Consumer use/
construction, paints,
electrical, and metal
products

Building/construction
materials - insulation

Roofing insulation

Oral

4.21E-04

Slight

Inhalation

4.50E-02

Dermal

8.11E-06

Acoustic ceiling

Oral

1.43E-05

Slight

Inhalation

3.63E-03

Dermal

2.76E-07

Building/construction
materials - wood and
engineered wood
products - wood resin
composites

Wood flooring

Oral

6.05E-04

Slight

Inhalation

3.92E-02

Dermal

1.20E-04

Wooden TV stand

Oral

4.93E-04

Moderate

Inhalation

2.56E-04

Dermal

2.52E-05

a Risk estimates are only presented for COUs, routes, and age groups that are below the non-cancer risk benchmarks or above the lifetime cancer benchmarks.

8200

Page 310 of 572

-------
8201

8202

8203

8204

8205

8206

8207

8208

8209

8210

8211

8212

8213

8214

8215

8216

8217

8218

8219

8220

8221

8222

8223

8224

8225

8226

8227

8228

8229

8230

8231

8232

8233

8234

8235

8236

8237

8238

8239

8240

8241

8242

8243

8244

8245

8246

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

5.3.2.2.2 COUs without Quantitative Risk Estimates
Paints and Coatings

Domestic retail production and manufacturing of paints and coatings containing TCEP has ceased, and
consumers can no longer purchase these products from store shelves in the United States. There remains
some possibility of exposure by consumers to TCEP from previous purchases, however. For example, in
the early 2000s, Ineerowski et al. (2001) detected TCEP in 85 percent of 983 household products in
Germany and reported TCEP in wood preservation coatings at 1.0 percent. Also, Haumann and
Thumulla (2002) detected TCEP in paints at a maximum of 840 mg/kg (0.084 percent) in Germany prior
to 2002 (TERA. 2013).

Exposure may occur from offgassing of old paint cannisters stored in homes or if these stored cannisters
are subsequently used to paint walls or other surfaces. Exposure is also possible from contact with and
off gassing from surfaces to which a paint or coating containing TCEP was previously applied, such as
in an older building. This dried paint scenario is similar to the acoustic ceilings/drywall scenario
assessed for the building/construction materials COU.

Despite the lack of a domestic market for consumer paints/coatings, it is possible that consumers could
buy commercial use products from the internet. These paints and coatings available for commercial use
have maximum weight fractions (25 percent) that is almost 4 times higher than weight fractions
available for consumer articles (6.8 percent).

Due to limited information regarding the use of paints and coatings and the uncertainties surrounding the
weight fraction, activity, and use patterns, and duration of use for consumers, EPA did not quantitatively
assess the consumer use of paints and coatings and has not made a conclusion regarding risk from this
COU. EPA's confidence in this exposure is indeterminate, and the Agency cannot characterize risk.

Disposal of Wastewater, Liquid Wastes, and Solid Wastes

Consumers may be exposed to articles containing TCEP during disposal and the handling of waste. The
removal of articles in DIY scenarios may lead to direct contact with articles and the dust generated from
the articles. Due to the difficulties in quantifying consumer disposal of products containing TCEP, it was
not quantitatively assessed for this risk evaluation. EPA's confidence in this exposure is indeterminate.

5.3.2.3 Summary of Risk Estimates for the General Population

5.3.2.3.1 COUs with Quantitative Risk Estimates

EPA quantitatively assessed human exposures to TCEP concentrations via oral ingestion of drinking
water, soil, and fish, dermal exposures to soil and surface water, and inhalation of ambient air. EPA
assessed risk associated with each of these exposure scenarios by comparing doses to acute, short-term,
and chronic human equivalent concentrations and doses. Furthermore, EPA assessed the lifetime cancer
risk from TCEP exposure via these routes. As noted previously, EPA uses a range of cancer benchmarks
from 1 in 10,000 to 1 in 1,000,000 to characterize lifetime cancer risks for the general population.

Table 5-60 and Table 5-61 summarize the MOEs used to characterize acute non-cancer risks for oral
exposures for the applicable COUs. Table 5-62 and Table 5-63 summarizes the chronic non-cancer
MOE estimates for the applicable COUs. Table 5-64 summarizes the lifetime cancer oral risk for the
applicable COUs. Oral ingestion non-cancer MOEs and cancer risks are presented for drinking water,
diluted drinking water, landfill leachate to groundwater and subsequent migration to drinking water,
incidental ingestion during swimming, fish ingestion, and soil ingestion for children playing with soil.

Page 311 of 572

-------
8247

8248

8249

8250

8251

8252

8253

8254

8255

8256

8257

8258

8259

8260

8261

8262

8263

8264

8265

8266

8267

8268

8269

8270

8271

8272

8273

8274

8275

8276

8277

8278

8279

8280

8281

8282

8283

8284

8285

8286

8287

8288

8289

8290

8291

8292

8293

8294

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table 5-65 summarizes the acute and chronic non-cancer dermal MOEs for incidental dermal exposures
during swimming and dermal ingestion of soils for children playing with soil associated with applicable
COUs.

Table 5-66 presents the general population chronic inhalation MOEs used to characterize risk for the
applicable COUs. Table 5-67 presents the general population lifetime cancer inhalation risk estimates
for the applicable COUs. Inhalation MOEs and risk estimates are provided for various distances from a
hypothetical facility for two meteorology conditions (Sioux Falls, South Dakota, for central tendency
meteorology; and Lake Charles, Louisiana, for higher-end meteorology).

Ingestion

Drinking Water and Incidental Surface Water Ingestion: Table 5-60 summarizes the acute drinking
water risk estimates for all COUs and life stages. The non-cancer MOE values for the acute drinking
water ingestion exposure by infants for four scenarios—Incorporation into paints and coatings (1-part
coatings), Incorporation into paints and coatings (2-part coatings), Use in paints and coatings at job sites,
and Formulation of TCEP containing reactive resin—are less than the benchmark MOE of 30. When
factoring in dilution, none of the life stages have acute drinking water MOE of less than the benchmark
for any scenario.

Because TCEP is recalcitrant to drinking water treatment removal processes, a 0 percent drinking water
treatment removal efficiency was used to calculate the oral drinking water exposure doses. The non-
diluted acute risk estimates assume the general population was drinking water at the site of the facility
outfall. To approximate a more typical drinking water concentration, distances between drinking water
intake locations and facilities based on SIC codes were used to calculate a dilution factor to estimate a
diluted drinking water concentration (See Section 5.1.3.4.1). All non-cancer MOEs from acute
incidental ingestion via swimming were larger than the benchmark MOE of 30 for adults, youth, and
children (Appendix H General Population).

None of the chronic MOEs from drinking water, diluted drinking water, incidental ingestion via
swimming, and drinking water contamination from landfill leachate were lower than the benchmark
MOE of 30. Drinking water MOEs are presented for both diluted and non-diluted surface water
concentrations. The diluted drinking water MOEs represent typical case scenarios, whereas MOEs based
on the non-diluted concentrations represent worst-case scenarios.

The DRAS Model described in Section 3.3.3.7 estimated TCEP groundwater concentrations from
landfill leachate. Only two industrial and commercial release scenarios had anticipated releases to
landfill (Incorporation into paints and coatings - 1-part coatings and processing into 2-part resin article).
The DRAS Model estimated groundwater concentrations by using production volume (2,500 lb) as the
input rather than the release estimate generated by the two industrial uses (21.5 kg/site-year for 1-part
coatings, and 42.9 kg/site-year for 2-part resin articles). Nevertheless, estimates via the full production
volume did not result in chronic oral MOEs below 30 for drinking water.

Lifetime (from birth) oral ingestion cancer risk greater than 1 in 1,000,000 is associated with releases
from four OESs: Incorporation into paints and coatings - 1-part coatings; Incorporation into paints and
coatings - resins/solvent-borne; Use in paints and coatings at job sites; and Processing into 2-part resin
article. There was also oral ingestion cancer risk greater than 1 in 1,000,000 for the adult lifetime for the
same scenarios, except for the use in paints and coatings at job sites. Under diluted drinking water
conditions, no lifetime risks from birth or for the adult lifetimes exceeded 1 in 1,000,000.

Page 312 of 572

-------
8295

8296

8297

8298

8299

8300

8301

8302

8303

8304

8305

8306

8307

8308

8309

8310

8311

8312

8313

8314

8315

8316

8317

8318

8319

8320

8321

8322

8323

8324

8325

8326

8327

8328

8329

8330

8331

8332

8333

8334

8335

8336

8337

8338

8339

8340

8341

8342

8343

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Fish Ingestion: For the adult general population, acute exposure estimates via fish ingestion using a
BAF of 2,198 L/kg showed MOEs less than 30 for all OESs except laboratory use of chemicals (Table
5-32). No OESs had an acute risk estimate less than 30 based on a BAF of 109 L/kg. For the adult
subsistence fisher, EPA only had one fish IR that resulted in the same doses for both acute and chronic
exposure. EPA estimated non-cancer MOEs by comparing that same dose with both the acute and
chronic HEDs. Exposure estimates based on a BAF of 2,198 L/kg showed MOEs less than the acute
benchmark for all OESs except laboratory use of chemicals. Using a BAF of 109, Laboratory use of
chemicals and import and repackaging showed MOEs less than the acute benchmark. For tribes, the
same approach was to estimate acute and chronic risks as the subsistence fisher. A BAF of 2,198
showed MOEs less than the acute benchmark for all OESs for both the current and heritage IR. A BAF
of 109 showed MOEs less than the acute benchmark for all COUs except Import and repackaging and
Laboratory use of chemicals based on the current mean IR (for the Suquamish Tribe). The BAF of 109
also had MOEs less than the acute benchmark for all COUs except Laboratory use of chemicals based
on the heritage IR (for the Kootenai Tribe).

Chronic exposure for the general population resulted in MOEs less than the chronic benchmark of 30 for
all OESs except Laboratory use of chemicals for both fish IRs and a BAF of 2,198/kg (Table 5-62). The
table presents adult general population risk estimates based on only the 90th percentile IR even though
two values were used, as discussed in Section 5.1.3.4.2. The MOEs based on the central tendency IR
will be 4.4 times higher. When estimating exposure and risks based on a BAF of 109 L/kg, there are
some differences in risks between the two IRs. The 90th percentile IR results in risks for three OESs:
Incorporation into paints and coatings - 1-part coating; Incorporation into paints and coatings - 2-part
reactive coatings; and Formulation of TCEP containing reactive resin. The central tendency IR did not
result in any OESs with risk estimates below their chronic benchmark.

Chronic exposure for the subsistence fisher and tribes resulted in MOEs less than 30 for all OESs based
on a BAF of 2,198 L/kg and all IRs. A BAF of 109 L/kg showed risk estimates less than the chronic
benchmark for all OESs except Laboratory use of chemicals.

Exposure estimates were not calculated for younger age groups. For younger age groups, acute and
chronic MOEs less than benchmark values are reasonably expected because these age groups generally
have higher fish ingestion rates per kilogram body weight (TableApx H-2). For tribes, adults were
reported to have the highest IR per kilogram of body weight (Section 2195.1.3.4.4).

For the adult general population, subsistence fisher, and tribe, cancer risk estimates are above 1 in
1,000,000 for all OESs and for both BAF values, as well as current and heritage IRs for tribes. Table
5-65 shows the lifetime cancer risk estimates for fish ingestion. Cancer risk estimates were not
calculated for fish ingestion among younger age groups. Similar to non-cancer risk, cancer risks for
younger age groups are reasonably expected to be higher than older groups because of the higher fish
ingestion rate per kilogram of body weight or because adults have the highest IR by body weight.
(TableApx H-2).

Soil Ingestion: Chronic oral non-cancer MOEs from soil were estimated for children 3 to 6 years of age
based on soil concentrations that were calculated from air deposition for various distances from a
hypothetical facility releasing TCEP (see Section 3.3.3.2). Oral doses were calculated for two exposure
scenarios: (1) a child conducting activities with soil, and (2) a child playing in mud (see Section
5.1.3.4.4). No MOEs were less than the benchmark of 30 for the children's soil ingestion scenario for
any of the COUs. In addition, there was no lifetime cancer risk for soil ingestion for any of the COUs.

Page 313 of 572

-------
8344

8345

8346

8347

8348

8349

8350

8351

8352

8353

8354

8355

8356

8357

8358

8359

8360

8361

8362

8363

8364

8365

8366

8367

8368

8369

8370

8371

8372

8373

8374

8375

8376

8377

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Dermal

Incidental Dermal from Swimming: Non-cancer MOEs were not lower than benchmark values for the
acute and chronic incidental dermal exposures swimming scenario for any of the COUs.

Children's Dermal Exposure from Playing in Soil: Dermal exposure estimates from soil were estimated
for children 3 to 6 years of age because these ages are expected to play in mud and perform activities
with soil. Soil concentrations were calculated via annual air deposition fluxes for various distances from
a hypothetical facility releasing TCEP (see Section 3.3.3.2). Dermal exposure doses were also calculated
for a child conducting activities with soil and a child playing in mud (see Section 5.1.3.3.2). No non-
cancer MOEs for chronic exposures were less than the benchmark MOE of 30 at 100 or 1000 m for
either scenario of children playing in mud or children conducting activities with soil.

Many uncertainties are associated with the dermal exposure estimate used for the chronic dermal MOE
that was less than the benchmark, including the lack of release information, site information, and
reasonableness of the exposure scenario. The source of the exposure is a hypothetical facility that
releases TCEP to the air for 2 days. Because no site information was available, EPA's release
assessment estimated a 50th percentile of 27 sites to a 95th percentile of 203 sites per the OES for the
commercial use of paints and coatings. To observe an MOE less than the benchmark, a child would have
to be playing in mud at 100 m from the hypothetical facility. TCEP would deposit to the soil after
deposition from air releases. Section 3.3.3.2 describes how EPA calculates soil concentrations from
annual modeled air deposition. No U.S. studies recorded TCEP in soil. Modeled soil concentrations at
100 m (4.15 x 103 ng/g) were two orders of magnitude higher than the TCEP concentrations found in
Germany (23.5 ng/g) (Mihailovic and Fries.: ). The study from Germany also indicated increased
soil concentration of TCEP due to snow melt (see Section 3.3.3.1).

Inhalation

Table 5-65 shows the COUs where EPA found lifetime inhalation cancer risk estimates greater than 1 in
1,000,000 for the 2,500 lb production volume, high-end release estimate, suburban forest scenario and
when using both central-tendency and high-end meteorological data. EPA found inhalation cancer risks
greater than the benchmark for the 50th percentile air concentrations for the use of paints and coatings at
job sites at distances as far as 60 m from the site. EPA also found cancer risk above this benchmark for
the 95th percentile air concentrations for the use of paints and coatings out to 100 m from the job site.

Page 314 of 572

-------
8378

8379

8380

8381

8382

8383

8384

8385

8386

8387

8388

8389

8390

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

displays the chronic inhalation non-cancer risk estimates for the 2,500 lb production volume, high-end
release estimate, suburban forest scenario, high-end meteorological data at 10 m from the facility. No
non-cancer inhalation MOEs were less than the acute (total UF = 30) or chronic (total UF = 30)
benchmark MOEs for any COUs. The lowest MOE for the chronic exposure scenario was 498 (the use
of paints and coatings scenario, high meteorological station data, at 10 m, 95th percentile). The lowest
MOE for the acute exposure scenario was 295,000 for the processing into 2-part resin article, high
meteorological station data, at 10 m, 95th percentile scenario (not shown). Ambient air is a minor
environmental compartment as described in Section 2.2.

It is unlikely that individual residences will be within 10 m of the stack or fugitive air release from these
facilities. However, these estimates suggest that fence line communities living within 100 m downwind
of facilities that use TCEP in paints and coatings at job sites may be at an increased risk of developing
cancer over their lifetimes.

Page 315 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

8391 Table 5-60. General Population Acute Drinking Water (Oral Ingestion) Non-cancer Risk Summary















Acute Oral Non-cancer IMOEs









cou













LJFs = 30















OES

Drinking Water

Drinking Water (Diluted)

Lifccvclc/

Sub-category



Adult

Infant

Youth

Youth

Child

Toddler

Adult

Infant

Youth

Youth

Child

Toddler

Category



(>21 yr)

(21 yr)

(
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

8394 Table 5-61. Acute Fish Ingestion Non-cancer Risk Summary

cou

OES

Acute Oral Non-cancer MOEs

UFs = 30

Life Cycle/
Category

Subcategory



General
Population

Subsistence
Fishers

Tribes
(Current
IR)a

Tribes
(Heritage IR)fc

BAF
2,198

BAF
109

BAF
2,198

BAF
109

BAF 2,198

BAF
109

BAF
2,198

BAF
109

Manufacturing/
import

Import

Repackaging

1.80E01

3.63E02

2.80E00

5.66E01

1.85E00

3.73E01

3.21E-01

6.47E00

Processing/
processing -
incorporation into
formulation,
mixture, or
reaction product

Flame retardant in:
paint and coating
manufacturing

Incorporation into paints
and coatings - 1-part
coatings

4.07E00

8.20E01

6.33E-01

1.28E01

4.17E-01

8.42E00

7.25E-02

1.46E00

Incorporation into paints
and coatings - 2-part
reactive coatings

4.49E00

9.05E01

6.98E-01

1.41E01

7.99E-02

9.28E00

7.99E-02

1.61E00

Polymers used in
aerospace equipment
and products

Formulation of TCEP
containing reactive resin

3.21E00

6.47E01

4.99E-01

1.01E01

5.71E-02

6.63E00

5.71E-02

1.15E00

Commercial use

Laboratory
chemicals

Use of laboratory
chemicals

4.50E02

9.07E03

7.00E01

1.41E03

8.01E00

9.30E02

8.01E00

1.62E02

Paints and coatings

Use of paints and
coatings at job sites

7.66E00

1.54E02

1.19E00

2.40E01

1.36E-0
1

1.58E01

1.36E-01

2.75E00

" Current fish consumption rate at 216 g/day based on survey of Suquamish Indian Tribe in Washington (Section 5.1.3.4.4).
6 Heritage fish consumption rate at 1,646 g/day based on study of Kootenai Tribe in Idaho (Section 5.1.3.4.4).

8395

8396

Page 317 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

8397 Table 5-62. General Populai

ion Chronic Water and Soil Ingestion Non-cancer Risk Summary

cou

OES

Chronic Non-cancer Oral IMOEs
II Fs =30

Life Cvcle/
Category

Subcategory

Drinking

Water
(Diluted)

Drinking

Water

Drinking
Water (via
Leaching to
Groundwater)

Ambient
Water
(Incidental
Ingestion)

Soil Intake

(50th)
at 100 m

Soil Intake

(95th)
at 100 m

Soil Intake

(50th)
at 1,000 m

Soil Intake

(95th)
at 1,000 m

Manufacturing/
import

Import

Repackaging

1.64E08

1.05E05

N/A

2.11E05

2.20E10

5.15E09

1.73E12

4.03E11

Processing/
processing -
incorporation
into formulation,
mixture, or
reaction product

Flame retardant
in: paint and
coating
manufacturing

Incorporation into
paints and
coatings - 1 -part
coatings

4.40E07

23,728

2.12E06

4.89E04

7.02E08

1.64E08

7.95E10

1.86E10

Incorporation into
paints and
coatings - 2-part
reactive coatings

4.85E07

26,171

N/A

5.39E04

4.85E09

1.13E09

3.68E11

8.59E10

Polymers used in
aerospace
equipment and
products

Formulation of
TCEP containing
reactive resin

9.89E06

18,706

N/A

4.62E04

4.41E09

1.03E09

3.46E11

8.07E10

Processing/
processing -
incorporation
into article

Aerospace
equipment and
products

Processing into 2-
part resin article

N/A

N/A

2.12E06

N/A

5.15E08

1.20E08

5.05E10

1.18E10

Commercial use

Laboratory
chemicals

Use of laboratory
chemicals

4.10E09

2.60E06

N/A

5.30E06

4.60E08

1.07E08

4.20E10

9.81E09

Paints and
coatings

Use of paints and
coatings at job
sites

6.96E07

4.47E04

N/A

8.98E04

2.98E05

6.96E04

5.72E07

1.34E07

8398

8399

Page 318 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

8400 Table 5-63. Chronic Fish Ingestion Non-cancer Risk Summary

cou



Gen Pop

Subsistence Fishers''

Tribes (Current)'

Tribes (Heritage)''

Life Cvclc/

Subcategory

OES

BAF 2,198"

BAF 109"

BAF

BAF

BAF

BAF

BAF

BAF

Category



CT

HE

CTe

HE

2,198

109

2,198

109

2,198

109

Manufacturing/i

Import

Repackaging

2.29E01

5.20E00

4.61E02

1.05E02

8.09E-01

1.63E01

5.34E-01

1.08E01

9.26E-02

1.87E00

mport





























Incorporation

5.16E00

1.17E00

1.04E02

2.37E01

1.83E-01

3.68E00

1.20E-01

2.43E00

2.09E-02

4.22E-01



Flame
retardant in:
paint and
coating
manufacturing

into paints and





















Processing/

coatings - 1-part
coatings





















processing -

incorporation

into

formulation,
mixture, or
reaction product

Incorporation
into paints and
coatings - 2-part
reactive coatings

5.69E00

1.29E00

1.15E02

2.61E01

2.02E-01

4.06E00

2.31E-02

2.68E00

2.31E-02

4.65E-01

Polymers used

Formulation of

4.07E00

9.26E-01

8.21E01

1.87E01

1.44E-01

2.90E00

1.65E-02

1.91E00

1.65E-02

3.32E-01

in aerospace
equipment and
products

TCEP containing
reactive resin





















Processing/

Aerospace

Processing into

N/A

N/A

N/A

N/A

N/A

N/A

N/A

N/A

N/A

N/A

Processing -
incorporation
into article

equipment and
products

2-part resin
article























Laboratory

Use of laboratory

5.71E02

1.30E02

1.15E04

2.62E03

2.62E01

4.07E02

2.31E00

2.68E02

2.31E00

4.66E01



chemicals

chemicals





















Commercial use

Paints and
coatings

Use of paints and
coatings at job
sites

9.72E00

2.21E00

1.96E02

4.46E01

3.44E-01

6.93E00

3.94E-02

4.57E00

3.94E-02

7.94E-01

" GP exposure estimates based on general population fish ingestion rate of 22.2 g/day.
h SF exposure estimates based on subsistence fisher ingestion rate of 142.2 g/day.

0 Current fish consumption rate at 216 g/day based on survey of Suquamish Indian Tribe in Washington (Section 5.1.3.4.4).
^Heritage fish consumption rate at 1,646 g/day based on study of Kootenai Tribe in Idaho (Section 5.1.3.4.4).
e Exposure estimates based on a general population mean fish ingestion rate of 5.04 g/day.

8401

8402

Page 319 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

8403 Table 5-64. General Population Lifetime Cancer Oral Ingestion Risk Summary Tab

cou



Lifetime Cancer Oral Risk Estimates

Drinking Water

Drinking Water
(Diluted)

Life Cycle/Category

Subcategory

OES

Lifetime
from Birth

Adult
Lifetime

Lifetime
from Birth

Adult
Lifetime

Manufacturing/import

Import

Repackaging

6.91E-07

2.70E-07

4.43E-10

1.73E-10

Processing/processing - incorporation into
formulation, mixture, or reaction product

Flame retardant in: paint
and coating manufacturing

Incorporation into
paints and coatings -
1-part coatings

3.06E-06

1.19E-06

1.65E-09

6.44E-10

Incorporation into
paints and coatings -
2-part reactive
coatings

2.77E-06

1.08E-06

1.50E-09

5.84E-10

Processing/processing -incorporation into
formulation, mixture, or reaction product

Polymers used in
aerospace equipment and
products

Formulation of
TCEP containing
reactive resin

3.88E-06

1.51E-06

7.35E-09

2.87E-09

Commercial use

Laboratory chemicals

Use of laboratory
chemicals

2.80E-08

1.10E-08

1.80-11

6.90E-12

Paints and coatings

Use of paints and
coatings at job sites

1.63E-06

6.34E-07

1.04E-09

4.07E-10

8404

8405

Page 320 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

8406 Table 5-65. Lifetime Cancer Risk Summary for General Population and Fish Consumption

cou

OES

Lifetime Cancer Oral Risk Estimates

Life Cvcle/
Category

Subcategory

Adult Fish Ingestion General Population"

Adult Subsistence
Fisher

Tribes
(Current IR)

Tribes
(Heritage IR)

BAF 2,198

BAF 109

BAF
2,198

BAF
109

BAF
2,198

BAF
109

BAF
2,198

BAF
109

CT"

HE

CT4

HE

Manufacturing/
import

Import

Repackaging

2.02E-03

8.90E-03

1.00E-04

4.42E-04

5.72E-02

2.84E-03

8.68E-02

4.30E-03

5.00E-01

2.48E-02

Processing/
processing -
incorporation
into formulation,
mixture, or
reaction product

Flame retardant
in: paint and
coating
manufacturing

Incorporation into
paints and coatings
- 1 -part coatings

8.97E-03

3.94E-02

4.45E-04

1.96E-03

2.53E-01

1.26E-02

3.84E-01

1.91E-02

2.21E00

1.10E-01

Incorporation into
paints and coatings
- 2-part reactive
coatings

8.13E-03

3.58E-02

4.03E-04

1.77E-03

2.30E-01

1.14E-02

2.01E00

1.73E-02

2.01E00

9.96E-02

Polymers used in
aerospace
equipment and
products

Formulation of
TCEP containing
reactive resin

1.14E-02

5.00E-02

5.64E-04

2.48E-03

3.22E-01

1.59E-02

2.81E00

2.42E-02

2.81E00

1.39E-01

Commercial use

Laboratory
chemicals

Use of laboratory
chemicals

8.12E-05

3.57E-04

4.02E-06

1.77E-05

2.29E-03

1.14E-04

2.00E-02

1.72E-04

2.00E-02

9.93E-04

Paints and
coatings

Use of paints and
coatings at job sites

4.77E-03

2.10E-02

2.36E-04

1.04E-03

1.35E-01

6.68E-03

1.18E00

1.01E-02

1.18E00

5.83E-02

" Cancer risk estimates for the adult general population are based on the high-end fish ingestion rate of 22.2 g/day.
4 Exposure estimates are based on a general population mean fish ingestion rate of 5.04 g/day.

8407

8408

Page 321 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

8409 Table 5-66. General Population Dermal Acute and Chronic Non-cancer Risk Summary

cou

OES

Acute MOEs
LJFs = 30

Chronic Non-cancer MOE"

UFs = 30

Life Cycle/Category

Subcategory

Surface Water
(Adult
Swimming)

Su rfacc
Water
(Adult
Swimming)

Child
Playing in
Mud at

100 m a

Child
Activities
with Soil at
100 m a

Child Playing
in Mud at
1,000 m "

Child
Activities
with Soil at
1,000 m "

Manufacturing/import

Import

Repackaging

6.82E03

4.55E05

6.95E06

1.43E09

5.44E08

1.12E11

Processing/processing -
incorporation into
formulation, mixture, or
reaction product

Flame retardant
in: paint and
coating
manufacturing

Incorporation into
paints and coatings - 1-
part coatings

1.54E03

1.05E05

2.21E05

4.55E07

2.51E07

5.15E09

Incorporation into
paints and coatings - 2-
part reactive coatings

1.70E03

1.14E05

1.53E06

3.14E08

1.16E08

2.39E10

Polymers used in
aerospace
equipment and
products

Formulation of TCEP
containing reactive
resin

1.21E03

9.75E04

1.39E06

2.86E08

1.09E08

2.24E10

Processing/processing -
incorporation into
article

Aerospace
equipment and
products

Processing into 2-part
resin article

N/A

N/A

1.62E05

3.34E07

1.59E07

3.27E09

Commercial use

Laboratory
chemicals

Use of laboratory
chemicals

1.70E05

1.13E07

L45E05

2.98E07

1.33E07

2.72E09

Paints and
coatings

Use of paints and
coatings at job sites

2.90E03

1.95E05

94E01

1.93E04

1.80E04

3.71E06

" A soil concentration based of annual air deposition fluxes is used to estimate the acute exposures scenario of a child playing with mud and conducting activities in soil.

8410

8411

Page 322 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

8412 Table 5-67. Lifetime Cancer Risk Summary for General Population and Fish Consumption"

cou

OES

Chronic Inhalation MOEs

UFs = 30

Life Cyclc/Catcgorv

Subcategory

Ambient Air 50th

Ambient Air 95th

Manufacturing/import

Import

Repackaging

9.34E07

5.10E07

Processing/processing -
incorporation into formulation,
mixture, or reaction product

Flame retardant in:
paint and coating
manufacturing

Incorporation into paints and coatings -
1-part coatings

3.66E06

1.49E06

Incorporation into paints and coatings -
2-part reactive coatings

2.22E07

7.18E06

Polymers used in
aerospace equipment and
products

Formulation of TCEP containing reactive
resin

1.98E07

6.41E06

Processing/processing -
incorporation into article

Aerospace equipment and
products

Processing into 2-part resin article

2.41E06

1.82E06

Commercial use

Laboratory chemicals

Use of laboratory chemicals

2.10E06

1.48E06

Paints and coatings

Use of paints and coatings at job sites

1.23E03

4.98E02

" 2,500 lb Production Volume - High-End Release Estimate, Suburban Forest Scenario at 10 m

8413

8414	Table 5-68. General Population Lifetime Cancer Inhalation Risk Summary Table"

COU

OES

Distances
(m)

Lifetime Cancer Inhalation Risk

Life Cycle/
Category

Subcategory

Central Tendency
Meteorological Data

High-End Meteorological Data

Cancer Risk Estimate
for 50th Percentile Air
Concentration

Cancer Risk Estimate
for 95th Percentile Air
Concentration

Cancer Risk Estimate
for 50th Percentile Air
Concentration

Cancer Risk Estimate
for 95th Percentile Air
Concentration

Commercial
Use

Paints and
coatings

Use in paints
and coatings
at job sites

10

2.06E-05

2.47E-05

2.29E-05

5.68E-05

30

6.32E-06

9.26E-06

6.03E-06

1.57E-05

30-60

2.98E-06

6.37E-06

2.83E-06

9.62E-06

60

2.10E-06

3.52E-06

1.94E-06

4.97E-06

100

7.48E-07

1.44E-06

6.86E-07

1.83E-06

a 2,500 lb Production Volume - High-End Release Estimate, Suburban Forest Scenario

8415

Page 323 of 572

-------
8416

8417

8418

8419

8420

8421

8422

8423

8424

8425

8426

8427

8428

8429

8430

8431

8432

8433

8434

8435

8436

8437

8438

8439

8440

8441

8442

8443

8444

8445

8446

8447

8448

8449

8450

8451

8452

8453

8454

8455

8456

8457

8458

8459

8460

8461

8462

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

5.3.2.3.2 COUs without Quantitative Risk Estimates
Distribution in Commerce

Distribution in commerce includes transporting TCEP or TCEP-containing products between work sites
or to final use sites, as well as loading and unloading from transport vehicles. The general population
may be in the proximity of vehicles that transport TCEP or TCEP-containing products.

Although TCEP production volumes have declined, recent reports (e.g., the 2020 CDR) indicate that
production volumes may be below reporting levels; therefore, the precise volume is unknown. The
general decline in production volume would logically lead to decreased distribution into commerce.
Therefore, exposure and risk would also likely have declined with time. Exposure is possible from
ongoing manufacturing, processing, industrial, and commercial uses. EPA has assessed some risks
related to distribution in commerce (e.g., based on fugitive releases from loading operations) within
other relevant COUs (e.g., manufacturing/repackaging). However, EPA lacks the data to assess the full
set of risks to the general population from this COU. Due to limited data for the full set of possible
exposures, EPA's confidence in these exposures is indeterminant. EPA cannot characterize risk for the
general population for this COU separately from the risks already estimated for other relevant COUs.

Processing — Recycling

EPA did not quantify risks to the general population from releases during recycling of either electronic
waste (e-waste) or recycled foam products due to limited information and limited use of TCEP in
electronics.

EPA did not find data to quantify releases of TCEP from e-waste recycling facilities. The total releases
are expected to be low for several reasons: The volume of TCEP in e-waste products is low; only a
fraction of the products is recycled; and recycling will likely be dispersed over many e-waste sites.
Although EPA located information on the presence of TCEP at e-waste recycling facilities during
systematic review, the data sources did not provide the volume of TCEP-contained electronics processed
at any of the facilities identified. Therefore, EPA's confidence in these exposures is indeterminant and
cannot characterize risk from e-waste recycling.

TCEP may be present within flexible foam, fabric, textile, and other applications that have been made
from recycled foam scraps generated during trimming of original TCEP-containing manufactured foam
products. EPA was not able to determine, with reasonable accuracy, the exact flame retardants that are
used in these products and did not locate information on releases during recycling of such foam.

Industrial and Commercial Use (Other) - Aerospace Equipment and Products: EPA does not expect
significant releases to the environment for the following COUs:

•	Industrial use - other use - aerospace equipment and products; OES: installing article
(containing 2-part resin) for aerospace applications (electronic potting); and

•	Commercial use - other use - aerospace equipment and products; OES: installing article
(containing 2-part resin) for aerospace applications.

After TCEP-containing resins have cured within products that are installed, EPA expects TCEP releases
and dermal exposures will be limited by TCEP being entrained into the hardened polymer matrix.

During installation it is possible that very small levels of dust could be generated, these were quantified
in Table 5-57 and do not indicate risk to workers from inhalation nor do they indicate the generation of
significant dust releases occurring. Releases may occur via the mechanism of blooming (volatilization
from the cured resin surface) during the service life of the aircraft or aerospace article, but EPA expects

Page 324 of 572

-------
8463

8464

8465

8466

8467

8468

8469

8470

8471

8472

8473

8474

8475

8476

8477

8478

8479

8480

8481

8482

8483

8484

8485

8486

8487

8488

8489

8490

8491

8492

8493

8494

8495

8496

8497

8498

8499

8500

8501

8502

8503

8504

8505

8506

8507

8508

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

that such releases during installation will be negligible (OE 09; NICNAS. 2001). Therefore, the
potential risk to workers and the general population from releases during installation of TCEP -
containing aircraft and aerospace articles is low.

Commercial Uses That Have Been Phased Out

EPA determined that the following commercial use COUs for TCEP are not ongoing uses:

•	Commercial use - furnishing, cleaning, treatment/care products - fabric and textile products;

•	Commercial use - furnishing, cleaning, treatment/care products - foam seating and bedding
products;

•	Commercial use - construction, paint, electrical, and metal products - building/construction
materials - insulation; and

•	Commercial use - construction, paint, electrical, and metal products - building/construction
materials - wood and engineered wood products - wood resin composites.

These COUs were phased out beginning in the late 1980s or early 1990s and replaced by other flame
retardants or flame-retardant formulations. EPA did not locate data to estimate (1) the amount of TCEP
that was historically used in these products, (2) the amounts of these products that have already reached
the end of their service life, or (3) the amounts of these products that have already been disposed. Based
on the years that the phase-out occurred, many of these products not likely to be in use because the end
of their service life was already reached (e.g., commercial roofing has an estimated lifespan of 17 to 20
years). EPA assumes that any of these products still used commercially represent a fraction of the
overall amount of TCEP previously used for these purposes. Therefore, releases to the environment from
these commercial uses would also represent only a fraction of previous release amounts.

Due to lack of information and possible low exposure, EPA has not quantified risks to the general
population from releases associated with these COUs. Therefore, EPA's confidence in these exposures
is indeterminant and cannot characterize risk for these COUs.

Disposal

Disposal is possible throughout the lifecycle of TCEP and TCEP-containing products, including waste
treatment and disposal resulting from manufacturing, processing, and commercial and consumer uses.

For processing COUs, EPA estimated releases to landfills or incinerators (see Section 5.3.2.1):

•	Incorporation into formulation, mixture, or reaction product - paint/coating manufacture - 1-part
coating OES (landfill)

•	Incorporation into articles - aerospace equipment and products - processing in two-part resin
article OES (landfill)

•	Incorporation into formulation, mixture, or reaction product - paint/coating manufacture - 2-part
coating OES (incineration)

•	Incorporation into formulation, mixture, or reaction product - polymers in aerospace equipment
and products - formulation of reactive resins OES (incineration)

Both releases to landfills and incinerators rely on inputs provided by ESDs or GSs, but the ESDs and
GSs do not specify the proportion of the throughput that goes to either of these two disposal practices.
Therefore, EPA was unable to further quantify environmental releases related to these two disposal
processes. For three of these processing COUs, EPA was able to perform quantitative risk
characterization for releases to surface water (which includes onsite wastewater treatment or discharge
to POTWs, where applicable) (see Table 3-2); any releases to on-site waste treatment or POTWs were

Page 325 of 572

-------
8509

8510

8511

8512

8513

8514

8515

8516

8517

8518

8519

8520

8521

8522

8523

8524

8525

8526

8527

8528

8529

8530

8531

8532

8533

8534

8535

8536

8537

8538

8539

8540

8541

8542

8543

8544

8545

8546

8547

8548

8549

8550

8551

8552

8553

8554

8555

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

combined with other exposures and this combined risk to the general population was quantified for these
processing COUs.

Waste treatment (POTW or onsite) or disposal (landfill or incineration) is expected to be negligible for
industrial and commercial uses related to installing articles for aerospace applications. For the COUs of
manufacturing/repackaging, commercial use of paints and coatings, and commercial use of laboratory
chemicals, disposal to landfills or incinerators is not expected but EPA estimated surface water releases
that could include release to wastewater treatment or POTWs and any resulting risks to the general
population were assessed for the individual COUs.

For the commercial uses that have been phased out, any currently used products that contain TCEP are
expected to be disposed in landfills but will represent just a fraction of previous amounts when TCEP
was used more widely. Landfills would likely contain TCEP in commercial articles from these COUs,
but data are lacking with which to estimate exposure and risk from disposal or waste treatment activities
for these COUs, and EPA has not quantified such risks. For e-waste recycling, there is also too little
information to estimate exposure from disposal and only a small portion of e-waste is expected to
contain TCEP.

There may be releases to the environment from consumer articles containing TCEP via end-of-life
disposal and demolition of consumer articles in the built environment, and the associated down-the-drain
release of TCEP from domestic laundry that removes TCEP containing dust from clothing to
wastewater. It is difficult for EPA to quantify these end-of-life and down-the-drain laundry exposures
due to limited information on source attribution of the consumer COUs. EPA's confidence in these
exposures is indeterminant. Therefore, EPA did not quantitatively assess these scenarios due to lack of
reasonably available information. Section 3.3 presents more information on TCEP presence in
wastewater and at landfill sites and modeling of releases to groundwater from landfills.

5.3.2.4 Summary of Risk Estimates for Infants from Human Milk

EPA estimated infant risks from milk ingestion based on milk concentrations modeled for maternal
exposures associated with consumer, occupational, and general population groups. Infant exposures
through milk were estimated for both mean (105 mL/kg-day) and upper (153 mL/kg-day) milk intake
rates. Risk estimates for short-term and chronic infant exposures through milk were calculated for both
cancer and non-cancer endpoints for each COU within each maternal group. Short-term risks, which
have an averaging time of 30 days or less, were estimated based on the infant's first month of life. The
first month of life generally had the highest doses because of the highest milk ingestion rate per
kilogram of body weight; thus, it is most protective for estimating shorter term risks. For chronic non-
cancer risks, exposure typically occurs over at least 10 percent of lifetime in adults. However, it cannot
be ruled out that continuous exposure during the first year of life will result in permanent health effects
through adulthood. Chronic risks were thus considered for infant doses in the first year of life. Similarly,
cancer risks were also estimated using the linear low-dose extrapolation even though exposure did not
occur over the lifetime.

Acute infant doses were not estimated because the Verner Model is designed to estimate milk
concentrations and doses from continuous exposure rather than an acute, 1-day dose. However, if short-
term or chronic doses result in risk estimates below their corresponding benchmark MOEs, EPA
estimated acute risks by comparing short-term and chronic doses with an acute POD. Appendix H.4.1
through Appendix H.4.5 presents risk estimates for all iterations that EPA considered.

Page 326 of 572

-------
8556

8557

8558

8559

8560

8561

8562

8563

8564

8565

8566

8567

8568

8569

8570

8571

8572

8573

8574

8575

8576

8577

8578

8579

8580

8581

8582

8583

8584

8585

8586

8587

8588

8589

8590

8591

8592

8593

8594

8595

8596

8597

8598

8599

8600

8601

8602

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

For the consumer exposure pathways, short-term and chronic infant risk estimates were above the
corresponding benchmark MOEs for all COUs. Infant cancer risk estimates are above 1 in 1,000,000 for
two consumer exposure scenarios regardless of milk intake rate: Building/construction materials not
covered elsewhere (roofing insulation) and Building/construction materials - wood and engineered
wood products (wood flooring). The infant cancer risk estimates for these two COUs range from
8.05xl0~6 to 1.22xl0~5. The maternal cancer risk estimates for the same COUs range from 8.11xl0~6 to
4.5 x 10~2 (Table 5-59). Although the lower bound of the cancer risk estimates for the mother and infant
are similar, it is important to note that maternal risks are calculated by separate exposure routes {i.e.,
oral, dermal, and inhalation). Dermal exposure to roofing insulation resulted in the lowest maternal
cancer risk estimates, and all other routes resulted in risk estimates that were two to four magnitudes
higher. Other COUs with cancer risk estimates above 1 in 1,000,000 for the mother were below this
level for the infant ingesting human milk. Therefore, infant risks are not proportionally higher than
maternal risks. Furthermore, the maternal risk estimates in Table 5-59 are based on doses for an adult
weighing 80 kg. If they were adjusted for women of reproductive age, the risk estimates for this
population will increase given the higher dose. This underscores the conclusion that minimizing
maternal exposure to TCEP is most important for protecting an infant, as the mother is more sensitive.

For the occupational exposure pathways, 1- and 2-day application of spray paints and coatings were not
evaluated because the Verner model is intended to estimate only continuous maternal exposure. Among
the evaluated OESs, short-term and chronic infant risk estimates were below their benchmark MOEs for
Commercial use - paints & coatings - spray (2-part coatings, 250-day application) regardless of the
maternal dose type (chronic or subchronic) and milk intake rate (mean or upper). For Laboratory
chemicals, a mean milk intake rate resulted in short-term risk estimates below their benchmark MOEs
based on a subchronic maternal dose. An upper milk intake rate for the same OES resulted in short-term
and chronic infant risk estimates below their benchmark MOEs regardless of the maternal dose type.
Lastly, for Incorporation into paints and coatings - 1-part coatings, a mean milk intake rate resulted in
short-term risk estimates below their benchmark MOEs based on a subchronic maternal dose. An upper
milk intake rate and subchronic maternal dose for the same OES resulted in short-term and chronic
infant risk estimates below the benchmark MOE. However, acute infant risk estimates were above the
MOE for all of the above OESs.

Cancer risk estimates vary depending on the maternal worker dose type and the milk intake rate. For
subchronic maternal doses, infant cancer risk estimates exceeded 1 in 1,000,000 for 8 out of the 10
OESs regardless of milk intake rate:

•	Import and repackaging;

•	Incorporation into paints and coatings - 1-part coatings;

•	Incorporation into paints and coatings - 2-part reactive coatings;

•	Processing - formulation of TCEP into 2-part reactive resins;

•	Processing - processing into 2-part resin article;

•	Commercial use - paints & coatings - spray (1-part, 250-day application);

•	Commercial use - paints & coatings - spray (2-part reactive coatings, 250-day application); and

•	Laboratory chemicals.

For the above OESs, infant cancer risk estimates ranged from 2.67x 10~6 to 6.06x 10~5. The OES that
showed short-term and chronic infant risks also showed the highest infant cancer risk estimates:
commercial use - paints and coatings - spray (2-part coatings, 250-day application). For this OES,
infant cancer risk estimates based on a mean and upper milk intake rate were 3.61 x 10~5 and 6.06x 10~5,
respectively.

Page 327 of 572

-------
8603

8604

8605

8606

8607

8608

8609

8610

8611

8612

8613

8614

8615

8616

8617

8618

8619

8620

8621

8622

8623

8624

8625

8626

8627

8628

8629

8630

8631

8632

8633

8634

8635

8636

8637

8638

8639

8640

8641

8642

8643

8644

8645

8646

8647

8648

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

For chronic maternal doses, infant cancer risk estimates exceeded 1 in 1,000,000 for 5 or 7 OESs,
depending on the milk intake rate:

•	Import and repackaging (only for upper milk intake rate);

•	Incorporation into paints and coatings - 1-part coatings;

•	Processing - formulation of TCEP into 2-part reactive resins (only for upper milk intake rate);

•	Processing - processing into 2-part resin article;

•	Commercial use - paints & coatings - spray (1-part coatings, 250-day application);

•	Commercial use - paints & coatings - spray (2-part reactive coatings, 250-day application); and

•	Laboratory chemicals.

For the above OESs, infant cancer risk estimates ranged from 1,06x 10~6 to 4.91 x 10~5. Again,
Commercial use - paints & coatings - spray (2-part coatings, 250-day application) had the highest infant
cancer risk estimate at 3.37x 10~5 and 4.91 x 10-5 for a mean and upper milk intake rate, respectively.
Overall, for occupational exposure pathways, the risk estimates for short-term, chronic, and cancer
effects are lower in the infants compared to the mothers.

EPA estimated risks to infants in tribal communities exposed to TCEP through fish ingestion. As
discussed in Section 5.1.3.4.4, a current mean ingestion rate (IR) and heritage IR was used. The milk
intake rate (mean vs upper) did not significantly change risk estimates. For the high BAF, both milk
intake rates and both fish IRs resulted in MOEs below the short-term and chronic benchmarks for all
COUs except Laboratory use of chemicals. All COUs had cancer risk estimates above 1 in 1,000,000.
The low BAF and current IR did not show any MOEs below the short-term and chronic benchmarks for
all COUs. However, cancer risks exceeded 1 in 1,000,000 for all COUs except Laboratory use of
chemicals. The low BAF, heritage IR, and mean milk intake rate resulted in risk estimates blow the
short-term and chronic benchmarks for the same three COUs, as well as cancer risks for all COUs
except Laboratory use of chemicals. The same results can be observed for the low BAF, heritage IR, and
upper milk intake rate; in addition, one COU showed short-term risks that the mean milk intake rate did
not. Lastly, the COUs that had MOEs below the short-term and chronic benchmarks were also compared
against the acute benchmark to determine if there are acute risks at that exposure level. A high BAF did
have MOEs below the acute benchmark (4 to 5 COUs depending on the IR type). A low BAF had no
risk estimates below the acute benchmark.

For the general population, EPA focused on maternal oral exposures because they resulted in
significantly higher doses than dermal or inhalation. Within the oral routes, ingestion of fish (at the
general population's 90th percentile IR of 22.2 g/day) and undiluted drinking water were among the
sentinel pathways for mothers. EPA estimated infant risks using these pathways and did not combine
across other routes. Using a low BAF, no OESs had short-term or chronic risk estimates below the MOE
based on the mean and upper milk uptake rate. Cancer risk estimates did not exceed 1 in 1,000,000 for
any of the OESs based on the mean intake rate. However, based on the upper milk intake rate, the cancer
risk estimate for Formulation of TCEP containing reactive resin did exceed 1 in 1,000,000 (1.21 xl0~6).

For the general population adult fish ingestion based on the high BAF, no OESs had risk estimates
below their short-term and chronic MOEs for both milk intake rates. Cancer risk estimates did exceed 1
in 1,000,000 for all OESs except Laboratory use of chemicals. Under the mean milk intake rate, cancer
risk estimates ranged from 2.96x 10~6 to 1,66x 10 5. Under the upper milk intake rate, cancer risk
estimates ranged from 4,32x ]0 6 to 2,43x10 5, The OES with the highest cancer risk estimate is
Formulation of TCEP containing reactive resin. Risk estimates for infants of subsistence fisher were not

Page 328 of 572

-------
8649

8650

8651

8652

8653

8654

8655

8656

8657

8658

8659

8660

8661

8662

8663

8664

8665

8666

8667

8668

8669

8670

8671

8672

8673

8674

8675

8676

8677

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

calculated but are expected to fall in between those for the adult general population and tribal
population.

Due to the uncertainties in estimating fish ingestion exposure as discussed in Section 5.3.2.3, EPA also
considered ingestion of undiluted drinking water. This pathway did not result in any non-cancer risk
estimates below the benchmark MOE or cancer risk estimates above 1 in 1,000,000. No maternal risks
were observed either. While it is possible that combining other exposure routes, such as dermal
absorption from swimming, can result in additional scenarios showing infant risk estimates below their
benchmark MOEs, results from consumer, occupational, and general population fish ingestion
demonstrated that the mothers are more sensitive than the infants. There are no COUs or OESs across all
maternal groups that showed higher risk estimates in the infants compared to the mothers. In fact, some
COUs resulted in maternal doses and risk estimates that are several magnitudes higher for the mothers
than the infants. Therefore, protecting the mother will also protect the infant from exposure via human
milk.

5,3.3 Risk Characterization for Potentially Exposed or Susceptible Subpopulations	

EPA considered PESS throughout the exposure assessment and throughout the hazard identification and
dose-response analysis. EPA has identified several PESS factors that may contribute to a group having
increased exposure or biological susceptibility. Examples of these factors include lifestage, occupational
and certain consumer exposures, nutrition, and lifestyle activities.

For the TCEP draft risk evaluation, EPA accounted for the following PESS groups: infants exposed
through human milk from exposed individuals, children and male adolescents who use consumer articles
or are among the exposed general population, subsistence fishers, tribal populations, pregnant women,
workers and consumers who experience aggregated or sentinel exposures, fenceline communities who
live near facilities that emit TCEP, and firefighters.

Table 5-69 summarizes how PESS were incorporated into the risk evaluation and also summarizes the
remaining sources of uncertainty related to consideration of PESS. Appendix D provides additional
details on PESS considerations for the TCEP risk evaluation.

Page 329 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

8678 Table 5-69. Summary of PESS Considerations Incorporated into the Risk Evaluation

PESS
Categories

Potentially Exposed Individuals

Susceptible Subpopulations

Potential Increased Exposures
Incorporated into Exposure
Assessment

Sources of Uncertainty for
Exposure Assessment

Potential Sources of Biological
Susceptibility Incorporated into
Hazard Assessment

Sources of Uncertainty for
Hazard Assessment

Lifestage

•	Lifestage-specific exposure
scenarios included infants
exposed through human
milk.

•	Exposure factors by age
group were applied to
calculate consumer oral and
dermal exposures.

•	Children scenarios of playing
in mud and activities with
soil considered for dermal
and oral soil ingestion.

•	Mouthing of consumer
articles considered for infants
and children.

•	The level of exposure via
milk is uncertain as described
in Section 5.1.3.7.2

•	Uncertainties regarding the
appropriateness for adjusting
inhalation values to younger
life stages for the consumer
analysis

•	There is potential susceptibility is
related to different lifestages using
adolescent male mice as the POD for
short-term and chronic exposure.
Potential differences in other
lifestages, such as older individuals,
which might relate toxicokinetic or
toxicodynamic differences was
addressed through a 10x UF for human
variability (see Section 5.2.8 for POD
and UFs).

•	The short-term/chronic POD is
expected to be protective of adolescent,
developmental, and adult outcomes
(including pregnant females) based on
comparison with existing
developmental and reproductive
studies and a 2-year bioassay for
TCEP. Pregnant females are the basis
of the acute POD.

•	The magnitude of
differences in toxicokinetics
and toxicodynamics for
some individuals may be
greater than accounted for
by the UFh of 10.

•	Inability to use some
reproductive/developmental
data due to errors in one
study results in uncertainty
regarding the magnitude of
some effects in offspring.

•	Some uncertainty exists
based on limited number of
studies and differences in
specific outcomes among
studies.

Pre-existing
Disease

• EPA did not identify pre-
existing disease factors
influencing exposure



•	Pre-existing diseases and conditions,
especially those that lead to
neurological and behavioral effects,
reproductive effects, and cancer may
increase susceptibility to the effects of
TCEP.

•	This greater susceptibility is addressed
through the 10x UF for human
variability.

• The increase in

susceptibility is not known
and is a source of
uncertainty; differences may
be greater than the UFh of
10.

Page 330 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023



Potentially Exposed Individuals

Susceptible Subpopulations

PESS
Categories

Potential Increased Exposures
Incorporated into Exposure
Assessment

Sources of Uncertainty for
Exposure Assessment

Potential Sources of Biological
Susceptibility Incorporated into
Hazard Assessment

Sources of Uncertainty for
Hazard Assessment

Lifestyle
Activities

• EPA evaluated exposures
resulting from subsistence
fishing and considered
increased intake of fish in
these populations, as well as
tribal populations.

• There is a high level of
uncertainty in the BAF
values because of limited
monitoring data. There is also
uncertainty in the modeled
surface water concentrations.

• EPA did not identify lifestyle factors
that specifically influence
susceptibility to TCEP and that could
be quantified. Generally, certain
factors (e.g., smoking, alcohol
consumption, diet) can affect health
outcomes.

• This is a remaining source
of uncertainty.

Occupational
and

Consumer
Exposures

•	Monitoring data suggest that
firefighters have elevated
TCEP exposures because of
firefighting activities
(indicated by elevated urine
concentrations of BCEP, a
metabolite of TCEP (Mayer et
aL 2021; Javatilaka et aL
2017)).

•	Consumer articles intended for
use by children (children's
play structures, toy foam
blocks) considered in the
assessment of COUs.

• Uncertainties in duration of
use of consumer articles in
the home.

• EPA did not identify occupational and
consumer exposures that influence
susceptibility.

• This is a remaining source
of uncertainty.

Socio-

demographic

• EPA did not evaluate exposure
differences between racial
groups.

• Monitoring literature

indicates TCEP levels in dust
are significantly associated
with the presence of
extremely worn carpets. This
may be relevant for lower
socioeconomic status
families (Castorina et al.
2017).

• EPA did not identify specific evidence
that sociodemographic factors
influence susceptibility to TCEP
although it is known that they can
affect susceptibility to disease.

• This is a remaining source
of uncertainty.

Page 331 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

PESS
Categories

Potentially Exposed Individuals

Susceptible Subpopulations

Potential Increased Exposures
Incorporated into Exposure
Assessment

Sources of Uncertainty for
Exposure Assessment

Potential Sources of Biological
Susceptibility Incorporated into
Hazard Assessment

Sources of Uncertainty for
Hazard Assessment

Nutrition

• EPA did not identify

nutritional factors influencing
exposure.



• Nutrition can affect susceptibility to
disease generally. EPA did not identify
specific evidence that nutritional
factors influence susceptibility to
TCEP.

• This is a remaining source
of uncertainty.

Genetics/
Epigenetics

• EPA did not identify genetic
or epigenetic factors
influencing exposure.



• Genetic disorders may increase
susceptibility to male reproductive
effects; this was addressed through a
10x UF for human variability (see
Section 5.2.6.1.2).

• The magnitude of the impact
of genetic disorders is
unknown and is a source of
uncertainty; differences may
be greater than the UFh of
10.

Unique
Activities

• EPA did not evaluate activities
that are unique to tribal
populations (e.g., sweat
lodges, powwows). The
evaluation of high fish
consumption among tribal
populations is included in the
category Lifestyle Activities.

• There is uncertainty in how
exposure factors (e.g., water
consumption rate) change for
specific tribal lifeways.

• EPA did not identify unique activities
that influence susceptibility.

• This is a remaining source
of uncertainty.

Page 332 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023



Potentially Exposed Individuals

Susceptible Subpopulations

PESS
Categories

Potential Increased Exposures
Incorporated into Exposure
Assessment

Sources of Uncertainty for
Exposure Assessment

Potential Sources of Biological
Susceptibility Incorporated into
Hazard Assessment

Sources of Uncertainty for
Hazard Assessment

Aggregate
Exposures

•	Occupational dermal and
inhalation exposures
aggregated.

•	Consumer inhalation, dermal,
and oral ingestion exposures
are presented by individual but
are aggregated in Appendix I.

Uncertainty is associated with
several exposures that EPA did
not aggregate (see Section
5.1.4):

•	Inhalation and drinking
water for the general
population from co-located
facilities due to the lack of
site-specific data for TCEP.

•	Across consumer,
commercial, or industrial
COUs due to a lack of data
indicating such co-exposures
exist for TCEP. Across
exposure scenarios based on
release estimates for the
general population because
such assumptions could
result in double-counting.
Across other exposure
scenarios (e.g., mouthing
consumer articles, drinking
water) due to a lack of data
indicating the co-exposure of
TCEP.

• Not relevant to susceptibility



Other
Chemical
and Non-
chemical
Stressors

• EPA did not identify factors
influencing exposure.



•	In vitro data on co-exposure with
benzo-a-pyrene showed increased
impacts on inflammation and
proliferation pathways.

•	TCEP showed anti-estrogenic activity
in vitro after co-exposure with 17(3-
estradiol.

• There is insufficient data to
quantitatively address
potential increased
susceptibility due to these
factors; this is a remaining
source of uncertainty.

8679

Page 333 of 572

-------
8680

8681

8682

8683

8684

8685

8686

8687

8688

8689

8690

8691

8692

8693

8694

8695

8696

8697

8698

8699

8700

8701

8702

8703

8704

8705

8706

8707

8708

8709

8710

8711

8712

8713

8714

8715

8716

8717

8718

8719

8720

8721

8722

8723

8724

8725

8726

8727

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

EPA considered susceptibility when conducting hazard identification and dose-response analysis for
TCEP. Limited human data are available on health effects of TCEP, and EPA did not identify
differences in susceptibility among human populations. However, animal studies identified
developmental effects CNTP. 1991a). as well as sensitive sexes for certain health outcomes (higher
incidence of neurotoxicity in female rats (NI	), greater sensitivity of male (vs. female) mice in

reproductive effects (Chen et ai. 2015a)). and EPA quantified risks based on these endpoints in the risk
evaluation. An acute POD based on neurotoxicity was identified for pregnant rats (Moser et ai. 2015).

As identified in Table 5-59, many other susceptibility factors are generally considered to increase
susceptibility of individuals to chemical hazards. These factors include pre-existing diseases, alcohol
use, diet, stress, among others. The effect of these factors on susceptibility to health effects of TCEP is
not known; therefore, EPA is uncertain about the magnitude of any possible increased risk from effects
associated with TCEP exposure.

For non-cancer endpoints, EPA used a default value of 10 for human variability (UFh) to account for
increased susceptibility when quantifying risks from exposure to TCEP. The Risk Assessment Forum, in
A Review of the Reference Dose and Reference Concentration Processes (	32b). discusses

some of the evidence for choosing the default factor of 10 when data are lacking and describe the types
of populations that may be more susceptible, including different lifestages (e.g., of children and elderly).

002b). however, did not discuss all the factors presented in Table Apx D-2. Thus,
uncertainty remains regarding whether these additional susceptibility factors would be covered by the
default UFh value of 10 chosen for use in the TCEP risk evaluation.

For cancer, the dose-response model applied to animal tumor data employed low-dose linear
extrapolation, and this assumes any TCEP exposure is associated with some positive risk of getting
cancer. EPA made this assumption in the absence of an established MOA for TCEP and according to
guidance from U.S. EPA's Guidelines for Carcinogen Risk Assessment (	'005b). Assuming

all TCEP exposure is associated with some risk is likely to be health conservative because EPA does not
believe that a mutagenic MOA is likely for TCEP and a threshold below which cancer does not occur is
expected to exist. However, information is lacking with which to determine an appropriate threshold.
Even though the cancer dose-response modeling assumes any exposure is associated with a certain risk,
EPA presents risk estimates in comparison with benchmark risk levels (1 in 1,000,000 to 1 in 10,000).

Although there is likely to be variability in susceptibility across the human population, EPA did not
identify specific human groups that are expected to be more susceptible to cancer following TCEP
exposure. Other than relying on animal tumor data for the more sensitive sex, the available evidence
does not allow EPA to evaluate or quantify the potential for increased cancer risk in specific
subpopulations, such as for individuals with pre-existing diseases or those who smoke cigarettes. Given
that a mutagenic mode of action is unlikely, EPA does not anticipate greater cancer risks from early life
exposure to TCEP. Therefore, EPA is not applying an age-dependent adjustment factor.

EPA also considered PESS throughout the exposure assessment. EPA estimated infant risks from milk
ingestion based on milk concentrations modeled for maternal exposures associated with consumer,
occupational, and general population groups. Infant exposures through milk were estimated for both
mean (105 mL/kg-day) and upper (153 mL/kg-day) milk intake rates. Risk estimates for short-term and
chronic infant exposures through milk were calculated for both cancer and non-cancer endpoints for
each COU within each maternal group. While EPA only had slight confidence in the exposure estimates
for infants for this pathway, EPA did determine that infants exposed through human milk ingestion are

Page 334 of 572

-------
8728

8729

8730

8731

8732

8733

8734

8735

8736

8737

8738

8739

8740

8741

8742

8743

8744

8745

8746

8747

8748

8749

8750

8751

8752

8753

8754

8755

8756

8757

8758

8759

8760

8761

8762

8763

8764

8765

8766

8767

8768

8769

8770

8771

8772

8773

8774

8775

8776

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

not more sensitive than the mothers. Protecting the mother will also protect the infant from exposure via
human milk. Results of that analysis are included in Section 5.3.2.4.

For the general population, EPA also identified subsistence fishers, children, infants, and fenceline
communities as PESS groups. In its evaluation, EPA considered the increased intake of fish in
subsistence fishers. Although there was not enough reasonably available information to assess exposures
for tribal populations specifically, EPA quantitatively evaluated the tribal fish ingestion pathway for
TCEP. Children, infants, and fenceline communities were also identified as a PESS group for the
general population through the drinking water pathway and soil ingestion pathways. The fish ingestion
analysis and the analysis of children's exposure through drinking water and soil can be found in Section
5.3.2.3.1.

For occupational exposures, EPA also conducted a qualitative assessment for firefighters. Monitoring
data suggests that firefighters have elevated TCEP exposures as a result of firefighting activities.
Elevated levels of flame retardants have been found in dust collected from fire stations and in firefighter
personal equipment (Shen et at.. 2018). A study on firefighters reported increased urine concentrations
of BCEP, a metabolite of TCEP, from pre-fire to 3- and 6-hour post fire collections. Although the results
were not statistically significant, pre-fire vs. post fire concentrations indicate that firefighters may be at
increased risk of TCEP exposures during structure fires (Mayer et at.. 2021). Researchers from the CDC
measured urine samples for BCEP in 76 members of the general population and 146 firefighters who
performed structure firefighting while wearing full protective clothing and SCBA respirators. BCEP was
detected in 10 percent of the general population at a median level that was below the detection limit and
in 90 percent of firefighters at a median of 0.86 ng/mL (Javatilaka et at.. 2017). TCEP was measured at
five fire stations across the United States (California, Minnesota, New Hampshire, New York, and
Texas) at median concentrations of 1,040 ng/g. In comparing chemical concentrations by vacuum use,
this study did not observe any differences in TCEP concentrations due to cleaning practices (vacuuming)
(Shen et at.. 2018). These levels are less than the median (2,700 ng/g) concentrations measured in 201 1
in California house dust (Podsom et at.. 2012). The US Fire Profile study states that the total number of
firefighters in 2020, 364,300 (35 percent) were career, while 676,900 (65 percent) were volunteers. The
US Fire Profile study also states that the number of fire departments for career firefighters is up to a total
of 5,244 establishments and a total of 24,208 establishments for volunteer firefighters (NFPA. 2022).

For consumer exposures, EPA identified and evaluated the exposure for PESS groups including children
and infants through exposure to consumer products. Risk estimates for these PESS groups can be found
in Section 5.3.2.2. EPA has moderate confidence in the fabric and textile products COU, and slight to
moderate confidence in the foam seating and bedding products and building/construction materials-
wood resin COUs. Confidence ratings are derived from consideration of variety of factors including
confidence in the model used, the default values, and the input parameters (e.g., density, use duration,
weight fraction, dermal parameters), and the corroborating monitoring data (see Table 5-18).

Limited information was available in the peer-reviewed and gray literature on the TCEP COUs.
However, the Ecology Washington database sampled consumer articles that children under 3 years of
age are expected to contact and/or mouthed. Of the 268 products related to TSCA COUs, 24 articles
were detected to have TCEP. Eleven out of twenty-four (4 percent of total) articles were related to fabric
and textiles uses, whereas 13 out of 24 (5 percent of total) were in foam articles. Products were sampled
in the summer of 2012 (WSDE. 2023).

Ion as et at. C sampled children's toys in Antwerp, Belgium, and reported an overall detection
frequency of 28 percent (32/114) of TCEP detected in children toys produced around the year 2007.

Page 335 of 572

-------
8777

8778

8779

8780

8781

8782

8783

8784

8785

8786

8787

8788

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Two out of eight articles were for wooden toys. Fang et al. (2013) reported a detection frequency of 95

percent (19/20) of V6/TCEP in vehicles with an average model year of 2004. Stapleton i

detected only one instance of V6/TCEP in 102 foam couches across the United States during 201 1-2012.

Table 5-70. Summary of Detection Frequencies and Sampling Dates for Relevant Consumer
Products Containing TCEP					

cou

Detection
Frequency

n

Source

Sampling Date

Life Cycle/
Category

Subcategory

Consumer Use/

Furnishing,

cleaning,

treatment/care

products

Fabric and
textile products

4%

268

Ecology Washington
database (WSDE.

2023)

2012

Foam seating
and bedding
products (Foam
Couches)

1%

102

(Staoleton et al.. 2012)

2011-2012

5%

268

Ecology Washington
database (WSDE.

2023)

2012

70%

20

Fane et al. (2013)

2009-2011

Foam seating
and bedding
products (Auto
Foam)

95%

20

Fane et al. (2013)

2009-2011
vehicle average
model year 2004

Construction,
paint, electrical,
and metal
products

Building/
construction
materials -
wood and
engineered
wood products -
wood resin
composites

100%

1

(8CHER. 2012)

1997

25%

8

(lonas et al., 2014)

2007

Table 5-70 provides a summary of the detection frequencies of the monitoring literature. It is significant
that all these frequency estimates are pre the implementation of California TB 117-2013, and it is
anticipated that manufacturers have phased out TCEP from their product due to the introduction of the
less stringent flammability standards for upholstered furniture (TB 117-2013).

Page 336 of 572

-------
8789

8790

8791

8792

8793

8794

8795

8796

8797

8798

8799

8800

8801

8802

8803

8804

8805

8806

8807

8808

8809

8810

8811

8812

8813

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table 5-71. Suggested Consumer Population Sizes Based on Characterization of Consumer Article
Detection Frequencies						

COU

Detection
Frequency

Adjusted
Detection
Frequency:
Current Use

Total U.S.
Population

(of

331,449,281)"

Total U.S.
Children
under 5 vears
(of

18,400,235)"

Total U.S.
Females of
Reproductive

Age (of
118,273,566)"

Life Cycle/
Category

Subcategory

Furnishing,
cleaning,
treatment/
care products

Fabric and
textile products

4%

0.4%

1,325,797

73,601

473,094

Foam seating
and bedding
products

5%

0.5%

1,657,246

92,001

591,368

Construction,
paint,

electrical, and

metal

products

Building/
construction
materials - wood
and engineered
wood products -
wood resin
composites

l%b

1%

3,314,493

184,002

1,182,736

a Values from the 2020 U.S. Census.

h Assessor judgement to overwrite literature detection frequency value. Only 9 samples presented TCEP use in
wooden products.

Table 5-71 assigns a detection frequency value for each COU above slight-moderate confidence. Four
percent is chosen for Fabric and Textile Products, and five percent is selected for foam seating and
bedding products. Although Fang	indicates higher detection frequencies in vehicles (95

percent), the vehicles selected in this study were from an average model year of 2003.5, and it is
understood that auto manufacturers have moved away from using V6/TCEP formulations in their
vehicles. A detection frequency value of 1 percent is selected for wood resin products, due to the scarce
number of examples indicating TCEP use in wood articles.

An order of magnitude correction to adjust the detection frequencies to current uses is applied for fabric
and textile products and foam seating and bedding products to adjust for TB 117-2013. The adjustment
is not applied to wood resin composites, as TB 117-2013 applies to upholstered furniture.

To characterize the population utilizing these consumer articles, the adjusted detection frequencies are
multiplied by the total US population, total U.S. population of children under 5 years of age, and total
US population of females of reproductive age from the 2020 LIS census. This calculation provides a
ballpark figure of the expected number of individuals who are exposed to current consumer articles.

Major assumptions in the characterization of this population include the idea that the use of these
consumer articles scale linearly with the detection frequency of detection among consumer articles, the
detection frequencies in the monitoring literature is representative of the use of TCEP compared to other
FRs in the marketplace, and that the order of magnitude adjustment is sufficient to reflect the phase
away from TCEP to other OPFRs.

Page 337 of 572

-------
8814

8815

8816

8817

8818

8819

8820

8821

8822

8823

8824

8825

8826

8827

8828

8829

8830

8831

8832

8833

8834

8835

8836

8837

8838

8839

8840

8841

8842

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

5,3,4 Risk Characterization for Aggregate and Sentinel Exposures

Section 2605(b)(4)(F)(ii) of TSCA requires EPA, as a part of the risk evaluation process, to describe
whether aggregate or sentinel exposures under the COUs were considered and the basis for their
consideration.

The term aggregate is defined as "the combined exposures to an individual from a single chemical
substance across multiple routes and across multiple pathways" in the Agency's final rule, Procedures
for Chemical Risk Evaluation Under the Amended Toxic Substances Control Act (82 FR 33726, July 20,
2017) (see also Appendix A.2 (Glossary of Select Terms).

In the procedural rule, EPA defines sentinel exposure as "the exposure to a single chemical substance
that represents the plausible upper bound of exposure relative to all other exposures within a broad
category of similar or related exposures" (40 CFR 702.33). In this evaluation, EPA considered sentinel
exposures by considering risks to populations who may have upper bound exposures, including workers
and ONUs who perform activities with higher exposure potential and fenceline communities. EPA
characterized high-end exposures using modeling approaches and if available, using monitoring data.
Where information on the distribution of exposures is available, EPA typically uses the 95th percentile
value of the available dataset to characterize high-end exposure for a given COU.

Across Routes

The Supplemental TCEP Consumer Modeling Results includes a figure that aggregates the consumer
exposure estimates by route (inhalation, dermal, ingestion) for each COU and life stage combination. In
addition, this supplemental file includes risk tables that indicate whether aggregation across routes result
in risk. Figure 5-18 and Figure 5-19 provide two examples where an aggregation across routes could
result in chronic and acute risk, where consideration from a single route would not result in risk. For
example, for Figure 5-18, if dermal, ingestion, and inhalation routes were considered individually the
exposure estimates do not exceed the chronic benchmark of (0.091 mg/kg/d). However, when
aggregating dermal and inhalation exposures, the chronic benchmark of (0.091 mg/kg/d) is exceeded.

Page 338 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Aggregate Chronic Average Daily Doses (CADDs)

TCEPCOUs

—¦ Chronic Benchmark (0.091 mg/kg/d)
q 15 I Dermal
Ingestion

-	| Inhalation

"O

COU Lifestage

8843

8844	Figure 5-18. Aggregate CADDs for Consumer Use of textiles in Outdoor Play Structures at Adult,

8845	Youth2, and Youthl Life Stages

8846

Page 339 of 572

-------
8847

8848

8849

8850

8851

8852

8853

8854

8855

8856

8857

8858

8859

8860

8861

8862

8863

8864

8865

8866

8867

8868

8869

8870

8871

8872

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Aggregate Acute Doses (ADRs)

TCEP COUs

0.6

	Chrcnit Benchmark (0.315 mg^kg/d)

„ _ ¦	Dermal
0.5

¦	Ingestion

I	Inhalation

ceaile-carpec Child 1	cextNe-ca rpet Infant2

COU Lifestage

Figure 5-19. Aggregate Acute Average Daily Doses (ADRs) for Carpet Back Coating, Childl, and
Infant2 Life Stages

There were no instances of aggregate lifetime risk for any COU where there was not already risk to the
COU from an individual route. The supplemental file includes risk tables that can further be toggled to
explore aggregate risks.

EPA combined exposures for the milk pathway across all routes for each COUs/OESs within workers
and consumers. However, for the general population, EPA only assessed the oral route when assessing
the milk pathway because exposure estimates showed that oral doses were several magnitudes higher
than dermal or inhalation doses. As a result, oral exposures will be the primary driver for infant risks via
the milk pathway. Furthermore, within the adult oral pathways that include fish ingestion, drinking
water ingestion, and incidental water ingestion from swimming, EPA only considered fish and drinking
water ingestion. These two pathways constitute the highest oral doses, thus having the greatest potential
to result in infant risks from human milk ingestion. Indeed, infant cancer risk estimates exceeded 1 in
1,000,000 for all COUs/OESs based on maternal fish ingestion (high BAF). Aggregating other exposure
scenarios will not further inform risk characterization.

Across Exposure Scenario

The confidence in the general population exposure scenarios for drinking water ingestion, fish ingestion
(lowBAF), and inhalation (100 m) is moderate. For the formulation of TCEP containing reactive resin
OES, chronic non-diluted drinking water exposure estimates are 1,46x 10 4 mg/kg/d. For the same OES,
chronic fish ingestion concentrations are two to three orders of magnitude higher for the general
population and subsistence fishers at 0.033 and 0.94 mg/kg/d, respectively. Chronic inhalation exposure
estimates are given in mg/m3 and do not exhibit risk—even at 10 m from a hypothetical facility.

Page 340 of 572

-------
8873

8874

8875

8876

8877

8878

8879

8880

8881

8882

8883

8884

8885

8886

8887

8888

8889

8890

8891

8892

8893

8894

8895

8896

8897

8898

8899

8900

8901

8902

8903

8904

8905

8906

8907

8908

8909

8910

8911

8912

8913

8914

8915

8916

8917

8918

8919

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Therefore, aggregate exposure across general population exposure scenarios does not result in an
appreciable difference as the exposure is dominated by the sentinel exposure (fish ingestion).
Furthermore, since the general population and subsistence fisher estimates result in chronic risk for all
COUs, aggregating additional exposure scenarios (e.g., consumer, occupational) with the general
exposure scenarios (fish ingestion) is uninformative in characterizing risks.

The confidence in the consumer COUs is moderate for the subcategories of carpet back coating, textile
in outdoor play structures, living room foam, automobile foam, and wooden TV stands. Chronic
ingestion estimates are above the chronic benchmark (0.091 mg/kg/d) for each of these subcategories
(carpet back coating, textile in outdoor play structures, living room foam, automobile foam, and wooden
TV stands), and chronic dermal estimates are above the benchmark for wooden TV stands. Since the
consumer exposure estimates result in chronic risk, aggregating additional exposure scenarios (e.g.,
general population, occupational) with the consumer exposure scenarios is uninformative in
characterizing risk.

The other consumer exposure scenario subcategories (e.g., insulation, mattress, wood resin) have slight
confidence. Aggregating these subcategories with additional exposure scenarios (e.g., general
population, occupational) would be uninformative in characterizing risk due to the slight confidence in
these scenarios.

5.3.5 Overall Confidence and Remaining Uncertainties in Human Health Risk
	Characterization	

EPA took fate, exposure (occupational, consumer, and general population), and human health hazard
considerations into account when characterizing the human health risks of TCEP. Human health risk
characterization evaluated confidence from occupational, consumer, and general population exposures
and human health hazards. Hazard confidence and uncertainty is represented by health outcome and
exposure duration as reported in Section 5.2.7, which presents the confidence, uncertainties, and
limitations of the human health hazards for TCEP. Confidence in the exposure assessment has been
synthesized in the respective weight of the scientific evidence conclusion sections for occupational
exposures (see Section 5.1.1.4), consumer exposures (see Section 5.1.2.4), and general population
exposures (see Section 5.1.3.7). Table 5-72 provides a summary of confidence for exposures and
hazards for non-cancer endpoints for the COUs that resulted in any non-cancer risks, and Table 5-73
provides a confidence summary for cancer for the COUs that resulted in cancer risks.

Uncertainties associated with the occupational exposure assessment include a lack of reported data from
databases such as TRI, NEI, DMR, and more recently, CDR. Site-specific data were only available for a
small number of current processors, and it is not clear if this data are representative of these industries
and workplace practices.

Uncertainties associated with the general population exposures assessment included the lack of site-
specific information, the incongruence between the modeled concentrations and doses with the
monitoring data, and the complexity of the assessed exposure scenarios. Section 5.1.3.7 illustrates the
confidence in the assessment of the general population exposure scenarios.

5.3.5.1 Occupational Risk Estimates
Exposure Monitoring Data and Use of Models

EPA only identified monitoring data for dust occurring within an electronic waste recycling facility.
Monitoring data for the remaining COUs/OESs was not found. Surrogate monitoring data were found to
assess potential exposure to TCEP during installation of aircraft and aerospace articles and this

Page 341 of 572

-------
8920

8921

8922

8923

8924

8925

8926

8927

8928

8929

8930

8931

8932

8933

8934

8935

8936

8937

8938

8939

8940

8941

8942

8943

8944

8945

8946

8947

8948

8949

8950

8951

8952

8953

8954

8955

8956

8957

8958

8959

8960

8961

8962

8963

8964

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

estimated inhalation exposure used TCEP monitoring data for furniture manufacturing (Makinen et al..
2009). Surrogate monitoring data are also used for the assessment of paints and coatings use during
spray application. It is unclear if these COUs have similar worker activities and if they are fully
representative of worker exposure for the OESs of installation of aircraft and aerospace articles and use
of paints and coatings. The remaining COUs/OESs used modelling approaches to estimate worker
exposures.

Where sufficient data were reasonably available, the 95th and 50th percentile exposure concentrations
were calculated using these data. The underlying distribution of the data, and the representativeness of
the reasonably available data, are not known. Where discrete data were not reasonably available, EPA
used reported statistics from the Monte Carlo simulations {i.e., 50th and 95th percentile). Because EPA
could not verify these values, there is an added level of uncertainty.

For OESs that do not have monitoring data, EPA used relevant GSs and/or ESDs to identify worker
activities and exposure routes that are reasonably expected to occur. Exposure distributions were then
created using Monte Carlo simulation with 100,000 iterations and the Latin hypercube sampling method.

EPA calculated ADC and LADC values assuming workers and ONUs are regularly exposed during their
entire working lifetime, which likely results in an overestimate. Individuals may change jobs during
their career such that they are no longer exposed to TCEP; therefore, actual ADC and LADC values
would be lower than the estimates presented.

While EPA has confidence in the models used, it is possible that they may not account for variability of
exact processes and practices at an individual site. Furthermore, there are no 2020 CDR reports for
TCEP and only one from 2016. Therefore, EPA made assumptions about pounds per site-year (2,500
presented in risk tables) that leads to uncertainty in these estimates.

Assumptions Regarding Occupational Non-users

Exposures for ONUs can vary substantially and most data sources do not sufficiently describe the
proximity of these employees to the TCEP exposure source. As such, exposure levels for the
"occupational non-user" category will have high variability depending on the work activity; therefore,
all ONU exposure estimates except for recycling of e-waste are considered to have only slight
confidence. For the OES of recycling of e-waste, monitoring data were available for workers conducting
activities consistent with the activities of ONUs, this results in a confidence rating of moderate to robust.

Modeled Dermal Exposures

The Fractional Absorption Model is used to estimate dermal exposure to TCEP in occupational settings.
The model also assumes a single exposure event per day based on existing framework of the EPA/OPPT
2-Hand Dermal Exposure to Liquids Model and does not address variability in exposure duration and
frequency. Additionally, the studies used to obtain the underlying values of the quantity remaining on
the skin {Qu) did not take into consideration the fact that liquid retention on the skin may vary with
individuals and techniques of application on and removal from the hands. Also, the data used were
developed from three kinds of oils; therefore, the data may not be applicable to other liquids. Based on
these uncertainties, EPA has a moderate level of confidence in the assessed baseline exposure.

Page 342 of 572

-------
8965

8966

8967

8968

8969

8970

8971

8972

8973

8974

8975

8976

8977

8978

8979

8980

8981

8982

8983

8984

8985

8986

8987

8988

8989

8990

8991

8992

8993

8994

8995

8996

8997

8998

8999

9000

9001

9002

9003

9004

9005

9006

9007

9008

9009

9010

9011

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Number of Workers

There are several uncertainties surrounding the estimated number of workers potentially exposed to
TCEP. Most are unlikely to result in a systematic underestimate or overestimate but could result in an
inaccurate estimate. CDR data were not available to estimate the number of workers associated with
manufacturing, processing, or use of TCEP. There are also uncertainties with BLS data, which are used
to estimate the number of workers for the remaining COUs. EPA had to use higher-level NAICS codes
(at 3- to 5-digit level) combined with assumptions from the U.S. Census' SUSB, which could result in
inaccuracies if the distribution of workers in occupations with TCEP exposure differs from the overall
distribution of workers in each NAICS. Also, EPA needed to designate which industries and occupations
have potential exposures, and this may result in over- or underestimation. However, any inaccuracies
would not be likely to systematically either overestimate or underestimate the number of exposed
workers.

5.3.5.2 Consumer Risk Estimates
Lack of Weight Fraction Data

No safety data sheets were available for consumer products containing TCEP. Monitoring literature and
databases suggest that TCEP is used in consumer articles (e.g., fabric and textiles, home furnishings,
automobile foams, childrens toys, and building materials such as insulation). Section 5.1.2.2 highlights
the available information on the consumer COUs and relevant exposure scenarios. EPA only had a few
U.S. studies and databases (Castorina et ai. 2017; Fane et ai. 2013). including the Ecology Washington
Database (WSDE. 2023). which provide information on article weight fractions for the consumer COUs.
Where there were gaps, EPA utilized foreign data (Ionas et ai. JO I I; Marklund et ai. 2003; Ingerowski
et ai. 2001) to help select values for product weight fraction data. EPA is unclear on how relevant the
foreign weight fraction data are for consumer articles used in the United States. Moreover, one of these
European studies ("Ingerowski et ai. 2001) had a low-quality data evaluation rating and was from the
early 2000s. In addition, there are limitations in the data integrity in the Washington State Database
(WSDE. 2023). There is a possibility that a chemical could be a contaminant rather than a component of
the formulation of the consumer article. In addition, there are some quality assurance and quality control
issues with the database suggesting that it might be unreliable.

Nevertheless, due to the paucity of information, EPA used low-quality information where higher quality
information was unavailable. In general, EPA has slight confidence in the building and construction
materials COUs (e.g., insulation and acoustic ceiling); slight-moderate confidence in the wood resin
products and foam seating and bedding products exposure scenarios; and moderate confidence in the
fabric and textile COUs (e.g., carpet back coating).

Complexity of Exposure Scenarios

The indoor air and indoor dust literature indicate that TCEP is present at higher values in indoor vs.
outdoor environments suggesting amplified exposures in the home. Uncertainties in the particle and gas
distribution (see Section 3.3.1.2.1) of TCEP builds further uncertainty on the reliability of direct
inhalation estimates vs. dust-mediated exposure via dermal absorption and oral ingestion.

SVOCs such as TCEP exhibit complex behaviors in the indoor environment. Shin et ai (2014) indicates
that TCEP has a relatively high emission rate compared to other semi volatile organic compounds. Shin
et ai (2014) observed that dust parameters such as removal rate from vacuuming, and dust loading onto
carpets and indoor furnishings are important variables that influence emission rates. CEM does
incorporate defaults for cleaning frequency and cleaning efficiency from settled floor dust; however,
EPA was not able to obtain data on dust loading onto carpets when assessing the consumer COUs. The

Page 343 of 572

-------
9012

9013

9014

9015

9016

9017

9018

9019

9020

9021

9022

9023

9024

9025

9026

9027

9028

9029

9030

9031

9032

9033

9034

9035

9036

9037

9038

9039

9040

9041

9042

9043

9044

9045

9046

9047

9048

9049

9050

9051

9052

9053

9054

9055

9056

9057

9058

9059

9060

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

uncertainties related to the behavior of TCEP in the indoor dust matrix further builds uncertainty into the
consumer risk estimates.

Model and Parameter Uncertainties

CEM 3.0 is a deterministic (rather than a population-based) model that provides point estimates of
TCEP exposure to population of interest. CEM is not equipped to model complex emission profiles or
activity patterns of residents other than those pre-populated within CEM. EPA used the CEM 3.0's
sensitivity mode to vary certain parameters to help understand which parameters influence the exposure
estimates. The initial concentration of SVOC in the article (a product of weight fraction and product
density) was the most important parameter for consumer modeling. Best judgments were used to
approximate product density of consumer articles where defaults were unavailable. The uncertainties in
the weight fraction and density information are reflected in EPA's overall confidence in consumer
modeling.

Dermal absorption parameter of fraction absorbed (Fabs) was estimated at 35.1 percent for all consumer
article scenarios from Abdallah et al. (2016). This value overrode the embedded CEM calculation for
dermal absorption. Estimates derived from the literature were of higher confidence then the CEM 3.0
calculated dermal absorption parameters. Nevertheless, there are uncertainties as to the applicability of
this one fraction absorbed value for all scenarios. Fraction absorbed can be a function of duration of
article or dust contact; however, because EPA was uncertain as to how often consumers, infants, and
children would wash their hands, EPA retained a conservative fraction absorbed value for the purposes
of consumer modeling.

Monitoring vs. Modeled Concentrations and Doses

The incongruence between modeled and measured concentrations and doses helps illustrate further
uncertainties in the consumer exposure assessment. Modeled indoor air concentrations for the
building/construction materials, insulation scenario (12.07 mg/m3) are six orders of magnitude higher
than the highest indoor air TCEP concentration observed in the United States (95th percentile of 35
ng/m3) (Dodson et al.. 2017). This discrepancy suggests major uncertainties in the insulation exposure
scenario.

The highest observed modeled dust intake in the reported modeled literature was 1.38 |ig/kg-day
reported for children at a kindergarten in Hong Kong (Deng et al.. 2018b). This value is within one to
two orders of magnitude of EPA's highest oral and dermal modeled intakes for children. EPA's highest
modeled oral intakes was 6,92x 10 2 mg/kg-day (69.2 |ig/kg-day) for the foam toy block scenario. EPA's
highest observed dermal intakes via dermal absorption was 3,07/10 1 mg/kg-day (307 |ig/kg-day) for
the wood flooring scenario. These comparisons suggest that the oral and dermal intakes are more like
values reported in the literature than the modeled inhalation estimates.

Timeseries of Inhalation Exposure Estimates

CEM 3.0 estimates a chronic inhalation exposure by averaging the exposure over 365 days. Chronic
consumer inhalation exposures from TCEP containing articles are initially dominated by the gas phase
concentrations (due to offgassing of TCEP). Figure 5-20 and Figure 5-21 display the time series air
concentrations for acoustic ceilings and wood flooring scenarios. After 4 weeks for the acoustic ceiling
scenario and 2 weeks for the wood flooring scenario, chronic consumer inhalation exposures are
dominated by the dust air concentrations. Chronic inhalation concentrations from insulation were
dominated by the gas phase concentrations; however, Figure 5-22 displays a precipitous drop in
concentration from the insulation article after the first few months.

Page 344 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Acoustic Ceiling Time Series Air Concentrations

9061

9062

9063

9064



c
o

!-h

c

(L)

o

c
o

u

00

o
—1

100
10

0.1
0.01
0.001
0.0001
0.00001

(N^ri^i'Ot^ONO'-Hm^ri^t^OOONOCNm^r'Ot^OOO'-HCNmi^'O
rHrHrtrHrHrHrtrHMMMMMMMmmmmmm

t-~ o

Days

¦Mean Gas Phase Cone (mg/m3)
¦Dust Air Cone (mg/m3)

¦TSP Air Cone (mg/m3)

Abraded Particle Air Cone (mg/m3)

Figure 5-20. Consumer Modeling Time Series Results for Acoustic Ceilings

Wood Flooring Time Series Air Concentrations

10000

Days

oO

c
'o

!-h

c

(L)
O

c
o

u

00

o
—1

9065

9066

9067

0.00001

^Tt^om'OONCNi^oO'-H^rt^om'OONCNi^oO'-H^rt^om'OONCNi^)
'-HfN^ri^i'Ot^ONO'-Hm^ri^t^OOONOCNm^r'Ot^OOO'-HCNmi^'O
rHrHrHrHrHrHrtrtMMMMMMMmmmnmm

¦Mean Gas Phase Cone (mg/m3)
Dust Air Cone (mg/m3)

¦TSP Air Cone (mg/m3)

Abraded Particle Air Cone (mg/m3)

Figure 5-21. Consumer Modeling Time Series Results for Wood Flooring

Page 345 of 572

-------
9068

9069

9070

9071

9072

9073

9074

9075

9076

9077

9078

9079

9080

9081

9082

9083

9084

9085

9086

9087

9088

9089

9090

9091

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Insulation Time Series Air Concentrations

CO

s

~St>

c
o

!-h

c

(L)
O

c
o

u

00

o
—1

1000
100
10
1

0.1
0.01
0.001
0.0001
0.00001
0.000001

(N

o

m vo On (N >/-> oo
>/-> vo	O

¦Mean Gas Phase Cone (mg/m3)
¦Dust Air Cone (mg/m3)

m^rir-)t^ooONO(Nm^riot^ooO'-H(Nr<-ii^),o

Days

TSP Air Cone (mg/m3)

^—Abraded Particle Air Cone (mg/m3)

Figure 5-22. Consumer Modeling Time Series Results for Insulation

Consumer articles containing TCEP are no longer manufactured in the United States. Consumers may
obtain new products containing TCEP only via import. Older articles in the home may have already
undergone offgassing of TCEP; thus, there is uncertainty as to the relevance of continued inhalation
exposure from older consumer articles containing TCEP as much of the exposure may have already
occurred in the first few weeks.

Risk Estimates for Conservative Scenarios

EPA did not utilize a range of estimates to model a central tendency and high-end for consumer
exposures. Detection frequencies of TCEP were low for various consumer products in the Washington
State Database and accompanying monitoring data, and rather than utilize a central tendency (that
potentially was below realistic detection limits), EPA selected plausible worst-case values for weight
fractions. Due to this approach, EPA has more confidence in scenarios that did not exhibit risk than
scenarios that exhibited risk.

5.3.5.3 General Population Risk Estimates
Location Information

Due to the lack of site-specific information, the exposures assessment relied on assumptions for location
specific model inputs. This lack of data results in uncertainties surrounding these location specific
parameters (e.g., flow parameters and meteorological data). The AERMOD Model included two
meteorological conditions (Sioux Falls, South Dakota for central tendency meteorology and Lake
Charles, Louisiana for higher-end meteorology), in addition to different land coverage scenarios
(Suburban Forests and Oceans) to characterize potential amounts of annual TCEP deposition to soil

Page 346 of 572

-------
9092

9093

9094

9095

9096

9097

9098

9099

9100

9101

9102

9103

9104

9105

9106

9107

9108

9109

9110

9111

9112

9113

9114

9115

9116

9117

9118

9119

9120

9121

9122

9123

9124

9125

9126

9127

9128

9129

9130

9131

9132

9133

9134

9135

9136

9137

9138

9139

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

from air. It is unclear how relevant these meteorological conditions and land cover scenarios are to
TCEP facilities as there are no available site-specific information.

EPA modeled air concentrations and deposition fluxes at various distances from the hypothetical facility
releasing TCEP. EPA selected various distances to calculate exposure doses and inhalation
concentrations for the general population (e.g., ambient air exposure to the general population, soil
dermal and oral intakes for children). In general, EPA has more confidence in risk estimates at further
distances from the hypothetical facility than risk estimates at closer distances. For example, EPA has
less confidence soil dermal exposure at 100 m of the facility than it does with soil dermal exposure at
1,000 m of the facility.

Due to the lack of site-specific information for industrial and commercial releases of TCEP, EPA could
not estimate the proximity of general population residents to drinking water intake locations. Drinking
water estimates were calculated for non-diluted (i.e., drinking water intake locations are at the site of the
surface water release) conditions as a worst-case scenario. Drinking water estimates were also calculate
for diluted conditions by estimating the distance between intake location and the site of release via
drinking water intake information available for various SIC codes. EPA has more confidence in these
estimates as they represent a more plausible distance from which the general population would receive
their drinking water.

Monitoring vs. Modeled Concentrations and Doses

The incongruence between modeled and measured concentrations and doses helps illustrate further
uncertainties in the general population. WQP data on surface water TCEP concentrations is three to five
orders of magnitude lower than modeled surface water concentrations (see Sections 3.3.2.4 and 3.3.2.5).
TCEP fish tissue concentrations within the Great Lakes (Guo et al. 2017b) are two to three orders of
magnitude lower than the TCEP tissue concentrations calculated using a whole organism BCF value
from another high-quality study ( ewe et al.. 2018). Modeled soil concentrations were within one
order of magnitude of a single study from published literature (Mihaitovic an	2012); however, it

is important to note that similarity with a single study is not enough to build confidence in the relevance
or accuracy of modeled results.

Complexity of Exposures Scenarios

The dermal absorption and ingestion from soil exposures scenarios require a complex understanding of
fate and transport of TCEP. Soil concentrations were calculated by modeling deposition fluxes of TCEP
at various distances from a hypothetical facility. Soil intakes were estimated for two exposures
scenarios—a child playing in mud and a child performing activities with soil. Parameters to calculate
these exposures, such as surface areas, absorption factors, and intake rates, were available in EPA's
Exposure Factors Handbook (\ c< < i1 \ JO \ ); however, there is high uncertainty in the scenario due
to the multiple unknowns (e.g., hypothetical facility, hypothetical release estimate, unknown distance
between homes and facility).

Model and Parameter Uncertainties

An additional uncertainty for the general population and consumer assessment are model uncertainties.
VVWM-PSC allowed for the application of a standard, conservative, set of parameters and adjust for
physical-chemical properties of TCEP. For example, stream reach was set to represent a shallow
waterway with a width of 5 m and depth of 1 m. There are uncertainties on the applicability of this
shallow water body volume.

Page 347 of 572

-------
9140

9141

9142

9143

9144

9145

9146

9147

9148

9149

9150

9151

9152

9153

9154

9155

9156

9157

9158

9159

9160

9161

9162

9163

9164

9165

9166

9167

9168

9169

9170

9171

9172

9173

9174

9175

9176

9177

9178

9179

9180

9181

9182

9183

9184

9185

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Ambient and drinking water estimates via VVWM-PSC and EFAST utilized a 0 percent drinking water
treatment removal efficiency (see Section E.2.5.3). While TCEP has been shown to be recalcitrant to
removal treatment processes, EPA is uncertain whether advanced treatment methods can remove TCEP
from water.

For AERMOD, EPA specified deposition parameters for such as the fraction of gas vs particle phase,
diffusivity in air, diffusivity in water, and the MMAD. Further sensitivity analysis can illustrate the
effects these parameters have on the deposition fluxes. Conflicting information in the peer-reviewed
literature creates uncertainties on the appropriate values of these parameters. Okeme (2018) has
described the complexities associated with the gas and particle partitioning of TCEP and has suggested
reported high concentrations of TCEP in particulates may be a result of sampling artifact (see Section
3.3.1.2.1).

A major uncertainty in fish ingestion exposure estimates was the selection of BAF values; Section
2.12.2 provides a review of BAFs found in the literature. The BAF of 2,198 for walleye {Sander vitreus)
from Guo et al. (2017a) was initially selected as a representative study of the U.S. population as it
sampled surface water and fish tissue concentrations in the Great Lakes. Walleye also represent a cool-
water top predator that serves as an important food fish. This species potentially preys on secondary and
tertiary consumers; however, it is uncertain what localized conditions affect BAF values within Guo et
al	. Furthermore, the surface water concentration and fish tissue concentrations were collected in

different years, thus it is difficult to hypothesize if TCEP surface water concentrations at the time of
sample collection influenced BAF values. A possible explanation for the resulting high oral risk
estimates could be an issue specific to BAFs for walleye (Sander vitreus) within the selected study Guo
et al. (2017a).

Risk Estimates for Conservative Scenarios

To help characterize risk EPA uses a range of central tendency and high-end estimates, as well as
varying scenarios. EPA has more confidence in a risk estimate when risk is observed using conservative
assumptions. In addition, EPA has more confidence in risk estimates when risk is not observed using
fewer conservative assumptions. No risk observed with conservative parameters can build confidence
that the OES/COU is not a risk to consumers or the general population. For example, drinking water
risks were estimated for drinking water, diluted drinking water, incidental ingestion via swimming and
drinking water contamination from landfill leachate. None of these scenarios resulted in chronic oral
risk. Lifetime cancer risks were found for a few OESs (Incorporation into 1-part and 2-part reactive
paints and coatings, Commercial use of paints and coatings, and Processing of 2-part resin articles);
however, when adjusting for dilution to drinking water intake locations, these OESs no longer show
lifetime cancer risk.

Due to the uncertainties in the BAF for walleye, EPA considered BAF values from all reviewed studies
to capture a range conditions (see Section 2.12.2). Liu et al. (2019a) measured BAFs for multiple aquatic
species in China and reported the lowest value of 109 to 202 L/kg for mud carp (iCirrhinus molitorella).
Samples were collected from an e-waste polluted pond in South China. Risk estimates using this lowest
BAF value (109 L/kg) still resulted in risks for fish consumption (see Table 5-60). Lastly, EPA's
modeled surface water concentrations are generally several magnitudes higher than measured
concentrations, thus resultant fish tissue concentrations and doses are high regardless of BAF. However,
EPA still relied on modeled data because of the paucity of measured data.

Page 348 of 572

-------
9186

9187

9188

9189

9190

9191

9192

9193

9194

9195

9196

9197

9198

9199

9200

9201

9202

9203

9204

9205

9206

9207

9208

9209

9210

9211

9212

9213

9214

9215

9216

9217

9218

9219

9220

9221

9222

9223

9224

9225

9226

9227

9228

9229

9230

9231

9232

9233

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

5.3.5.4 Hazard Values

EPA has moderate confidence in all hazard values used to modeled risks from TCEP. There are
uncertainties that are common to all values. All are based on animal toxicity data, TCEP-specific
information related to differences between animals and humans is lacking, and TCEP values are from
oral toxicity studies that required extrapolation to inhalation and dermal hazard values. The impact of
these assumptions on the direction of risk (under- or overprediction) is unknown. Additional
uncertainties specific to individual hazard values are described below, with details presented in Section
5.2.7.

Acute HED and HEC

Based on the weight of the scientific evidence analysis of the reasonably available toxicity studies from
animals, the key acute exposure effect is neurotoxicity. EPA identified a POD from high-quality acute
animal toxicity study to calculate risks for acute exposure scenarios for TCEP. Tilson et 90)
identified neurotoxicity in female rats, and EPA concluded that these types of effects are likely to be
caused by TCEP. EPA did not identify human data or other animal toxicity data using acute exposure
durations, and there is uncertainty because the POD does not account for all the effects associated with
acute exposure.

Short- Term/Chronic HED and HEC

EPA concluded that reproductive and developmental toxicity in humans is likely to be caused by TCEP
and identified a high-quality 35-day study in adolescent male mice that identified decreases in
seminiferous tubule numbers as the non-cancer POD for both short-term and chronic exposure scenarios
(Chen et al. ). The observed effect is adverse and fertility due to male reproductive effects is
known to be sensitive in humans. Using Chen et al. (2015a) for the POD is expected to be protective of
other hazards (e.g., neurotoxicity) for these exposure durations. There is uncertainty about the precision
of the doses because Chen et al. (2015a) is a dietary study and the authors did not state the amount of
food consumed. Using a 35-day toxicity study for chronic exposure durations adds some uncertainty
(e.g., the POD for the same effect may be lower after chronic exposure) but based on the weight of the
scientific evidence for other studies with male reproductive toxicity at higher doses and limited data
from an unobtainable inhalation study that identified effects related to male reproductive toxicity and
fertility, EPA believes the use of this study is relevant for the chronic duration.

Cancer CSF and IUR

Integrating evidence from humans, animals, and mechanistic studies resulted in a conclusion that TCEP
is likely to cause cancer in humans under relevant exposure circumstances. EPA used a sensitive
endpoint, kidney tumors in male rats, from a high-quality study CNTP. 1991b) to estimate cancer risks
from exposure to TCEP. The increased incidence of renal tubule adenomas and carcinomas is
considered adverse, relevant to humans, and representative of a continuum of benign to malignant
tumors. Increased incidence of tumors was identified in one epidemiological study that identified an
association between TCEP and thyroid tumors (Hoffman et al.. 2017). Because	identified

primarily benign kidney tumors (adenomas), the incidence of malignant tumors is less certain. However,
humans may be more sensitive and develop malignancies sooner than rats. Use of linear low dose
extrapolation is also uncertain because direct mutagenicity is not likely to be the predominant MO A;
thus, risks may be overpredicted using linear low dose extrapolation. Use of only kidney tumors could
result in some underestimation of risk.

Page 349 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

9234	Table 5-72. Overall Confidence for Acute, Short-Term, and Chronic Human Health Non-cancer Risk Characterization for COUs

9235	Resulting in Risks" h					

cou

Route/Exposed
Group

Exposure
Confidence

Hazard
Confidence

Risk
Characterization
Confidence

Life Cycle
Stage

Category

Subcategory

Occupational

Manufacturing

Import

Import

Dermal/Worker

++

++

Moderate

Processing

Processing - incorporation
into formulation, mixture, or
reaction product

Paint and coating
manufacturing

Dermal/Worker

++

++

Moderate

Processing - incorporation
into formulation, mixture, or
reaction product

Polymers used in
aerospace equipment and
products

Dermal/Worker

++

++

Moderate

Processing - incorporation
into article

Aerospace equipment
and products

Dermal/Worker

++

++

Moderate

Commercial
Use

Paints and coatings

Paints and coatings

Inhalation/W orker

++

++

Moderate

Inhalation/ONU

+

++

Slight

Dermal/Worker

++

++

Moderate

Laboratory chemicals

Laboratory chemical

Dermal/Worker

++

++

Moderate

( onsumci'

Consumer Use

Painls and coalings

Painls and coalings

N A

N A

++

N A

Furnishing, cleaning,
treatment/care products

Fabric and textile
products

Oral

++

++

Moderate

Inhalation

++

++

Moderate

Furnishing, cleaning,
treatment/care products

Foam seating and
bedding products

Oral

++

++

Moderate

Construction, paint,
electrical, and metal
products

Building/construction
materials

Inhalation

+

++

Slight





Oral

++

++

Moderate

Page 350 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

cou

Route/Exposed
Group

Exposure
Confidence

Hazard
Confidence

Risk
Characterization
Confidence

Life Cycle
Stage

Category

Subcategory



Construction, paint,
electrical, and metal
products

Building/construction
materials - wood and
engineered wood
products - wood resin
composites

Inhalation

++

++

Moderate

Dermal

++

++

Moderate

Disposal

Disposal

Disposal

N/A

N/A

++

N/A

( ieiK'ial popukilion exposures

Manufacturing

Import

Import

Oral

+

++

Slight

Processing

Processing - incorporation
into formulation, mixture, or
reaction product

Paint and coating
manufacturing

Oral

++

++

Moderate

Processing - incorporation
into formulation, mixture, or
reaction product

Polymers used in
aerospace equipment and
products

Oral

++

++

Moderate

Processing - incorporation
into article

Aerospace equipment and
products

Oral

+

++

Slight

Commercial Use

Paints and coatings

Paints and coatings

Oral

++

++

Moderate

Dermal

++

++

Moderate

Laboratory chemicals

Laboratory chemical

Oral

+

++

Slight

a This table identifies COUs that have any non-cancer risk (acute, short-term, or chronic) and the route associated with the risk.
b Short-term risks were evaluated for workers only, not consumers or the general population.

9236

9237

Page 351 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

9238	Table 5-73. TCEP Evidence Table Summarizing Overall Confidence for Lifetime Human Health Cancer Risk Characterization for

9239	CPUs Resulting in Risks					

COUs

Route/Exposed

Group

Exposure
Confidence

Hazard
Confidence

Risk
Characterization
Confidence

Life Cycle
Stage

Category

Subcategory

OcCU|\lllOIKll

Manufacturing

Import

Import

Dermal/Worker

++

++

Moderate

Processing

Processing - incorporation
into formulation, mixture, or
reaction product

Paint and coating
manufacturing

Dermal/Worker

++

++

Moderate

Processing - incorporation
into formulation, mixture, or
reaction product

Polymers used in
aerospace equipment and
products

Dermal/Worker

++

++

Moderate

Processing - incorporation
into article

Aerospace equipment
and products

Dermal/Worker

++

++

Moderate

Commercial
Use

Paints and coatings

Paints and coatings

Inhalation/W orker

++

++

Moderate

Inhalation/ONU

+

++

Slight

Dermal/Worker

++

++

Moderate

Laboratory chemicals

Laboratory chemical

Dermal/Worker

++

++

Moderate

( onsunvi'

Consumer Use

Paints and coatings

Paints and coatings

N/A

N/A

++

N/A

Furnishing, cleaning,
treatment/care products

Fabric and textile
products

Oral

++

++

Moderate

Inhalation

++

++

Moderate

Furnishing, cleaning,
treatment/care products

Foam seating and
bedding products

Oral

++

++

Moderate

Construction, paint, electrical,
and metal products

Building/construction
materials

Inhalation

+

++

Slight

Construction, paint, electrical,
and metal products

Building/construction
materials - wood and
engineered wood

Oral

++

++

Moderate

Inhalation

++

++

Moderate

Page 352 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

COUs

Route/Exposed
Group

Exposure
Confidence

Hazard
Confidence

Risk

Life Cycle
Stage

Category

Subcategory

Characterization
Confidence





products - wood resin
composites

Dermal

++

++

Moderate



Paints and coatings

Paints and coatings

N/A

N/A

++

N/A



Furnishing, cleaning,
treatment/care products

Fabric and textile
products

Oral

++

++

Moderate



Inhalation

++

++

Moderate



Dermal

++

++

Moderate



Furnishing, cleaning,
treatment/care products

Foam seating and
bedding products

Oral

++

++

Moderate



Inhalation

++

++

Moderate

Consumer Use

Dermal

++

++

Moderate





Oral

+

++

Slight



Construction, paint, electrical,
and metal products

Building/construction
materials

Inhalation

+

++

Slight







Dermal

+

++

Slight





Building/construction
materials - wood and
engineered wood
products - wood resin
composites

Oral

++

++

Moderate



Construction, paint, electrical,
and metal products

Dermal

++

++

Moderate

Disposal

Disposal

Disposal

N/A

N/A

++

N/A

(ieneral population exposures

Manufacturing

Import

Import

Oral

+

++

Slight

Processing

Processing - incorporation
into formulation, mixture, or
reaction product

Paint and coating
manufacturing

Oral

++

++

Moderate

Processing - incorporation
into formulation, mixture, or
reaction product

Polymers used in
aerospace equipment and
products

Oral

++

++

Moderate

Page 353 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

COUs

Route/Exposed
Group

Exposure
Confidence

Hazard
Confidence

Risk
Characterization
Confidence

Life Cycle
Stage

Category

Subcategory



Processing - incorporation
into article

Aerospace equipment and
products

Oral

+

++

Slight

Commercial
Use

Paints and coatings

Paints and coatings

Oral

++

++

Moderate

Inhalation

++

++

Moderate

Laboratory chemicals

Laboratory chemical

Oral

+

++

Slight

9240

Page 354 of 572

-------
9241

9242

9243

9244

9245

9246

9247

9248

9249

9250

9251

9252

9253

9254

9255

9256

9257

9258

9259

9260

9261

9262

9263

9264

9265

9266

9267

9268

9269

9270

9271

9272

9273

9274

9275

9276

9277

9278

9279

9280

9281

9282

9283

9284

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

6 UNREASONABLE RISK DETERMINATION

EPA has determined that TCEP presents an unreasonable risk of injury to human health and the
environment under the COUs. This draft unreasonable risk determination is based on the information in
previous sections of this draft risk evaluation and the appendices and supporting documents in
accordance with TSCA section 6(b), as well as TSCA's best available science (TSCA section 26(h)),
weight of the scientific evidence standards (TSCA section 26(i)), and relevant implementing regulations
in 40 CFR 702.

Twenty COUs were evaluated for TCEP and are listed in Table 1-1. The following COUs contribute to
the unreasonable risk, considered singularly or in combination with other exposures:

•	Manufacturing (import);

•	Processing - incorporation into formulation, mixture, or reaction product - paint and coating
manufacturing;

•	Processing - incorporation into formulation, mixture, or reaction product - polymers used in
aerospace equipment and products;

•	Processing - incorporation into article - aerospace equipment and products;

•	Commercial use - paints and coatings;

•	Commercial use - laboratory chemicals;

•	Consumer use - furnishing, cleaning, treatment/care products - fabric and textile products;

•	Consumer use - furnishing, cleaning, treatment/care products - foam seating and bedding
products; and

•	Consumer use - construction, paint, electrical, and metal products - building/construction
materials - wood and engineered wood products - wood resin composites.

The following COUs are not expected to contribute to the unreasonable risk:

•	Processing - recycling;

•	Distribution in commerce;

•	Industrial use - other use - aerospace equipment and products;

•	Commercial use - other use - aerospace equipment and products; and

•	Consumer use - construction, paint, electrical, and metal products - building/construction
materials - insulation.

EPA did not have sufficient information to determine whether the following COUs contribute to the
unreasonable risk:

•	Commercial use - furnishing, cleaning, treatment/care products - fabric and textile products;

•	Commercial use - furnishing, cleaning, treatment/care products - foam seating and bedding
products;

•	Commercial use - construction, paint, electrical, and metal products - building/construction
materials - insulation;

•	Commercial use - construction, paint, electrical, and metal products - building/construction
materials - wood and engineered wood products - wood resin composites;

•	Consumer use - paints and coatings; and

•	Disposal.

Because TCEP production volumes and uses have declined, and no companies reported manufacture or
import of TCEP in the 2020 CDR, EPA had limited data available to evaluate certain COUs. For those

Page 355 of 572

-------
9285

9286

9287

9288

9289

9290

9291

9292

9293

9294

9295

9296

9297

9298

9299

9300

9301

9302

9303

9304

9305

9306

9307

9308

9309

9310

9311

9312

9313

9314

9315

9316

9317

9318

9319

9320

9321

9322

9323

9324

9325

9326

9327

9328

9329

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

COUs, EPA made a risk determination by integrating reasonably available information in a qualitative
risk characterization. Analyses of those COUs with limited data are provided in Sections 4.3.6.2 and
5.3.2.1.2 of this draft risk evaluation.

The COUs that contribute to unreasonable risk from TCEP are based on risk estimates that assume a
production volume of 2,500 lb, which EPA estimates, based on the data available, is reflective of current
domestic TCEP use. However, TCEP's production volume was in the tens of thousands of pounds as
recently as 2015, and there are no existing federal limits on the use of TCEP in the United States. EPA
anticipates that unreasonable risk from TCEP will increase if production volumes increase from 2,500
lb; risk estimates associated with a 25,000 lb production volume are presented in Appendix G and the
Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Supplemental Information File: Risk
Calculator for Occupational Exposures.

Whether EPA makes a determination of unreasonable risk for a particular chemical substance under
amended TSCA depends upon risk-related factors beyond exceedance of benchmarks, such as the
endpoint under consideration, the reversibility of effect, exposure-related considerations (e.g., duration,
magnitude, or frequency of exposure, or population exposed), and the confidence in the information
used to inform the hazard and exposure values. The Agency generally has a moderate or robust degree
of confidence in its characterization of risk where the scientific evidence weighed against the
uncertainties is robust enough to characterize hazards, exposures, and risk estimates, as well as where
the uncertainties inherent in all risk estimates do not undermine EPA's confidence in its risk
characterization. This draft risk evaluation discusses important assumptions and key sources of
uncertainty in the risk characterization, and these are described in more detail in the respective weight of
the scientific evidence conclusions sections for fate and transport, environmental release, environmental
exposures, environmental hazards, and human health hazards. It also includes overall confidence and
remaining uncertainties sections for human health and environmental risk characterizations.

In the TCEP unreasonable risk determination, EPA considered risk estimates with an overall confidence
rating of slight, moderate, or robust. In general, the Agency makes an unreasonable risk determination
based on risk estimates that have an overall confidence rating of moderate or robust, since those
confidence ratings indicate the scientific evidence is adequate to characterize risk estimates despite
uncertainties or is such that it is unlikely the uncertainties could have a significant effect on the risk
estimates (see Appendix F.2.3.1). For TCEP, one COU, Consumer use - construction, paint, electrical,
and metal products - building/construction materials - insulation, had only slight confidence for all risk
estimates; therefore, the Agency is concluding that this COU does not contribute to the unreasonable
risk of TCEP.

Following issuance of a final risk evaluation for TCEP, EPA will initiate risk management for TCEP by
applying one or more of the requirements under TSCA section 6(a) to the extent necessary so that TCEP
no longer presents an unreasonable risk. Under TSCA section 6(a), EPA is not limited to regulating the
specific activities found to drive unreasonable risk and may select from among a suite of risk
management options related to manufacture, processing, distribution in commerce, commercial use, and
disposal to address the unreasonable risk. For instance, EPA may regulate upstream activities (e.g.,
processing, distribution in commerce) to address downstream activities that drive unreasonable risk
(e.g., use) — even if the upstream activities are not unreasonable risk drivers.

Page 356 of 572

-------
9330

9331

9332

9333

9334

9335

9336

9337

9338

9339

9340

9341

9342

9343

9344

9345

9346

9347

9348

9349

9350

9351

9352

9353

9354

9355

9356

9357

9358

9359

9360

9361

9362

9363

9364

9365

9366

9367

9368

9369

9370

9371

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

6.1 Unreasonable Risk to Human Health

Calculated risk estimates (MOEs or cancer risk estimates) can provide a risk profile of TCEP by
presenting a range of estimates for different health effects for different COUs. When characterizing the
risk to human health from occupational exposures during risk evaluation under TSCA, EPA conducts
assessments of risk and makes its determination of unreasonable risk from a scenario that does not
assume use of respiratory protection or other PPE.47 A calculated MOE that is less than the benchmark
MOE is a starting point for supporting a determination of unreasonable risk of injury to health, based on
non-cancer effects. Similarly, a calculated cancer risk estimate that is greater than the cancer benchmark
is a starting point for supporting a determination of unreasonable risk of injury to health from cancer. It
is important to emphasize that these calculated risk estimates alone are not bright-line indicators of
unreasonable risk, and factors must be considered other than whether a risk estimate exceeds a
benchmark.

6.1.1	Populations and Exposures EPA Assessed to Determine Unreasonable Risk to
Human Health

EPA evaluated risk to workers, including ONUs and male and female adolescents and adults (>16 years
old); consumer users; general population; and infants via human milk from exposed individuals using
reasonably available monitoring and modeling data for inhalation, dermal, and ingestion exposures, as
applicable. EPA evaluated risk from inhalation and dermal exposure of TCEP to workers as well as
inhalation exposures to ONUs. The Agency also evaluated risk from oral, dermal, and inhalation
exposures to consumers. For the general population, EPA evaluated risk from (1) ingestion exposures
via drinking water, incidental surface water ingestion, fish ingestion (including subsistence fishers), and
soil ingestion by children; (2) dermal exposures to swimmers and children playing in the mud and other
activities with soil; and (3) chronic inhalation exposure. For infants consuming the human milk of
exposed individuals, EPA evaluated risk from milk ingestion based on milk concentrations modeled for
maternal exposures associated with occupational, consumer, and general population COUs. Descriptions
of the data used for human health exposure and human health hazards are provided in Sections 5.1 and
5.2 of this draft risk evaluation. Uncertainties for overall exposures and hazards are presented in Section
5.3.5 and are summarized in Table 5-66 and Table 5-67 and are considered in the unreasonable risk
determination. Note that Table 5-52 of this draft risk evaluation presents TCEP exposure durations by
population.

6.1.2	Summary of Unreasonable Risks to Human Health

EPA determined that the unreasonable risks presented by TCEP are due to

•	non-cancer effects and cancer in workers from dermal and inhalation exposures;

•	non-cancer effects and cancer in consumers from ingestion, dermal, and inhalation exposures;

•	non-cancer effects and cancer in infants from exposure through human milk ingestion; and

•	non-cancer effects and cancer in the general population (including subsistence fishers, tribal
populations, and children) from fish consumption and, to a lesser extent, the general population
from inhalation exposure.

With respect to health endpoints upon which EPA is basing this unreasonable risk determination, the
Agency has moderate overall confidence in the following PODs: (1) acute neurotoxicity, (2) short-term
and chronic reproductive effects, and (3) kidney cancer. EPA's exposure and overall risk
characterization confidence levels varied and are summarized in Table 5-63.

47 It should be noted that, in some cases, baseline conditions may reflect certain mitigation measures, such as engineering
controls, in instances where exposure estimates are based on monitoring data at facilities that have engineering controls in
place.

Page 357 of 572

-------
9372

9373

9374

9375

9376

9377

9378

9379

9380

9381

9382

9383

9384

9385

9386

9387

9388

9389

9390

9391

9392

9393

9394

9395

9396

9397

9398

9399

9400

9401

9402

9403

9404

9405

9406

9407

9408

9409

9410

9411

9412

9413

9414

9415

9416

9417

9418

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

The health risk estimates for workers, ONUs, consumers, the general population, and infants through the
milk pathway are presented in Section 5.3.2. For consumer and general population exposures, risk
estimates are provided in Sections 5.3.2.2 and 5.3.2.3 of this draft risk evaluation only when margins of
exposure (MOEs) were smaller than benchmark MOEs for non-cancer effects or when cancer risks
exceeded benchmark risk levels of 1 in 1,000,000 (lxl0~6). A complete list of health risk estimates for
consumers and the general population is in the following supplemental files of the draft risk evaluation
(see also Appendix C): Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Supplemental
Information File: E-FAST Modeling Results, Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate
(TCEP) - Supplemental Information File: Exposure Air Concentration Risk Calculations, and Draft
Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Supplemental Information File: TCEP
Consumer Modeling Results, Risk Calculations and Sensitivity Analysis.

6.1.3 Basis for EPA's Determination of Unreasonable Risk to Human Health

In developing the exposure and hazard assessments for TCEP, EPA analyzed reasonably available
information to ascertain whether some human populations may have greater exposure and/or
susceptibility than the general population to the hazard posed by TCEP. For the TCEP draft risk
evaluation, EPA identified the following groups as PESS: pregnant women, infants exposed through
human milk from exposed individuals, children and male adolescents who use consumer articles or
among the exposed general population, subsistence fishers, tribal populations, workers and consumers
who experience aggregated or sentinel exposures, fenceline communities who live near facilities that
emit TCEP, and firefighters (see Section 5.3.3, Table 5-62, and Appendix D.l).

Risk estimates based on high-end exposure levels (e.g., 95th percentile) are generally intended to cover
individuals with sentinel exposure levels whereas risk estimates at the central tendency exposure are
generally estimates of average or typical exposure. EPA aggregated exposures across certain routes for
consumers and identified at least two COUs where aggregating exposures across routes resulted in risk
where there was not risk when considering a single route. EPA did not aggregate exposures across
consumer COUs, since each COU already presented chronic risk to consumers. Since risk to the general
population was driven by sentinel exposures via fish ingestion, EPA did not aggregate risk across routes
or exposure scenarios for this population. EPA did not characterize aggregate risk to workers. The
uncertainty factor (UF) of 10 for human variability that EPA applied to MOEs accounts for increased
susceptibility of populations such as children and elderly populations. EPA also generally relies on high-
end exposure levels to make an unreasonable risk determination to capture vulnerable populations that
are expected to have higher exposures. Additionally, the non-cancer PODs are based on susceptible
populations. The acute POD is based on effects observed during pregnancy and the short-term and
chronic POD is based on reproductive effects observed in adolescent males.

For cancer, although there is likely to be variability in susceptibility across the human population, EPA
did not identify specific human groups that are expected to be more susceptible to cancer following
TCEP exposure. More information on how EPA characterized sentinel and aggregate risks is provided in
Section 5.3.4. For infants consuming human milk from exposed individuals, EPA calculated risk
estimates based on the upper and mean human milk intake rate. Because the risk estimates for infants via
human milk from exposed individuals did not differ significantly when the mean human milk intake was
used vs. the upper human milk intake rate, EPA's unreasonable risk determination is based on the upper
human milk intake rate.

For the COUs listed below, the Agency had limited data available and was not able to quantify risks to
human health:

Page 358 of 572

-------
9419

9420

9421

9422

9423

9424

9425

9426

9427

9428

9429

9430

9431

9432

9433

9434

9435

9436

9437

9438

9439

9440

9441

9442

9443

9444

9445

9446

9447

9448

9449

9450

9451

9452

9453

9454

9455

9456

9457

9458

9459

9460

9461

9462

9463

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

•	Processing - recycling (for general population only);

•	Distribution in commerce;

•	Commercial use - furnishing, cleaning, treatment/care products - fabric and textile products;

•	Commercial use - furnishing, cleaning, treatment/care products - foam seating and bedding
products;

•	Commercial use - construction, paint, electrical, and metal products - building/construction
materials - insulation;

•	Commercial use - construction, paint, electrical, and metal products - building/construction
materials - wood and engineered wood products - wood resin composites;

•	Consumer use - paints and coatings; and

•	Disposal.

For the COU listed below, the Agency anticipated that human exposures would be negligible and did not
quantify risk to human health:

•	Distribution in commerce;

•	Commercial use - other use - aerospace equipment and products

6.1.4	Unreasonable Risk in Occupational Settings

Based on the occupational risk estimates and related risk factors, EPA is determining that cancer and
non-cancer effects from worker dermal exposure to TCEP in occupational settings for all COUs with
quantified risk estimates except for recycling, and from worker inhalation exposure to TCEP from one
COU (commercial use of paints and coatings), contribute to unreasonable risk. More information on
occupational risk estimates is in Section 5.3.2.1 of this draft risk evaluation.

EPA is using a Fractional Absorption Model to estimate dermal exposure to TCEP in occupational
settings. The model assumes a single exposure event per day and does not address variability in
exposure duration and frequency. However, even with these uncertainties and limitations, EPA still
considers the weight of the scientific evidence for dermal risk estimates generated by the model to be
sufficient for determining whether a COU contributes to unreasonable risk. More information on EPA's
confidence in these risk estimates and the uncertainties associated with them can be found in Section
5.1.1.4 of this draft risk evaluation.

6.1.5	Unreasonable Risk to Consumers

Based on the consumer risk estimates and related risk factors, EPA finds unreasonable risk of non-
cancer and cancer effects to infants and young children through age 5 from mouthing of articles covered
by the Consumer use - furnishing, cleaning, treatment/care products - foam seating and bedding
products COU and the Consumer use - furnishing, cleaning, treatment/care products - fabric and textile
products COU and from ingesting dust contaminated with TCEP from other articles in the home covered
by the remaining consumer COUs.

Additionally, dermal contact with TCEP from the Consumer use - construction, paint, electrical, and
metal products - building/construction materials - wood and engineered wood products - wood resin
composites COU contribute to acute and chronic risk for infants, children, adolescents, and adults.
Inhalation of TCEP from this COU contributes to acute and chronic risks for adults; however, inhalation
by consumers from this COU are primarily from the first few weeks of exposure via offgassing of
TCEP. Thus, EPA does not anticipate there to be unreasonable risk via inhalation from TCEP-containing
products since these products have already been in commerce for longer than the offgassing period.

Page 359 of 572

-------
9464

9465

9466

9467

9468

9469

9470

9471

9472

9473

9474

9475

9476

9477

9478

9479

9480

9481

9482

9483

9484

9485

9486

9487

9488

9489

9490

9491

9492

9493

9494

9495

9496

9497

9498

9499

9500

9501

9502

9503

9504

9505

9506

9507

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Additionally, inhalation of TCEP from the Consumer use - furnishing, cleaning, treatment/care products
- fabric and textile products COU contributes to acute inhalation risk for adults and cancer risks for
adults and children.

EPA's overall confidence in the acute, short-term, and chronic consumer inhalation, ingestion, and
dermal exposure estimates used to make a determination of unreasonable risk is moderate. More
information on the consumer analysis can be found in Sections 3.2.1, 3.4, 5.1.2, and 5.3.2.2 of the draft
risk evaluation.

6,1,6 Unreasonable Risk to the General Population	

EPA identified the following exposure routes as contributing to the unreasonable risk of TCEP for the
following sub-populations:

Fish Ingestion

Based on the risk estimates and related risk factors for fishers among the general population, subsistence
fishers and fishers who are members of tribes48 who eat fish contaminated with TCEP, EPA determined
that all COUs contribute to unreasonable risk of cancer. Additionally, based on the risk estimates and
related risk factors, the following is a summary of COUs that contribute to risks of non-cancer effects
for subsistence fishers and fishers who are members of tribes:

•	Three COUs contribute to unreasonable risk of acute non-cancer effects for subsistence fishers.

•	Four COUs contribute to unreasonable risk of chronic non-cancer effects for subsistence fishers.

•	Three COUs contribute to the unreasonable risk of acute non-cancer effects for tribes at their
current intake rate of fish; assuming a heritage intake rate of fish, a fourth COU contributes to
the unreasonable risk of acute non-cancer effects.

•	Four COUs contribute to the unreasonable risk of chronic non-cancer effects for tribes at both
intake rates of fish.

To make a determination of unreasonable risk based on fish consumption, EPA used the mean intake
rate for fishers among the general population, since the potentially exposed and susceptible population
of subsistence fishers and fishers who are tribe members have risk estimates based on their intake rates
of fish. Additionally, to determine unreasonable risk, EPA used a bioaccumulation factor (BAF) of 109
L/kg and an ingestion rate of 5.04 g/day (142.4 g/day for subsistence fishers and 216 g/day or 1,646
g/day for fishers who are members of tribes) for adults aged 16 to less than 70 years to calculate risk
estimates (Section 5.1.3.4.4). EPA's confidence in the risk estimates using the BAF of 109 L/kg is
moderate. Acute and chronic non-cancer risk estimates to the general population for oral fish ingestion
are in Table 5-60 and Table 5-61 of this draft risk evaluation. Cancer risk estimates for oral fish
ingestion are in Table 5-62.

Based on the risk estimates for adults, EPA estimates that TCEP presents unreasonable risk of acute and
chronic non-cancer effects and cancer for children aged 15 years old or less who consume fish tissue
contaminated with TCEP, due to their higher rate of ingestion per kg of body weight.

Inhalation

EPA estimates that one COU contributes to the unreasonable risk of TCEP via inhalation. EPA's
confidence in inhalation risk estimates is moderate at 100 m and is robust at 1,000 m. Chronic inhalation
non-cancer risk estimates indicating no risk for even the very conservative distance of 10 m are in Table

48 Subsistence fishers and fishers who are members of tribes represent a PESS group for TCEP due to their increased
exposure via fish ingestion.

Page 360 of 572

-------
9508

9509

9510

9511

9512

9513

9514

9515

9516

9517

9518

9519

9520

9521

9522

9523

9524

9525

9526

9527

9528

9529

9530

9531

9532

9533

9534

9535

9536

9537

9538

9539

9540

9541

9542

9543

9544

9545

9546

9547

9548

9549

9550

9551

9552

9553

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

5-64. Cancer risk estimates are very close to the benchmark of 1x 10~6 at 100 m for one COU
(Commercial use - paints and coatings), based on modeled concentrations without any analysis of land
use around facilities to identify if there are exposures to general population. Cancer inhalation risk
estimates are presented in Table 5-65.

Additionally, in this draft risk evaluation, EPA evaluated the following sub-populations and routes of
exposure but did not identify any contribution to the unreasonable risk of TCEP from these routes:

Drinking Water and Incidental Surface Water Ingestion

EPA does not estimate that ingestion of drinking water (diluted), drinking water from groundwater
contaminated with TCEP leaching from landfills, or incidental surface water ingestion during swimming
contribute to the unreasonable risk of TCEP for any COU. Acute oral non-cancer risk estimates for
drinking water and drinking water (diluted) ingestion for any age group {i.e., adults >21, youths 16-20,
youths 11-15, children 6-10, toddlers 1-5, and infants from birth to <1 year) are presented in Table 5-59
of this draft risk evaluation. Chronic non-cancer risk estimates for drinking water and incidental surface
water ingestion are provided in Table 5-61; cancer risk estimates from drinking water are presented in
Table 5-62.

Soil Ingestion

EPA does not estimate that chronic soil ingestion contributes to the unreasonable risk of TCEP for any
COU. Risk estimates were calculated for a child conducting activities with soil and playing in mud.
EPA's confidence in the risk estimates at 1,000 m is moderate. Chronic non-cancer risk estimates for
soil ingestion are presented in Table 5-61 of this draft risk evaluation.

Incidental Dermal from Swimming

EPA does not estimate that incidental dermal exposure to an adult swimming contributes to the
unreasonable risk of TCEP for any COU. Dermal acute and chronic non-cancer risk estimates for
swimming are provided in Table 5-63 of this draft risk evaluation. EPA's confidence in the risk
estimates is moderate.

Children '.s Dermal Exposure from Playing in Mud and Soil Activities

EPA does not estimate that chronic dermal exposure to children 3 to 6 years old playing in mud and
conducting soil activities contributes to the unreasonable risk of TCEP for any COU. EPA's confidence
in the risk estimates at 1,000 m is moderate. Dermal, chronic non-cancer risk estimates for children
playing in mud and soil activities are included in Table 5-63.

6.1,7 Unreasonable Risk to Infants from Human Milk

EPA evaluated risk to infants who ingest human milk from individuals exposed to TCEP under the
conditions of use for which the Agency was able to estimate risks. EPA concludes that risk for infants
ingesting human milk is less than the risk TCEP presents to workers, consumers, and the general
population under its COUs. Based on the risk estimates for this population, and EPA's confidence in
them (moderate), EPA determined that the human milk pathway contributes to the unreasonable risk of
TCEP for seven COUs (Section 5.3.2.4 and Appendix H.4.5).

6.2 Unreasonable Risk to the Environment

Calculated risk quotients (RQs) can provide a risk profile by presenting a range of estimates for different
environmental hazard effects for different COUs. An RQ equal to 1 indicates that the exposures are the
same as the concentration that causes effects. An RQ less than 1, when the exposure is less than the

Page 361 of 572

-------
9554

9555

9556

9557

9558

9559

9560

9561

9562

9563

9564

9565

9566

9567

9568

9569

9570

9571

9572

9573

9574

9575

9576

9577

9578

9579

9580

9581

9582

9583

9584

9585

9586

9587

9588

9589

9590

9591

9592

9593

9594

9595

9596

9597

9598

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

effect concentration, generally indicates that there is not risk of injury to the environment that would
support a determination of unreasonable risk for the chemical substance. An RQ greater than 1, when the
exposure is greater than the effect concentration, generally indicates that there is risk of injury to the
environment that would support a determination of unreasonable risk for the chemical substance.
Additionally, if a chronic RQ is 1 or greater, the Agency evaluates whether the chronic RQ is 1 or
greater for 14 or more days before making a determination of unreasonable risk.

6.2.1	Populations and Exposures EPA Assessed to Determine Unreasonable Risk to the
Environment

For aquatic organisms, EPA evaluated exposures via surface water and sediment (including pore water).
For terrestrial organisms, EPA evaluated exposures via soil, air, and surface water. The Agency did not
directly assess terrestrial organism exposures from air due to soil and terrestrial food web being the
driver of exposures to terrestrial organisms; however, EPA assessed terrestrial organism exposures from
air deposition of TCEP to soil. Additionally, EPA estimated terrestrial organism exposures from trophic
transfer of TCEP from soil and surface water.

6.2.2	Summary of Unreasonable Risks to the Environment

EPA quantitatively assessed risk for five COUs and determined that all five contribute to the
unreasonable risk to the environment presented by TCEP due to:

• chronic growth and development effects to the Japanese medaka fish in surface water and
sediment (including pore water).

Risks to terrestrial organisms and risks from trophic transfer from the five COUs quantitatively assessed
do not contribute to the unreasonable risk to the environment presented by TCEP.

6.2.3	Basis for EPA's Determination of Unreasonable Risk of Injury to the Environment

Consistent with EPA's determination of unreasonable risk to human health, the RQ is not treated as a
bright-line and other risk-based factors may be considered (e.g., confidence in the hazard and exposure
characterization, duration, magnitude, uncertainty) for purposes of making an unreasonable risk
determination. TCEP is described as a "ubiquitous" contaminant because it is commonly found in
various environmental compartments such as outdoor air, surface water, drinking water, groundwater,
soil, sediment, biota, and precipitation all over the world (see Section 3). Additionally, TCEP is
persistent in water, soil and sediment, and EPA has robust confidence that TCEP can undergo long-
range transport.

EPA has moderate confidence in the chronic aquatic hazards and aquatic exposures contributing to
unreasonable risk. Additionally, the Agency has moderate to robust confidence in the terrestrial
exposures and hazards, which do not contribute to unreasonable risk. Because exposure via soil and the
terrestrial food web was determined to be the driver of exposure, EPA does not expect exposure to
TCEP via air or surface water to contribute to unreasonable risk to terrestrial organisms. Similarly, EPA
does not expect exposure to TCEP via biosolids to contribute to unreasonable risk to the environment.
The Agency's overall environmental risk characterization confidence levels were varied and are
summarized in Table 4-23.

In making a determination of unreasonable risk, EPA considered aggregating environmental exposures
for aquatic and terrestrial organisms but did not because the surface water and sediment pathways for
aquatic organisms and the soil pathway for terrestrial organisms were such large contributors to
unreasonable risk (see Section 4.3.6.1).

Page 362 of 572

-------
9599

9600

9601

9602

9603

9604

9605

9606

9607

9608

9609

9610

9611

9612

9613

9614

9615

9616

9617

9618

9619

9620

9621

9622

9623

9624

9625

9626

9627

9628

9629

9630

9631

9632

9633

9634

9635

9636

9637

9638

9639

9640

9641

9642

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

For the COUs listed below, the Agency had limited data available and was not able to fully quantify
risks to the environment:

•	Processing - recycling;

•	Distribution in commerce;

•	Commercial use - furnishing, cleaning, treatment/care products - fabric and textile products;

•	Commercial use - furnishing, cleaning, treatment/care products - foam seating and bedding
products;

•	Commercial use - construction, paint, electrical, and metal products - building/construction
materials - insulation;

•	Commercial use - construction, paint, electrical, and metal products - building/construction
materials - wood and engineered wood products - wood resin composites;

•	Consumer use - furnishing, cleaning, treatment/care products - fabric and textile products;

•	Consumer use - furnishing, cleaning, treatment/care products - foam seating and bedding
products;

•	Consumer use - construction, paint, electrical, and metal products - building/construction
materials - insulation

•	Consumer use - construction, paint, electrical, and metal products - building/construction
materials wood and engineered wood products - wood resin composites.

•	Consumer use - paints and coatings; and

•	Disposal.

For the COUs listed below, the Agency anticipated that there would be no releases to the environment
and did not quantify risks to the environment:

•	Industrial use - other use - aerospace equipment and products

•	Commercial use - other use - aerospace equipment and products

6.3 Additional Information Regarding the Basis for the Unreasonable Risk
Determination

Table 6-1, Table 6-2, and Table 6-3 summarize the basis for this draft unreasonable risk determination
of injury to human health and the environment (Table 6-4) presented in this draft TCEP risk evaluation.
In these tables, a checkmark (V) indicates how the COU contributes to the unreasonable risk by
identifying the type of effect (e.g., non-cancer and cancer for human health; acute or chronic
environmental effects) and the exposure route to the population or receptor that results in such
contribution. Not all COUs, exposure routes, or populations or receptors evaluated are included in the
tables. The tables only include the relevant exposure route, or the population or receptor that supports
the conclusion that the COU contributes to the TCEP unreasonable risk determination. As explained in
Section 1, for this draft unreasonable risk determination, EPA considered the effects of TCEP to human
health at the central tendency and high-end, as well as effects of TCEP to human health and the
environment from the exposures associated from the condition of use, risk estimates, and uncertainties in
the analysis. See Section 5.3.2.1 of this draft risk evaluation for a summary of risk estimates.

6.3.1 Additional Information about COUs Characterized Qualitatively

As explained earlier in this section, EPA did not have enough data to calculate risk estimates for all
COUs, and EPA characterized the risk by integrating limited amounts of reasonably available
information in a qualitative characterization. While the Agency is concluding that TCEP, as a whole
chemical, presents unreasonable risk to human health and the environment, at this time, (1) EPA does
not have enough information to quantify with enough weight of the scientific evidence how much of the

Page 363 of 572

-------
9643

9644

9645

9646

9647

9648

9649

9650

9651

9652

9653

9654

9655

9656

9657

9658

9659

9660

9661

9662

9663

9664

9665

9666

9667

9668

9669

9670

9671

9672

9673

9674

9675

9676

9677

9678

9679

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

unreasonable risk of TCEP may be contributed by these COUs, or (2) EPA does not expect these COUs
to contribute to the unreasonable risk of TCEP due to negligible environmental releases or negligible
human exposures. EPA has summarized the basis for its conclusion about these COUs below.

For Processing - recycling, EPA did not find data to quantify environmental releases of TCEP from e-
waste facilities. The total releases are expected to be low since TCEP is not typically used in electronics.
While EPA cannot calculate risk estimates for processing - recycling, given the expected total releases,
EPA concludes that processing - recycling does not contribute to TCEP's unreasonable risk to the
environment.

In addition, EPA characterized distribution in commerce qualitatively since EPA had limited data about
exposures from these COUs besides those exposures from other COUs already quantified with release
estimates. While EPA cannot calculate risk estimates for distribution in commerce separately from the
risk related to loading and unloading from transport vehicles already estimated for other relevant COUs,
and because of the decline in TCEP production volumes, EPA has concluded that distribution in
commerce does not contribute to TCEP's unreasonable risk.

For disposal, releases to landfills, incinerators, air, and surface water are integrated as part of each OES
(including loading and unloading activities) used to evaluate each COU quantified, as opposed to a
standalone disposal COU. However, EPA is unable to determine if disposal contributes to TCEP's
unreasonable risk.

For Industrial use - other use - aerospace equipment and products, and Commercial use - other use -
aerospace equipment and products, EPA does not expect significant releases to the environment to occur
and does not expect these COUs to contribute to the unreasonable risk of TCEP to the environment (see
Section 5.3.2.3.2). Additionally, EPA did not quantify dermal exposures from these two COUs but does
not anticipate dermal exposures from these two COUs to contribute to the unreasonable risk of TCEP to
human health.

Finally, for commercial and consumer COUs evaluated qualitatively, according to literature sources,
TCEP was used for these commercial and consumer COUs in the past, but manufacturing and
processing was phased out starting in the late 1980s or early 1990s in favor of other flame retardants or
flame-retardant formulations. The Agency assumes that commercial and consumer products with TCEP
that are still in use, but are no longer manufactured or processed, represents a fraction of the overall
amount of TCEP previously used. Therefore, TCEP releases for these COUs are expected to be lower
than those associated with COUs already quantified in this draft risk evaluation; however, EPA is unable
to determine if these COUs contribute to TCEP's unreasonable risk.

Page 364 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

9680 Table 6-1. Supporting Basis for the Draft Unreasona

)le Risk Determination for Human Health (Occupational CPUs)

cou

Population

Exposure
Route

Human Health Effects

Life Cycle
Stage

Category

Subcategory

Acute
Non-cancer

Short-Term
Non-cancer

Chronic
Non-cancer

Lifetime
Cancer

Manufacturing

Import

Import

Worker

Dermal

Va

Va

~



General Population

Fish Ingestion



N/A



V

General Population -
Subsistence Fishers

Fish Ingestion



N/A

~

V

Tribes - Current IR

Fish Ingestion



N/A

~

V

Tribes - Heritage IR

Fish Ingestion

V

N/A

V

s

Processing

Processing -
incorporation into
formulation, mixture, or
reaction product

Paint and coating
manufacturing

Worker

Dermal*

V

~

V

s

General Population

Fish Ingestion



N/A



s

General Population -
Subsistence Fishers

Fish Ingestion

V

N/A

V

s

Tribes - Current IR

Fish Ingestion

V

N/A

V

s

Tribes - Heritage IR

Fish Ingestion

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

con

Population

Exposure
Route

Human Health Effects

Life Cycle
Stage

Category

Subcategory

Acute
Non-cancer

Short-Term
Non-cancer

Chronic
Non-cancer

Lifetime
Cancer







General Population -
Subsistence Fishers

Fish Ingestion



N/A



¦/

Tribes - Current IR

Fish Ingestion



N/A



¦/

Tribes - Heritage IR

Fish Ingestion



N/A



¦/

" The risk estimate exceeded the benchmark for both the central tendency and the high-end.

h The risk estimate exceeded the benchmark for the high-end and is based on the most conservative OES (1-part coatings).
c The risk estimate exceeded the benchmark for the high-end.

d The risk estimate exceeded the benchmark for the high-end and is based on the most conservative OES (2-part coatings, 250-day).

e The risk estimate exceeded the benchmark for both the high-end and central tendency and is based on the most conservative OES (2-part coatings, 250-day).

9681

Page 366 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

9682 Table 6-2. Supporting Basis for the Draft Unreasonable Ris

j. Determination for Human Health (Consumer CPUs)

cou





Acute
Non-cancer

Short-Term/
Chronic Non-cancer



Life Cycle
Stage

Category

Subcategory

Population"

Exposure Route

Cancer







Adult

Inhalation

~



V



Furnishing, cleaning,

treatment/care

products

Fabric and textile

Child

Ingestion - Dust and Mouthing



V

V



products

Infant

Ingestion - Dust and Mouthing

~

V







Child

Inhalation





V







Adult

Ingestion - Dust





V



Furnishing, cleaning,

treatment/

care products

Foam seating and

Child

Ingestion - Dust and Mouthing

~

S

S

Consumer
Use

bedding products

Infant

Ingestion - Dust and Mouthing

~

S





Child

Dermal





V





Adult

Dermal





V





Building/construction
materials - wood and

Adult

Ingestion - Dust





V



Construction, paint,
electrical, and metal

Adult

Inhalation





V



engineered wood

Child

Ingestion - Dust



V

V



products

products - wood resin
composites

Inhalation





V





Infant

Ingestion - Dust



S









Dermal



S



" "Child" represents ages 3 through 20 years, and "Infant" represents ages 0 through 2 years

9683

9684

Page 367 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

9685	Table 6-3. Supporting Basis for the Draft Unreasonable Risk Determination for Human Health (Infant Risks from Human Milk

9686	Ingestion, Upper Milk Intake Rate)					

COL

Maternal Exposure Route

Maternal Exposure
Duration

Short-
Term

Chronic

Cancer

Life Cycle
Stajjc

Category

Subcategory

Maternal occupational exposures

Manufacturing

Import

Import

Dermal, Inhalation (High-
End)

Chronic





~

Subchronic





~

Processing

Processing -
incorporation into
formulation, mixture,
or reaction product

Paint and coating manufacturing

Chronic





~

Subchronic





~

Polymers used in aerospace
equipment and products

Chronic





~

Subchronic





~

Processing -
incorporation into
article

Aerospace equipment products

Chronic







Subchronic







Commercial
Use

Paints and coatings

Paints and coatings

Chronic

~

~

~

Subchronic

~

~

~

Laboratory chemicals

Laboratory chemicals

Chronic





~

Subchronic

V

~

V

Maternal ueueral population exposures

Processing

Processing -
incorporation into
formulation, mixture,
or reaction product

Formulation of TCEP containing
reactive resin

General Population Fish
Ingestion (Low BAF)

N/A





V

Manufacturing

Import

Import

Tribal Fish Ingestion (Low
BAF)

Current IR





V

Heritage IR





V

Processing

Processing -
incorporation into
formulation, mixture,
or reaction product

Paint and coating manufacturing

Current IR





V

Heritage IR

V

~

V

Polymers used in aerospace
equipment and products

Current IR





~

Heritage IR

~



~

Commercial
Use

Paints and coatings

Paints and coatings

Current IR





~

Heritage IR

~



~

Laboratory chemicals

Laboratory chemicals

Current IR







Heritage IR







Page 368 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

con

Maternal Exposure Route

Maternal Exposure
Duration

Short-
Term

Chronic

Cancer

Life Cycle
Stajjc

Category

Subcategory

Maternal consumer exposures

Consumer Use

Construction, paint,
electrical, and metal
products

Construction, paint,
electrical, and metal
products

Building/construction materials -
materials not covered elsewhere -
wood resin composites
Building/construction materials -
materials not covered elsewhere -
wood resin composites

N/A

N/A





~

9687

Page 369 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

9688 Table 6-4. Supporting Basis for the Draft Unreasonable Risk Determination for the Environment

COIJ

Population/
Receptor

Compartment

Environmental Effects

Life Cycle Stage

Category

Subcategory

Acute

Chronic

Manufacturing

Import

Import

Aquatic

Surface water





Sediment



V

Processing

Processing - incorporation into
formulation, mixture, or reaction product

Paint and coating
manufacturing

Aquatic

Surface water





Sediment



V

Processing - incorporation into
formulation, mixture, or reaction product

Polymers used in aerospace
equipment and products

Aquatic

Surface water





Sediment



V

Commercial Use

Paints and coatings

Paints and coatings

Aquatic

Surface water





Sediment



V

Laboratory chemicals

Laboratory chemical

Aquatic

Surface water



V

Sediment



V

9689

9690

Page 370 of 572

-------
9691

9692

9693

9694

9695

9696

9697

9698

9699

9700

9701

9702

9703

9704

9705

9706

9707

9708

9709

9710

9711

9712

9713

9714

9715

9716

9717

9718

9719

9720

9721

9722

9723

9724

9725

9726

9727

9728

9729

9730

9731

9732

9733

9734

9735

9736

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

REFERENCES

Abaft Martincigh. BS. (2019). Concentrations, sources and human exposure implications of
organophosphate esters in indoor dust from South Africa. Chemosphere 230: 239-247.

http://dx.doi.ore 10 101 i. chemosphere!^I I
Abdallah. MA-E; Pawa arrad. S. (2016). Human dermal absorption of chlorinated

organophosphate flame retardants; implications for human exposure. Toxicol Appl Pharmacol
291: 28-37. http://dx.doi.oi ^ 10 101 'j.taap.201 ^ I J 004
Abdollahi. A; Eng. A; J an tun en. LM; Ahrens. L; Shoeib. M; Parnis. JM; Hamer. T. (2017).

Characterization of polyurethane foam (PUF) and sorbent impregnated PUF (SIP) disk passive
air samplers for measuring organophosphate flame retardants. Chemosphere 167: 212-219.
http://dx.doi.ore/10.1016/i .chemosphere!01 0° I I I
Alexander. GR. (1977). Food of vertebrate predators on trout waters in north central lower Michigan.

Mich Acad 10: 181-195.

Ali. N; Dirtu. AC; van Den Eede. N; Goosey. E; Harrad. S: Neels. H; T Mannetit pakley. J;

Douwes. aci. A. (2012). Occurrence of alternative flame retardants in indoor dust from
New Zealand: Indoor sources and human exposure assessment. Chemosphere 88: 1276-1282.
http://dx.doi.ore/10.1016/i.chemosphere.-01J 0 '< 100
Alonso-Amelot. ME. (2018). Milk as a natural product: foreign natural and anthropogenic organic

compounds in it. In R Atta ur (Ed.), Studies in Natural Products Chemistry, vol 56 (pp. 335-435).
Amsterdam, Netherlands: Elsevier, http://dx.doi 10 101 U978-0-444-64058-1.00010 N
Alzualde \ <-'^hl. M; Sipes. NS: Hsieh. 3U	\ ^ KK I i^les. RS; Muriaiu \ < H livedo. C.

(2018). Toxicity profiling of flame retardants in zebrafish embryos using a battery of assays for
developmental toxicity, neurotoxicity, cardiotoxicity and hepatotoxicity toward human
relevance. Neurotoxicol Teratol 70: 40-50. http://dx.doi.ore/10.1016/i.ntt..01S 10 00J
Andreseri J \. Muir. D; Lleno. D; Darling 1 I'lieokil*! 'x >' stcr. K. (2007). Emerging pollutants in the
North Sea in comparison to Lake Ontario, Canada, data. Environ Toxicol Chem 26: 1081-1089.
http://dx.doi.ore/10.1S97/0 11 R.l
Arnot. J; Gob as. F. (2006). A review of bi oconcentrati on factor (BCF) and bioaccumulation factor
(BAF) assessments for organic chemicals in aquatic organisms. Environ Rev 14: 257-297.
http://dx.doi.ore	106-005

Arukwe. A; Carteny. CC: Eeeen. T; Moder. M. (2018). Novel aspects of uptake patterns, metabolite
formation and toxicological responses in salmon exposed to the organophosphate esters-tris(2-
butoxyethyl)- and tris(2-chloroethyl) phosphate. Aquat Toxicol 196: 146-153.
http://dx.doi.ore 10 101 i .aquatox _01 \ 01 01 I
ASTM. (2014). Standard guide for conducting acute toxicity tests on test materials with fishes,
macroinvertebrates, and amphibians. (E 29-96(2014)). West Conshohocken, PA.
https://compass.astm.ore/document/?contentCode=ASTM%7CE0729-96R14%7Cen-US
Aston. LS; Nod; ber. IN; Reece. CA. (1996). Organophosphate flame retardants in needles of
Pinus ponderosa in the Sierra Nevada foothills. Bull Environ Contam Toxicol 57: 859-866.
http://dx.doi.ore	001289900269

AT SDR. (2012). Toxicological profile for phosphate ester flame retardants. (PB2013-105796). Atlanta,
GA: U.S. Department of Health and Human Services, Public Health Service.
http://www.atsdr.cdc.eov/toxprofiles/tp202.pdf
Awonaikc R, 1 tn \ I * \\ ania. F. (2021). Precipitation-induced transport and phase partitioning of
organophosphate esters (OPEs) in urban and rural watersheds. Environ Sci (Camb) 7: 2106-
2115. http://dx.doi.cnv 10 |Q',*> mrW00329A

Page 371 of 572

-------
9737

9738

9739

9740

9741

9742

9743

9744

9745

9746

9747

9748

9749

9750

9751

9752

9753

9754

9755

9756

9757

9758

9759

9760

9761

9762

9763

9764

9765

9766

9767

9768

9769

9770

9771

9772

9773

9774

9775

9776

9777

9778

9779

9780

9781

9782

9783

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Bacaloni. \ 1 ucci. t ^ trine Nazzari. M; Samper ^ < teana. A. (2008). Occurrence of
organophosphorus flame retardant and plasticizers in three volcanic lakes of central Italy.
Environ Sci Technol 42: 1898-1903.

Barnes. KK; Christenson. SC; Kolpin. * *\\ t o- izio. M; Furloi^ * I itiee. SD; Meyer. M 1 *' .rber.

(2004). Pharmaceuticals and other organic waste water contaminants within a leachate
plume downgradient of a municipal landfill. Ground Water Monit Remediat 24: 119-126.
http://dx.doi.orE	>592.2004.tb00720.x

Barnthouse. LW; PeAneelis. PL; Gardne	. RV; Suter. GW: Vaughan. PS. (1982).

Methodology for Environmental Risk Analysis. (ORNL/TM-8167). Oak Ridge, TN: Oak Ridge
National Laboratory.

Beckwitl (1999). Human renal carcinoma — Pathogenesis and biology. In CC Capen; E Pybing; JM
Rice; JP Wilbourn (Eds.), IARC Scientific Publication No 147 (pp. 81-93). Lyon, France:
International Agency for Research on Cancer.

Bekel' I b io. H; Warn «} (2021). Tissue distribution and bioaccumulation of organophosphate
esters in wild marine fish from Laizhou Bay, North China: Implications of human exposure via
fish consumption. J Hazard Mater 401: 123410. http://dx.doi.o	/i.ihazmat.2020.123410

Bekel	io. H; Wane. Q: Chen. J. (2019). Bioaccumulation and trophic transfer of emerging

organophosphate flame retardants in the marine food webs of Laizhou Bay, North China.

Environ Sci Technol 53: 13417-13426. http://dx.doi.org/10.102 l/acs.est.9b03687
Benotti. MI; Trenholm. RA; Vanderfoul <' 1 Unladv. JC: Stanford •!' I * v	v \ (2009).

Pharmaceuticals and endocrine disrupting compounds in U.S. drinking water. Environ Sci
Technol 43: 597-603. http://dx.doi.ore/10.1021/es801845a
Beth-Hubner. M. (1999). Toxicological evaluation and classification of the genotoxic, carcinogenic,
reprotoxic and sensitising potential of tris(2-chloroethyl)phosphate. Int Arch Occup Environ
Health 72: M17-M23. http://dx.doi.ore/10.1007/PL00014210
Beyer V Mackih U Vlatthies. M; Wani.t t \\ ¦ •'.¦ster. \i. (2000). Assessing long-range transport
potential of persistent organic pollutants. Environ Sci Technol 34: 699-703.

(1977). Microbial mutagenicity test with trichloroethyl phosphate and trichloropropyl
phosphate. (314/1/77). Flushing, NY: Aceto Corporation.
arprises. (2017). Safety data sheet: TC-800 part A. Tustin, CA: BJB Enterprises, Inc.

Blessineer. TP; Euling. SY; Wane. L; Hogan. KA; Cai. C; Klinefelter. G; Saillenfait. AM. (2020).

Ordinal dose-response modeling approach for the phthalate syndrome. Environ Int 134: 105287.
http://dx.doi.ore 10 101 i.envint.-^M .87
Blum hi. M; Birnbaum. LS; Piamond. ML; Phillips. A; Sinela. V: Sipes. NS: Stapleton. HM;

Venier. M. (2019). Organophosphate ester flame retardants: Are they a regrettable substitution
for polybrominated diphenyl ethers? Environ Sci Technol Lett 6: 638-649.
http://dx.doi.ore	.cs.estlett.9b00582

Bonifas. J; Blomekt (2015). N-acetylation of aromatic amines: implication for skin and immune
cells [Review], Frontiers in Bioscience (Elite Edition) 7: 267-281.
http://dx.doi.ore/10.2741/E733
Borga. K; Kidd. KA; Muu in r^'relumi t citder. JM; Gob a t V Kxicklick. J; Malm. O;

Powell. PE. (2012). Trophic magnification factors: Considerations of ecology, ecosystems, and
study design. Integr Environ Assess Manag 8: 64-84. http://dx.doi.ore	;am.244

Bradman \ i astorina. R; Gaspar I ^ ishioka. M; Colon. M; Weathers. W; Egeghy. PP; Maddalena.
R; Williams. cins. PL; McKone. TE. (2014). Flame retardant exposures in California early
childhood education environments. Chemosphere 116: 61-66.
http://dx.doi.ore/10.1016/i.chemosphere.-01 I OJ 0 J

Page 372 of 572

-------
9784

9785

9786

9787

9788

9789

9790

9791

9792

9793

9794

9795

9796

9797

9798

9799

9800

9801

9802

9803

9804

9805

9806

9807

9808

9809

9810

9811

9812

9813

9814

9815

9816

9817

9818

9819

9820

9821

9822

9823

9824

9825

9826

9827

9828

9829

9830

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Bradman. A; Caspar I', ("astorina. R; Tone-Lin. E; McKotv I' *>laddalcna, R. (2012). Environmental
Exposures in Early Childhood Education Environments (pp. 423). (08-305). Berkeley, CA:
University of California, Berkeley, https://www.arb.ca.eov/research/apr/past/08-305.pdf
Brandsma. S nards. P; Leslie. HA; de Boer. J. (2015). Tracing organophosphorus and brominated
flame retardants and plasticizers in an estuarine food web. Sci Total Environ 505: 22-31.
http://dx.doi.ore 10 101 i.scitotenv.201 I OS.072
Broadwater. K; Ceballos. D; Paee. E; Croteau. G; Mueller. C. (2017). Health hazard evaluation report:
HHE-2014-0131-3298, evaluation of occupational exposure to flame retardants at four
gymnastics studios. Broadwater, K; Ceballos, D; Page, E; Croteau, G; Mueller, C.
https://www.cdc.eov/niosh/hhe/reports/pdfs/	S8.pdf

Brod;	iesser. J; Bauer. C; Roembke. J. (1997). The environmental fate of six existing

chemicals in laboratory tests. Chemosphere 34: 515-538. http://dx.doi.ore/10.1016/S0045-
6535(96)00390-6

Bukowski. K; Wvsokinski. D; Mokra. K; Wozniak. K. (2019). DNA damage and methylation induced
by organophosphate flame retardants: Tris(2-chloroethyl) phosphate and tris(l-chloro-2-propyl)
phosphate in human peripheral blood mononuclear cells. Hum Exp Toxicol 38: 724-733.

http://dx.doi.ore/10.1177/096C 3839174
Burb'i 1 i Sanders. JM; Hn i 1 *\\ \Linhews. HB. (1991). Metabolism of tris(2-chloroethvD phosphate

in rats and mice. Drug Metab Dispos 19: 443-447.

Buszka. PM; Yeskis J'1 IvoU'in. DW; Furl on ^ * I aiee. SD; Meyer. MT. (2009). Waste-indicator
and pharmaceutical compounds in landfill-leachate-affected ground water near Elkhart, Indiana,
2000-2002. Bull Environ Contam Toxicol 82: 653-659. http://dx.doi.ore/10.100T/s00128-009~
9702-z

Caldwell (1999). Review of mononuclear cell leukemia in F-344 rat bioassays and its significance
to human cancer risk: A case study using alkyl phthalates [Review], Regul Toxicol Pharmacol
30: 45-53. http://dx.doi.ors 36/rtph. 1999.1305
Caldwell Idridee. SR; Lineton. A.W; McKee. RH. (1999). Retrospective evaluation of alpha 2u-
globulin accumulation in male rat kidneys following high doses of diisononyl phthalate. Toxicol
Sci 51: 153-160. http://dx.doi.org/10.1093/toxsci I I i '

Canb -an Velzen. Ml; Hallner. E; Zwinderman. AH; Wickman. M; Leonards. PE; van Ree. R; van
Riit LS. (2015). Exposure to organophosphate and polybrominated diphenyl ether flame
retardants via indoor dust and childhood asthma. Indoor Air 26: 403-413.
http://dx.doi.ore	na. 12221

C,	^ < \\ \ 1 t 1 iane. K; Corcoran. MB; Hosseini. S; Bonina. SM; Rockne. KJ;

Sturchio. NC; Giesv. IP; Lu> < 1 \ \ hang. G. (2017). Organophosphate Esters in Sediment of
the Great Lakes. Environ Sci Technol 51: 1441-1449. http://dx.doi.org/10.102 l/acs.est.6b05484
Cao. S; Zeng. X; Song. Ha; Li. H; Yu. Z; Sheng. G; Fu. J. (2012). Levels and distributions of

organophosphate flame retardants and plasticizers in sediment from Taihu Lake, China. Environ
Toxicol Chem 31: 1478-1484. http://dx.doi.ore/10.1002/etc. 1872
Cjo 'u, 'Im<. H bang. YC; Ren. MH; Zham I h« ^ o	\ < i CH; Shen. MH;

B , (2019). Influence of Air Pollution on Inhalation and Dermal Exposure of Human to
Organophosphate Flame Retardants: A Case Study During a Prolonged Haze Episode. Environ
Sci Technol 53: 3880-3887. http://dx.doi.org/10.102 l/acs.est.8b07053
Castorina. R; Butt. C; Stapleton. HM; Avery. D; Harley. KG; Holland. N; Eskenazi. B; Bradman. A.
(2017). Flame retardants and their metabolites in the homes and urine of pregnant women
residing in California (the CHAMACOS cohort). Chemosphere 179: 159-166.
http://dx.doi.ore 10 101 i.chemosphere.101 0 '< 0

Page 373 of 572

-------
9831

9832

9833

9834

9835

9836

9837

9838

9839

9840

9841

9842

9843

9844

9845

9846

9847

9848

9849

9850

9851

9852

9853

9854

9855

9856

9857

9858

9859

9860

9861

9862

9863

9864

9865

9866

9867

9868

9869

9870

9871

9872

9873

9874

9875

9876

9877

9878

9879

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

CDC. (2020). CDC Health Topics A-Z: Healthy food environments: Improving access to healthier food
[Website], https://www.cdc.gov/nutrition/healthv-food-environments/improving-access-to-
healthier-food.html
CDC. (2021). CDC Health Topics A-Z: Micronutrients [Website],

http s ://www. cdc. gov/nutriti on/mi cronutri ent-

malnutrition/index.1	>ttps%3 A%2F%2Fwww.cdc. gov%2Fimmpact%2Fin

dex.html

CDC. (2022). CDC Health Topics A-Z: Physical activity [Website],

https://www.cdc.eov/phvsicalactivity/index.html
CDC. (2023a). CDC Health Topics A-Z: Cancer [Website], https://www.cdc.gov/cancer/

CDC. (2023b). CDC Health Topics A-Z: Infertility FAQs [Website],
https://www.cdc.gov/reproductivehealth/infertilitv/index.htm
CDC. (2023c). CDC Health Topics A-Z: Nutrition [Website], https://www.cdc.eov/nutrition/index.html
CDC. (2023d). CDC Health Topics A-Z: Stress at work [Website],

https://www.cdc.eov/niosh/topics/stress/

C (2014). Chronic Hazard Advisory Panel on phthalates and phthalate alternatives (with

appendices). Bethesda, MD: U.S. Consumer Product Safety Commission, Directorate for Health
Sciences. https://www.cpsc.eov/s3fs-piiblic/CHAP-REPORT-With-Appendices.pdf
Chapman. DE; Michener. SR; Powis. G (1991). Metabolism of the flame retardant plasticizer tri s(2-
chloroethyl)phosphate by human and rat liver preparations. Fundam Appl Toxicol 17: 215-224.
http://dx.doi.ors )3/toxsci/l 7.2.215
Chen. G: Jin N \\ u \ 1 >u. L; Fu. Z. (2015a). Exposure of male mice to two kinds of organophosphate
flame retardants (OPFRs) induced oxidative stress and endocrine disruption. Environ Toxicol
Pharmacol 40: 310-318. http://dx.doi.ore/10.1016/i.etap.2015.06.021
Chen, u, 'bang. S: Jin. Y; Wu \ 1 ut 1 < Han. U i u (2015b). TPP and TCEP induce oxidative
stress and alter steroidogenesis in TM3 Leydig cells. Reprod Toxicol 57: 100-110.
http://dx.doi.ore 10 101 i reprotox.201 ^ 0s 01 I
Chen. *	« H \ * •¦¦¦van • (2019a). Sorption oftri-n-butyl phosphate and tris(2-chloroethyl)

phosphate on polyethylene and polyvinyl chloride microplastics in seawater. Mar Pollut Bull
149: 110490. http://dx.doi.ore/10 101 | marpolbul.-O I" I 10 190
Chen. Y; Ye. J; Chen. Ya; Hu. Ha; Zhang. H; Ou. H. (2019b). Degradation kinetics, mechanism and
toxicology of tris(2-chloroethyl) phosphate with 185 nm vacuum ultraviolet. Chem Eng J 356:
98-106. http://dx.doi.ore/10.1016/i.cei.2018.09.007
Chene. W; Sun. L; Huang. W; Rui	^hang. P; Dine. R: Li. M. (2013a). Detection and

distribution of Tris(2-chloroethyl) phosphate on the East Antarctic ice sheet. Chemosphere 92:
1017-1021. http://dx.doi.oi	1 .chemosphere.2013.03.038

Cheng. W; Xie. Z; Blais. J.M; Zhang. P; Li. M; Yam lang. W; Ding. R; Sun. L. (2013b).

Organophosphorus esters in the oceans and possible relation with ocean gyres. Environ Pollut
180: 159-164. http://dx.doi.ore/10.1016/i.envpol.2013.05.032
Chokwe. TB; Okonkwo. JO. (2019). Occurrence, distribution and ecological risk assessment of

organophosphorus flame retardants and plasticizers in sediment samples along the Vaal River
catchment, South Africa. Emerging Contaminants 5: 173-178.
http://dx.doi.ore 10 101 i .emcon _0 r" 0 00 '<

Cj	o. HS: Park. K; Lee. JW; Kim. P; Oh. JE. (2018). Tissue-specific distribution and

bioaccumulation potential of organophosphate flame retardants in crucian carp. Environ Pollut
239: 161-168. http://dx.doi.ore/10.1016/i.envpol.2018.03.104
Cj	.. JE. (2020). Seasonal occurrence and removal of organophosphate esters in conventional

and advanced drinking water treatment plants. Water Res 186: 116359.
http://dx.doi.ore 10 101 t .watres.2020 II

Page 374 of 572

-------
9880

9881

9882

9883

9884

9885

9886

9887

9888

9889

9890

9891

9892

9893

9894

9895

9896

9897

9898

9899

9900

9901

9902

9903

9904

9905

9906

9907

9908

9909

9910

9911

9912

9913

9914

9915

9916

9917

9918

9919

9920

9921

9922

9923

9924

9925

9926

9927

9928

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Confidential. (1973). Toxicology laboratory report - tris (2-chloroethyl) phosphate [TSCA. Submission],
(OTS0516689. 86-890000118S. TSCATS/401616).

https://ntrl.ntis.eov/NTRL/dashboard/searchResults/titleDetail/OTS0516689.xhtml
Cristale. J; Alvarez-Martin. A; Rodriguez-Cruz. S; Sanchez-Martin. Ml; Lacorte. S. (2017). Sorption
and desorption of organophosphate esters with different hydrophobicity by soils. Environ Sci
Pollut Res Int 24: 27870-27878. http://dx.doi.on	60-0

Cristale. J; Araeao Bele. TG; Lacorte. S; de Marchi. MKR. (2019). Occurrence of flame retardants in
landfills: A case study in Brazil. Environ Res 168: 420-427.
http://dx.doi.ore 10 101 i.envres._01 \ 10 010
Cristale. J; Garcia Vazquez \ I'grata t 1 acorte. S. (2013). Priority and emerging flame retardants in
rivers: occurrence in water and sediment, Daphnia magna toxicity and risk assessment. Environ
Int 59: 232-243. http://dx.doi.cn v hM 5/i.envint.2Q t'< 0 Oil
Cristale. J: Ramos. DP: Dantas. RF; Machulek Junici \ Lacorte. S: Sans. C; Esplueas. S. (2016). Can
activated sludge treatments and advanced oxidation processes remove organophosphorus flame
retardants? Environ Res 144: 1 1-18. http://dx.doi.orv 10 101 i.envre? .01 10 008
Delumyea. R; Petel. R. (1979). Deposition velocity of phosphorus-containing particles over southern

Lake Huron, April-October, 1975. Atmos Environ 13: 287-294. http://dx.doi.ore/10.1016/0004-
6'^h <\VVl I ^

Dene. M; Kuo. DTF; Wu. O: Zhang. Y; Liu. X; Liu. S: Hu. X; Ma u. Z: Zhang. H. (2018a).

Organophosphorus flame retardants and heavy metals in municipal landfill leachate treatment
system in Guangzhou, China. Environ Pollut 236: 137-145.
http://dx.doi.ore 10 101 i.envpol.20.1 s 01 0 L
Dene. WJ; Li. N: Wu. R; Richard. WKS; Wa Ho. W. (2018b). Phosphorus flame retardants and
Bisphenol A in indoor dust and PM2.5 in kindergartens and primary schools in Hong Kong.
Environ Pollut 235: 365-371. http://dx.doi.ore/10.1016/i.envpol JO I IJ 093
Descartei (2020). Bills of Lading (USA Bills Import HOUSES), http://www.datamyne.com
Dobrv. A; Keller. R. (1957). Vapor pressures of some phosphate and phosphonate esters. J Phys Chem

61: 1448-1449. http://dx.doi.ore/10.1021/il50556a052
Dodson. RE; Perovich. 1 < 1 < »vaci \ \ an den Eede. N: Ion as. AC; Dirtu. AC; Brod\ Kudel. RA.
(2012). After the PBDE phase-out: A broad suite of flame retardants in repeat house dust
samples from California. Environ Sci Technol 46: 13056-13066.
http://dx.doi.ore/10.102 i/es303879n
Dodson.	lesk	;on. MP; McCaulev. M; Camann. DE; Yau.	' nkiewicz. G; Rudel.

RA. (2017). Chemical exposures in recently renovated low-income housing: Influence of
building materials and occupant activities. Environ Int 109: 114-127.
http://dx.doi.ore 10 101 i.envint. JO I 0 00
Dodson. RE; Van den Eede 1 ovau \ 1 erovich 1 1	1 ><.udel. RA. (2014). Urinary

Biomonitoring of Phosphate Flame Retardants: Levels in California Adults and
Recommendations for Future Studies. Environ Sci Technol 48: 13625-13633.
http://dx.doi.ore/10.102 i/es503445c
DOE. (2016). Table 1: Chemicals of concern and associated chemical information. PACs. Washington,
DC.

Duncan. M. (2000). Fish consumption survey of the Suquamish Indian Tribe of the Port Madison Indian
Reservation, Puget Sound Region. Suquamish, WA: The Suquamish Tribe, Port Madison Indian
Reservation, http://www.deq.state.or.iis/wq/standards/docs/toxics/siiqiiamish2000report.pdf
Duratec. (2018). Safety Data Sheet: Grey fire-resistant primer. Bloomington, CA: Dura Technologies,
Inc.

Dybine. E; Sanner. T. (1999). Species differences in chemical carcinogenesis of the thyroid gland,
kidney and urinary bladder. In CC Capen; E Dybing; RM Rice; JD Wilbourn (Eds.), IARC

Page 375 of 572

-------
9929

9930

9931

9932

9933

9934

9935

9936

9937

9938

9939

9940

9941

9942

9943

9944

9945

9946

9947

9948

9949

9950

9951

9952

9953

9954

9955

9956

9957

9958

9959

9960

9961

9962

9963

9964

9965

9966

9967

9968

9969

9970

9971

9972

9973

9974

9975

9976

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Scientific Publications, vol 147 (pp. 15-32). Lyon, France: International Agency for Research on
Cancer. https://piiblications.iarc.fr/Book-And-Report-Series/Iarc-Scientific-Publications/Species-
Differences-In-Thyroid-Kidnev-And-Urinarv-Bladder-Carcinogenesis-1999
EC. (2009). Screening Assessment for the Challenge: Ethanol, 2-chloro-, phosphate (3:1) (Tris(2-
chloroethyl) phosphate [TCEP]). http://www.ec.gc.ca/ese-
ees/default. asp?lang=En&xml=C3 78778 A-D834-54E0-7F69-E6E2944A74FC
EC/HC. (2011). Screening assessment report on hexabromocyclododecane. Ottawa, Canada.

http://www.ec.gc.ca/ese-ees/default. asp?lang=En&n=7882C148-l#a4
ECB. (2003). Technical guidance document on risk assessment: Part II. (EUR 20418 EN/2).
Luxembourg: Office for Official Publications of the European Communities.
http://ihcp.irc.ec.eiiropa.eii/our activities/public-
health/risk assessment of Biocides/doc/tgd/tgdpart2 2ed.pdf
ECB. (2009). European Union risk assessment report: Tris(2-chloroethy 1) phosphate, TCEP (pp. 213).
Italy: European Commission, Office for Official Publications of the European Communities.
https://echa.eiiropa.eii/documents/10162/2663989d-1795-44al-8f30 I UiM I
ECHA. (2008). Guidance on information requirements and chemicals safety assessment: Supporting
reference guidelines, chapters R7-R11. Helsinki, Finland.

http://echa.europa.eu/web/guest/guidance-documents/guidance-on-information-requirements-
and-chemical-safetv-assessment
ECHA.. (2018). Screening report: An assessment of whether the use of TCEP, TCPP and TDCP in

articles should be restricted. Helsinki, Finland: European Union, European Chemicals Agency.
https://echa.europa.eu/documents/101 „ l'< II screening report tcep tepp td-
cp en.pdf/e0960aa7-f703-499c-24ff-fba627060698
Ekpe. 01): Choc trcelo. D; Oh. IE. (2020). Chapter One - Introduction of emerging halogenated
flame retardants in the environment. In JE Oh (Ed.), (pp. 1-39). Amsterdam, The Netherlands:
Elsevier, http://dx.doi.oi \ 10 101 5/bs.coac.2Q 19.11.002
Fang. M; Webster. TF; Gooden. D; Cooper. EM; McClean. MP; Carignan. C; Makev. C: Stapleton.

HM. (2013). Investigating a novel flame retardant known as V6: measurements in baby products,
house dust, and car dust. Environ Sci Technol 47: 4449-4454.
http://dx.doi.org/10.102 i/es400032v
FDRL. (1972). Cholinesterase studies on rats and rabbits with Olin's intermediate for Chemical 5898 1
[TSCA Submission], (1-11286-F. OTS0517372. 86-890000046. TSCATS/310359). Olin
Chemicals. https://ntrl.ntis.gov/NTRL/dashboard/searchResults/titlePetail/OTSQ517372.xhtml
Fernie. KJ; Palace. V: Peters. LE; Basu. N: Letcher. RJ; Karouna-Renier. NK; Schultz. SL: Lazarus. RS;
Rattn	(2015). Investigating endocrine and physiological parameters of captive American

kestrels exposed by diet to selected organophosphate flame retardants. Environ Sci Technol 49:
7448-7455. http://dx.doi.org/10.1021/acs.est.5b00857
Follmann. W: Wober. J. (2006). Investigation of cytotoxic, genotoxic, mutagenic, and estrogenic effects
of the flame retardants tris-(2-chloroethyl)-phosphate (TCEP) and tris-(2-chloropropyl)-
phosphate (TCPP) in vitro. Toxicol Lett 161: 124-134.
http://dx.doi.org 10 101 | .toxlet.2005.08.008
E; Mihailovic. 1. (201 1). Pollution of soils with organophosphorus flame retardants and

plasticizers. J Environ Monit 13: 2692-2694. http://dx.doi.org/10.1039/c 1 em 10538h
E; Puttmann. W. (2001). Occurrence of organophosphate esters in surface water and ground water

in Germany. J Environ Monit 3: 621 -626. http://dx.doij	9/b 105072a

E; Puttmann. W. (2003). Monitoring of the three organophosphate esters TBP, TCEP and TBEP in
river water and ground water (Oder, Germany). J Environ Monit 5: 346-352.
http://dx.doi.org/10.1039/b210342g

Page 376 of 572

-------
9977

9978

9979

9980

9981

9982

9983

9984

9985

9986

9987

9988

9989

9990

9991

9992

9993

9994

9995

9996

9997

9998

9999

10000

10001

10002

10003

10004

10005

10006

10007

10008

10009

10010

10011

10012

10013

10014

10015

10016

10017

10018

10019

10020

10021

10022

10023

10024

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Ut > Ut U ' nio. K; Li. H; Xue. Q; Chen "S . Wane. L; Shi J, I'm J, Zhang O, Zham \ Hang. G.
(2020). Occurrence and trophic magnification of organophosphate esters in an Antarctic
ecosystem: Insights into the shift from legacy to emerging pollutants. J Hazard Mater 396:
122742. http://dx.doi.org/l /i.ihazmat.2020.122742
Galloway. SM; Armstrong. Ml; Reuben. C; Colman. S; Brown r% i .tnnoii i	\U ^ ,ikamura.

imed. M; Duk. S; Rimpo. J; Margoli	snick. MA; Anderson. ger. E. (1987).

Chromosome aberrations and sister chromatid exchanges in Chinese hamster ovary cells
evaluations of 108 chemicals [Review], Environ Mol Mutagen 10: 1-175.
http://dx.doi.org/10.1002/em. 2850100502
Gao. Z; Den. N \\un \\ U N .trig. S: Sun. C. (2014). Determination of organophosphorus flame
retardants in fish by pressurized liquid extraction using aqueous solutions and solid-phase
microextraction coupled with gas chromatography-flame photometric detector. J Chromatogr A
1366: 31-37. http://dx.doi.cnv 10 1016/i.chroma. JO I I 0° 028
Giannini. EG; Testa. R; Savarino. V. (2005). Liver enzyme alteration: A guide for clinicians [Review],

CMAJ 172: 367-379. http://dx.doi.on 10 I -03/cmai 10 10 ¦>_

Giorgino. MI; Rasmussen. RB; Pfeifle. CM. (2007). Occurrence of organic wastewater compounds in

selected surface-water supplies, Triangle Area of North Carolina, 2002-2005 (pp. 29). (Scientific
Investigations Report 2007-5054). U.S. Geological Survey.
http://dx.doi.orE	ir20075054

Grimes. G; Beaucham. C; Grant. M; Ramsey. J. (2019). Health hazard evaluation report: HHE-2016-
0257-3333, May 2019, evaluation of exposure to metals and flame retardants at an electronics
recycling company. Grimes, G; Beaucham, C; Grant, M; Ramsey, J.
https://www.cdc.gov/niosh/hhe/reports/pdfs/	)df

Gulati. DK; Russell. S. (1985). Tris(2-chloroethyl) phosphate: Sperm morphology vaginal cytology
evaluations in rodents. (Contract No. N01-ES-3-5026, Study No. SMVCE-82-001).
Environmental Health Research and Testing, Inc.

Guo. J; Romanak. K; Westenbroek. S; Hites. RA; Venier. M. (2017a). Current-use flame retardants in
the water of Lake Michigan tributaries. Environ Sci Technol 51: 9960-9969.
http://dx.doi.org/10J02 i/acs.est.7h0 L'M
G	er. M; Sal am ova. A; Hites. RA. (2017b). Bioaccumulation of Dechloranes,

organophosphate esters, and other flame retardants in Great Lakes fish. Sci Total Environ 583: 1-
9. http://dx.doi.oi ^ 10 101 S/i. sci totem _ 01 I I 0 '<

Guo. JH; Simon. K; Romanak. werman. W; Venier. M. (2018). Accumulation of flame retardants
in paired eggs and plasma of bald eagles. Environ Pollut 237: 499-507.
http://dx.doi.ore	.envpol.2018.02.056

Hahladat r < \ 'Hr t \ \\ ^Ivn k L. ;ovidou  \ 1 \ \ \ ang. R; Wau^ I o bang bang. O; Jiang. G. (2020). Occurrence and
distribution of organophosphate esters in the air and soils of Ny-Alesund and London Island,
Svalbard, Arctic. Environ Pollut 263: 114495. http://dx.doi.org/10.1016/i .envpol.2020.114495
H 1 (2018). Mechanisms of Rodent Renal Carcinogenesis Revisited [Review], Toxicol Pathol 46:
956-969. http://dx.doi.c	7/0192623318797071

Page 377 of 572

-------
10025

10026

10027

10028

10029

10030

10031

10032

10033

10034

10035

10036

10037

10038

10039

10040

10041

10042

10043

10044

10045

10046

10047

10048

10049

10050

10051

10052

10053

10054

10055

10056

10057

10058

10059

10060

10061

10062

10063

10064

10065

10066

10067

10068

10069

10070

10071

10072

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Harpi larding. A; Harris. S; Berger. P. (2012). Subsistence Exposure Scenarios for Tribal
Applications. Hum Ecol Risk Assess 18: 810-831.
http://dx.doi.orE 30/10807039.2012.688706
Haumant lumulla. J. (2002). Semi volatile organochemicals in indoor environment - Chlorinated
phosphorus and organotin compounds in material und house dust samples [Paper], In 9th
International Conference on Indoor Air Quality and Climate (Indoor Air 2002). Retrieved from
http s ://www. irbn et. de/daten/i conda/ C	£

Haworth. S: Lawlor. T; Mortelmans. K; Speck. W; Zeiger. E. (1983). Salmonella mutagenicity test

results for 250 chemicals. Environ Mutagen 5: 3-142. http://dx.doi.off 32/em.2860050703
Havnes. WM. (Ed.). (2014). Tris(2-chloroethyl) phosphate. In CRC handbook of chemistry and physics

(95 ed.). Boca Raton, FL: CRC Press. Taylor & Francis Group.

Hazleton Laboratories. (1983). Screening of priority chemicals for potential reproductive hazard (final
report) with attachments and cover sheet [TSCA Submission], (Hazleton Study No. 6125-101
through 6125-110. OTS0516205. 86-870001624. TSCATS/400290). Shell Oil Company.
https://ntrl.ntis. gov/NTRL/dashboard/searchResults.xhtml?searchOuerv=OTS0516205
He. C; Cova> i \ ileffemaii \1 r^iduel t i Ltrdeii i \ \Uiellei < oms. LML; Nele Van Den. E;
Hobson. P; Thai. P; Wang. X; Li. Y. (2018a). Urinary metabolites of organophosphate esters:
Concentrations and age trends in Australian children. Environ Int 111: 124-130.
http://dx.doi.ore 10 101 i.envint.-Oi I I 0rs
He. C; Wang. X; Tang. S: Phom i It ^ady^i t _ Vlueller. IF. (2018b). Concentrations of

Organophosphate Esters and Their Specific Metabolites in Food in Southeast Queensland,
Australia: Is Dietary Exposure an Important Pathway of Organophosphate Esters and Their
Metabolites? Environ Sci Technol 52: 12765-12773. http://dx.doi.org/10.102 l/acs.est.8b03043
He. Ml, \ am I Vang. ZH; Li. Q; Wei. SO. (2017). Occurrence and Distribution of Organophosphate
Esters in Surface Soil and Street Dust from Chongqing, China: Implications for Human
Exposure. Arch Environ Contam Toxicol 73: 349-361. http://dx.doi.ore/10.1007/

2	iane. P; Li. C; Cui. XY; Ma. LQ. (2018c). Impact of particle size on distribution and

human exposure of flame retardants in indoor dust. Environ Res 162: 166-172.
http://dx.doi.ore 10 101 i .envres. JO I IJ 01 I
Health Canada. (2020). Screening assessment - Phthalate substance grouping. (En 14-393/2019E-PDF).
Environment and Climate Change Canada, https://www.canada.ca/en/environment-climate-
change/services/evaluating-existing-substances/screening-assessment-phthalate-substance-
grouping.html

Heindel. II; Lamb JC. Cliapin. RE; Gulati. DK; Hope. E; George. J; Jameson	igue. J; Schwetz.

(1989). Reproductive toxicity testing by continuous breeding test protocol in Swiss (CD-I)
mice. (PB89152425; NTP-89054). Springfield, VA: U.S. Department of Commerce.
https://ntrl.ntis.gov/NTRL/dashboard/searchResiilts/titleDetail/PB:	A tint

Herr. DW; Sanders. JM; Matthews. HB. (1991). Brain distribution and fate of tris(2-chloroethyl)

phosphate in Fischer 344 rats. Drug Metab Dispos 19: 436-442.

Herrlich. T; Steib. C; Lang. A. (2013). [Flame-resistant coating for the rear side of a carpet] [Patent],

Clariant International Ltd. https://patents.google.com/patent/WQ2014026741 Al/en
H	(1965). The environment and disease: Association or causation? Proc R Soc Med 58: 295-300.

http://dx.doi.ore 10 I I 00	505800503

Hoffman. K; Lorenzo. A; Butt. CM; Hammel. SC; Henders	)man. SA; Scheri. RP; Stapleton.

HM; Sos (2017). Exposure to flame retardant chemicals and occurrence and severity of
papillary thyroid cancer: A case-control study. Environ Int 107: 235-242.
http://dx.doi.ore 10 101 i.envint..01 0> • 0J I

Page 378 of 572

-------
10073

10074

10075

10076

10077

10078

10079

10080

10081

10082

10083

10084

10085

10086

10087

10088

10089

10090

10091

10092

10093

10094

10095

10096

10097

10098

10099

10100

10101

10102

10103

10104

10105

10106

10107

10108

10109

10110

10111

10112

10113

10114

10115

10116

10117

10118

10119

10120

10121

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Hon. R; 1 ^ .o \ mi \ ha. J; Wane. Z. (2017). Accumulation and distribution of

organophosphate flame retardants (PFRs) and their di-alkyl phosphates (DAPs) metabolites in
different freshwater fish from locations around Beijing, China. Environ Pollut 229: 548-556.

http://dx.doi.ore	.emvpot.2017.06.097

H.SDB. (2015). Tris(2-chloroethyl) phosphate, CASRN: 1 15-96-8. U.S. Department of Health and
Human Services, National Institutes of Health, National Library of Medicine.
http://toxnet.nlm .nih.gov/cgi-bin/sis/search2/r?dbs+hsdb:@term+@DOC]

Hurtado. C; Montano-Chavez. YN; Dominguez. C; Bavona. J.M. (2017). Degradation of Emerging

Organic Contaminants in an Agricultural Soil: Decoupling Biotic and Abiotic Processes. Water
Air Soil Pollut228: 243-p. 243. http://dx.doi.ors 10 100 ^ I I. 0 01 '< |tL
Hutchins. SR.; Tom son. MB; Wilson. IT; Ward. CH. (1984). Fate of trace organics during rapid

infiltration of primary waste water at Fort Devens, Massachusetts (USA). Water Res 18: 1025-
1036. http://dx.doi.<	0255-0

U (1990). Tris(2-chloroethyl) phosphate [LARC Monograph], In I ARC Monographs on the
Evaluation of Carcinogenic Risks to Humans, vol 48 (pp. 109-120). Lyon, France.
https://piiblications.iarc.fr/Book-And-Report-Series/Iarc-Monoeraphs-On-The-Identification-Of-
Carcinoeenic-Hazards-To-Humans/Some-Flame-Retardants-And-Textile-Chemicals-And-
Exposures-In-T' tile-Manufacturine-Industrv-1990
'jj (2019). Report of the Advisory Group to Recommend Priorities for the I ARC Monographs
during 2020-2024. Geneva, Switzerland: World Health Organization.

https://www.iarc.who.int/news-events/report-of-the-advisorY-eroiip-to-recommend-priorities-for-
the-iarc-monographs-during-2020-2024/

ILO. (2019). International Chemical Safety Card: 1677. Tris(2-chloroethy 1) phosphate.

http://www.ilo.ore/dyn/icsc/showcard.displaY7p lang=en&p c	iv version=2

Ingerows	imulla. J. (2001). Chlorinated ethyl and isopropyl phosphoric acid triesters

in the indoor environment—an inter-laboratory exposure study. Indoor Air 11: 145-149.
http://dx.doi.ore 10 1034/i. 1600-0668.2001 01 100.'. i h \

Ion as U . 1 nt in U _ \mhonisseii < 'x eels. H; Covaci. A. (2014). Downsides of the recycling process:
Harmful organic chemicals in children's toys. Environ Int 65: 54-62.
http://dx.doi.ore 10 101 i.envint.-O1 ; I. 0rs
IPCS. (1998). Flame retardants: Tris(chloropropy 1) phosphate and tris(2-chloroethyl) phosphate.
(Environmental Health Criteria 209). Geneva, Switzerland: World Health Organization.
http: //www, in chem.org/ docum ents/ehc/eh c/eh c209. htm
Jackson. J; Sutton. R. (2008). Sources of endocrine-disrupting chemicals in urban wastewater, Oakland,

CA. Sci Total Environ 405: 153-160. http://dx.doi.o	/i.scitotenv.20Q8.Q6.033

Jayatilaka. NK; Restrepo. P; Williams. L; Ospina. M; Valentin-Blasin: ilafat. AM. (2017).
Quantification of three chlorinated dialkyl phosphates, diphenyl phosphate, 2,3,4,5-
tetrabromobenzoic acid, and four other organophosphates in human urine by solid phase
extraction-high performance liquid chromatography-tandem mass spectrometry. Anal Bioanal
Chem 409: 1323-1332. http://dx.doi.ore/lO 100 i00_l 01 00 I I
Ji. Y;Waiiv \ \ \ t Um i ,mg. X: Zhang. K: Sim \\ M.l > ^ ' vn. H. (2019).

Occurrence of organophosphate flame retardants in farmland soils from Northern China: Primary
source analysis and risk assessment. Environ Pollut 247: 832-838.
http://dx.doi.ore/10.1016/i .envpol.20.1 •> 01 036
Johnson. PD; Goldb ^ v < \ lavs. MZ; Dawson. BY (2003). Threshold of trichloroethylene

contamination in maternal drinking waters affecting fetal heart development in the rat. Environ
Health Perspect 111: 289-292. http://dx.doix ^ ^. - )/eht> > I _ >

Kademoglou. K; Xi' t ^ .dilla-Sanchez. J A; Haue. LS; Covaci \ i ollins. CD. (2017). Legacy and
alternative flame retardants in Norwegian and UK indoor environment: Implications of human

Page 379 of 572

-------
10122

10123

10124

10125

10126

10127

10128

10129

10130

10131

10132

10133

10134

10135

10136

10137

10138

10139

10140

10141

10142

10143

10144

10145

10146

10147

10148

10149

10150

10151

10152

10153

10154

10155

10156

10157

10158

10159

10160

10161

10162

10163

10164

10165

10166

10167

10168

10169

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

exposure via dust ingestion. Environ Int 102: 48-56.

http://dx.doi.ore 10 101 i.envint.^M l_

Kaiiwara. N: Noma. kigarni. H. (2011). Brominated and organophosphate flame retardants in
selected consumer products on the Japanese market in 2008. J Hazard Mater 192: 1250-1259.
http://dx.doi.ore	jhazmat.2011.06.043

Kawaeoshi. Y; Nakamura. S: Fukunaga. I. (2002). Degradation of organophosphoric esters in leachate
from a sea-based solid waste disposal site. Chemosphere 48: 219-225.
http://dx.doi.org/i 0. i 0 i 6/50045-6535(02)000 •» I
Kawashima. K; Tanaka. S: Nakaura. S: Nagao. S: En do oda. K; Takanaka. tori. Y. (1983).
[Effect of oral administration of tris(2-chloroethyl) phosphate to pregnant rats on prenatal and
postnatal development], Eisei Shikenjo Hokoku 101: 55-61.

Kerle ;hoenherr. J. (1988). Permeation of lipophilic chemicals across plant cuticles prediction
from partition coefficients and molar volumes. Arch Environ Contam Toxicol 17: 7-12.
http://dx.doi.ore

Kim. J; Isobt uto. M; Nguyen Minh. T; Katsura. K; Malarvannai idarvanto. A; Chang. KH;
Prudente. M; Pham Hu	ahashi. S: Tanabe. S. (2014). Organophosphorus flame

retardants (PFRs) in human breast milk from several Asian countries. Chemosphere 116: 91-97.
http://dx.doi.ore 10 101 i.chemosphere. 101 I 0. 0 '< '<

Kim. v ^ ^ h<> ( t nu IS; Vanderfonl •!%» Snyder. SA. (2007). Occurrence and removal of

pharmaceuticals and endocrine disruptors in South Korean surface, drinking, and waste waters.
Water Res 41: 1013-1021. http://dx.doi.o	--atres.2006.06.034

Kim. Ill; Kannan. K. (2018). Occurrence and distribution of organophosphate flame

retardants/plasticizers in surface waters, tap water, and rainwater: Implications for human
exposure. Environ Sci Technol 52: 5625-5633. http://dx.doi.org/10.1021/acs.est.8b00727
Kim. Ill; Oh. IK; Kannan. K. (2017). Occurrence, removal, and environmental emission of

organophosphate flame retardants/plasticizers in a wastewater treatment plant in New York
State. Environ Sci Technol 51: 7872-7880. http://dx.doi.org/10J02 l/acs.est.?b02035
Kim. Un; Jo. H; Lee. In	)h. J. (2015). Investigation of bioaccumulation and biotransformation

of polybrominated diphenyl ethers, hydroxylated and methoxylated derivatives in varying
trophic level freshwater fishes. Chemosphere 137: 108-114.
http://dx.doi.ore 10 101 i.chemosphere.-01 "i 0s 10 i
Koiadinovic. J; Potier.'	~e. M; Cosson. RP; Bustamante. P. (2007). Bioaccumulation of trace

elements in pelagic fish from the Western Indian Ocean. Environ Pollut 146: 548-566.
http://dx.doi.org/10J 016/i.envpol.200 0 01
Kotpi*> I UN gg. ET; Mever. MT; Thurman. EM; Zaugg. SO. Barber. LB; Buxton. H.T. (2002).
Pharmaceuticals, hormones, and other organic wastewater contaminants in US streams, 1999-
2000: A national reconnaissance. Environ Sci Technol 36: 1202-1211.
http://dx.doi.org/10J02 1/esOl 1055_i
Krivoshii	gmster. GTS; Sprangers. K; Blust. R; Husson. SJ. (2018). A toxicogenomics

approach to screen chlorinated flame retardants tris(2-chloroethyl) phosphate and tris(2-
chloroisopropyl) phosphate for potential health effects. J Appl Toxicol 38: 459-470.
http://dx.doi.org 10 I00_\j;u
Krivoshif r%\ 1 urdenne. F, Covad. A; Blust. R; Husson. SJ. (2016). Assessing in-vitro estrogenic
effects of currently-used flame retardants. Toxicol In Vitro 33: 153-162.
http://dx.doi.ore 10 101 i.tiv.^01 0 '< 006
Kurt-Karakus. P; Alegria. H; Birgul. A; Gungormus. E; Jan tun en. L. (2018). Organophosphate ester
(OPEs) flame retardants and plasticizers in air and soil from a highly industrialized city in
Turkey. Sci Total Environ 625: 555-565. http://dx.doi.oi ^ 10 101 >/i. sci totem JO I I J .'>07

Page 380 of 572

-------
10170

10171

10172

10173

10174

10175

10176

10177

10178

10179

10180

10181

10182

10183

10184

10185

10186

10187

10188

10189

10190

10191

10192

10193

10194

10195

10196

10197

10198

10199

10200

10201

10202

10203

10204

10205

10206

10207

10208

10209

10210

10211

10212

10213

10214

10215

10216

10217

10218

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

La Guardia. MI; Hale. RC. (2015). Halogenated flame-retardant concentrations in settled dust,

respirable and inhalable particulates and polyurethane foam at gymnastic training facilities and
residences. Environ Int 79: 106-114. http://dx.doi.oi	lenvint.2015.02.014

Lai. S: Xie. Z: Sons l' > iham \ Vli. W: Pens i k.o \ * < Unghaus. R. (2015V
Occurrence and dry deposition of organophosphate esters in atmospheric particles over the
northern South China Sea. Chemosphere 127: 195-200.
http: //dx. doi. or g/10.1016/i. ch em osph ere.:

Lannim 1 1 i reas's i liapin. RE; Mann. PC: Barlow. NJ: Regan. KS; Goodman. DG. (2002).
Recommended approaches for the evaluation of testicular and epididymal toxicity. Toxicol
Pathol 30: 507-520. http://dx.doi.org/i0J0SC m . . 0290105695
Lapwortl iaran. N: Stuart. ME; Ward. RS. (2012). Emerging organic contaminants in

groundwater: A review of sources, fate and occurrence [Review], Environ Pollut 163: 287-303.
http://dx.doi.org 10 101 | .envpol JO I I i J 034
Lee. E; Shon. HK; Cho. J. (2014). Role of wetland organic matters as photosensitizer for degradation of
micropollutants and metabolites. J Hazard Mater 276: 1-9.
http://dx.doi.ore	jhazmat.2014.05.001

Lee. RE. Jr.; Patterson. RK. (1969). Size determination of atmospheric phosphate, nitrate, chloride, and

ammonium particulate in several urban areas. Atmos Environ 3: 249-255.

Lee. S; Cho. HI; Choi. W; Moon. (2018). Organophosphate flame retardants (OPFRs) in water and
sediment: Occurrence, distribution, and hotspots of contamination of Lake Shihwa, Korea. Mar
Pollut Bull 130: 105-112. http://dx.doi.or	j .marpolbul.2018.03.009

Lehncr \ I . S;nnsii> t ^.umbeiha. WK. (2010). Organophosphate ester flame retardant-induced acute
intoxications in dogs. J Med Toxicol 6: 448-458. http://dx.doi.t	6-1

tnia. F. (2004). Is rain or snow a more efficient scavenger of organic chemicals? Atmos
Environ 38: 3557-3571. http://dx.doi.org/10.1016/i.atmosenv.2004.03.039
Lewis. RJ. Sr.; Hawlev. GG. (2007). Hawley's condensed chemical dictionary (15th ed.). Hoboken, NJ:

John Wiley & Sons, http://dx.doi.org/i 0. i 002/9780470114735
i i ( t lien. J; Xie. HB; Zh \ \t . I* i 1 t \ » H to. X. (2017a). Effects of Atmospheric Water
on OH-initiated Oxidation of Organophosphate Flame Retardants: A DFT Investigation on
TCPP. Environ Sci Technol 51: 5043-5051. http://dx.doi.org/10.1021/acs.est.7b00347

;. Z; Mi. W; Lai. S; Tian. C; Emeis. KC: Ebinghaus. R (2017b). Organophosphate esters in air,
snow, and seawater in the North Atlantic and the arctic. Environ Sci Technol 51: 6887-6896.
http://dx.doi.org/10J02 i/acs.est.7h01 „ v)

Li. M; Mathieu. J; Yam \ I torenza. S; Pern \ He. Z; Zhou j, Uvare (2013). Widespread

distribution of soluble di-iron monooxygenase (SDIMO) genes in Arctic groundwater impacted
by 1,4-dioxane. Environ Sci Technol 47: 9950-9958. http://dx.doi.org, ,l/es402228x
Li. M; Van Orden. ET; Devrie \unig. Z; Hincli^' K \L.tt x IJ (2015). Bench-scale

biodegradation tests to assess natural attenuation potential of 1,4-dioxane at three sites in
California. Biodegradation 26: 39-50. http://dx.doi.c
Li. TY; Bao. LI; Wu. CC; Liu. LY; Wong. CS; Zens (2019a). Organophosphate flame retardants
emitted from thermal treatment and open burning of e-waste. J Hazard Mater 367: 390-396.
http://dx.doi.org 10 101 t.ihazmat JO IN IJ 0 11
Li. W; Wans groan. K. (2019b). Occurrence, distribution and human exposure to 20

organophosphate esters in air, soil, pine needles, river water, and dust samples collected around
an airport in New York state, United States. Environ Int 131: 105054.
http://dx.doi.org 10 101 i.envint.-0 I" 10^054
Liang. K; Liu. J. (2016). Understanding the distribution, degradation and fate of organophosphate esters
in an advanced municipal sewage treatment plant based on mass flow and mass balance analysis.
Sci Total Environ 544: 262-270. http://dx.d0i.0rg/l 0.1016/i .scitotenv.^ > I * II II.

Page 381 of 572

-------
10219

10220

10221

10222

10223

10224

10225

10226

10227

10228

10229

10230

10231

10232

10233

10234

10235

10236

10237

10238

10239

10240

10241

10242

10243

10244

10245

10246

10247

10248

10249

10250

10251

10252

10253

10254

10255

10256

10257

10258

10259

10260

10261

10262

10263

10264

10265

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Life Sciences Research Ltd. (1990a). Fyrol CEF: Acute toxicity to rainbow trout [TSCA Submission],
(90/AKL021/0161. OTS0528355. 86-910000061. TSCATS/413338). Life Science Research Ltd.
Akzo Chemicals Inc.

https://ntrl.ntis.eov/NTRL/dashboard/searchResults/titleDetail/OTS0528355.xhtml
Life Sciences Research Ltd. (1990b). Fyrol CEF: Assessment of its ready biodegradability. Modified
Sturm test [TSCA Submission], (90/AKL023/0230. OTS0528355. 86-910000061.
TSCATS/413338). Akzo Chemicals Inc.

https://ntrl.ntis.eov/NTRL/dashboard/searchResiilts/titleDetail/OTSQ528355.xhtml
Life Sciences Research Ltd. (1990c). Fyrol FR-2: Assessment of its ready biodegradability. Modified
Sturm test [TSCA Submission], (90/AKL029/0232. OTS0528355. 86-910000061.
TSCATS/413338). Akzo Chemicals Inc.

https://ntrl.ntis.eov/NTRL/dashboard/searchResults/titleDetail/OTS0528355.xhtml
Lipworth (1996). Pharmacokinetics of inhaled drugs [Review], Br J Clin Pharmacol 42: 697-705.

http://dx.doi.ore/10.1046/i. 1365-2125.1996.00493 ,x
Liu 1 \ t Hit"- R\ 'vd am ova. A. (2016). Hair and nails as noninvasive biomarkers of human
exposure to brominated and organophosphate flame retardants. Environ Sci Technol 50: 3065-
3073. http://dx.doi.ore/10.102 l/acs.est.5b05073
Li	am ova. A; He. K; Hites. RA. (2015). Analysis of polybrominated diphenyl ethers and

emerging halogenated and organophosphate flame retardants in human hair and nails. J
Chromatogr A 1406: 251-257. http://dx.doi.oi^ 10 101 5/i.chroma.2015.06.003
Liu. Y; Liggio. J; Harner. T; Jan tun en. L; Shoeib. M; Li. SM. (2014). Heterogeneous OH initiated

oxidation: A possible explanation for the persistence of organophosphate flame retardants in air.
Environ Sci Technol 48: 1041-1048. http://dx.doi.ore/10.1021 /es4 ,

Liu. Y; L	ella. PZ; Zeng. Y; Mai. BH. (2019a). Organophosphorus flame retardants in a

typical freshwater food web: Bioaccumulation factors, tissue distribution, and trophic transfer.
Environ Pollut 255 Pt. 2: 113286. http://dx.doi.ore/10 101 >/i.envpol.2^ I II _ • 6
Liu. YE; Luo. XI; Huang. LQ; Zene. YH; M. (2019b). Organophosphorus flame retardants in fish
from Rivers in the Pearl River Delta, South China. Sci Total Environ 663: 125-132.
http://dx.doi.ore/10.1016/i .scitotenv.2019.01.344
Lown. JW; Josh	cLauehlin. LW. (1980). Novel antitumor nitrosoureas and related compounds

and their reactions with DNA. J Med Chem 23: 798-805.
http://dx.doi.ore	m00181a018

Luo i	1 i uo. Y; Li. S\l '^1 \ (2015). Size-dependent atmospheric deposition and

inhalation exposure of particle-bound organophosphate flame retardants. J Hazard Mater 301:
504-511. http://dx.doi.ore/10.1016/i jhazmat _01 0° 01 I
Ma. Y; Vojta. S: Becanova. J; Habtemichael. AZ; Adelman vluir. D; Lohmann. R (2021). Spatial
distribution and air-water exchange of organophosphate esters in the lower Great Lakes. Environ
Pollut 286: 117349. http://dx.doi.(	5/i.envpol.2021.117349

M	mann. R; Mi. W; Gao. G. (2017). Organophosphate ester flame retardants and

plasticizers in ocean sediments from the north pacific to the arctic ocean. Environ Sci Technol
51: 3809-3815. http://dx.doi.org/10.1021 /acs.est.7b(

Mack it ih c .rdo. \ l\»erson. S: Cowan. CE. (1996). Evaluating the environmental fate of a
variety of types of chemicals using the EQC model. Environ Toxicol Chem 15: 1627-1637.
http://dx.doi.ore/10.1002/etc.5620150929
Maddela. NR.; Venkateswarlu. K; Megharai. M. (2020). Tris(2-chloroethy 1) phosphate, a pervasive

flame retardant: Critical perspective on its emissions into the environment and human toxicity
[Review], Environ Sci Process Impacts 22: 1809-1827. http://dx.doi.ore/10.1039/d0em00222d

Page 382 of 572

-------
10266

10267

10268

10269

10270

10271

10272

10273

10274

10275

10276

10277

10278

10279

10280

10281

10282

10283

10284

10285

10286

10287

10288

10289

10290

10291

10292

10293

10294

10295

10296

10297

10298

10299

10300

10301

10302

10303

10304

10305

10306

10307

10308

10309

10310

10311

10312

10313

10314

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Mage. DT; Allen. RH; Kodali. A. (2008). Creatinine corrections for estimating children's and adult's

pesticide intake doses in equilibrium with urinary pesticide and creatinine concentrations. J Expo
Sci Environ Epidemiol 18: 360-368. http://dx.doi.ore/10.1038/si.ies.7500614

Makinen. MSE; Makinen. MR.A; Koistinen. JTB; Pasanen. AL; Pasanen. PO; Kalliokosl	pi.

AM. (2009). Respiratory and dermal exposure to organophosphorus flame retardants and
tetrabromobisphenol A at five work environments. Environ Sci Technol 43: 941-947.
http://dx.doi.ore/10J02 i/es802593t
Mansouri. K; Grulke. CM; Judson. RS; William; (2018). OPERA models for predicting
physicochemical properties and environmental fate endpoints. J Cheminform 10: 10.
http://dx.doi.ore/10.1186/s I ;; _ I 0 i s 0.63-1
Marklund. A; Anderssc [aelund. P. (2003). Screening of organophosphorus compounds and their
distribution in various indoor environments. Chemosphere 53: 1137-1146.
http://dx.doi.ore/10.1016/50045-6535(03)00666-0
Marklund \ \nderssoii t laulund. P. (2005a). Organophosphorus flame retardants and plasticizers
in Swedish sewage treatment plants. Environ Sci Technol 39: 7423-7429.
http://dx.doi.ore/10.102 l/ev0 101 H
Marklund. A; Anderssc [aelund. P. (2005b). Traffic as a source of organophosphorus flame
retardants and plasticizers in snow. Environ Sci Technol 39: 3555-3562.
http://dx.doi.ore/10.102 l/es0482177
Maronpot. RR; Nvsi j \ I«odder. NG; Seneupta. A; Smith MS 1 \ ons. JM; Heil. AT; Drewes. IE. (2016).

Multimedia screening of contaminants of emerging concern (CECS) in coastal urban watersheds
in southern California (USA). Environ Toxicol Chem 35: 1986-1994.
http://dx.doi.ore 32/etc.3348
Masoner. JR.; KoIhii 1 *\\ uidciiv { ; \ o :zarelli. IM; Gu\ (2016). Landfill leachate as a mirror
of today's disposable society: Pharmaceuticals and other contaminants of emerging concern in
final leachate from landfills in the conterminous United States. Environ Toxicol Chem 35: 906-
918. http://dx.doi.ore/10.1002/etc
Masoner. JR; Kolprii I *\\ Uttleitv \l ; \ o zarelli. IM; Gu\ " • hwab. I-A. (2014a). Contaminants
of emerging concern in fresh leachate from landfills in the conterminous United States. Environ
Sci Process Impacts 16: 2335-2354. http://dx.doi.orj	;4em00124a

Masoner. JR; Kolpin. DW; Furlong. ET; Cozzarelli. IM; Gray. XL; Schwab. EA. (2014b). Contaminants
of emerging concern in fresh leachate from landfills in the conterminous United States:
Supplementary information [Supplemental Data], Environ Sci Process Impacts 16.

Mato. "S . Ihobe. T; Takada. H; Kanehiro. H; Ohtake. C; Kaminuma. T. (2001). Plastic resin pellets as a
transport medium for toxic chemicals in the marine environment. Environ Sci Technol 35: 318-
324. http://dx.doi.ore/10.1021 /esOO 10498
Matsson. P; Kihlberg. J. (2017). How big is too big for cell permeability? J Med Chem 60: 1662-1664.

http://dx.doi.ore/10.102 l/acs.imedchem.7b00237
Matsukami. H; Nguyen Minh. T; Suzuki. Go; Someva. M, i e Huu I * bam Hung. V; Takahashi. S;
Tanabe. S; Takigami. H. (2015). Flame retardant emission from e-waste recycling operation in
northern Vietnam: Environmental occurrence of emerging organophosphorus esters used as
alternatives for PBDEs. Sci Total Environ 514: 492-499.
http://dx.doi.ore/10.1016/i.scitotenv.201 ^ 0J.008
Matthews. HB; Dixon. D; Hen	>on. H. (1990). Subchronic toxicity studies indicate that tris(2-

chloroethyl)phosphate administration results in lesions in the rat hippocampus. Toxicol Ind
Health 6: 1-15. http://dx.doi.oiv 10 I I '074823379000 00101

Page 383 of 572

-------
10315

10316

10317

10318

10319

10320

10321

10322

10323

10324

10325

10326

10327

10328

10329

10330

10331

10332

10333

10334

10335

10336

10337

10338

10339

10340

10341

10342

10343

10344

10345

10346

10347

10348

10349

10350

10351

10352

10353

10354

10355

10356

10357

10358

10359

10360

10361

10362

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Matthews. HB; Eustis. SL; Haseman. J. (1993). Toxicity and carcinogenicity of chronic exposure to
tris(2-chloroethyl)phosphate. Fundam Appl Toxicol 20: 477-485.

http://dx.doi.ore 36/faat. 1993.1058
Mave	t. KW: Chen. IC: Sammons. D; Toennis. C; Robertson. S: Kerber. S: Horn. GP; Smith.

PL; Calafat. AM; Ospina. M; Siodin. A. (2021). Characterizing exposures to flame retardants,
dioxins, and furans among firefighters responding to controlled residential fires. Int J Hyg
Environ Health 236: 113782. http://dx.doi.orv 10 101 'i.iiheh.-OJ I II ; 82
McDonoueh. CA; De Silva U* Sun. C; Cabrerizo. \ \ .0101007
Mihailovic. I; Fries. E. (2012). Atmospheric deposition of chlorinated organophosphate flame retardants
(OFR) onto soils. Atmos Environ 56: 177-183. http://dx.doi.or:	i.atmosenv.2012.03.054

Mihailovic. I; Miloradov. ? .es. E. (2011). Application of Twisselmann extraction, SPME, and
GC-MS to assess input sources for organophosphate esters into soil. Environ Sci Technol 45:
2264-2269. http://dx.doi.ore/10.1021/esl03870f
Mokra. K; Bukowski. K; Wozniak. K. (2018). Effects of tris( 1 -chloro-2-propyl )phosphate and tris(2-
chloroethyl)phosphate on cell viability and morphological changes in peripheral blood
mononuclear cells (in vitro study). Hum Exp Toxicol 37: 1336-1345.
http://dx.doi.ore 10 II 0960 I IN S3529
Moller. A; Sturm. R; Xie. Z; Cai. M; He. J; Ebinehaus. R (2012). Organophosphorus flame retardants

and plasticizers in airborne particles over the Northern Pacific and Indian Ocean toward the Polar
Regions: Evidence for global occurrence. Environ Sci Technol 46: 3127-3134.
http://dx.doi.ore/10.102 l/es204272v
Moller. A; Xie. Z; Cab; urm. R; Ebinehaus. R (2011). Organophosphorus flame retardants and
plasticizers in the atmosphere of the North Sea. Environ Pollut 159: 3660-3665.
http://dx.doi.off	.envpol.2011.07.022

Mose	lips. PM; Hedee. JM; McDaniel. KL. (2015). Neurotoxicological and thyroid

evaluations of rats developmentally exposed to tris(l,3-dichloro-2-propyl)phosphate (TDCIPP)
and tris(2-chloro-2-ethyl)phosphate (TCEP). Neurotoxicol Teratol 52: 236-247.
http://dx.doi.off 10 101 i.ntt.JO I "i 08.004
Muir. DCG. (1984). Phosphate esters. In J Schoer (Ed.), (pp. 41-66). Berlin, Germany: Springer-Verlag.
http://dx.doi.ore/10.1007/978-3-540-38819-7 2

Page 384 of 572

-------
10363

10364

10365

10366

10367

10368

10369

10370

10371

10372

10373

10374

10375

10376

10377

10378

10379

10380

10381

10382

10383

10384

10385

10386

10387

10388

10389

10390

10391

10392

10393

10394

10395

10396

10397

10398

10399

10400

10401

10402

10403

10404

10405

10406

10407

10408

10409

10410

10411

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

N . lion. C; Li. R; Sh> <. . . H; Jin. S; Gao "S . Jiao. L; Cai. Y. (2020). Occurrence, distribution,
air-seawater exchange and atmospheric deposition of organophosphate esters (OPEs) from the
Northwestern Pacific to the Arctic Ocean. Mar Pollut Bull 157: 111243.
http://dx.doi.org/10 J 016/i .marpolbul.2020 I I I J l'<

Nakamura. A; Tateno. N: Koiima. S: Kaniwa. MA; Kawamura. T. (1979). The mutagenicity of

halogenated alkanols and their phosphoric acid esters for Salmonella typhimurium. Mutat Res
66: 373-380. http://dx.dou	'9190048-X

NCBI. (2020). PubChem database: Compound summary: Tris(2-chloroethy 1) phosphate.

https://pubchem.ncbi.nlm.nih.gov/compound/Tris 2-chloroethyl-phosphate
NFPA. (2022). US Fire Department Profile 2020. Washington, DC.

NIC'NAS. (2001). Priority existing chemical assessment report no. 17: Trisphosphates. (PEC17).

Sydney, Australia. https://www.industrialchernicals.gov.au/sites/default/files/PEi

Trisphosphates.pdf

NIQSH. (2018). Health Hazard Evaluation Report: HHE-2015-0050-3308, May 2018. Evaluation of

Exposure to Metals, Flame Retardants, and Nanomaterials at an Electronics Recycling Company.
Beaucham, CC; Ceballos, D; Page, EH; Mueller, C; Calafat, AM; Sjodin, A; Ospina, M; La
Guardia, M; Glassford, E.

NIVA. (2008). Screening of selected metals and new organic contaminants 2007: Phosphorus flame

retardents, polyfluorinated organic compounds, nitro-PAHs, silver, platinum and sucralose in air,
wastewater treatment falcilities, and freshwater and marine recipients. (1014/2008; TA-
2367/2008). Oslo, Norway, https://niva.brage.unit.n0/niva-xmlui/handle/l 1250/213998
Noack. (1993). Unveroffentl. Unters. im auftrag der Hoechts AG [Unpublished investigation on behalf

of Hoechts AG], (Test No. AAE37081, Project No. 930818HH). Noack.

NTP. (1991a). Final report on the reproductive toxicity of tris(2-chloroethyl) phosphate in CD-I Swiss
mice. (RACB92040). Research Triangle Park, NC: National Institute of Environmental Health
Sciences, National Toxicology Program.

https://ntrl.ntis.gov/NTRL/dashboard/searchResiilts/titleDetail/PB92129170.xhtml
NTP. (1991b). Toxicology and carcinogenesis studies of tris(2-chloroethyl) phosphate (CAS No. 1 15-
96-8) in F344/N rats and B6C3F1 mice (gavage studies) [NTP] (pp. 1-233). (TR-391). Research
Triangle Park, NC: U.S. Department of Health and Human Services, National Institutes of
Health, National Toxicology Program.
https://search.proquest.com/docview/18593949697accoimti
NTP. (1992). CAMEO Chemicals Database of Hazardous Materials [Website],

http://cameochemicals.noaa. gov./

ODPHP. (2023a). Healthy People 2030 - Social determinants of health literature summaries:

Neighborhood and built environment [Website], https://health.gov/healthypeople/priority-
areas/social-determinants-health/literature-summaries#neighborhood
ODPHP. (2023b). Healthy People 2030 - Social determinants of health literature summaries: Poverty
[Website], https://health.gov/healthvpeople/prioritv-areas/social-determinants-health/literature-
summ aries/povertv

ODPHP. (2023c). Healthy People 2030 - Social determinants of health literature summaries: Social and
community context [Website], https://health.gov/healthvpeople/prioritv-areas/social-
determinants-health/literature-summaries#social
OECD. (2006). SIDS Initial Assessment Profile for SLAM 23: Tris(2-chloroethy 1 )phosphate (CAS no.
115-96-8) [OECD SIDS], Paris, France: Organization for Economic Co-operation and
Development. https://hpvchemicals.oecd.org/UI/handler.axd7id b I s dQ5-d( > I	1 *

603 a

OECD. (2007). Detailed review paper on cell transformation assays for detection of chemical

carcinogens. In Series on Testing and Assessment No 31. (ENV/JM/MONO(2007)18). Paris,

Page 385 of 572

-------
10412

10413

10414

10415

10416

10417

10418

10419

10420

10421

10422

10423

10424

10425

10426

10427

10428

10429

10430

10431

10432

10433

10434

10435

10436

10437

10438

10439

10440

10441

10442

10443

10444

10445

10446

10447

10448

10449

10450

10451

10452

10453

10454

10455

10456

10457

10458

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

France.

http://www.oecd. org/officialdocuments/publicdisplavdocumentpdf/?doclanguage=en&cote=env/

im/mono(2007)18

OECD. (2009). Emission scenario document on plastic additives. (JT03267870). Paris, France: OECD
Environmental Health and Safety Publications.

http://www.oecd.ore/officialdocuments/publicdisplavdocumentpdf7?cote=env/im/mono(2004)8/r
ev 1 &doclanguage=en

OECD. (2012). Test No. 305: Bioaccumulation in fish: Aqueous and dietary exposure. In

9789264185296. Paris, France: OECD Publishing, http://dx.doi.org/10.1787/9789264185296-en
OECD. (2016a). Test No. 473: In vitro mammalian chromosomal aberration test. Paris, France.

http://dx.doi.orE 37/9789264264649-en
OECD. (2016b). Test no. 474: Mammalian erythrocyte micronucleus test [Standard], In OECD
Guidelines for the Testing of Chemicals. Paris, France: OECD Publishing.
http://dx.doi.orE	>789264264762~en

OECD. (2017). Overview of the set of OECD Genetic Toxicology Test Guidelines and updates
performed in 2014-2015. Series on Testing & Assessment No. 238 - 2nd edition.
(ENV/JM/MONO(2016)33/REV1). Paris, France.
http://www.oecd.org/officialdocuments/publicdisplavdocumentpdf
OECD. (2022). Series on Testing & Assessment, No. 156: Guidance notes on dermal absorption studies
(Second edition). (ENV/JM/MONO(2011)36/REV1). Paris, France: Organisation for Economic
Co-operation and Development (OECD).

https://www.oecd.org/officialdocuments/publicdisplavdocumentpdf/7cote	;

MONO(! 5%20&doclanguage=en
Oken (2018). Examining the gas-particle partitioning of organophosphate esters: How reliable are
air measurements? Environ Sci Technol 52: 13834-13844.
http://dx.doi.orE	ics.est.8b04588

Oketp t4 >> ls y tiven. LV; Lorenzo. M; Dhal. S: Pico \ \iundale. VH; Diamond. M L (2018).

Polydimethylsiloxane (silicone rubber) brooch as a personal passive air sampler for semi-volatile
organic compounds. Chemosphere 208: 1002-1007.
http://dx.doi.or£ 10 101 |.chemosphere.101 \ 0 r\

Okerti N 1 ^ ns. TFM; Parnis. JM; Diamond. ML; Bidteman. TF: Jan tun en. LM. (2020). Gas
chromatographic estimation of vapor pressures and octanol-air partition coefficients of
semivolatile organic compounds of emerging concern. Journal of Chemical and Engineering
Data 65: 2467-2475. hti|- 
-------
10459

10460

10461

10462

10463

10464

10465

10466

10467

10468

10469

10470

10471

10472

10473

10474

10475

10476

10477

10478

10479

10480

10481

10482

10483

10484

10485

10486

10487

10488

10489

10490

10491

10492

10493

10494

10495

10496

10497

10498

10499

10500

10501

10502

10503

10504

10505

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Pane. L; Yai	o. JH; Zhang. HZ. (2016). Comparison of wastewater treatment processes on the

removal efficiency of organophosphate esters. Water Sci Technol 74: 1602-1609.

http://dx.doi. ore/10.2166/wst.2016.356
Park. H; Che	. IE. (2018). Evaluation of the current contamination status of PFASs and

OPFRs in South Korean tap water associated with its origin. Sci Total Environ 634: 1505-1512.
http://dx.doi.ore 10 101 |.scitotenv.201 \ 0 I 0 8
Peveih \\ \ \ enier. M; Rodenbm^ v ik. SN: Hombuckle. KC: Hites. RA. (2015).

Variations of flame retardant, polycyclic aromatic hydrocarbon, and pesticide concentrations in
Chicago's atmosphere measured using passive sampling. Environ Sci Technol 49: 5371-5379.
http://dx.doi.ore 10 I0JI acs.est.5luW I
Phillips. AL; Hammel. SC; He	enzo. AM; Chen. A; Webster. TF; Stapleton. HM. (2018).

Children's residential exposure to organophosphate ester flame retardants and plasticizers:
Investigating exposure pathways in the TESIE study. Environ Int 116: 176-185.
http://dx.doi.ore 10 101 i .envint..01 \ 0 I 01 '<

Poet. TS: Corlev. RA; Thrall t I* I >1\\ jrds. J A; Tanoio. H; Weitz. KK; Hui. X; Maibach. HI; Wester.
RC. (2000). Assessment of the percutaneous absorption of trichloroethylene in rats and humans
using MS/MS real-time breath analysis and physiologically based pharmacokinetic modeling.
Toxicol Sci 56: 61-72. http://dx.doi.ore/10.1093/toxsci/

Pom a v .les. C; Bmylami Christia. C; Goscinm c. \ on Loco. J; Covaci. A. (2018). Occurrence
of organophosphorus flame retardants and plasticizers (PFRs) in Belgian foodstuffs and
estimation of the dietary exposure of the adult population. Environ Sci Technol 52: 2331-2338.
http://dx.doi.ore/10J02 l/acs.est.7b06395
PPG. (2010). Safety data sheet: PITT-CHAR XP EP 97-194 component A. Borgerhout, Belgium: PPG
Coatings SPRL/BVBA.

PPG. (2016). Safety data sheet: PITT-CHAR XP part A off white. Pittsburgh, PA: PPG Industries, Inc.
Price. K; Haddad. S; Krishnan. K. (2003). Physiological modeling of age-specific changes in the

pharmacokinetics of organic chemicals in children. J Toxicol Environ Health A 66: 417-433.
http://dx.doi.orE 30/15287390306450
Prival. Ml; McCoy. EC; Gutte 3senkranz. HS. (1977). Tris(2,3-dibromopropyl) phosphate:
Mutagenicity of a widely used flame retardant. Science 195: 76-78.
http://dx.doi.on;	cience.318761

Propp. VR; De Silva. AO; Spencer. C; Brown. SI; Catinean. SD; Smith. JE; Roy. JW. (2021). Organic
contaminants of emerging concern in leachate of historic municipal landfills. Environ Pollut 276:
116474. http://dx.doi.ore/10 101 /i.envpol.20, i II I I
Qi. C; Yu. G; Zhone. M; Peng. G; Huane. J; Wane, B (2019). Organophosphate flame retardants in
leachates from six municipal landfills across China. Chemosphere 218: 836-844.
http://dx.doi.ore 10 101 i.chemosphere.101 s II 1^0
Raimondo. SDNV; Barron. MG. (2010). Web-Based Interspecies Correlation Estimation (Web-ICE) for
Acute Toxicity: User Manual Version 3.1. (600R10004). Raimondo, S., D.N. Vivian, and M.G.
Barron, http://nepis.epa.eov/exe/ZyPIJRL.cei?Dockey=P 10068ND.txt
Rantakokko. P; Kur	ber. J; Huane. T; Kiviranta. H; Cequier. E; Thomsen. C. (2019).

Concentrations of brominated and phosphorous flame retardants in Finnish house dust and
insights into children's exposure. Chemosphere 223: 99-107.
http://dx.doi.ore 10 101 i.chemosphere.-O rs 0J 0J
Rauert. C; Schuster. JK; Eng. A; Hamer. T. (2018). Global atmospheric concentrations of brominated
and chlorinated flame retardants and organophosphate esters. Environ Sci Technol 52: 2777-
2789. http://dx.doi.ore/10.102 l/acs.est.7b06239

Page 387 of 572

-------
10506

10507

10508

10509

10510

10511

10512

10513

10514

10515

10516

10517

10518

10519

10520

10521

10522

10523

10524

10525

10526

10527

10528

10529

10530

10531

10532

10533

10534

10535

10536

10537

10538

10539

10540

10541

10542

10543

10544

10545

10546

10547

10548

10549

10550

10551

10552

10553

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Reemtsm.t I' Keren i \1p it WVH; Gatimi I \ Knepper. TP; Neumann. M; Quintana. IB; de Vooet. P.
(2016). Mind the gap: Persistent and mobile organic compounds—Water contaminants that slip
through. Environ Sci Technol 50: 10308-10315. http://dx.doi.ore/10.1021 /acs.est.6b
Reers. AR: Erie. ML; Williams. TP; Elliott ,11'. Cox. ME; Beischlae. TV. (2016). The Flame-Retardant
Tris(l,3-dichloro-2-propyl) Phosphate Represses Androgen Signaling in Human Prostate Cancer
Cell Lines. J Biochem Mol Toxicol 30: 249-257. http://dx.doi.ore/10.1002/ibt.21786
Regtv	;tmann. W. (2009). Organophosphorus flame retardants and plasticizers in rain and snow

from middle Germany. CLEAN - Soil, Air, Water 37: 334-342.
http://dx.doi.ore/10.1002/clen.200900050
Reen	mann. W. (2010a). Occurrence and fate of organophosphorus flame retardants and

plasticizers in urban and remote surface waters in Germany. Water Res 44: 4097-4104.
http://dx.doi.ore 10 101 i.watres.JO 10 0^ 0J I
Reen-	mann. W. (2010b). Seasonal fluctuations of organophosphate concentrations in

precipitation and storm water runoff. Chemosphere 78: 958-964.
http://dx.doi.ore	.chemosphere.2009.12.027

Ren. G; Chu. X; Zhang. J; Zheng. K; Zhou. X; Zene. X; Yu. Z. (2019). Organophosphate esters in the
water, sediments, surface soils, and tree bark surrounding a manufacturing plant in north China.
Environ Pollut 246: 374-380. http://dx.doi.oi> 10 101 'i.envpol l - l _

Ren. X; Han. HI; Lee. YJ; Lee. SI	ae. KJ; Kim. IS. (2012). Proapoptotic effect of a

micropollutant (tris-(2-chloroethyl)-phosphate) at environmental level in primary cultured renal
proximal tubule cells. J Water Health 10: 522-530. http://dx.doi.or	wh.2012.180

R	in. HI; Kim. IS. (2008). Effect of tris-(2-chloroethy 1 )-phosphate (TCEP) at

environmental concentration on the levels of cell cycle regulatory protein expression in primary
cultured rabbit renal proximal tubule cells. Chemosphere 74: 84-88.
http://dx.doi.ore/10.1016/i .chemosphere.2008.09.022
R	i. IS. (2009). Cellular effect evaluation of micropollutants using

transporter functions of renal proximal tubule cells. Chemosphere 77: 968-974.
http://dx.doi.ore 10 101 i .chemosphere.2009.OK 01J
Reynold?	>n. KA. (1984). T gamma-lymphoproliferative disease and related disorders in

humans and experimental animals: A review of the clinical, cellular, and functional
characteristics [Review], Blood 64: 1146-1158.
http://dx.doi.ore 10 I 182/bloo
-------
10554

10555

10556

10557

10558

10559

10560

10561

10562

10563

10564

10565

10566

10567

10568

10569

10570

10571

10572

10573

10574

10575

10576

10577

10578

10579

10580

10581

10582

10583

10584

10585

10586

10587

10588

10589

10590

10591

10592

10593

10594

10595

10596

10597

10598

10599

10600

10601

10602

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Sal a. M; Gu. ZG; Moen houroulinkov. I. (1982). In vivo and in vitro biological effects of the

flame retardants tris(2,3-dibromopropyl) phosphate and tris(2-chlorethyl)orthophosphate. Eur J
Cancer Clin Oncol 18: 1337-1344. http://dx.doij	277-5379(82190138-9

Sal am ova \. Mb V \ enier. M; Hites. RA. (2014). High levels of organophosphate flame retardants in
the great lakes atmosphere. Environ Sci Technol Lett 1: 8-14.
http://dx.doi.ore/10.102 l/ez400034n
Sal am ova \ Poverh \\ \ onier. M; Hites. K \ (2016). Spatial and temporal trends of particle phase
organophosphate ester concentrations in the atmosphere of the great lakes. Environ Sci Technol
50: 13249-13255. http://dx.doi.org/10.102 l/acs.est.6b04789
Sasaki. K; Suzul	da. M; Uchiyama. M. (1982). Bioconcentration and excretion of phosphoric

acid triesters by killifish (Oryzeas latipes). Bull Environ Contam Toxicol 28: 752-759.
http://dx.doi.ore	5648

Schenk. A; Ghallab. A; Hofmann. U; Hassan. R; Schwarz. M; Schuppi Schwen. LO; Braeuning.

nico. D; Henestler. JG; Kuepfer. L. (2017). Physiologically-based modelling in mice
suggests an aggravated loss of clearance capacity after toxic liver damage. Sci Rep 7: 6224.
http://dx.doi.ore 10 10 'A -415^\ 01 0 I ¦> I
SCHER. (2012). Opinion on tris(2-chloroethy 1 )phosphate (TCEP) in toys. European Commission
Scientific Committee on Health and Environmental Risks.
http://ec.eiiropa.eu/health/scientific committees/environmental risks/doc s/scher
Schreder. ED; La Guardia. Ml. (2014). Flame retardant transfers from U.S. households (dust and
laundry wastewater) to the aquatic environment. Environ Sci Technol 48: 11575-11583.
http://dx.doi.ore/10.102 l/es502227h
Schreder. ED; Udine. N: La Guardia. Ml. (2016). Inhalation a significant exposure route for chlorinated
organophosphate flame retardants. Chemosphere 150: 499-504.
http://dx.doi.ore/10.1016/i .chemosphere.lO I \ \ 0S4
Scott. Hi:'. Sverko. E; Maeim-' M (1996). Determination of benzothiazole and alkylphosphates in water
samples from the Great Lakes drainage basin by gas chromatography/atomic emission detection.
Water Qual Res J Can 31: 341-360. http://dx.doi.ore/10.2166/wqij. 1996.021
Selerade. MK; Gilmour. MI. (2010). Suppression of pulmonary host defenses and enhanced

susceptibility to respiratory bacterial infection in mice following inhalation exposure to
trichloroethylene and chloroform. J Immunotoxicol 7: 350-356.
http://dx.doi.ore/10.3109/1 •» i \ 1010 ->.01 ^

Sengupta ons. JM; Smitl	ves. IE; Snyder. SA; Heil. A; Maruva. KA. (2014). The

occurrence and fate of chemicals of emerging concern in coastal urban rivers receiving discharge
of treated municipal wastewater effluent. Environ Toxicol Chem 33: 350-358.
http: //dx. doi. or e/10.1002/etc.245 7
Sharma. R; Sharma. S. (2023). StatPearls: Physiology, blood volume. Treasure Island, FL: StatPearls

Publishing. https://www.ncbi.nlm.nih.eov/books/NBK526077/

Sharn	tad. R. (2014). Alkaline phosphatase: An overview [Review], Indian J Clin

Biochem 29: 269-278. http://dx.doi.ore/10.1007/sl2	^

Shen. B; Whitehead. TP; Gill. R; Dha	'own. FR; Petreas. M; Patton. S; Hammond. SK. (2018).

Organophosphate flame retardants in dust collected from United States fire stations. Environ Int
112: 41-48. http://dx.doi.orv 10 101 1 en vim JO I I J 00°

Shepel'skaia. NR; Dyshginevich. NE. (1981). [Experimental study of the gonadotoxic effect of tri-

(chloroethyl)-phosphate]. Gig Sanit20-21.

Shin. HM; McKone. TE; Nishioka. MG; Fallin. MP; Croen. LA; Hertz-Picciotto. I; Newschaffer. CI;
Benn . (2014). Determining source strength of semi volatile organic compounds using
measured concentrations in indoor dust. Indoor Air 24: 260-271.
http://dx.doi.ore	ia. 12070

Page 389 of 572

-------
10603

10604

10605

10606

10607

10608

10609

10610

10611

10612

10613

10614

10615

10616

10617

10618

10619

10620

10621

10622

10623

10624

10625

10626

10627

10628

10629

10630

10631

10632

10633

10634

10635

10636

10637

10638

10639

10640

10641

10642

10643

10644

10645

10646

10647

10648

10649

10650

10651

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Shoeib. M; Ahrens. L; Jan tun en. L; Harner. T. (2014). Concentrations in air of organobromine,

organochlorine and organophosphate flame retardants in Toronto, Canada. Atmos Environ 99:
140-147. http://dx.doi.ofe/' 3/i.atmosenv.2014.09.040
Sigma-Aldrich. (2019). Tris(2-chloroethy 1) phosphate. Sigma-Aldrich Inc.

https://www.siemaaldrich.eom/cataloe/product/aldrich/l 19660?lang=en®ion=US
Simmon. auhanen. K; Tardiff. RG. (1977). Mutagenic activity of chemicals identified in drinking
water. In D Scott; B Bridges; F Sobel (Eds.), Developments in Toxicology and Environmental
Science, 2 (pp. 249-258). New York, NY: Elsevier/North Holland Press.

Siodin. A; Carls son. H; Thuresson. K; Sjolin. S: Bergman. A; Ostman. C. (2001). Flame retardants in

indoor air at an electronics recycling plant and at other work environments. Environ Sci Technol
35: 448-454. http://dx.doi.c	000077n

Smyth. HF. Jr.: Carpenter. CP; Weil. CS. (195 1). Range-finding toxicity data: List IV. AM A Arch Ind

Hyg Occup Med 4: 119-122.

Snyder. SA; Wert. EC; Rexing O] 'egei1 ' : I '!.:. (2006). Ozone Oxidation of Endocrine
Disruptors and Pharmaceuticals in Surface Water and Wastewater. Ozone: Science and
Engineering 28: 445-460. http://dx.doi.ore/10.1080/	>01039726

Sonich-Mullin. C; Fielder. R; Wilt^-c i, Baetcke. t 1 ^'mpsey. J; Fenner-Crisp. P; Grant. D; Hartley. M;
Knaai' \ i n*'se. 1); Mangelsdorf. I; Meek. E; Rice. JM N »nines. M. (2001). IPCS conceptual
framework for evaluating a mode of action for chemical carcinogenesis. Regul Toxicol
Pharmacol 34: 146-152. http://dx.doi.	!6/rtph.2001.1493

Stachel. B; Jantzen. E; Knoth. W; Kruger. iom. P; Oetken. M; Reincke. H; Sawal. G; Schwartz. R;
Uhlig. S. (2005). The Elbe Flood in August 2002—Organic Contaminants in Sediment Samples
Taken After the Flood Event. J Environ Sci Health A Tox Hazard Subst Environ Eng 40: 265-
287. http://dx.doi.org/10.1081/ESE-2I
Stackelbc	rlong. ET; Meyer. MT; Zaugg. SD; Henderson. AK; and Reissman. DB. (2004).

Persistence of pharmaceuticals compounds and other organic wastewater contaminants in a
conventional drinking-water-treatment plant. Sci Total Environ 329: 99-113.
http://dx.doi.ore	scitotenv.2004.03.015

Stapleton. HM; Klosterhaus. S; Keller rguson. P; van Bergen. S; Cooper. E; Webster. TF; Blum.
A. (2011). Identification of Flame Retardants in Polyurethane Foam Collected from Baby
Products. Environ Sci Technol 45: 5323-533 1. http://dx.doi.on	;s2007462

Stapleton. HM; Sharma. S; Getzinger. u. Tciguson. P; Gabriel. M; Websfrn i1	A. (2012).

Novel and High Volume Use Flame Retardants in US Couches Reflective of the 2005 PentaBDE
Phase Out. Environ Sci Technol 46: 13432-13439. http://dx.doi.org/10.1021/es303471d
Stubbing	o. JH; Simon. K; Romanak. K; Bowerman. W; Venier. M. (2018). Flame retardant

metabolites in addled bald eagle eggs from the Great Lakes region. Environ Sci Technol Lett 5:
354-359. http://dx.doi.org/10.1021/acs.estlett.8b00163
Si ;tcher. RJ; Yu. H. (2016). Organophosphate flame retardants and plasticizers in aqueous

solution: pH-dependent hydrolysis, kinetics, and pathways. Environ Sci Technol 50: 8103-8111.
http://dx.doi.org/10J02 i/acs.est.6b02187
Silhring. R; Diamond. ML; Scheringer. M; Wong I', lVcko. M; Sfrn 11 _ ^; Mum \ I king. H; Fellin. P;
Li. H; J an tun en. LM. (2016). Organophosphate esters in Canadian Arctic air: Occurrence, levels
and trends. Environ Sci Technol 50: 7409. http://dx.doi.org/10.102 l/acs.est.6b00365
Sun. L; Tan. H; Peng. T; Wang. S; Xu. W; Oian.	(2016). Developmental neurotoxicity

of organophosphate flame retardants in early life stages of Japanese medaka (Oryzias latipes).
Environ Toxicol Chem 35: 293 1-2940. http://dx.doi.org/10.1002/etc.3477
Sundkvist. AM; Olofsson. LI; Haglund. P. (2010). Organophosphorus flame retardants and plasticizers in
marine and fresh water biota and in human milk. J Environ Monit 12: 943-951.
http: //dx. doi. or g/10.103 9/b 921910b

Page 390 of 572

-------
10652

10653

10654

10655

10656

10657

10658

10659

10660

10661

10662

10663

10664

10665

10666

10667

10668

10669

10670

10671

10672

10673

10674

10675

10676

10677

10678

10679

10680

10681

10682

10683

10684

10685

10686

10687

10688

10689

10690

10691

10692

10693

10694

10695

10696

10697

10698

10699

10700

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Sutton. R; Chen. D; Sui> 1 K'ig Oi.Wu. Y. (2019). Characterization of brominated, chlorinated, and
phosphate flame retardants in San Francisco Bay, an urban estuary. Sci Total Environ 652: 212-
223. http://dx.doi.org/10.1016/i .scitotenv.20.1 S 10 096
Takada. H; Karapanagioti. H. (2019). Hazardous chemicals associated with plastics in the marine

environment. In The Handbook of Environmental Chemistry (HEC), vol 78. Cham, Switzerland:
Springer, http://dx.doi.org/10.1007/978-3-319-95568-1
Takada. K; Yasuhara. K; Nakaii \ N >^ Uimoto. H; Momm.t < Uurokawa. Y; Ahli \ I'che. M.

(1989). [Carcinogenicity study of tris (2-chloroethyl) phosphate in ddY mice], J Toxicol Pathol
2: 213-222. http://dx.doi.Org/10.1293/tox.2.213
Tarookou. J; Kuete. V. (2014). Chapter 10: Mutagenicity and carcinogenicity of African medicinal
plants. In V Kuete (Ed.), (pp. 277-322). Amsterdam, Netherlands: Elsevier.
http://dx.doi.ore hM r 8-0-12-800018-2.00010-8
Tan. I1) \ jnv 1 \\i \ uan. Q; Liu. X; Li. L; Chen. I ) (2019). Co-existence of organophosphate di-
and tri-esters in house dust from South China and Midwestern United States: Implications for
human exposure. Environ Sci Technol 53: 4784-4793. http://dx.doi.org/10.1021 /acs.est.9b00229
Tang H, Poena. G; Bastiaensen. M; Yin. SS: Luo. XI; M o \ • ovaci. A. (2019). Bioconcentration
and biotransformation of organophosphorus flame retardants (PFRs) in common carp (Cyprinus
carpio). Environ Int 126: 512-522. http://dx.doi.org/10 J 016/i.envint.201'" 0. 0 3
Taniai. E; Havashi.	line. A; Watanabe. M; Akane. H; Suzuki. K; Mitsumori. K; Shibutani. M.

(2012a). Cellular distribution of cell cycle-related molecules in the renal tubules of rats treated
with renal carcinogens for 28 days: Relationship between cell cycle aberration and
carcinogenesis. Arch Toxicol 86: 1453-1464. http://dx.doi.org/10.1007/s0020 I 0 i J 0829-z
Taniai. E; Yafune. A; Havashi. H; Itahashi. M; Hara-Kudo. Y; Suzuki. K; Mitsumori. K; Shibutani. M.
(2012b). Aberrant activation of ubiquitin D at G2 phase and apoptosis by carcinogens that evoke
cell proliferation after 28-day administration in rats. J Toxicol Sci 37: 1093-1 111.
http://dx.doi.ore	ts. 37.1093

Tj ;rica. (2018). Safety Data Sheet: Tris(2-chloroethy 1) phosphate, P0268. Portland, OR.

TERA. (2013). Toxicity review of tris(2-chloroethyl) phosphate (TCEP). U.S. Consumer Product Safety
Commission, Directorate for Hazard Identification and Reduction.

https://web.archive.Org/web/20190320055405/https://www. cpsc.gov/s3fs-public/pdfs/TCEP-
contract-report-with -cover-letter, pdf
TERA. (2015). Environmental concentrations and consumer exposure data for tris(2-chloroethyl)

phosphate (TCEP). Cincinnati, OH: Toxicology Excellence for Risk Assessment Center at the
University of Cincinnati.

https://web.archive.org/web/20170207003923/https://www.cpsc. gov/s3fs-
public/pdfs/CPSCStaffStatementToxicologyExcellenceRiskAssessmentsReportExposureDataTC

EP.pdf

Thomas. J; Haseman. JK; Goodman. II; Ward. JM; Loughran. TP. Jr.; Spencc (2007). A review of
large granular lymphocytic leukemia in Fischer 344 rats as an initial step toward evaluating the
Implication of the endpoint to human cancer risk assessment [Review], Toxicol Sci 99: 3-19.
http://dx.doi.org	oxsci/kfm098

Tilson. HA; Verone feLamb. RL; Matthews. HB. (1990). Acute exposure to tris(2-

chloroethyl)phosphate produces hippocampal neuronal loss and impairs learning in rats. Toxicol
Appl Pharmacol 106: 254-269. http://dx.doij nv i o i o t < \)4i-003X(90)90245-P
Toscano. WA; Coleman. KP. (2012). Patty's Toxicology: Esters of carbonic and orthocarbonic acid,
organic phosphorous, monocarboxylic halogenated acids, haloalcohols, and organic silicon (6
ed.). New York, NY: John Wiley & Sons.

L. (2016). May 2016 Occupational Employment and Wage Estimates: National Industry-
Specific Estimates [Website], http://www.bls.gov/oes/tables.htm

Page 391 of 572

-------
10701

10702

10703

10704

10705

10706

10707

10708

10709

10710

10711

10712

10713

10714

10715

10716

10717

10718

10719

10720

10721

10722

10723

10724

10725

10726

10727

10728

10729

10730

10731

10732

10733

10734

10735

10736

10737

10738

10739

10740

10741

10742

10743

10744

10745

10746

10747

10748

10749

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

sus Bureau. (2015). Statistics of U.S. Businesses (SUSB).
https://www.census.eov/data/tables/2015/econ/susb/2015-susb-annual.html

(1987).	Household solvent products: A national usage survey. (EPA-OTS 560/5-87-005).
Washington, DC: Office of Toxic Substances, Office of Pesticides and Toxic Substances.
https://ntrl.ntis. eov/NTKL/dashboard/ searchResults.xhtml?searchQuery=PB S |i

(1988).	National survey of solid waste (municipal) landfill facilities [EPA Report],
(EPA/530-SW88-034). Washington, DC: Office of Solid Waste and Emergency Response, U.S.
Environmental Protection Agency.

(1991).	Guidelines for developmental toxicity risk assessment. Fed Reg 56: 63798-63826.

(1992).	Guidelines for exposure assessment. Federal Register 57(104):22888-22938 [EPA
Report], (EPA/600/Z-92/001). Washington, DC.
http://cfpiib.epa.eov/ncea/cfm/recordisplay.cfm7deii 3

(1993a). Reference Dose (RfD): description and use in health risk assessments background
document 1A, March 15, 1993. Washington, DC: U.S. Environmental Protection Agency,
Integrated Risk Information System, https://www.epa.gov/iris/reference-dose-rfd-description-
and-use-health-risk-assessments
U.S. EPA. (1993b). Wildlife exposure factors handbook [EPA Report], (EPA/600/R-93/187).
Washington, DC: Office of Research and Development.
http://cfpub.epa. eov/ncea/cfm/recordisplav.cfm?deid=2799

(1996). Guidelines for reproductive toxicity risk assessment [EPA Report], (EPA/630/R-
96/009). Washington, DC: U.S. Environmental Protection Agency, Risk Assessment Forum.
https://nepis.epa. eov/Exe/ZyPURL.cei?Dockev=3 0004YQB.txt

(1998a). Assessment of thyroid follicular cell tumors [EPA Report] (pp. 1-51). (EPA/630/R-
97/002). Washington, DC: U.S. Environmental Protection Agency, Risk Assessment Forum.
https://www.epa.eov/sites/prodiiction/files/2i 'docum ents/thyroi d.pdf
1	\ (1998b). Guidelines for ecological risk assessment [EPA Report], (EPA/630/R-95/002F).

Washington, DC: U.S. Environmental Protection Agency, Risk Assessment Forum.
https://www.epa.eov/risk/euidelines-ecoloeical-risk-assessment
U.S. EPA. (2000a). Interim guidance for using ready and inherent biodegradability tests to derive input
data for multimedia models and wastewater treatment plants (WWT) models (9/1/2000).
Washington, DC. https://www.epa.gov/tsca-screening-tools/interim-guidance-using-readv-and-
inherent-biodegradabilitv-tests-derive-0
U.S. EPA. (2000b). Methodology for deriving ambient water quality criteria for the protection of human
health (2000). (EPA/822/B-00/004). Washington, DC: U.S. Environmental Protection Agency,
Office of Water, http://www.epa.gov/waterscience/criteria/humanhealth/method/complete.pdf
U.S. EPA. (2000c). Supplementary guidance for conducting health risk assessment of chemical mixtures
(pp. 1-209). (EPA/630/R-00/002). Washington, DC: U.S. Environmental Protection Agency,

Risk Assessment Forum, http://cfpub.epa.eov/ncea/cfm/recordisplav.cfm?deid=20533
U.S. EPA. (2000d). The use of data on cholinesterase inhibition for risk assessments of

organophosphorous and carbamate pesticides. Washington, DC: U.S. Environmental Protection
Agency, Office of Pesticide Programs, https://www.epa.gov/sites/default/files/2015-
07/docum ents/ cholin. pdf

U.S. EPA. (2002a). Guidance on cumulative risk assessment of pesticide chemicals that have a common
mechanism of toxicity [EPA Report], Washington, D.C.

(2002b). A review of the reference dose and reference concentration processes.
(EPA630P02002F). Washington, DC. https://www.epa.gov/sites/production/files/2014-

)cum ents/ rfd-final.p df

U.S. EPA. (2003a). Attachment 1-3 Guidance for Developing Ecological Soil Screening Levels (Eco-
SSLs): Evaluation of Dermal Contact and Inhalation Exposure Pathways for the Purpose of

Page 392 of 572

-------
10750

10751

10752

10753

10754

10755

10756

10757

10758

10759

10760

10761

10762

10763

10764

10765

10766

10767

10768

10769

10770

10771

10772

10773

10774

10775

10776

10777

10778

10779

10780

10781

10782

10783

10784

10785

10786

10787

10788

10789

10790

10791

10792

10793

10794

10795

10796

10797

10798

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Setting Eco-SSLs. (OSWER9285755E). Washington, DC: .S. Environmental Protection Agency,
Office of Solid Waste and Emergency Response.

https://www.epa.eov/sites/prodiiction/files/2015-09/dociiments/ecossl attachment l-3.pdf
U.S. EPA. (2003b). Attachment 1 -4. Guidance for developing ecological soil screening levels (Eco-
SSLs): Review of background concentration for metals. (OSWER Directive 92857-55).
Washington, DC. https://www.epa.gov/sites/default/files/2Q15-
09/docum ents/ecossl attachmt	f

U.S. EPA. (2003c). Methodology for deriving ambient water quality criteria for the protection of human
health (2000), technical support document. Volume 2: Development of national bioaccumulation
factors [EPA Report], (EPA/822/R-03/030). Washington, DC.
http://www.epa.eov/waterscience/criteria/humanhealth/method/tsdvol2.pdf
U.S. EPA. (2005a). Guidance for developing ecological soil screening levels [EPA Report], (OSWER
Directive 92857-55). Washington, DC: U.S. Environmental Protection Agency, Office of Solid
Waste and Emergency Response, http://www.epa.gov/chemical-research/guidance-developing-
ecoloeical-soil-screenine-levels
U.S. EPA. (2005b). Guidelines for carcinogen risk assessment [EPA Report], (EPA630P0300IF).
Washington, DC. https://www.epa.gov/sites/production/files/2013-
09/documents/cancer guidelines final 3-25-Q5.pdf
U.S. EPA. (2005c). Supplemental guidance for assessing susceptibility from early-life exposure to
carcinogens [EPA Report], (EPA/630/R-03/003F). Washington, DC: U.S. Environmental
Protection Agency, Risk Assessment Forum, https://www.epa.gov/risk/supplemental-guidance-
assessine-susceptibilitv-early-life-exposure-carcinoeens
U.S. EPA. (2007a). Attachment 4-3 Guidance for Developing Ecological Soil Screening Levels (Eco-
SSLs) Eco-SSL Standard Operating Procedure (SOP) #4: Wildlife Toxicity Reference Value
Literature Review, Data Extraction and Coding. (OSWER9285755F).
http://nepis.epa.gov/exe/ZyPURL.cei?Dockev=P 100CDHC.txt
U.S. EPA. (2009). Provisional Peer-Reviewed Toxicity Values for tris(2-chloroethyl)phosphate (TCEP)
(CASRN 115-96-8) [EPA Report], (EPA/690/R-09/069F). Cincinnati, OH.
http://hhpprtv.ornl.eov/issue papers/Tris2chloroethylphosphate.pdf
U.S. EPA. (201 la). Exposure factors handbook: 201 1 edition [EPA Report], (EPA/600/R-090/052F).

Washington, DC: U.S. Environmental Protection Agency, Office of Research and Development,
National Center for Environmental Assessment.
https://nepis.epa. eov/Exe/ZyPURL.cei?Dockey=P 100F2QS.txt
U.S. EPA. (201 lb). Exposure factors handbook: 201 1 edition (final) [EPA Report], (EPA/600/R-

090/052F). Washington, DC: U.S. Environmental Protection Agency, Office of Research and
Development, National Center for Environmental Assessment.
http://cfpub.epa. gov/ncea/cfm/recordisplav.cfm?deid=236252
U.S. EPA. (201 lc). Recommended Use of Body Weight 3/4 as the Default Method in Derivation of the
Oral Reference Dose.

(2012a). Advances in inhalation gas dosimetry for derivation of a reference concentration
(RfC) and use in risk assessment (pp. 1-140). (EPA/600/R-12/044). Washington, DC.

https://cfpi.ib.epa.gov/ncea/risk/recordisplay.cfm?deid=244650> \ t 11 * 0.1 „*, CFTOKEN=
17139189

U.S. EPA. (2012b). Benchmark dose technical guidance [EPA Report], (EPA 100R12001). Washington,
DC: U.S. Environmental Protection Agency, Risk Assessment Forum.
https://www.epa.gov/risk/benchmark-dose-technical-guidance
U.S. EPA. (2012c). Estimation Programs Interface Suite™ for Microsoft® Windows, v 4.1 1 [Computer
Program], Washington, DC. Retrieved from https://www.epa.gov/tsca-screening-tools/epi-
suitetm-estimation-program-interface

Page 393 of 572

-------
10799

10800

10801

10802

10803

10804

10805

10806

10807

10808

10809

10810

10811

10812

10813

10814

10815

10816

10817

10818

10819

10820

10821

10822

10823

10824

10825

10826

10827

10828

10829

10830

10831

10832

10833

10834

10835

10836

10837

10838

10839

10840

10841

10842

10843

10844

10845

10846

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

U.S. EPA. (2014a). Estimated fish consumption rates for the U.S. population and selected

subpopulations (NHANES 2003-2010) [EPA Report], (EPA-820-R-14-002). Washington, DC.

https://www.epa.eov/sites/prodiiction/files/2015-01/dociiments/fish-consiimption-rates-! if
U.S. EPA. (2014b). Framework for human health risk assessment to inform decision making. Final
[EPA Report], (EPA/100/R-14/001). Washington, DC: U.S. Environmental Protection, Risk
Assessment Forum, https://www.epa.gov/risk/framework-human-health-risk-assessment-inform-
decisi on-making

U.S. EPA. (2015a). Flame retardants used in flexible polyurethane foam: An alternatives assessment

update. (EPA 744-R-l5-002). Washington, DC: U.S. Environmental Protection Agency, Design
for the Environment.

https://nepis.epa. eov/Exe/ZvPDF.cei/P100MXUQ.PDF?Dockev=P 100MXUQ.PDF
U.S. EPA. (2015b). TSCA. Work Plan Chemical Problem Formulation and Initial Assessment.

Chlorinated Phosphate Ester Cluster Flame Retardants. (EPA 740-R1-5001). Washington, DC.
http ://www. epa. gov/ sites/production/files/2015 -
09/documents/cpe fr cluster problem formulation.pdf
U.S. EPA. (2016a). Ecological effects test guidelines: OCSPP 850.1075: Freshwater and saltwater fish
acute toxicity test [EPA Report], (EPA 712-C-16-007; EPA-HQ-OPPT-2009-0154-0035).
Washington, DC: U.S. Environmental Protection Agency, Office of Chemical Safety and
Pollution Prevention. https://www.reeiilations.eov/dociiment/EPA-HQ-OPPT-2009-0154-0035
U.S. EPA. (2016b). Guidance for conducting fish consumption surveys. (823B 16002).

https://www.epa.eov/sites/prodiiction/files/2i 'documents/fcsurvey euidance.pdf

(2016c). Integrated review plan for the national ambient air quality standards for particulate
matter [EPA Report], (EPA-452/R-16-005). Research Triangle Park, NC.
https://nepis.epa. eov/Exe/ZyPURL.cei?Dockev=P 100R5VE.txt
U.S. EPA. (2016d). Public database 2016 chemical data reporting (May 2017 release). Washington, DC:
US Environmental Protection Agency, Office of Pollution Prevention and Toxics. Retrieved
from https://www.epa.eov/chemical-data-reportine
U.S. EPA. (2016e). Weight of evidence in ecological assessment [EPA Report], (EPA/100/R-16/001).
Washington, DC: Office of the Science Advisor.
https://nepis.epa. eov/Exe/ZyPURL.cei?Dockey=P 100SFXR.txt
U.S. EPA. (2017a). Human Health Benchmarks for Pesticides: Updated 2017 Technical Document (pp.
5). (EPA 822-R -17 -001). Washington, DC: U.S. Environmental Protection Agency, Office of
Water, https://www.epa.gov/sites/production/files/2Q15-lQ/documents/hh-benchmarks-
techdoc.pdf

U.S. EPA. (2017c). Update for Chapter 5 of the Exposure Factors Handbook: Soil and dust ingestion
[EPA Report], (EPA/600R-17/384F). Washington, DC: National Center for Environmental
Assessment, Office of Research and Development.

https://nepis.epa. eov/Exe/ZyPURL.cei?Dockey=P 100TTX4.txt
U.S. EPA. (2018). User's Guide for the A MS/EPA Regulatory Model (AERMOD). (EPA Document
Number: EPA-454/B-18-001). U.S. EPA.

https://hero.epa.gov/hero/index.cfm?action=search.view&reference id=6277143U.S. EPA.
(2019b). Chemistry dashboard information for tris(2-chloroethyl) phosphate. 115-96-8
[Website], https://comptox.epa.gov/dashboard/dsstoxdb/results?search=DTX
U.S. EPA. (2019d). Consumer Exposure Model (CEM) 2.1 User Guide. (EPA Contract # EP-W-12-
010). Washington, DC.

(2019e). Consumer Exposure Model (CEM) 2.1 User Guide - Appendices. (EPA Contract #
EP-W-12-010). Washington, DC: U.S. EPA. https://www.epa.gov/tsca-screening-
tools/consumer-exposure-model-cem-version-20-users-guide

Page 394 of 572

-------
10847

10848

10849

10850

10851

10852

10853

10854

10855

10856

10857

10858

10859

10860

10861

10862

10863

10864

10865

10866

10867

10868

10869

10870

10871

10872

10873

10874

10875

10876

10877

10878

10879

10880

10881

10882

10883

10884

10885

10886

10887

10888

10889

10890

10891

10892

10893

10894

10895

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

U.S. EPA. (20191). Point Source Calculator: A Model for Estimating Chemical Concentration in Water
Bodies. Washington, DC: U.S. Environmental Protection Agency, Office of Chemical Safety and
Pollution Prevention.

U.S. EPA. (2019g). User's Guide: Integrated Indoor-Outdoor Air Calculator (IIOAC). Washington, DC:
U.S. EPA. https://hero.epa.eov/hero/index.cfm?action=search.view&reference id=6277143U.S.
EPA. (2020b). Final Scope of the Risk Evaluation for Tris(2-chloroethy 1) Phosphate (TCEP);
CASRN 115-96-8. (EPA 740-R-20-009). Washington, DC: Office of Chemical Safety and
Pollution Prevention, https://www.epa.gov/sites/default/files/2020-09/documents/casrn 115-96-
8 tris2-chloroethyl phosphate tcep final scope.pdf

U.S. EPA. (2020c). Final Scope of the Risk Evaluation for T ri s( 2-chl oroethy 1) Phosphate (TCEP);

CASRN 115-96-8, Supplemental File: Data Extraction and Data Evaluation Tables for Physical
and Chemical Property Studies. Washington, DC: Office of Chemical Safety and Pollution
Prevention.

U.S. EPA. (2021). Draft systematic review protocol supporting TSCA risk evaluations for chemical
substances, Version 1.0: A generic TSCA systematic review protocol with chemical-specific
methodologies. (EPA Document #EPA-D-20-031). Washington, DC: Office of Chemical Safety
and Pollution Prevention. https://www.reeiilations.eov/dociiment/EPA-HQ-OPPT-2'

0005

U.S. EPA. (2022a). Consumer Exposure Model (CEM) user guide, Version 3.0. (EPA Contract #EP-W-
12-010). Washington, DC: U.S. Environmental Protection Agency, Office of Pollution
Prevention and Toxics.

https://hero.epa.gov/hero/index.cfm?action=search.view&referenc	U.S. EPA.

(2023a). Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Supplemental File
Folder: Supplemental Information on Human Milk PBPK Verner Modeling Results. Washington,
DC: Office of Pollution Prevention and Toxics, Office of Chemical Safety and Pollution
Prevention.

(2023b). Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Supplemental
Information File: Benchmark Dose Modeling Results for TCEP. Washington, DC: Office of
Pollution Prevention and Toxics, Office of Chemical Safety and Pollution Prevention.

U.S. EPA. (2023c). Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Supplemental
Information File: Consumer Exposure Modeling Inputs. Washington, DC: Office of Pollution
Prevention and Toxics, Office of Chemical Safety and Pollution Prevention.

U.S. EPA. (2023d). Draft Risk Evaluation for Tris(2-chloroethy 1) Phosphate (TCEP) - Supplemental
Information File: Consumer Exposure Modeling Results. Washington, DC: Office of Pollution
Prevention and Toxics, Office of Chemical Safety and Pollution Prevention.

(2023e). Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Supplemental
Information File: E-FAST Modeling Results. Washington, DC: Office of Pollution Prevention
and Toxics, Office of Chemical Safety and Pollution Prevention.

U.S. EPA. (2023f). Draft Risk Evaluation for T ri s( 2-chl oroethy 1) Phosphate (TCEP) - Supplemental

Information File: Environmental Monitoring and Biomonitoring Concentrations Summary Table.
Washington, DC: Office of Pollution Prevention and Toxics, Office of Chemical Safety and
Pollution Prevention.

U.S. EPA. (2023g). Draft Risk Evaluation for Tri s( 2-chl oroethy 1) Phosphate (TCEP) - Supplemental
Information File: Environmental Monitoring Concentrations Reported by Media Type.
Washington, DC: Office of Pollution Prevention and Toxics, Office of Chemical Safety and
Pollution Prevention.

U.S. EPA. (2023h). Draft Risk Evaluation for Tri s( 2-chl oroethy 1) Phosphate (TCEP) - Supplemental
Information File: Exposure Air Concentration Risk Calculations. Washington, DC: Office of
Pollution Prevention and Toxics, Office of Chemical Safety and Pollution Prevention.

Page 395 of 572

-------
10896

10897

10898

10899

10900

10901

10902

10903

10904

10905

10906

10907

10908

10909

10910

10911

10912

10913

10914

10915

10916

10917

10918

10919

10920

10921

10922

10923

10924

10925

10926

10927

10928

10929

10930

10931

10932

10933

10934

10935

10936

10937

10938

10939

10940

10941

10942

10943

10944

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

U.S. EPA. (2023i). Draft Risk Evaluation for Tris(2-chloroethy 1) Phosphate (TCEP) - Supplemental
Information File: Human Health Hazard Points of Departure Comparison Tables. Washington,
DC: Office of Pollution Prevention and Toxics, Office of Chemical Safety and Pollution
Prevention.

(2023j). Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Supplemental
Information File: IIOAC Modeling Input and Results. Washington, DC: Office of Pollution
Prevention and Toxics, Office of Chemical Safety and Pollution Prevention.

U.S. EPA. (2023 k). Draft Risk Evaluation for Tri s( 2-chl oroethy 1) Phosphate (TCEP) - Supplemental
Information File: Risk Calculator for Occupational Exposures. Washington, DC: Office of
Pollution Prevention and Toxics, Office of Chemical Safety and Pollution Prevention.

(20231). Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Supplemental
Information File: Supplemental Information on Environmental Release and Occupational
Exposure Assessment. Washington, DC: Office of Pollution Prevention and Toxics, Office of
Chemical Safety and Pollution Prevention.

(2023m). Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Supplemental
Information File: Water Quality Portal Processed Water Data. Washington, DC: Office of
Pollution Prevention and Toxics, Office of Chemical Safety and Pollution Prevention.

U.S. EPA. (2023n). Draft Risk Evaluation for Tri s( 2-chl oroethy 1) Phosphate (TCEP) - Systematic
Review Protocol. Washington, DC: Office of Pollution Prevention and Toxics, Office of
Chemical Safety and Pollution Prevention.

U.S. EPA. (2023o). Draft Risk Evaluation for Tri s( 2-chl oroethy 1) Phosphate (TCEP) - Systematic

Review Supplemental File: Data Extraction Information for Environmental Hazard and Human
Health Hazard Animal Toxicology and Epidemiology. Washington, DC: Office of Pollution
Prevention and Toxics, Office of Chemical Safety and Pollution Prevention.
https://www.reeulations.eov/docket/EPA~HQ~QPPT~2018~

U.S. EPA. (2023p). Draft Risk Evaluation for Tri s( 2-chl oroethy 1) Phosphate (TCEP) - Systematic

Review Supplemental File: Data Extraction Information for General Population, Consumer, and
Environmental Exposure. Washington, DC: Office of Pollution Prevention and Toxics, Office of
Chemical Safety and Pollution Prevention.

U.S. EPA. (2023q). Draft Risk Evaluation for Tri s( 2-chl oroethy 1) Phosphate (TCEP) - Systematic

Review Supplemental File: Data Quality Evaluation and Data Extraction Information for Dermal
Absorption. Washington, DC: Office of Pollution Prevention and Toxics, Office of Chemical
Safety and Pollution Prevention.

U.S. EPA. (2023r). Draft Risk Evaluation for T ri s( 2-chl oroethy 1) Phosphate (TCEP) - Systematic
Review Supplemental File: Data Quality Evaluation and Data Extraction Information for
Environmental Fate and Transport. Washington, DC: Office of Pollution Prevention and Toxics,
Office of Chemical Safety and Pollution Prevention.

U.S. EPA. (2023s). Draft Risk Evaluation for Tris(2-chloroethy 1) Phosphate (TCEP) - Systematic
Review Supplemental File: Data Quality Evaluation and Data Extraction Information for
Environmental Release and Occupational Exposure. Washington, DC: Office of Pollution
Prevention and Toxics, Office of Chemical Safety and Pollution Prevention.

U.S. EPA. (2023t). Draft Risk Evaluation for T ri s( 2-chl oroethy 1) Phosphate (TCEP) - Systematic
Review Supplemental File: Data Quality Evaluation and Data Extraction Information for
Physical and Chemical Properties. Washington, DC: Office of Pollution Prevention and Toxics,
Office of Chemical Safety and Pollution Prevention.

U.S. EPA. (2023u). Draft Risk Evaluation for Tri s( 2-chl oroethy 1) Phosphate (TCEP) - Systematic
Review Supplemental File: Data Quality Evaluation Information for Environmental Hazard.
Washington, DC: Office of Pollution Prevention and Toxics, Office of Chemical Safety and
Pollution Prevention.

Page 396 of 572

-------
10945

10946

10947

10948

10949

10950

10951

10952

10953

10954

10955

10956

10957

10958

10959

10960

10961

10962

10963

10964

10965

10966

10967

10968

10969

10970

10971

10972

10973

10974

10975

10976

10977

10978

10979

10980

10981

10982

10983

10984

10985

10986

10987

10988

10989

10990

10991

10992

10993

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

U.S. EPA. (2023v). Draft Risk Evaluation for Tris(2-chloroethy 1) Phosphate (TCEP) - Systematic
Review Supplemental File: Data Quality Evaluation Information for General Population,
Consumer, and Environmental Exposure. Washington, DC: Office of Pollution Prevention and
Toxics, Office of Chemical Safety and Pollution Prevention.

U.S. EPA. (2023w). Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Systematic
Review Supplemental File: Data Quality Evaluation Information for Human Health Hazard
Animal Toxicology. Washington, DC: Office of Pollution Prevention and Toxics, Office of
Chemical Safety and Pollution Prevention. https://www.reeutations.eov/docket/EPA-HQ-OPPT-
2018-0476

U.S. EPA. (2023x). Draft Risk Evaluation for Tri s( 2-chl oroethy 1) Phosphate (TCEP) - Systematic
Review Supplemental File: Data Quality Evaluation Information for Human Health Hazard
Epidemiology. Washington, DC: Office of Pollution Prevention and Toxics, Office of Chemical
Safety and Pollution Prevention.

Ulsamer. AG; Osterber McLaughlin. J. (1980). Flame retardant chemicals in textiles. Clin Toxicol
17: 101-132.

Ume/ii l' »tniemoto. J; Soma \ ¦iztiki. T. (1998). Tri s( 2-chl oroethy 1) phosphate increases

ambulatory activity in mice: Pharmacological analyses of its neurochemical mechanism. Toxicol
Appl Pharmacol 148: 109-116. http://dx.doi.oiv iO 1006/tajp rs° \ ; I _

Van den Eede. N: Maho. W; Erratico. C; Neels. H; Covaci. A. (2013). First insights in the metabolism
of phosphate flame retardants and plasticizers using human liver fractions. Toxicol Lett 223: 9-
15. http://dx.doi.ore/10 J016/i.toxlet.i01 ; OS 0 i J
van der Veet> ^ » r x J. (2012). Phosphorus flame retardants: properties, production, environmental
occurrence, toxicity and analysis [Review], Chemosphere 88: 1119-1153.
http://dx.doi.ore 10 101 i.chemosphere.1 ^ I _ ^ ^ >7
Venier. M; Dove. A; Romanak. K; Backus. S: Hites. R. (2014). Flame retardants and legacy chemicals
in Great Lakes' water. Environ Sci Technol 48: 9563-9572. http://dx.doi.c	501509r

Verbruggen. EMI; Rila. JP; Traas. TP; Posthuma-Doodeman. CJAM; Posthumus. R. (2005).

Environmental risk limits for several phosphate esters, with possible application as flame
retardant. (RIVM report 601501024/2005). Bilthoven, The Netherlands: Netherlands National
Institute for Public Health and the Environment.

https://rivm.openrepositorv.eom/bitstream/handle/l 0029/255802/601501024.pdf? sequence=3&is
Allowed=v

Verner. MA; Ayotte. P; Mtickl harbonneau. M; Haddad. S. (2009). A physiologically based

pharmacokinetic model for the assessment of infant exposure to persistent organic pollutants in
epidemiologic studies. Environ Health Perspect 117: 481-487.

http://dx.doi.ore S9/ehp.0800047
Verner. ft larbonneau. M; Lopez-Carrillo. L; Haddad. S. (2008). Physiologically based

pharmacokinetic modeling of persistent organic pollutants for lifetime exposure assessment: A
new tool in breast cancer epidemiologic studies. Environ Health Perspect 116: 886-892.

http://dx.doi.ore 1.39/elin KVM
Wallner. P; Kundi. M; Moshammer. H; Piegler. K; Hohenblum. P; Scharf. S; Frohtich. M; Dambereer.

ere. P; Hutter. HP. (2012). Indoor air in schools and lung function of Austrian school
children. J Environ Monit 14: 1976-1982. http://dx.doij	9/c2em30059a

Wambaueh. IF; Wetmeh' M \ King. CL; Nicolas. C; Pearce. RG; Honda. GS; Dinallc K \ii\us. D;
Gilben .<	« rsgdrinaravanan. A; Snodgrass. r% r%mckman. A; Strock. C; Setzer. RW;

Thomas. RS. (2019). Assessing toxicokinetic uncertainty and variability in risk prioritization.
Toxicol Sci 172: 235-251. http://dx.doi.ore/10.1093/toxsci/kfz205
Wane. C; Wane. P; Zhao. J; Fu. M; Zham 1 1 \ \ \ k Zhu \ ^ ^ 1 b mg. O; Jiar^

(2020a). Atmospheric organophosphate esters in the Western Antarctic Peninsula over 2014-

Page 397 of 572

-------
10994

10995

10996

10997

10998

10999

11000

11001

11002

11003

11004

11005

11006

11007

11008

11009

11010

11011

11012

11013

11014

11015

11016

11017

11018

11019

11020

11021

11022

11023

11024

11025

11026

11027

11028

11029

11030

11031

11032

11033

11034

11035

11036

11037

11038

11039

11040

11041

11042

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

2018: Occurrence, temporal trend and source implication. Environ Pollut 267: 115428.

http://dx.doi.ore 10 101 i.envpol.201 ^ I I L *

Wane «* Wang. P; Wain \ iuin \\ 'iiu. C; Sun. H; Matsiko. J; Zhu \ « >. Y; Mem \\ bane.
Q; Jiang. G. (2019a). Temporal variations of PM2.5-bound organophosphate flame retardants in
different microenvironments in Beijing, China, and implications for human exposure. Sci Total
Environ 666: 226-234. http://dx.doi.ore/1 /i.scitotenv.2019.02.076
Wane v Ui. H; Du. Z; Chen. H; Pern f ao. S. (2017a). Bioaccumulation mechanism of
organophosphate esters in adult zebrafish (Danio rerio). Environ Pollut 229: 177-187.
http://dx.doi.ore/10.1016/i.envpol .01 0 0
Wane <15 'hao. H; Wain \ «* t lien. J; Ouan. X. (2017b). Determination and prediction of

octanol-air partition coefficients for organophosphate flame retardants. Ecotoxicol Environ Saf
145: 283-288. http://dx.doi.ore/10.1016/i.ecoen\ JO I 0 0 10
Wane. RY; Needham. LL. (2007). Environmental chemicals: from the environment to food, to breast
milk, to the infant [Review], J Toxicol Environ Health B Crit Rev 10: 597-609.
http://dx.doi.ore 30/10937400701389891
Wang. X; Chen. P; Zha hu. L; Wu. F. (2021). Transplacental Behaviors of Organophosphate Tri-
and Diesters Based on Paired Human Maternal and Cord Whole Blood: Efficiencies and Impact
Factors. Environ Sci Technol 55: 3091-3100. http://dx.doi.org/10.102 l/aes.est.0c06095
Wane. X; Zhong. W; Xiao. <' < m. Q; Yam < 1 <»vaci. A; Zhu 1 (2019b). Bioavailability and

biomagnification of organophosphate esters in the food web of Taihu Lake, China: Impacts of
chemical properties and metabolism. Environ Int 125: 25-32.
http://dx.doi.ore 10 101 i.envint.-0 rs 01 01 s
Wane. X; Zhu. L; Zhou	e. L. (2018a). Partition and source identification of organophosphate

esters in the water and sediment of Taihu Lake, China. J Hazard Mater 360: 43-50.
http://dx.doi.ore	jhazmat.2018.07.082

Wane ^ . Kannan. P; Hal den. K1! i annan. K. (2019c). A nationwide survey of 31 organophosphate
esters in sewage sludge from the United States. Sci Total Environ 655: 446-453.
http://dx.doi.ore/10.1016/i.scitotenv.2^I * I I __4
Warn . W; Martinez-Moral. MP; Sun. H; Kannan. K. (2019d). Metabolites of organophosphate
esters in urine from the United States: Concentrations, temporal variability, and exposure
assessment. Environ Int 122: 213-221. http://dx.doi.ore/10.1016/i envint.1'01S I I 00
Wane. Y; Li. Z; Tan I mi \ , "hao. H; Chen i (2020b). Occurrence and air-soil exchange of

organophosphate flame retardants in the air and soil of Dalian, China. Environ Pollut 265 Pt. A:
114850. http://dx.doi.ore/l /i.envpol.2020.114850
Wane \ YUi \ 'hanv * * hao. H; Ho^ U	\ Ion J (2018b). Occurrence, distribution, and

air-water exchange of organophosphorus flame retardants in a typical coastal area of China.
Chemosphere 211: 335-344. http://dx.doi.ore/10.1016/i.chemosphere..0 i s 0 0 32
Watts. MI; Linden. KG. (2008). Photooxidation and subsequent biodegradability of recalcitrant tri-alkyl
phosphates TCEP and TBP in water. Water Res 42: 4949-4954.
http://dx.doi.ore	.watres.2008.09.020

Watts. Ml; Linden. (2009). Advanced oxidation kinetics of aqueous trialkyl phosphate flame
retardants and plasticizers. Environ Sci Technol 43: 2937-2942.
http://dx.doi.ore	;s8031659

Weemann. F; Cavin. L; Macleod. M; Scheringer. M; Huneerbuhler. K. (2009). The OECD software tool
for screening chemicals for persistence and long-range transport potential. Environ Modell Softw
24: 228-237. http://dx.doi.cn ^ 10 1016/i .envsoft.20Q\ 0 01 I
Wei. GL; Li. DO; Zhuo. MN; Liao \ v	\	U li.j.i 'iiang. SY; Liane o (2014).

Organophosphorus flame retardants and plasticizers: Sources, occurrence, toxicity and human
exposure [Review], Environ Pollut 196C: 29-46. http://dx.doi.org/10.1016/i .envpol.201 10° 01 J

Page 398 of 572

-------
11043

11044

11045

11046

11047

11048

11049

11050

11051

11052

11053

11054

11055

11056

11057

11058

11059

11060

11061

11062

11063

11064

11065

11066

11067

11068

11069

11070

11071

11072

11073

11074

11075

11076

11077

11078

11079

11080

11081

11082

11083

11084

11085

11086

11087

11088

11089

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Weil. ED. (2001). Kirk-Othmer Encyclopedia of Chemical Technology Flame Retardants, Phosphorus.
New York, NY: John Wiley & Sons.

http://dx.doi.Org/https://onlinelibrary.wilev.com/doi/abs/10.1002/0 I I _ '< v\ I I <»i i . 05091
2l b2

Welke. B; Ettlinger. K; Riederer. M. (1998). Sorption of volatile organic chemicals in plant surfaces.

Environ Sci Technol 32: 1099-1104.

Weschler. CI; Nazaroff. WW. (2010). SVOC partitioning between the gas phase and settled dust indoors

[Review], Atmos Environ 44: 3609-3620. http://dx.doi.org/10.1016/i.atmosenv.2010 0 0.9
Wesely. ML; Doske	tnnon. JD. (2002). Deposition parameterizations for the Industrial Source

Complex (ISC3) model. (ANL/ER/TR-01/003). Argonne, IL: Argonne National Lab.

Westerhoff. ^ n. Y; Snyder. S: Wert. E. (2005). Fate of endocrine-disruptor, pharmaceutical, and
personal care product chemicals during simulated drinking water treatment processes. Environ
Sci Technol 39: 6649-6663. http://dx.doi.org/10.102 l/es0484799
White. PR; Widdowson. EM; Woodai'i U"15 1 h \ -Tsori i\\ « (1991). The composition of body tissues

(II) Fetus to young adult. Br J Radiol 64: 149-159.

Whittaker. C; Ri»^' I VtcKernan. L; Danko\io D, 1 out « Xlacmahon. K; Kuemp^'l « umwalde. R;
Schu (2016). Current Intelligence Bulletin 68: NIOSH Chemical Carcinogen Policy. US
Department of Health and Human Services.

https://ntrl.ntis.gov/NTRL/dashboard/searchResults/titleDetail/PB:	html

Willming. MM; Lilavois. ( ron. MG; Raimondo. S. (2016). Acute toxicity prediction to

threatened and endangered species using Interspecies Correlation Estimation (ICE) models.
Environ Sci Technol 50: 10700-10707. http://dx.doi.org/10.102 i/acs.est.6b03009
Wolschke. H; Suehring. R; Mi. W; Moell	Ebinghaus. R. (2016). Atmospheric occurrence

and fate of organophosphorus flame retardants and plasticizer at the German coast. Atmos
Environ 137: 1-5. hu
-------
11090

11091

11092

11093

11094

11095

11096

11097

11098

11099

11100

11101

11102

11103

11104

11105

11106

11107

11108

11109

11110

11111

11112

11113

11114

11115

11116

11117

11118

11119

11120

11121

11122

11123

11124

11125

11126

11127

11128

11129

11130

11131

11132

11133

11134

11135

11136

11137

11138

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Yadav. IC: Devi. ML; I tang. G. (2018). Organophosphate ester flame retardants in Nepalese soil:
Spatial distribution, source apportionment and air-soil exchange assessment. Chemosphere 190:
114-123. http://dx.doi.cnv 10 101 5/i.chemosphere.201 0° II.

Yaman. B; Dumanoglu. abasi. M. (2020). Measurement and modeling the phase partitioning of
organophosphate esters using their temperature-dependent octanol-air partition coefficients and
vapor pressures. Environ Sci Technol 54: 8133-8143. http://dx.doi.org/10.102 l/acs.est.0c02823
Yang ig. J; Huang. W; Xie. W; Liu. W. (2014). Particle size-specific distributions and preliminary
exposure assessments of organophosphate flame retardants in office air particulate matter.
Environ Sci Technol 48: 63-70. http://dx.doi.org	s403186z

Yang. W; Zhao. F; Fang \ . 1 i 1 . 1 i. C". I Ni (2018a). lH-nuclear magnetic resonance metabolomics
revealing the intrinsic relationships between neurochemical alterations and neurobehavioral and
neuropathological abnormalities in rats exposed to tris(2-chloroethyl)phosphate. Chemosphere
200: 649-659. http://dx.doi.org/10.1016/i.chemosphere.2018.02.056
Yang. X; Sun. L; Xiang. J; Hu. S: Su. S. (2013). Pyrolysis and dehalogenation of plastics from waste
electrical and electronic equipment (WEEE): A review. Waste Manag 33: 462-473.
http://dx.doi.ore	.wasman.2012.07.025

Yang \ \iao. Y; Chang. Y; Cui. Y; Klobuc a Li. M. (2018b). Intestinal damage, neurotoxicity and
biochemical responses caused by tris (2-chloroethyl) phosphate and tricresyl phosphate on
earthworm. Ecotoxicol Environ Saf 158: 78-86. http://dx.doi.o	/i.ecoenv.2018.04.012

Yasuhara. A. (1995). Chemical components in leachates from hazardous wastes landfills in Japan.

Toxicol Environ Chem 51: 113-120. http://dx.doi.org/10.1080/027722495Q9358229
Yasuhara \ c.liiraishi. H; Nishikawa. M; Yamamotc < 'x akasugi. O; Okumura. T; Kenmotsu. K;

Fukui. H; Nagase. M; Kawagoshi. Y. (1999). Organic components in leachates from hazardous
waste disposal sites. Waste Manag Res 17: 186-197. http://dx.doi.org/10.103- i | I
3070.1999.00038.x

Yonemoto. J; Shiraishi. H; Soma. Y; In aba. K; Sone. H; Kobavashi. S. (1997). Use of rat embryo limb
bud cell cultures to screen organochlorine compounds detected in the water and sediment of
rivers in Tokyo metropolis for developmental toxicity. Toxicol Environ Chem 62: 125-133.
http://dx.doi.orE 30/02772249709358502

Zhang. H; ZIhhi < * \t^ ' 'Ihhi \ i u t _ i u t Mi. \\ « Innghaus. R; Christie. P; Luo. Y. (2018a).
Occurrences of organophosphorus esters and phthalates in the microplastics from the coastal
beaches in north China. Sci Total Environ 616-617: 1505-1512.
http://dx.doi.org 10 101 |.scitotenv.201 10 I '<

Zhang. S: Gitungo. S: /	ksen. IE; Raczko. RF. (2016a). A pilot plant study using conventional

and advanced water treatment processes: Evaluating removal efficiency of indicator compounds
representative of pharmaceuticals and personal care products. Water Res 105: 85-96.
http://dx.doi.org 10 101 i.watres.-O I 08.033
Zhang. S: Rosat irrison. Ml. (2009). Detection of multidecadal oceanic variability by ocean data
assimilation in the context of a "perfect" coupled model. J Geophys Res Oceans 114.
http://dx.doi.orE 29/2008JC005261
Zhang. W; C	i. X; Xu. J. (2021). Organophosphate esters in sediment from Taihu Lake,

China: Bridging the gap between riverine sources and lake sinks. Front Env Sci Eng 16: 30.
http://dx.doi.ore 10 100 ^ I I '83-0. Ill I °

Zhang. W: Zham \ U	i W; Xiong. W; Lu. W; Zheng. H: Chen, i \\iqn. J. (2017aY

Tris (2-chloroethyl) phosphate induces senescence-like phenotype of hepatocytes via the
p21Wafl/Cipl-Rb pathway in a p53-independent manner. Environ Toxicol Pharmacol 56: 68-75.
http://dx.doi.ore	.etap.2017.08.028

Zhang. W; Zham \ Wmv \u < ituang. C; Yin. W; Wang. J; Xiong. W; Lu. W; Zheng. H; Yuan.
J. (2016b). Tris(2-chloroethyl)phosphate-induced cell growth arrest via attenuation of SIRT1-

Page 400 of 572

-------
1139

1140

1141

1142

1143

1144

1145

1146

1147

1148

1149

1150

1151

1152

1153

1154

1155

1156

1157

1158

1159

1160

1161

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

independent PI3K/Akt/mTOR pathway. J Appl Toxicol 36: 914-924.

http://dx.doi.ore	at.3223

Zhang. W: Zham \ i .i* 1 Wang 1 \hm>v W 1 W: Zheng. H: Yuan. J. (2016cV

Involvement of ROS-mediated mitochondrial dysfunction and SIRT3 down-regulation in tris(2-
chloroethyl) phosphate-induced cell cycle arrest. Toxicology Research 5: 461-470.
http://dx.doi.ore	;5tx00229i

Zhan: hang. W; Hon. J; Wang. X; Zheng. H; Xiong. We; Yuan. J. (2017b). Combined effect of
tris(2-chloroethyl) phosphate and benzo (a) pyrene on the release of IL-6 and IL-8 from HepG2
cells via the EGFR-ERK1/2 signaling pathway. RSC Adv 7: 54281-54290.
http://dx.doi.orE	;7ral 1273d

Zham \ heng. X; Wei. L; Sun. K uo. H; Liu \ < ,iu. S; Li. Y; M.ti «' (2018b). The distribution
and accumulation of phosphate flame retardants (PFRs) in water environment. Sci Total Environ
630: 164-170. http://dx.doi.ore 10 101 i.scitotenv.20lS 0^.215
Zham mi \ Wang \ It N jng. C; Rodgers. TFM; Tail t (2022). Occurrence and distribution
of organophosphate flame retardants in the typical soil profiles of the Tibetan Plateau, China. Sci
Total Environ 807: 150519. http://dx.doi.org/10.1016/j.scitotenv.2^. * s 0->Jo
Zhao. ;n. M; Gao. F; Shen. H; Hu. J. (2017). Organophosphorus flame retardants in pregnant
women and their transfer to chorionic villi. Environ Sci Technol 51: 6489-6497.
http: //dx. doi. or g/10.1021 /acs. est. / I
Zhao. H; Zha ^ 1 m ham v \U> I* \w 1 \\ .. . Y; 1 hi. J. (2018). Trophic transfer of
organophosphorus flame retardants in a lake food web. Environ Pollut 242: 1887-1893.
http://dx.doi.ore 10 101 i.envpol.20:is 0 0

Page 401 of 572

-------
1162

1163

1164

1165

1166

1167

1168

1169

1170

1171

1172

1173

1174

1175

1176

1177

1178

1179

1180

1181

1182

1183

1184

1185

1186

1187

1188

1189

1190

1191

1192

1193

1194

1195

1196

1197

1198

1199

1200

1201

1202

1203

1204

1205

1206

1207

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

APPENDICES

Appendix A ABBREVIATIONS, ACRONYMS, AND GLOSSARY OF
SELECT TERMS

A,1 Abbreviations and Acronyms

AC

Acute exposure concentrations

AChE

Acetyl cholinesterase

ADC

Average daily concentrations

ADME

Absorption, distribution, metabolism, and elimination

AERMOD

American Meteorological Society (AMS)/EPA Regulatory Model

AF

Assessment factor

ALP

Alkaline phosphatase

ALT

Alanine transferase

AST

Aspartate transaminase

ATSDR

Agency for Toxic Substances and Disease Registry

BAF

Bioaccumulation factor

BCCP

Bis(2-chloroethyl) carboxymethyl phosphate

BCF

Bioconcentration factor

BCGP

Bis(2-chloroethyl) 2-hydroxyethyl phosphate

BCHP

Bis(2-chloroethyl) hydrogen phosphate

BLS

Bureau of Labor Statistics

BMD

Benchmark dose

BMDL

Benchmark dose lower confidence limit

BMF

Biomagnification factor

BMR

Benchmark response

BSAF

Biota-sediment accumulation factor

CASRN

Chemical Abstracts Service Registry Number

CBI

Confidential business information

CDR

Chemical Data Reporting (Rule)

CEPA

Canadian List of Toxic Substances

CERCLA

Comprehensive Environmental Response, Compensation and Liability Act

CFR

Code of Federal Regulations

ChV

Chronic health value

CI

Confidence interval

coc

Concentration(s) of concern

CoCAP

Cooperative Chemicals Assessment Program

CPS A

Consumer Product Safety Act

CPSC

Consumer Product Safety Commission

CSCL

Chemical Substances Control Law

CSF

Cancer slope factor

CSHO

Certified Safety and Health Official

CTD

Characteristic travel distance

DIY

Do-it-yourself

DMR

Discharge Monitoring Report

DOT

Department of Transportation

DRAS

(Hazardous Waste) Delisting Risk Assessment Software (EPA model)

Page 402 of 572

-------
11208

11209

11210

11211

11212

11213

11214

11215

11216

11217

11218

11219

11220

11221

11222

11223

11224

11225

11226

11227

11228

11229

11230

11231

11232

11233

11234

11235

11236

11237

11238

11239

11240

11241

11242

11243

11244

11245

11246

11247

11248

11249

11250

11251

11252

11253

11254

11255

11256

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

DWTP

EC50

ECHA

ECOSAR

EPA

EPCRA

ESD

EU

FIR

GS

HC05

HEC

HED

HERO

HHE

IARC

IMAP

IR

IRIS
IUR

Koc
Kow
Kp

LADC

LADD

LCD

LC50

LD50

LOAEL

LOD

LOEC

LOQ

Log Koc

Log Kow

LRAT

MOA

MOE

MSW

MSWLF

NAICS

NATA

ND

NEI

NHANES

NICNAS

NIH

NIOSH

NITE

Drinking water treatment plant

Effect concentration at which 50 percent of test organisms exhibit an effect

European Chemicals Agency

Ecological Structure Activity Relationships (model)

Environmental Protection Agency

Emergency Planning and Community Right-to-Know Act

Emission Scenario Document

European Union

Food intake rate

Generic Scenario

Hazard concentration that is protective of 95 percent of the species in the sensitivity
distribution

Human equivalent concentration
Human equivalent dose

Health and Environmental Research Online (Database)

Health hazard evaluation
International Agency for Research on Cancer
Inventory Multi-Tiered Assessment and Prioritisation
Ingestion rate

Integrated Risk Information System
Inhalation unit risk

Soil organic carbon: water partitioning coefficient
Octanol: water partition coefficient
Permeability coefficient
Lifetime average daily concentrations
Lifetime average daily dose
Lifecycle diagram

Lethal concentration at which 50 percent of test organisms die
Lethal dose at which 50 percent of test organisms die
Lowest-observable-adverse-effect level
Limit of detection

Lowest-observed-effect concentration
Limit of quantification

Logarithmic organic carbon: water partition coefficient

Logarithmic octanol: water partition coefficient

Long-range transport via long-range atmospheric transport

Mode of action

Margin of exposure

Municipal solid waste

Municipal solid waste landfills

North American Industry Classification System

National Scale Air-Toxics Assessment

Non-detect

National Emissions Inventory

National Health and Nutrition Examination Survey

National Industrial Chemicals Notification and Assessment Scheme

National Institutes of Health

National Institute for Occupational Safety and Health
National Institute of Technology and Evaluation

Page 403 of 572

-------
11257

11258

11259

11260

11261

11262

11263

11264

11265

11266

11267

11268

11269

11270

11271

11272

11273

11274

11275

11276

11277

11278

11279

11280

11281

11282

11283

11284

11285

11286

11287

11288

11289

11290

11291

11292

11293

11294

11295

11296

11297

11298

11299

11300

11301

11302

11303

11304

11305

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

NMAM

NIOSH Manual of Analytical Methods

NO A A

National Oceanic and Atmospheric Administration

NOEL

No-observed-effect level

NOAEL

No-observed-adverse-effect level

NPDES

National Pollutant Discharge Elimination System

NTP

National Toxicology Program

NWIS

National Water Information System

OCSPP

Office of Chemical Safety and Pollution Prevention

OECD

Organisation for Economic Co-operation and Development

OES

Occupational exposure scenario

ONU

Occupational non-user

OPP

Office of Pesticide Programs

OPPT

Office of Pollution Prevention and Toxics

OSHA

Occupational Safety and Health Administration

PBPK

Physiologically based pharmacokinetic

PBZ

Personal breathing zone

PECO

Population, exposure, comparator, and outcome

PEL

Permissible exposure limit (OSHA)

PESS

Potentially exposed or susceptible subpopulations

PMOC

Persistent mobile organic compound

POD

Point of departure

POTW

Publicly owned treatment works

PPE

Personal protective equipment

PV

Production volume

QSAR

Quantitative structure-activity relationship (model)

RCRA

Resource Conservation and Recovery Act

REACH

Registration, Evaluation, Authorisation and Restriction of Chemicals (European Union)

RP

Respirable particle

RQ

Risk quotient

SCADC

Subchronic average daily concentration

SCE

Sister chromatid exchange

SDS

Safety data sheet

SIDS

Screening Information Dataset

SOC

Standard Occupational Classification (BLS codes)

SSD

Species sensitivity distribution

STEL

Short-term exposure limit

STORET

STOrage and RETrieval and Water Quality exchange

SVOC

Semi-volatile compound

TE

Transfer efficiency

TESIE

Toddler's Exposure to SVOCs in the Indoor Environment (study)

TGD

Technical Guidance Document (European Commission)

TCEP

Tris(2-chloroethyl) phosphate

TMF

Trophic magnification factor

TRI

Toxics Release Inventory

TRV

Toxicity reference value

TSCA

Toxic Substances Control Act

TWA

Time-weighted average

UF

Uncertainty factor

U.S.

United States

Page 404 of 572

-------
11306

11307

11308

11309

11310

11311

11312

11313

11314

11315

11316

11317

11318

11319

11320

11321

11322

11323

11324

11325

11326

11327

11328

11329

11330

11331

11332

11333

11334

11335

11336

11337

11338

11339

11340

11341

11342

11343

11344

11345

11346

11347

11348

11349

11350

11351

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

USGS	United States Geological Survey

V6	2,2-Bi s(chloromethyl)-propane-1,3 -diyltetraki s(2-chloroethyl) bi sphosphate

VOC	Volatile organic compound

VP	Vapor pressure

Web-ICE	Web-based Interspecies Correlation Estimation

WHO	World Health Organization

WQP	Water Quality Portal

WWTP	Wastewater treatment plant

7Q10	The lowest 7-day average flow that occurs (on average) once every 10 years

30Q5	The lowest 30-day average flow that occurs (on average) once every 5 years

A.2 Glossary of Select Terms

Best available science (	02.33): "means science that is reliable and unbiased. Use of best

available science involves the use of supporting studies conducted in accordance with sound and
objective science practices, including, when available, peer reviewed science and supporting studies and
data collected by accepted methods or best available methods (if the reliability of the method and the
nature of the decision justifies use of the data). Additionally, EPA will consider as applicable:

(1)	The extent to which the scientific information, technical procedures, measures, methods,
protocols, methodologies, or models employed to generate the information are reasonable for and
consistent with the intended use of the information;

(2)	The extent to which the information is relevant for the Administrator's use in making a decision
about a chemical substance or mixture;

(3)	The degree of clarity and completeness with which the data, assumptions, methods, quality
assurance, and analyses employed to generate the information are documented;

(4)	The extent to which the variability and uncertainty in the information, or in the procedures,
measures, methods, protocols, methodologies, or models, are evaluated and characterized; and

(5)	The extent of independent verification or peer review of the information or of the procedures,
measures, methods, protocols, methodologies or models."

Condition of use (COU) (15 U.S.C. § 2602(4)): "means the circumstances, as determined by the
Administrator, under which a chemical substance is intended, known, or reasonably foreseen to be
manufactured, processed, distributed in commerce, used, or disposed of."

Margin of exposure (MOE) (U.S. EPA. 2002a): "a numerical value that characterizes the amount of
safety to a toxic chemical-a ratio of a toxicological endpoint (usually a NOAEL [no observed adverse
effect level]) to exposure. The MOE is a measure of how closely the exposure comes to the NOAEL."

Mode of action (MOA) ( EPA. 2000c): "a series of key events and processes starting with
interaction of an agent with a cell, and proceeding through operational and anatomical changes causing
disease formation."

Point of departure (POD) (	302a): "dose that can be considered to be in the range of

observed responses, without significant extrapolation. A POD can be a data point or an estimated point
that is derived from observed dose-response data. A POD is used to mark the beginning of extrapolation
to determine risk associated with lower environmentally relevant human exposures."

Page 405 of 572

-------
11352

11353

11354

11355

11356

11357

11358

11359

11360

11361

11362

11363

11364

11365

11366

11367

11368

11369

11370

11371

11372

11373

11374

11375

11376

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Potentially exposed or susceptible subpopulations (PESS) (	02(12)): "means a group of

individuals within the general population identified by the Agency who, due to either greater
susceptibility or greater exposure, may be at greater risk than the general population of adverse health
effects from exposure to a chemical substance or mixture, such as infants, children, pregnant women,
workers, or the elderly."

Reasonably available information (40 CFR 702.33): "means information that EPA possesses or can
reasonably generate, obtain, and synthesize for use in risk evaluations, considering the deadlines
specified in TSCA section 6(b)(4)(G) for completing such evaluation. Information that meets the terms
of the preceding sentence is reasonably available information whether or not the information is
confidential business information, that is protected from public disclosure under TSCA section 14."

Routes (40 CFR 702.33): "means the particular manner by which a chemical substance may contact the
body, including absorption via ingestion, inhalation, or dermally (integument)."

Sentinel exposure (40 CFR 702.33): "means the exposure from a single chemical substance that
represents the plausible upper bound of exposure relative to all other exposures within a broad category
of similar or related exposures."

Weight of the scientific evidence (	): "means a systematic review method, applied in a

manner suited to the nature of the evidence or decision, that uses a pre-established protocol to
comprehensively, objectively, transparently, and consistently, identify and evaluate each stream of
evidence, including strengths, limitations, and relevance of each study and to integrate evidence as
necessary and appropriate based upon strengths, limitations, and relevance."

Page 406 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

11377	Appendix B REGULATORY AND ASSESSMENT HISTORY

11378	B.l Federal Laws and Regulations

1 1379

11380	Table Apx B-l. Federal Laws and Regulations		

Statutes/Regulations

Description of Authority/Regulation

Description of Regulation

LPA sUiluk's ivuukilions

TSCA - section 5

Provides EPA with authority to determine
a significant new use for a chemical
substance; conduct a review of a notice of
a significant new use; and make a
determination whether the chemical
substance or significant new use presents
an unreasonable risk of injury to health or
the environment.

EPA proposed a significant new use rule
(SNUR) for TCEP (88 FR 40741. June 22.
2023).

TSCA - section 6(b)

EPA is directed to identify high-priority
chemical substances for risk evaluation;
and conduct risk evaluations on at least 20
high priority substances no later than three
and one-half years after the date of
enactment of the Frank R. Lautenberg
Chemical Safety for the 21st Century Act.

TCEP is one of the 20 chemicals EPA
designated as a High-Priority Substance for
risk evaluation under TSCA (84 FR 71.924.
December 30, 2019). Designation of TCEP
as high-priority substance constitutes the
initiation of the risk evaluation on the
chemical.

TSCA - section 8(a)

The TSCA section 8(a) CDR Rule requires
manufacturers (including importers) to
give EPA basic exposure-related
information on the types, quantities and
uses of chemical substances produced
domestically and imported into the United
States.

TCEP manufacturing (including importing),
processing and use information is reported
under the CDR rule (85 FR 201.22. April 2.
2020).

TSCA - section 8(b)

EPA must compile, keep current and
publish a list (the TSCA Inventory) of each
chemical substance manufactured,
processed, or imported in the United
States.

TCEP was on the initial TSCA Inventory
and therefore was not subject to EPA's new
chemicals review process under TSCA
Section 5 (60 FR .1.6309. March 29. 1995V
The chemical is on the active inventory.

TSCA - section 8(d)

Provides EPA with authority to issue rules
requiring producers, importers, and (if
specified) processors of a chemical
substance or mixture to submit lists and/or
copies of ongoing and completed,
unpublished health and safety studies.

Two submissions received in 2021 (U.S.
EPA, Chemical Data Access Tool, accessed
November 25, 2022).

TSCA - section 4

Provides EPA with authority to issue rules
and orders requiring manufacturers
(including importers) and processors to test
chemical substances and mixtures.

Three chemical data submissions from test
rules received for TCEP: all three were
monitoring reports (1978, 1980, and 1981)
(U.S. EPA. ChemView. accessed April 3.
2019).

Page 407 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Statutes/Regulations

Description of Authority/Regulation

Description of Regulation

EPCRA - section 313

Requires annual reporting from facilities in
specific industry sectors that employ 10 or
more full-time equivalent employees and
that manufacture, process or otherwise use
a TRI-listed chemical in quantities above
threshold levels. A facility that meets
reporting requirements must submit a
reporting form for each chemical for which
it triggered reporting, providing data across
a variety of categories, including activities
and uses of the chemical, releases, and
other waste management (e.g., quantities
recycled, treated, combusted) and pollution
prevention activities (under section 6607 of
the Pollution Prevention Act). These data
include on- and off-site data as well as
multimedia data (i.e., air, land, and water).

TCEP is a listed substance subject to
reporting requirements under 40 CFR
372.65 effective as of November 30, 2022.

11381	B.2 State Laws and Regulations

11382

11383	Table Apx B-2. State Laws and Regulations

State Actions

Description of Action

State Prohibitions

Three states have adopted prohibitions for the use of TCEP in children's products,
including Marvland ( altli Gen § 24-306). New York (TRIS-free Children and
Babies Act 0 ir Conser § 37-0701 el sea.)), Minnesota (Four flame Retardants in
Furniture Foam and Children's Products (Minn. Stat. § 325F.071)).

California adopted a prohibition, effective on January 1, 2020, on the selling and
distribution in commerce of new, not previously owned juvenile products, mattresses, or
upholstered furniture that contains, or a constituent component of which contains,
covered flame retardant chemicals at levels above 1.000 parts per million (A.B. 2998.
Legislative Council. Sess. 2017-2018. C.A. 2018).

State Drinking Water
Standards and
Guidelines

Minnesota developed a health-based guidance value for TCEP in drinking water (Minn

R Chan. 4720).

Chemicals of High
Concern to Children

Several states have adopted reporting laws for chemicals in children's products
containing TCEP. including Maine ("38 MRSA Chanter 16-D). Minnesota ("Toxic Free
Kids Act Minn. Stat. 116.9401 to 116.9407). Oregon (Toxic-Free Kids Act, Senate Bill

478, 2015). Vermont ( x ) and Washington State (Wash. Admin. Code

I'- -i 1 n>).

Other

California listed TCEP on Proposition 65 in 1992 due to cancer (Cal Code Rees. Title 27.
S 27001).

California issued a Health Hazard Alert for TCEP (Hazard Evaluation Svstem and

Information Service, ).

California lists TCEP as a designated prioritv chemical for biomonitoring (California SB

1379).

Page 408 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

State Actions

Description of Action



TCEP is listed as a Candidate Chemical under California's Safer Consumer Products
Program (Health and Safety Code § 25252 and 25253). The regulation for Children's
Foam-Padded Sleeping Products containing TCEP as a Priority Product went into effect
on July 1, 2017: Manufacturers of this product must notify the Department by September
1, 2017 (California Department of Toxic Substances Control, Accessed April 12, 2019).

11384	B,3 International Laws and Regulations

11385

11386	Table Apx B-3. International Laws and Regulations

Country/ Organization

Requirements and Restrictions

Canada

TCEP (Ethanol, 2-chloro-, phosphate (3:1)) is on the Canadian List of
Toxic Substances ( chedule 1).

TCEP was added to Schedule 2 of the Canada Consumer Product Safety
Act (CCPSA), based on concerns for carcinogenicity and impaired fertility.
(Government Canada Chemical Safety portal. Accessed April 10. 201.9).

In January 2013, a Significant New Activity was adopted for TCEP
(Canada Gazette, April 3. 2014; Vol. 148. No. 9).

European Union

In June 2017, TCEP was added to Annex XIV of REACH (Authorisation
List) with a sunset date of August 21. 2015 (European Chemicals Agency
(ECHA. 2019) database. Accessed April 10. 2019).

In 2010, TCEP was listed on the Candidate list as a Substance of Very
High Concern (SVHC) under regulation (EC) No 1907/2006 - REACH

(Registration, Evaluation, Authorization and Restriction of Chemicals due
to its reproductive toxicity (category 57C)).

Australia

Ethanol, 2-chloro-, phosphate (3:1) (TCEP) was assessed under Human
Health Tier II and III of the Inventory Multi-Tiered Assessment and
Prioritisation (IMAP). Uses reported include commercial: (NICNAS.

2016, Ethanol, 2-chloro-, phospha man health tier II
assessment. Accessed April 8. 2019) (NIC lanol, 2-chloro
phosphate (3:1): Human health tier III assessment. Accessed April 8,
2019).

Japan

TCEP is regulated in Japan under the following legislation:

•	Act on the Evaluation of Chemical Substances and Regulation of Their
Manufacture, etc. (Chemical Substances Control Law; CSCL),

•	Act on Confirmation, etc. of Release Amounts of Specific Chemical
Substances in the Environment and Promotion of Improvements to the
Management Thereof,

•	Air Pollution Control Law

(National Institute of Technology and Evaluation fNITE] Chemical Risk
Information Platform TCHRIP'I. April 8. 2019).

Basel Convention

Waste substances and articles containing or contaminated with
polychlorinated biphenyls (PCBs) and/or polychlorinated terphenyls

Page 409 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Country/ Organization

Requirements and Restrictions



(PCTs) and/or polybrominated biphenyls (PBBs) are listed as a category of
waste under the Basel Convention. Although the United States is not
currently a party to the Basel Convention, this treaty still affects U.S.
importers and exporters.

fatt]3://www.baselint/Portals/4/Basel%2()Oonvention/docs/text/BaselCoriye
ntionText-e.pdf.

11387	B.4 Assessment History

11388

11389	Table Apx B-4. Assessment History of TCEP

Authoring Organization

Publication

EPA publications

U.S. EPA, Superfund Health Risk Technical Support
Center, Office of Research and Development (ORD)

Provisional Peer-Reviewed Toxicity Values (PPRTV)
for Tris(2-chloroethvl)phosphate (TCEP) (CASI
96-8) U.S. EPA (2009)

U.S. EPA, Design for the Environment Program

Oilier I S -Ixisv.

Design for the Environment (DfE) Alternatives
Assessments

d oi'gani/.alions

Agency for Toxic Substances and Disease Registry
(ATSDR)

Toxicological Profile for Phosphate Ester Flame

Retardants (2012)

National Toxicology Program (NTP), National
Institutes of Health (NIH)

Technical Report on Toxicology and Carcinogenesis
Studies of Tris(2-chloroethvl) Phosphate (CASI
96-8) in F344/N Rats and B6C3F1 Mice (Gavaae
Studies) (1991)

1 n Iciiuil ional

Organisation for Economic Co-operation and
Development (OECD), Cooperative Chemicals
Assessment Program (CoCAP)

chloroethvl)phosphate (CAS no. 115-96-8) (2006)



International Agency for Research on Cancer (IARC)

Monographs on the Evaluation of Carcinogenic Risks to
Humans Volume 71 (1999)

European Union, European Chemicals Agency (ECHA)

European Union Risk Assessment Repoi >-
96-8: Tris (2-chloroethvl) phosphate, TCEP (2009)

Government of Canada, Environment Canada, Health
Canada

Screening Assessment for the Challenge EthanoL 2-
chloro-, phosphate (3:1) (Tris(2-chlrorethvl) phosphate
iPl) (2009)

National Industrial Chemicals Notification and
Assessment Scheme (NICNAS), Australian
Government

EthanoL 2-chloro-. phosphate (3:1): Human health tier

it cessment (2016), and EthanoL 2-chltn^ phosphate

(3:1): Human health tier III assessme

11390

11391

Page 410 of 572

-------
11392

11393

11394

11395

11396

11397

11398

11399

11400

11401

11402

11403

11404

11405

11406

11407

11408

11409

11410

11411

11412

11413

11414

11415

11416

11417

11418

11419

11420

11421

11422

11423

11424

11425

11426

11427

11428

11429

11430

11431

11432

11433

11434

11435

11436

11437

11438

11439

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Appendix C LIST OF SUPPLEMENTAL DOCUMENTS

Appendix C incudes a list and citations for all supplemental documents included in the Draft Risk
Evaluation for TCEP. See Docket EPA-HQ-QPPT-2018-0476 for all publicly released files associated
with this draft risk evaluation package; see Docket EPA-HQ-OPPT-2023-0265 for all publicly released
files associated with peer review and public comments.

Associated Systematic Review Protocol and Data Quality Evaluation and Data Extraction

Documents - Provide additional detail and information on systematic review methodologies used as
well as the data quality evaluations and extractions criteria and results.

Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Systematic Review Protocol
(	Z023n) - In lieu of an update to the Draft Systematic Review Protocol Supporting TSCA

Risk Evaluations for Chemical Substances, also referred to as the "2021 Draft Systematic Review
Protocol" (	021), this systematic review protocol for the Draft Risk Evaluation for TCEP

describes some clarifications and different approaches that were implemented than those described
in the 2021 Draft Systematic Review Protocol in response to (1) SACC comments, (2) public
comments, or (3) to reflect chemical-specific risk evaluation needs. This supplemental file may also
be referred to as the "TCEP Systematic Review Protocol."

Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Systematic Review Supplemental
File: Data Quality Evaluation and Data Extraction Information for Physical and Chemical
Properties (U.S. EPA. 2023f) - Provides a compilation of tables for the data extraction and data
quality evaluation information for TCEP. Each table shows the data point, set, or information
element that was extracted and evaluated from a data source that has information relevant for the
evaluation of physical and chemical properties. This supplemental file may also be referred to as the
"TCEP Data Quality Evaluation and Data Extraction Information for Physical and Chemical
Properties."

Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Systematic Review Supplemental
File: Data Quality Evaluation and Data Extraction Information for Environmental Fate and
Transport (	k) - Provides a compilation of tables for the data extraction and data

quality evaluation information for TCEP. Each table shows the data point, set, or information
element that was extracted and evaluated from a data source that has information relevant for the
evaluation for Environmental Fate and Transport. This supplemental file may also be referred to as
the "TCEP Data Quality Evaluation and Data Extraction Information for Environmental Fate and
Transport."

Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Systematic Review Supplemental
File: Data Quality Evaluation and Data Extraction Information for Environmental Release and
Occupational Exposure (	023 s) - Provides a compilation of tables for the data extraction

and data quality evaluation information for TCEP. Each table shows the data point, set, or
information element that was extracted and evaluated from a data source that has information
relevant for the evaluation of environmental release and occupational exposure. This supplemental
file may also be referred to as the "TCEP Data Quality Evaluation and Data Extraction Information
for Environmental Release and Occupational Exposure."

Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Systematic Review Supplemental
File: Data Quality Evaluation and Data Extraction Information for Dermal Absorption (U.S. EPA.

Page 411 of 572

-------
11440

11441

11442

11443

11444

11445

11446

11447

11448

11449

11450

11451

11452

11453

11454

11455

11456

11457

11458

11459

11460

11461

11462

11463

11464

11465

11466

11467

11468

11469

11470

11471

11472

11473

11474

11475

11476

11477

11478

11479

11480

11481

11482

11483

11484

11485

11486

11487

11488

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

2023q) - Provides a compilation of tables for the data extraction and data quality evaluation
information for TCEP. Each table shows the data point, set, or information element that was
extracted and evaluated from a data source that has information relevant for the evaluation for
Dermal Absorption. This supplemental file may also be referred to as the "TCEP Data Quality
Evaluation and Data Extraction Information for Dermal Absorption."

Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Systematic Review Supplemental
File: Data Quality Evaluation Information for General Population, Consumer, and Environmental
Exposure. (U.S. EPA. 2023v) - Provides a compilation of tables for the data quality evaluation
information for TCEP. Each table shows the data point, set, or information element that was
evaluated from a data source that has information relevant for the evaluation of general population,
consumer, and environmental exposure. This supplemental file may also be referred to as the "TCEP
Data Quality Evaluation Information for General Population, Consumer, and Environmental
Exposure."

Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Systematic Review Supplemental
File: Data Extraction Information for General Population, Consumer, and Environmental Exposure
(	Z023p) - Provides a compilation of tables for the data extraction for TCEP. Each table

shows the data point, set, or information element that was extracted from a data source that has
information relevant for the evaluation of general population, consumer, and environmental
exposure. This supplemental file may also be referred to as the "TCEP Data Extraction Information
for General Population, Consumer, and Environmental Exposure."

Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Systematic Review Supplemental
File: Data Quality Evaluation Information for Human Health Hazard Epidemiology (U.S. EPA.
2023x) - Provides a compilation of tables for the data quality evaluation information for TCEP.

Each table shows the data point, set, or information element that was evaluated from a data source
that has information relevant for the evaluation of epidemiological information. This supplemental
file may also be referred to as the "TCEP Data Quality Evaluation Information for Human Health
Hazard Epidemiology."

Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Systematic Review Supplemental
File: Data Quality Evaluation Information for Human Health Hazard Animal Toxicology (U.S.
EPA. 2023w) - Provides a compilation of tables for the data quality evaluation information for
TCEP. Each table shows the data point, set, or information element that was evaluated from a data
source that has information relevant for the evaluation of human health hazard animal toxicity
information. This supplemental file may also be referred to as the "TCEP Data Quality Evaluation
Information for Human Health Hazard Animal Toxicology."

Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Systematic Review Supplemental
File: Data Quality Evaluation Information for Environmental Hazard (	)23u) - Provides

a compilation of tables for the data quality evaluation information for TCEP. Each table shows the
data point, set, or information element that was evaluated from a data source that has information
relevant for the evaluation of environmental hazard toxicity information. This supplemental file may
also be referred to as the "TCEP Data Quality Evaluation Information for Environmental Hazard."

Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Systematic Review Supplemental
File: Data Extraction Information for Environmental Hazard and Human Health Hazard Animal
Toxicology and Epidemiology (	23 o) - Provides a compilation of tables for the data

Page 412 of 572

-------
11489

11490

11491

11492

11493

11494

11495

11496

11497

11498

11499

11500

11501

11502

11503

11504

11505

11506

11507

11508

11509

11510

11511

11512

11513

11514

11515

11516

11517

11518

11519

11520

11521

11522

11523

11524

11525

11526

11527

11528

11529

11530

11531

11532

11533

11534

11535

11536

11537

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

extraction for TCEP. Each table shows the data point, set, or information element that was extracted
from a data source that has information relevant for the evaluation of environmental hazard and
human health hazard animal toxicology and epidemiology information. This supplemental file may
also be referred to as the "TCEP Data Extraction Information for Environmental Hazard and Human
Health Hazard Animal Toxicology and Epidemiology."

Associated Supplemental Information Documents - Provide additional details and information on
exposure, hazard, and risk assessments.

Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Supplemental Information
File: Supplemental Information on Environmental Release and Occupational Exposure
Assessment (	231).

Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Supplemental Information
File: E-FAST Modeling Results (U.S. EPA. 2023e).

Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Supplemental Information
File: IK)AC Modeling Input and Results (	23}).

Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Supplemental Information
File: Environmental Monitoring Concentrations Reported by Media Type (	323 e).

Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Supplemental Information
File: Environmental Monitoring and Biomonitoring Concentrations Summary Table (

2023a

Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Supplemental Information
File: Consumer Exposure Modeling Inputs (	2023c).

Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Supplemental File Folder:
Supplemental Information on Consumer Exposure Modeling Results (	2023d).

Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Supplemental Information
File: Human Health Hazard Points of Departure Comparison Tables (U.S. EPA. 2023i) -
Provides an Excel spreadsheet of PODs for all studies and hazard outcomes resulting in likely or
suggestive evidence integration conclusions. Basic study details as well as the PODs from each
study and associated HEDs, HECs, and total UFs for non-cancer endpoints, as well as CSFs and
IURs for cancer endpoints are presented.

Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Supplemental Information
File: Benchmark Dose Modeling Results for TCEP (U.S. EPA. 2023b) - Provides inputs to BMD
modeling as well as outputs for individual health effects associated with hazard outcomes that
have likely evidence integration conclusions. Information includes goodness of fit details for all
models that were run, as well as BMD and BMDL values for the selected BMR and any
comparison BMRs. Graphs of the chosen models are also presented.

Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Supplemental Information
File: Risk Calculator for Occupational Exposures (U.S. EPA. 2023k).

Page 413 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

11538	Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Supplemental Information

11539	File: Exposure Air Concentration Risk Calculations (U.S. EPA. 2023h).

11540

11541	Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Supplemental Information

11542	File: Water Quality Portal Processed Water Data (U.S. EPA. 2023m).

11543

11544	Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Supplemental File Folder:

1 1545	Supplemental Information on Human Milk PBPK Verner Modeling Results (	023 a)

11546

Page 414 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

11547	Appendix D DETAILED EVALUATION OF POTENTIALLY

11548	EXPOSED OR SUSCEPTIBLE SUBPOPULATIONS

11549	D.l PESS Based on Greater Exposure

11550	In this section, EPA addresses the following potentially exposed populations expected to have greater

11551	exposure to TCEP. Table Apx D-l presents the quantitative data sources that were used in the PESS

11552	exposure analysis for incorporating increased background and COU-specific exposures.

11553

Page 415 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

11554 Table Apx D-l. PESS Evidence Crosswalk for Increased Exposure

Category

Subcategory

Increased Background Exposure

Increased COU or Pathway Specific
Exposures

Quantitative Data Sources

Lifestage

Embryo/fetus

•	Transfer of exposure from the parent
(placenta to fetus)

•	Ratio of placenta: maternal serum (Rpm)
concentrations shown to range from 0.76
for TCEP



• (Wang et al., 2021)

Children

(infants, toddlers)

• EPA did not identify sources of increased
background exposure anticipated for this
lifestage

•	Hand to mouth behavior leads to
increased ingestion of household dust

•	Age-appropriate behavior patterns
(elevated soil ingestion exposure
(children's activities with soil, children
playing mud)

•	Human milk exposure from maternal
doses derived from TSCA sources

•	Different exposure factors

•	Drinking water exposure from TSCA
sources

•	EPA Age Grouping
Guidance

•	Exposure Factors
Handbook (US. EPA.
2017c)

•	See Section 5.1.3.4.7

Geriatric

• Older populations that generally use
supplements may be at higher exposure to
TCEP due to use of Fish oil supplements

• EPA did not identify sources of
increased COU or pathway specific
exposure for this lifestage

• Pom a et al. (2018)

Sociodemo-
graphic /
Lifestyle

Race/Ethnicity

• EPA did not identify sources of increased
background exposure anticipated for this
lifestage

•	TCEP levels in dust are significantly
associated with the presence of
extremely worn carpets; lower
socioeconomic status (SES) populations
are more prone to having homes with
older carpets due to their cost of
replacement

•	Fenceline populations (typically lower
SES) may live closer to emitting sources

• (Castorina et aL, 2017).

Subsistence
Fishing

• EPA did not identify sources of increased
background exposure anticipated for this
lifestage

• Subsistence fishing populations that
consumer more fish have elevated levels
of TCEP exposure

• See Section 5.1.3.4.3

Occupational

Firefighters

• Firefighters may be at increased risk of
TCEP exposures during structure fires
(Mayer et aL, 2021).

• EPA did not identify sources of
increased COU or pathway specific
exposure for firefighters

•	See qualitative discussion
Section 5.3.3

•	(Javatilaka et aL, 2017).

Page 416 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Category

Subcategory

Increased Background Exposure

Increased COU or Pathway Specific
Exposures

Quantitative Data Sources

Consumer

High frequency
consumers

• Non-TSCA source such as dietary

exposures through food, food packaging,
drugs, and personal care products that
contain TCEP

• Consumer products designed for
children (e.g., children's outdoor play
structures, toy foam blocks) may lead to
elevated exposures for children and
infants.

•	Use Report

•	EPA's Exposure Factors
Handbook (Ch. 17)

•	See Sections 5.1.2.2 and
5.1.3.4.8

High duration
consumers

11555

Page 417 of 572

-------
11556

11557

11558

11559

11560

11561

11562

11563

11564

11565

11566

11567

11568

11569

11570

11571

11572

11573

11574

11575

11576

11577

11578

11579

11580

11581

11582

11583

11584

11585

11586

11587

11588

11589

11590

11591

11592

11593

11594

11595

11596

11597

11598

11599

11600

11601

11602

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

D.2 PESS Based on Greater Susceptibility

In this section, EPA addresses subpopulations expected to be more susceptible to TCEP exposure than
other populations. TableApx D-2 presents the data sources that were used in the PESS analysis
evaluating susceptible subpopulations and identifies whether and how the subpopulation was addressed
quantitatively in the risk evaluation of TCEP.

Several conclusions can be made regarding factors that may increase susceptibility to the effects of
TCEP. Limited human data are available on health effects of TCEP and EPA did not identify differences
in susceptibility among human populations. Animal studies identified developmental effects (NTP.
1991a) as well as sensitive sexes for certain health outcomes—higher incidence of neurotoxicity in
female rats (	) and greater sensitivity of male (vs. female) mice in reproductive effects (Chen

et ai. 2015a)—and EPA quantified risks based on these endpoints in the risk evaluation. It is possible
that these differences in rodents reflect differences in humans. However, if sex differences in
susceptibility among rodents are due solely to differences in toxicokinetics, there is uncertainty for
humans given a lack of metabolic differences among sexes in experiments using human liver tissues
(Chapman et ai. 1991).

As identified in Table Apx D-2, many other susceptibility factors that are generally considered to
increase susceptibility of individuals to chemical hazards. These factors include pre-existing diseases,
alcohol use, diet, stress, among others. The effect of these factors on susceptibility to health effects of
TCEP is not known; therefore, EPA is uncertain about the magnitude of any possible increased risk from
effects associated with TCEP exposure.

For non-cancer endpoints, EPA used a default value of 10 for human variability (UFh) to account for
increased susceptibility when quantifying risks from exposure to TCEP. The Risk Assessment Forum, in
A Review of the Reference Dose and Reference Concentration Processes (	32b). discusses

some of the evidence for choosing the default factor of 10 when data are lacking and describe the types
of populations that may be more susceptible, including different lifestages (e.g., of children and elderly).

002b). however, did not discuss all the factors presented in Table Apx D-2. Thus,
uncertainty remains regarding whether these additional susceptibility factors would be covered by the
default UFh value of 10 chosen for use in the TCEP risk evaluation.

For cancer, the dose-response model applied to animal tumor data employed low-dose linear
extrapolation, and this assumes any TCEP exposure is associated with some positive risk of getting
cancer. EPA made this assumption in the absence of an established MOA for TCEP and according to
guidance from U.S. EPA's Guidelines for Carcinogen Risk Assessment (	'005b). Assuming

all TCEP exposure is associated with some risk is likely to be health conservative because EPA does not
believe that a mutagenic MOA is likely for TCEP and a threshold below which cancer does not occur is
expected to exist. However, information is lacking with which to determine an appropriate threshold.
Even though the cancer dose-response modeling assumes any exposure is associated with a certain risk,
EPA presents risk estimates in comparison with benchmark risk levels (1 in 1,000,000 to 1 in 10,000).

Although there is likely to be variability in susceptibility across the human population, EPA did not
identify specific human groups that are expected to be more susceptible to cancer following TCEP
exposure. Other than relying on animal tumor data for the more sensitive sex, the available evidence
does not allow EPA to evaluate or quantify the potential for increased cancer risk in specific
subpopulations, such as for individuals with pre-existing diseases or those who smoke cigarettes. Given

Page 418 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

11603	that a mutagenic mode of action is unlikely, EPA does not anticipate greater cancer risks from early life

11604	exposure to TCEP. Therefore, EPA is not applying an age-dependent adjustment factor.

11605

Page 419 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table Apx

)-2. PESS Evidence Crosswalk for Biological Suscepi

tibility Considerations

Susceptibility
Category

Examples of
Specific
Factors

Direct Evidence this Factor
Modifies Susceptibility to TCEP

Indirect Evidence of Interaction with Target
Organs or Biological Pathways Relevant to
TCEP

Susceptibility Addressed in Risk
Evaluation?

Description of Interaction

Key Citations

Description of Interaction

Key Citation(s)

Lifestage

Embryos/
fetuses/infants

Direct quantitative animal
evidence for developmental
toxicity (e.g., decreased
fertility and live births with
some increased severity in the
second generation).

Lack of effects on
neurodevelopment (doses up
to 90 mg/kg-day)

NTP (1991a)
Moseret al. (2015)





POD for male reproductive endpoints
protective of effects in offspring "

Pregnancy/
lactating status

Rodent dams not particularly
susceptible during pregnancy
and lactation except in one
prenatal study, in which 7 of
30 dams died at 200 mg/kg-
day

NTP (199la)
Hazleton

Laboratories (1983)
Moseret al. (2015)





POD for male reproductive endpoints
protective of effects in dams

Males of

reproductive

age

Reproductive outcomes
(effects on seminiferous
tubules) in adolescent male
mice

Chen et al, (2015a)

Possible contributors to male
reproductive

effects/infertility (see also
factors in other rows):

•	Enlarged veins of testes

•	Trauma to testes

•	Anabolic steroid or illicit
drug use

•	Cancer treatment

CDC (2023b)

POD for this endpoint and study used
to calculate non-cancer risks

Children

Reproductive outcomes
(effects on seminiferous
tubules) in adolescent male
mice

Chen et al, (2015a)





Adolescent animal POD used to
calculate non-cancer risks; other
variability and uncertainty addressed
using default UFh

Elderly

No direct evidence identified







Use of default UFh

Pre-existing
disease or
disorder

Health
outcome/
target organs

No direct evidence identified



Several conditions may
contribute to male
reproductive
effects/infertility:
• Hormone disorders
(hypothalamus/ pituitary
glands)

CDC (2023b)
CDC (2023a)

Use of default UFh

Page 420 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Susceptibility
Category

Examples of
Specific
Factors

Direct Evidence this Factor
Modifies Susceptibility to TCEP

Indirect Evidence of Interaction with Target
Organs or Biological Pathways Relevant to
TCEP

Susceptibility Addressed in Risk
Evaluation?

Description of Interaction

Key Citations

Description of Interaction

Key Citation(s)









• Diabetes, cystic fibrosis,
autoimmune disorders,
certain infections

Viruses such as human
papilloma virus can increase
susceptibility to cancer





Toxicokinetics

Sex differences in
toxicokinetic parameters
might have resulted in
differences in susceptibility.

Herr et al, (1991)
Burka et al, (1991)
Chapman et al,

0





Use of PODs for the more sensitive
sex; Use of default UFh

Lifestyle
activities

Smoking

No direct evidence identified



Heavy smoking may increase
susceptibility for
reproductive outcomes and
cancer.

CDC (2023 a)
CDC (2023b)

Qualitative discussion in this section
(D.2) and this table

Alcohol
consumption

No direct evidence identified



Heavy alcohol use may affect
susceptibility to reproductive
outcomes and cancer.

CDC (2023b)

Qualitative discussion in this section
(D.2) and this table

Physical
Activity

No direct evidence identified



Insufficient activity may
increase susceptibility to
multiple health outcomes.

Overly strenuous activity
may also increase
susceptibility.

CDC (2022)

Qualitative discussion in this section
(D.2) and this table

Sociodemo-
graphic status

Race/ethnicity

No direct evidence identified
(e.g., no information on
polymorphisms in TCEP
metabolic pathways or
diseases associated
race/ethnicity that would lead
to increased susceptibility to
effects of TCEP by any
individual group)







Qualitative discussion in this section
(D.2) and this table

Socioeconomic
status

No direct evidence identified



Individuals with lower
incomes may have worse
health outcomes due to social
needs that are not met,

ODPHP (2023b)

Qualitative discussion in this section
(D.2) and this table

Page 421 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Susceptibility
Category

Sociodemo-
graphic status

Examples of
Specific
Factors

Direct Evidence this Factor
Modifies Susceptibility to TCEP

Indirect Evidence of Interaction with Target
Organs or Biological Pathways Relevant to
TCEP

Susceptibility Addressed in Risk
Evaluation?

Description of Interaction

Key Citations

Description of Interaction

Key Citation(s)







environmental concerns, and
barriers to health care access.





Sex/gender

Males (mice): Potentially
more sensitive regarding
reproductive effects

Females (rats): More sensitive
for neurotoxicity

Metabolism experiments using
liver slices and microsomes
show differences in
metabolism by sex for rats,
but not for humans. Thus,
there is uncertainty regarding
whether human females and
males are susceptible
subpopulations.

NTP (1991a)

Chen et al, (2015a)
Chapman et al,

0





PODs are used in the risk evaluation
for both endpoints.

Nutrition

Diet

No direct evidence identified



Poor diets can lead to chronic
illnesses such as heart
disease, type 2 diabetes, and
obesity.

Obesity can increase
susceptibility to cancer.

c:

DC (2023 a)

Qualitative discussion in this section
(D.2) and this table

c:

DC (2020)

c:

DC (2023c)



Malnutrition

No direct evidence identified



Micronutrient malnutrition
can lead to multiple
conditions that include birth
defects, maternal and infant
deaths, preterm birth, low
birth weight, poor fetal
growth, childhood blindness,
undeveloped cognitive
ability.

Thus, malnutrition may
increase susceptibility to

CDC (2021)

CDC (2023c)

Qualitative discussion in this Section
(D.2) and this table

Page 422 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Susceptibility
Category

Examples of
Specific

Direct Evidence this Factor
Modifies Susceptibility to TCEP

Indirect Evidence of Interaction with Target
Organs or Biological Pathways Relevant to
TCEP

Susceptibility Addressed in Risk
Evaluation?



Factors

Description of Interaction

Key Citations

Description of Interaction

Key Citation(s)











some/all health outcomes
associated with TCEP.





Genetics/

Target organs

No direct evidence identified



Genetic disorders, such as
Klinefelter's syndrome, Y-
chromosome microdeletion,
myotonic dystrophy can
affect male
reproduction/fertility

CDC (2023b)

Use of default UFh to assess
variability among humans

epigenetics

Toxicokinetics

No direct evidence identified



Specific enzymes have not
been identified for TCEP's
metabolic pathways.
Therefore, potential
polymorphisms are not
known.



Use of default UFh to assess
variability among humans



Built

environment

No direct evidence identified



Poor-quality housing is
associated with a variety of
negative health outcomes.

ODPHP (2023a)

Qualitative discussion in this Section
(D.2|| and this table

Other chemical
and

nonchemical
stressors

Social

environment

No direct evidence identified



Social isolation and other
social determinants (e.g.,
decreased social capital,
stress) can lead to negative
health outcomes.

CDC (2023d)
ODPHP (2023c)

Qualitative discussion in this Section
(D.2) and this table

Other chemical
and

nonchemical
stressors

Chemical co-
exposures

An in vitro study of liver cells
co-exposed to TCEP and
benzo-a-pyrene activated
pathways associated with cell
proliferation and inflammation
and increased expression of
pro-inflammatory cytokines,
whereas exposure to TCEP
alone did not.

TCEP showed anti-estrogenic
activity (32 percent inhibition)
in vitro using the breast

Zhang et al, (2017b)
Krivoshiev et al,

(2016)





Qualitative discussion in this Section
(D.2) and this table

Page 423 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Susceptibility
Category

Examples of
Specific
Factors

Direct Evidence this Factor
Modifies Susceptibility to TCEP

Indirect Evidence of Interaction with Target
Organs or Biological Pathways Relevant to
TCEP

Susceptibility Addressed in Risk
Evaluation?

Description of Interaction

Key Citations

Description of Interaction

Key Citation(s)



adenocarcinoma cell line,
MCF-7 after co-expo sure with
17B-estradiol.









"An error in reporting the results in NTP (199 la) precluded using sex ratio; use of this endpoint would have resulted in using a LOAEL of 175 mg/kg-day with an HED of 23.3
mg/kg-day and a benchmark MOE of 300. This would have resulted in similar but slightly greater risk.

11607

11608

Page 424 of 572

-------
11609

11610

11611

11612

11613

11614

11615

11616

11617

11618

11619

11620

11621

11622

11623

11624

11625

11626

11627

11628

11629

11630

11631

11632

11633

11634

11635

11636

11637

11638

11639

11640

11641

11642

11643

11644

11645

11646

11647

11648

11649

11650

11651

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Appendix E PHYSICAL AND CHEMICAL PROPERTIES AND
FATE AND TRANSPORT DETAILS

E.l Physical and Chemical Properties Evidence Integration

The physical and chemical property values selected for use in the risk evaluation for TCEP are given in
Table 2-1. These values were taken from the Final Scope of the Risk Evaluation for Tris(2-chloroethyl)
Phosphate (TCEP) CASRN115-96-8 (	20b). except for physical form, vapor density,

autoflammability, flashpoint, Henry's Law constant, and octanol:air partition coefficient (log Koa).

In the final scope (	3b), no vapor density, log Koa, and autoflammability data were

reported and a flashpoint value from a medium-quality study was provided. After the final scope was
published, vapor density, autoflammability data, and log Koa data were identified in the systematic
review process along with high-quality flashpoint data.

E.l.l Physical Form

In the final scope (	1020b). physical state and physical properties were 2 of 17 endpoints

provided. As provided in the Final Scope of Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP)
Supplemental File - Data Extraction and Data Evaluation Tables for Physical and Chemical Property
Studies (U ,S. EPA. 2020c). only one source was identified and evaluated as a high-quality data for the
physical state endpoint. Ultimately, "liquid" was used in the risk evaluation. For physical properties, two
sources were identified and evaluated as high-quality studies. The reason was not provided, but "clear,
transparent liquid" was preferred and reported over "low odor." For this risk evaluation, both endpoints
were combined and re-named to physical form. After the systematic review process was completed, six
high-quality data were identified and extracted while a medium-quality study was excluded. TCEP is
identified as a clear, transparent liquid with slight odor (DOH. . 'i , I v «« \ b, .009;
Lewis and Hawlev. 2007; Weil. 2001). These descriptions agree with the qualitative description given in
the final scope (' v >1 V \ .^iOb).

E.1.2 Vapor Density

A vapor density data was identified through systematic review. It was from a secondary source, NCBI
(2020) and rated it high-quality. Therefore, the vapor density of 9.8 was included in the risk evaluation.
The primary source of the data is

E.1.3 OctanolrAir Partition Coefficient (Log Koa)

Two high-quality log Koa data were identified through systematic review. Okeme et al. (2020) gave a
log Koa range of 7.85 to 7.93. Yam an et al. (2020) gave a log KOA value of 7.91. Because 7.91 is
within the range of 7.85 to 7.93, the Okeme et al. (2020) data was selected for use in the risk evaluation.

E.l.4 Henry's Law Constant (HLC)

A Henry's Law constant (HLC) of 2.55 10 * atm nrVmol at 25 °C was reported in the final scope (U.S.
E 20b). It was estimated using the Bond method in HENRYWIN™, which is an estimation
method that splits a compound into a summation of the individual bonds that comprise the compound
(	2012d). However, when measured HLC values are not available, a calculated value based on

high-quality measured water solubility and vapor pressure data are typically preferred over an estimated
value (Meylam and Howard. 1991). With a high-quality measured vapor pressure of 0.0613 nimHg and a
water solubility of 7,820 mg/L, the revised HLC is 2,945 x ] 0 6 atmm3/mol at 25 °C. Systematic review
identified two HLC data: one high-quality (Ekpe et al.. 2020) and one medium-quality data (IPCS.
1998). Both data were not included in this draft risk evaluation because a calculated HLC value based on

Page 425 of 572

-------
11652

11653

11654

11655

11656

11657

11658

11659

11660

11661

11662

11663

11664

11665

11666

11667

11668

11669

11670

11671

11672

11673

11674

11675

11676

11677

11678

11679

11680

11681

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

high-quality measured water solubility and vapor pressure data are available for use in the risk
evaluation.

E.1.5 Flash Point

Eight high-quality and four medium-quality flash point data were identified through systematic review.
The flash point data ranged from 200 to 252 °C. In general, flash point is measured using either an open
cup or closed cup technique. The closed cup technique normally gives lower values for the flash point
than open cup (approximately 5 to 10 °C lower). The extracted flash point data include values measured
using both closed cup and open cup techniques and some sources not reporting the technique used. Four
medium-quality data were excluded for this risk evaluation because high-quality flash point data are
available. The 216 °C datum extracted from	and Lewis and Hawlev (2007) was

excluded because the analytical method was not provided and there was no indication that a reliable
method was used. The 202 °C datum extracted from IK |8) was excluded because the data were
extracted from a secondary source without peer review and did not provide a reference of the original
source. The 200 °C datum extracted from	} was excluded because the test sample

appeared to catch fire at approximately 200 °C, but did not show a distinct flash point as defined by the
ASTM D93 method. The 232 °C datum extracted from Toscano and Coleman (2012) and Siema-Aldrich
(2 was excluded because the analytical method used was not reported. Between the remaining two
high-quality flash point data, the 225 °C datum extracted from	t was selected for use in

this draft risk evaluation because flash point is defined as "the lowest temperature at which a chemical
will ignite with an ignition source."

E.1.6 Autoflammability

Three medium-quality autoflammability data were identified through systematic review. The 480 °C
datum extracted from ECB (2009) and	was selected for use in this risk evaluation because

autoflammability is defined as "the lowest temperature at which a chemical will spontaneously combust
without an ignition source." Therefore, the 1,115 °F (-602 °C) datum extracted from	was

excluded.

A composite plot comprising box and whisker plots of reported high-, medium-, and low-quality
physical and chemical property data values are shown in Figure Apx E-l.

Page 426 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE

December 2023



i





-40-







O

O





O

0

lure

on
0





CD



D

2









CTJ

CD
0



~



CD
Q.

E

^ -60-



«





E

CD
1-

R



-70-

350-

300-

250-

-200-

150 -

100-

1.42-

E

o

¦3,1.41-

c

CD

Q 1.40-

1.39-

9.850-

ra 9.825-

C
<1)

9.800-

o

Q.

CD
>

CD

9.775-

J5

CD

cr

Melting Point

Boiling Point

Density

9.750-

Vapor Density

0.5-

0.4 -

I '

E
E

0 0.3 -
zi

0.2-

o

Q.

rc 0.1 -

0.0-

8000-

7000-

_Q

o

0
00

6000-

5000-

~

1.5-

£
o

CD
O

Vapor Pressure

1.0-

0.5-

Water Solubility

log Kow

o
§

_ 1e-06 -

Henry's Law Constant

90th Percentile

75th Percentile

Selected Value

Median
Mean

Datum

25th Percentile
10th Percentile
Legend

250-

.240-

230-

ra
a

^•220-



o

03

| 1.4725-

I

1

1.4700-

Refractive Index

FigureApx E-l„ Box and Whisker Plots of Reported Physical and Chemical Property Data Values

E.2 Fate and Transport

E.2.1 Approach and Methodology

EPA conducted a Tier I assessment to identify the environmental compartments (i.e., water, sediment,
biosolids, soil, groundwater, air) of major and minor relevance to the fate and transport of TCEP. Next, a
Tier II assessment was conducted to identify the fate pathways and media most likely to cause exposure
to environmental releases. Media-specific fate analyses were performed as described in Sections E.2.2,
E.2.3, and E.2.4.

Page 427 of 572

-------
11691

11692

11693

11694

11695

11696

11697

11698

11699

11700

11701

11702

11703

11704

11705

11706

11707

11708

11709

11710

11711

11712

11713

11714

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE

December 2023

E.2.1.1 EPI Suite™ Model Inputs

To set up EPI Suite™ for estimating fate properties of TCEP, the physical and chemical properties were
input based on the values in Table 2-1. EPI Suite™ was run using default settings {i.e., no other
parameters were changed or input) (Figure_Apx E-2).

$$ EPI Suite

^JnJxJ

1

Output

Fugacity

Help

EPI Suite - Welcome Screen

Cleai Input Fields

r Full
<* Summary

l|0=P(0CCCI)(0CCCI)0CCCI

Input Chem Name:|E«hano'. 2-chloro-. phosphate (3:1 J
Name Lookup

Henry LC: | 2.945E-06 atm-m /mole Watei Solubility: j	7820 n»g/L

Melting Point |	li5 Celsius	Vapoi Pressuie: |	0613 mm Hg

Boiling Point: |	330 Celsius	Log Kow:	|	TT78

330 Celsius
Lake

HYDROWIN

Water Depth: |
Wind Velocity: f
Current Velocity: [~~

~r

5 r
~r

meters

meters/sec

meters/sec

EPI Links

The Estimation Programs Interface [EPI) SuiteTM was developed by the US Environmental Protection Agency's Office of Pollution Prevention
and Toxics and Syracuse Research Corporation (SRC). It is a screening-level tool, intended for use in applications such as to quickly screen
chemicals for release potential and "bin" chemicals by priority for future work. Estimated values should not be used when experimental
(measured) values are available.

EPI SuiteTM cannot be used for all chemical substances. The intended application domain is organic chemicals. Inorganic and organometallic
chemicals generally are outside the domain.

Important information on the performance, development and application of EPI SuiteTM and the individual programs within it can be
found under the Help tab. Copyright 2000-2012 United States Environmental Protection Agency for EPI SuiteTM and all component
programs except BioHCWIN and KOAWIN.

FigureApx E-2. Screen Capture of EPI Suite™ Parameters Used to Calculate Fate and Physical
and Chemical Properties for TCEP

E.2.1.2 Fugacity Modeling

Because no current data were being reported to the TRI or DMR, TCEP releases to the environment
could not be estimated. The approach described by Mackav et al. (19961 using the Level III Fugacity
Model in EPI Suite1X1 (LEV3EPIIM) was used for this Tier II analysis. LEV3EPI1M is described as a
steady-state, non-equilibrium model that uses a chemical's physical and chemical properties and
degradation rates to predict partitioning of the chemical between environmental compartments and its
persistence in a model environment ( J.S. EPA. 2012d). TCEP's physical and chemical properties were
taken directly from Table 2-1. Environmental release information is useful for fugacity modeling
because the emission rates will predict a real-time percent mass distribution for each medium. Instead,
environmental degradation half-lives were taken from high-quality studies that were identified through
systematic review to reduce levels of uncertainties. Based on TCEP's environmental half-lives,
partitioning characteristics, and the results of Level III Fugacity modeling (Figure Apx E-3), TCEP is
expected to be found predominantly in water or soil, depending on the media of release. The
LEV3EPI™ results were consistent with environmental monitoring data. Further discussion of TCEP
partitioning can be found in Sections E.2.2, E.2.3, and E.2.4.

Page 428 of 572

-------
11715

11716

11717

11718

11719

11720

11721

11722

11723

11724

11725

11726

11727

11728

11729

11730

11731

11732

11733

11734

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

100%

100% Soil Release 100% Air Release 100% Water Release Equal Releases
Air ¦ Water ¦Soil 1 Sediment

FigureApx E-3. EPI Suite™ Level III Fugacity Modeling Graphical Result for TCEP

E.2.1.3 OECD Pov and LRTP Screening Tool

TCEP's long-range transport potential (LRTP) was evaluated by using OECD's Overall Environmental
Persistence (Pov) and LRTP Screening Tool (Version 2.2) (Wegmann et al.. 2009). The OECD POV and
LRTP Tool is in a spreadsheet format containing multimedia chemical fate models that were designed
based on the recommendations of the OECD expert group to estimate environmental persistence and
LRTP of organic chemicals at a screening level. With a chemical's physical and chemical properties, the
OECD POV and LRTP Tool will be able to predict its Pov, characteristic travel distance (CTD), and
transfer efficiency (TE). Pov is the overall persistence in the whole environment in days, CTD quantifies
the distance in kilometers (km) from the point of release to the point at which the concentration has
dropped to 1/e, or approximately 37 percent of its initial value, and TE estimates the percentage of
emitted chemical that is deposited to surface media after transport away from the region of release. The
OECD Pov and LRTP Screening Tool calculates two emission scenarios specific CTD values, for
emissions to air and water. Only transport in the medium that receives the emission is considered, thus
CTD in air is calculated from the emission-to-air scenario and CTD in water is calculated from the
emission-to-water scenario. No CTD is calculated for emissions to soil because soil is not considered to
be mobile. The physical and chemical properties were input based on the values in Table 2-1 and Table
2-2 (Figure_Apx E-4). The modeling results will be discussed further in Sections E.2.2 and E.2.3.1.

Page 429 of 572

-------
11735

11736

11737

11738

11739

11740

11741

11742

11743

11744

11745

11746

11747

11748

11749

11750

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

OECD POV & LRTP
Screening Tool*

Main Menu

Help

Preferences

Select chemicals to evaluate

Simultaneous runs of one database and one chemical are possible.

Databases



Reference Chemicals
Generic PCB Hcnnologues^

A





V







Deselect | Manage DB» |

Database Status:

~

Single Chemical







Monte Carlo Parameters

Name

TCEP





Dispersion factors

Molecular mass

285.49





for each property

Log K3„

-3.919



~

~~5 "

Log Kbk

1/78



~

5 "

Half life in air (h)

5.80E+00



~

10

Half life in water (h)

1.00E+04



~

10

Half life in soil (h)

4.25E+02



~

10



Clear

I



Reset



Chemical Status:

~



Calculate

Include Monte Carlo Analysis for Single Chemical

Color Codes



H

Results already present

~

No warnings: calculation possible

~

Warnings: calculation still possible

¦

Errors: calculation impossible

~

No data entered

' A manual describing this software is provided or the Help page.

FigureApx E-4. Screen Capture of OECD Pov and LRTP Screening Tool Parameters Used to
Calculate TCEP's LRTP

E.2.1.4 Evidence Integration

A brief description of evidence integration for fate and transport is available in the 2021 Draft
Systematic Review Protocol (U.S. EPA. 2021). Additional details on fate and transport evidence
integration are provided here.

The environmental fate characteristics given in Appendix C of the Final Scope of the Risk Evaluation
for Tris(2-chloroethyl) Phosphate (TCEP) CASRN115-96-8 (U.S. EPA. 2020b) were identified prior to
completing the systematic review. The following sections summarize the findings and provide the
rationale for selecting the environmental fate characteristics that was given in Table 2-2.

E.2.2 Air and Atmosphere

TCEP in its pure form is a liquid at environmental temperatures with a melting point of-55 °C (DOE.
2016; U.S. EPA. 2015a. b; Toscano and Coleman. 2012) and a vapor pressure of 0.0613 mmHg at 25 °C
(U.S. EPA. 2019b; Dobry and Keller. 1957). The log Koa range of 7.5 to 7.98 indicates that TCEP is

Page 430 of 572

-------
11751

11752

11753

11754

11755

11756

11757

11758

11759

11760

11761

11762

11763

11764

11765

11766

11767

11768

11769

11770

11771

11772

11773

11774

11775

11776

11777

11778

11779

11780

11781

11782

11783

11784

11785

11786

11787

11788

11789

11790

11791

11792

11793

11794

11795

11796

11797

11798

11799

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

expected to adsorb to the organic carbon present in airborne particles (Okeme et ai. 2020; Ji et ai. 2019;

Wang etal. 2017b).

As an SVOC, TCEP will exist in both the gas and particle phases (Wane et ai. 2020a; Okeme. 2018;
TERA. 2015). Results from air monitoring studies reported concentrations of gaseous TCEP up to 6,499
pg/m3 (Ma et ai. 2021; Wu et ai. 2020) and particle bound TCEP up to 2,100 pg/m3 in North America
(Wu et ai. 2020; Abdollahi et ai. JO I ; Sal am ova et ai. 2016; Sal am ova et ai. 2014; Shoeib et ai.
2014). Multiple studies have identified urban sources as sources of TCEP in the environment through
fugitive emissions to air (Abdollahi et ai. 2017; Luo et ai. JO I \ Vtotter et ai. 2011). Although the
exact sources of TCEP emissions from urban environment are unknown, they are likely the articles that
were treated with or containing TCEP (Abdollahi et ai. JO I ; Luo et ai. 2015; Wei et ai. 2014; Moller
et ai. 2011; Aston et ai. 1996).

Compared to outdoor air, TCEP concentrations are significantly higher in indoor air because TCEP has
the potential to volatilize from treated products and diffuse into air, as well as partition onto dust due to
its use as an additive (Oi et ai. 2019;.h » \ n , Liu et ai j'li, \ r Ok :0C; H 2009; NICNAS.
2001). In northern California, indoor air concentrations of TCEP were detected up to 15,340 pg/m3
(Bradman et ai. 2014) and dust concentrations was measured up to 6.84 |ig/g (Bradr	). In

addition, TCEP is a known impurity in 2,2-bis(chloromethyl)-propane-l,3-diyltetrakis(2-chloroethyl)
bisphosphate (V6) commercial mixtures that are primarily used in furniture and automobile foam.

Higher concentrations of TCEP (up to 50.12 jug/g) were found in dust samples that were collected from
the surfaces of the front and back seats of automobiles in Boston, MA (Fane et ai. 2013).

TCEP is not expected to undergo significant direct photolysis in the atmosphere because its chemical
structure does not absorb light at wavelengths greater than 290 nm (H.SDB. 2015). TCEP in the gaseous
phase is expected to degrade rapidly by reaction with photochemically produced hydroxyl radicals
(•OH) in the atmosphere. A half-life of 5.8 hours was calculated from the AOPWIN module in EPI
Suite™ using an estimated rate constant of 2,2/10 " cm3/molecules-second at 25 °C, assuming an
atmospheric hydroxyl radical concentration of 1.5><106 molecules/cm3 and a 12-hour day (

2012d). The atmospheric half-life of TCEP does not pertain to indoor environments due to lower
hydroxyl radical concentrations, less mixing of air, and lower sunlight intensity.

TCEP has been detected in air and snow in remote locations such as the Arctic and Antarctica (Na et ai.
2020; Wane et ai. 2020a; Xie et ai. 2020; Rauert et ai. 2018; Li et ai. 2017b; Silhrine et ai. 2016;
Chene et ai. 2013b; Moller et ai. 2012; NIVA. 2008). Particle-bound TCEP was found to be highly
persistent in the atmosphere and had slower rates for the reaction with hydroxyl radicals due to the
presence of atmospheric water (Wu et ai. 2020; Li et ai. 2017a; Liu et ai. 2014). Particle-bound TCEP
is primarily removed from the atmosphere by wet or dry deposition. Based on its physical and chemical
properties and short half-life in the atmosphere (ti/2 = 5.8 hours), TCEP was assumed to be not persistent
in the air (	1). The OECD Pov and LRTP Screening Tool was run to get additional

information on TCEP's long-range transport potential in the air. For TCEP emissions in air, a Pov of 11
days, CTD of 118 km (-73 miles), and TE of 0.0142 percent were given using a molecular mass of
285.49 g/mol, log Kaw of-3.919, and log Kow of 1.78 along with atmospheric half-life of 5.8 hours,
water half-life of 10,000 hours, and soil half-life of 424.8 hours (Figure Apx E-4). A CTD of 118 km
(-73 miles) suggests that TCEP does not have the potential to undergo long-range transport in the air
and a TE of 0.0142 percent suggests that negligible fraction of TCEP emitted to air will be deposited to
surface media such as water. CTD can also be calculated using the LEV3EPI™ module in EPI Suite™
without considerations for advection (U.S. EPA. 2012d; Beyer et ai. 2000). After entering TCEP's
physical and chemical properties (Figure Apx E-2), a CTD of 238 km (-148 miles) was calculated.

Page 431 of 572

-------
11800

11801

11802

11803

11804

11805

11806

11807

11808

11809

11810

11811

11812

11813

11814

11815

11816

11817

11818

11819

11820

11821

11822

11823

11824

11825

11826

11827

11828

11829

11830

11831

11832

11833

11834

11835

11836

11837

11838

11839

11840

11841

11842

11843

11844

11845

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Particle-bound TCEP has the potential to undergo long-range atmospheric transport (LRAT) and it is
likely the reason why TCEP is found in the Arctic and other remote locations with no source of releases.
TCEP's LRTP could be crucially underestimated when using gaseous phase atmospheric half-life in
multimedia models like the OECD Pov and LRTP Screening Tool.

E.2.3 Aquatic Environments

Wastewater treatment effluent, atmospheric deposition, air-water gaseous exchange, and runoff have
been identified as sources of TCEP detected in aquatic and marine environments, especially in urban
areas (Ma et at.. 2021; Crisul' ^ at.. 2019; Guo et at.. 2017a; Kim et at.. 201 ).

E.2.3.1 Surface Water

TCEP is not expected to undergo abiotic degradation processes such as hydrolysis and photolysis in
aquatic environments under environmentally relevant conditions. The rate of hydrolysis will be highly
dependent on pH and temperature. TCEP showed no significant hydrolysis over 35 days at pH levels of
7, 9, and 11 at 20 °C, but an extensive degradation occurred when the pH level was adjusted to 13 (ti/2 =
0.083 days) (Su et at.. 2016). A hydrolysis study by Saint-Hilaire et at. (2011) observed the pH-
dependent hydrolysis of TCEP between pH 8 to 13 at 50 °C and confirmed that TCEP's hydrolysis rates
increased as pH levels increased. TCEP's hydrolysis half-life was estimated to be approximately 2 years
at pH level of 8 at 25 °C. In addition, TCEP's hydrolysis rates also increased in the presence of reduced
sulfur species. The calculated half-lives for TCEP after reacting with 5.6 mM bisulfide (HS~) and 0.33
mM polysulfides (S were 90 and 30 days, respectively. The results also indicated that the three
reduced sulfur species reacted with TCEP in a nucleophilic substitution reaction with bis(chloroethyl)
phosphate (BCEP) being the major transformation product. The hydrolysis half-lives estimated by
QSAR models were found to be inconsistent with experimental values. HYDROWIN™, an aqueous
hydrolysis rate program in EPI Suite™, estimated TCEP's half-life to be approximately 20 days at pH 5
to 9 and approximately 17 days at pH 10 (	). However, the half-life values from

HYDROWIN™ were not included in this draft risk evaluation because the half-life values from high-
quality hydrolysis studies mentioned above are available. In addition, it is unlikely for TCEP undergo
indirect photolysis. No photolytic degradation was observed after exposing TCEP to natural sunlight for
15 days in lake water (Regnery and Puttmann. 2010a). Other experimental studies also observed no
photolytic degradation (Chen et at.. 2019b; Lee et at.. 2014; Watts and Linden. 2009. 2008).

For biotic degradation in water, TCEP is not readily biodegradable under aerobic conditions. In a ready
biodegradability test using the Modified Sturm test (OECD 301B), TCEP showed a minimal degradation
after 28 days and the cumulative carbon dioxide production was negligible (Life Sciences Research Ltd.
1990b). In another ready biodegradability test using the Closed Bottle test (OECD 301D), TCEP was not
readily biodegradable (Life Sciences Research Ltd. 1990c). Based on these two biodegradation studies,
rapid biodegradation of TCEP is not likely when it is released to surface water.

A limited number of test results on anaerobic biodegradability of TCEP were available. Previous
assessments of TCEP reported that no degradation was observed for TCEP in an anaerobic
biodegradation study after 58 days using ISO DIS 11734, which is equivalent to OECD 311 (

2015a; EC. 2009). This result was not selected for use in the risk evaluation because the original study
by Noac	was published in German; therefore, it did not undergo the systematic review process.

Another study, Kawaeoshi et at. (2002) reported that TCEP did not undergo biodegradation under
anaerobic condition after 60 days using leachate from a sea-based solid waste disposal site in Japan.

This study was not selected for use in the risk evaluation because it was rated as a medium-quality study
since critical information on test conditions was not included and there was insufficient evidence to

Page 432 of 572

-------
11846

11847

11848

11849

11850

11851

11852

11853

11854

11855

11856

11857

11858

11859

11860

11861

11862

11863

11864

11865

11866

11867

11868

11869

11870

11871

11872

11873

11874

11875

11876

11877

11878

11879

11880

11881

11882

11883

11884

11885

11886

11887

11888

11889

11890

11891

11892

11893

11894

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

confirm that TCEP disappearance was not likely due to other processes. Due to lack of anaerobic
biodegradation studies on TCEP, no anaerobic biodegradation data were selected for this risk evaluation.

Two studies showed that TCEP was able to undergo volatilization from oceans and had the highest
water-to-air emission flux in two monitoring studies. In Li et al. (201 M, TCEP volatilization from
seawater to air was seen in all samples across the North Atlantic and the Arctic, and equilibrium was
reached in some samples that was caused by relatively low TCEP concentrations in seawater. A similar
result was seen in another air-water gaseous exchange study on a coastal site where TCEP had the
highest emission flux in water (Wane et al.. 2018b). Both studies suggest that the air-water gaseous
exchange is an important process for TCEP to transport between the air and water, causing a secondary
pollution. TCEP's volatilization behavior did not align with its physical and chemical properties and
modeling prediction. A low Henry's Law constant of 2.945x 10~6 atmm3/mol at 25 °C (Table 2-1)
indicates that TCEP is not expected to volatilize from surface water (TERA. 2015; Toscano and
Coleman. 2012; Reenerv and Puettmann. 2009; Dobry and Keller. 1957). HLC is equivalent to an
air:water partitioning coefficient (Kaw) of 1.21 /10 4 or log Kaw of -3.19 at 25 °C, which indicates that
TCEP will favor water over air (1 c. « ^ \ JO I :
-------
11895

11896

11897

11898

11899

11900

11901

11902

11903

11904

11905

11906

11907

11908

11909

11910

11911

11912

11913

11914

11915

11916

11917

11918

11919

11920

11921

11922

11923

11924

11925

11926

11927

11928

11929

11930

11931

11932

11933

11934

11935

11936

11937

11938

11939

11940

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

end up as litter where TCEP are released into the open environment. Extreme events such as storms,
floods, cyclones, tidal waves, and tsunamis, are also a significant immediate source of land-based plastic
debris. Plastic wastes containing TCEP can potentially migrate from the plastic product to water by the
weathering of microplastics (Hahladakis et al. 2018). Because TCEP has primarily been used as an
additive flame retardant and plasticizers, they can easily leach from plastic wastes. Furthermore, plastic
debris (e.g., macroplastics, microplastics) could act as carriers for TCEP. The high specific surface areas
of microplastics make them a good sorbent for hydrophobic and hydrophilic organic chemicals (Zhang
et al.. 2018a). Widely used plastics such as polyvinyl chloride (PVC) and polyethylene (PE) sorb
organic pollutants from seawater after they are exposed to environmental conditions (Takada and
Karapanagioti. 2019). In Chen et al. (2019a) TCEP was seen to sorb onto PVC and PE microplastics in
seawater. When the temperature was in the range of 5 to 15 °C, the adsorption capacity of TCEP
increased with increasing temperature, but when the temperature was greater than 15 °C, the adsorption
capacity decreased with increasing temperature. Through adsorbing pollutants from surrounding
seawater, microplastics can accumulate and increase the concentrations of pollutants up to the order of
106 (Mato et al.. 2001). Plastic wastes are found in the ocean all over the world and they can travel long
distances, especially to remote regions.

Based on the findings provided above, TCEP has the potential undergo long-range transport in water and
its LRTP could be underestimated when using multimedia models like the OECD Pov and LRTP
Screening Tool.

E.2.3.2 Sediments

TCEP can be transported to sediment from overlying surface water by advection and dispersion of
dissolved TCEP and by deposition of suspended solids containing TCEP. However, it is likely that
TCEP concentrations in overlying water would be higher than in sediment due to its high water
solubility (7,820 mg/L) (Lee et al.. 2018; Ma et al.. 201 ; Kundsma et al.. 2015; Cao et al.. JO IJ).
Higher concentrations of TCEP in sediment are expected to be found at potential source locations (e.g.,
near urban and industrialized areas) (Chokwe and Okonkwo. 2019; Tan et al.. 2019; Lee et al.. 2018;
Wane et al.. 2018a; Cao et al.. .01 ; Maruya et al.. l^t . i uMale -n jI J0l'<).

No anaerobic biodegradation studies were identified. The rate of biodegradation in sediments can be
estimated by extrapolation from aerobic biodegradation testing or estimated by considering that the rate
of anaerobic degradation is typically at least four times slower (64 FR 60197) and up to 9 times slower
than aerobic degradation (	2012d). For the water compartment, TCEP did not pass a ready

biodegradability test (OECD 30IB) (Life Sciences Research 1 id K^Ob) (Table 2-2), so a water half-life
of 10,000 hours was given (	00a). Considering that the rate of anaerobic degradation is 4 to

9 times slower than aerobic biodegradation, the predicted half-life of TCEP would be 40,000 to 90,000
hours in the sediment compartment.

E.2.4 Terrestrial Environments

TCEP is released to terrestrial environments via land application of biosolids, disposal of solid waste to
landfills, and atmospheric deposition.

E.2.4.1 Soil

Based on its range of log Koc values (Table 2-2), TCEP accumulation in soil is expected to be unlikely.
Due to its high water solubility (7,820 mg/L), dissolved TCEP in the soil may be mobile and eventually
migrate to groundwater (see Section E.2.4.2). TCEP in the soil was seen to be vertically transported to
deeper soil horizons, causing TCEP concentration in the surface soil to be lower (He et al.. 201 ;
Bacaloni et al.. 2008). Zhang et al. (2022) reported that higher levels of TCEP was found deeper in the

Page 434 of 572

-------
11941

11942

11943

11944

11945

11946

11947

11948

11949

11950

11951

11952

11953

11954

11955

11956

11957

11958

11959

11960

11961

11962

11963

11964

11965

11966

11967

11968

11969

11970

11971

11972

11973

11974

11975

11976

11977

11978

11979

11980

11981

11982

11983

11984

11985

11986

11987

11988

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

soil (30 to 80 cm) compared to the surface soil samples (0 to 20 cm). Mihailovic and Fries (
reported a similar result in its study.

The estimated log Koc value for TCEP is 2.59, using the molecular connectivity index (MCI) method in
KOCWIN™ (	d). The estimated value from EPI Suite™ was not included in this risk

evaluation because the log Koc values from high-quality field studies are available.

There was only one high-quality study on TCEP degradation in soil. Hurtado et al. (1 studied the
degradation of TCEP in an agricultural soil from Spain. The soil had a sandy texture (90 percent sand, 8
percent silt, and 2 percent clay) and a total organic carbon content of 5 g/kg. After 40 days, 78 percent of
TCEP degraded under aerobic conditions at test substance concentration of 50 |ig/kg. A half-life of 17.7
days (Table 2-2) was estimated based on second-order kinetics. Another soil degradation study was
identified, but this study was evaluated as low-quality (YECB. 2009). citing (Brodsky et al.. 1997)). The
primary degradation of TCEP at a concentration of 5 mg/kg soil was conducted in a laboratory test
system with standard soil for 100 days. The degradation kinetic curve was fitted to a 2nd order square
root function resulting in a DT50 of 167 days and DT90 of »100 days. In addition, TCEP was seen to
be slightly mobile in a leaching test. However, this study was not included in this risk evaluation
because the testing conditions, inoculum information, sampling and analytical methods were not
reported and the omissions likely had an impact on the study results.

TCEP in soil can re-volatilize from contaminated soil into the atmosphere causing a secondary pollution.
A Henry's Law constant of 2.945x 10~6 atmm3/mol at 25 °C, calculated based on a vapor pressure of
0.0613 mmHg and a water solubility of 7,820 mg/L at 25 °C, indicates that TCEP is not expected to
volatilize from dry soil but possibly from moist soil (ATSDR. 2012; Toscano and Coleman. 2012;
Regnery and Puettmann. 2009; Dobry and Keller. 1957). Yet, there are field studies showing that TCEP
underwent an air-soil exchange. In Wang et al. (2020b). the air-soil exchange behavior of TCEP varied
between locations. TCEP was observed to be at an air-soil exchange equilibrium in the suburban and
rural areas, but net volatilization occurred in the urban area. The highest volatilization flux was found at
a site near a bus terminal. Yadav et al. (2018) reported net volatilization from soil to the air as TCEP's
principal process in air-soil exchange. Han et al. (2020) reported a net volatilization in a sampling site
located in the Arctic.

Also, several studies have reported that atmospheric deposition of TCEP may have contributed to soil
contamination since there were no point sources nearby (Ji et al.. 2019; Ren et al.. 2019; Fries and
Mihailovic. 201 I; Mihailovic et al.. 2011). In Bacaloni et al. (2008). lake water samples were collected
from three remote volcanic lakes in central Italy. The three lakes were specifically chosen because there
were no local contamination sources (e.g., tributaries, industries, sewage treatment plants) nearby.
Therefore, the possible sources of contamination would be from local anthropogenic activities, long-
range transport and deposition from rainfall, or runoff processes. TCEP was detected in all three lakes at
the ng/L level and the maximum concentrations occurred during the late summer to autumn months
(August to October), which coincides with higher tourism activity and vehicular traffic at all three
locations. In Han et al. (2020). the net deposition from air to soil was found to be predominant in four
out of five sampling sites in the Arctic.

E.2.4.2 Groundwater

There are two sources of TCEP in the environment that may contaminate groundwaters. Point sources
include wastewater effluents and landfill leachates and are discussed in Sections E.2.5.2 and E.2.4.3.
Diffuse sources include storm water runoff and runoff from biosolids applied to agricultural land and are
discussed in sections E.2.3.1 and E.2.4.4.

Page 435 of 572

-------
11989

11990

11991

11992

11993

11994

11995

11996

11997

11998

11999

12000

12001

12002

12003

12004

12005

12006

12007

12008

12009

12010

12011

12012

12013

12014

12015

12016

12017

12018

12019

12020

12021

12022

12023

12024

12025

12026

12027

12028

12029

12030

12031

12032

12033

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Municipal solid waste landfills (MSWLFs) can be a source of TCEP groundwater contamination.
Historic landfills are more likely to lack the infrastructure of modern landfills, such as liners, leachate
collection systems, and reactive barriers, which would prevent leachate from entering the groundwater
system (Propp et ai. 2021; Lapworth et ai. 2012; Barnes et ai. 2004).

Propp et al. (2021) assessed contaminants of emerging concern in leachate-impacted groundwater from
20 closed MSWLFs in Ontario, Canada. Those "historic" landfills had been closed for at least three
decades. High concentrations of TCEP were reported in groundwater up to 2.92 |ig/L. In addition,
Buszka et al. (2009) collected groundwater samples from a domestic well located in a neighborhood east
of the Himco Dump, which is an unlined landfill that was used for commercial, industrial, medical, and
general waste disposal from 1960 to 1976 in Elkhart, Indiana. TCEP concentration ranged from 0.65 to
0.74 |ig/L. Both studies suggests that TCEP in landfill impacted groundwater was resistant to biotic and
abiotic degradation processes and is very persistent. Barnes et al. (2004) collected groundwater samples
from a historic landfill in central Oklahoma. The landfill was unlined and built adjacent to the Canadian
River in 1920, then covered with a clay cap and vegetated when it was permanently closed in 1985.
TCEP concentration of 0.36 |ig/L was measured in a well that was 3.28 feet away from the landfill.
However, TCEP concentration of 0.74 |ig/L was measured in a well that was 305 feet away from the
landfill. This shows that TCEP has the potential to be transported away from point sources and enter the
groundwater.

E.2.4.3 Landfills

TCEP is not considered a hazardous waste, so it is not listed under Subtitle C of the Resource
Conservation and Recovery Act (RCRA) (40 CFR 261). Solid waste containing TCEP can be disposed
in MSWLFs or industrial waste landfills {i.e., construction and demolition [C&D] debris landfills).
MSWLFs that were built after 1991 are required to use a composite liner and a leachate collection
system. The composite liner includes a minimum of 30-mil flexible membrane liner (FML) overlaying a
two-foot layer of compacted soil lining the bottom and sides of the landfill (40 CFR 258.40). It is
expected that solid waste containing TCEP will be disposed to a lined landfill with a leachate collection
system. However, historic landfills are likely to lack the infrastructure of modern landfills, such as
liners, leachate collection systems, and reactive barriers (Propp et al.. 2021; Lapworth et al.. 2012;
Barnes et al.. 2004). Leachate-impacted groundwater in historic landfills is discussed in Section E.2.4.2.

As mentioned in Section 2.2.2, TCEP is primarily used as an additive plasticizer and flame retardant.
When used as an additive, TCEP is added to manufactured materials via physical mixing rather than
chemical bonding (Oi et al. 2019; Liu et al. 2014; ATSDR. 2012; EC. 2009; NICNAS. 2001).
Consequently, it is highly likely that TCEP will be released from the solid wastes and enter the leachate.
Leachates from 11 landfill sites in Japan reported TCEP concentrations in the range of 6 to 30,100 ng/L
(Yasuhara et al. 1999). The maximum concentration of TCEP was reported in a landfill that consisted
of waste plastics, waste combustion residue, plants, and domestic incombustible wastes. Several other
studies also showed high concentrations of TCEP in leachate samples collected from MSWLFs in the
United States and China (Oi et al. 2019; Deng et al. 2018a; Masoner et al. 2016; Masoner et al.
2014b).

Landfill leachate can be discharged to WWTPs and the release of TCEP to surface water from treated
landfill leachate will depend on the removal of TCEP during wastewater treatment (see Section
E.2.5.2.). The fate and transport of TCEP entering the surface water is discussed in Section E.2.3.1.

Page 436 of 572

-------
12034

12035

12036

12037

12038

12039

12040

12041

12042

12043

12044

12045

12046

12047

12048

12049

12050

12051

12052

12053

12054

12055

12056

12057

12058

12059

12060

12061

12062

12063

12064

12065

12066

12067

12068

12069

12070

12071

12072

12073

12074

12075

12076

12077

12078

12079

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

E.2.4.4 Biosolids

Sludge is defined as the solid, semi-solid, or liquid residue generated by wastewater treatment processes.
The term "biosolids" refers to treated sludge that meet the EPA pollutant and pathogen requirements for
land application and surface disposal (40 CFR 503).

Because TCEP is resistant to degradation in wastewater treatment, some residual concentrations of
TCEP may be present in biosolids and transferred to surface soil during land application. TCEP
concentrations up to 317 ng/g dry weight were detected in sewage sludge collected from wastewater
treatment plants located in the United States (Wane et ai. 2019c; Kim et al. 2017). An anaerobic
digestion study using sewage sludge showed that TCEP was persistent under anaerobic conditions (Pane
et al.. 2018). It is likely that dissolved TCEP will eventually reach surface water via runoff after the land
application of biosolids due to its persistence.

E.2.4.5 Key Sources of Uncertainty

There are significant differences between the predicted and the field observed log Koc values. The
predicted log Koc values are generally lower than the ones reported from field studies. The log Koc
reported in previous assessments of TCEP were in the range of 2.04 to 2.59 ( h ^ \	\ T

2012; EC. 2009; ECB. 2009; NICNAS. 2001). Koc values within this range are associated with low
sorption to soil and will be able to migrate to groundwater. However, a range of 2.5 to 4.3 was obtained
from several field studies (Awonaike et al.. 2021; Zhang et al.. 2021; Wane et al.. 2018a; Zhang et al..
2018b). Log Koc within this range are associated with moderate to strong sorption to soil, sediment, and
suspended solids.

E.2.5 Persistence Potential of TCEP

Biotic and abiotic degradation studies have shown TCEP to be persistent. In the atmosphere, TCEP in
the gaseous phase will be degraded by reacting with hydroxyl radicals (*OH), but particle-phase TCEP
will not be degraded (see Section E.2.2). TCEP does not undergo hydrolysis under environmentally
relevant conditions and is persistent in water (see Section E.2.3.1), sediment (see Section E.2.3.2), and
soil (see Section E.2.4.1). Using the Level III Fugacity model in EPI Suite™ (LEV3EPI™) (see Section
E.2.1.2), TCEP's overall environmental half-life was estimated to be approximately 168 days (U.S.

). Therefore, TCEP is expected to be persistent in the atmosphere as well as aquatic and
terrestrial environments.

E.2.5.1 Destruction and Removal Efficiency

Destruction and removal efficiency is a percentage that represents the mass of a pollutant removed or
destroyed in a thermal incinerator in relative to the mass that entered the system. EPA requires that
hazardous waste incineration systems destroy and remove at least 99.99 percent of each harmful
chemical in the waste, including treated hazardous waste (46 FR 7684).

Only one study was identified in regard to thermal treatment and open burning of articles containing
TCEP. Li et al. ( ; reported that the articles released TCEP in the range of 9,800 to 49,000 ng/g
after undergoing thermal treatment at 300 °C for 150 minutes. For open burning, the articles released
TCEP in the range of 1,000 to 2,600 ng/g after being exposed to an open flame for three minutes at 800
to 1,350 °C. These results showed that TCEP was not completely destroyed. This was to be expected
since flame retardant-containing materials are known to have reduced flammability, which can result in
incomplete combustion.

When undergoing thermal degradation in air at 220 °C and higher, TCEP will rapidly decompose to
produce numerous toxic byproducts, including 1,2-dichloroethane (C2H4CI2), vinyl chloride (C2H3CI),

Page 437 of 572

-------
12080

12081

12082

12083

12084

12085

12086

12087

12088

12089

12090

12091

12092

12093

12094

12095

12096

12097

12098

12099

12100

12101

12102

12103

12104

12105

12106

12107

12108

12109

12110

12111

12112

12113

12114

12115

12116

12117

12118

12119

12120

12121

12122

12123

12124

12125

12126

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

hydrogen chloride (HC1), carbon monoxide (CO), and acetaldehyde (C2H4O), among others (

2015a: NIC/HAS. 2001; Ml 4; Pactorek et al. 1978V

Because open burning can contribute to the emission of TCEP or other toxic byproducts to the
surrounding environment (Matsukami et al.. 2015). thermal treatment and open burning are not
favorable options for the disposal of TCEP.

E.2.5.2 Removal in Wastewater

Wastewater treatment is performed to remove contaminants from wastewater using physical, biological,
and chemical processes. Generally, municipal wastewater treatment facilities apply primary and
secondary treatments. During the primary treatment, screens, grit chambers, and settling tanks are used
to remove solids from wastewater. After undergoing primary treatment, the wastewater undergoes a
secondary treatment. Secondary treatment processes can remove up to 90 percent of the organic matter
in wastewater using biological treatment processes such as trickling filters or activated sludge.
Sometimes an additional stage of treatment such as tertiary treatment is utilized to further clean water
for additional protection using advanced treatment techniques (e.g., ozonation, chlorination,
disinfection). A negative removal efficiency can be reported if the pollutant concentration is higher in
the effluents than the pollutant concentration in the influents.

Because TCEP is not readily biodegradable under aerobic conditions based on two ready
biodegradability tests (Life Sciences Researc	3b, c), it is not expected to be removed from

wastewater by biodegradation. This conclusion is supported by STPWIN™, an EPI Suite™ module that
estimates chemical removal in sewage treatment plants. STPWIN™ estimated that a total of 2.23
percent of TCEP in wastewater will be removed: 0.08 percent by biodegradation, 0.17 percent by air
stripping, and 1.99 percent by sorption to sludge (	i). STPWIN™ simulates a

conventional wastewater treatment plant that uses activated sludge secondary treatment. The
biodegradation half-life parameter was set to 10,000 hours for the primary clarifier, aeration vessel, and
settling tank, which is a default for recalcitrant chemicals. The physical and chemical properties for
TCEP given in Table 2-1 were used (Figure Apx E-2). The results from STPWIN™ were not included
in this draft risk evaluation because high-quality wastewater treatment studies are available.

A total of 19 wastewater treatment studies were identified during systematic review. Seven studies were
evaluated and rated as medium-quality studies. These studies were not included in this draft risk
evaluation. Numerous high-quality wastewater treatment studies reported either a negative removal
efficiency or a removal of less than 10 percent for TCEP after undergoing primary and secondary
treatments. An overall TCEP removal of-60.2 percent was calculated for a municipal wastewater
treatment in Frankfurt, Germany (Fries and Puttmann. 2001). An average overall TCEP removal of
-32.2 percent was calculated from the removals reported for five activated sludge treatment plants in
Catalonia, Spain (Cristale et al.. 2016).

An TCEP removal of-18.9 percent removal was calculated for a municipal wastewater treatment plant
in Beijing, China (Liang and Liu. 2016). TCEP was not removed (0 percent) in two activated sludge
treatment plants in western Germany (Meyer and Bester. 2004) and an activated sludge treatment plant
in South Korea (Kim et al.. 2007). An overall TCEP removal of 9 percent was calculated from the
removals reported for two small-sized, three medium-sized, and two large-sized municipal sewage
treatment plants in Sweden (Marklund et al.. 2005a). An overall TCEP removal of-19.1 percent was
reported from an activated sludge plant in Albany, New York, based on measured concentrations in
wastewater and suspended particle matter (Kim et al.. 2017). This study was selected for use in this risk

Page 438 of 572

-------
12127

12128

12129

12130

12131

12132

12133

12134

12135

12136

12137

12138

12139

12140

12141

12142

12143

12144

12145

12146

12147

12148

12149

12150

12151

12152

12153

12154

12155

12156

12157

12158

12159

12160

12161

12162

12163

12164

12165

12166

12167

12168

12169

12170

12171

12172

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

evaluation because this is the best representative of the full-scale wastewater treatment processes that are
used in the United States.

Several high-quality studies observing the efficacy of advanced (tertiary) treatment techniques were
identified. Cristale et al. (2016) reported a low TCEP removal rate (<38 percent) after a several series of
advanced treatment techniques such as chlorination, ozonation, ultraviolet (UV) radiation, and
UV/hydrogen peroxide (UV/H2O2). Liang and Liu	reported an overall TCEP removal of-30.1

percent after undergoing tertiary treatment that consisted of hyperfiltration, ozonation, and chlorination.
Pane et al. (1 reported an overall TCEP removal of 0.3 percent and 12.3 percent using UV filters in
two activated sludge plants in China.

Overall, because TCEP has a high water solubility and remains in treated wastewater, negligible to low
accumulation of TCEP will be found in sewage sludge and will not significantly contribute to the
removal of TCEP in wastewater treatments (Kim et al.. . ^ l , 1 u1 tale et al.. 2016; Liang and Liu. 2016;
Marklund et al.. 2005a). In addition, biodegradation and air stripping are not expected to be significant
removal processes. Therefore, TCEP is expected to pass through wastewater treatment systems and be
discharged into the receiving waters.

E.2.5.3 Removal in Drinking Water Treatment

In the United States, drinking water typically comes from surface water (i.e., lakes, rivers, reservoirs)
and groundwater. The source water then flows to a drinking water treatment plant (DWTP) where it
undergoes a series of water treatment steps before being dispersed to homes and communities. In the
United States, public water systems often use conventional treatment processes that include coagulation,
flocculation, sedimentation, filtration, and disinfection, as required by law.

Five U.S. studies were identified and reviewed on the removal of TCEP in DWTPs. Those DWTPs
consisted of both conventional and advanced treatment processes and used river water as the source. In
all five studies, TCEP was found to be either minimally removed or not removed at all after undergoing
pre-ozonation (or coagulation), flocculation, sedimentation, ozonation, filtration, and chlorination (Choo
and Oh. 2020; Zhang et al.. 2016a; Benotti et al.. 2009; Snyder et al.. 2006; Westerhoff et al.. 2005;
Stackelberg et al.. 2004).

Several studies have demonstrated that granular activated carbon (GAC) and powdered activated carbon
(PAC) enhanced the removal of TCEP when added to conventional treatment methods (Choo and Oh.
2020; Padhye et al.. 2014; Westerhoff et al.. 2005; Stackelberg et al.. 2004). A South Korean drinking
water treatment study reported a removal efficiency of-52 percent after undergoing coagulation and
ultrafiltration. After undergoing the GAC step, 73.7 percent of TCEP was removed (Kim et al.. 2007). A
high level of uncertainty exists about TCEP's carbon usage rate. The higher the carbon usage rate, the
more expensive the treatment costs will be to achieve high levels of TCEP removal. Higher treatment
costs may determine that GAC nor PAC is not an economically feasible method for removing TCEP
from drinking water. In addition, the use of activated carbon filtration, such as PAC and GAC, is not
mandatory for drinking water treatment facilities in the United States.

E.2.6 Bioaccumulation Potential of TCEP

Information on bioconcentration and bioaccumulation in aquatic and terrestrial organisms are important
to understand the behavior of TCEP in the environment and a key component in assessing its risk to all
living organisms, including humans.

Page 439 of 572

-------
12173

12174

12175

12176

12177

12178

12179

12180

12181

12182

12183

12184

12185

12186

12187

12188

12189

12190

12191

12192

12193

12194

12195

12196

12197

12198

12199

12200

12201

12202

12203

12204

12205

12206

12207

12208

12209

12210

12211

12212

12213

12214

12215

12216

12217

12218

12219

12220

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Bioconcentration is the uptake and retention of a chemical by an aquatic organism from ambient water
only (	33c). Bioconcentration does not include chemical exposure through diet, but rather

its uptake by respiratory and dermal surfaces (Arnnot and Gobas. 2006). The bioconcentration factor
(BCF) is the ratio of the concentration of a chemical in the tissue of an organism to its concentration in
the ambient water once a steady state has been achieved (OECD. 2012). The resulting BCF value
provides an indication of the potential for a chemical to bioconcentrate in lipids of organisms.

Three high-quality semi-static tests were identified and selected for use in the risk evaluation. Tame et al.
(2 reported steady-state BCF values of 1.0 in the muscle, 1.6 in the gill, 2.6 in the brain, 1.6 in the
kidney, and 4.3 in the liver in juvenile common carp (Cyprinus carpio) after 28 days of exposure to
TCEP at 9.1 jug/L. Wane et al (2017a) reported steady-state BCF values of 0.8 in the muscle, 1.9 in the
gill, 2.2 in the brain, and 2.4 in liver of adult zebrafish (Danio rerio) after 19 days of exposure to TCEP
at 893 |ig/L. The concentration of TCEP in all tissue compartments achieved steady-state in 3 days and
the depuration half-life was <5.3 hours. Another high-quality semi-static test reporting BCF values in
fish was identified and selected. Arukwe et al. (2 reported BCF values of 0.3 1, 0.16, and 0.34 in the
muscle in juvenile Atlantic salmon (Salmo salaf) after 7 days of exposure to TCEP at concentrations of
0.04, 0.2, 1 mg/L, respectively.

A continuous flow-through test was identified during systematic review. Sasaki et al. (1982) reported
BCF values of 1.1 and 1.3 in killifish (Oryzias latipes) after 5 and 11 days of exposure to TCEP at
concentrations of 12.7 and 2.3 mg/L, respectively. The depuration half-life was 0.7 hour, which
indicates that the killifish eliminated TCEP rapidly. This study was evaluated as a medium-quality study
because insufficient information was available on the test conditions and study design. This added
uncertainty on whether its BCF values would be a good representation of TCEP's bioconcentration
potential and thus will not be considered in this risk evaluation.

The range of experimental BCF values provided above agrees with the calculated BCF values of 1.04
L/kg given by the BCFBAF™ module in EPI Suite™ (I c. < ^ \ rV I !) and 1.29 by another QSAR
model, OPEn structure-activity/property Relationship App (OPERA) (' v «« \	, Mansouri et al..

2018). The calculated values from EPI Suite™ and OPERA are not included in this risk evaluation
because the BCF values from high-quality studies cited above are available.

Bioaccumulation is the net accumulation of a chemical by an organism by all possible routes of
exposure (e.g., respiration, dietary, dermal) from all surrounding environmental media (e.g., air, water,
sediment, and diet) (ECHA. 2008). The bioaccumulation factor (BAF) can be expressed as the steady-
state ratio of the chemical concentration in an organism to the concentration in the ambient water. The
concentration of a chemical in an organism can be measured and reported on wet weight (ww), dry
weight (dw), or lipid weight (lw) basis. In order to reduce any variability and uncertainty, lipid-
normalized BAFs in whole fish and fish tissues were used in this risk evaluation. Lipid weight BAF
values were converted to wet weight BAF values by using EquationApx E-l.

EquationApx E-l

There are multiple wet weight BAF values reported for aquatic organisms collected from water bodies
that contained TCEP. A mean BAF value (L/kg wet weight) of 794 in the muscle and 1,995 in the liver,
kidney, and gill, respectively, were reported for pelagic and benthic fish collected from Laizhou Bay in
China (Bekele et al.. 2021). A mean BAF value (L/kg wet weight) of 30.7 in the muscle and 70.7 in the
liver was reported for crucian carp (Carassius auratus) collected from Nakdong River in South Korea

Page 440 of 572

-------
12221

12222

12223

12224

12225

12226

12227

12228

12229

12230

12231

12232

12233

12234

12235

12236

12237

12238

12239

12240

12241

12242

12243

12244

12245

12246

12247

12248

12249

12250

12251

12252

12253

12254

12255

12256

12257

12258

12259

12260

12261

12262

12263

12264

12265

12266

12267

12268

12269

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

(Choo et ai. ). A mean BAF value (L/kg wet weight) of 2,198 was reported in walleye {Sander
vitreus) collected from the Great Lakes (Guo et al.J ). Mean whole body BAF values (L/kg wet
weight) ranging from 109 to 1,248 were reported for aquatic organisms collected from a freshwater pond
containing electronic wastes (e-waste) in South China (Liu et ai. 2019a). Mean BAF values of 6,3 10 in
benthic invertebrates, 2,690 in pelagic fish, and 4,270 in benthic fish were reported for fish collected
from Zhushan Bay in Lake Taihu, China (Wane et ai. 2019b).

Zhang et ai (2018b) reported a median BAF value (L/kg wet weight) of 21,380 in the muscle of fishes
collected from a site that was less than 1 km away from the outfall of a wastewater treatment plant
located in Pearl River Delta, China. Fish species included catfish (Clarias batrachus), common carp
(Cyprinus carpio), bream (Parabramispekinensis), and white semiknife-carp (Hemiculter leucisculus).
This BAF value is not included in this draft risk evaluation because this study was evaluated as a
medium quality. Surface water samples were collected from 11 different sites, while fish samples were
collected from only 1 site. Because the TCEP concentrations in surface water were reported as a range,
independent calculation of the BAF could not be conducted. In addition, the reported BAF value could
not be verified whether it was a lipid-normalized BAF value. Hon	reported a mean whole

body BAF value (L/kg wet weight) of 34.7 for topmouth gudgeon, (Pseudorasboraparva), crucian carp
(Carassius auratus), and loach (Misgurnus anguillicaudatus) collected from urban surface water in
Beijing, China. Because this study was evaluated as a medium quality, these BAF data are not included
in this risk evaluation. The tissue-specific values were based on average water concentrations; however,
the study did not specify which of the nine rivers the tissue concentrations in the fish were from and not
all loach samples have reported corresponding concentrations in several rivers, which adds uncertainty
in the study's calculations. Sutton et ai (1 measured TCEP in the blubber of harbor seals (Phoca
vitulina) from San Francisco Bay. This study was not included in this draft risk evaluation because
upper trophic fish are the focus of this bioaccumulation assessment.

The upper-trophic fish BAF value of 6.3 and a biotransformation half-life of 0.0798 days (~1 hour and
55 minutes) were estimated using a log Kow value of 1.78 in the BCFBAF™ Model (	I).

The biotransformation half-life of 0.219 days (-5.3 hours) was estimated by OPERA 0 v •! P \ M c;
Mansouri et ai. 2018). These estimated values were not included in this draft risk evaluation because
data from high-quality monitoring studies are available.

Bioaccumulation from soil to terrestrial or benthic organisms is expressed by the biota-sediment
accumulation factor (BSAF), which is the ratio of concentrations of a chemical in the tissue of a
sediment-dwelling organism to the concentration of a chemical in sediment. Wane et ai (2019b)
reported a BSAF value of 2.19x 10~3 and 1,48x 10~3 for invertebrates and benthic fishes, respectively,
from Zhushan Bay in Lake Taihu, China. Liu et ai (2019a) reported a BSAF range of 0.015 to 0.171 for
aquatic organisms collected from freshwater pond polluted with e-wastes in South China. Choo et ai
(2 reported a mean BSAF value of 1.09 in the muscle and 2.49 in the liver of crucian carp
{Carassius auratus). Zhang et ai (2018b) reported a BSAF value of 1.38 10 3 in fish muscles collected
from a site that was less than 1 km away from the outfall of a wastewater treatment plant located in Pearl
River Delta, China. This BSAF value is not included in this draft risk evaluation because this study was
evaluated as a medium quality. Sediment samples were collected from 11 different sites, while fish
samples were collected from only 1 site. Because the TCEP concentration in sediment was reported as a
range, independent calculation of BSAF could not be conducted.

Biomagnification describes the potential of a chemical to be transferred through the food web. It is
defined as an increase of a chemical concentration in the tissue of an organism compared to the tissue
concentration of its prey. The biomagnification potential of a chemical can be expressed as either a

Page 441 of 572

-------
12270

12271

12272

12273

12274

12275

12276

12277

12278

12279

12280

12281

12282

12283

12284

12285

12286

12287

12288

12289

12290

12291

12292

12293

12294

12295

12296

12297

12298

12299

12300

12301

12302

12303

12304

12305

12306

12307

12308

12309

12310

12311

12312

12313

12314

12315

12316

12317

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

biomagnification factor (BMF) or trophic magnification factor (TMF). Generally, TMF is preferred over
BMF because TMF represents the average value of the prey-to-predator magnification factor over a food
chain rather than just a specific predator-prey relationship (Fu et ai. 2020). When a trophic dilution
occurs, the concentration of a pollutant decreases as the trophic level increases. It could be a result of a
net balance of ingestion rate, uptake from food, internal transformation, or elimination processes
favoring loss of pollutant that enters the organism via food.

In Brandsma et al. ( , TMFs were calculated for organophosphate flame retardants (OPFRs) in two
food webs (benthic and pelagic) and in total food web of Western Scheldt in Netherlands. No significant
relationship was observed between TCEP and pelagic food web and total food web. It is possible that the
trophic dilution in the pelagic food web occurred because TCEP was likely to be adsorbed to particles,
and thus were likely to be more abundant in the sediment than in the water column. However, a TMF
value of 2.6 was reported for benthic food web. It was determined that the trophic magnification in the
benthic food web of TCEP was due to high levels of TCEP emission and the organisms' substantial
exposure. Fu et al. (2020) studied the trophic magnification behavior of organophosphate esters in the
Antarctic ecosystem that included algae (.Halymenia floresia), archaeogastropoda (Nacella concinna),
neogastropoda (Trophon geversianus), black rockcod (Notothenia coriiceps), and penguins (Pygoscelis
papua). The TMF of TCEP was 5.2, which indicated that TCEP can be magnified through this food
chain. Zhao et al. (2018) studied the trophic transfer of OPFRs in a lake food web from Taihu Lake,
China, that included plankton, five invertebrate species, and eleven fish species. There was no
significant correlation between TCEP and trophic level. Trophic dilution was likely to be a result of
rapid metabolism in sampled fishes.

E.2.6.1 Key Sources of Uncertainty

There is a significant disparity between the BCF and BAF values reported for TCEP. It was observed
that field-measured BAFs were much higher than laboratory-measured BCFs. In controlled laboratory
studies, the exposure time is short, reaching equilibrium is challenging, and the exposure pathway is
limited (lack of dietary intake). A field-measured BAF considers an organism's exposure to a chemical
through all exposure routes in a natural aquatic ecosystem and incorporates chemical biomagnification
and metabolism, making it the most direct measure of bioaccumulation (	1003c). TCEP has

the ability to quickly bioaccumulate in fish tissue if it is exposed to high TCEP concentration in the
surrounding water for a period of time. For example, TCEP concentration in the muscle of juvenile
Atlantic salmon (Salmo salar) increased 10-fold when the water concentration of TCEP increased from
0.2 to 1 mg/L in 7 days (Anikwe et al.. 2018).

Overall, a significantly higher concentration of TCEP was observed in liver than in the muscle (Tame et
al.. 2019; Choo et al.. 2018; Hon et al.. 2017; Wane et i	). Hon et al. (2017) showed that

metabolically active tissues, such as liver and kidney, accumulate more than metabolically inactive
tissue like muscle. The liver is the first tissue to be perfused by trace pollutants and it has a higher lipid
contents and assimilation rate than in muscles (Kim et al.. 2015; Koiadinovic et al.. 2007). Several
studies showed that a significant correlation was observed between lipid contents and TCEP
concentrations, indicating that lipid content is an important factor determining TCEP bioaccumulation in
aquatic organisms (Bekele et al.. 2019; Wane et al.. IVI j; * \io et al.. 2014). However, some studies
showed no significant correlations between TCEP concentrations and lipid contents (Liu et al.. JO r\i;
Liu et al.. 2019b; Brandsma et al.. 2015). The accumulative potential of TCEP can vary greatly due to
several factors such as fish species, feeding habits, and temporal and spatial factors (U.S. EPA. 2003c).
When taken as a whole, studies provided above indicate that TCEP could have the potential to
bioaccumulate and biomagnify in benthic food webs.

Page 442 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

12318	The reported TMF reported by Brandsma	was reported as "tentative" because the sample

12319	size was small (n = 15). As a general rule, a number of samples between 30 and 60 are recommended to

12320	achieve statistical reliable TMFs (Borga et ai. 2012). The small sample size adds some uncertainty with

12321	the use of this TMF value in this draft risk evaluation.

12322

Page 443 of 572

-------
12323

12324

12325

12326

12327

12328

12329

12330

12331

12332

12333

12334

12335

12336

12337

12338

12339

12340

12341

12342

12343

12344

12345

12346

12347

12348

12349

12350

12351

12352

12353

12354

12355

12356

12357

12358

12359

12360

12361

12362

12363

12364

12365

12366

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Appendix F ENVIRONMENTAL HAZARD DETAILS

F,1 Approach and Methodology

For aquatic species, EPA estimates hazard by calculating a concentration of concern (COCs) for a
hazard threshold. COCs can be calculated using a deterministic method by dividing a hazard value by an
assessment factor (AF) according to EPA methods as shown in EquationApx F-l (	2016a.

2014b. 2012b).

Equation Apx F-l

COC = toxicity value/AF

COCs can also be calculated using probabilistic methods. For example, an SSD can be used to calculate
a hazardous concentration for 5 percent of species (HC05). The HC05 estimates the concentration of a
chemical that is expected to protect 95 percent of aquatic species. This HC05 can then be used to
calculate a COC. For TCEP, Web-ICE (Version 3.3; Appendix F.2.1.1) followed by SSD probabilistic
method (Appendix F.2.1.2) was used to calculate the acute COC. The deterministic method was used to
calculate at chronic COC

For terrestrial species, EPA estimates hazard by using a hazard value for soil invertebrates, a
deterministic approach, or by calculating a TRV for mammals (Appendix F.2.2). The TRV is expressed
as doses in units of mg/kg-bw/day. Although the TRV for TCEP is derived from laboratory mice and rat
studies, body weight is normalized; therefore, the TRV can be used with ecologically relevant wildlife
species to evaluate chronic dietary exposure to TCEP (	2007a).

F.2 Hazard Identification

F.2.1 Aquatic Hazard Data

F.2.1.1 Web-Based Interspecies Correlation Estimation (Web-ICE)

Results from the systematic review process indicated three studies with empirical data meeting
evaluation criteria on aquatic species for TCEP with two studies producing LC50 endpoint data. To
supplement the empirical data, EPA used a modeling approach, Web-ICE. Web-ICE predicts toxicity
values for environmental species that are absent from a dataset and can provide a more robust dataset to
estimate toxicity thresholds. Specifically, EPA used Web-ICE to supplement empirical data for aquatic
organisms for acute exposure durations. EPA also considered ECOSAR predictions. However, after
comparing predictions with empirical data available for TCEP, EPA had more confidence in the Web-
ICE predictions. Therefore, Web-ICE predictions were used quantitatively during evidence integration.
Note that within the ECOSAR dataset there are measured TCEP toxicity data for acute exposure to fish
and daphnia, chronic exposure to daphnia, and exposure to algae (	2022c). These data

originate from studies within the Japan Chemicals Collaborative Knowledge database (J-CHECK) and
will be potentially integrated into EPA's analysis once the studies become available, are translated, and
are evaluated through systematic review.

Acute dose-response assays for fish and aquatic invertebrates create useful hazard endpoints for risk
assessments. Calculated endpoints such as EC50 or LC50 values and associated descriptors (confidence
interval, NOEC, and LOEC values) are often comparable across taxa when standardized methodologies
and statistical analysis are employed and documented. Two studies in the TCEP dataset had 96-hour
LC50 data for rainbow trout and zebra fish ( lalde et ai. 2018; Life Sciences Research Ltd. 1990a).

Page 444 of 572

-------
12367

12368

12369

12370

12371

12372

12373

12374

12375

12376

12377

12378

12379

12380

12381

12382

12383

12384

12385

12386

12387

12388

12389

12390

12391

12392

12393

12394

12395

12396

12397

12398

12399

12400

12401

12402

12403

12404

12405

12406

12407

12408

12409

12410

12411

12412

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

This limited dataset for aquatic organisms contained data gaps that EPA looked to fill using other lines
of evidence {i.e., modeling approaches).

The Web-ICE application was developed by EPA and collaborators to provide interspecies extrapolation
models for acute toxicity (Raimondo and Barron. 2010). Web-ICE models estimate the acute toxicity
(LC50/LD50) of a chemical to a species, genus, or family with no test data (the predicted taxon) from
the known toxicity of the chemical to a species with test data (the commonly tested surrogate species).

Web-ICE models are log-linear least square regressions of the relationship between surrogate and
predicted taxon based on a database of acute toxicity values; that is, median effect or lethal water
concentrations for aquatic species (EC50/LC50). Separate acute toxicity databases are maintained for
aquatic animals (vertebrates and invertebrates), aquatic plants (algae), and wildlife (birds and
mammals), with 1,440 models for aquatic taxa and 852 models for wildlife taxa currently included in
Web-ICE version 3.3 (Willmine et ai. 2016). Open-ended toxicity values {i.e., >100 mg/kg or <100
mg/kg) and duplicate records among multiple sources are not included in any of the databases.

The aquatic animal database within Web-ICE comprises of 48- or 96-hour EC50/LC50 values based on
death or immobility. This database is described in detail in the Aquatic Database Documentation found
on the Download Model Data page of Web-ICE and describes the data sources, normalization, and
quality and standardization criteria {e.g., data filters) for data used in the models. Data used in model
development adhered to standard acute toxicity test condition requirements of the ASTM International
(ASTM. 2014) and EPA's OCSPP {e.g., (	)).

EPA used the 96-hour LC50 toxicity data from rainbow trout and zebrafish studies in Table 4-2 as
surrogate species to predict LC50 toxicity values using the Web-ICE application (Raimondo and Barron.
2010). The Web-ICE Model estimated toxicity values for 77 species. For model validation, the model
results are then screened by the following quality standards to ensure confidence in the model
predictions. If a predicted species did not meet all the quality criteria listed below, the species was
eliminated from the dataset (Willmine et ai. 2016):

•	High R2 (> -0.6)

o The proportion of the data variance that is explained by the model. The closer the R2
value is to 1.0, the more robust the model is in describing the relationship between the
predicted and surrogate taxa.

•	Low mean square error (MSE; < ~0. 95)

o An unbiased estimator of the variance of the regression line.

•	High slope (> -0.6)

o The regression coefficient represents the change in log 10 value of the predicted taxon
toxicity for every change in loglO value of the surrogate species toxicity.

Previously published guidance on the Web-ICE Model did not include quantitative guidance on
confidence intervals, so the following was also required to be included in the TCEP database:

•	Narrow 95 percent confidence intervals

o One order of magnitude between lower and upper limit

After screening, the acute toxicity values for 18 additional aquatic organisms (16 fish, 1 amphibian, and
1 aquatic invertebrate species) were added to the rainbow trout and zebrafish 96-hour LC50 data
(Table Apx F-l). The toxicity data were then used to calculate the distribution of species sensitivity to
TCEP exposure through the SSD toolbox as shown in Figure Apx F-4 and Table 4-4 (Etterson. 2020).

Page 445 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

12413 Table Apx F-l. Web-ICE Predicted Species that Met Model Selection Criteria	

Predicted Species

Surrogate Species

LC50 mg/L

95% CI

R2

MSE

Slope



Rainbow trout

249.00











Zebrafish embryo

279.1









Bluegill

Rainbow trout

231.66

183.96-291.74

0.88

0.21

0.93

Channel catfish

Rainbow trout

172.56

100.50-296.30

0.79

0.4

0.82

Fathead minnowa

Rainbow trout

298.23

192.71-461.53

0.83

0.32

0.86

Fathead minnowa

Zebrafish embryo

258.53

135.59-492.96

0.84

0.54

0.91

Goldfish

Rainbow trout

392.66

153.72-1,003.00

0.86

0.42

0.85

Atlantic salmon

Rainbow trout

260.09

104.18-649.31

0.95

0.12

1.01

Brook trout

Rainbow trout

258.84

127.67-524.75

0.94

0.11

1.02

Brown trout

Rainbow trout

252.60

117.39-543.51

0.95

0.1

0.99

Bullfrog

Rainbow trout

333.44

159.02-699.16

0.97

0.15

0.88

Chinook salmon

Rainbow trout

229.96

123.72-427.44

0.96

0.07

0.94

Coho salmon

Rainbow trout

319.44

220.61-462.56

0.98

0.04

0.98

Common carp

Rainbow trout

304.89

104.50-889.57

0.87

0.3

0.89

Cutthroat trout

Rainbow trout

168.04

99.52-283.74

0.94

0.09

0.93

Daphnid

Rainbow trout

337.13

298.97-380.16

0.99

0

0.98

Green sunfish

Rainbow trout

314.52

107.19-922.86

0.94

0.13

0.92

Lake trout

Rainbow trout

98.63

51.81-187.73

0.93

0.08

0.86

Largemouth bass

Rainbow trout

143.43

52.46-392.13

0.86

0.24

0.94

Sheepshead minnow

Rainbow trout

101.21

47.14-217.30

0.65

0.56

0.75

Yellow perch

Rainbow trout

201.80

78.71-517.39

0.94

0.14

0.98

a The geometric mean of LC50 data for multiple predictions from different surrogate species are used for the species sensitivity distribution (SSD).

12414

Page 446 of 572

-------
12415

12416

12417

12418

12419

12420

12421

12422

12423

12424

12425

12426

12427

12428

12429

12430

12431

12432

12433

12434

12435

12436

12437

12438

12439

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

F.2.1.2 Species Sensitivity Distribution (SSD)

The SSD Toolbox is a resource created by EPA's Office of Research and Development (ORD) that can
fit SSDs to environmental hazard data (Etterson. 2020). The SSD Toolbox runs on Matlab 2018b (9.5)
for Windows 64 bit. For the TCEP Risk Evaluation, EPA calculated an SSD with the SSD Toolbox
using acute LC50 hazard data from systematic review and estimated data from the Web-ICE application
(Appendix F.2.1.1) that included 18 fish, one amphibian, and one invertebrate species. The SSD is used
to calculate a hazardous concentration for 5 percent of species (HC05). The HC05 estimates the
concentration of TCEP that is expected to be protective for 95 percent of species.

The SSD toolbox contains functions for fitting six distributions (normal, logistic, triangular, Gumbel,
Weibull, and Burr). Maximum likelihood was used to assess the goodness-of-fit of the data distribution
based on P-values. The larger the deviation of the p-value from 0.5 the greater the indication of lack of
fit. The Weibull distribution (HC05 = 121.49 mg/L, P = 0.66) had the best goodness-of-fit using the
maximum likelihood method (Figure Apx F-l). The sample-size corrected Akaike Information
Criterion (AICc) model selection was then used with maximum likelihood, which also indicated Weibull
as the best fit model (Figure Apx F-2). Because numerical methods may lack statistical power for small
sample sizes, a visual inspection of the data were also used to assess goodness-of-fit. A Q-Q plot was
used to assess the goodness-of-fit for the Weibull distribution (Figure Apx F-3). For the Q-Q plot, the
horizontal axis gives the empirical quantiles, and the vertical axis gives the predicted quantiles (from the
fitted distribution). The Q-Q plot demonstrates a good model fit with the data points in close proximity
to the line across the data distribution. The SSD plot shows the distribution of species sensitivity to
TCEP exposure. The calculated HC05 was 121.5 mg/L with a 95 percent CI of 85.0 mg/L to 170.6 mg/L
(Figure_Apx F-4).

Page 447 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

SSD Toolbox
File Plot

0

X

C:\Users\jbresset\OneDrive - Environmental Protection Agency (EPA)\Desktop\Flame Retardants\TCEP\TCEP.SSD.20220916.xlsx

Fit Distribution

Distribution:
burr

Fitting method
maximum likelihood

Goodness of Fit:
Iterations:

Scaling parameters
~ Scale to Body Weight

1000

Scaling factor:
Target weight;

1.15

100

Toolbox

Results:

Distribution Method

HC05

normal

logistic

triangular

gumbel

weibull

burr

125.4109
132.5977
117.6288
118.7111
121.4923
106.2081

0.0639
0.1798
0.0559
0.0060
0.6633
0.9141

FigureApx F-l. SSD Toolbox Interface Showing HC05s and P-Values for Each Distribution Using Maximum Likelihood Fitting
Method Using TCEP's Acute Aquatic Hazard Data (Etterson, 2020)

Page 448 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

* ModelSeioction

Percentile of interest: 5

Model-averaqed HCp:

Model-averaqed SE of HCp; ""
CV of HCp:

AICc Table



Distribution

AICc

delta AICc

Wt

HCp

SE HCp

1

weibull

234.9938

0

0.8202

121.4923

21.5791

2

burr

237.2150

2.2161

0.2048

108.2081

29.8403

•>

logistic

239.5421

4,5433

0.0640

132.5977

19.5104

4

normal

239.8332

4.8344

0.0553

125,4109

18.1083

5

triangular

239.8672

4.8884

0.0544

117,8288

9,4898

6

gumbe!

247.3237

12.3249

0,0013

118,7111

12.2030

12444

12445	FigureApx F-2. AICc for the Six Distribution Options in the SSD Toolbox for TCEP's Acute Aquatic Hazard Data (Etterson. 2020)

12446

Page 449 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE

December 2023

•A Weibul! Quantile Plot (Q-G)	— ~ X

File

ic3 id ^ G Id

Predicted Quantiles

12448	FigureApx F-3. Q-Q Plot of TCEP Acute Aquatic Hazard Data with the Weibull Distribution

12449	(Etterson, 2020)

12450

Page 450 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

1

0.9
0.8
£.0.7

1 0.6

o

h_

CL



CD

"5 0.4
E

° 0.3
0.2
0.1
0

-—weibull-ML
~ HC05

¦— 95% CL HC05

Goldfish Carassius alfratus
Daphnid Simocephalus vetulu^

Bullfrog Lithobates calesbeia^is
Coho salmon Oncorhynchus kisu/:h
Green sunfish Lepomis cyanfttus
Common carp Cyprinus ca/rpk
Zebrafish Danio rerio- embrvp
Fathead minnow Pimephales promelas J
Atlantic salmon Salmo sal/r
Brook trout Salvelinus fontil^lis

/Brown tmut Salmo trutta
/ Rainbow trout Orrcorhynchus myklss
SUuegill Lepomis macrochirus
Chinook salmon Oncorhynchus tshawytscha

Yellow perch Perca ftavescens
Channel ca^sh Ictalurus punctatus
• f Cutthroat /out Oncorhynchus clarkii
ftrgemovth bas^Micropterus salmoides
i Cyprinodprwariegatus



1.7

1.8

1.9

Wr Sii^nne^pamdvcush

2 2.1 2.2 2.3 2.4
Toxicity Value (Log 10 [LC50]) mg/L)

2.5

2.6

2.7

FigureApx F-4. SSD Distribution for TCEP's Acute Hazard Data (Etterson, 2020).

The HC05 is 121.5 mg/L, 95% CI = 85.0 to 170.6 mg/L.

Page 451 of 572

-------
12454

12455

12456

12457

12458

12459

12460

12461

12462

12463

12464

12465

12466

12467

12468

12469

12470

12471

12472

12473

12474

12475

12476

12477

12478

12479

12480

12481

12482

12483

12484

12485

12486

12487

12488

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

F.2.2 Terrestrial Hazard Data	

For calculation of the mammal TRV, an a priori framework for selection of the TRV value based on the
results of the NOAEL and LOAEL data (FigureApx F-5.). The minimum dataset required to calculate a
TRV consists of three results with NOEL or LOEL values for reproduction, growth, or mortality for at
least two species. If these minimum results are not available, then a TRV is not calculated.

For mammalian species, EPA estimates hazard by calculating a TRV. The TRV is expressed as doses in
units of mg/kg-bw/day. Although the TRV for TCEP is derived from laboratory mice and rat studies,
body weight is normalized; therefore, the TRV can be used with ecologically relevant wildlife species to
evaluate chronic dietary exposure to TCEP. Representative wildlife species chronic hazard threshold
will be evaluated in the trophic transfer assessments using the TRV. The flow chart in Figure Apx F-5.
was used to select the data to calculate the TRV with NOEL and/or LOEL data and described below
(I	>007a).

Step 1: At least three results and two species tested for reproduction, growth, or mortality general
end points.

For rats, a 2-year NOEL/LOEL (NTP. 1991b). a 16-week NOEL/LOEL for males, and a 16-
week NOEL/LOEL for females for mortality were used (Matthews et ai. 1990).

For mice, a 16-week NOEL/LOEL for reproduction (Matthews et ai. 1990) and an 8-day LOEL
for mortality were used (Hazleton Laboratories. 1983).

Step 2: Are there three or more NOELs in reproduction or growth effect groups?

Because there was only a single reproduction effect result and no growth effect results, then
proceed to step 3.

Step 3: If there is at least one NOEL result for the reproduction or growth effect groups?
The NOEL for reproduction is 175 mg/kg-bw/day

Then the TRV is equal to the lowest reported NOEL for any effect group (reproduction, growth,
or mortality), except in cases where the NOEL is higher than the lowest bounded LOEL.
The lowest bounded LOEL for mortality is 88 mg/kg-bw/day

Then the TRV is equal to the highest bounded NOEL below the lowest bounded LOEL.
The highest NOEL below the lowest NOEL is 44 mg/kg-bw/day.

The TRV for TCEP is 44 mg/kg-bw/day.

Page 452 of 572

-------
12489

12490

12491

12492

12493

12494

12495

12496

12497

12498

12499

12500

12501

12502

12503

12504

12505

12506

12507

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE

December 2023

FigureApx F-5. TRY Flow Chart

F.2.3 Evidence Integration

Data integration includes analysis, synthesis, and integration of information for the draft risk evaluation.
During data integration, EPA considers quality, consistency, relevancy, coherence, and biological
plausibility to make final conclusions regarding the weight of the scientific evidence. As stated in the
Draft Systematic Review Protocol Supporting TSCA Risk Evaluations for Chemical Substances (U.S.
EPA. 20211 data integration involves transparently discussing the significant issues, strengths, and
limitations as well as the uncertainties of the reasonably available information and the major points of
interpretation.

The general analytical approaches for integrating evidence for environmental hazard is discussed in
Section 7.4 of the 2021 Draft Systematic Review Protocol (J.S. EPA. 2021).

The organization and approach to integrating hazard evidence is determined by the reasonably available
evidence regarding routes of exposure, exposure media, duration of exposure, taxa, metabolism and
distribution, effects evaluated, the number of studies pertaining to each effect, as well as the results of
the data quality evaluation.

Page 453 of 572

-------
12508

12509

12510

12511

12512

12513

12514

12515

12516

12517

12518

12519

12520

12521

12522

12523

12524

12525

12526

12527

12528

12529

12530

12531

12532

12533

12534

12535

12536

12537

12538

12539

12540

12541

12542

12543

12544

12545

12546

12547

12548

12549

12550

12551

12552

12553

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

The environmental hazard integration is organized around effects to aquatic and terrestrial organisms as
well as the respective environmental compartments (e.g., pelagic, benthic, soil). Environmental hazard
assessment may be complex based on the considerations of the quantity, relevance, and quality of the
available evidence.

For TCEP, environmental hazard data from toxicology studies identified during systematic review have
used evidence that characterizes apical endpoints; that is, endpoints that could have population-level
effects such as reproduction, growth, and/or mortality. Additionally, mechanistic data that can be linked
to apical endpoints will add to the weight of the scientific evidence supporting hazard thresholds. EPA
also considered predictions from Web-ICE and ECOSAR to supplement the empirical data found during
systematic review.

F.2.3.1 Weight of the Scientific Evidence

After calculating the hazard thresholds that were carried forward to characterize risk, a narrative
describing the weight of the scientific evidence and uncertainties was completed to support EPA's
decisions. The weight of the scientific evidence fundamentally means that the evidence is weighed (i.e.,
ranked) and weighted (i.e., a piece or set of evidence or uncertainty may have more importance or
influence in the result than another). Based on the weight of the scientific evidence and uncertainties, a
confidence statement was developed that qualitatively ranks (i.e., robust, moderate, slight, or
indeterminate) the confidence in the hazard threshold. The qualitative confidence levels are described
below.

The evidence considerations and criteria detailed within (	321) guides the application of

strength-of-evidence judgments for environmental hazard effect within a given evidence stream and
were adapted from Table 7-10 of the 2021 Draft Systematic Review Protocol (	>21).

EPA used the strength-of-evidence and uncertainties from (	2021) for the hazard assessment

to qualitatively rank the overall confidence using evidence Table 4-6 for environmental hazard.
Confidence levels of robust (+ + +), moderate (+ +), slight (+), or indeterminant are assigned for each
evidence property that corresponds to the evidence considerations (	21). The rank of the

Quality of the Database consideration is based on the systematic review overall quality determination
(High, Medium, or Low) for studies used to calculate the hazard threshold, and whether there are data
gaps in the toxicity dataset. Another consideration in the Quality of the Database is the risk of bias (i.e.,
how representative is the study to ecologically relevant endpoints). Additionally, because of the
importance of the studies used for deriving hazard thresholds, the Quality of the Database consideration
may have greater weight than the other individual considerations. The high, medium, and low systematic
review overall quality determinations ranks correspond to the evidence table ranks of robust (+ + +),
moderate (+ +), or slight (+), respectively. The evidence considerations are weighted based on
professional judgment to obtain the overall confidence for each hazard threshold. In other words, the
weights of each evidence property relative to the other properties are dependent on the specifics of the
weight of the scientific evidence and uncertainties that are described in the narrative and may or may not
be equal. Therefore, the overall score is not necessarily a mean or defaulted to the lowest score. The
confidence levels and uncertainty type examples are described below.

Confidence Levels

• Robust (+ + +) confidence suggests thorough understanding of the scientific evidence and

uncertainties. The supporting weight of the scientific evidence outweighs the uncertainties to the

Page 454 of 572

-------
12554

12555

12556

12557

12558

12559

12560

12561

12562

12563

12564

12565

12566

12567

12568

12569

12570

12571

12572

12573

12574

12575

12576

12577

12578

12579

12580

12581

12582

12583

12584

12585

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

point where it is unlikely that the uncertainties could have a significant effect on the exposure or
hazard estimate.

•	Moderate (+ +) confidence suggests some understanding of the scientific evidence and
uncertainties. The supporting scientific evidence weighed against the uncertainties is reasonably
adequate to characterize exposure or hazard estimates.

•	Slight (+) confidence is assigned when the weight of the scientific evidence may not be adequate
to characterize the scenario, and when the assessor is making the best scientific assessment
possible in the absence of complete information. There are additional uncertainties that may need
to be considered.

•	Indeterminant (N/A) corresponds to entries in evidence tables where information is not available
within a specific evidence consideration.

Types of Uncertainties

The following uncertainties may be relevant to one or more of the weight of the scientific evidence
considerations listed above and will be integrated into that property's rank in the evidence table (Table
4-6):

•	Scenario Uncertainty: Uncertainty regarding missing or incomplete information needed to fully
define the exposure and dose.

o The sources of scenario uncertainty include descriptive errors, aggregation errors, errors
in professional judgment, and incomplete analysis.

•	Parameter Uncertainty: Uncertainty regarding some parameter.

o Sources of parameter uncertainty include measurement errors, sampling errors,
variability, and use of generic or surrogate data.

•	Model Uncertainty: Uncertainty regarding gaps in scientific theory required to make predictions
on the basis of causal inferences.

o Modeling assumptions may be simplified representations of reality.

Table Apx F-2 summarizes the weight of the scientific evidence and uncertainties, while increasing
transparency on how EPA arrived at the overall confidence level for each exposure hazard threshold.
Symbols are used to provide a visual overview of the confidence in the body of evidence, while de-
emphasizing an individual ranking that may give the impression that ranks are cumulative (e.g., ranks of
different categories may have different weights).

Page 455 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

12586	TableApx F-2. Considerations that Inform Evaluations of the Strength of the Evidence within an Evidence Stream Apical

12587	Endpoints, Mechanistic, or Field Studies)		

Consideration

Increased Evidence Strength (of the Apical
Endpoints, Mechanistic, or Field Studies
Evidence)

Decreased Evidence Strength (of the Apical Endpoints, Mechanistic, or
Field Studies Evidence)

The evidence considerations and criteria laid out here guide the application of strength-of-evidence judgments for an outcome or environmental hazard effect
within a given evidence stream. Evidence integration or synthesis results that do not warrant an increase or decrease in evidence strength for a given
consideration are considered "neutral" and are not described in this table (and, in general, are captured in the assessment-specific evidence profile tables).

Quality of the database"
(risk of bias)

•	A large evidence base of high- or medium-qaa[ity
studies increases strength.

•	Strength increases if relevant species are
represented in a database.

•	An evidence base of mostly low-quality studies decreases strength.

•	Strength also decreases if the database has data gaps for relevant species,
i.e., a trophic level that is not represented.

•	Decisions to increase strength for other considerations in this table should
generally not be made if there are serious concerns for risk of bias; in other
words, all the other considerations in this table are dependent upon the
quality of the database.

Consistency

Similarity of findings for a given outcome (e.g., of a
similar magnitude, direction) across independent
studies or experiments increases strength,
particularly when consistency is observed across
species, life stage, sex, wildlife populations, and
across or within aquatic and terrestrial exposure
pathways.

•	Unexplained inconsistency (i.e., conflicting evidence; see U.S. EPA
(2005b) decreases strength.)

•	Strength should not be decreased if discrepant findings can be reasonably
explained by study confidence conclusions; variation in population or
species, sex, or life stage; frequency of exposure (e.g., intermittent or
continuous); exposure levels (low or high); or exposure duration.

Strength (effect magnitude)
and precision

•	Evidence of a large magnitude effect (considered
either within or across studies) can increase strength.

•	Effects of a concerning rarity or severity can also
increase strength, even if they are of a small
magnitude.

•	Precise results from individual studies or across the
set of studies increases strength, noting that
biological significance is prioritized over statistical
significance.

•	Use of probabilistic model (e.g., Web-ICE, SSD)
may increase strength.

Strength may be decreased if effect sizes that are small in magnitude are
concluded not to be biologically significant, or if there are only a few
studies with imprecise results.

Biological gradient/dose-
response

•	Evidence of dose-response increases strength.

•	Dose-response may be demonstrated across studies
or within studies and it can be dose- or duration-
dependent.

• A lack of dose-response when expected based on biological
understanding and having a wide range of doses/exposures evaluated in the
evidence base can decrease strength.

Page 456 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Consideration

Increased Evidence Strength (of the Apical
Endpoints, Mechanistic, or Field Studies
Evidence)

Decreased Evidence Strength (of the Apical Endpoints, Mechanistic, or
Field Studies Evidence)



•	Dose response may not be a monotonic dose-
response (monotonicity should not necessarily be
expected, e.g., different outcomes may be expected
at low vs. high doses due to activation of different
mechanistic pathways or induction of systemic
toxicity at very high doses).

•	Decreases in a response after cessation of exposure
(e.g., return to baseline fecundity) also may increase
strength by increasing certainty in a relationship
between exposure and outcome (this particularly
applicable to field studies).

•	In experimental studies, strength may be decreased when effects resolve
under certain experimental conditions (e.g., rapid reversibility after
removal of exposure).

•	However, many reversible effects are of high concern. Deciding between
these situations is informed by factors such as the toxicokinetics of the
chemical and the conditions of exposure, see (U.S. EPA. 1998b). enduoint
severity, judgments regarding the potential for delayed or secondary
effects, as well as the exposure context focus of the assessment (e.g.,
addressing intermittent or short-term exposures).

•	In rare cases, and typically only in toxicology studies, the magnitude of
effects at a given exposure level might decrease with longer exposures
(e.g., due to tolerance or acclimation).

•	Like the discussion of reversibility above, a decision about whether this
decreases evidence strength depends on the exposure context focus of the
assessment and other factors.

•	If the data are not adequate to evaluate a dose-response pattern, then
strength is neither increased nor decreased.

Biological relevance

Effects observed in different populations or
representative species suggesting that the effect is
likely relevant to the population or representative
species of interest (e.g., correspondence among the
taxa, life stages, and processes measured or observed
and the assessment endpoint).

An effect observed only in a specific population or species without a clear
analogy to the population or representative species of interest decreases
strength.

Physical/chemical relevance

Correspondence between the substance tested and
the substance constituting the stressor of concern.

The substance tested is an analogue of the chemical of interest or a mixture
of chemicals which include other chemicals besides the chemical of
interest.

Environmental relevance

Correspondence between test conditions and
conditions in the region of concern.

The test is conducted using conditions that would not occur in the
environment.

" Database refers to the entire dataset of studies integrated in the environmental hazard assessment and used to inform the strength of the evidence. In this context,
database does not refer to a computer database that stores aggregations of data records such as the ECOTOX Knowledgebase.

12588

12589

Page 457 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Appendix G ENVIRONMENTAL RISK DETAILS
G.l Risk Estimation for Aquatic Organisms

TableApx G-l. Calculated Risk Quotients Based on TCEP Sediment Concentrations (ppb) as

Calculated Using Mode

ed Data for Air Deposition to Sediment

Exposure Scenario

Production
Volume
(lb/year)"

Meteorological
Model6

Sediment
Concentration (ppb)
at 1,000 mc

Chronic RQ (Hazard
Value: 55.9 ppb)

Import and repackaging

2,500

MetCT

6.05E-04

1.08E-05

MetHIGH

7.35E-04

1.31E-05

25,000

MetCT

2.15E-03

3.85E-05

MetHIGH

2.98E-03

5.33E-05

Incorporation into paints
and coatings - 1-part
coatings

2,500

MetCT

1.32E-02

2.36E-04

MetHIGH

2.10E-02

3.76E-04

25,000

MetCT

3.00E-02

5.37E-04

MetHIGH

3.18E-02

5.69E-04

Incorporation into paints
and coatings - 2-part
reactive coatings

2,500

MetCT

3.38E-03

6.05E-05

MetHIGH

4.88E-03

8.73E-05

25,000

MetCT

9.31E-03

1.67E-04

MetHIGH

1.48E-02

2.65E-04

Use in paints and
coatings at job sites

2,500

MetCT

7.85E01

9.39E-02

MetHIGH

1.25E02

1.36E-01

25,000

MetCT

5.25E00

1.40E00

MetHIGH

7.58E00

2.24E00

Formulation of TCEP-
containing reactive resins
(for use in 2-part
systems)

2,500

MetCT

1.57E-02

2.81E-04

MetHIGH

1.49E-02

2.67E-04

25,000

MetCT

1.17E-02

2.09E-04

MetHIGH

1.08E-02

1.93E-04

Processing into 2-part
resin article

2,500

MetCT

3.78E-03

6.76E-05

MetHIGH

5.46E-03

9.77E-05

25,000

MetCT

1.11E-02

1.99E-04

MetHIGH

1.76E-02

3.15E-04

Laboratory chemicals

2,500

MetCT

1.93E-02

3.45E-04

MetHIGH

1.79E-02

3.20E-04

25,000

MetCT

1.11E-02

1.99E-04

MetHIGH

1.02E-02

1.82E-04

a Production volume of 2,500 lb TCEP/year uses high-end estimates (95th percentile). Production volume of 25,000
lb TCEP/yr uses central tendency estimates (median).

b The ambient air modeled concentrations and deposition values are presented for two meteorology conditions (Sioux
Falls, South Dakota, for central tendency meteorology; and Lake Charles, Louisiana, for higher-end meteorology).
c Estimated concentrations of TCEP (90th percentile) that could be in sediment via air deposition at a community
(1,000 m from the source) exposure scenario.

Page 458 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

12596	TableApx G-2. Environmental Risk Quotients by Exposure Scenario with Production Volumes of 2,500 lb/year for Aquatic

12597	Organisms with TCEP Surface Water Concentration (ppb) Modeled by VVWM-PSC"	

Exposu rc
Scenario

Production
Volume (lb/vear)fc

Days of
Release

Release

(kg/day)

Modeled Using VVWM-PSC

Max Day Avg
(ppb)'

COC Type

COC (ppb)

Days of Exceedance
(days per year)

RQ

Import and
repackaging

2,500

4

9.88

2,380

Acute

85,000

N/A

0.03

680

Chronic

55.9

5

12.16

Incorporation into
paints and
coatings - 1-part
coatings

2,500

2

35.17

10,200

Acute

85,000

N/A

0.12

1,480

Chronic

55.9

4

26.48

Incorporation into
paints and
coatings - 2-part
reactive coatings

2,500

1

31.89

8,250

Acute

85,000

N/A

0.10

670

Chronic

55.9

3

11.99

Use in paints and
coatings at job
sites

2,500

2

23.25

5,570

Acute

85,000

NA

0.07

800

Chronic

55.9

3

14.31

Formulation of
TCEP into 2-part
reactive resins

2,500

1

31.53

9,150

Acute

85,000

N/A

0.11

785

Chronic

55.9

3

14.04

Laboratory
chemicals

2,500

182

0.39

95

Acute

85,000

N/A

1.12E-03

95

Chronic

55.9

179

1.70

" Model input parameter for K0c utilized the mean (2.82).

b Production volume of 2,500 lb TCEP/year uses high-end estimates (95th percentile for all COUs except the laboratory chemicals COU uses the 1st percentile).
cMax day average represents the maximum concentration over a 1- or 14-day average period corresponding with the acute or chronic COC used for the RQ estimate.
d VVWM-PSC Model input parameter for K0c utilized the mean (2.82).

N/A = Days of exceedance are modeled for the application of chronic COCs and do not apply for acute COCs and corresponding RQs.

12598

12599

12600

Page 459 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

TableApx G-3. Environmental Risk Quotients by Exposure Scenario with Production Volumes of 2,500 lb/year for Aquatic

Exposure Scenario

Production
Volume
(Ib/vear)h

Days of
Release

Release
(kg/day)

Benthic Pore

Water
Concentration
(ppb)c

Benthic Pore Water Concentration ''

COC Type

COC

(ppb)

Days of
Exceedance

RQ

Import and repackaging

2,500

4

9.88

154

Acute

85,000

N/A

1.8E-03

138

Chronic

55.9

49

2.47

Incorporation into paints and coatings -
1-part coatings

2,500

2

35.17

337

Acute

85,000

N/A

3.96E-03

302

Chronic

55.9

82

5.4

Incorporation into paints and coatings -
2-part reactive coatings

2,500

1

31.89

154

Acute

85,000

N/A

1.81E-03

138

Chronic

55.9

48

2.47

Use in paints and coatings at job sites

2,500

2

23.25

184

Acute

85,000

N/A

2.16E-03

164

Chronic

55.9

56

2.93

Formulation of TCEP into 2-part
reactive resins

2,500

1

31.53

179

Acute

85,000

N/A

2.11E-03

161

Chronic

55.9

55

2.88

Laboratory chemicals

2,500

182

0.39

66

Acute

85,000

N/A

7.76E-04

66

Chronic

55.9

82

1.18

" Model input parameter for K0c utilized the mean (2.82).

h Production volume of 2,500 lb TCEP/year uses high-end estimates (95th percentile for all COUs except the laboratory chemicals COU uses the 1st percentile).
c Max day average represents the maximum concentration over a 1- or 14-day average period corresponding with the acute or chronic COC used for the RQ estimate.
'WWM-PSC Model input parameter for K0c utilized the mean (2.82).

N/A = Days of exceedance are modeled for the application of chronic COCs and do not apply for acute COCs and corresponding RQs.

Page 460 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

12605	TableApx G-4. Environmental RQs by Exposure Scenario with Production Volumes of 2,500 lb/year for Aquatic Organisms with

12606	TCEP Sediment Concentration (ppb) Modeled by VVWM-PSC			

Exposure Scenario

Production
Volume
(lb/vear)''

Days of
Release

Release
(kg/day)

Sediment
Concentration
(ppb)'

Sediment''

COC
Type

COC

(ppb)

Days of
Exceedance

RQ

Import and repackaging

2,500

4

9.88

4,130

Acute

85,000

N/A

0.05

3,690

Chronic

55.9

168

66.01

Incorporation into paints and coatings - 1-part
coatings

2,500

2

35.17

9,020

Acute

85,000

N/A

0.11

8,090

Chronic

55.9

187

144.72

Incorporation into paints and coatings - 2-part
reactive coatings

2,500

1

31.89

4,120

Acute

85,000

NA

0.05

3,690

Chronic

55.9

167

66.01

Use in paints and coatings at job sites

2,500

2

23.25

4,930

Acute

85,000

N/A

0.06

4,390

Chronic

55.9

171

78.53

Formulation of TCEP into 2-part reactive
resins

2,500

1

31.53

4,800

Acute

56

N/A

0.06

4,320

Chronic

85,000

171

77.28

Laboratory chemicals

2,500

182

0.39

1,760

Acute

85,000

NA

0.02

1,760

Chronic

55.9

249

31.48

" Model input parameter for K0c utilized the mean (2.82)

h Production volume of 2,500 lb TCEP/year uses high-end estimates (95th percentile for all COUs except the laboratory chemicals COU uses the 1st percentile).
c Max day average represents the maximum concentration over a 1 - or 14-day average period corresponding with the acute or chronic COC used for the RQ estimate.
J VVWM-PSC Model input parameter for K0c utilized the mean (2.82).

N/A = Days of exceedance are modeled for the application of chronic COCs and do not apply for acute COCs and corresponding RQs.

12607

12608

Page 461 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

TableApx G-5. Environmental RQs by Exposure Scenario with Production Volumes of 25,000 lb/year for Aquatic Organisms with

Exposure Scenario

Production
Volume
(lb/vear)"

Days of
Release

Release
(kg/day)

Modeled Using VVWM-PSC

Max 1-Dav Avg
(ppb)''

COC Type

COC

(ppb)

Days of Exceedance
(days per year)

RQ

Import and repackaging

25,000

39

7.13

1,730

Acute

85,000

N/A

0.02

1,730

Chronic

55.9

40

30.7

Incorporation into paints
and coatings - 1-part
coatings

25,000

57

10.97

3,250

Acute

85,000

N/A

0.04

3,250

Chronic

55.9

58

58.1

Incorporation into paints
and coatings - 2-part
reactive coatings

25,000

4

65.89

19,500

Acute

85,000

N/A

0.23

5,560

Chronic

55.9

6

99.5

Use in paints and
coatings at job sites

25,000

1

2.31

559

Acute

85,000

N/A

0.01

40

Chronic

55.9

1

0.7

Formulation of TCEP
into 2-part reactive resins

25,000

6

45.5

15,900

Acute

85,000

N/A

0.19

6,830

Chronic

55.9

9

122.2

Laboratory chemicals

25,000

229

2.74

664

Acute

85,000

N/A

0.01

664

Chronic

55.9

229

11.9

Risk to aquatic organisms is indicated by scenarios with an acute RQ > 1, or a chronic RQ > 1 and 14 days or more of exceedance for the chronic COC.
a Model input parameter for Koc utilized the mean (2.13).

b Production volume of 25,000 lb TCEP/yr uses central tendency estimates (median).

cMax day average represents the maximum concentration over a 1- or 14-day average period corresponding with the acute or chronic COC used for the RQ
estimate.

N/A = Days of exceedance are modeled for the application of chronic COCs and do not apply for acute COCs and corresponding RQs.

Page 462 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

TableApx G-6. Environmental RQs by Exposure Scenario with Production Volumes of 25,000 lb/year for Aquatic Organisms with

Exposure Scenario

Production
Volume
(lb/vear)''

Days of
Release

Release
(kg/day)

Benthic Pore Water
Concentration (ppb)'

Benthic Pore Water

COC
Type

COC

(ppb)

Days of
Exceedance

RQ

Import and repackaging

25,000

39

7.13

793

Acute

85,000

N/A

9.3E-03

745

Chronic

55.9

138

13.3

Incorporation into paints and coatings - 1-
part coatings

25,000

57

10.97

1,850

Acute

85,000

N/A

2.2E-02

1,770

Chronic

55.9

175

31.7

Incorporation into paints and coatings - 2-
part reactive coatings

25,000

4

65.89

1,260

Acute

85,000

N/A

1.5E-02

1,130

Chronic

55.9

132

20.2

Use in paints and coatings at job sites

25,000

1

2.31

9.3

Acute

85,000

N/A

1.1E-04

8

Chronic

55.9

0

0.14

Formulation of TCEP into 2-part reactive
resins

25,000

6

45.5

1,510

Acute

85,000

N/A

1.8E-02

1,360

Chronic

55.9

139

24.3

Laboratory chemicals

25,000

229

2.74

457

Acute

85,000

N/A

5.4E-03

456

Chronic

55.9

255

8.2

Risk to aquatic organisms is indicated by scenarios with an acute RQ > 1, or a chronic RQ > 1 and 14 days or more of exceedance for the chronic COC.
a model input parameter for Koc utilized the mean (2.13).

b Production volume of 25,000 lb TCEP/yr uses central tendency estimates (median).

c Max day average represents the maximum concentration over a 1- or 14-day average period corresponding with the acute or chronic COC used for the RQ
estimate.

N/A = Days of exceedance are modeled for the application of chronic COCs and do not apply for acute COCs and corresponding RQs.

Page 463 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

TableApx G-7. Environmental RQs by Exposure Scenario with Production Volumes of 25,000 lb/year for Aquatic Organisms with

Exposure Scenario

Production
Volume
(lb/vear)''

Days of
Release

Release
(kg/day)

Sediment
Concentration
(ppbf

Sediment

COC
Type

COC
(ppb)

Days of
Exceedance

RQ

Import and repackaging

25,000

39

7.13

4,570

Acute

85,000

N/A

0.1

4,300

Chronic

55.9

189

76.9

Incorporation into paints and coatings - 1-part
coatings

25,000

57

10.97

10,700

Acute

85,000

N/A

0.1

10,200

Chronic

55.9

214

182.5

Incorporation into paints and coatings - 2-part
reactive coatings

25,000

4

65.89

7,240

Acute

85,000

NA

0.1

6,500

Chronic

55.9

182

5.6

Use in paints and coatings at job sites

25,000

1

2.31

54

Acute

85,000

N/A

0

48

Chronic

55.9

0

0.9

Formulation of TCEP into 2-part reactive
resins

25,000

6

45.5

8,720

Acute

55.9

N/A

0.1

7,850

Chronic

85,000

187

140.4

Laboratory chemicals

25,000

229

2.74

2,640

Acute

85,000

N/A

0.1

2,630

Chronic

55.9

308

47.1

Risk to aquatic organisms is indicated by scenarios with an acute RQ > 1, or a chronic RQ > 1 and 14 days or more of exceedance for the chronic COC.

" Model input parameter for K0c utilized the mean (2.13).

h Production volume of 25,000 lb TCEP/yr uses central tendency estimates (median).

c Max day average represents the maximum concentration over a 1 - or 14-day average period corresponding with the acute or chronic COC used for the RQ estimate.
N/A = Days of exceedance are modeled for the application of chronic COCs and do not apply for acute COCs and corresponding RQs.

Page 464 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

12620	G.2 Risk Estimation for Terrestrial Organisms

12621

12622	Table Apx G-8. Calculated RQs Based on TCEP Soils Concentrations (mg/kg) as Calculated

12623	Using Modeled Data for Air Deposition to Soil			

Exposure Scenario

Production
Volume (lb/year)"

Meteorological
Model''

Soil Concentration
(mg/kg) at 1,000 mc

Chronic RQ (Hazard
Value: 612 mg/kg)

Import and Repackaging

2,500

MetCT

1.49E-06

2.43E-09

MetHIGH

1.92E-06

3.14E-09

25,000

MetCT

5.43E-06

8.87E-09

MetHIGH

7.59E-06

1.24E-08

Incorporation into paints
and coatings - 1-part
coatings

2,500

MetCT

3.33E-05

5.44E-08

MetHIGH

5.67E-05

9.27E-08

25,000

MetCT

7.59E-05

1.24E-07

MetHIGH

8.24E-05

1.35E-07

Incorporation into paints
and coatings - 2-part
reactive coatings

2,500

MetCT

1.1IE—05

1.82E-08

MetHIGH

2.41E-05

3.94E-08

25,000

MetCT

2.19E-05

3.59E-08

MetHIGH

3.68E-05

6.01E-08

Use in paints and coatings
at job sites

2,500

MetCT

3.97E-03

6.49E-06

MetHIGH

5.58E-03

9.11E-06

25,000

MetCT

5.59E-02

9.14E-05

MetHIGH

8.65E-02

1.41E-04

Formulation of TCEP-
containing reactive resins
(for use in 2-part systems)

2,500

MetCT

3.89E-05

6.35E-08

MetHIGH

3.85E-05

6.30E-08

25,000

MetCT

2.93E-05

4.79E-08

MetHIGH

2.82E-05

4.60E-08

Processing into 2-part
resin article

2,500

MetCT

1.21E-05

1.97E-08

MetHIGH

2.57E-05

4.20E-08

25,000

MetCT

2.71E-05

4.42E-08

MetHIGH

4.58E-05

7.48E-08

Laboratory chemicals

2,500

MetCT

4.84E-05

7.90E-08

MetHIGH

4.65E-05

7.59E-08

25,000

MetCT

2.75E-05

4.50E-08

MetHIGH

2.68E-05

4.37E-08

a Production volume of 2,500 lb TCEP/yr uses high-end estimates (95th percentile). Production volume of 25,000 lb
TCEP/yr uses central tendency estimates (median).

h The ambient air modeled concentrations and deposition values are presented for two meteorology conditions (Sioux
Falls, South Dakota, for central tendency meteorology; and Lake Charles, Louisiana, for higher-end meteorology).
c Estimated concentrations of TCEP (90th percentile) that could be in soil via air deposition at a community (1,000 m
from the source) exposure scenario.

12624

Page 465 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

12625	G.3 Trophic Transfer Analysis Results

12626

12627	Table Apx G-9. RQs Based on Potential Trophic Transfer of TCEP in Terrestrial Ecosystems Using EPA's Wildlife Risk Model for

12628	Eco-SSLs (Equation 4-1)						

Exposure Scenario

PV (lb/vear)"

Model6

Soil Concentration
(mg/kg) at 1,000 m':

Nematode

Mammal

S h o rt-Tailed Sh rew

TCEP in
biota
(mg/kg/dav)

RQ

TCEP in
Biota
(mg/kg/dav)

RQ

TCEP in
Biota
(mg/kg/dav)

RQ

Import and Repackaging

2,500

MetCT

1.49E-06

1.5E-06

2.4E-09

1.2E-06

2.7E-08

1.2E-06

1.8E-06

MetHIGH

1.92E-06

1.9E-06

3.1E-09

1.5E-06

3.5E-08

1.5E-06

2.3E-06

25,000

MetCT

5.43E-06

5.4E-06

8.9E-09

4.3E-06

9.8E-08

4.3E-06

6.5E-06

MetHIGH

7.59E-06

7.6E-06

1.2E-08

6.0E-06

1.4E-07

6.0E-06

9.1E-06

Incorporation into paints and
coatings - 1-part coatings

2,500

MetCT

3.33E-05

3.3E-05

5.4E-08

2.6E-05

6.0E-07

2.6E-05

4.0E-05

MetHIGH

5.67E-05

5.7E-05

9.3E-08

4.5E-05

1.0E-06

4.5E-05

6.8E-05

25,000

MetCT

7.59E-05

7.6E-05

1.2E-07

6.0E-05

1.4E-06

6.0E-05

9.1E-05

MetHIGH

8.24E-05

8.2E-05

1.3E-07

6.5E-05

1.5E-06

6.5E-05

9.9E-05

Incorporation into paints and
coatings - 2-part reactive
coatings

2,500

MetCT

1.1IE—05

1.1E-05

1.8E-08

8.8E-06

2.0E-07

8.8E-06

1.3E-05

MetHIGH

2.41E-05

2.4E-05

3.9E-08

1.9E-05

4.4E-07

1.9E-05

2.9E-05

25,000

MetCT

2.19E-05

2.2E-05

3.6E-08

1.7E-05

4.0E-07

1.7E-05

2.6E-05

MetHIGH

3.68E-05

3.7E-05

6.0E-08

2.9E-05

6.6E-07

2.9E-05

4.4E-05

Use in paints and coatings at
job sites

2,500

MetCT

0.004

0.004

6.4E-06

0.003

6.8E-05

0.003

0.005

MetHIGH

0.006

0.0056

9.0E-06

0.004

9.8E-05

0.004

0.007

25,000

MetCT

0.056

0.059

9.6E-05

0.044

1.0E-03

0.044

0.067

MetHIGH

0.086

0.086

1.4E-04

0.068

1.5E-03

0.068

0.103

Formulation of TCEP -
containing reactive resins (for
use in 2-part systems)

2,500

MetCT

3.89E-05

3.9E-05

6.4E-08

3.1E-05

7.0E-07

3.1E-05

4.7E-05

MetHIGH

3.85E-05

3.9E-05

6.3E-08

3.1E-05

7.0E-07

3.1E-05

4.6E-05

25,000

MetCT

2.93E-05

2.9E-05

4.8E-08

2.3E-05

5.3E-07

2.3E-05

3.5E-05

MetHIGH

2.82E-05

2.8E-05

4.6E-08

2.2E-05

5.1E-07

2.2E-05

3.4E-05

Page 466 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Exposure Scenario

PV (lb/vear)"

Model6

Soil Concentration
(mg/kg) at 1,000 mc

Nematode

Mammal

Short-Tailed Shrew

TCEP in
biota
(mg/kg/dav)

RQ

TCEP in
Biota
(mg/kg/dav)

RQ

TCEP in
Biota
(mg/kg/dav)

RQ

Processing into 2-part resin
article

2,500

MetCT

1.21E-05

1.2E-05

2.0E-08

9.6E-06

2.2E-07

9.6E-06

1.5E-05

MetHIGH

2.57E-05

2.6E-05

4.2E-08

2.0E-05

4.6E-07

2.0E-05

3.1E-05

25,000

MetCT

2.71E-05

2.7E-05

4.4E-08

2.2E-05

4.9E-07

2.2E-05

3.3E-05

MetHIGH

4.58E-05

4.6E-05

7.5E-08

3.6E-05

8.3E-07

3.6E-05

5.5E-05

Laboratory chemicals

2,500

MetCT

4.84E-05

4.8E-05

7.9E-08

3.8E-05

8.7E-07

3.8E-05

5.8E-05

MetHIGH

4.65E-05

4.6E-05

7.6E-08

3.7E-05

8.4E-07

3.7E-05

5.6E-05

25,000

MetCT

2.75E-05

2.8E-05

4.5E-08

2.2E-05

5.0E-07

2.2E-05

3.3E-05

MetHIGH

2.68E-05

2.7E-05

4.4E-08

2.1E-05

4.8E-07

2.1E-05

3.2E-05

" PV = Production volume of 2,500 lb TCEP/yr uses high-end estimates (95th percentile); PV of 25,000 lb TCEP/yr uses central tendency estimates (median).
h The ambient air modeled concentrations and deposition values are presented for two meteorology conditions (Sioux Falls, South Dakota, for central tendency
meteorology; and Lake Charles, Louisiana, for higher-end meteorology).

c Estimated concentrations of TCEP (90th percentile) that could be in soil via air deposition at a community (1,000 m from the source) exposure scenario.

12629

Page 467 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

TableApx G-10. RQs Based on Potential Trophic Transfer of TCEP from Fish to American Mink

as a Model Aquatic Pret

ator Using EPA's Wildlife Risk Model for Eco-SS

^s (Equation 4-1)

Scenario Name

Production
Volume
(lb/year)"

Release
Distribution

SWC6
WL)

Fish
Concentration
(mg/kg)

American Mink

TCEP in Biota
(mg/kg/day)

RQ

Import and repackaging

2,500

High-end

2,370

0.81

0.51

0.02

Incorporation into paints
and coatings - 1-part
coatings

2,500

High-end

10,300

3.50

2.21

0.08

Incorporation into paints
and coatings - 2-part
reactive coatings

2,500

High-end

9,340

3.18

2.01

0.07

Use in paints and coatings
at job sites

2,500

High-end

5,580

1.90

1.20

0.04

Formulation of TCEP
containing reactive resin

2,500

High-end

10,900

3.71

2.34

0.08

Laboratory chemicals

2,500

High-end

96

3.2E-02

0.02

7.0E-04

Import and repackaging

25,000

Central
tendency

1,720

0.58

0.37

0.01

Incorporation into paints
and coatings - 1-part
coatings

25,000

Central
tendency

3,230

1.10

0.69

0.02

Incorporation into paints
and coatings - 2-part
reactive coatings

25,000

Central
tendency

19,300

6.56

4.15

0.14

Use in paints and coatings
at job sites

25,000

Central
tendency

555

0.19

0.12

4.1E-03

Processing into 2-part resin
article

25,000

Central
tendency

15,800

5.37

3.39

0.12

Laboratory chemicals

25,000

Central
tendency

663

0.23

0.14

5.0E-03

a Production volume of 2,500 lb TCEP/yr uses high-end estimates (95th percentile for all COUs except the laboratory
chemicals COU uses the 1st percentile). Production volume of 25,000 lb TCEP/yr uses central tendency estimates
(median).

b TCEP Surface Water Concentration (SWC) calculated using WWM-PSC.

Page 468 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

12633	Appendix H GENERAL POPULATION EXPOSURE DETAILS

12634	H.l Exposure Factors

12635

12636	Table Apx H-l. Body Weight by Age Group	

Age Group"

Mean Body Weight (kg)''

Infant (<1 year)

7.83

Young toddler (1 to <2 years)

11.4

Toddler (2 to <3 years)

13.8

Small child (3 to <6 years)

18.6

Child (6 to <11 years)

31.8

Teen (11 to <16 years)

56.8

Adults (16 to <70 years)

80.0

a Age group weighted average
U.S. EPA ('. .Table 8-1

12637

12638	Table Apx H-2. Fish Ingestion Rates by Age Group

Age Group

Fish Ingestion Rate
(g/kg-day)"

50th Percentile

90th Percentile

Infant (<1 year)6

N/A

N/A

Young toddler (1 to <2 years)b

0.053

0.412

Toddler (2 to <3 years)6

0.043

0.341

Small child (3 to <6 years)6

0.038

0.312

Child (6 to <11 years)6

0.035

0.242

Teen (11 to <16 years)6

0.019

0.146

Adult (16 to <70 years)c

0.063

0.277

Subsistence fisher (adult)'7

1.78

a Age group weighted average, using body weight from Table Apx H-l above
feU.S. EPA (2014a). Table 20a
CU.S. EPA (2014a). Table 9a

dU.S. EPA (2000b)

12639

Page 469 of 572

-------
12640

12641

12642

12643

12644

12645

12646

12647

12648

12649

12650

12651

12652

12653

12654

12655

12656

12657

12658

12659

12660

12661

12662

12663

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

H.2 Water Pathway

H.2.1 Surface Water and Groundwater Monitoring Database Retrieval and Processing

The complete set of TCEP monitoring results stored in the WQP was retrieved in March 2023, with no
filters applied other than the chemical name (NWQMC. 2022). This raw dataset included 17,521
samples. To filter down to only the desired surface water samples to include in this analysis, only
samples with the "ActivityMediaSubdivisionName" attribute of "Surface Water" were kept. The dataset
removed values that that were below the detection limit.

After these steps, a total of 466 surface water samples and 51 groundwater samples remained in the
dataset. This monitoring dataset is attached as the Draft Risk Evaluation for Tris(2-chloroethyl)
Phosphate (TCEP) - Supplemental Information File: Water Quality Portal Processed Water Data (

1 23m).

H.2.1.1 Water Plots and Figures Generated in R

Exploratory analysis of the WQP data were conducted in R. An Rmarkdown file summarizing the steps
taken to explore, wrangle and visualize this dataset is available at	xessible Link to Interactive

Figure.

The Water Media Maps and Time Series Graphs are interactive plots made with the leaflet and plotly
packages. Clicking on the points in the water media maps displays summary information of the
associated data point. Similarly hovering over the data points in the Time Series Graphs provides
summary information of the plotted data point. Media can be selected and de-selected in the legend to
display and remove select media from the figures. The tiles to the left in the media maps allow for
different map layers (Esri.WorldGrayCanvas, OpenStreetMap, Esri.WorldTopoMap) and allows users to
select and deselect the underlying datasets.

Page 470 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE

December 2023

Map of Water monitoring in the United States (excludes non-detects)

CANADA

610 00
ug/l

08S 19E 02DCCD1
Well
Dissolved

OrganizationUSGS Idaho Water Science Center
Provider	NWIS

ft

Media

Effluent
Finished Water
Groundwater
Hyporheic zone
Landfill effluent
Leachate
Surface Water
Missing

1 7j^

UNITED STATES



Leaflet | Tiles © Esri — Esri, DeLorme, NAVTEQ

Time Series Graphs

Plot of Water in the United States by Time (excluding non-detects on log scale)

(Apr 2002, 2.78533)

I Groundwater in i. ® fj ~ xf

610 ug/l





Idaho





•

08S 19E 02DCCD1





•

Well





•

Dissolved





•

USGS Idaho Water Science Center





•

NWIS





•

Hyporheic zone
Landfill effluent
Leachate

•	Surface Water

•	NA

c

-------
12667

12668

12669

12670

12671

12672

12673

12674

12675

12676

12677

12678

12679

12680

12681

12682

12683

12684

12685

12686

12687

12688

12689

12690

12691

12692

12693

12694

12695

12696

12697

12698

12699

12700

12701

12702

12703

12704

12705

12706

12707

12708

12709

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

H.2.2 Methodology for Obtaining New Flow Data (2015 to 2020)

The following steps were utilized to retrieve more recent flow data for the TCEP environmental
assessment (flow values for the 2015 to 2020 are summarized in Draft Risk Evaluation for Tris(2-
chloroethyl) Phosphate (TCEP) - Supplemental Information File: E-FASTModeling Results (

2023 e):

1.	SIC codes assigned to TCEP were provided: 2851, 4952, 2821, 2823, 2824.

2.	Wastewater discharge facility information was obtained for all facilities assigned to each of the
SIC codes using the "echoWaterGetFacilitylnfo" function in the echor package in R. This results
in -47,000 facilities.

3.	A data field was added to categorize the SIC codes into new industrial sector names as described
in Table 3 of Versar's "Facility and Stream Flow Database" document. These include "Paint
Formulation," "POTWs—All facilities," and "Adhesives, Sealants, Plastics, Resins, Rubber, and
Manufacturing."

4.	For the 4952 SIC code, only facilities with a "POTW" indicator in the permit component data
field were included. This results in a list of-19,000 facilities. This step was taken in parallel to
one described in EPA Contractor Versar's "Facility and Stream Flow Database" document,
where instead of acquiring facilities with a 4,952 SIC designation, all NPDES with a POTW
permit component were retrieved from the water facility search tool in ECHO. Note: Versar also
created a subset "Industrial POTW" category by extracting NPDES permits with a "Y" pre-
treatment indicator from the "POTW—All facilities" category, using the ICI-NPDES database
on the ECHO website.

5.	Any duplicate NPDESs were excluded.

6.	Four hundred facilities were selected at random without replacement from each industrial sector
group. This step was taken because 19,000 facilities is too many to acquire NHD flow
information for in a timely manner.

7.	NHD 14-digit reach codes were retrieved from the ECHO
"dmr_rest_services.get_facility_report" backend server for each unique NPDES/permit that was
active between 2015 to 2020, thus narrowing the facilities to only those with active permits
during this time.

8.	Facilities where a NPDES identifier could not be matched with a NHD reach code were
excluded. 877 facilities had active permits during this time period and which also included
reported NHD reach codes.

9.	For each unique NPDES-reach code combination, mean and monthly average flow data were
retrieved from the NHD flowline database. Exposure related flow metrics (e.g., 7Q10 and 30Q5)
were then calculated using methods established by the 1,4-D and 1,1-DCA teams.

10.	The distribution of flows was plotted

11.	A summary statistics table was created for each of the industrial SIC categories.

H.2.3 E-FAST: Predicted Flowing Surface Water Concentrations (First Tier Modeling)

EPA's E-FAST, Version 2.0, was specifically developed to support EPA assessments of potential
environmental exposures. The E-FAST Model contains default parameter values that allow for exposure
estimations of a chemical in the surface water after a source emits the chemical into a water body
considering simple dilution. EPA uses H-l to estimate surface water concentrations in E-FAST.

Page 472 of 572

-------
12710

12711

12712

12713

12714

12715

12716

12717

12718

12719

12720

12721

12722

12723

12724

12725

12726

12727

12728

12729

12730

12731

12732

12733

12734

12735

12736

12737

12738

12739

12740

12741

12742

12743

12744

12745

12746

12747

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

EquationApx H-l

Where:

SWC
R

CF1

T

SF

Surface water concentration in |ig/L
Release kg/site/day
Conversion factor (109 |ig/kg)

Percent removal, typically from wastewater treatment
Flow of receiving river (million liters per day)

CF2 = Conversion factor (106 L/day/MLD)

Inputs

Release (kg/site/day): As discussed in Section 3.2, the daily release values (kg/site/day) were calculated
using a production volume of 2,500 lb/year, 25,000 lb/yr, emission factors (kg TCEP released/kg TCEP
handled), and number of release days per year. Refer to Table 3-3 for a summary of the release values
by COU, and for sub-scenario-specific release values.

Removalfrom Wastewater Treatment (%): Removal from wastewater treatment is the percentage of the
chemical removed from wastewater during treatment before discharge to a body of water. Although
removal from wastewater treatment for TCEP was estimated as 0 percent. This is a conservative
estimate relative to what is indicated in Table 2-2 that indicates wastewater removal to be 5 percent for
primary treatment and 19.1 percent for complete treatment (Kim et ai. 2017). EPA assumed that "on-
site WWT," "POTW" release types and direct releases to water did not receive wastewater treatment and
no wastewater treatment removal was applied. This is a conservative assumption that results in the total
amount of TCEP released to wastewater treatment at a direct discharging site being released to surface
water. It reflects the uncertainty of the type of wastewater treatment that may be in use at a direct
discharging facility and the TCEP removal efficiency in that treatment.

Flow of Receiving River (Million L/Day): E-FAST requires the selection of a receiving stream flow from
the E-FAST 2014 database. For site-specific assessments, the stream flow is selected by searching for a
facility's NPDES permit number, name, or the known discharging waterbody reach code. As no specific
facilities were identified for the TCEP assessment for water releases, stream flows were selected using
the "SIC Code Option" within E-FAST. This option uses the 10th and 50th percentile stream flows of all
facilities in a given industry sector, as defined by the SIC codes of the industry sector. The associated
SIC Codes for the COU/OES are organized as presented in Table Apx H-3 below:

Page 473 of 572

-------
12748

12749

12750

12751

12752

12753

12754

12755

12756

12757

12758

12759

12760

12761

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table Apx H-3. Crosswalk of CPU and PES, Abbreviations, and Relevant SIC Codes

cou

OES

Abbreviation

SIC Code

Manufacturing - import - import

Repackaging of import
containers

MFG-IMP

POTW All

Processing - incorporation into
formulation, mixture, or reaction
product - flame retardant in: Paint
and coating manufacturing

Incorporation into paints and
coatings - 1-part coatings

PAINT-WB

Paint Formulation

Processing - incorporation into
formulation, mixture, or reaction
product - flame retardant in: Paint
and coating manufacturing

Incorporation into paints and
coatings - 2-part reactive
coatings

PAINT-SB

Paint Formulation

Commercial use - paints and
coatings

Use in paints and coatings at
job sites

COM

POTW All

Processing - incorporation into
formulation, mixture, or reaction
product - flame retardant in:
Polymers

Formulation of TCEP
containing reactive resin

PROC

Plastic Resins and
Synthetic Fiber
Manufacture

Use of laboratory chemicals

Wastewater to onsite
treatment or discharge to
POTW (with or without
pretreatment)

LAB

POTW All

These SIC Code stream flows were selected because they were thought to best represent the industrial
activity associated with the COUs and release type.

The flow of rivers is highly variable and is dependent on many factors such as weather patterns and
effluent released from different facilities. The volume of a river varies over time with different flows
expected seasonally and from year to year. The 50th percentile 7Q10 flows represent the lowest
expected weekly flow over a 10-year period and were selected for use in the ecological risk assessment.
The flows for the selected industry sector/SIC Code are shown in Table Apx H-4. Although not used in
the ecological assessment, harmonic means are also shown since they were used to calculate surface
water concentrations for the scenario specific fish ingestion scenario in the highly exposed human
exposure assessment. Harmonic mean flow values represent long-term average flow conditions.

Page 474 of 572

-------
12762

12763

12764

12765

12766

12767

12768

12769

12770

12771

12772

12773

12774

12775

12776

12777

12778

12779

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

TableApx H-4. Harmonic Mean, 30Q5, 7Q10, and 1Q10 50th Percentile Flows for Relevant
TCEP SIC Codes

Sector within E-FAST

Year(s)

Harmonic Mean

Flow MLD
(50th Percentile)

30Q5 Flow
MLD
(50th
Percentile)

7Q10 Flow
MLD
(50th
Percentile)

1Q10 Flow
MLD
(50th
Percentile)

SIC Code - POTW -
All Facilities

2009

1.11E01

1.94E00

1.06E00

9.60E-01

2015-2020

1.15E01

7.23E00

4.13E00

3.47E00

SIC Code - Paint
Formulation

2009

3.54E01

1.25E01

7.29E00

6.10E00

2015-2020

9.21E00

5.95E00

3.38E00

2.84E00

SIC Code - Plastic
Resins and Synthetic
Fiber Manufacture

2009

4.45E01

1.37E01

8.02E00

7.44E00

2015-2020

6.51E00

5.05E00

2.85E00

2.40E00

Outputs

Draft_RE_Exp_EFAST_Modeling 20230626.xlsx provides the inputs, outputs, and equations that were
utilized for calculating surface water concentrations of TCEP, drinking water estimates, diluted drinking
water estimates, incidental oral ingestion estimates from swimming and incidental dermal absorption
estimates from swimming.

Advantages to the E-FAST Model are that it requires minimal input parameters, and it has undergone
extensive peer review by experts outside of EPA. The limitations associated with use of the E-FAST
Model relate to the assumptions made regarding use of sector-based flow information as a surrogate for
site-specific flow information, as well as lack of partitioning (between dissolved and suspended
sediment within the water column or between the water column and the benthic environment) and
degradation parameters that were employed in the PSC model. Additionally, note that low-flow stream
inputs combined with high-release estimates may yield overly conservative surface water concentrations
greater than the water solubility of TCEP.

Page 475 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

H.2.3.1 E-FAST Exposure Activity Parameters

Table Apx H-5. Incidental Dermal (Swimming) JV

odeling Parameters

Input

Description
(Units)

Adult

(>21
years)

Youth
(11-15
years)

Child
(6-10
years)

Notes

Reference

BW

Body weight (kg)

80

56.8

31.8

EPA Exposure Factors Handbook Chapter
8 (2011), Table 8-1 mean body weight

U.S. EPA.
2> ;5096

SA

Skin surface area
exposed (cm2)

19,500

15,900

10,800

U.S. EPA Swimmer Exposure Assessment
Model (SWIMODEL), 2015

U.S. EPA.
2< 1

ET

Exposure time
(hr/day)

3

2

1

High-end default short-term duration from
U.S. EPA Swimmer Exposure Assessment
Model (SWIMODEL), 2015.

U.S. EPA.

7



ED

Exposure duration
(years for ADD)

33

5

5

Number of years in age group, up to the
95th percentile residential occupancy
period. EPA Exposure Factors Flandbook
Chapter 16(2011), Table 16-5.

U.S. EPA.

15096

AT

Averaging time
(years for ADD)

33

5

5

Number of years in age group, up to the
95th percentile residential occupancy
period. EPA Exposure Factors Flandbook
Chapter 16(2011), Table 16-5.

U.S. EPA.

15096

Kp

Permeability
coefficient (cm/hr)

2.20E-03

CEM estimate aqueous Kp based on log
Kow of 1.25

Abdallali et al
2016.3120332

Table Apx H-6. Incidental Oral Ingestion (Swimming) Modeling Parameters

Input

Description
(Units)

Adult

(>21
years)

Youth
(11-15

years)

Child
(6-10
years)

Notes

Reference

IRinc

Ingestion rate (L/hr)

0.092

0.152

0.096

EPA Exposure Factors Handbook
Chapter 3 (2019), Table 3-7, upper
percentile ingestion while swimming.

U.S. EPA.
7267482

BW

Body weight (kg)

80

56.8

31.8

EPA Exposure Factors Handbook
Chapter 8 (2011), Table 8-1 mean body
weight.

U.S. EPA.

2011,

7485096

ET

Exposure time
(hr/day)

3

2

1

High-end default short-term duration
from U.S. EPA Swimmer Exposure
Assessment Model (SWIMODEL),
2015; based on competitive swimmers
in the age class.

U.S. EPA.

2015,

6811897

IRinc-

daily

Incidental daily
ingestion rate
(L/day)

0.276

0.304

0.096

Calculation: ingestion rate x exposure
time



IR/BW

Weighted incidental
daily ingestion rate
(L/kg-day)

0.0035

0.0054

0.0030

Calculation: ingestion rate/body weight



ED

Exposure duration
(years for ADD)

33

5

5

Number of years in age group, up to the
95th percentile residential occupancy
period. EPA Exposure Factors
Handbook Chapter 16 (2011), Table 16-
5.

U.S. EPA,

2011,

7485096

Page 476 of 572

-------
12785

12786

12787

12788

12789

12790

12791

12792

12793

12794

12795

12796

12797

12798

12799

12800

12801

12802

12803

12804

12805

12806

12807

12808

12809

12810

12811

12812

12813

12814

12815

12816

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Input

Description
(Units)

Adult

(>21
years)

Youth
(11-15

years)

Child
(6-10
years)

Notes

Reference

AT

Averaging time
(years for ADD)

33

5

5

Number of years in age group, up to the
95th percentile residential occupancy
period. EPA Exposure Factors
Handbook Chapter 16 (2011), Table 16-
5.

U.S. EPA.
2011.

7485096

CF1

Conversion factor
(mg/^g)

1.00E-03





CF2

Conversion factor
(days/year)

365





H.2.4 VVWM-PSC: Predicted Flowing Surface Water Concentrations (Second Tier
Modeling)

Site-specific parameters influence how partitioning occurs over time. For example, the concentration of
suspended sediments, water depth, and weather patterns all influence how a chemical may partition
between compartments. Physical and chemical properties of the chemical itself also influence
partitioning and half-lives into environmental media. TCEP has a Koc greater than 100, indicating a high
potential to sorb to suspended particles in the water column and settled sediment in the benthic
environment.

EPA conducted higher tier modeling with PSC-VVWM to estimate benthic concentrations (porewater
and sediment).

H.3 Ambient Air Pathway

This section provides an overview of EPA's screening level methodology for the ambient air pathway.
Where reasonably available, fugitive and stack air release data from the 2019 TRI are used to quantify
environmental releases. No TRI data were available for TCEP. EPA used estimated releases from a
hypothetical facility using TCEP for the COUs (Figure Apx H-2).

AERMOD is used to estimate ambient air concentrations and exposures to human populations at various
distances from the emission source. Distances of up to 10,000 m are evaluated to capture potential
exposures and associated risks to fenceline communities. A distance of 10,000 m is used for this
methodology to capture populations nearer to releasing facilities than may otherwise be evaluated under
other EPA administered laws. Additionally, professional knowledge and experience regarding exposures
associated with the ambient air pathway find risks frequently occur out to approximately 1,000 m from a
releasing facility and quickly decrease farther out. Although 10,000 m is an order of magnitude farther
out than where risks are expected to occur, 10,000 m provides an opportunity to capture other factors
related to potential exposure and associated potential risks via the ambient air pathway (like multiple
facilities impacting a single individual) providing flexibility for screening level analyses for future risk
evaluations. While 10,000 m is used for the outer distance in the screening level analysis, the
methodology is not limited to 10,000 m. If risks are identified out to 10,000 m, then additional analysis
using the screening level methodology can be extended to farther distances for purposes of identifying
where risks may fall below levels of concern.

Page 477 of 572

-------
12817

12818

12819

12820

12821

12822

12823

12824

12825

12826

12827

12828

12829

12830

12831

12832

12833

12834

12835

12836

12837

12838

12839

12840

12841

12842

12843

12844

12845

12846

12847

12848

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

• Stack and * Ambient air • Inhalation ~ Fenceline
fugitive air concentrations	Community

releases	from

iv

Source	Pathway	Route ^ Receptors

modeling

Distance
from
Source

FigureApx H-2. Overview of EPA's Screening Level Ambient Air Pathway Methodology

H.3.1 Modeling Approach for Estimating Concentrations in Ambient Air

EPA applied a tiered approach to estimate ambient air concentrations and exposures for members of the
general population that are in proximity (between 10 to 10,000 m) to emissions sources emitting the
chemicals being evaluated to the ambient air. All exposures were assessed for the inhalation route only.
For TCEP, multi-year release data were not available.

Step 1: Ambient Air: IIOAC Methodology

Methodology is scenario-specific. Analysis evaluates ambient air concentrations and associated
exposures/risks resulting from facility-specific releases at three pre-defined distances (100, 100 to
1,000, and 1,000 m) from a releasing facility.

Step 2: Ambient Air: AERMOD Methodology

Methodology is scenario-specific. Analysis evaluates ambient air concentrations and associated
exposures/risks, and deposition concentrations to land and water, resulting from facility-specific
releases at eight finite distances (10, 30, 60, 100, 1,000, 2,500, 5,000, and 10,000 m) and two area
distances (30 to 60 m and 100 to 1,000 m) from each releasing facility (or generic facility for
alternative release estimates).

H.3.2 Ambient Air: Screening Methodology

The Ambient Air: IIOAC Methodology identifies, at a high level, if there are inhalation exposures to
select human populations from a chemical undergoing risk evaluation that indicates a potential risk. This
methodology inherently includes both estimates of exposures as well as estimates of risks to inform the
need, or potential need, for further analysis. If findings from the Ambient Air: IIOAC Methodology
indicate any potential risk (acute non-cancer, chronic non-cancer, or cancer) for a given chemical above
(or below as applicable) typical Agency benchmarks, EPA generally will conduct a higher tier analysis
of exposures and associated risks for that chemical. If findings from the Ambient Air: IIOAC
Methodology do not indicate any potential risks for a given chemical above (or below as applicable)
typical agency benchmarks, EPA would not expect a risk would be identified with higher tier analyses,
but may still conduct a limited higher tier analysis at select distances to ensure potential risks are not
missed (e.g., at distances <100 m to ensure risks do not appear very near a facility where populations
may be exposed).

Page 478 of 572

-------
12849

12850

12851

12852

12853

12854

12855

12856

12857

12858

12859

12860

12861

12862

12863

12864

12865

12866

12867

12868

12869

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Model

EPA's IIOAC model49 was used to estimate high-end and central tendency (mean) exposures to select
human populations at three pre-defined distances from a facility releasing a chemical to the ambient air
(100, 100 to 1,000, and 1,000 m). IIOAC is a spreadsheet-based tool that estimates indoor and outdoor
air concentrations using pre-run results from a suite of dispersion scenarios run in a variety of
meteorological and land-use settings within EPA's AERMOD. As such, IIOAC is limited by the
parameterizations utilized for the pre-run scenarios within AERMOD (meteorologic data, stack heights,
distances, populations, etc.) and any additional or new parameterization would require revisions to the
model itself. Readers can learn more about the IIOAC model, equations within the model, detailed input
and output parameters, pre-defined scenarios, default values used, and supporting documentation by
reviewing the IIOAC users guide (	19g).

Releases

EPA modeled exposures for the following list of COUs/OES that had air releases. EPA ran two
scenarios for each release scenario:

1.	Central Tendency (50th percentile) Estimate for High Production Volume (25,000 lb) - HIGH-
CT; and

2.	High End (95th percentile) Estimate for Low Production Volume (2,500 lb) - LOW HE.

TableApx H-7. Ambient Air Release Inputs Utilized for Ambient Air Modeling: IIOAC and
AERMOD Methodology for TCEP					

Scenario Name

Production
Volume

Estimate

Fugitive/
Stack

Release Duration
(hours/day)

Release
Frequency
(days/year)

Release
Amount
(kg/site/day)

COM-Paints-USE

LOW

HE

Fugitive

8 hr/day (8-4 pm)

2

1.14E02

IND-LabChem-USE

LOW

HE

Fugitive

8 hr/day (8-4 pm)

235

2.32E-04

IND-LabChem-USE

LOW

HE

Stack

8 hr/day (8-4 pm)

235

2.32E-04

MFG-Repack

LOW

HE

Fugitive

1 hr/day (12-1 pm)

4

3.43E-04

MFG-Repack

LOW

HE

Stack

1 hr/day (1 pm)

4

3.43E-04

PROC-Article-PROC-
twopart-resin

LOW

HE

Fugitive

8 hr/day (8-4 pm)

109

4.22E-04

PROC-Article-PROC-
twopart-resin

LOW

HE

Stack

8 hr/day (8-4 pm)

109

4.22E-04

PROC-Paints-INC-2-part
reactive coatings

LOW

HE

Fugitive

8 hr/day (8-4 pm)

1

7.90E-03

PROC-Paints-INC-2-part
reactive coatings

LOW

HE

Stack

8 hr/day (8-4 pm)

1

1.99E-02

PROC-Paints-INC-1 -part

LOW

HE

Fugitive

8 hr/day (8-4 pm)

4

9.60E-03

PROC-Paints-INC-1 -part

LOW

HE

Stack

8 hr/day (8-4 pm)

4

9.60E-03

PROC-Polymer-FORM-
reactive-resin

LOW

HE

Fugitive

8 hr/day (8-4 pm)

1

8.83E-03

49 The IIOAC website is available at https://www.epa.gov/tsca-screening-tools/iioac-integrated-indoor-ontdoor-air-calailator.

Page 479 of 572

-------
12870

12871

12872

12873

12874

12875

12876

12877

12878

12879

12880

12881

12882

12883

12884

12885

12886

12887

12888

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Scenario Name

Production
Volume

Estimate

Fugitive/
Stack

Release Duration
(hours/day)

Release
Frequency
(days/year)

Release
Amount
(kg/site/day)

PROC-Polymer-FORM-
reactive-resin

LOW

HE

Stack

8 hr/day (8-4 pm)

1

2.07E-02

COM-Paints-USE

HIGH

CT

Fugitive

8 hr/day (8-4 pm)

1

1.23E01

IND-LabChem-USE

HIGH

CT

Fugitive

8 hr/day (8-4 pm)

230

1.35E-04

IND-LabChem-USE

HIGH

CT

Stack

1 hr/day (1 pm)

230

1.35E-04

MFG-Repack

HIGH

CT

Fugitive

1 hr/day (12-1 pm)

39

1.88E-04

MFG-Repack

HIGH

CT

Stack

1 hr/day (1 pm)

39

1.88E-04

PROC-Article-PROC-
twopart-resin

HIGH

CT

Fugitive

8 hr/day (8-4 pm)

231

1.43E-04

PROC-Article-PROC-
twopart-resin

HIGH

CT

Stack

8 hr/day (8-4 pm)

231

1.43E-04

PROC-Paints-INC-2-part
reactive coatings

HIGH

CT

Fugitive

8 hr/day (8-4 pm)

4

6.77E-03

PROC-Paints-INC-2-part
reactive

HIGH

CT

Stack

8 hr/day (8-4 pm)

4

5.63E-03

PROC-Paints-INC-1 -part

HIGH

CT

Fugitive

8 hr/day (8-4 pm)

52

1.63E-03

PROC-Paints-INC-1 -part

HIGH

CT

Stack

8 hr/day (8-4 pm)

52

1.63E-03

PROC-Polymer-FORM-
reactive-resin

HIGH

CT

Fugitive

8 hr/day (8-4 pm)

6

5.36E-03

PROC-Polymer-FORM-
reactive-resin

HIGH

CT

Stack

8 hr/day (8-4 pm)

8

3.72E-03

Exposure Scenarios

EPA modeled exposure scenarios for two source types: stack (point source) and fugitive (area source)
releases. These source types have different plume and dispersion characteristics accounted for
differently within the IIOAC model. All COUs had stack and fugitive emissions except for the
commercial use of paints and coatings (COM-Paints-USE).

The topography represents an urban or rural population density and certain boundary layer effects (like
heat islands in an urban setting) that can affect turbulence and resulting concentration estimates at
certain times of the day. EPA ran both urban and rural population density for all scenarios.

IIOAC includes 14 pre-defined climate regions (each with a surface station and upper-air station). Since
release data used for the Ambient Air: IIOAC Methodology was not facility- or location-specific, EPA
selected 1 of the 14 climate regions to represent a high-end (South [Coastal]) climate region. This
selection was based on a sensitivity analysis of the average concentration and deposition predictions.
This climate regions selected represents the meteorological dataset that tended to provide high-end
concentration estimates relative to the other stations within IIOAC. The meteorological data within the
IIOAC Model are from years 2011 to 2015 as that is the meteorological data utilized in the suite of pre-
run AERMOD exposure scenarios during development of the IIOAC model (see (	;)).

Page 480 of 572

-------
12889

12890

12891

12892

12893

12894

12895

12896

12897

12898

12899

12900

12901

12902

12903

12904

12905

12906

12907

12908

12909

12910

12911

12912

12913

12914

12915

12916

12917

12918

12919

12920

12921

12922

12923

12924

12925

12926

12927

12928

12929

12930

12931

12932

12933

12934

12935

12936

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

While this is older meteorological data, sensitivity analyses related to different years of meteorological
data found that although the data does vary, the variation is minimal across years so the impacts to the
model outcomes remain relatively unaffected.

The release scenarios were informed by the release duration and release frequency that were provided in
Section 3.2.

Results

TCEP_IIOAC_04272023.xlsx presents the overall inputs and outputs for IIOAC. In IIOAC, all
calculated air concentrations of fine and coarse particles are capped by an upper limit equal to the
National Ambient Air Quality Standards (NAAQS) for particulate matter (PM) (U.S. EPA. 2016c).
These limits are 35 and 150 [j,g/m3 for fine and coarse particles {i.e., the NAAQS for PM2.5 and PM10),
respectively. For the IIOAC results, these limits were met for all the COU/OES releases with stack
emissions. In addition, this limit reach was reached for the fine, fugitive emissions, LOW-HE release
scenario for the commercial use of paints and coatings.

A further limitation of IIOAC is that it does not model for gaseous deposition. Due to the inability to
model gaseous deposition, and due to the initial screening results meeting the NAAQS caps, EPA
decided to run a higher tier model (AERMOD) for the ambient air pathway.

H.3.3 Ambient Air: AERMOD Methodology

The Ambient Air: AERMOD Methodology was developed to allow EPA to conduct a higher tier
analysis of releases, exposures, and associated risks to human populations around releasing facilities at
multiple distances when EPA has site-specific data like reported releases, facility locations (for local
meteorological data), source attribution, and other data when reasonably available. This methodology
can also incorporate additional site-specific information like stack parameters (stack height, stack
temperature, plume velocity, etc.), building characteristics, release patterns, different terrains, and other
parameters when reasonably available. AERMOD can be performed independent of the Tier 1 modeling
described above, provides a more thorough analysis, can include wet and dry deposition estimates, and
allows EPA to fully characterize identified risks for chemicals undergoing risk evaluation. The
application of this methodology can be applied to single or multiple years of data. TCEP had no TRI or
NEI data. Thus, air releases from the release assessment were used to estimated ambient air
concentrations for a single year.

Model

The Ambient Air: AERMOD Methodology for this draft risk evaluation utilizes AERMOD to estimate
TCEP exposures to fenceline communities at user defined distances from a facility releasing TCEP.
AERMOD is a steady-state Gaussian plume dispersion model that incorporates air dispersion based on
planetary boundary layer turbulence structure and scaling concepts, including treatment of both surface
and elevated sources and both simple and complex terrain. AERMOD can incorporate a variety of
emission source characteristics, chemical deposition properties, complex terrain, and site-specific hourly
meteorology to estimate air concentrations and deposition amounts at user-specified population
distances and at a variety of averaging times. Readers can learn more about AERMOD, equations within
the model, detailed input and output parameters, and supporting documentation by reviewing the
AERMOD Users Guide (	).

Releases

EPA modeled exposures using the release data developed as described in Section 3.2. Release data were
provided (and modeled) on a COU-by-COU basis as no facility information was available for TCEP.

Page 481 of 572

-------
12937

12938

12939

12940

12941

12942

12943

12944

12945

12946

12947

12948

12949

12950

12951

12952

12953

12954

12955

12956

12957

12958

12959

12960

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Exposure Points

The Ambient Air: AERMOD Methodology evaluated exposures to exposure points at eight finite
distances (10, 30, 60, 100, 1,000, 2,500, 5,000, and 10,000 m) and two area distances (30 to 60 m, and
100 to 1,000 m) from each releasing facility (or generic facility for alternative release estimates).
Exposure points for each of the eight finite distances were placed in a polar grid every 22.5 degrees
around the respective distance ring. This results in a total of 16 exposure points around each finite
distance ring for which exposures are modeled. FigureApx H-3 provides a visual depiction of the
placement of exposure points around a finite distance ring. Although the visual depiction only shows
exposure points locations around a single finite distance ring, the same placement of exposure points
occurred for all eight finite distance rings.

Receptor Locations around each Finite Distance Ring

100 -1,000 m

2,500 m

10,000 m

22.5 '

Releasing Facility

Figure Apx H-3. Modeled Exposure Points Locations for Finite Distance Rings

Exposure points for the area distance 30 to 60 m evaluated were placed in a cartesian grid at equal
distances between 40 and 50 m around each releasing facility (or generic facility for alternative release
estimates) were placed at 10-meter increments.

Exposure points for the area distance 100 to 1,000 m evaluated were placed in a cartesian grid at equal
distances between 200 and 900 m around each releasing facility (or generic facility for alternative
release estimates) were placed at 100-meter increments. This results in a total of 456 exposure points for
which exposures are modeled. Figure Apx H-4 provides a visual depiction of the placement of exposure
points (each dot) around the 100 to 1,000 m area distance ring.

Page 482 of 572

-------
12961

12962

12963

12964

12965

12966

12967

12968

12969

12970

12971

12972

12973

12974

12975

12976

12977

12978

12979

12980

12981

12982

12983

12984

12985

12986

12987

12988

12989

12990

12991

12992

12993

12994

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

All exposure points were at 1.8 m above ground, as a proximation for breathing height for ambient air
concentration estimations. A duplicate set of exposure points was at ground level (0 m) for deposition
estimations.

Meteorological Data

Meteorological data for EPA estimated releases (where TRI or city data were not available) were
modeled with the two meteorological stations utilized in the pre-screen methodology (Sioux Falls, South
Dakota, for central-tendency meteorology; Lake Charles, Louisiana, for higher-end meteorology). These
two meteorological stations represent meteorological datasets that tended to provide high-end and
central tendency concentration estimates relative to the other stations within IIOAC based on a
sensitivity analysis of the average concentration and deposition predictions conducted in support of
IIOAC development. These two meteorological stations are based on 5 years of meteorological data
(2011 to 2015) and provide high-end and central tendency exposure concentrations utilized for risk
calculation purposes to identify potential risks. The "ADJ U*" option was not used for the 2011 to 2015
data as this could lead to model overpredictions of ambient concentrations during those particular
conditions.

All processing also used automatic substitutions for small gaps in data for cloud cover and temperature.
Urban/Rural Designations

Urban/rural designations of the area around a facility are relevant when considering possible boundary
layer effects on concentrations.

Air emissions taking place in an urbanized area are subject to the effects of urban heat islands,
particularly at night. When sources are set as urban in AERMOD, the model will modify the boundary
layer to enhance nighttime turbulence, often leading to higher nighttime air concentrations. AERMOD
uses urban-area population as a proxy for the intensity of this effect.

Where TRI or city data were not available for a facility requiring modeling, there was no way for EPA
to determine an appropriate urban or rural designation. Instead, EPA modeled each such facility once as

Page 483 of 572

-------
12995

12996

12997

12998

12999

13000

13001

13002

13003

13004

13005

13006

13007

13008

13009

13010

13011

13012

13013

13014

13015

13016

13017

13018

13019

13020

13021

13022

13023

13024

13025

13026

13027

13028

13029

13030

13031

13032

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

urban and once as not urban.50 There is no recommended default urban population for AERMOD
modeling, so for these facilities EPA assumed an urban population of 1 million people, which is
consistent with the estimated populations used with IIOAC. Although slightly higher, the assumed urban
population is close to the average of all the urban populations used for the TRI reporting facilities
(which was 847,906 people).

For the TCEP risk evaluation EPA selected the urban air concentrations vs. rural air concentrations as
urban concentrations were generally more conservative. Rural air concentrations may be relevant for
facilities located in rural areas, and because TCEP has long range transport potential. However due to
lack of site-specific information for facilities, this risk evaluation used the more conservative urban air
estimates from AERMOD.

Physical Source Specifications for Alternative Release Estimates

EPA estimated releases (where TRI or city data were not available) were modeled centering all
emissions on one location and using IIOAC default physical parameters. Stack emissions were modeled
from a point source at 10 meters above ground from a 2-meter inside diameter, with an exit gas
temperature of 300 Kelvin and an exit gas velocity of 5 m/sec (Table 6 of the IIOAC User Guide).
Fugitive emissions were modeled at 3.05 m above ground from a square area source of 10 m on a side
(Table 7 of the IIOAC User Guide).

Deposition Parameters

AERMOD was used to model daily (g/m2/day) and annual (g/m2/year) deposition rates from air to land
and water at eight finite distances (10, 30, 60, 100, 1,000, 2,500, 5,000, and 10,000 m) and two area
distances (30 to 60 m and 100 to 1,000 m) from each releasing facility (or generic facility for alternative
release estimates).

AERMOD can model both gaseous and particle deposition. For TCEP, EPA considered both gaseous
and particle deposition. There is conflicting literature on whether TCEP is present in particulates vs. gas.
Section 3.3.1.2.1 discusses these differences. Input parameter values for AERMOD deposition modeling
are shown in TableApx H-8.

EPA provided the parameter values and settings for AERMOD deposition modeling, as indicated in
Table Apx H-8 and Table Apx H-9. The particle deposition utilized the "METHOD2" option in
AERMOD, which is recommended when particle size distributions are not well known and when less
than 10 percent of particles (by mass) are 10 |im or larger. Note that we modeled each scenario twice—
once with gaseous deposition utilizing land cover of "suburban area, forested" and once with "bodies of
water."

50 While this may be viewed as a potential double counting of these releases, EPA only utilized the highest estimated releases
from a single exposure scenario from the suite of exposure scenarios modeled for surrogate/estimated facility releases as
exposure estimates and for associated risk calculations.

Page 484 of 572

-------
13033

13034

13035

13036

13037

13038

13039

13040

13041

13042

13043

13044

13045

13046

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table Apx H-8. Settings for Gaseous Deposition

Parameter

Value

Source

Diffusivitv in air

5.67E-02 cm2/sec

Utilizing with the chemical properties
from Table 1 of Shin et al. (2014)

Diffusivity in
water

2.70E-05 cm2/sec

Paae 2310 ofMelnikovaet al. (2019)

Henry's Law
Constant

2.95E-06 Pa m3/mol

Not specified

rci: Cuticular
resistance to
uptake by lipids
for individual
leaves

3.26E03 sec/cm

Based on vapor pressure (Vp=8.13 Pa), empirical
relationships described bv Welke et; "») and (Kerler
and Schoenherr, 1988, dd. author-vear) and the values of rci

and of Vp available for numerous chemicals in Weselv et al.
(2002) —together, these implv a relationship of loe(rcl) =
0.4892*log(Vp in Pa) + 3.0682

Seasons

DJF = winter with no snow;
MAM = transitional spring
with partial green coverage
or short annuals; JJA =
Midsummer with lush
vegetation; SON. = Autumn
with unharvested cropland

Assumption

Land Cover

Option 1: Suburban areas,
forested; Option 2: Bodies
of water

A limited set of AERMOD tests suggested suburban-forest
was a reasonable and appropriately health-protective default
land-cover selection when land-cover analysis is not
possible. Bodies of water typically led to the highest
deposition values (ICF unpublished data).

Notes: Pa = Pascal; mol = mole; DJF = December-February; MAM = March-May; JJA = June-August; SON
= September-November.

Table Apx H-9. Sei

ttings for Particle Deposition

Parameter

Value

Source

Mass fraction 2.5
|im or smaller

0.4 |im

Based on ranges found for phosphates in (Delumyea

and Petel. 1979)12 and (Lee and Patterson. 1969)13

Mass-mean
diameter

2.2 |im

Based on a default for phosphates (source not
specified)

Cuticular Resistance

The cuticular resistance (rci) value represents the resistance of a chemical to uptake by individual leaves
in a vegetative canopy. For TCEP, rci was not readily available in literature. For chemicals for which the
rci value is not readily available in literature, EPA developed three methods to estimate the rci value. For
TCEP, EPA used rci value estimated using Method 2.

Method 1: Approximation of Rci Value as a Function of Vapor Pressure: Data from the literature
indicate that rci value varies as a function of the vapor pressure (VP, units of Pa) of a chemical (Welke et
ai. 1998; Kerler and Schoenherr. 1988). A high VP indicates that chemical has a high propensity for the
vapor phase relative to the condensed phase, and therefore, would have high resistance to uptake from

Page 485 of 572

-------
13047

13048

13049

13050

13051

13052

13053

13054

13055

13056

13057

13058

13059

13060

13061

13062

13063

13064

13065

13066

13067

13068

13069

13070

13071

13072

13073

13074

13075

13076

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

the atmosphere into leaves (i.e., high rci). Furthermore, Wesely et al. (2002) provides a large database of
VP and rci values.

Analysis of the Wesley et al. data reveals that there is a linear correlation between log(VP) and log(rci),
as illustrated in FigureApx H-5 and EquationApx H-2. Linear regression yields rci as a function of VP

(R2 = 0.606):

Equation Apx H-2

log(rcl) = 0.489 log (VP) + 3.068
rcl = 1170 VP0A98

16
14
12
10

8

o

4
2
0
-2
-4

-12 -10 -8 -6 -4 -2 0	2	4	6	8

log (VP)

Figure Apx H-5. Cuticular Resistance as a Function of Vapor Pressure

Method 2: Empirical Calculation of Cuticular Resistance: Method 2 estimates rci value using various
empirical equations found in literature. This method assumes the vapor pressure of the chemical at 20 to
25 °C is equal to the saturation vapor pressure. For VOCs, using the equations collectively provided
under Equation Apx H-3 (Welke et al.,) the polymer matrix-air partition coefficient (K\[Xa) can be
calculated as follows:

Equation Apx H-3

log(Kmxo) = 6.290 - 0.892 log(7P)

Next, KMxa can be converted to the cuticular membrane-air partition coefficient, Kcma:

Kcmci = 0.77 KMXa

Welke, et al. also provide an empirical relationship between the polymer matric-water partition
coefficient and the air-water partition coefficient, Kmxw. Recognizing the air-water partition coefficient
is the Henry's law constant, HLC (unitless), yields,

Kmxw = KMXa HLC

y = 0.4892x +3.0682
R2 = 0.6058

.

/ 	

	 * ,

Page 486 of 572

-------
13077

13078

13079

13080

13081

13082

13083

13084

13085

13086

13087

13088

13089

13090

13091

13092

13093

13094

13095

13096

13097

13098

13099

13100

13101

13102

13103

13104

13105

13106

13107

13108

13109

13110

13111

13112

13113

13114

13115

13116

13117

13118

13119

13120

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

This relationship can be generalized from the polymer matrix to the cuticular membrane.

KcMw — KcMa HLC

In a separate study, Kerler and Schoenhei 0 have developed an empirical relationship that equates
Kcmw to the permeance coefficient for cuticular membranes, Pcm. However, this relationship was
developed using data for non-volatile chemicals. Consequently, applying it to volatile organic chemicals
introduces a large amount of uncertainty to the analysis and may not be scientifically justifiable.

In the above equation, MV is the molecular volume of the chemical in question, which can be calculated
from the molar mass, m (units of g/mol), and density, d (units of g/cm3):

Finally, rci is understood to be the inverse of Pcm. The above relationships can be put together and
simplified to yield a single equation for rcl as a function of vapor pressure, molar mass, and density:

Method 3: Read across of Cuticular Resistance from an Analog: This method assumes that chemicals
that have structural similarity, physical and chemical similarity, and exhibit similar vapor pressures will
also exhibit similar rci values. Available data in literature (Wesety et al. 2002) can be used as a
crosswalk for read across determination of rci. The unknown rci value is then assumed to be equal to the
rci of the analog.

Ambient Air Exposure Concentration Outputs

Hourly-average concentration outputs were provided from AERMOD for each exposure points around
each distance ring (each of 16 exposure points around a finite distance ring or each exposure points
within the area distance ring). Daily and Period averages were then calculated from the modeled hourly
data. Daily averages for the finite distance rings were calculated as arithmetic averages of all hourly data
for each day modeled for each v around each ring. Daily averages for the area distance ring were
calculated as the arithmetic average of the hourly data for each day modeled across all exposure points
within the area distance ring. This results in the following number of daily average concentrations at
each distance modeled.

1. Daily averages for EPA estimated releases: Average concentrations for each of 365 (or 366) days
for each of 16 exposure points around each finite distance ring.

Period averages were calculated from all the daily averages for each exposure points for each distance
ring over one year for facilities where releases were estimated. This results in a total of 16 period
average concentration values for each finite distance ring. This is derived from either averaging the daily

-238 d

Page 487 of 572

-------
13121

13122

13123

13124

13125

13126

13127

13128

13129

13130

13131

13132

13133

13134

13135

13136

13137

13138

13139

13140

13141

13142

13143

13144

13145

13146

13147

13148

13149

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

averages across the single year of meteorological data used for TRI reporting facilities or across the
multi-year meteorological data used for EPA estimated releases.

Daily and period average Outputs were stratified by different source scenarios, such as urban/not urban
setting or emission-strengths where needed. Outputs from AERMOD are provided in units of
micrograms per cubic meter (|ig/m3) requiring conversion to parts per million (ppm) for purposes of
calculating risk estimates for 1,4-dioxane. The following formula was used for this conversion:

EquationApx H-4

CPPm= (24.45*(Caermod)/1,000)/MW

Where:

Cppm
24.45
Caermod
MW

Concentration (ppm)

Molar volume of a gas at 25 °C and 1 atmosphere pressure

Concentration from AERMOD (|ig/m3)

Molecular weight of the chemical of interest (g/mole)

Post-processing scripts were used to extract and summarize the output concentrations for each facility,
release, and exposure scenario. The following statistics for daily- and period-average concentrations
were extracted or calculated from the results for each of the modeled distances {i.e., each ring or grid of
exposure points) and scenarios:

•	Minimum

•	Maximum

•	Average

•	Standard deviation

•	10th, 25th, 50th, 75th, and 95th percentiles

Table Apx H-10. Description of Daily or Period Average and Air Concentration Statistics

Statistic

Description

Minimum

The minimum daily or period average concentration estimated at any exposure point on
any day at the modeled distance.

Maximum

The maximum daily or period average concentration estimated at any exposure point on
any day at the modeled distance.

Average

Arithmetic mean of all daily or period average concentrations estimated at all exposure
points locations on all days at the modeled distance. This incorporates lower values
(from days when the exposure point largely was upwind from the facility) and higher
values (from days when the exposure point largely was downwind from the facility).

Percentiles

The daily or period average concentration estimate representing the numerical percentile
value across the entire distribution of all concentrations at all exposure point locations
on any day at the modeled distance. The 50th percentile represents the median of the
daily or period average concentration across all concentration values for all exposure
point locations on any day at the modeled distance.

Page 488 of 572

-------
13150

13151

13152

13153

13154

13155

13156

13157

13158

13159

13160

13161

13162

13163

13164

13165

13166

13167

13168

13169

13170

13171

13172

13173

13174

13175

13176

13177

13178

13179

13180

13181

13182

13183

13184

13185

13186

13187

13188

13189

13190

13191

13192

13193

13194

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Deposition from Ambient Air to Soil and Water Exposure Concentration Outputs

As previously mentioned, AERMOD was used to model daily (g/m2/day) and annual (g/m2/year)
deposition rates {i.e., deposition flux) from air releases to water body catchment areas. EPA
quantitatively evaluated the risk to aquatic (pelagic and benthic) and terrestrial organisms from exposure
to soil, surface water bodies and sediment via air deposition resulting from the manufacturing,
processing, use, or disposal of TCEP. The following equations and parameters are based on the generic
farm pond scenario from models, such as the GENEEC2 (Generic Estimated Environmental
Concentration) and EXAM (Exposure Analysis Modeling System) used by EPA' Office of Pesticide
Programs (OPP) Environmental Fate and Effects Division (EFED). Total deposition for each media
(soil, water body, and sediment) were derived using the deposition rate modeled by AERMOD to
calculate media (soil, water body, and sediment) concentrations using the generic farm pond parameters
for area, mixing depths, and densities, respectively:

Soil:

EquationApx H-5.

Total Deposition to Soil Catchment {ug) = Deposition flux x Area x CF

Where:

Deposition flux
Area

CF

Annual deposition flux to water body catchment (g/m2)

Area of soil catchment (area of water body catchment - area of
water body) or 100,000 m2 - 10,000 m2 = 90,000 m2
g to (j,g; 1,000,000

Soil Catchment Concentration

{Total Deposition to Soil Catchent)

kg/ {Area of soil catchment x mix depth x soil density)

Where:

Area

Mix depth
Soil density

90,000 m2
0.1 m

1,700 kg/m3

Water Body:
Equation Apx H-6

T otal Deposition to Water Body {ug) = Deposition flux x Area x CF

Where:

Deposition flux

Area

CF

Annual deposition flux to water body catchment (g/m2)
Area of water body; 10,000 m2
gtoug; 1,000,000

mg

Water Body Concentration ( —

Total Deposition to Water Body

{Area x Pond Depth x CF)

Where:

Area

Pond depth
CF

area of water body; 10,000 m2
2 m

m3 to L; 1,000

Page 489 of 572

-------
13195

13196

13197

13198

13199

13200

13201

13202

13203

13204

13205

13206

13207

13208

13209

13210

13211

13212

13213

13214

13215

13216

13217

13218

13219

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Sediment:

EquationApx H-7

Sediment Concentration

(ug\
\kg)

Total Deposition to Water Body

(Area x mix depth x sediment density)

Where:

Area

Mix depth
Sediment density

Area of water body; 10,000 m2
0.1 m

1,300 kg/m

3

AERMOD Air Concentrations and Deposition Results

Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Supplemental Information File:
Exposure Air Concentration Risk Calculations (U.	2023h) includes the ambient air

concentrations, deposition concentrations (soil, water body, and sediment) for all OESs, and the
associated risk calculations.

H.4 Human Milk Pathway

TCEP is predicted to passively accumulate in human milk because it has a small mass (285.48 Da), is
slightly lipophilic (Log P = 1.78), and is a weak base (thus less likely to be ionized or protein bound).
The key chemical characteristics of TCEP are shown below in Table Apx H-l 1. Furthermore,
biomonitoring data confirmed TCEP's presence in human milk (He et ai. 2018a; Kim et ai. 2014;
Sundkvist et ai. 2010). Because of TCEP's potential to transfer to human milk and infants'
susceptibility to its health effects, a quantitative analysis of the milk pathway is necessary to predict
potential risks to infants. Milk concentrations were estimated based on the maternal doses using a multi-
compartment physiologically based pharmacokinetic (PBPK) model identified by EPA as the best
available model ("Verner et ai. 2009; Verner et ai. 2008). hereafter referred to as the Verner Model.

Page 490 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table Apx H-ll. Key Chemica

Characteristics of TCE

>

Key Question or
Decision

Result

Chemical Property or
Population

Current Value Used
for Analysis

Rcference(s)

Is the chemical
lipophilic (log P>1) and
less than 800 Da?

Yes

Average mass

285.49 Da

ConioTox Dashboard
(eoa.aov) 1 Tris(2-
cMoroetlivl) oliosoliate

Log Kow (Log P) from
Scoping review (Measured)

1.78

U.S. EPA (2020b)

Log Kow (Log P) from
other EPA sources

1.44, 1.78, 0.54-1.4

EPA, personal
communication

Log Kow (Log P, Predicted)

1.44108

ConioTox Dashboard
(eoa.aov) 1 Tris(2-
cliloroetlwl) oliosoliate

Is the chemical
hydrophilic and less than
200 Da?

No

Average mass

285.49 Da

ConioTox Dashboard
(eoa.eov) 1 Tris(2-
cliloroetlwl) oliosoliate

Water solubility (measured)

7,820 mg/L at 20 °C

U.S. EPA (2020b)

Is the chemical a weak
base?

Yes

pKa a

-9.1

Chemaxon

(littos://clieniaxon.coni/)li
ttos://clieniaxon.coni/

Phosphorus esters
hydrolysis rates available

NR

U.S. EPA (2020b)

Passive Diffusion
Prediction

Yes

Also supported by
topological polar surface
area (calculated)h

44.8 A

PubChem (nih.gov) 1

conioound/8295

Is there evidence of
passive diffusion in
peer-reviewed literature?

No

N/A

NR

N/A

Active Transport
Prediction

No

N/A

NR

N/A

Is there evidence of
active transport?

No

N/A

NR

N/A

Has the chemical been
detected in human milk?

Yes

Women in Australia, Japan,
Philippines, Vietnam, and
Sweden

Range: ND to 0.47
ng/mL

He et al. (2018a)

Central tendency: 0.14
ng/g to 42 ng/g lw

Kim et al. (2014)

Central tendency: 4.9
ng/g lw

Sundkvist et al. (2010)

Is there a measured
value for human milk
partition coefficient?

No

N/A

N/R

N/A

" The httD://www.t3db.ca/ database was searched for oka. but the oriuinal source for most chemicals was Chemaxon. a
proprietary software package. Efforts are underway to update pKa source data using published sources and/or QSAR
approaches using open-source code.

b The topological polar surface area of a molecule is defined as the surface sum over all polar atoms in a molecule.
Membrane Dcrmcabilitv is tvoicallv limited when oolar surface area (PSA) exceeds 140 A2. (Matsson and Kihlberg. 2017).

Page 491 of 572

-------
13221

13222

13223

13224

13225

13226

13227

13228

13229

13230

13231

13232

13233

13234

13235

13236

13237

13238

13239

13240

13241

13242

13243

13244

13245

13246

13247

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

H.4.1 Verner Model

The solubility of TCEP in the water of tissue and blood must be considered because it is slightly
lipophilic (log P = 1.78). EPA identified the Verner Model, a multi-compartment PBPK model that
distributes a chemical between different tissue compartments, as appropriate for evaluating infant
exposure to less lipophilic chemicals like TCEP. The Verner Model accounts for every female life stage
and includes data on maternal height, weight, and age. It also integrates several concurrent physiologic
events that are relevant to infant exposure from milk (e.g., pre- and postpartum changes in maternal
physiology, lactation, infant growth) and inputs physiological parameters, including organ volume,
composition, and blood flow throughout a woman's entire life. Note that the Verner Model was
validated using only data on persistent organic pollutants levels measured in mothers and infants from a
Northern Quebec Inuit population (Verner et al.. 2009). It was not validated using data on TCEP, which
were not available.

The Verner Model describes the period from the beginning of the mother's life to the first year of the
infant's life. As shown in FigureApx H-6, the model consists of a total of 14 compartments: 9 maternal
(uterus, brain, richly perfused tissue, poorly perfused tissue, adipose tissue, mammary tissue, liver,
placenta, and fetus) and 5 infantile (brain, richly perfused tissue, poorly perfused tissue, adipose tissue,
and liver). Distribution of the chemical is driven by blood flow and the partitioning between the blood
and the tissues.

Initial Body Burden





u

Richly Perfused

¦1

¦11





Poorly Perfused



Bj







Adipose Tissue

^¦1









Mammary Tissue



¦

ml



Liver

m

Richly Perfused

Poorly Perfused

Adipose Tissue

Figure Apx H-6. Compartments and Exposure Routes for Verner Model

Figure adapted from (Verner et al.. 2009).

EPA implemented the Verner Model in the R programming language to enable running the model using
modern R packages. The model was written as three systems of ordinary differential equations (ODEs),
corresponding to preconception, pregnancy, and breastfeeding. The number of compartments included in

Page 492 of 572

-------
13248

13249

13250

13251

13252

13253

13254

13255

13256

13257

13258

13259

13260

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

preconception, pregnancy, and breastfeeding are 7, 9, and 12, respectively. In addition, the following
additional updates were introduced into the R code:

•	Discontinuities related to physiological terms at ages 3 and 18 were corrected.

•	Mass balance tables were introduced for quality assurance evaluation.

•	Brain volume parameters were added (personal communication) (Verner et al. 2008).

•	A batch version of the code was developed to run several exposure scenarios consecutively.

•	Graphics were elaborated to visualize three key stages: conception, birth, and lactation.

•	Milk intake rates updated using EPA's Exposure Factors Handbook (	).

•	Model output expanded to include daily infant dose.

•	Model computes peak and average infant dose for each age group within the first year of life.
The model inputs are shown in Table Apx H-12 below.

Table Apx H-12.

)ata Input Requirements for the Multi-compartment Model

Input

Organs or Data

Data Source(s)

Blood flow

Mother: fetus, placenta, uterus, brain, richly
perfused tissue, poorly perfused tissue,
adipose tissue, mammary tissue, liver, heart

Infant: brain, richly perfused tissue, poorly
perfused tissue, adipose tissue, liver, heart

Calculated from eauations in (Verner et al..
2009; Verner et al.. 2008); blood flow to brain
was not published and estimated based on
correspondences with author

Organ volume

Mother: fetus, placenta, uterus, brain, richly
perfused tissue, poorly perfused tissue,
adipose tissue, mammary tissue, liver

Infant: brain, richly perfused tissue, poorly
perfused tissue, adipose tissue, liver

Calculated from eauations in (Verner et al..
2009; Verner et al.. 2008). Changes made to
skeletal muscles (part of poorly perfused
tissue) and extra fat, mammary, and uterine
volume at end of pregnancy to keep
parameters continuous

Fraction of lipid or
water in tissue

Mother: blood, brain, liver, adipose tissue,
richly perfused tissue, poorly perfused tissue,
mammary tissue, uterus, placenta

Infant: blood, adipose tissue, liver, richly
perfused tissue, poorly perfused tissue, brain

(Verner et al.. 2009; Verner et al.. 2008; Price
et al.. 2003; White et al.. 1991)

Tissue :blood

partition

coefficients

Mother: fetus, placenta, uterus, brain, richly
perfused tissue, poorly perfused tissue,
adipose tissue, mammary tissue, liver

Infant: brain, richly perfused tissue, poorly
perfused tissue, adipose tissue, liver

Calculated from Kow, fraction of lipid or
water in tissue of interest, and equation in
( ler et al.. 2008)

Milk:blood partition
coefficient

Same formula used for tissue:blood
coefficients

Calculated from Kow, fraction of lipid or
water in milk, and eauations in (Verner et al..
2008)

Fraction of lipids in
milk

Function of number of days post-partum, or
age of the child

( ler et al.. 2008)

Half-life (TCEP)

17.64 hours

Half-life is used to calculate a hepatic
extraction ratio that varies by age because it

Half-life value estimated from a one-
compartment model

litti3s://coniDtox.eDa.aov/dasliboard/clieniical/
adme-ivive-subtab/DTXSID:

Page 493 of 572

-------
13261

13262

13263

13264

13265

13266

13267

13268

13269

13270

13271

13272

13273

13274

13275

13276

13277

13278

13279

13280

13281

13282

13283

13284

13285

13286

13287

13288

13289

13290

13291

13292

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Input

Organs or Data

Data Source(s)



considers blood and tissue volumes that
change by age.



Oral dose

Default/User input

Derived from occupational, consumer, and
general population doses adjusted for body
weight representative of women of
reproductive age

Duration of
breastfeeding

Default/user input

One year is the default.

Volume of
breastfeeding

Default/user input

( ler et al.. 2009)

Description of Absorption, Distribution, and Excretion Parameters

The model is composed of three different stages: pre-conception, pregnancy, and breastfeeding. Each

dAf

model solves the rate of change of the amount — of the chemical in compartment t (tissue) as listed in,

Table Apx H-13 where At denotes the amount of chemical in the tissue. These rates of change are given
in terms of the blood flow to the tissue Qt, the compartment concentration Ct, the tissue:blood partition
coefficient Pt.b, and the arterial blood concentration Ca, as collectively defined under EquationApx H-8
below. The distribution of the chemical can be described by mass balance equations for tissue t as
described in Vernier et al. (2008) as

Equation Apx H-8

— = Qt (ca-—)•

dt vt\ a Pf.bJ
The arterial blood concentration is computed as

	 y QtCyt

ka Zit n 5
vc

with this sum being taken over all tissues. Here, Qc denotes the cardiac blood flow and Cvt denotes the
tissue venous blood concentration. The tissue:blood partition coefficients can be computed according to

Verner et al. (2008) by

= Kow ¦ Flt + Fwt
vb ~ Kow ¦ Flb + Fwb'

where Kow denotes the octanol-water partition coefficient of the chemical under consideration, Flt and
Fwt denote the time-varying percentages of lipid and water, respectively, in compartment t. Flb and
Fwb denote the percentages of lipid and water, respectively, in blood.

The mass balance equation for the liver compartment has a slightly different form, as it has an
absorption and metabolism term. It is given by Verner et al. (2008) as

dAi	/ Ci \

— = Intake + Qt j^Ca - — j - RAM

Page 494 of 572

-------
13293

13294

13295

13296

13297

13298

13299

13300

13301

13302

13303

13304

13305

13306

13307

13308

13309

13310

13311

13312

13313

13314

13315

13316

13317

13318

13319

13320

13321

13322

13323

13324

13325

13326

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

where Qi is the blood flow to the liver and RAM represents the metabolism in |ig/day. To compute this,
the volume of distribution is first calculated.

Vdage — Vblood Prp-.b ' ^rp Ppp-.b ' ^pp Pit: 6 ' Ki Pf-.b ' Pl-.b ' ^1 Pmam-.b ' ^mam Pbrain.b

Vbrain?

where VMood denotes the volume of blood in the mother, computed according to the Nadler equation
(Sharma and Sharma. 2023). This is used to compute additional parameters defined in (Verner et al.
2008). The clearance is

n (ln(2A vh

LLage ~ I ^ I v aage>

where HL denotes the half-life of the chemical in days. This is used to compute the quantity Ehaqe as

Ehnne =

age

CLage

age	Ql ,

which in turn is used to compute the intrinsic clearance value

CLintc=-- (EH2S£±\

VI \1 EhageJ

From here, the hepatic extraction is computed by

CLintr-Vl

Lj iX — 	

CLintc-Vl+Ql5

which is used to compute the metabolism rate measured in |ig/day.

RAM = Ql • Eh • Ca,

To solve this system of differential equations, organ volumes and blood flows are required for all time.
The system is solved numerically using the ODE function in the deSolve package in R. The output of
the model is a chemical amount and concentration in each organ compartment, as well as the milk
concentration for the entire time period of the simulation.

H.4.2 Milk Ingestion Rates by Age	

Milk ingestion rates by age are provided in Table 15-1 of the Exposure Factors Handbook (U.S. EPA.
201 la) and presented in Table Apx H-13.

Page 495 of 572

-------
13327

13328

13329

13330

13331

13332

13333

13334

13335

13336

13337

13338

13339

13340

13341

13342

13343

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table Apx H-13. M

ean and Upper Milk Ingestion Rates by Age

Age Group

Milk Ingestion (mL/kg day)

Mean

Upper (95th percentile)

Birth to <1 month

150

220

1 to <3 month

140

190

3 to <6 month

110

150

6 to <12 month

83

130

Birth to <1 year

104.8

152.5

H.4.3 Modeled Milk Concentrations

Three non-U.S. biomonitoring studies demonstrated the presence of TCEP in human milk. Two of the
studies measured lipid weight concentrations that ranged from non-detect to 512 ng/g (average 0.14-42
ng/g) in (Kim et al. 2014) and 2.1 to 8.2 ng/g (median 4.9 ng/g) in (Sundkvist et al. ). One study
by (He et al.. 2018a) measured wet weight concentrations from three milk samples collected in
Australia, and concentrations ranged from non-detect to 0.47 ng/mL (4,70/10 7mg/mL), Because the
Verner Model estimates wet weight concentrations, modeled concentrations can only be compared with
measured concentrations by (He et al.. 2018a). The range of the wet weight concentrations across each
COU/OES for each maternal group is presented in TableApx H-14. In general, the lower and upper
bound of the modeled concentrations are three magnitudes below and four magnitudes above measured
concentrations, respectively.

Table Apx H-14. Range of Modeled Milk Concentrations by
Maternal Group		

Maternal Group

Milk Concentrations
(mg/mL)

Consumer

3.96E-08 to 2.62E-04

Occupational

1.96E-10 to 1.13E-03

General population

1.83E-10 to 5.22E-04

H.4.4 Infant Exposure Estimates

Page 496 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table Apx H-15. Average Infant Doses via

luman Mil

i. Exposure from Maternal Consumer Use Scenarios

COU Subcategory and Consumer
Exposure Scenarios

Maternal Dose
(|ii»/k«-(lav)""

Milk Intake
Rate Type

Birth to <1 Month

(mg/kg-day)

1 to <3 Month

(mg/kg-day)

3 to <6 Month
(mg/kg-day)

6 to 12 Month
(m«/k«-day)

Birth to 12 Month
(mg/kg-day)

Fabric textile, leather products not
covered elsewhere (carpet back
coating)

6.08E00

Mean

1.01E-04

1.02E-04

9.16E-05

8.33E-05

9.00E-05

Fabric textile, leather products not
covered elsewhere (textile for
children's play structures)

8.94E01

Mean

1.48E-03

1.50E-03

1.35E-03

1.22E-03

1.32E-03

Building/construction materials not
covered elsewhere (roofing
insulation)

1.73E03

Mean

2.87E-02

2.91E-02

2.61E-02

2.37E-02

2.56E-02

Building/construction materials not
covered elsewhere (acoustic ceiling)

1.40E02

Mean

2.31E-03

2.35E-03

2.11E-03

1.92E-03

2.07E-03

Foam seating and bedding product
(foam automobile)

6.86E00

Mean

1.13E-04

1.15E-04

1.03E-04

9.40E-05

1.02E-04

Foam seating and bedding product
(foam living room)

1.53E01

Mean

2.53E-04

2.57E-04

2.30E-04

2.10E-04

2.26E-04

Foam seating and bedding product
(mattress)

7.54E00

Mean

1.25E-04

1.27E-04

1.14E-04

1.03E-04

1.12E-04

Foam seating and bedding product
(foam - other - toy block)

2.73E-01

Mean

4.52E-06

4.59E-06

4.11E-06

3.74E-06

4.04E-06

Building/construction materials -
wood and engineered wood products
(wood flooring)

1.80E03

Mean

2.97E-02

3.02E-02

2.71E-02

2.46E-02

2.66E-02

Building/construction materials -
wood and engineered wood products
(wooden tv stand)

1.03E02

Mean

1.70E-03

1.73E-03

1.55E-03

1.41E-03

1.53E-03

Fabric textile, leather products not
covered elsewhere (carpet back
coating)

6.08E00

Upper

1.47E-04

1.38E-04

1.25E-04

1.30E-04

1.31E-04

Fabric textile, leather products not
covered elsewhere (textile for
children's play structures)

8.94E01

Upper

2.16E-03

2.03E-03

1.83E-03

1.90E-03

1.93E-03

Building/construction materials not
covered elsewhere (roofing
insulation)

1.73E03

Upper

4.19E-02

3.94E-02

3.55E-02

3.69E-02

3.74E-02

Building/construction materials not
covered elsewhere (acoustic ceiling)

1.40E02

Upper

3.39E-03

3.18E-03

2.87E-03

2.98E-03

3.02E-03

Page 497 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

COU Subcategory and Consumer
Exposure Scenarios

Maternal Dose
(|ii«/k»-(lav)""

Milk Intake
Rate Type

Birth to <1 Month
(mg/kg-day)

1 to <3 Month

(mg/kg-day)

3 to <6 Month
(mg/kg-day)

6 to 12 Month

(mg/kg-day)

Birth to 12 Month

(mg/kg-day)

Foam seating and bedding product
(foam automobile)

6.86E00

Upper

1.66E-04

1.56E-04

1.41E-04

1.46E-04

1.48E-04

Foam seating and bedding product
(foam living room)

1.53E01

Upper

3.70E-04

3.48E-04

3.13E-04

3.26E-04

3.30E-04

Foam seating and bedding product
(mattress)

7.54E00

Upper

1.82E-04

1.72E-04

1.54E-04

1.61E-04

1.63E-04

Foam seating and bedding product
(foam - other - toy block)

2.73E-01

Upper

6.61E-06

6.21E-06

5.59E-06

5.82E-06

5.89E-06

Building/construction materials -
wood and engineered wood products
(wood flooring)

1.80E03

Upper

4.35E-02

4.09E-02

3.68E-02

3.83E-02

3.88E-02

Building/construction materials -
wood and engineered wood products
(wooden tv stand)

1.03E02

Upper

2.49E-03

2.35E-03

2.11E-03

2.20E-03

2.23E-03

" Consumer maternal doses were combined across oral, dermal, and inhalation routes. For inhalation, no extrapolation using Equation 5-22 was necessary because the
CEM already calculates a dose in mg/kg-day, as shown in Section 5.1.2.3 for consumers.

b Chronic maternal doses are the most relevant durations for building and construction materials, fabric and textile products, and foam seating and bedding products
because they are typically used over a longer time frame than other types of consumer products with direct applications (e.g., household cleaners, solvents).

13345

Page 498 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

13346 Table Apx H-16. Average Infant Doses from Maternal Workers Based on Mean Milk Intake Rate

OES

Route

Maternal
Exposure
Duration

Maternal Dose
(jig/kg-day)

Birth to <1
Month

(mg/kg-day)

1 to <3
Month
(mg/kg-day)

3 to <6
Month
(mg/kg-day)

6 to 12
Month

(mg/kg-day)

Birth to 12
Month

(mg/kg-day)

Import and repackaging

Dermal,

Inhalation

(High-end)

Chronic

1.57E02

2.59E-03

2.63E-03

2.36E-03

2.15E-03

2.32E-03

Incorporation into paints and coatings - 1-part
coatings

Chronic

8.38E02

1.39E-02

1.41E-02

1.26E-02

1.15E-02

1.24E-02

Incorporation into paints and coatings - 2-part
reactive coatings

Chronic

8.53E01

141E-03

1.43E-03

1.29E-03

1.17E-03

1.26E-03

Processing - formulation of TCEP into 2-part
reactive resins

Chronic

1.73E02

2.86E-03

2.90E-03

2.60E-03

2.37E-03

2.56E-03

Processing - processing into 2-part resin article

Chronic

2.18E03

3.60E-02

3.66E-02

3.28E-02

2.98E-02

3.22E-02

Processing - recycling electronics

Chronic

1.37E-01

2.26E-06

2.30E-06

2.06E-06

1.87E-06

2.03E-06

Commercial use - paints & coatings - spray (1-
part, 250-day application)

Chronic

1.45E03

240E-02

2.44E-02

2.18E-02

1.99E-02

2.14E-02

Commercial use - paints & coatings - spray (2-
part reactive, 250-day application)

Chronic

7.25E03

1.20E-01

1.22E-01

1.09E-01

9.93E-02

1.07E-01

Laboratory chemicals

Chronic

4.35E03

7.20E-02

7.32E-02

6.56E-02

5.96E-02

644E-02

Industrial/commercial use - installation of
aerospace products, chronic, inhalation

Inhalation
(High-end)

Chronic

1.35E-03

2.23E-08

2.27E-08

2.04E-08

1.85E-08

2.00E-08

Import and repackaging

Dermal,

Inhalation

(High-end)

Subchronic

1.86E03

3.07E-02

3.12E-02

2.80E-02

2.55E-02

2.75E-02

Incorporation into paints and coatings - 1-part
coatings

Subchronic

5.84E03

9.65E-02

9.81E-02

8.79E-02

8.00E-02

8.64E-02

Incorporation into paints and coatings - 2-part
reactive coatings

Subchronic

5.74E02

9.50E-03

9.65E-03

8.65E-03

7.87E-03

8.50E-03

Processing - formulation of TCEP into 2-part
reactive resins

Subchronic

1.63E03

2.70E-02

2.75E-02

2.46E-02

2.24E-02

242E-02

Processing - processing into 2-part resin article

Subchronic

2.33E03

3.86E-02

3.92E-02

3.51E-02

3.19E-02

345E-02

Processing - recycling electronics

Subchronic

1.47E-01

242E-06

2.46E-06

2.21E-06

2.01E-06

2.17E-06

Commercial use - paints & coatings - spray (1-
part, 250-day application)

Subchronic

1.55E03

2.57E-02

2.61E-02

2.34E-02

2.13E-02

2.30E-02

Commercial use - paints & coatings - spray (2-
part reactive, 250-day application)

Subchronic

7.76E03

1.28E-01

1.30E-01

1.17E-01

1.06E-01

1.15E-01

Laboratory chemicals

Subchronic

5.83E03

9.63E-02

9.79E-02

8.78E-02

7.98E-02

8.62E-02

Industrial/commercial use - installation of
Aerospace products, chronic, inhalation

Inhalation
(High-end)

Subchronic

1.45E-03

2.39E-08

2.43E-08

2.18E-08

1.98E-08

2.14E-08

Page 499 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table Apx H-17. Average Infant Doses from Maternal Workers Based on Upper Milk

OES

Route

Maternal
Exposure
Duration

Maternal Dose
(fi$j/kjj-day)

Birth to <1
Month

(mjj/kjj-day)

1 to <3
Month
(m«/k«-(lay)

3 to <6
Month
(mjj/kjj-day)

6 to 12
Month

(m^/kji-day)

Birth to 12
Month

(mjj/kj*-day)

Import and repackaging



Chronic

1.57E02

3.79E-03

3.56E-03

3.21E-03

3.34E-03

3.38E-03

Incorporation into paints and coatings - 1-part
coatings



Chronic

8.38E02

2.03E-02

1.91E-02

1.72E-02

1.79E-02

1.81E-02

Incorporation into paints and coatings - 2-part
reactive coatings



Chronic

8.53E01

2.06E-03

1.94E-03

1.75E-03

1.82E-03

1.84E-03

Processing - formulation of TCEP into 2-part
reactive resins

Dermal,

Chronic

1.73E02

4.18E-03

3.93E-03

3.54E-03

3.68E-03

3.73E-03

Processing - processing into 2-part resin article

Inhalation
(High-end)

Chronic

2.18E03

5.27E-02

4.96E-02

4.46E-02

4.64E-02

4.70E-02

Processing - recycling electronics

Chronic

1.73E-01

3.31E-06

3.11E-06

2.80E-06

2.92E-06

2.95E-06

Commercial use - paints & coatings - spray (1-
part, 250-day application)



Chronic

1.45E03

3.51E-02

3.30E-02

2.97E-02

3.09E-02

3.13E-02

Commercial use - paints & coatings - spray (2-
part reactive, 250-day application)



Chronic

7.25E03

1.75E-01

1.65E-01

1.48E-01

1.54E-01

1.56E-01

Laboratory chemicals



Chronic

4.35E03

1.05E-01

9.91E-02

8.92E-02

9.28E-02

9.40E-02

Industrial/commercial use - installation of
aerospace products, chronic, inhalation

Inhalation
(High-end)

Chronic

1.35E-03

3.27E-08

3.08E-08

2.77E-08

2.88E-08

2.92E-08

Import and repackaging



Subchronic

1.86E03

4.50E-02

4.23E-02

3.81E-02

3.96E-02

4.62E-02

Incorporation into paints and coatings - 1-part
coatings



Subchronic

5.84E03

1.41E-01

1.33E-01

1.20E-01

1.24E-01

1.45E-01

Incorporation into paints and coatings - 2-part
reactive coatings



Subchronic

5.74E02

1.39E-02

1.31E-02

1.18E-02

1.22E-02

1.43E-02

Processing - formulation of TCEP into 2-part
reactive resins

Dermal,

Subchronic

1.63E03

3.95E-02

3.72E-02

3.35E-02

3.48E-02

4.07E-02

Processing - processing into 2-part resin article

Inhalation
(High-end)

Subchronic

2.33E03

5.65E-02

5.31E-02

4.78E-02

4.97E-02

5.80E-02

Processing - recycling electronics

Subchronic

1.47E-01

3.55E-06

3.33E-06

3.00E-06

3.12E-06

3.65E-06

Commercial use - paints & coatings - spray (1-
part, 250-day application)



Subchronic

1.55E03

3.76E-02

3.53E-02

3.18E-02

3.31E-02

3.86E-02

Commercial use - paints & coatings - spray (2-
part reactive, 250-day application)



Subchronic

7.76E03

1.88E-01

1.77E-01

1.59E-01

1.65E-01

1.93E-01

Laboratory chemicals



Subchronic

5.83E03

1.41E-01

1.33E-01

1.19E-01

1.24E-01

1.45E-01

Industrial/commercial use - installation of
aerospace products, chronic, inhalation

Inhalation
(High-end)

Subchronic

1.45E-03

3.50E-08

3.29E-08

2.96E-08

3.08E-08

3.60E-08

ntake Rate

Page 500 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

TableApx H-18. Average Infant Doses via Human Milk Exposure from Maternal General Population Oral Exposures Based on
Mean Milk Intake Rate

COLs/OES

Route

Maternal Dose
(jig/kg-day)

Birth to <1
Month

(mg/kg-day)

1 to <3
Month

(mg/kg-day)

3 to <6
Month
(mg/kg-day)

6 to 12
Month

(mg/kg-day)

Birth to 12
Month

(mg/kg-day)

Import and repackaging

Gen Pop Fish Ingestion,
High B AF

6.37E02

1.05E-02

1.07E-02

9.60E-03

8.73E-03

9.43E-03

Incorporation into paints and coatings - 1-part
coatings

Gen Pop Fish Ingestion,
High B AF

2.82E03

4.66E-02

4.74E-02

4.25E-02

3.86E-02

4.17E-02

Incorporation into paints and coatings - 2-part
reactive coatings

Gen Pop Fish Ingestion,
High B AF

2.56E03

4.23E-02

4.30E-02

3.86E-02

3.51E-02

3.79E-02

Use in paints and coatings at job sites

Gen Pop Fish Ingestion,
High B AF

1.50E03

2.48E-02

2.52E-02

2.26E-02

2.05E-02

2.22E-02

Formulation of TCEP containing reactive resin

Gen Pop Fish Ingestion,
High B AF

3.58E03

5.92E-02

6.02E-02

5.39E-02

4.90E-02

5.30E-02

Laboratory chemicals

Gen Pop Fish Ingestion,
High B AF

2.55E01

4.22E-04

4.29E-04

3.84E-04

3.49E-04

3.77E-04

Import and repackaging

Gen Pop Fish Ingestion,
Low BAF

3.16E01

5.23E-04

5.31E-04

4.76E-04

4.33E-04

4.68E-04

Incorporation into paints and coatings - 1-part
coatings

Gen Pop Fish Ingestion,
Low BAF

1.40E02

2.32E-03

2.35E-03

2.11E-03

1.92E-03

2.07E-03

Incorporation into paints and coatings - 2-part
reactive coatings

Gen Pop Fish Ingestion,
Low BAF

1.27E02

2.10E-03

2.13E-03

1.91E-03

1.74E-03

1.88E-03

Use in paints and coatings at job sites

Gen Pop Fish Ingestion,
Low BAF

7.44E01

1.23E-03

1.25E-03

1.12E-03

1.02E-03

1.10E-03

Formulation of TCEP containing reactive resin

Gen Pop Fish Ingestion,
Low BAF

1.78E02

2.94E-03

2.99E-03

2.68E-03

2.44E-03

2.63E-03

Laboratory chemicals

Gen Pop Fish Ingestion,
Low BAF

1.27E00

2.10E-05

2.13E-05

1.91E-05

1.74E-05

1.88E-05

Import and repackaging

Undiluted Drinking Water

3.16E-02

5.23E-07

5.31E-07

4.76E-07

4.33E-07

4.68E-07

Incorporation into paints and coatings - 1-part
coatings

Undiluted Drinking Water

1.40E-01

2.31E-06

2.35E-06

2.11E-06

1.92E-06

2.07E-06

Incorporation into paints and coatings - 2-part
reactive coatings

Undiluted Drinking Water

1.26E-01

2.08E-06

2.12E-06

1.90E-06

1.73E-06

1.86E-06

Use in paints and coatings at job sites

Undiluted Drinking Water

7.42E-02

1.23E-06

1.25E-06

1.12E-06

1.02E-06

1.10E-06

Formulation of TCEP containing reactive resin

Undiluted Drinking Water

1.77E-01

2.93E-06

2.97E-06

2.67E-06

2.42E-06

2.62E-06

Laboratory chemicals

Undiluted Drinking Water

1.26E-03

2.08E-08

2.12E-08

1.90E-08

1.73E-08

1.86E-08

Page 501 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

13350	TableApx H-19. Average Infant Doses via Human Milk Exposure from Maternal General Population Oral Exposures Based on

13351	Upper Milk Intake Rate								

COUs/OES

Route

Maternal Dose
(fig/kg-day)

Birth to <1
Month

(mg/kg-day)

1 to <3
Month
(mg/kg-day)

3 to <6
Month
(mg/kg-day)

6 to 12
Month

(mg/kg-day)

Birth to 12
Month

(mg/kg-day)

Import and repackaging

Gen Pop Fish Ingestion,
High B AF

6.37E02

1.54E-02

1.45E-02

1.30E-02

1.36E-02

1.38E-02

Incorporation into paints and coatings - 1-part
coatings

Gen Pop Fish Ingestion,
High B AF

2.82E03

6.83E-02

6.42E-02

5.78E-02

6.01E-02

6.09E-02

Incorporation into paints and coatings - 2-part
reactive coatings

Gen Pop Fish Ingestion,
High B AF

2.56E03

6.20E-02

5.83E-02

5.24E-02

5.45E-02

5.53E-02

Use in paints and coatings at job sites

Gen Pop Fish Ingestion,
High B AF

1.50E03

3.63E-02

3.41E-02

3.07E-02

3.20E-02

3.24E-02

Formulation of TCEP containing reactive resin

Gen Pop Fish Ingestion,
High B AF

3.58E03

8.66E-02

8.15E-02

7.33E-02

7.63E-02

7.73E-02

Laboratory chemicals

Gen Pop Fish Ingestion,
High B AF

2.55E01

6.17E-04

5.80E-04

5.22E-04

5.43E-04

5.51E-04

Import and repackaging

Gen Pop Fish Ingestion,
Low BAF

3.16E01

7.65E-04

7.19E-04

6.47E-04

6.73E-04

6.82E-04

Incorporation into paints and coatings - 1-part
coatings

Gen Pop Fish Ingestion,
Low BAF

1.40E02

3.39E-03

3.19E-03

2.87E-03

2.98E-03

3.02E-03

Incorporation into paints and coatings - 2-part
reactive coatings

Gen Pop Fish Ingestion,
Low BAF

1.27E02

3.07E-03

2.89E-03

2.60E-03

2.71E-03

2.74E-03

Use in paints and coatings at job sites

Gen Pop Fish Ingestion,
Low BAF

7.44E01

1.80E-03

1.69E-03

1.52E-03

1.59E-03

1.61E-03

Formulation of TCEP containing reactive resin

Gen Pop Fish Ingestion,
Low BAF

1.78E02

4.31E-03

4.05E-03

3.65E-03

3.79E-03

3.84E-03

Laboratory chemicals

Gen Pop Fish Ingestion,
Low BAF

1.27E00

3.07E-05

2.89E-05

2.60E-05

2.71E-05

2.74E-05

Import and repackaging

Undiluted Drinking Water

3.16E-02

7.65E-07

7.19E-07

6.47E-07

6.73E-07

6.82E-07

Incorporation into paints and coatings -1-part
coatings

Undiluted Drinking Water

1.40E-01

3.39E-06

3.19E-06

2.87E-06

2.98E-06

3.02E-06

Incorporation into paints and coatings - 2-part
reactive coatings

Undiluted Drinking Water

1.26E-01

3.05E-06

2.87E-06

2.58E-06

2.68E-06

2.72E-06

Use in paints and coatings at job sites

Undiluted Drinking Water

7.42E-02

1.80E-06

1.69E-06

1.52E-06

1.58E-06

1.60E-06

Formulation of TCEP containing reactive resin

Undiluted Drinking Water

1.77E-01

4.28E-06

4.03E-06

3.63E-06

3.77E-06

3.82E-06

Laboratory chemicals

Undiluted Drinking Water

1.26E-03

3.05E-08

2.87E-08

2.58E-08

2.68E-08

2.72E-08

Page 502 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

13352	TableApx H-20. Average Infant Doses via Human Milk Exposure from Maternal Tribal Fish Ingestion Based on Mean Milk Intake

13353	Rate

COUs/OES

Route

Maternal Dose
(fig/kg-day)

Birth to <1
Month

(mg/kg-day)

1 to <3
Month
(mg/kg-day)

3 to <6
Month
(mg/kg-day)

6 to 12
Month

(mg/kg-day)

Birth to 12
Month

(mg/kg-day)

Import and repackaging

Current IR, High B AF

6.21E03

1.03E-01

1.04E-01

9.36E-02

8.51E-02

9.19E-02

Incorporation into paints and coatings - 1-part
coatings

Current IR, High B AF

2.75E04

4.55E-01

4.62E-01

4.14E-01

3.77E-01

4.07E-01

Incorporation into paints and coatings - 2-part
reactive coatings

Current IR, High B AF

1.44E05

2.38E00

2.42E00

2.17E00

1.97E00

2.13E00

Use in paints and coatings at job sites

Current IR, High B AF

8.42E04

1.39E00

1.42E00

1.27E00

1.15E00

1.25E00

Formulation of TCEP containing reactive resin

Current IR, High B AF

2.01E05

3.32E00

3.38E00

3.03E00

2.75E00

2.97E00

Laboratory chemicals

Current IR, High B AF

1.43E03

2.36E-02

2.40E-02

2.15E-02

1.96E-02

2.12E-02

Import and repackaging

Current IR, High B AF

3.08E02

5.09E-03

5.18E-03

4.64E-03

4.22E-03

4.56E-03

Incorporation into paints and coatings - 1-part
coatings

Current IR, High B AF

1.36E03

2.25E-02

2.29E-02

2.05E-02

1.86E-02

2.01E-02

Incorporation into paints and coatings - 2-part
reactive coatings

Current IR, High B AF

1.24E03

2.05E-02

2.08E-02

1.87E-02

1.70E-02

1.83E-02

Use in paints and coatings at job sites

Current IR, High B AF

7.25E02

1.20E-02

1.22E-02

1.09E-02

9.93E-03

1.07E-02

Formulation of TCEP containing reactive resin

Current IR, High B AF

1.73E03

2.86E-02

2.91E-02

2.61E-02

2.37E-02

2.56E-02

Laboratory chemicals

Current IR, High BAF

1.23E01

2.03E-04

2.07E-04

1.85E-04

1.68E-04

1.82E-04

Import and repackaging

Heritage IR, High BAF

3.58E04

5.92E-01

6.02E-01

5.39E-01

4.90E-01

5.30E-01

Incorporation into paints and coatings -1-part
coatings

Heritage IR, High BAF

1.58E05

2.61E00

2.66E00

2.38E00

2.16E00

2.34E00

Incorporation into paints and coatings - 2-part
reactive coatings

Heritage IR, High BAF

1.44E05

2.38E00

2.42E00

2.17E00

1.97E00

2.13E00

Use in paints and coatings at job sites

Heritage IR, High BAF

8.42E04

1.39E00

1.42E00

1.27E00

1.15E00

1.25E00

Formulation of TCEP containing reactive resin

Heritage IR, High BAF

2.01E05

3.32E00

3.38E00

3.03E00

2.75E00

2.97E00

Laboratory chemicals

Heritage IR, High BAF

1.43E03

2.36E-02

2.40E-02

2.15E-02

1.96E-02

2.12E-02

Import and repackaging

Heritage IR, Low BAF

1.77E03

2.93E-02

2.97E-02

2.67E-02

2.42E-02

2.62E-02

Incorporation into paints and coatings -1-part
coatings

Heritage IR, Low BAF

7.86E03

1.30E-01

1.32E-01

1.18E-01

1.08E-01

1.16E-01

Incorporation into paints and coatings - 2-part
reactive coatings

Heritage IR, Low BAF

7.13E03

1.18E-01

1.20E-01

1.07E-01

9.77E-02

1.05E-01

Page 503 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

COUs/OES

Route

Maternal Dose
(jig/kg-day)

Birth to <1
Month

(mg/kg-day)

1 to <3
Month

(mg/kg-day)

3 to <6
Month
(mg/kg-day)

6 to 12
Month

(mg/kg-day)

Birth to 12
Month

(mg/kg-day)

Use in paints and coatings at job sites

Heritage IR, Low BAF

4.18E03

6.91E-02

7.02E-02

6.30E-02

5.73E-02

6.18E-02

Formulation of TCEP containing reactive resin

Heritage IR, Low BAF

9.97E03

1.65E-01

1.68E-01

1.50E-01

1.37E-01

148E-01

Laboratory chemicals

Heritage IR, Low BAF

7.11E01

1.18E-03

1.19E-03

1.07E-03

9.74E-04

1.05E-03

13354

13355

13356

13357

TableApx H-21. Average Infant Doses via Human Milk Exposure from Maternal Tribal Fish Ingestion Based on Upper Milk Intake
Rate

COUs/OES

Route

Maternal Dose
(fig/kg-day)

Birth to <1
Month

(mg/kg-day)

1 to <3
Month
(mg/kg-day)

3 to <6
Month
(mg/kg-day)

6 to 12
Month

(mg/kg-day)

Birth to 12
Month

(mg/kg-day)

Import and repackaging

Current IR, High BAF

6.21E03

1.50E-01

1.41E-01

1.27E-01

1.32E-01

1.34E-01

Incorporation into paints and coatings - 1-part
coatings

Current IR, High BAF

2.75E04

6.66E-01

6.26E-01

5.63E-01

5.86E-01

5.94E-01

Incorporation into paints and coatings - 2-part
reactive coatings

Current IR, High BAF

144E05

349E00

3.28E00

2.95E00

3.07E00

3.11E00

Use in paints and coatings at job sites

Current IR, High BAF

842E04

2.04E00

1.92E00

1.72E00

1.79E00

1.82E00

Formulation of TCEP containing reactive resin

Current IR, High BAF

2.01E05

4.86E00

4.57E00

4.12E00

4.28E00

4.34E00

Laboratory chemicals

Current IR, High BAF

143E03

346E-02

3.25E-02

2.93E-02

3.05E-02

3.09E-02

Import and repackaging

Current IR, High BAF

3.08E02

745E-03

7.01E-03

6.31E-03

6.56E-03

6.65E-03

Incorporation into paints and coatings - 1-part
coatings

Current IR, High BAF

1.36E03

3.29E-02

3.10E-02

2.79E-02

2.90E-02

2.94E-02

Incorporation into paints and coatings - 2-part
reactive coatings

Current IR, High BAF

1.24E03

3.00E-02

2.82E-02

2.54E-02

2.64E-02

2.68E-02

Use in paints and coatings at job sites

Current IR, High BAF

7.25E02

1.75E-02

1.65E-02

1.48E-02

1.54E-02

1.57E-02

Formulation of TCEP containing reactive resin

Current IR, High BAF

1.73E03

4.19E-02

3.94E-02

3.54E-02

3.69E-02

3.73E-02

Laboratory chemicals

Current IR, High BAF

1.23E01

2.98E-04

2.80E-04

2.52E-04

2.62E-04

2.66E-04

Import and repackaging

Heritage IR, High BAF

3.58E04

8.66E-01

8.15E-01

7.33E-01

7.63E-01

7.73E-01

Incorporation into paints and coatings -1-part
coatings

Heritage IR, High BAF

1.58E05

3.82E00

3.60E00

3.24E00

3.37E00

341E00

Incorporation into paints and coatings - 2-part
reactive coatings

Heritage IR, High BAF

144E05

349E00

3.28E00

2.95E00

3.07E00

3.11E00

Page 504 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

COUs/OES

Route

Maternal Dose
(jig/kg-day)

Birth to <1
Month

(mg/kg-day)

1 to <3
Month

(mg/kg-day)

3 to <6
Month
(mg/kg-day)

6 to 12
Month

(mg/kg-day)

Birth to 12
Month

(mg/kg-day)

Use in paints and coatings at job sites

Heritage IR, High B AF

8.42E04

2.04E00

1.92E00

1.72E00

1.79E00

1.82E00

Formulation of TCEP containing reactive resin

Heritage IR, High B AF

2.01E05

4.86E00

4.57E00

4.12E00

4.28E00

4.34E00

Laboratory chemicals

Heritage IR, High B AF

1.43E03

3.46E-02

3.25E-02

2.93E-02

3.05E-02

3.09E-02

Import and repackaging

Heritage IR, Low BAF

1.77E03

4.28E-02

4.03E-02

3.63E-02

3.77E-02

3.82E-02

Incorporation into paints and coatings -1-part
coatings

Heritage IR, Low BAF

7.86E03

1.90E-01

1.79E-01

1.61E-01

1.67E-01

1.70E-01

Incorporation into paints and coatings - 2-part
reactive coatings

Heritage IR, Low BAF

7.13E03

1.73E-01

1.62E-01

1.46E-01

1.52E-01

1.54E-01

Use in paints and coatings at job sites

Heritage IR, Low BAF

4.18E03

1.01E-01

9.51E-02

8.56E-02

8.91E-02

9.02E-02

Formulation of TCEP containing reactive resin

Heritage IR, Low BAF

9.97E03

2.41E-01

2.27E-01

2.04E-01

2.12E-01

2.15E-01

Laboratory chemicals

Heritage IR, Low BAF

7.11E01

1.72E-03

1.62E-03

1.46E-03

1.51E-03

1.53E-03

13358	H.4.5 Infant Risk Estimates

13359

13360	Table Apx H-22. Infant Risks via Human Milk Exposure from Maternal Consumer Use Scenarios

COU Subcategory and Consumer Exposure Scenarios

Milk Intake
Rate Type

Short-Term

Chronic

Cancer

Fabric textile, leather products not covered elsewhere (carpet back coating)

Mean

2.71E04

3.03E04

2.83E-08

Fabric textile, leather products not covered elsewhere (textile for children's play structures)

Mean

1.85E03

2.06E03

4.15E-07

Building/construction materials not covered elsewhere (roofing insulation)

Mean

9.53E01

1.06E02

8.05E-06

Building/construction materials not covered elsewhere (acoustic ceiling)

Mean

1.18E03

1.32E03

6.50E-07

Foam seating and bedding product (foam automobile)

Mean

2.41E04

2.69E04

3.19E-08

Foam seating and bedding product (foam living room)

Mean

1.08E04

1.21E04

7.11E-08

Foam seating and bedding product (mattress)

Mean

2.19E04

2.45E04

3.50E-08

Foam seating and bedding product (foam - other - toy block)

Mean

6.05E05

6.76E05

1.27E-09

Building/construction materials - wood and engineered wood products (wood flooring)

Mean

9.19E01

1.03E02

8.35E-06

Building/construction materials - wood and engineered wood products (wooden TV stand)

Mean

1.60E03

1.79E03

4.79E-07

Fabric textile, leather products not covered elsewhere (carpet back coating)

Upper

1.85E04

2.08E04

4.12E-08

Page 505 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

COU Subcategory and Consumer Exposure Scenarios

Milk Intake
Rate Type

Short-Term

Chronic

Cancer

Fabric textile, leather products not covered elsewhere (textile for children's play structures)

Upper

1.26E03

1.42E03

6.06E-07

Building/construction materials not covered elsewhere (roofing insulation)

Upper

6.51E01

7.30E01

1.18E-05

Building/construction materials not covered elsewhere (acoustic ceiling)

Upper

8.06E02

9.04E02

9.49E-07

Foam seating and bedding product (foam automobile)

Upper

1.64E04

1.84E04

4.65E-08

Foam seating and bedding product (foam living room)

Upper

7.37E03

8.27E03

1.04E-07

Foam seating and bedding product (mattress)

Upper

1.50E04

1.68E04

5.11E-08

Foam seating and bedding product (foam - other - toy block)

Upper

4.13E05

4.63E05

1.85E-09

Building/construction materials - wood and engineered wood products (wood flooring)

Upper

6.28E01

7.04E01

1.22E-05

Building/construction materials - wood and engineered wood products (wooden TV stand)

Upper

1.09E03

1.23E03

6.99E-07

13361

13362

13363	TableApx H-23. Infant Risks via Human Milk Exposure from Maternal Occupational Use Scenarios Based on Mean Milk Intake

13364	Rate

OES

Route

Maternal Exposure
Duration

Short-Term

Chronic

Cancer

Import and repackaging

Dermal,

Inhalation

(High-end)

Chronic

1.05E03

1.18E03

7.28E-07

Incorporation into paints and coatings - 1-part coatings

Chronic

1.97E02

2.20E02

3.89E-06

Incorporation into paints and coatings - 2-part reactive coatings

Chronic

1.94E03

2.16E03

3.97E-07

Processing - formulation of TCEP into 2-part reactive resins

Chronic

9.56E02

1.07E03

8.03E-07

Processing - processing into 2-part resin article

Chronic

7.58E01

8.47E01

1.01E-05

Processing - Recycling Electronics

Chronic

1.21E06

1.35E06

6.36E-10

Commercial use - paints & coatings - spray (1-part, 250-day
application)

Chronic

1.14E02

1.27E02

6.74E-06

Commercial use - paints & coatings - spray (2-part reactive, 250-
day application)

Chronic

2.28E01

2.55E01

3.37E-05

Laboratory chemicals

Chronic

3.79E01

4.24E01

2.02E-05

Industrial/commercial use - installation of aerospace products

Inhalation
(High-end)

Chronic

1.22E08

1.37E08

6.28E-12

Import and repackaging



Subchronic

8.88E01

9.93E01

8.64E-06

Page 506 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

OES

Route

Maternal Exposure
Duration

Short-Term

Chronic

Cancer

Incorporation into paints and coatings - 1-part coatings

Dermal,

Inhalation

(High-end)

Subchronic

2.83E01

3.16E01

2.71E-05

Incorporation into paints and coatings - 2-part reactive coatings

Subchronic

2.87E02

3.21E02

2.67E-06

Processing - formulation of TCEP into 2-part reactive resins

Subchronic

1.01E02

1.13E02

7.59E-06

Processing - processing into 2-part resin article

Subchronic

7.08E01

7.91E01

1.08E-05

Processing - recycling electronics

Subchronic

1.13E06

1.26E06

6.81E-10

Commercial use - paints & coatings - spray (1-part, 250-day
application)

Subchronic

1.06E02

1.19E02

7.21E-06

Commercial use - paints & coatings - spray (2-part reactive, 250-
day application)

Subchronic

2.13E01

2.38E01

3.61E-05

Laboratory chemicals

Subchronic

2.83E01

3.17E01

2.71E-05

Industrial/commercial use - installation of aerospace products

Inhalation
(High-end)

Subchronic

1.14E08

1.28E08

6.72E-12

13365

13366

13367	TableApx H-24. Infant Risks via Human Milk Exposure from Maternal Occupational Use Scenarios Based on Upper Milk Intake

13368	Rate

OES

Route

Maternal Exposure
Duration

Short-Term

Chronic

Cancer

Import and repackaging

Dermal,

Inhalation

(High-end)

Chronic

7.20E02

8.08E02

1.06E-06

Incorporation into paints and coatings - 1-part coatings

Chronic

1.35E02

1.51E02

5.68E-06

Incorporation into paints and coatings - 2-part reactive coatings

Chronic

1.32E03

1.48E03

5.78E-07

Processing - formulation of TCEP into 2-part reactive resins

Chronic

6.53E02

7.32E02

1.17E-06

Processing - processing into 2-part resin article

Chronic

5.18E01

5.80E01

1.48E-05

Processing - recycling electronics

Chronic

8.24E05

9.24E05

9.28E-10

Commercial use - paints & coatings - spray (1-part, 250-day
application)

Chronic

7.78E01

8.73E01

9.83E-06

Commercial use - paints & coatings - spray (2-part reactive, 250-day
application)

Chronic

1.56E01

1.75E01

4.91E-05

Laboratory chemicals

Chronic

2.59E01

2.90E01

2.95E-05

Page 507 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

OES

Route

Maternal Exposure
Duration

Short-Term

Chronic

Cancer

Industrial/commercial use - installation of aerospace products

Inhalation
(High-end)

Chronic

8.35E07

9.35E07

9.17E-12

Import and repackaging

Dermal,

Inhalation

(High-end)

Subchronic

6.07E01

5.90E01

1.45E-05

Incorporation into paints and coatings - 1-part coatings

Subchronic

1.93E01

1.88E01

4.56E-05

Incorporation into paints and coatings - 2-part reactive coatings

Subchronic

1.96E02

1.91E02

4.49E-06

Processing - formulation of TCEP into 2-part reactive resins

Subchronic

6.90E01

6.71E01

1.28E-05

Processing - processing into 2-part resin article

Subchronic

4.84E01

4.70E01

1.82E-05

Processing - recycling electronics

Subchronic

7.70E05

7.49E05

1.15E-09

Commercial use - paints & coatings - spray (1-part, 250-day
application)

Subchronic

7.27E01

7.07E01

1.21E-05

Commercial use - paints & coatings - spray (2-part reactive, 250-day
application)

Subchronic

1.45E01

1.41E01

6.06E-05

Laboratory chemicals

Subchronic

1.94E01

1.88E01

4.55E-05

Industrial/commercial use - installation of aerospace products

Inhalation
(High-end)

Subchronic

7.80E07

7.58E07

1.13E-11

13369

13370

13371	TableApx H-25. Infant Risks via Human Milk Exposure from Maternal General Population Oral Exposures Based on Mean Milk

13372	Intake Rate

COUs/OESs

Route

Short-Term

Chronic

Cancer

Import and Repackaging

Gen Pop Fish Ingestion, High BAF

2.59E02

2.90E02

2.96E-06

Incorporation into paints and coatings - 1-part coatings

Gen Pop Fish Ingestion, High BAF

5.85E01

6.54E01

1.31E-05

Incorporation into paints and coatings - 2-part reactive coatings

Gen Pop Fish Ingestion, High BAF

6.45E01

7.21E01

1.19E-05

Use in paints and coatings at job sites

Gen Pop Fish Ingestion, High BAF

1.10E02

1.23E02

6.97E-06

Formulation of TCEP containing reactive resin

Gen Pop Fish Ingestion, High BAF

4.61E01

5.15E01

1.66E-05

Laboratory chemicals

Gen Pop Fish Ingestion, High BAF

6.47E03

7.24E03

1.19E-07

Import and Repackaging

Gen Pop Fish Ingestion, Low BAF

5.22E03

5.84E03

1.47E-07

Incorporation into paints and coatings - 1-part coatings

Gen Pop Fish Ingestion, Low BAF

1.18E03

1.32E03

6.51E-07

Page 508 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

COUs/OESs

Route

Short-Term

Chronic

Cancer

Incorporation into paints and coatings - 2-part reactive coatings

Gen Pop Fish Ingestion, Low BAF

1.30E03

1.45E03

5.90E-07

Use in paints and coatings at job sites

Gen Pop Fish Ingestion, Low BAF

2.22E03

2.48E03

3.46E-07

Formulation of TCEP containing reactive resin

Gen Pop Fish Ingestion, Low BAF

9.27E02

1.04E03

8.27E-07

Laboratory chemicals

Gen Pop Fish Ingestion, Low BAF

1.30E05

1.45E05

5.90E-09

Import and Repackaging

Undiluted Drinking Water

5.22E06

5.84E06

1.47E-10

Incorporation into paints and coatings - 1-part coatings

Undiluted Drinking Water

1.18E06

1.32E06

6.51E-10

Incorporation into paints and coatings - 2-part reactive coatings

Undiluted Drinking Water

1.31E06

1.46E06

5.86E-10

Use in paints and coatings at job sites

Undiluted Drinking Water

2.23E06

2.49E06

3.45E-10

Formulation of TCEP containing reactive resin

Undiluted Drinking Water

9.33E05

1.04E06

8.23E-10

Laboratory chemicals

Undiluted Drinking Water

1.31E08

1.46E08

5.86E-12

13373

13374

13375

13376

TableApx H-26. Infant Risks via Human Milk Exposure from Maternal General Population Oral Exposures Based on Upper Milk
Intake Rate

COUs/OESs

Route

Short-Term

Chronic

Cancer

Import and Repackaging

Gen Pop Fish Ingestion, High BAF

1.77E02

1.99E02

4.32E-06

Incorporation into paints and coatings - 1-part coatings

Gen Pop Fish Ingestion, High BAF

4.00E01

4.48E01

1.91E-05

Incorporation into paints and coatings - 2-part reactive coatings

Gen Pop Fish Ingestion, High BAF

4.41E01

4.94E01

1.74E-05

Use in paints and coatings at job sites

Gen Pop Fish Ingestion, High BAF

7.52E01

8.43E01

1.02E-05

Formulation of TCEP containing reactive resin

Gen Pop Fish Ingestion, High BAF

3.15E01

3.53E01

2.43E-05

Laboratory chemicals

Gen Pop Fish Ingestion, High BAF

4.42E03

4.96E03

1.73E-07

Import and Repackaging

Gen Pop Fish Ingestion, Low BAF

3.57E03

4.00E03

2.14E-07

Incorporation into paints and coatings - 1-part coatings

Gen Pop Fish Ingestion, Low BAF

8.06E02

9.03E02

9.49E-07

Incorporation into paints and coatings - 2-part reactive coatings

Gen Pop Fish Ingestion, Low BAF

8.88E02

9.96E02

8.61E-07

Use in paints and coatings at job sites

Gen Pop Fish Ingestion, Low BAF

1.52E03

1.70E03

5.05E-07

Formulation of TCEP containing reactive resin

Gen Pop Fish Ingestion, Low BAF

6.34E02

7.10E02

1.21E-06

Laboratory chemicals

Gen Pop Fish Ingestion, Low BAF

8.88E04

9.96E04

8.61E-09

Page 509 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

COUs/OESs

Route

Short-Term

Chronic

Cancer

Import and Repackaging

Undiluted Drinking Water

3.57E06

4.00E06

2.14E-10

Incorporation into paints and coatings - 1-part coatings

Undiluted Drinking Water

8.06E05

9.03E05

9.49E-10

Incorporation into paints and coatings - 2-part reactive coatings

Undiluted Drinking Water

8.95E05

1.00E06

8.54E-10

Use in paints and coatings at job sites

Undiluted Drinking Water

1.52E06

1.70E06

5.03E-10

Formulation of TCEP containing reactive resin

Undiluted Drinking Water

6.37E05

7.14E05

1.20E-09

Laboratory chemicals

Undiluted Drinking Water

8.95E07

1.00E08

8.54E-12

13377

13378

13379	Table Apx H-27. Infant Risks via Human Milk Exposure from Tribal Maternal Fish Exposures Based on Mean Milk Intake Rate

COUs/OESs

Route

Short-term

Chronic

Acute based on
Short-term
Dose

Acute based
on Chronic

Dose

Cancer

Import and Repackaging

Current IR, High BAF

2.66E01

2.97E01

9.21E01

1.03E02

2.89E-05

Incorporation into paints and coatings - 1-part coatings

Current IR, High BAF

6.00E00

6.71E00

2.08E01

2.32E01

1.28E-04

Incorporation into paints and coatings - 2-part reactive coatings

Current IR, High BAF

1.15E00

1.28E00

3.97E00

4.44E00

6.69E-04

Use in paints and coatings at job sites

Current IR, High BAF

1.96E00

2.19E00

6.79E00

7.59E00

3.91E-04

Formulation of TCEP containing reactive resin

Current IR, High BAF

8.21E-01

9.18E-01

2.85E00

3.18E00

9.34E-04

Laboratory chemicals

Current IR, High BAF

1.15E02

1.29E02

NA

NA

6.65E-06

Import and Repackaging

Current IR, Low BAF

5.36E02

5.99E02

NA

NA

1.43E-06

Incorporation into paints and coatings - 1-part coatings

Current IR, Low BAF

1.21E02

1.36E02

NA

NA

6.32E-06

Incorporation into paints and coatings - 2-part reactive coatings

Current IR, Low BAF

1.33E02

1.49E02

NA

NA

5.76E-06

Use in paints and coatings at job sites

Current IR, Low BAF

2.28E02

2.54E02

NA

NA

3.37E-06

Formulation of TCEP containing reactive resin

Current IR, Low BAF

9.54E01

1.07E02

NA

NA

8.04E-06

Laboratory chemicals

Current IR, Low BAF

1.34E04

1.50E04

NA

NA

5.72E-08

Import and Repackaging

Heritage IR, High BAF

4.61E00

5.15E00

1.60E01

1.79E01

1.66E-04

Incorporation into paints and coatings - 1-part coatings

Heritage IR, High BAF

1.04E00

1.17E00

3.62E00

4.05E00

7.34E-04

Page 510 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

COUs/OESs

Route

Short-term

Chronie

Aeute based on
Short-term
Dose

Aeute based
on Chronie

Dose

Caneer

Incorporation into paints and coatings - 2-part reactive coatings

Heritage IR, High BAF

1.15E00

1.28E00

3.97E00

4.44E00

6.69E-04

Use in paints and coatings at job sites

Heritage IR, High BAF

1.96E00

2.19E00

6.79E00

7.59E00

3.91E-04

Formulation of TCEP containing reactive resin

Heritage IR, High BAF

8.21E-01

9.18E-01

2.85E00

3.18E00

9.34E-04

Laboratory chemicals

Heritage IR, High BAF

1.15E02

1.29E02

4.00E02

4.47E02

6.65E-06

Import and Repackaging

Heritage IR, Low BAF

9.33E01

1.04E02

NA

NA

8.23E-06

Incorporation into paints and coatings - 1-part coatings

Heritage IR, Low BAF

2.10E01

2.35E01

7.28E01

8.13E01

3.65E-05

Incorporation into paints and coatings - 2-part reactive coatings

Heritage IR, Low BAF

2.32E01

2.59E01

8.02E01

8.97E01

3.31E-05

Use in paints and coatings at job sites

Heritage IR, Low BAF

3.95E01

4.41E01

NA

NA

1.94E-05

Formulation of TCEP containing reactive resin

Heritage IR, Low BAF

1.66E01

1.85E01

5.74E01

6.41E01

4.63E-05

Laboratory chemicals

Heritage IR, Low BAF

2.32E03

2.60E03

NA

NA

3.30E-07

13380

13381

13382	Table Apx H-28. Infant Risks via Human Milk Exposure from Tribal Maternal Fish Exposures Based on Upper Milk Intake Rate

COUs/OESs

Route

Short-term

Chronie

Aeute based on
Short-term
Dose

Aeute based
on Chronie

Dose

Caneer

Import and Repackaging

Current IR, High BAF

1.82E01

2.04E01

6.29E01

7.06E01

4.21E-05

Incorporation into paints and coatings - 1-part coatings

Current IR, High BAF

4.10E00

4.60E00

1.42E01

1.59E01

1.86E-04

Incorporation into paints and coatings - 2-part reactive coatings

Current IR, High BAF

7.83E-01

8.78E-01

2.71E00

3.04E00

9.76E-04

Use in paints and coatings at job sites

Current IR, High BAF

1.34E00

1.50E00

4.64E00

5.20E00

5.71E-04

Formulation of TCEP containing reactive resin

Current IR, High BAF

5.61E-01

6.29E-01

1.94E00

2.18E00

1.36E-03

Laboratory chemicals

Current IR, High BAF

7.89E01

8.84E01

NA

NA

9.70E-06

Import and Repackaging

Current IR, Low BAF

3.66E02

4.11E02

NA

NA

2.09E-06

Incorporation into paints and coatings - 1-part coatings

Current IR, Low BAF

8.29E01

9.30E01

NA

NA

9.22E-06

Page 511 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

COUs/OESs

Route

Short-term

Chronie

Aeute based on
Short-term
Dose

Aeute based
on Chronie

Dose

Caneer

Incorporation into paints and coatings - 2-part reactive coatings

Current IR, Low BAF

9.10E01

1.02E02

NA

NA

8.41E-06

Use in paints and coatings at job sites

Current IR, Low BAF

1.56E02

1.74E02

NA

NA

4.92E-06

Formulation of TCEP containing reactive resin

Current IR, Low BAF

6.52E01

7.31E01

NA

NA

1.17E-05

Laboratory chemicals

Current IR, Low BAF

9.17E03

1.03E04

NA

NA

8.34E-08

Import and Repackaging

Heritage IR, High BAF

3.15E00

3.53E00

1.09E01

1.22E01

2.43E-04

Incorporation into paints and coatings - 1-part coatings

Heritage IR, High BAF

7.14E-01

8.00E-01

2.47E00

2.77E00

1.07E-03

Incorporation into paints and coatings - 2-part reactive coatings

Heritage IR, High BAF

7.83E-01

8.78E-01

2.71E00

3.04E00

9.76E-04

Use in paints and coatings at job sites

Heritage IR, High BAF

1.34E00

1.50E00

4.64E00

5.20E00

5.71E-04

Formulation of TCEP containing reactive resin

Heritage IR, High BAF

5.61E-01

6.29E-01

1.94E00

2.18E00

1.36E-03

Laboratory chemicals

Heritage IR, High BAF

7.89E01

8.84E01

NA

NA

9.70E-06

Import and Repackaging

Heritage IR, Low BAF

6.37E01

7.14E01

NA

NA

1.20E-05

Incorporation into paints and coatings - 1-part coatings

Heritage IR, Low BAF

1.44E01

1.61E01

4.97E01

5.57E01

5.33E-05

Incorporation into paints and coatings - 2-part reactive coatings

Heritage IR, Low BAF

1.58E01

1.77E01

5.48E01

6.15E01

4.83E-05

Use in paints and coatings at job sites

Heritage IR, Low BAF

2.70E01

3.03E01

9.35E01

NA

2.83E-05

Formulation of TCEP containing reactive resin

Heritage IR, Low BAF

1.13E01

1.27E01

3.92E01

4.40E01

6.76E-05

Laboratory chemicals

Heritage IR, Low BAF

1.59E03

1.78E03

NA

NA

4.82E-07

13383

Page 512 of 572

-------
13384

13385

13386

13387

13388

13389

13390

13391

13392

13393

13394

13395

13396

13397

13398

13399

13400

13401

13402

13403

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

H.4.6 Sensitivity Analysis	

EPA conducted a sensitivity analysis for TCEP to evaluate the effect of chemical and biological
considerations on modeled milk concentrations, as shown in TableApx H-29. Sensitivity was measured
using elasticity, which is defined as the ratio of percent change in each result to the corresponding
percent change in model input. A positive elasticity means that an increase in the model parameter
resulted in an increase in the model output, whereas a negative elasticity had an associated decrease in
the model output. Table Apx H-7 shows the results of the sensitivity analysis.

Table Apx H-29. Variables and Values Used in Sensitivity Analysis

Variable

Base/Default Values

Sensitivity Values

Half-Life

17.64

15.87, 19.40 (increased and decreased from base value
by 10%)

Kow17

60.26

66.28 and 54.23 (increased and decreased from base
value by 10%)

Lipid fraction in milk

0.038 + 0.000095*age

Multiplied the function by 1.1 and 0.9 to increase and
decrease from base value by 10%, respectively

Age at pregnancy

25

40 (increased to reflect an alternate scenario)

11 The analysis varied Kow rather than log Kow because the partition coefficient equations used are based on Kow.
Kow is not used elsewhere in the model equations.

-0.5

Half-life
Half-life 4-
KOW t

KOW 4/

Milk Lipid Fraction

0.5

1.5

Milk Lipid Fraction vj,

FigureApx H-7. Sensitivity Analysis of Model Inputs Measured as Elasticity

The elasticity for half4ife is close to one. For the relatively short half-life (<24 hours) of TCEP, a ±10
percent change in half4ife reflected a near equivalent percent change in the infant milk dose. In contrast,
a ±10 percent change to Kow resulted in a smaller change in the infant milk dose. Half-life and Kow
parameters are independent values in the model. The half-life is used to estimate the liver compartment's
elimination rate while Kow is used to estimate the partition coefficients. For a slightly lipophilic
compound like TCEP, an increase in Kow (and calculated partition coefficient) leads to a relatively
larger increase in the blood:lipid partition coefficient than for other compartments such as mammary

Page 513 of 572

-------
13404

13405

13406

13407

13408

13409

13410

13411

13412

13413

13414

13415

13416

13417

13418

13419

13420

13421

13422

13423

13424

13425

13426

13427

13428

13429

13430

13431

13432

13433

13434

13435

13436

13437

13438

13439

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

tissue. Thus, more TCEP will be stored in lipids and less in the mammary tissue, causing a decrease in
infant milk dose. If half-life increases, more TCEP is available in the body and each compartment at a
given time, including the mammary tissue, causing an increase in infant milk dose. TCEP infant doses
were insensitive to alterations of milk lipid fractions. Milk concentrations were similarly insensitive
(data not shown). This insensitivity may reflect the relatively low Kow for TCEP.

Although the model treats Kow and half-life independently, these parameters are linked from a
toxicokinetic perspective. The Kow of the chemical likely influences both the partition coefficient (the
lipid compartments in particular) and the half-life. More lipophilic compounds tend to have larger
lipid:blood partition coefficient and longer half-lives than less lipophilic compounds. Thus, a 10 percent
change in Kow might also cause a percent change in the half life, and that correlation is not captured in
the model or sensitivity analysis.

Neither maternal age nor infant sex (results not shown) affected milk doses, indicating this model is not
sensitive to these parameters for TCEP. For infant sex, the only parameter differentiating male and
females in this model are growth curves, which are considered in the dose calculation.

H.5 Landfill Analysis Using DRAS

DRAS is an efficient tool developed by EPA Region 6 to provide a multipath risk assessment for the
evaluation of Resource Conservation and Recovery Act (RCRA) hazardous waste delisting. For the
TCEP Risk Evaluation, DRAS was specifically applied to model groundwater concentration estimates
from disposing TCEP to a hypothetical RCRA Subtitle D landfill at a range of loading rates and leachate
concentrations. A comprehensive description of the assumptions and calculations applied in DRAS can
be found in the Technical Support Document for the Hazardous Waste Delisting Risk Assessment
Software (https://www.epa.eov/hw/technical-siipport-dociiment-hazardoiis-waste-delistine-risk-
assessment-software-drasY

Because DRAS derives calculations based on a survey of drinking water wells located downgradient
from waste management units (U.S. EPA. 1988). the model may provide the closest estimate to real
world scenarios available. Alhough there is some uncertainty inherent to applying the model as an
assessment tool under amended TSCA for risk evaluations, few other tools are available to effectively
address this pathway. This appendix will provide the input variables and calculations used to apply the
model determine potential groundwater concentrations. TableApx H-30 and TableApx H-31 provide
the input values used for each parameter in the model. Note that loading volumes were based on the
range of estimated production volumes (2,500 to 25,000 lb) and were calculated based on the density of
TCEP (1.39 g/cm3). For each loading volume, the range of leachate concentrations was applied.

Page 514 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table Apx H-30. Input Variables for Chemical of

Concern

Input Variable for Chemical of Concern

Value

Chem Name

TCEP

CASRN

115-96-8

Maximum Contaminant Level

0

Oral Slope Cancer Factor

0.1a

Inhalation Slope Cancer Factor (1/mg kg day)

0.018a

Oral Reference Dose (mg/kg day)

0.03a

Inhalation Reference Dose (mg/kg day)

0.03a

Bioconcentration Factor (1/kg)

0

Soil Saturation Level

0

Toxicity Regulatory Rule regulatory level (mg/L)

0 a

Henry's Law Constant (atm -m3/mol)

2.95E-06

Diffusion coefficient in Water (cm2/s)

5.07E-06

Diffusion coefficient in Air (cm2/s)

0.044a

Water Solubility (mg/L)

7,820

Landfill Dilution Attenuation Factor

15.4

Surface Impoundment Dilution Attenuation Factor

3.18

Time to Skin Attenuation (hr/event)

0

Skin permeability constant (cm/hr )

0.00022a

Lag time (hr)

0.28a

Bunge constant

4.1E-05a

Organic

Yes

Bioaccumulation Factor (L/kg)

6,016a

Chronic Ecological Value (mg/L)

85a

Carcinogen

No

Molecular Weight (g/mol)

285.49

Vapor Pressure (atm)

8.07E-5

Suspended sediment-surface water partitioning
coefficient (mg/L)

298.725

log Kow (log[mg/l])

1.78

Chemical Class

svoca

Analytical Method

8,260Da

Version Description

Nonea

Create Date

Nonea

Creator

Nonea

Cancer Risk Level

1.00E-06a

Hazard Quotient

^a

a Input variables do not directly or indirectly affect groundwater concentrations

Page 515 of 572

-------
13442

13443

13444

13445

13446

13447

13448

13449

13450

13451

13452

13453

13454

13455

13456

13457

13458

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table Apx H-31. Waste Management Unit (WMU) Properties

Input Variable for WMU Properties

Value(s)

Waste Management Unit Type

Landfill

Loading Volume (m3)

8.17E-01

8.17E00

Cancer Risk Level

1.00E-06

Hazard Quotient

1.0

Detection Limit

0.5

Waste Management Active Life (years)

20

TCLP Concentration (mg/L)/Total
Concentration (mg/kg)

0.0001

0.001

0.01

0.1

1

Once the model was executed for each loading rate and leachate concentration scenario, the groundwater
concentration was calculated using the leachate concentration and the 90th percentile weight-adjusted
dilatation attenuation factor using:

Equation Apx H-9

r-.j 	 Leachate Concentration

(j Wr —	,

Weight-Adjusted. DAF

Where:

GWC	= Groundwater concentration

Leachate concentration = Input variable for the waste management unit

Weight-Adjusted DAF = Weight- adjusted dilution attenuation factor

The results of these analyses are provided in Table 3-7.

Page 516 of 572

-------
13459

13460

13461

13462

13463

13464

13465

13466

13467

13468

13469

13470

13471

13472

13473

13474

13475

13476

13477

13478

13479

13480

13481

13482

13483

13484

13485

13486

13487

13488

13489

13490

13491

13492

13493

13494

13495

13496

13497

13498

13499

13500

13501

13502

13503

13504

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Appendix I CONSUMER EXPOSURE DETAILS

1.1 Approach and Methodology

EPA evaluated TCEP exposure resulting from the use of consumer products and industrial processes.
The Agency utilized a modeling approach to evaluate exposure because chemical-specific personal
monitoring data attributable to the COUs was not identified for consumers during data gathering and
literature searches performed as part of systematic review using the evaluation strategies described in the
Draft Systematic Review Protocol Supporting TSCA Risk Evaluations for Chemical Substances (U.S.
EPA. 2021) and in the Systematic Review Protocol for the Draft Risk Evaluation for Tris(2-chloroethyl)
Phosphate (TCEP) (U.S. EPA. 2023n).

There are a limited number of consumer articles that still contain TCEP, because many manufacturers
have reformulated them to remove TCEP. Consumer products containing TCEP are readily available via
the internet as finished articles (e.g., furniture and foam products). Use of these products can result in
exposures of the consumer user to TCEP during and after article use. Consumer exposure can occur via
inhalation, dermal, and oral routes.

Consumer products containing TCEP were identified through review and searches of a variety of
sources, including the National Institutes of Health (NTH) Household Products Database, various
government and trade association sources for products containing TCEP, company websites for safety
data sheets (SDSs), Kirk-Othmer Encyclopedia of Chemical Technology, and the internet. In general,
information on the consumer uses of TCEP was sparse and many manufacturers reported changes in
formulation and ceasing the use of TCEP in favor of other chemicals.

Identified consumer products (see Table l-l) were then categorized into six consumer use groups
considering (1) consumer use patterns, (2) information reported in SDSs, (3) product availability to the
public, and (4) potential risk to consumers.

Readers are referred to each model's user guide and associated user guide appendices for details on each
model, as well as information related to equations used within the models, default values, and the basis
for default values. Each model is peer reviewed. Default values within CEM are a combination of high
end and mean or central tendency values derived from EPA's Exposure Factors Handbook (
2017c). literature, and other studies.

1.1.1 Consumer Exposure Model (CEM)	

CEM 3.0 is a deterministic model that utilizes user provided input parameters and various assumptions
(or defaults) to generate exposure estimates. In addition to pre-defined scenarios, which align well with
the consumer uses identified in Table 1-1, CEM is peer reviewed, provides flexibility to the user
allowing modification of certain default parameters when chemical-specific information is available and
does not require chemical-specific emissions data (which may be required to run more complex
indoor/consumer models).

CEM predicts indoor air concentrations from consumer product use through a deterministic, mass-
balance calculation derived from emission calculation profiles within the model. There are six emission
calculation profiles within CEM (E1-E6) that are summarized in the CEM users guide and associated
appendices https://www.epa.eov/tsca-screenine4ools. If selected, CEM provides a time series air
concentration profile for each run. These are intermediate values produced prior to applying pre-defined
activity patterns.

Page 517 of 572

-------
13505

13506

13507

13508

13509

13510

13511

13512

13513

13514

13515

13516

13517

13518

13519

13520

13521

13522

13523

13524

13525

13526

13527

13528

13529

13530

13531

13532

13533

13534

13535

13536

13537

13538

13539

13540

13541

13542

13543

13544

13545

13546

13547

13548

13549

13550

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

CEM uses a two-zone representation of the building of use when predicting indoor air concentrations.
Zone 1 represents the room where the consumer product is used. Zone 2 represents the remainder of the
building. Each zone is considered well-mixed. CEM allows further division of Zone 1 into a near field
and far field to accommodate situations where a higher concentration of product is expected very near
the product user when the product is used. Zone 1-near field represents the breathing zone of the user at
the location of the product use while Zone 1-far field represents the remainder of the Zone 1 room.
Inhalation exposure is estimated in CEM based on zones and pre-defined activity patterns. The
simulation run by CEM places the product user within Zone 1 for the duration of product use while the
bystander is placed in Zone 2 for the duration of product use. Following the duration of product use, the
user and bystander follow one of three pre-defined activity patterns established within CEM, based on
modeler selection. The selected activity pattern takes the user and bystander in and out of Zone 1 and
Zone 2 for the period of the simulation. The user and bystander inhale airborne concentrations within
those zones, which will vary over time, resulting in the overall estimated exposure to the user and
bystander.

CEM contains two methodologies for estimating dermal exposure to chemicals in products—the
permeability method (P-DER1) and the fraction absorbed method (A-DER1). Each of these
methodologies further has two model types, one designed for dermal exposure from use of a product (P-
DERla and A-DERla) and the other designed for dermal exposure from use of an article (P-DERlb and
A-DERlb). Each methodology has associated assumptions, uncertainties, and data input needs within
the CEM model. Both methodologies factor in the dermal surface area to body weight ratio and weight
fraction of chemical in a consumer product.

The permeability model is based on the ability of a chemical to penetrate the skin layer once contact
occurs. The permeability model assumes a constant supply of chemical, directly in contact with the skin,
throughout the exposure duration. The ability to use the permeability method can be beneficial when
chemical-specific skin permeability coefficients are available in the scientific literature. However, the
permeability model within CEM does not consider evaporative losses when it estimates dermal exposure
and therefore may be more representative of a dermal exposure resulting from a constant supply of
chemical to the skin due to a barrier or other factor that may restrict evaporation of the chemical of
interest from the skin such as a product soaked rag against the hand while using a product), or
immersion of a body part into a pool of product. Either of these examples has the potential to cause an
increased duration of dermal contact and permeation of the chemical into the skin resulting in dermal
exposure.

The fraction absorbed method is based on the absorbed dose of a chemical. This method essentially
measures two competing processes, evaporation of the chemical from the skin and penetration of the
chemical deeper into the skin. This methodology assumes the application of the chemical of concern
occurs once to an input thickness and then absorption occurs over an estimated absorption time. The
fraction absorbed method can be beneficial when chemical specific fractional absorption measurements
are available in the scientific literature. The consideration of evaporative losses by the fraction absorbed
method within CEM may make this model more representative of a dermal exposure resulting from
scenarios that allow for continuous evaporation and typically would not involve a constant supply of
product for dermal permeation. Examples of such scenarios include spraying a product onto a mirror and
a small amount of mist falling onto an unprotected hand. For TCEP, literature values for fraction
absorbed were used from Abdallah et al. (2016). rather than the faction absorbed estimation via CEM.

Page 518 of 572

-------
13551

13552

13553

13554

13555

13556

13557

13558

13559

13560

13561

13562

13563

13564

13565

13566

13567

13568

13569

13570

13571

13572

13573

13574

13575

13576

13577

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

1.1.1 Inputs

1.1.1.1 Consumer Exposure Modeling and Sensitivity Analysis

Inputs for the each of the CEM 3.0 base and sensitivity runs are provide in

TCEP_Draft_Exp_Consumer_Inputs_May_2023.xlsx. Where available, EPA relied on the Exposure
Factors Handbook (	) and the peer-reviewed and gray literature to inform input

parameters. For article-specific parameters (e.g., product density, thickness of article surface layer,
surface area) that were unavailable in the handbook or the peer-reviewed or gray literature, EPA used
professional judgment to determine whether the CEM default values were appropriate, or whether there
should be an alternative value for the parameter based on professional judgment. All the input
parameters and their rationale are provided in the Draft Risk Evaluation for Tris(2-chloroethyl)
Phosphate (TCEP) - Supplemental Information File: Consumer Exposure Modeling Inputs (U.S. EPA.
2023c). Inputs for the sensitivity analysis are provided in the "Sensitivity Analysis" tab of the Consumer
Exposure Modeling Inputs (	|23c).

1.1.1 Results	

Raw Consumer Modeling results are available in pdf and xlsx format in

TCEP_Consumer_Modeling_Results.zip. Results from the consumer modeling have been visualized in
bar charts, and risk tables in the Supplemental TCEP Consumer Modeling Results.

1.1.1.1 Navigating Supplemental Consumer Modeling Results

Consumer Modeling Results were tabulated in R and have been displayed in an "Rmarkdown file." The
associated R script uses a workflow that loads the input data from the consumer modeling results,
cleans, filters, and wrangles the relevant data, and displays the modeling results in the form of bar plots
and risk tables.

Bar plots are interactive, and reviewers are able to pan and select certain data fields to help compare the
results from the various consumer COUs (see FigureApx 1-1 through FigureApx 1-4). Hovering over
the data bars provides a tool tip that indicates the value of the bar.

Page 519 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE

December 2023

Lifetime Average Daily Doses (LADDs)

TCEP CQl



1.S
1..6
1.4

db

-L.

E

* O.S
c

cj

3 06

0.4
0.2
0-*

Inhalation
Ingestion
Dermal



£



Inhalation

%

''%•
^ »«¦ %



% % %,

r"- tath





S>r„

"»S

cou



¦«*>

%

'n

Hr

"rhi

13578

13579

13580

13581

13582

13583

13584

FigureApx 1-1. Screenshot of Lifetime Average Daily Doses (LADDs) Bar
Chart Displaying Tool Tip for Acoustic Ceiling, Inhalation Estimate

Source: Supplemental TCEP Consumer Modeling Results

The toolbar at the top also has various functionalities that can allow for more exploration of the data. For
example, simply hover and select the outlined double bars to compare data.

Lifetime Average Daily Doses (LADDs)

TCEP COUs ,

M
£
E

0.6

13585

13586

13587

13588

lnih3laton

(=)e

Inhalation
Ingestion
Dermal



Figure Apx 1-2. Screenshot of Lifetime Average Daily Doses (LADDs) Bar
Chart Displaying Function to Compare Data on Hover, for Insulation Estimates

Source: Supplemental TCEP Consumer Modeling Results.

Page 520 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE

December 2023

Or to select and deselect data, the viewer can click the legend to remove data from the accompanying
bar plot.

Lifetime Average Daily Doses (LADDs)

TCEPCOUs

0.4

- 0.3
•5b

"3> 0.25

tt 0.15

0.1

A"

Ingestion
Dermal





s'i



cou

"k

FigureApx 1-3. Screenshot of Lifetime Average Daily Doses (LADDs) Bar Chart Displaying Bar
Chart that Deselects Inhalation Estimate and Selects Ingestion and Dermal Estimates

Source: Supplemental TCEP Consumer Modeling Results

Or the viewer can drag and select a certain section of the plot to view it in greater detail:

Lifetime Average Daily Doses (LADDs)

TCEP COUs

Lifetime Average Daily Doses (LADDs)

TCEP COUs

0.25

ab
E

Bf 0.2

ingestion
Dermal

wood-tvsrand

Figure Apx 1-4. Screenshots of Lifetime Average Daily Doses (LADDs) Bar Chart Displaying a
Cropped Subsection of the Figure

Source: from Supplemental TCEP Consumer Modeling Results.

Page 521 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

13604	1.1.1.1 CEM 3.0 User Guide and Appendices

13605	The CEM 3.0 user guide and appendices provide the underlying equations and default parameters that are

13606	used in CEM 3.0. The Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Supplemental
13 607 Information File: Consumer Exposure Modeling Inputs (U,	2023 c) gives the inputs and
13608 assumptions used for consumer modeling.

Page 522 of 572

-------
13609

13610

13611

13612

13613

13614

13615

13616

13617

13618

13619

13620

13621

13622

13623

13624

13625

13626

13627

13628

13629

13630

13631

13632

13633

13634

13635

13636

13637

13638

13639

13640

13641

13642

13643

13644

13645

13646

13647

13648

13649

13650

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Appendix J HUMAN HEALTH HAZARD DETAILS

J,1 Toxicokinetics and PBPK Models

J.l.l Absorption

EPA did not identify in vivo human studies that evaluated absorption, distribution, metabolism, or
elimination (ADME) of TCEP by any route of exposure.

Oral

Following oral exposures to radiolabeled TCEP, in vivo ADME studies in rats and mice found that
TCEP is rapidly and extensively absorbed. More than 90 percent of 14C-labeled TCEP was absorbed
based on radioactivity found in urine, feces, volatiles, and CO: after 2 hours post-dose (Burka et al.
1991; Heir et al.. 1991). For input to the draft risk evaluation, EPA will assume that absorption is 100
percent.

Inhalation

EPA did not identify any in vivo animal data for absorption of TCEP by the inhalation route of exposure.
For input to the draft risk evaluation, EPA will assume that absorption is 100 percent, equivalent to oral
exposure.

Dermal

EPA did not locate any in vivo studies of dermal absorption in humans or animals but identified an in
vitro study using excised human skin that evaluated the dermal absorption of TCEP (Abdallah et al..
2016V

Although no dermal in vivo toxicokinetic studies are available, EPA identified Abdallah et al.
which measured dermal absorption using excised human skin in multiple in vitro experiments conducted
according to OECD TG 428, Skin Absorption: In Vitro Method. The experiments used exposures of
either 24 or 6 hours; acetone or 20 percent Tween 80 in water as the vehicle; 500 or 1,000 ng/cm2
application to skin; and finite (depletable) or infinite dose. EPA gave each of the finite dose experiments
overall quality determinations of medium. For the experiment that claimed to investigate an infinite
dose, EPA assigned a low overall quality determination scenario, because conditions for infinite dosing
(use of neat or large body of material) were not met and the results did not reflect steady-state flux
throughout the experiment (e.g., applied dose was depletable).

EPA used the 500 ng/cm2 24-hour finite dose application in acetone (0.005 percent solution) to estimate
absorption for workers because this was the only experiment for which the authors reported absorption
at multiple time points. Because EPA assumes workers wash their hands after an 8-hour shift, EPA used
the value of 16.5 percent, which is the amount of TCEP absorbed at 8 hours. In accordance with OECD
Guidance Document 156 (OECD. 2022). EPA also added the quantity of material remaining in the skin
(6.8 percent) at the end of the experiment as potentially absorbable.51 Therefore, EPA assumes workers
absorb 23.3 percent TCEP through skin and used this value to calculate risks for workers (see Section
5.1.1.3).

51 EPA used 6.8 percent (the total amount remaining in skin after washing) because the authors did not conduct tape
stripping.

Page 523 of 572

-------
13651

13652

13653

13654

13655

13656

13657

13658

13659

13660

13661

13662

13663

13664

13665

13666

13667

13668

13669

13670

13671

13672

13673

13674

13675

13676

13677

13678

13679

13680

13681

13682

13683

13684

13685

13686

13687

13688

13689

13690

13691

13692

13693

13694

13695

13696

13697

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

For consumer exposures and exposure to soil scenarios that assume hand washing does not occur for 24
hours, EPA used the value at 24 hours (28.3 percent) plus the amount remaining in skin (6.8 percent)
from the same experiment used for workers (500 ng/cm2 24-hour finite dose application in acetone);
total absorption was 35.1 percent absorption and was used to calculate risks (see Sections 5.1.2.2.3 and
5.1.3.3.2).

The estimates identified above apply to finite exposure scenarios for which the TCEP dose is depleted
over time. For exposure scenarios such as swimming in which a maximum absorption rate is expected to
be maintained {i.e., the dose is not depletable during the exposure duration), EPA used the dermal
permeability coefficient (Kp) of 2.2 10 2 cm/h derived by Abdallah et al. (2016) from the experiment
that used the 24-hour 1,000 ng/cm2 TCEP skin application to calculate risks (see Section 5.1.3.3.1).

023 q) presents quality determinations for individual experiments conducted by Abdallah et
al. (20161 with EPA comments for each of the data quality metrics. Data extraction tables with details
on methods and results of the experiments are also presented in	)23q).

J.1.2 Distribution

Oral

TCEP distributes widely throughout the body. At 2 hours following the oral exposure, there was TCEP-
derived 14C in all brain regions of male and female rats. Also, the increasing levels of TCEP-derived 14C
were observed with increasing TCEP doses. There were no significant differences in TCEP-derived 14C
levels in blood and brain (including cerebellum, brainstem, caudate, hypothalamus, cortex,
hippocampus, and midbrain) in male and female rats and 24 hours following a single dose. The
concentration of 14C-labeled TCEP in blood was significantly more increased with dose in males than
females after 2 hours (p < 0.05). However, there was no significant difference in the amount of TCEP
present in blood and all brain regions after 24 hours of exposure (Burka et al.. 1991; Heir et al.. 1991).
Oral administration studies in rats by NTP found that TCEP produced sex-specific seizures and lesions
in the hippocampal brain regions in some animals receiving the higher doses (NI	). Results

reported by Heir et	observed similar sex-specific clinical signs of toxicity in animals receiving

the higher doses.

Inhalation

No in vivo animal data evaluating the distribution of TCEP following inhalation route exposures were
identified.

Dermal

EPA did not identify in vivo animal data that evaluated the distribution of TCEP following dermal route
exposures.

J.1.3 Metabolism

Oral

TCEP is predominantly metabolized in the liver in laboratory animals and urinary excretion is the
primary route of elimination for metabolites. In the liver, two pathways are involved in the metabolism
of TCEP (Burka et al.. 1991; Heir	). First pass biotransformation occurs via oxidative and

hydrolytic pathways. Some oxidative metabolites can undergo secondary biotransformation via the
glucuronidation pathway. Burka et al. (1991) conducted a study to detect variations in metabolism of
TCEP between male mice and male and female rats. The results showed that TCEP underwent extensive
metabolism in all three groups. TCEP was excreted primarily in the form of metabolites in urine and
feces of both species and were identified as hydrogen phosphate (BCHP), bis(2-chloroethyl) 2-

Page 524 of 572

-------
13698

13699

13700

13701

13702

13703

13704

13705

13706

13707

13708

13709

13710

13711

13712

13713

13714

13715

13716

13717

13718

13719

13720

13721

13722

13723

13724

13725

13726

13727

13728

13729

13730

13731

13732

13733

13734

13735

13736

13737

13738

13739

13740

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

hydroxyethyl phosphate (BCGP), andbis(2-chloroethyl) carboxymethyl phosphate (BCCP) (Burkaet
ai. 1991). In other toxicological studies in rats and mice, TCEP has been shown to cause neurotoxicity
at lower doses in females than in males (Yame et ai. 2018a; NTP. 1991b; Matthews et ai. 1990). Burka
et ai (1991) examined whether there was any relationship between acute neurotoxicity and metabolism.
Male and female rats were pretreated with aldehyde dehydrogenase inhibitors to alter the urinary
metabolic profile. The relative amount of the hydrolytic metabolite (BCHP) was increased compared to
the oxidative metabolite (BCCP). Because aldehyde dehydrogenase inhibitors interfere with the
metabolic pathway leading to the oxidative metabolite (BCCP), increased levels of the reactive
metabolite may possibly account for increased neurotoxicity (Burka et ai. 1991).

Inhalation

No in vivo animal data for metabolism of TCEP by the inhalation route of exposure was identified.
Dermal

EPA did not identify in vivo animal data that evaluated metabolism of TCEP by the dermal route of
exposure.

J.1.4 Elimination

Oral

TCEP is primarily eliminated in the urine following oral exposure. Burka	and Herr et ai

0.lllii) reported that more than 75 percent of 14C-labeled TCEP was eliminated in 24 hours for both rats
and mice, with less than 10 percent excreted in feces (Burka et ai. 1991). There was little to no sex-
specific difference in the rate of elimination of TCEP for rats. However, male mice eliminated TCEP at
3 times the rate observed for rats during the first 8 hours (Burka et ai. 1991). Urinary excretion is the
primary route of elimination for metabolites (Burka et ai. 1991; Herr et ai. 1991).

Inhalation

No in vivo animal data for metabolism of TCEP by the inhalation route of exposure was identified.
Dermal

EPA did not identify in vivo animal data that evaluated elimination of TCEP by the dermal route of
exposure.

J. 1.5 PBPK Modeling Approach

EPA did not identify any PBPK models specific to TCEP but is using the Verner Model (Verner et ai.
2009; Verner et ai. 2008) to predict milk concentrations used to assess infant exposure through
ingestion of human milk. The model is described in Appendix H.4.1.

J.2 Detailed Mode of Action Information

EPA has determined that TCEP is likely to cause tumors in kidneys under exposure circumstances
relevant to human health. For blood cancer (mononuclear cell leukemia); thyroid cancer (follicular cell
adenoma or carcinoma); Harderian gland cancer (adenoma or carcinoma); and liver cancer
(hepatocellular adenomas or carcinomas), evidence of carcinogenicity is slight. EPA summarizes
biochemical, cellular, and mechanistic data that may be relevant to induction of kidney tumors—the
target organ with the strongest weight of the scientific evidence conclusion.

Page 525 of 572

-------
13741

13742

13743

13744

13745

13746

13747

13748

13749

13750

13751

13752

13753

13754

13755

13756

13757

13758

13759

13760

13761

13762

13763

13764

13765

13766

13767

13768

13769

13770

13771

13772

13773

13774

13775

13776

13777

13778

13779

13780

13781

13782

13783

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Although EPA did not specifically investigate other possible mechanisms related to other tumor types
following TCEP exposure, conclusions for induction of kidney tumors may be relevant for induction of
other tumors.

3.2.1 Mutagenicity

EPA did not identify in vivo studies that evaluated any of the following relevant effects specifically in
kidneys, the target of tumors likely to be caused by TCEP: (1) oncogene or tumor suppressor gene
mutations, (2) other gene mutations and chromosomal aberrations, (3) DNA adducts, or (4) DNA
damage. However, one in vivo micronucleus assay in Chinese hamsters via intraperitoneal (i.p.)
administration did identify the presence of micronuclei in bone marrow (Sala	2) and EPA

considered this to be equivocal/weakly positive.52 Also, EPA did not identify any additional in vivo
studies that evaluated DNA damage, DNA adducts or other measures of DNA damage and/repair in
surrogate tissues.

Most bacterial reverse mutation assays using Salmonella typhimurium strains showed that TCEP was
negative for direct gene mutations (Follmann and Wober. 2006; NTP. 1991b; Haworth et ai. 1983;
Prival et al. 1977; Simmon et ai. 1977). TCEP was also negative in a study of forward gene mutations
in Chinese hamster lung fibroblasts (Sala	I).53

However, Nakamura et al. (1979) identified positive dose-response trends in two S. typhimurium strains:
in TA100, the response was less than two-fold higher than the negative control at the highest non-toxic
dose, but in TA1535 (with metabolic activation), TCEP induced an increase of more four- to seven-fold
over controls. It is not clear why the results of Nakamura et ;il \ lv">79) differed from other studies, but
Nakamura et al. (1979) used Kanechlor 500 to induce enzymes in the S9 fraction whereas other studies
used Aroclor 1254 or did not use a method to induce enzymes.

Two studies of TCEP induction of SCEs identified equivocal results in Chinese hamster ovary cells
(positive in one of two trials with S9, negative without S9) and positive results without a dose-response
in Chinese hamster lung fibroblasts (Galloway et al.. 1987; Sala et al.. 1982). suggesting some genetic
damage. These results are not definitive for direct mutagenic effects because there is a lack of
understanding of SCEs mechanism(s) of action (OB	).

TCEP was not considered to be an alkylating agent in an in vitro DNA binding assay (Lown et al..

1980).

Bukowski et al. (2019) conducted in vitro comet assays in peripheral mononuclear blood cells (PMBCs)
and identified DNA damage at the highest concentration tested (1 mM); however, there is uncertainty
regarding whether cytotoxicity occurred at this concentration. Another comet assay did not identify
DNA damage in Chinese hamster fibroblasts at TCEP concentrations up to 1 mM with or without
metabolic activation (Follmann and Wober. 2006).

Sala et al. (1982) identified a high level of cell transformation in Syrian hamster embryo (SHE) cells but
a lower level using C3H10T1/2 cells with metabolic activation. OECD (2007). p. 24, states that "cell
transformation has been related to structural alterations and changes in the expression of genes involved

52	Two additional micronucleus tests in mice (one via the oral route and one via i.p.) were negative (Beth-Hubner. 1999") but
the studies were not available for review by EPA.

53	Beth-Hubner (.1.999) reported negative results in a reverse gene mutation assay using Saccharomyces cerevisiae D4 and in
two mouse lymphoma assays (using the thymidine kinase locus).

Page 526 of 572

-------
13784

13785

13786

13787

13788

13789

13790

13791

13792

13793

13794

13795

13796

13797

13798

13799

13800

13801

13802

13803

13804

13805

13806

13807

13808

13809

13810

13811

13812

13813

13814

13815

13816

13817

13818

13819

13820

13821

13822

13823

13824

13825

13826

13827

13828

13829

13830

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

in cell cycle control, proliferation and differentiation." The genomic changes may result from direct or
indirect genetic interactions or non-genotoxic mechanisms.

EPA did not identify in vitro studies of DNA adducts.

Although there is uncertainty regarding reasons for equivocal/weakly positive results, EPA concludes
that TCEP is not likely to induce tumors via a mutagenic MOA.

3.2.2 Other Modes of Action

Biochemical and mechanistic information that may suggest TCEP could act via MO As other than a
mutagenic MOA. Several in vivo and in vitro studies have evaluated tissue changes, gene transcription,
and protein activities among other activities that identified tumor precursors or possible key events in
mechanisms of tumor induction.

Taniai et al. (2012a) dosed male F344/NSIc rats daily via oral gavage with 0 or 350 mg/kg-bw/day
TCEP and examined effects on proximal tubular epithelial cells of the outer stripe of the outer medulla
(OSOM) of the kidney as well as the whole cortex. TCEP exposure resulted in scattered proximal
tubular regeneration, likely associated with cells in the quiescent GO-phase of the cell cycle. TCEP did
not induce karyomegaly (enlarged nuclei) in the tubular epithelia. TCEP also led to a significant increase
in Ki-67 immunoreactive cells vs. controls (p < 0.01); Ki-67 nuclear antigen is a marker of cell
proliferation expressed in cells in the G1 to M phase of the cell cycle. However, TCEP exposure did not
result in aberrant expression of cell cycle-related molecules except for topoisomerase Ila (Topo Ha),
which acts from the late S to G2 and M phase; TCEP significantly increased Topo Ila-immunoreactive
cells in the cortex and OSOM (p < 0.01), which may signify increased cell proliferation (Taniai et al..
2012a). It is also possible that DNA damage may have been a precipitating factor in the increase of
Topo Ila (Taniai et al.. 2012a).

Using the same protocol {i.e., male rats dosed via oral gavage at 0 or 350 mg/kg-day TCEP for 28 days),
Taniai et al. (2012b) observed that TCEP exposure increased cells immunoreactive for markers of cell
proliferation (Mcm3), apoptosis (Ubd) and deregulation of the G2/M phase of the cell cycle (TUNEL) (p
< 0.01). Carcinogens that increase cell proliferation may increase cell populations undergoing M phase
disruption that leads to chromosomal instability linked to cancer (Taniai et al.. 2012b).

In vitro studies show that TCEP exposure of primary rabbit renal proximal tubule cells (PTCs) resulted
in cytotoxicity, reduced DNA synthesis, altered expression of cell cycle regulatory proteins, and
inhibition of ion- and non-ion-transport functions. Increased expression of pro-apoptotic regulatory
proteins and decreased expression of proteins that inhibit apoptosis were also observed (Ren et al.. 2012;
Ren et al.. 2009. 2008).

Additional in vivo and in vitro studies identified several biochemical changes in tissues and cell of other
organs. Male ICR mice exposed to TCEP in the diet for 35 days exhibited increased markers of
oxidative stress (hepatic antioxidant enzyme activities and their gene expression) in livers (Chen et al..
2015a). Liver cells or cell lines cultured with TCEP exhibited reduced viability, cell cycle arrest, cellular
and mitochondrial oxidative stress, impaired mitochondrial function, and perturbation of cell signaling
pathways (Mennillo et al.. 2019; Zhang et al.. 2017b; Zhang et A JO I j; bang et al.. 2016c; Zhang et
al.. 2016b). TCEP exposure of human peripheral blood mononuclear cells resulted in cytotoxicity
(Mokra et al.. 2018) and decreased DNA methylation (Bukowski et al.. 2019).

Page 527 of 572

-------
13831

13832

13833

13834

13835

13836

13837

13838

13839

13840

13841

13842

13843

13844

13845

13846

13847

13848

13849

13850

13851

13852

13853

13854

13855

13856

13857

13858

13859

13860

13861

13862

13863

13864

13865

13866

13867

13868

13869

13870

13871

13872

13873

13874

13875

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

In	, the authors reported no hyperplasia in rats at the 66-week interim sacrifice in the

narrative (data tables not included). Although focal hyperplasia was observed and can be expected to be
a precursor to tumors, the only related finding regarding kidney tumors at the 66-week sacrifice was a
single renal tubule adenoma seen in a female rat. Therefore, evidence of temporal progression from
hyperplasia to adenoma and then carcinoma is not available. At two-years, hyperplasia was observed in
male rats but incidence was slightly lower (0, 2, and 24) than adenomas (1, 5, and 24) compared with
hyperplasia at 0, 44, and 88 mg/kg-day. The lack of temporality and limited information on pre-cursor
lesions and their relationship with tumors leads to uncertainty regarding dose-response progression from
hyperplasia to adenomas and carcinomas in males. Female rats did have higher rates of hyperplasia (0,
3, 16) than adenomas (0, 2, 5), at 0, 44, and 88 mg/kg-day, respectively.

Conclusion

J.2.3 Mode of Action Conclusions	

EPA concluded that a mutagenic MOA is not likely from exposure to TCEP. Several studies have
investigated biochemical and cellular changes in kidneys or renal cells that may be associated with steps
in other MO As for kidney cancer. However, EPA has not performed a formal analysis on postulated
MO As (e.g., as in Sonich-Mullin et al. (2> ).

There is sparse information on temporality and dose-response of potential pre-cursor events within the in
vivo studies and no clear NOAEL regarding tumor response to be able to model tumor incidence with a
nonlinear/threshold dose response analysis.

U.S. EPA's PPRTV (	309) concluded that the overall weight of evidence for mutagenicity is

negative and that no mechanistic data identify specific potential key events in an MOA for kidney or
other tumors induced by TCEP exposure other than a general association with known proliferative and
preneoplastic lesions.

J.3 Dose-Response Derivation

EPA evaluated data for health outcomes with the strongest weight of the scientific evidence and from
studies with sufficient sensitivity and adequate quantitative information to characterize the dose-
response relationships of TCEP (see Section 5.2.6.1).

J.3.1 Adjustments for All PODs (Non-cancer and Cancer)

For TCEP, all data considered for PODs are obtained from oral animal toxicity studies in rats or mice.
For consistency and easier comparison of sensitivity across health effects, EPA converted all doses to
daily doses before conducting benchmark dose (BMD) modeling. For example, if the toxicity study
dosed animals via gavage for five days per week at 22 mg/kg-day, EPA multiplied that value by 5/7 to
obtain an equivalent daily value of 15.7 mg/kg-day. Studies in which animals were dosed every day did
not require conversion. Any adjustments for different frequency of exposure (e.g., five days per week
for workers) are made in the exposure calculations specific to exposure scenarios.

Because toxicity values for TCEP are from oral animal studies, EPA must use an extrapolation method
to estimate equivalent human doses (HEDs) and cancer slope factors (CSFs). The preferred method
would be to use chemical-specific information for such an extrapolation. However, there are no TCEP-
specific PBPK models and EPA did not locate other TCEP information to conduct a chemical-specific
quantitative extrapolation. In the absence of such data, EPA relied on the guidance from U.S. EPA.
(201 \ which recommends scaling allometrically across species using the three-quarter power of body

Page 528 of 572

-------
13876

13877

13878

13879

13880

13881

13882

13883

13884

13885

13886

13887

13888

13889

13890

13891

13892

13893

13894

13895

13896

13897

13898

13899

13900

13901

13902

13903

13904

13905

13906

13907

13908

13909

13910

13911

13912

13913

13914

13915

13916

13917

13918

13919

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

weight (BW3/4) for oral data. Allometric scaling accounts for differences in physiological and
biochemical processes, mostly related to kinetics.

For application of allometric scaling in risk evaluations, EPA uses dosimetric adjustment factors
(DAFs), which can be calculated using EquationApx J-l.

EquationApx J-l. Dosimetric Adjustment Factor (DAF)

presents DAFs for extrapolation to humans from several species. However, because
those DAFs used a human body weight of 70 kg, EPA has updated the DAFs using a human body
weight of 80 kg for the TCEP risk evaluation (	a). EPA used the body weights of 0.025

and 0.25 kg for mice and rats, respectively, as presented in 1 c. « ^ \ (.VI I. I The resulting DAFs for
mice and rats are 0.133 and 0.236, respectively.

For this draft risk evaluation, EPA assumes absorption for oral and inhalation routes is 100 percent and
no adjustment was made when extrapolating to the inhalation route. This is supported by oral
toxicokinetics data that shows greater than 90 percent absorption via the oral route (Burka et ai. 1991).

J.3.2 Non-cancer Dose-Response Modeling

EPA concluded that TCEP likely causes neurotoxicity, reproductive, developmental, and kidney effects
in humans under relevant exposure circumstances. For these outcomes (as well as suggestive evidence
integration conclusions), EPA conducted BMD modeling (	2023b) and compared PODs

among these two categories of evidence integration conclusion categories to determine the sensitivity of
individual health affects (	0231). Although EPA conducted BMD modeling for the non-

cancer hazard outcomes with suggestive evidence integration conclusions, the focus of the evaluation
was on the likely endpoints. Section 5.2.6.1 describes how EPA chose the sensitive studies and
individual health effects within these health outcome categories for the non-cancer HED and HEC
derivations.

As noted above, EPA converted doses for each study to daily doses before conducting BMD modeling.
If data were not amenable to BMD modeling (e.g., there was only one treatment group) or data did not
fit BMD models, NOAELs or LOAELs were also converted to daily values, as needed.

Use of allometric scaling for oral animal toxicity data to account for differences among species allows
EPA to decrease the default intraspecies uncertainty factor (UFa) used to set the benchmark margin of
exposure (MOE); the default value of 10 can be decreased to 3, which accounts for any toxicodynamic
differences that are not covered by use of BW3 4 Using the appropriate DAF from Equation Apx J-l,
EPA adjusts the POD to obtain the daily HED:

Where:

DAF
BWa
BWh

Dosimetric adjustment factor (unitless)

Body weight of species used in toxicity study (kg)

Body weight of adult human (kg)

J.3.2.1 Calculating Daily Oral Human Equivalent Doses (HEDs)

Page 529 of 572

-------
13920

13921

13922

13923

13924

13925

13926

13927

13928

13929

13930

13931

13932

13933

13934

13935

13936

13937

13938

13939

13940

13941

13942

13943

13944

13945

13946

13947

13948

13949

13950

13951

13952

13953

13954

13955

13956

13957

13958

13959

13960

13961

13962

13963

13964

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

EquationApx J-2. Daily Oral Human Equivalent Dose (HED)

HEDDaiiy = PODDany X DAF

Where:

HEDDaiiy = Human equivalent dose assuming daily doses (mg/kg-day)

PODDaiiy = Oral POD assuming daily doses (mg/kg-day)
DAF	= Dosimetric adjustment factor (unitless)

J.3.2.2 Use of Oral HED as Dermal HED

llj	• I k - recommends the BW3/4 approach only for oral PODs, and there is no established

guidance for dosimetric adjustments of dermal PODs. However, EPA only extrapolated between species
from oral animal toxicity values because the only acceptable data were from oral studies. EPA
extrapolated to the dermal HED from the oral HED after the oral species extrapolation and accounted for
differences in absorption in the dermal exposure estimate, not within the HEDs.

EPA used a value of 23.3 percent (hand washing after 8 hours) for workers as described in Section
5.1.1.3. EPA used a value of 35.1 percent (no handwashing for 24 hours) for dermal absorption in
calculations of consumer exposure and exposure to soil, which are described in Sections 5.1.2.2.3 and
5.1.3.3.2, respectively. For dermal exposure from swimming (a nondepletable source), EPA uses the
dermal permeability coefficient (Kp) of 2.2><10~2 cm/hr as described in Section 5.1.3.3.1. The same
uncertainty factors are used in the benchmark MOE for both oral and dermal scenarios.

J.3.2.3 Extrapolating to Inhalation Human Equivalent Concentrations (HECs)

For the inhalation route, EPA extrapolated the daily oral HEDs to inhalation human equivalent
concentrations (HECs) using a human body weight and breathing rate relevant to a continuous exposure
of an individual at rest, as follows:

Equation Apx J-3. Extrapolating from Oral HED to Inhalation HEC

_ „rn	w /• BWH

HECoaily, continuous ~ H E D Dauy X (—	)

i * £j L/q

Where:

HECoaily continuous = Inhalation HEC based on continuous daily exposure (mg/m3)
HEDDaiiy	= Oral HED based on daily exposure (mg/kg-day)

BWh	= Body weight of adult humans (kg) = 80

IRr	= Inhalation rate for an individual at rest (m3/hr) = 0.6125

EDc	= Exposure duration for a continuous exposure (hr/day) = 24

Based on information from	1 la). EPA assumes an at rest breathing rate of 0.6125 nrVhr.

Adjustments for different breathing rates required for individual exposure scenarios are made in the
exposure calculations, as needed.

It is often necessary to convert between ppm and mg/m3 due to variation in concentration reporting in
studies and the default units for different OPPT models. Therefore, EPA presents all PODs in
equivalents of both units to avoid confusion and errors. Equation Apx J-4 presents the conversion of the
HEC from mg/m3 to ppm.

Page 530 of 572

-------
13965

13966

13967

13968

13969

13970

13971

13972

13973

13974

13975

13976

13977

13978

13979

13980

13981

13982

13983

13984

13985

13986

13987

13988

13989

13990

13991

13992

13993

13994

13995

13996

13997

13998

13999

14000

14001

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

EquationApx J-4. Converting Units for HECs (mg/m3 to ppm)

mg 24.45
X ppm = Y —5- x

m3 MW

Where:

24.45 = Molar volume of a gas at standard temperature and pressure (L/mol), default
MW = Molecular weight of the chemical

J.3.2.4 TCEP Non-cancer HED and HEC Calculations for Acute Exposures

Moser et al. (2015) identified neurotoxicity in pregnant female rats at 125 mg/kg-day via oral gavage in
a prenatal study. The POD is based on a NOAEL of 40 mg/kg-day (tremors within a few days of
dosing). EPA used Equation Apx J-l to determine a DAF specific to rats (0.236), which was in turn
used in the following calculation of the daily HED using Equation Apx J-2:

mq mq
9.46 		— = 40-	— X 0.236

kg — day kg — day

EPA then calculated the continuous HEC for an individual at rest using Equation Apx J-3:

mg	mg	80 kg

51.5 —^ = 9.46-	x(-	,	,

m	kg day 0.6125 * 24 hr

hr

Equation Apx J-4 was used to convert the HEC from mg/m3 to ppm:

mg 24.45

4.41 ppm = 51.5 —- x	—

HH	m3 285

J.3.2.5 TCEP Non-cancer HED and HEC Calculations for Short-Term and Chronic
Exposures

Chen et al. (2015a) identified decreased numbers and degeneration of seminiferous tubules in male mice
in a 35-day study in which TCEP was administered in the diet. This endpoint is directly applicable to
short-term exposure scenarios and because it is more sensitive than endpoints from the chronic studies,
EPA also uses it for chronic exposure scenarios. The POD is based on a BMDLs of 21.0 mg/kg-day.
EPA used Equation Apx J-l to determine a DAF specific to rats, which was in turn used in the
following calculation of the daily HED using Equation Apx J-2:

mq	mq

2.79—— = 21.0—— X 0.133

kg	kg

EPA then calculated the continuous HEC for an individual at rest using Equation Apx J-3:

, 80 kg
15.2 mg/m3 = 2.79 mg/kg x (	^	)

0.6125 * 24 hr
hr

Page 531 of 572

-------
14002

14003

14004

14005

14006

14007

14008

14009

14010

14011

14012

14013

14014

14015

14016

14017

14018

14019

14020

14021

14022

14023

14024

14025

14026

14027

14028

14029

14030

14031

14032

14033

14034

14035

14036

14037

14038

14039

14040

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

EquationApx J-4 was used to convert the HEC from mg/m3 to ppm:

ma 24.45
1.30 ppm = 15.2 —- x	—

HH	m3 285

J.3.3	Cancer Dose-Response Modeling	

EPA concludes that TCEP is likely to be carcinogenic to humans based on considerations outlined in
U.S. EPA's Guidelines for Carcinogen Risk Assessment (	i. 2005b). EPA modeled the dose

response for the target organ with the most robust data - kidney tumors. For tumors in several other
target organs, see the evidence integration tables in Appendix K.

J.3.3.1 Calculating Daily Oral Cancer Slope Factors (CSFs)

Like non-cancer data, all cancer data are obtained from oral animal toxicity studies (	).

Because an MOA has not been established for TCEP, EPA assumed linear low dose extrapolation (U.S.
El 35b). EPA conducted BMD modeling of kidney tumors for both male and female rats to obtain
the CSF for TCEP (U.S. EPA. 2023b). EPA adjusted the CSF using the DAF (see Equation Apx J-l) to
account for allometric scaling between species. Equation Apx J-5 shows the calculation to obtain the
DAF-adjusted CSF:

Equation Apx J-5. Daily Oral Cancer Slope Factor (CSF)

C S ^Human,Daily ~ ^Animal, Daily /DAF

Where:

CSFnuman,Daily = Human equivalent daily oral cancer slope factor (mg/kg-day ')

CSFAnimai, Daily = Animal daily oral cancer slope factor (mg/kg-day ')
DAF	= Dosimetric adjustment factor (unitless)

Because EPA has not concluded that TCEP acts via a mutagenic MOA, an age-dependent adjustment
factor (ADAF) (	)5c) was not applied. EPA did not use CSFs for combined tumors (across

multiple target organs) for the risk evaluation but focused on the tumors with the most robust evidence
from the animal data.

J.3.3.2 Use of Oral CSF as Dermal CSF

The BW34 approach is only recommended for oral toxicity data extrapolation, and there is no established
guidance for dosimetric adjustments of dermal PODs. In the absence of available guidance, and when
the dermal CSFs are extrapolated from oral CSFs that incorporated BW3 4 scaling, EPA uses the oral
CSF for the dermal route of exposure because it has already been converted to a human dose. EPA
accounts for dermal absorption in the dermal exposure estimate, which can then be directly compared to
this HED. Sections 5.1.2.2.3 and 5.1.3.3.2 describe how EPA uses dermal absorption in calculations of
consumer exposure and exposure to soil, respectively; Section 5.1.1.3 describes dermal exposure for
workers; and Section 5.1.3.3.1 describes dermal exposure from swimming (an infinite, nondepletable
source).

Page 532 of 572

-------
14041

14042

14043

14044

14045

14046

14047

14048

14049

14050

14051

14052

14053

14054

14055

14056

14057

14058

14059

14060

14061

14062

14063

14064

14065

14066

14067

14068

14069

14070

14071

14072

14073

14074

14075

14076

14077

14078

14079

14080

14081

14082

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

J.3.3.3 Extrapolating to Inhalation Unit Risks (IURs)

For the inhalation route, EPA extrapolated the daily oral HEDs to inhalation HECs using a human body
weight and breathing rate relevant to a continuous exposure of an individual at rest. For this draft risk
evaluation, EPA assumes absorption for oral and inhalation routes is equivalent and no adjustment was
made when extrapolating from the oral to the inhalation route. The equation to convert to the inhalation
route is as follows:

EquationApx J-6. Extrapolating from the Oral CSF to an Inhalation IUR

I URnuman,continuous ~ CSFHuman,daily ^ ("

IRn * EDr

Where:

IURtluman, continuous

CSFHuman, daily

IRr

EDc

BWh

BW,



H

Human equivalent continuous daily inhalation unit risk ((mg/m3) ')
Human equivalent daily oral cancer slope factor (mg/kg-day ')
Inhalation rate for an individual at rest (m3/hr) = 0.6125
Exposure duration for a continuous exposure (hr/day) = 24
Body weight of adult humans (kg) = 80

Based on information presented in	'1 la). EPA assumes an at rest breathing rate of 0.6125

m3/hr.

EPA may need to convert between mg/m3 and ppm due to variation in concentration reporting in studies
and the default units for different OPPT models. Therefore, all PODs are presented in equivalents of
both units to avoid confusion and errors. Equation Apx J-7 identifies how to convert the IUR from
(mg/m3)-1 to (ppm)-1.

Equation Apx J-7. Converting Units for IURs (mg/m3 to ppm)

mg MW
X per ppm = Y per —- x ttt-tz
?n

Where:

24.45 = Molar volume of a gas at standard temperature and pressure (L/mol), default
MW = Molecular weight of the chemical

J.3.3.4 TCEP CSF and IUR Calculations for Lifetime Exposures

The most sensitive CSF was estimated as a risk of 0.0058 per mg/kg-day using BMD modeling software
to model the dose-response for renal tubule adenomas and carcinomas in male rats from the NTP
(1 2-year cancer bioassay. EPA then used this CSF and the rat-specific DAF (0.24) (Equation Apx
J-l) to obtain a human relevant CSF using Equation Apx J-5. The calculations specific to TCEP are as
follows:

mq	mq

0.0245 per= 0.0058 per/0.236
kg	kg

Page 533 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

14083	Using EquationApx J-6, EPA converted the oral CSF to an IUR:

14084

ma	0.6125 m3/hr * 24 hr

14085	0.00451 per	= 0.0245 per mg/kg x (			)

m3	80 kg

14086

14087	EPA used Equation Apx J-7 to convert the IUR from units of mg/m3 to ppm:

14088

14089

mg 285

14090	0.0526 per ppm = 0.00451 per —7 x

ms 24.45

14091

Page 534 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

14092	Appendix K EVIDENCE INTEGRATION FOR HUMAN HEALTH

14093	OUTCOMES

This appendix presents evidence integration tables for the major health outcomes associated with TCEP
(see TableApx K-l through TableApx K-6). It also presents a section with short evidence integration
summaries for health outcomes with limited data (Section K.2).

Page 535 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

14097	K.l Evidence Integration Tables for Major Human Health Hazard Outcomes

14098

14099	Table Apx K-l. Evidence Integration for Neurotoxicity			

Database Summary

Factors that Increase
Strength

Factors that
Decrease Strength

Summary of Key
Findings and Within-
stream Strength of the
Evidence Judgment

Inferences across
Evidence Streams and

Overall Evidence
Integration Judgment

Evidence integration summary judgment on neurotoxicity

Evidence in studies of exposed humans considered for deriving toxicity values (none)

Overall judgment for
neurological/
behavioral effects based
on integration of
information across
evidence streams:
Evidence indicates that
TCEP likely causes
neurological/
behavioral effects in
humans under relevant
exposure circumstances.

Evidence from in vivo mammalian animal studies considered for deriving toxicity values

NTP studies (Matthews et al.. 1993; NTP. 1991b:
Matthews et al.. 1990). Rats and mice exposed bv
gavage; evaluated brain/hippocampal lesions, clinical
signs of toxicity, serum cholinesterase activity. Overall
quality determination: High

Brain/hiDDOcaniDal lesions (historatholoev) (16 weeks,
and two years [rats only])

•	Female rats: brain weight decrease observed at the
highest dose.

•	Male rats: necrosis of the neurons of the
hippocampus,

•	Female rats: necrosis of the neurons of the
hippocampus. Neuronal necrosis was also observed
in the thalamus.

•	Female rats: in over 40% of female rats receiving
the highest dose showed focal gliosis, hemorrhage,
mineralization, and pigmentation, and hemosiderin
in the brain stem and cerebellum after 2 years.

Clinical sisns of toxicity (16 davs. andl6 weeks)

•	Female rats: occasionally appeared hyperactive and
exhibited resistance to handling. Seizures were
observed during week 12 of dosing.

•	Male rats: no clinical signs of toxicity were
observed in male rats.

Effect size/orecision:

•	Histopathology, serum
cholinesterase activity,
behavioral changes in
female rats were
significantly increased
over controls.

Dosc-rcsDonsc eradient:

•	Decrease in serum
cholinesterase activity
appears to increase with
dose in female rats.
Incidences of brain
histopathology findings
increased with dose in
male and female rats.

Consistency:

•	Brain weight,
brain/hippocampal
lesions, clinical signs of
toxicity, serum
cholinesterase activity,
and behavioral findings
were observed in
female rats across
different studies.

Consistency:

• Effects seen
primarily in female
rats

Key findings'.

Results across available
animal toxicological
studies showed
neurotoxicity in female
rats in a dose-response
manner. Effects do not
suggest increased severity
or frequency after
developmental exposure.
Overall judgment for
neurotoxicity based on
animal evidence:
• Robust

Page 536 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Database Summary

Factors that Increase
Strength

Factors that
Decrease Strength

Summary of Key
Findings and Within-
strcam Strength of the
Evidence Judgment

Inferences across
Evidence Streams and

Overall Evidence
Integration Judgment

•	Male and female mice exhibited convulsive
movements and reduced ability to keep balance
during the first three days of dosing at the two
highest doses.

Serum cholinesterase activity

•	Female rats: serum cholinesterase activity was
decreased at the highest doses after 14 days.

•	Female rats: serum cholinesterase activity in female
rats receiving the higher were 75% and 59%,
respectively, of the control animals. The 88
mg/kg/day animals were decreased 9.3% compared
to control animals.

•	There were no treatment-related effects on serum
cholinesterase activity in both male and female
mice

Tilson et al. (1990). 1-dav savase studv in rats:
evaluated hippocampal lesions and behavioral
findings. Overall quality determination: High

•	Treatment produced consistent damage to CA1
pyramidal cells with lesser damage to CA4, CA3,
and CA2 pyramidal cells. Significant damage was
also seen in dentate granule cells.

•	Treated rats were mildly impaired in the acquisition
of the water maze task that had a reference memory
component. However, in the repeated acquisition
task, the rats were clearly deficient.

Yang et al. (2018a). 60-dav savase studv in rats:

evaluated clinical signs of toxicity hippocampal

lesions, and behavioral findings. Overall quality

determination: High

Clinical sisns of toxicity

Coherence across
endDoints:

• Signs of neurotoxicity
and neurobehavioral
effects corresponded
to histopathology
changes in female
rats.







Page 537 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Database Summary

Factors that Increase
Strength

Factors that
Decrease Strength

Summary of Key
Findings and Within-
strcam Strength of the
Evidence Judgment

Inferences across
Evidence Streams and

Overall Evidence
Integration Judgment

•	Occasional periods of hyperactivity and periodic
convulsions in female rats. There were not
treatment-related effects observed in male rats

Behavioral findings

•	Remarkably higher escape latencies to find the
hidden platform than the vehicle controls (p <
0.01). Significantly shorter cumulative distances
from the original platform than the controls.
Significantly fewer cross-times were noted in the
highest dose for female rats. Male rats were not
tested.

Hazleton Laboratories (1983). A single dose during
GD 7-14. Overall quality determination: High

•	There was a low incidence of maternal animals
with clinical signs of OP toxicity (up to 2/50
animals on GD 7-14).

Developmental Neurotoxicity.

Moser et al. (2015) Overall aualitv determination:

High

Assessment of neurobehavioral and related hormonal
responses after dosing pregnant Long-Evans rats from
GD 10 through PND 22 via oral gavage up to 90
mg/kg-day No TCEP-related adverse effects in T3, T4,
brain or serum AChE in dams or offspring. In addition,
no effects on brain weight in offspring at PND 6 and
sporadic behavioral changes do not suggest
biologically relevant adverse outcomes or
developmental toxicity.









1 ideuce in mechanistic studies and supplemental information

In vivo:

Yang et al. (2018a). Compared to those in the control,
the major metabolites that had increased in the aqueous

• None

• None

Overall judgment for
neurotoxicity based on
mechanistic evidence:

Page 538 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Database Summary

Factors that Increase
Strength

Factors that
Decrease Strength

Summary of Key
Findings and Within-
strcam Strength of the
Evidence Judgment

Inferences across
Evidence Streams and

Overall Evidence
Integration Judgment

phase of TCEP-treated groups were N-acetyl aspirate
(NAA), glutamine (GLU), glutamic acid, glucose,
taurine, choline, creatine, and myo-inositol levels,
whereas those that had decreased were lactate, g-amino
butyric acid (GABA), glycine, and two unknown
compounds. In the lipid phase, the major metabolites
that were different between the control and TCEP-
treated groups were cholesterol ester and glycerol,
which were increased, whereas free cholesterol, total
cholesterol, lipid (CH2CH2CO), fatty acid,
polyunsaturated fatty acid, and phosphatidylcholine
levels were decreased.





• Indeterminate



14100

Page 539 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table Apx K-2. Evidence Integration for Reproductive Effeci

ts

Database Summary

Factors that Increase Strength

Factors that Decrease Strength

Summary of Key
Findings and
Within-stream
Strength of the
Evidence Judgment

Inferences across
Evidence Streams and

Overall Evidence
Integration Judgment

Evidence integration summary judgment on reproductive effects

Evidence in studies of exposed humans considered for deriving toxicity values (none)

Overall judgment for
reproductive effects
based on integration of
information across
evidence streams:

Evidence indicates that
TCEP exposure likely
causes reproductive
effects in humans under
relevant exposure
circumstances.

Evidence from apical endpoints in in vivo mammalian animal studies considered for deriving toxicity values

•	Short-term, subchronic, and
chronic gavage studies in male and
female rats and mice and a
subchronic dietary study in male
mice examined testes weight and/or
histology of the reproductive
orsans NTP (1991b) and Chen et
al. (2015a). Overall aualitv
determination: High

•	The Reproductive Assessment by
Continuous Breeding (RACB)
Protocol54 was used to evaluate
fertility, litters/pair, live pups/litter,
proportion of pups born alive, sex
of live pups, pup weights at birth,
sperm morphology, vaginal
cytology, and/or reproductive organ
weights and histology in mice
treated via savase (NTP. 1991a).
Overall quality determination:

High.

•	Biolosical eradient/dose-

rcsdorise: The magnitude and
severity of histological changes
in the testes (changes in the
number and appearance of
seminiferous tubules)
increased with increasing dose
in the subchronic dietary study
in ICR mice.

•	Fertility index, number of
litters/pair decreased in a dose-
related manner during the
continuous F0 breeding phase
of the RACB.

Consistencv:

•	Decreased testes weight was
observed in gavage and dietary
subchronic studies in mice.

•	Decreased fertility index was
observed during continuous F0
breeding and crossover mating
phases of the RACB.

•	Sperm effects (decreases on
sperm concentration and
percent motile sperm,
increased sperm abnormalities)
identified during crossover

Consistencv:

•	Changes in testes histology were
observed in a subchronic dietary
study in ICR mice, but no
histological changes to
reproductive organs were
observed in short-term,
subchronic, or chronic gavage
studies in F344 rats and CD-I
andB6C3Fl mice.

Oualitv of the database:

•	Testes weights were assessed in
subchronic, but not chronic, NTP
studies in rats and mice.

Key findings'.
Available animal
toxicological studies
showed decreased
testes weight,
histological changes
in the testes of ICR
mice, sperm effects,
and/or reduced
fertility and
fecundity.

Overall judgment for
reproductive effects
based on animal
evidence:
• Moderate

54 The RACB protocol consists of 4 phases: (1) dose range-finding, (2) continuous (F0) breeding, (3) crossover mating; and (4) assessment of fertility in F1 offspring.

Page 540 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Database Summary

Factors that Increase Strength

Factors that Decrease Strength

Summary of Key
Findings and
Within-stream
Strength of the
Evidence Judgment

Inferences across
Evidence Streams and

Overall Evidence
Integration Judgment



mating correlated with
decreased fertility index when
treated males were bred with
untreated females.

•	Mechanistic changes from in
vivo and in vitro studies
(decreased testicular
testosterone, altered gene
expression related to
steroidogenesis, and decreased
testosterone secretion) are
consistent with observed
effects on testes and sperm.

Oualitv of the database:

•	Effects were observed in high-
quality studies.







l\\ idence mi mcdianMic sludicsaiid Mipplcmcnlal iiil'iimialkiii

•	A subchronic dietary study in male
mice evaluated testicular
testosterone and gene expression
related to testosterone synthesis
(Chen et aL 2015a").

•	An in vitro study using TM3
Leydig cells evaluated testosterone
secretion and gene expression
related to steroidogenesis and
oxidative stress ("Chen et al..

2015b).

•	Three in vitro studies evaluated
estrogenic, anti-estrogenic,
androgenic, and/or anti-androgenic
activity using a yeast reporter assay
or human (endometrial, prostate
and breast) cancer cell lines
(Krivoshiev et aL, 2016; Reers et

Bioloeical eradient/dose-

rc sdo rise:

•	In vivo data showed decreased
testicular testosterone and
altered gene expression related
to testosterone synthesis at the
dose in which decreased testes
weight and testicular damage
were observed.

•	An in vitro study showed
decreased testosterone
secretion and/or changes in
gene expression related to
steroidogenesis and oxidative
stress at both tested
concentrations.

Consistency:

•	Altered gene expression

Consistency:

•	There was inconsistency across
studies with respect to estrogen
receptor and androgen receptor
agonist and/or antagonist activity
in human (endometrial, prostate,
and breast) cancer cell lines.

Oualitv of the database:

•	Few potential mechanisms were
investigated in available studies.

Bioloeical olausibilitv/relevance to

humans:

•	Oxidative stress is a nonspecific
mechanism.

Key findings: Limited
available mechanistic
data indicate that
TCEP may induce
oxidative stress and
endocrine disruption
via altered expression
of genes involved in
steroidogenesis.
Overall judgment for
reproductive effects
based on mechanistic
evidence:

• Slight

Page 541 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Database Summary

Factors that Increase Strength

Factors that Decrease Strength

Summary of Key
Findings and
Within-stream
Strength of the
Evidence Judgment

Inferences across
Evidence Streams and

Overall Evidence
Integration Judgment

al., 2016; Follinann and Weber,
2006).

related to steroidogenesis
correlated with decreased
testosterone in vivo and in
vitro.

Bioloeical olausibilitv/relevance
to humans:

• Endocrine disruption, via
altered expression of genes
involved in testosterone
synthesis, is a plausible
mechanism for infertility,
sperm effects, and testicular
damage that is relevant to
humans.







GD = gestation day

14102

Page 542 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

14103 Table Apx K-3. Evidence Integration for Developmental Effects

Database Summary

Factors that Increase Strength

Factors that Decrease Strength

Summary of Key
Findings and within-
Stream Strength of
the Evidence
Judgment

Inferences across
Evidence Streams and

Overall Evidence
Integration Judgment

Evidence integration summary judgment on developmental effects

Evidence in studies of exposed humans considered for deriving toxicity values (none)

Overall judgment for
developmental effects
based on integration of
information across
evidence streams:
Evidence indicates that
TCEP exposure likely
causes developmental
effects in humans under
relevant exposure
circumstances.

Evidence from apical endpoints in in vivo mammalian animal studies considered for deriving toxicity values

•	An oral gavage study evaluated
uterine parameters, number of
pups, pup weight, and viability
following gestational exposure
(GDs 7-14) in female mice
(Hazleton Laboratories. 1983).
Overall quality determination: High

•	Assessment of neurobehavioral and
related hormonal responses after
dosing pregnant Long-Evans rats
from GD 10 through PND 22 via
oral gavage up to 90 mg/kg-day.
No adverse effects in T3, T4, brain
or serum ACliE in dams or
offspring. No effects on brain
weight in offspring at PND 6.
Sporadic behavioral changes do not
suggest biologically relevant
adverse outcomes or developmental
toxicity. Moseret al. (2015).

Overall quality determination: High

•	Biolosical eradient/dose-

rcsdorise: number of littcrs/oair
and number of live pups/litter
decreased in a dose-related
manner during the continuous
F0 breeding phase of the
RACB.

•	Supporting reproductive
effects: Magnitude and severity
of testes histological changes
increased with dose in the
subchronic dietary study in
ICR mice.

Consistency:

•	Decreased numbers of live
pups/litter were observed
during continuous F0 breeding
and crossover mating phases of
the RACB.

•	Decreased number of live
pups/litter was observed at the
same dose in F0 and F1
breeding phases of the RACB,
with greater severity in the
second generation.

Consistency of suDDortina

reproductive effects:

•	Decreased testes weight was
observed in gavage and dietary
subchronic studies in mice.

Magnitude and precision:

•	The developmental gavage
studies in mice used only one
dose group and no
developmental effects were
observed.

•	The developmental neurotoxicity
study in rats did not result in
effects in offspring.

Key findings'.
Available animal
toxicological studies
resulted in decreased
live pups per litter.
Overall judgment for
developmental effects
based on animal
evidence:

• Moderate

Page 543 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Database Summary

Factors that Increase Strength

Factors that Decrease Strength

Summary of Key
Findings and within-
Strcam Strength of
the Evidence
Judgment

Inferences across
Evidence Streams and

Overall Evidence
Integration Judgment



•	Sperm effects identified during
crossover mating correlated
with decreased fertility index
when treated males were bred
with untreated females.

•	Mechanistic changes from in
vivo and in vitro studies
(decreased testosterone,
altered steroidogenesis gene
expression) consistent with
effects on testes and sperm.

Oualitv of the database:

•	Effects were observed in high-
quality studies.







l\\ idence mi incdiaiiislic sludicsaiid siipplcincnl;il iiil'iimialkiii

•	Yonemoto et al. (1997) evaluated
inhibitory concentrations for cell
proliferation (IP50) and
differentiation (ID50) in rat embryo
limb bud cells.

•	Reproductive: A subchronic dietaiy
study in male mice evaluated
testicular testosterone and gene
expression related to testosterone
synthesis (Chen et al. 2015a).

•	An in vitro study using TM3
Leydig cells evaluated testosterone
secretion and gene expression
related to steroidogenesis and
oxidative stress (Chen et al.

2015b).

•	Three in vitro studies evaluated
estrogenic, anti-estrogenic,
androgenic, and/or anti-androgenic
activity using a yeast reporter assay

Bioloeical eradient/dose-response

(reproductive effects):

•	In vivo data showed decreased
testicular testosterone and
altered gene expression related
to testosterone synthesis at the
dose in which decreased testes
weight and testicular damage
were observed.

•	An in vitro study showed
decreased testosterone
secretion and/or changes in
gene expression related to
steroidogenesis and oxidative
stress at both tested
concentrations.

Consistency (Reproductive):

•	Altered gene expression
related to steroidogenesis
correlated with decreased

Consistency (Reproductive):

•	There was inconsistency across
studies with respect to estrogen
receptor and androgen receptor
agonist and/or antagonist activity
in human (endometrial, prostate,
and breast) cancer cell lines.

Oualitv of the database:

•	Few potential mechanisms were
investigated in available studies.

Bioloeical plausibility/relevance to

humans:

•	Oxidative stress is a possible
nonspecific mechanism.

Key findings: Limited
available mechanistic
data indicate that
TCEP may induce a
ratio of inhibition of
proliferation and
differentiation
resulting in concern
for development;
oxidative stress; and
endocrine disruption
via altered expression
of genes involved in
steroidogenesis.
Overall judgment for
developmental effects
based on mechanistic
evidence:

• Slight

Page 544 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Database Summary

Factors that Increase Strength

Factors that Decrease Strength

Summary of Key
Findings and within-
Strcam Strength of
the Evidence
Judgment

Inferences across
Evidence Streams and

Overall Evidence
Integration Judgment

or human (endometrial, prostate
and breast) cancer cell lines
(Krivoshiev et aL 2016; Reers et

aL 2016; Follmann and Wober,

testosterone in vivo and in
vitro.

Bioloeical olausibilitv/relevance
to humans:







2006).

•	Yonemoto et al. (1997)
identified an IP5o of 3600 |iIVI
of TCEP using rat embryo
limb bud cells. The ID5o was
1570 nM; the ratio of
concentrations suggested
possible developmental
toxicity.

•	Reproductive: Endocrine
disruption, via altered
expression of genes involved
in testosterone synthesis, is a
plausible mechanism for
infertility, sperm effects, and
testicular damage that is
relevant to humans.

GD = gestation day

14104

Page 545 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

14105 Table Apx K-4. Evidence Integration Table for Kidney Effects

Database Summary

Factors that Increase
Strength

Factors that Decrease
Strength

Summary of Key
Findings and
within-Stream
Strength of the
Evidence Judgment

Inferences across
Evidence Streams and

Overall Evidence
Integration Judgment

Evidence integration summary judgment on kidney effects

Evidence in studies from exposed humans for deriving toxicity values (none)

Overall judgment for
renal effects based on
integration of
information across
evidence streams:

Evidence indicates that
TCEP exposure likely
causes kidney effects in
humans under relevant
exposure circumstances.

Evidence from in vivo mammalian animal studies considered for deriving toxicity values

NTP (1991b): Rats and mice cxooscd bv eavaee:

evaluated kidney weights and histopathology. Overall

quality determination: High

Kidnev weiehts (16 davs. 16 weeks, and 66 weeks 1 rats

only])

•	Male rats: increased kidney weights at all time
points.

•	Female rats: no change after 16 days, dose-related
increases in kidney weights after 16 weeks, and no
change after 66 weeks.

•	Male mice: no change after 16 days and decreased
kidney weight after 16 weeks.

•	Female mice: increased kidney weight after 16 days
and no change after 16 weeks.

Histooatholoev (16 davs. 16 weeks, and 104 weeks)

•	No changes in rats or mice after 16 days or in rats
after 16 weeks.

•	Male rats: renal tubule hyperplasia and renal tubule
adenomas after 104 weeks at 88 mg/kg/day; one
adenoma occurred as early as 66 weeks at 88
mg/kg/day; increase in combined adenomas or
carcinomas at 88 mg/kg/day (see also TableApx
K-6 for cancer endpoints).

•	Female rats: renal tubule hyperplasia and renal tubule
adenomas after 104 weeks at 88 mg/kg/day (see also
Table Apx K-6 for cancer endpoints).

•	Male mice: epithelial cytomegaly after 16 weeks at
700 mg/kg-day; karyomegaly after 104 weeks at

Effect sizc/orecision:
• Histopathology
changes in rats and
mice of both sexes
were significantly
increased over controls
by both pairwise and
trend tests.
Dosc-rcsDonsc eradient:
Incidences of kidney
histopathology findings
increased with dose in rats
and mice of both sexes.
TcniDoralitv:
Histopathology findings
were more prevalent and
occurred at lower doses as
exposure duration
increased.

Consistencv:

Renal histopathology
changes were observed in
rats and mice of both
sexes and in studies in
two different laboratories.
Coherence across
endDoints:

Kidney weight changes
corresponded to

Inconsistencv
Kidney weight changes
did not occur at all time
points in female rats or
mice of either sex.
Incoherence:

Kidney weight changes
did not correspond to
histopathology changes in
female rats or mice of
either sex.

Imprecision:

•	Dosing errors occurred
in 16-week studies in
rats and mice.

•	Treatment-related
deaths occurred in 16-
week study in rats.

•	Survival was
decreased at the high
dose in both sexes of
rat in 104-week study.

Key findings'.

Results across
available animal
toxicological studies
showed renal toxicity
in rats and mice.
Overall judgment for
renal effects based on
animal evidence:
• Moderate

Page 546 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Database Summary

Factors that Increase
Strength

Factors that Decrease
Strength

Summary of Key
Findings and
within-Strcam
Strength of the
Evidence Judgment

Inferences across
Evidence Streams and

Overall Evidence
Integration Judgment

>	175 mg/kg-day; one adenocarcinoma and three
adenomas at 350 mg/kg-day (see also Table Apx
K-6 for cancer endpoints).

• Female mice: epithelial cytomegaly after 16 weeks at
700 mg/kg-day; karyomegaly after 104 weeks at

>	175 mg/kg-day.

Taniai et al. (2012a) 28-dav eavaee studv in rats:
evaluated histopathology. Overall quality determination:
Medium

Historatholoev

Male rats: scattered proximal tubular regeneration in the
cortex and outer stripe of the outer medulla (OSOM) at
350 mg/kg-day.

histopathology changes in
male rats.







1 idence in mechanistic studies and supplemental mlonniilion

In vivo:

Markers for cell proliferation and aooDtosis (Taniai et
al. 2012b") and reeeneratine tubules (Taniai et al.
2012s) were increased in kidnevs (OSOM and cortex)
of rats after 28 days (gavage)

In vitro:

TCEP exposure of primary rabbit renal proximal tubule
cells (PTCs) resulted in cytotoxicity, reduced DNA
synthesis, altered expression of cell cycle regulatory
proteins, and inhibition of ion- and non-ion-transport
functions. Increased expression of pro-apoptotic
regulatory proteins and decreased expression of proteins
that inhibit aoootosis were also observed (Ren et al..
2012; Ren et al, 2009, 2008).

Dose response sradient:
Across the in vitro
studies, dose-related
changes in the endpoints
were observed.

Consistent with related
aoical endooints: Results
from mechanistic studies
are consistent with in vivo
histopathology findings in
the renal tubules.

IniDrccision/Inconsistcnc

•	There are few studies of
mechanistic endpoints
in the kidneys.

•	In vitro studies used
only one cell model and
all were conducted in
the same laboratory.

Key findings'.
Apoptosis and altered
cell cycle regulation
may contribute to
renal effects of TCEP
in animals.

Overall judgment for
renal effects based on
mechanistic
evidence:

• Slight

Page 547 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

14106 Table Apx K-5. Evidence Integration Table for Liver Effects

Database Summary

Factors that Increase Strength

Factors that Decrease Strength

Summary of Key
Findings and
Within-strcam
Strength of the
Evidence Judgment

Inferences across
Evidence Streams and

Overall Evidence
Integration Judgment

1 !\ idciiee iiilcmnlkiu suiiininiA iiiduuiciil on li\cidTccls



1 :\ idcuec mi studies i)l"c.\pi)scd 1 li111kiiis considered fur den\ iiiu u<\ieil\ \ nines (in>iiei
1 idenee IVoni apical eiidpoiiils in in vi\" ni;inini;ili;ni ;imni;il studies Ibr deri\ um lo\icil\ \ nines

• liidclcrnuunlc

Overall judgment for
liver effects based on
integration of
information across
evidence streams:

Evidence suggests but is
not sufficient to
conclude that TCEP
causes hepatic effects in
humans under relevant
exposure circumstances.

•	NTP (1991b): Subchronic and
chronic gavage studies in rats and
mice that examined liver weights,
clinical chemistry, and
histopathology. Overall quality
determination: High

•	One 35-day dietary exposure study
in male mice that examined liver
weishts (Chen et al. 2015a).
Overall quality determination: High

Bioloeical eradient/dose-

rc sdo rise:

•	A dose-related trend in
hepatocellular adenoma was
observed in male mice in the
chronic study.

•	Increases in liver weights in
male rats occurred at lower
doses as duration increased.

•	Dose-related increases in liver
weights were seen in female
rats and female mice at 16
weeks and in male rats at 66
weeks.

Oualitv of the database:

•	Effects observed in high-
quality studies.

Magnitude and precision:

•	The incidence of eosinophilic
foci in male mice was
statistically significantly
increased at only the top dose
after 2 years.

Consistency:

•	There were no histopathology
findings in rats or female mice,
including no hypertrophy.

•	Liver weight increases were seen
in female rats after 16 days and
16 weeks, but not 66 weeks of
exposure.

•	Increased liver weight was not
seen in the 35-day study.

•	No biologically relevant changes
in serum enzymes were seen in
the 2-year bioassay and not
measured in shorter studies.

Oualitv of the database:

•	Liver weights were not assessed
in mice exposed longer than 16
weeks.

Key findings'.
Available animal
toxicological studies
showed increased
liver weights in rats
and mice in the
absence of relevant
clinical chemistry
findings;
histopathology
changes in the liver
were observed only in
male mice.

Overall judgment for
liver effects based on
animal evidence:
• Slight

Page 548 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Database Summary

Factors that Increase Strength

Factors that Decrease Strength

Summary of Key
Findings and
Within-strcam
Strength of the
Evidence Judgment

Inferences across
Evidence Streams and

Overall Evidence
Integration Judgment

l\\ idence in mechanistic studies and supplemental information



•	One in vivo 3 5-day dietary
exposure study in male mice
examining markers of oxidative
stress (Chen et al. 2015a).

•	Five in vitro studies examining
viability, cell cycle, cellular and
mitochondrial oxidative stress,
mitochondrial function, and cell
signaling pathways in liver cells
and/or cell lines (Mennillo et al..
2019; Zhang et al.. 2017b; Zhang et
al.. 2017a; Zhang et al.. 2016c;

Bioloeical eradient/dose-

rc sdo rise:

•	In vivo data showed induction
of hepatic oxidative stress
occurring earlier than apical
endpoints.

•	Across the in vitro studies,
dose-related changes in
viability, oxidative stress, and
impaired mitochondrial
functioning were observed.

Bioloeical olausibilitv/relevance

Oualitv of the database:

•	Few potential mechanisms were
investigated in available studies.

Bioloeical eradient/dose response:

•	Oxidative stress was
demonstrated in vivo at higher
doses than those associated with
liver lesions in chronic study.

Bioloeical olausibilitv/relevance to

humans:

•	Oxidative stress is a nonspecific
mechanism and was seen only at
doses higher than those
associated with liver lesions.

Key findings: Limited
available mechanistic
data indicate that
TCEP may induce
oxidative stress, alter
cellular energetics,
and/or influence cell
signaling related to
proliferation, growth,
and survival in the
liver.

Overall judgment for
liver effects based on
mechanistic evidence:
• Slight

Zhang et al.. 2016b").

to humans:

• Oxidative stress is a plausible
mechanism for eosinophilic
foci and tumor formation that
is relevant to humans.

14107

14108

Page 549 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

14109 Table Apx K-6. Evidence Integration Table for Cancer

Database Summary

Factors that Increase
Strength

Factors that Decrease
Strength

Summary of Key
Findings and
within-Stream
Strength of the
Evidence Judgment

Inferences across
Evidence Streams and

Overall Evidence
Integration Judgment

Evidence integration summary judgment on cancer

Evidence in studies of exposed humans considered for deriving toxicity values

Overall judgment for
cancer effects based on
integration of information
across evidence streams:
EPA concludes that
TCEP is likely to be
carcinogenic to humans
using guidance from U.S.
EPA's Guidelines for
Carcinogen Risk
Assessment (U.S. EPA.
2005b).

Hoffman et al. (2017)

Case-control study of thyroid cancer and
TCEP in household dust. Overall quality
determination: High

• Significant increase in adjusted OR for
TCEP (in dust) above median level among
papillary thyroid cancer cases compared to
controls. TCEP in dust in homes associated
with more aggressive tumors in sample (n
= 70 cases, 70 controls)

Biological Plausibility
• Thyroid cancers also
reported in female rats
exposed to TCEP orally.

Oualitv of the database:

•	One epidemiological study
of cancer (high-quality); no
studies of renal cancers in
humans.

Biolosical eradient/dose-
rcsDonsc:

•	Exposure was measured
after outcome.

Magnitude and Precision

•	Dust used as proxy for
TCEP exposure;
corresponding biological
samples were not collected
to match with dust samples

Key findings'.
Available
epidemiological
study of cancer was
limited.

Overall judgment for
cancer effects based
on human evidence:
• Indeterminate

Evidence from apical endpoints in in vivo mammalian animal studies

Kidney cancer

NTP (1991b): F344 rats and B6C3F1 mice
exposed by gavage for 104 weeks. Overall
quality determination: High
• Increased incidences of adenomas and
adenomas or carcinomas in male rats (one
adenoma occurred at week 66) and
increased incidences of adenomas in
female rats.

Oualitv of the database:

•	Evidence in high-quality
study in rats and mice

Magnitude and precision:

•	Significant pairwise
comparisons in male and
female rats.

•	Renal tubule tumors are
rare in F344/N rats and
B6C3F1 mice.

Biolosical eradient/dose-

rcsDonsc:

Magnitude and precision:

•	Survival was decreased at
the high dose in both sexes
of rat in 104-week study.

Consistencv:

•	No significantly increased
incidence of tumors was
seen in two strains of
female mice or in male
B6C3F1 mice.

Key findings'.
Dose-related
increased renal tumor
incidences
demonstrated in a
high-quality study in
rats of both sexes
Overall judgment for
kidney cancer effects
based on animal
evidence:

• Robust

Page 550 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Database Summary

Factors that Increase
Strength

Factors that Decrease
Strength

Summary of Key
Findings and
within-Strcam
Strength of the
Evidence Judgment

Inferences across
Evidence Streams and

Overall Evidence
Integration Judgment



•	Significant dose-related
trends in male and female
rats.

Consistency:

•	Effects seen in both sexes
of rat.









NTP (1991b): Overall aualitv determination:
High

•	Increased incidence of mononuclear cell
leukemia (MNCL) in male and female rats

•	No increased incidence of MNCL or other
hematologic cancer in male or female mice

Oualitv of the database:

•	Evidence in high-quality
studies in rats and mice.

Magnitude and precision:

•	Significant pairwise
comparisons in male and
female rats.

Biolosical sradient/dose-

rcsDonsc:

•	Significant dose-related
trends in male and female
rats.

Consistency:

•	Evidence in two sexes.

Magnitude and precision:
MNCL is common in F344
rats, its spontaneous incidence
varies widely, and incidences
in male rats exposed to TCEP
were within historical
controls.

Biolosical

olausibilitv/relevance to
humans:

Occurrence of MNCL is rare
in mice and other strains of
rats (Thomas et al. 2007).
MNCL may be similar to
large granular lymphocytic
leukemia (LGLL) in humans
(Caldwell et al. 1999;
Caldwell 1999; Reynolds and
Footi. 1984). particularly an
aggressive form of CD3- LGL
leukemia known as aggressive
natural killer cell leukemia
(ANK.CL) (Thomas et al.
2007). However, Maronpot et
al. (2016) note that ANKCL is
extremely rare with less than
98 cases reported worldwide,
and the authors contend that

Key findings'.
Dose-related
increases in MNCL
incidences
demonstrated in a
high-quality study in
rats of both sexes, but
this is a common
spontaneous cancer
in rats and only the
incidence in high
dose female rats was
outside the historical
control range.

Overall judgment for
hematopoietic system
cancer effects based
on animal evidence:
• Slight

Page 551 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Database Summary

Factors that Increase
Strength

Factors that Decrease
Strength

Summary of Key
Findings and
within-Strcam
Strength of the
Evidence Judgment

Inferences across
Evidence Streams and

Overall Evidence
Integration Judgment





ANKCL has an etiology
related to infection with
Epstein-Barr virus, not
chemical exposure.















NTP (1991b): Overall aualitv determination:

High

•	Nonsignificant increase in incidence of
follicular cell adenoma or carcinoma in
male rats.

•	Significantly increased incidences of
follicular cell carcinomas and adenoma or
carcinoma in female rats.

•	No increased incidence of thyroid tumors
in male or female mice.

Oualitv of the database:

•	Evidence in high-quality
studies in rats and mice.

Magnitude and precision:

•	Significant pairwise
comparison in female rats.

Biolosical sradient/dose-

rcsDonsc:

•	Significant dose-related
trend in female rats;
borderline significant trend
in males.

Consistency:

Effect seen in both sexes of

rats.

Magnitude and precision:

•	Survival was decreased at
the high dose in both sexes
of rat in 104-week study.

Consistency:

•	Effect seen in only one
species (rats).

Biolosical

Dlausibilitv/rclcvancc to

humans:

U.S. i and Dvbing

Key findings'.
Dose-related
increases in thyroid
follicular cell tumor
incidences were
demonstrated in a
high-quality study in
female rats. Rodents
may be more
sensitive than
humans to thyroid
follicular cell tumors.
Overall judgment for
thyroid cancer effects
based on animal
evidence:

• Slight





and Sanner (1999) concluded
that rodents are more sensitive
than humans to thyroid
follicular tumors induced by
thyroid-pituitary gland
disruption and thyroid
stimulating hormone (TSH)
hyperstimulation. NTP
(1991b) did not measure TSH
in the chronic rat study.



Page 552 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Database Summary

Factors that Increase
Strength

Factors that Decrease
Strength

Summary of Key
Findings and
within-Strcam
Strength of the
Evidence Judgment

Inferences across
Evidence Streams and

Overall Evidence
Integration Judgment





NTP (1991b): Overall aualitv determination:
High

• Increased incidence of adenoma or
carcinoma in female mice (when interim
sacrifice groups included); no increased
incidence of Harderian gland tumors in rats
or male mice.

Oualitv of the database:
• Evidence in high-quality
studies in rats and mice.

Magnitude and precision:

•	Increased incidence of
tumors in female B6C3F1
mice was statistically
significant only when
interim sacrifice groups
were included.

Bioloeical eradient/dose-

rcsDonsc:

•	Increased incidence in
female B6C3F1 mice
occurred only at highest
tested dose.

Consistency

•	No increased incidence of
tumors in male B6C3F1
mice, or rats of either sex.

Key findings'.
Increased tumor
incidence was only
seen in one sex of
one species (female
B6C3F1 mice).
Overall judgment for
Harderian gland
cancer effects based
on animal evidence:
• Slight



Page 553 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Database Summary

Factors that Increase
Strength

Factors that Decrease
Strength

Summary of Key
Findings and
within-Strcam
Strength of the
Evidence Judgment

Inferences across
Evidence Streams and

Overall Evidence
Integration Judgment





NTP (1991b): Overall aualitv determination:
High

• Dose-related trend for adenomas,
borderline significant increase in male
mice at high dose; no effects on female
mice or rats of either sex.

Oualitv of the database:

•	Evidence in high-quality
studies in rats and mice.

Biolosical eradient/dose-

rcsDonsc:

•	Significant dose-related
trend in male B6C3F1
mice.

Magnitude and precision:

•	Increased incidence of
adenomas in male B6C3F1
mice was not statistically
significant by pairwise
comparison.

Consistency

•	No increase in liver tumor
incidence in female mice or
in rats of either sex.

Key findings'.
Dose-related trend in
tumor incidence was
seen only in one sex
of one species (male
B6C3F1 mice).

Overall judgment for
liver cancer effects
based on animal
evidence:

• Slight



1 !\ idence mi niecliaiiisiic studies and supplciiiciilal iiilonniiiion



Page 554 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Database Summary

Factors that Increase
Strength

Factors that Decrease
Strength

Summary of Key
Findings and
within-Strcam
Strength of the
Evidence Judgment

Inferences across
Evidence Streams and

Overall Evidence
Integration Judgment

Genotoxicity

In vivo:

•	Weakly positive/equivocal for
micronucleus induction in Chinese
hamsters (Sala et al. 1982).

In vitro:

•	Positive for bacterial mutagenicity in one
S. typhimurium strains, and weakly
positive in another (Nakamura et al.

1979).

•	Negative for bacterial mutagenicity in
several studies using multiple strains of S.
typhimurium with and without metabolic
activation (Follmann and Wober, 2006);
negative for mutagenicity and DNA strand
breaks in hamster V79 cells (Follmann and
Wober, 2006; Sala et al, 1982).

•	Positive for SCEs in hamster V79 cells
(Sala et al. 1982) and DNA strand breaks
in human PBMCs (Bukowski et al, 2019).

•	Positive/weak positive for cell
transformation (may not be a genotoxic
mechanism) in two cell tvDes (Sala et al.
1982)

Oualitv of the database:
• Tests of bacterial
mutagenicity in multiple
strains, large concentration
range, and assays with and
without metabolic
activation.

Oualitv of the database:

•	Few studies in mammalian
cells and limited in vivo
data.

Magnitude and precision/

Biolosical eradient/dose-

rcsDonsc:

•	Few positive findings, lack
of information on
cytotoxicity in at least one
and weak/equivocal in one.

Consistencv:

•	DNA strand break findings
were not consistent across
studies/cell types.

Key findings'.
Available data
indicate that TCEP
has little genotoxic
potential. Limited
available data
indicate that TCEP
may induce oxidative
stress, alter cellular
energetics, and/or
influence cell
signaling related to
proliferation, growth,
and survival in
kidney, liver, and
blood cells.

Overall judgment for
cancer effects based
on mechanistic
evidence:

• Slight



Other (non-genotoxic) mechanistic studies"

Kidnev:

•	Markers for cell proliferation and apoptosis
(Taniai et al, 2012b) and regenerating
tubules (Taniai et al.. 2012a) were
increased in kidneys (OSOM and cortex)
of rats after 28 days (gavage)

•	TCEP exposure of primary rabbit renal
proximal tubule cells (PTCs) resulted in
cytotoxicity, reduced DNA synthesis,
altered expression of cell cycle regulatory

Biolosical eradient/dose-

rcsDonsc:

• Across the in vitro studies,
dose-related changes were
observed.

Oualitv of the database:

•	There are few studies in
relevant tissue types and
only two in vivo studies.

•	Available studies were not
directly focused on cancer
mechanisms.

Page 555 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Database Summary

Factors that Increase
Strength

Factors that Decrease
Strength

Summary of Key
Findings and
within-Strcam
Strength of the
Evidence Judgment

Inferences across
Evidence Streams and

Overall Evidence
Integration Judgment

proteins, and inhibition of ion- and non-
ion-transport functions. Increased
expression of pro-apoptotic regulatory
proteins and decreased expression of
proteins that inhibit apoptosis were also
observed (Ren et aL 2012; Ren et aL
2009. 2008).

Hematopoietic:

•	TCEP exposure of human peripheral blood
mononuclear cells resulted in cytotoxicity
(Mokra et aL 2018) and decreased DNA
methvlation (Bukowski et aL 2019).

Liver:

•	Markers of oxidative stress (hepatic
antioxidant enzyme activities and their
gene expression) were increased in the
livers of male ICR mice after 35 days of
dietary exposure to TCEP (Chen et aL
2015a).

•	Liver cells and/or cell lines cultured with
TCEP exhibited reduced viability, cell
cycle arrest, cellular and mitochondrial
oxidative stress, impaired mitochondrial
function, and perturbation of cell signaling
pathways (Mennillo et aL 2019; Zhang et
aL 2017b; Zhang et aL. 2017a; Zhang et
lik 2016c; Zhang ei.ik 2016b).









" No tissue-specific mechanistic data related to harderian gland or thyroid follicular cell cancers were identified in the available literature.

14110

Page 556 of 572

-------
14111

14112

14113

14114

14115

14116

14117

14118

14119

14120

14121

14122

14123

14124

14125

14126

14127

14128

14129

14130

14131

14132

14133

14134

14135

14136

14137

14138

14139

14140

14141

14142

14143

14144

14145

14146

14147

14148

14149

14150

14151

14152

14153

14154

14155

14156

14157

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

K.2 Evidence Integration Statements for Health Outcomes with Limited
Data

Skin and Eye Irritation

The human evidence is indeterminate for skin and eye irritation. The two readily available dermal
irritation studies in animals showed inconsistent results and the single eye irritation study of medium-
quality showed that TCEP is not irritating; these studies are indeterminate. Although one study was
uninformative, EPA considered that these results are not affected by the lack of statistical analysis.
Overall, the currently available evidence is inadequate to assess whether TCEP causes irritation in
humans.

Mortality

Human evidence is indeterminate for mortality because there are no human epidemiological studies.
There is modest evidence in animal studies that shows higher mortality in rats than mice in oral studies
at the same doses and uncertain potential for mortality via the dermal route given conflicting results.
Overall, evidence suggests but is not sufficient to conclude that TCEP exposure causes mortality in
humans under relevant exposure circumstances. This conclusion is based on oral studies in rats and mice
that assessed dose levels between 12 and 700 mg/kg-day and dermal studies in rabbits at approximately
279 and 556 mg/kg-day.

Immune/Hematological

Evidence from an epidemiological study did not identify an association between TCEP and childhood
asthma and was indeterminate for immune and hematological effects; the study evaluated only a single
type of immune effect. Animal studies did not identify histopathological changes in immune-related
organs or in hematological parameters. A statistically significant increased trend in mononuclear cell
leukemia with increasing dose was seen in rats. In mechanistic studies, TCEP was associated with
decreases in an inflammatory cytokine and altered gene expression of inflammatory proteins in two
studies, but a third study identified inflammatory changes only after co-exposure with benzo-a-pyrene.

Available evidence is indeterminate and therefore, is inadequate to assess whether TCEP may cause
immunological or hematological effects in humans under relevant exposure circumstances.

Thyroid

Hoffman et al. f: identified an association between TCEP exposure and thyroid cancer in humans
and	identified increased incidences of thyroid neoplasms in rats in a 2-year cancer

bioassay but with uncertainty regarding its association with TCEP exposure. However, Moser et al.
(2 found no changes in serum thyroid hormone levels in rat dams and offspring in a
prenatal/postnatal study. Based on these data, human evidence for thyroid effects is slight and animal
evidence is indeterminate. Overall, the currently available evidence is inadequate to assess whether
TCEP may cause thyroid changes in humans under relevant exposure circumstances.

Endocrine (Other)

Based on indeterminate human and animal evidence and lack of mechanistic support, the currently
available evidence is inadequate to assess whether TCEP may cause endocrine changes other than
thyroid and reproductive hormones in humans.

Page 557 of 572

-------
14158

14159

14160

14161

14162

14163

14164

14165

14166

14167

14168

14169

14170

14171

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Lung/Respiratory

Based on a lack of epidemiological studies, human evidence is indeterminate. In addition, animal data
are indeterminate (no relevant histopathological effects, lung weight changes in studies with high and
uninformative overall data quality determinations) based on high-quality studies. Therefore, the
currently available evidence is inadequate to assess whether TCEP may cause lung or respiratory effects
in humans under relevant exposure circumstances.

Body Weight

EPA identified no human studies that had information on body weight changes and therefore, human
evidence is indeterminate. In animal toxicity studies, TCEP effects on body weight were not consistent
across multiple studies. When body weight changes were observed, they were not consistently increased or
decreased. Therefore, the animal data are indeterminate. Overall, the currently available evidence is
inadequate to assess whether TCEP may cause changes in body weight in humans under relevant
exposure circumstances.

Page 558 of 572

-------
14172

14173

14174

14175

14176

14177

14178

14179

14180

14181

14182

14183

14184

14185

14186

14187

14188

14189

14190

14191

14192

14193

14194

14195

14196

14197

14198

14199

14200

14201

14202

14203

14204

14205

14206

14207

14208

14209

14210

14211

14212

14213

14214

14215

14216

14217

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Appendix L GENOTOXICITY DATA SUMMARY

TableApx L-3 summarizes the database of studies on chromosomal aberrations, gene mutations, and
other genotoxicity endpoints for TCEP. Although EPA did not evaluate these studies using formal data
quality criteria, selected studies were reviewed by comparing against current OECD test guidelines and
important deviations are noted below. When interpreting the results of these studies, EPA also consulted

OECD C

EPA did not retrieve all original studies for one or more of the following reasons: (1) they were not
readily available, (2) they were in a foreign language, (3) they evaluated effects other than chromosomal
aberrations or gene mutations, and (4) there were multiple studies of the same type (e.g., bacterial
reverse mutation assays). EPA also referred to some studies cited in the 2009 European Union Risk
Assessment Report (ECB. 2009) and Beth-Hubni 9) for some studies that were not obtained.

L.l.l Chromosomal Aberrations

EPA located one in vivo micronucleus assay using Chinese hamsters (Sala et ai. 1982) that was
equivocal/weakly positive for micronuclei. Two additional in vivo micronucleus studies in mice cited in
E 09) and Beth-Hubner (1999) were not readily available. EPA also identified an in vitro assay
that did not find chromosomal aberrations to be associated with TCEP exposure in Chinese hamster
ovary cells (Galloway et ai. 1987).

L.l.1.1 In Vivo Data

Sala et al. (1982) report results of an in vivo micronucleus assay in which Chinese hamsters were treated
with a single i.p. dose at 0, 62.5, 125, or 250 mg/kg bw and bone marrow was evaluated for presence of
micronuclei. The authors conducted a Student's T-test to determine whether the means differed between
dose groups and the DMSO negative control. In females, the two lowest doses exhibited a statistically
significant increase in micronuclei compared with controls. Males had increased micronuclei at the
highest dose. However, only two hamsters per sex per dose were used, which would have made
statistical significance difficult to detect. When results for both sexes were combined, the two highest
doses showed differences from controls (see Table Apx L-l). The authors also conducted linear
regression to evaluate the dose response but did not report those results. The authors describe the results
as a slight effect that is difficult to interpret due to different responses between sexes and "variation with
the doses." EPA conducted a comparison of the means of each sex for each of the doses and considered
the dose-response for the combined sexes to be valid.

The study methods deviated from OECD Test Guideline 474 ((	) in several ways.

Specifically, the authors used an exposure route that is not recommended and scored fewer erythrocytes
than recommended (2,000 vs. 4,000). Furthermore, the study did not provide information to ensure that
the test substance reached the bone marrow, although positive effects suggest TCEP likely reached the
target tissue (Sala et al.. 1982). In addition, when using both sexes, the guidelines recommend using five
animals per sex, not two per sex. Despite these deviations, some of which might decrease the ability to
detect a response (e.g., numbers of animals/sex and number of erythrocytes scored, lack of verification
that the chemical reached the bone marrow), the results are consistent with an equivocal/ weak positive
response.

The 2009 European Union Risk Assessment Report (ECB. 2009) and Beth-Hubni 9) reference two
additional micronucleus studies that reported negative results. The cited studies were an oral study using
NMRI mice with dosing for one time at 1,000 mg/kg and an i.p. injection study with doses up to 700
mg/kg using CD-I mice (ECB. 2009).

Page 559 of 572

-------
14218

14219

14220

14221

14222

14223

14224

14225

14226

14227

14228

14229

14230

14231

14232

14233

14234

14235

14236

14237

14238

14239

14240

14241

14242

14243

14244

14245

14246

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Table Apx L-l. Results of In Vivo Micronucleus Test

Dose (mg/kg-bw)

Mean (Standard Deviation)'"

d

Males

Females

Both Sexes

Qa

4(1.3)

3 (0.58)

3.5 (1.0)

62.5

4 (0.82)

6.5 (1.4)*

5.25 (1.4)

125

6.25 (1.1)

7.0 (1.3)**

6.63 (1.1)***

250

7.25 (0.35)*

6.75 (3.0)

7.0(2.0)**

a DMSO solvent control (2,200 mg/kg-bw); * p < 0.05; ** p < 0.01; *** p < 0.001
b Standard deviation is in parentheses is equal to the standard error reported in the study x square-root of n
(2/sex/dose for individual sexes and 4/dose for combined sexes)
c Number of micronuclei per 1,000 polychromatic erythrocytes

d Comparison of sexes for each does was done with the following program that compared means:

littDs://www.medcalc.ors/calc/comDarison of means.olio; the p values for 0. 62.5. 125. and 250 me/ke were

0.4252, 0.1612, 0.5969, and 0.8367, demonstrating that outcomes were not significantly different between the
sexes and the results could be combined.

Source: Salaet al. (1982)

L.l.1.2 In Vitro Data

Galloway et al. (1987) evaluated chromosomal aberrations in Chinese hamster ovary cells. Many study
methods were consistent with OECD Test Guideline 473 (OECD. 2016a). except that the authors scored
only 100 cells per concentration compared with the recommended 300 per concentration needed to
conclude that a test is clearly negative. Aberrations at 0, 160, 500 and 1,600 jag/m L were observed in 6,
10, 10 and 9 percent of cells without activation, respectively, and 4, 10, 7 and 8 percent with activation.
Neither trend test was statistically significant (p < 0.05).

L.1.2 Gene Mutations	

A forward gene mutation study using Chinese hamster lung fibroblasts (Sala et al.. 1982) and multiple
bacterial reverse gene mutation assays (Follmann and Wober. 2006; Haworth et al.. 198. , * ^ \ \
Prival et al.. 1977; Simmon et al.. 1977) were all negative for the induction of gene mutations. Beth-
Hubn 9) also reported negative results in a reverse gene mutation assay yeast and in two mouse
lymphoma assays. A single study (Nakamura et; 9) induced a four- to seven-fold increase in gene
mutations in one Salmonella typhimurium strain with metabolic activation and less than a doubling in a
second strain.

L. 1.2.1 In Vitro Studies

Sala et al. (1982) evaluated the effect of TCEP exposure in a forward gene mutation assay that measured
induction of 6-thioguanine-resistant mutants using Chinese hamster lung fibroblasts (V79 cells) in the
presence and absence of metabolic activation. The authors used a negative control (acetone) as well as
two positive controls. Although the incubation times and solvents followed OECD Test Guideline 476
(2016) recommendations, the experiment did not report use of an enzyme-inducing agent for the S9
fraction and the S9 fraction was used at 20 percent (vs. <10 percent as recommended by OECD 476).
The experiment also employed three instead of a recommended four concentrations. Furthermore, it is
not clear whether the OECD 476 recommended 20x 106 cells were grown by the time the cells were
treated with TCEP. The positive control run without S9 was not one of the OECD 476 recommended
controls. TCEP exposure did not result in increased mutations with or without S9; the authors noted that
the results were confirmed in several independent experiments.

Page 560 of 572

-------
14247

14248

14249

14250

14251

14252

14253

14254

14255

14256

14257

14258

14259

14260

14261

14262

14263

14264

14265

14266

14267

14268

14269

14270

14271

14272

14273

14274

14275

14276

14277

14278

14279

14280

14281

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

TCEP tested negative for gene mutations in many bacterial reverse mutation assays using multiple S.
typhimurium strains (Follmann and Wober. 2006; Haworth et ai. 1983; Prival et ai. 1977; Simmon et
ai. 1977) (see TableApx L-3). Beth-Hubner (1999) references two additional studies that reported
negative results in reverse mutation assays using S. typhimurium strains TA98, TA100, TA1535,
TA1537, and TA1538.

A single study (Nakamura et at.. 1979) identified increased mutations using S. typhimurium TA100 both
with and without metabolic activation and for TA1535 in the presence of metabolic activation
(Table Apx L-3). In S. typhimurium TA100, none of the concentrations showed a doubling of revertants
compared with the negative control response. However, the TA1535 response was approximately 4
times greater than controls at 3 |iM (-860 |ig/plate) and more than 7 times higher at 10 |iM (-2,900
Hg/plate) (Nakamura et at.. 1979). The study did not present statistical analyses. Therefore, EPA
modeled the dose-response to confirm the findings. It is not clear why the Nakamura et; 9) results
were inconsistent with other studies. Concentrations were comparable to other studies that showed
negative results. One difference in this study compared with others is in the method of enzyme induction
used to prepare the S9 fraction; Nakamura et I \ lv">79) used a mixture of PCBs (Kanechlor 500) for this
induction, whereas others used Aroclor 1254 or did not appear to induce enzymes in the S9 fractions.

Table Apx L-2. Resu

ts of Bacterial Reverse Mutation Test in Salmonella typhimurium

Concentration
(jiMol)

His+ Revertants/Plate

TA100

TA1535

-S9

+S9

-S9

+S9

0

141

140

9

14

1

158

191

14

31

3

161

192

8

57

10

111

246

6

107

30

8

86

1

7

Source: Nakamura et al. (1979)

None of the bacterial reverse mutation assays used Escherichia coli WP2 uvrA or E. coli WP2 uvrA
(PKM101), which should more likely identify oxidizing or alkylating mutagens than the Salmonella
strains used in the majority of TCEP studies. However, Follmann and Wober (2006) did test TCEP using
S. typhimurium TA102, which can also identify such mutagens, and found that TCEP did not induce
reverse mutations with this strain.

Beth-Hubner (1999) also reported negative results in a reverse gene mutation assay using
Saccharomyces cerevisiae D4 and in two mouse lymphoma assays (using the thymidine kinase locus).

L.1.3 Other Genotoxicity Assays

Table Apx L-3 summarizes two sister chromatid exchange (SCE) assays (Galloway et ai I * , v ala et
al. 1982). in vitro comet assays measuring DNA damage and repair (Bukowski et al. 2019; Follmann
and Wober. 2006). two cell transformation assays (Sala et a I), and a DNA binding assay using
TCEP (Lown et al.. 1980). Beth-Hubner (1999) also summarized an eye mosaic test (somatic mutation
and recombination) using Drosophila melanogaster.

Page 561 of 572

-------
14282

14283

14284

14285

14286

14287

14288

14289

14290

14291

14292

14293

14294

14295

14296

14297

14298

14299

14300

14301

14302

14303

14304

14305

14306

14307

14308

14309

14310

14311

14312

14313

14314

14315

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

These assays test for potentially harmful effects on genetic material such as DNA damage, cell
transformation, DNA alkylation and chromosomal damage. However, unlike gene mutation and
chromosomal aberrations studies, the changes measured in these assays may not be persistent and
transmissible.

Two studies of TCEP induction of SCEs identified equivocal results in Chinese hamster ovary cells
(positive in one of two trials with S9, negative without S9) and positive results without a dose-response
in Chinese hamster lung fibroblasts (Galloway et ai. 1987; Sala et ai. 1982). suggesting some genetic
damage, but without an understanding of the mechanism of action for this damage. The OECD test
guideline related to evaluation of SCEs (OECD 479) was deleted in 2014 because the mechanism for
this effect is not known (OECD. 2017).

TCEP was not considered to be an alkylating agent in an in vitro DNA binding assay (Lown et ai.

1980).

Bukowski et al. (2019) conducted in vitro comet assays (alkaline and neutral) in peripheral mononuclear
blood cells (PMBCs) and identified DNA damage at the highest concentration of TCEP tested (1 mM).
Cell toxicity was not evaluated in the study, but previous results identified viability of PMBCs to be 92
percent of controls at 1 mM TCEP. DNA damage to the PMBCs was repaired within 2 hours (Bukowski
et al.. 2019). Another comet assay did not identify DNA damage in Chinese hamster fibroblasts at TCEP
concentrations up to 1 mM with or without metabolic activation (Follmann and Wober. 2006).

Sala et al. (1982) identified a high level of cell transformation in Syrian hamster embryo (SHE) cells but
a lower level with metabolic activation when using C3H10T1/2 cells. OE( 07). p. 24, states that
"cell transformation has been related to structural alterations and changes in the expression of genes
involved in cell cycle control, proliferation and differentiation." The genomic changes may result from
direct or indirect genetic interactions or non-genotoxic mechanisms. Tamokou and Kuete (2 notes
that the SHE assay is believed to detect early steps in the process of carcinogenesis, and that C3H10 cell
assays related to later changes.

Taniai et al. (2012a) found no statistically significant increase in immunoreactive cells associated with
repair of double-strand DNA double-strand breaks or regulation of cell cycle checkpoints after such
DNA damage in kidneys of male rats dosed with 350 mg/kg-day TCEP for 28 days.

Page 562 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

14316 Table Apx L-3. TCEP Genotoxicity Studies

Test Type

Exposure

Metabolic
Activation

Positive
Controls

Outcome

Rct'crcncc(s)

Spceics (Sex)/
Route

Conecntration/Dosc/
Duration

( ln\iiikiMim;il ;ihcii;ilions in vivn

Micronucleus

Chinese
hamsters
(M+F)/
intraperitoneal

0, 62.5, 125, 250
mg/kg

Single administration

NA

Yes

Equivocal, weakly
positive for micronuclei

Sala et al. (1982)

('hix)iiu
-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Test Type

Exposure

Mctabolie
Activation

Positive
Controls

Outcome

Rcfcrcncc(s)

Spceics (Sex)/
Route

Conecntration/Dosc/
Duration



TA1535,
TA1537,
TA1538

8,599.5 ng/plate]









In vitro

bacterial reverse mutation assay

Salmonella
typhimurium
strains TA100,
TA1535,
TA1538

1,390 and 13,900 ng/
plate "

± S9 from
normal
Sprague-
Dawley rats
and from rats
induced by
Aroclor 1254

None stated

Negative for mutagenicity

[No statistical methods
cited; visual inspection
showed lack of dose
response]

Prival et al. (1977)

In vitro

bacterial reverse mutation assay

Salmonella

typhimurium

strains TA98,

TA100,

TA1535,

TA1537,

TA1538

Compounds were
tested up to 5
mg/plate or toxic
dose, whichever was
lower

+ S9 from
rats induced
by Aroclor
1254
[unclear
whether
TCEP was
tested
without S9]



Negative for mutagenicity

Simmon et al. (1977)

In vitro

bacterial reverse mutation assay

Salmonella

typhimurium

strains TA 98,

TA100,

TA1535,

TA1537,

TA1538

0,0.1, 10, 100, 500,
2000 ng/plate; No
cytotoxicity observed

± S9 from
rats induced
by Aroclor
1254



Negative for mutagenicity

BIBRA (1977)

()llicr ueiKi|ii\icil> assa\ s

In vitro Sister chromatid exchange

Chinese
hamster ovary
cells

Without S9: One trial.
26 hr incubation
5,16,50, 160 ng/mL;
With S9. Two trials, 2
hr incubation; Trial 1:
160, 500, 1,600
Hg/mL; Trial 2: 1200,
1400, 1600 ng/mL

+/-

S9 from rats

Yes

Equivocal overall
Without activation -
negative;

With activation - Trial 1
had significant responses
at the two highest doses;
Trial 2 was negative at all
doses; lowest
concentration with stat
significant increase was

Galloway et al. (1987)
andN

Page 564 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Test Type

Exposure

Metabolic
Activation

Positive
Controls

Outcome

Rcfcrcncc(s)

Spceics (Sex)/
Route

Conecntration/Dosc/
Duration











500 ug/mL; Trial 1
reached a 20% increase in
SCEs [No mention
whether cytotoxicity was
observed.]



In vitro

Sister chromatid exchanges

V79 cells
Chinese
hamster lung
fibroblasts

343, 490, 700, 1,000
Hg/ml (experiment I);
2,000, 3,000 ng/mL
(experiment II)





SCEs induced with no
clear dose response (toxic
observed at 3000 ug/mL,
with mitosis partially
inhibited)

Sala et al. (1982)

In vitro comet assay:
DNA damage

Human:

peripheral

blood

mononuclear
cells

1 to 1,000 nM
(alkaline version)
10 to 1,000 nM
(neutral version)



Yes -

H202

(alkaline

version);

9 Gy

(neutral

version)

DNA damage observed at
1 mM in both assays
(single and double strand
breaks in alkaline version;
double strand breaks in
the neutral version).

Cell viability was not
assessed in the current
assav but Mokra et al.
Q identified
viability as slightly
decreased at 1 mM TCEP
(92% of controls)

Bukowski et al. (2019)

In vitro comet assay:
DNA repair

Human:

peripheral

blood

mononuclear
cells

100, 500, 1,000 nM
(alkaline)

500, 1,000 nM
(neutral) for 24 hr to
induce damage; 60-
120 min for repair
assay





Single and double strand
breaks and alkali-labile
sites occurred observed at
1,000 |iIVI were repaired
after 2 hr (alkaline)
Double strand breaks at
1,000 |iIVI were repaired
after 2 hr (neutral)

Bukowski et al. (2019)

In vitro comet assay

V79 Chinese
hamster
fibroblast cells

1 to 1,000 for 24
hr

+/- S9

Yes -

potassium

dichromate

No DNA strand breaks
observed with or without
S9

Follmann and Wober

(2006)

In vitro cell transformation

Syrian hamster
embryo cells

400, 500, 600, 800
Hg/mL





High level of
transformation

Sala et al. (1982)

Page 565 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Test Type

Exposure

Metabolic
Activation

Positive
Controls

Outcome

Rcfcrcncc(s)

Spceics (Sex)/
Route

Conecntration/Dosc/
Duration















In vitro cell transformation

C3H10T1/2
cells

900 and 1,500 ng/mL

Yes



Low incidence of
transformed foci with
metabolic activation (S9)

Sala et al. (1982)

DNA binding

In vitro

PM2-CCC-

DNA

5 mM in 180 min





No alkylation observed

Lown et al. (1980)

14317

Page 566 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE

December 2023

14318	Appendix M EXPOSURE RESPONSE ARRAY FOR HUMAN

14319	HEALTH HAZARDS

14320	The following exposure response array (Figure_Apx M-l) presents HEDs for all studies and hazard

14321	endpoints that yielded likely or suggestive evidence integration conclusions. The information is arrayed

14322	by lowest to highest HED for NOAELs and BMDLs; all PODs based on LOAELs are listed separately.

























ANOAEL HED



Kidney; Rel. kidney wt; 16 wk; rat (F); NTP 1991b























¦ BMDL HED



Reproductive; No. of seminiferous tubules; 35 d; mouse (M); Chen et al. 2015

¦























Liver; Rel. liver wt; 16 wk; rat (F); NTP 1991b



.....





















~ LOAEL HED



Kidney; Abs. kidney wt; 16 wk; rat (F); NTP 1991b

__

.....

























Liver; Abs. liver wt; 16 wk; mouse (F); NTP 1991b

























NOAELs and BMDLs: UF = 30

Liver; Abs. liver wt; 66 wk; rat (M); NTP 1991b



¦



~











	





LOAELs: UF

= 300

Kidney; Hyperplasia; 2 yr; rat (F); NTP 1991b





¦

























Kidney; Karyomegaly; 2 yr; mouse (M); NTP 1991b





























Kidney; Hyperplasia; 2 yr; rat (M); NTP 1991b





¦























Neurotoxicity; Brain lesions; 2 yr; rat (F); NTP 1991b





"a

_





















Mortality; 2 yr; rat (M,F); NTP 1991b





A























Liver; Rel. liver wt; 66 wk; rat (M); NTP 1991b





"A

	





















Kidney; Abs. & rel. kidney wt; 66 wk; rat (M); NTP 1991b





A

	





















Kidney; Rel. kidney wt; 16 wk; mouse (M); NTP 1991b





A























Liver; Rel. liver wt; 16 wk; mouse (F); NTP 1991b



























Developmental; Task 2: Live male Fl pups/litter; Up to 18 wk; mouse (M); NTP 1991a





























Developmental; Task4: Live F2 pups/litter; 14 wk; mouse (M,F); NTP 1991a





























Neurotoxicity; Hippocampal lesions; 60 d; rat (F); Yang et al. 2018







r



















N eu rotoxi city;

Brain (hippocampal) necrosis; 16 wk; rat (F); NTP 1991b; Matthews et al. 1990







¦



















Neurotoxicity; Changes in path length, Morris water maze; 60 d; rat (F); Yang et al. 2018







¦





















Reproductive; Testiscular testosterone; 35 d; mouse (M); Chen et al. 2015







~





















Liver; Eosinophilic liver foci; 2 yr; mouse (M); NTP 1991b







¦





















Kidney; Karyomegaly; 2 yr; mouse (F); NTP 1991b







¦



















Liver; Abs. liver wt; 16 wk; rat (F); NTP 1991b







~



















Neurotoxicity; Ataxia, convulsions; 16 d; mouse (NS); NTP 1991b







A~



















Neurotoxicity; Brain lesions; 2 yr; mouse (M); NTP 1991b







~



















Reproductive; Abs. & rel. testes wt; 16 wk; mouse (M); NTP 1991b







~



















Reproductive; Sperm count; 16 wk; mouse (M); Matthews et al. 1990







A.

















Neurotoxicity; Serum cholinesterase activity; 16 wk; rat (F); NTP 1991b; Matthews et al. 1990























Developmental; Task 2: FO mean litters/pair; Live total F1 pups/litter; Live female F1 pups/litter; Up to 18 wk;

mniKP riUI.FV MTP 1 QQ1 3







A

















Reproductive; Task 4: Fertility & pregnancy index in Fl; 14 wk; mouse (M,F); NTP 1991a







A



















Neurotoxicity; Clinical observations; 60 d; rat (F); Yanget al. 2018









A



















Reproductive; Testes wt; 35 d; mouse (M); Chen et al. 2015





























Liver; Abs. & rel. liver wt; 16 wk; rat (M); NTP 1991b







A

















Kidney; Abs. & rel. kidney wt; 16 wk; rat (M); NTP 1991b







~















Reproductive; Task 2: Fertility, litter 5 in FO; Up to 18 wk; mouse (M,F); NTP 1991a









¦

















Liver; Rel. liver wt; 16 wk; mouse (M); NTP 1991b









~

















Kidney; Abs. kidney wt; 16 wk; mouse (M); NTP 1991b









j

k

















Kidney; Rel. kidney wt; 16 d; mouse (F); NTP 1991b









A

i













Reproductive; Task 2: Days to litter 2 & days to litter 3 in FO; Up to 18 wk; mouse (M,F); NTP 1991a











A











Liver; Task 4: Cytomegaly, hepatitis, & hepatocellular degeneration in Fl; 14 wk; mouse (M); NTP 1991a











~













Kidney; Regenerating tubules, other histopathological changes; 28 d; rat (M); Taniai et al. 2012













4













Liver; Task 3: Abs. & rel. liver wt in FO; Cytomegaly & hepatitis in FO; 18 wk; mouse (M); NTP 1991a
Developmental; Task 3: Live female Fl pups/litter (treated FO males); Live male & total Fl pups/litter

(treated FO males or females); 18 wk; mouse (M,F); NTP 1991a













....

























.....











Reproductive; Task 3: Organ wt changes & histopathology; Sperm parameters; Pregnancy & fertility indices;

1 R wk- mnncp flWI PV MTP 1 QQ1 sa Ml













...ill













Neurotoxicity; Prostration, jerking movements, languidity; 8 d; mouse (F); Hazleton Labs 1983





































LO





1

oo

ID

OO







Dose (i

ng/kg-day)

[1] Task 3: Abs. epididymis wt in FO; Fluid & degenerated cells in epididymis in FO; Abs. & rel. testes wt in FO; Interstitial

eel 1

hyper

plasia

in

testes in FO; Ovarian cysts in FO; Sperm concentration, % motile, & % abnormal sperm in FO; Pregnancy index & fertility index (treated FO

males)



















14323

14324	FigureApx M-l. Exposure Response Array for Likely and Suggestive Human Health Hazard

14325	Outcomes

Page 567 of 572

-------
14326

14327

14328

14329

14330

14331

14332

14333

14334

14335

14336

14337

14338

14339

14340

14341

14342

14343

14344

14345

14346

14347

14348

14349

14350

14351

14352

14353

14354

14355

14356

14357

14358

14359

14360

14361

14362

14363

14364

14365

14366

14367

14368

14369

14370

14371

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

Appendix N DRAFT EXISTING CHEMICAL EXPOSURE LIMIT
(ECEL) DERIVATION

EPA has calculated a draft 8-hour existing chemical occupational exposure value to summarize the
occupational exposure scenario and sensitive health endpoints into a single value. This calculated draft
value may be used to support risk management efforts for TCEP under TSCA section 6(a), 15 U.S.C.
§2605. EPA calculated the draft value rounded to 0.09 mg/m3 for inhalation exposures to TCEP as an 8-
hour time-weighted average (TWA) and for consideration in workplace settings (see Appendix N.l)
based on the lifetime cancer inhalation unit risk (IUR) for kidney cancer.

TSCA requires risk evaluations to be conducted without consideration of costs and other non-risk
factors, and thus this draft occupational exposure value represents a risk-only number. If risk
management for TCEP follows the final risk evaluation, EPA may consider costs and other non-risk
factors, such as technological feasibility, the availability of alternatives, and the potential for critical or
essential uses. Any existing chemical exposure limit (ECEL) used for occupational safety risk
management purposes could differ from the draft occupational exposure value presented in this
appendix based on additional consideration of exposures and non-risk factors consistent with TSCA
section 6(c).

This calculated draft value for TCEP represents the exposure concentration below which workers and
occupational non-users are not expected to exhibit any appreciable risk of adverse toxicological
outcomes, accounting for potentially exposed and susceptible populations (PESS). It is derived based on
the most sensitive human health effect {i.e., cancer) relative to benchmarks and standard occupational
scenario assumptions of 8 hours per day, 5 days per week exposures for a total of 250 days exposure per
year, and a 40-year working life.

EPA expects that at the draft occupational exposure value of 0.008 ppm (0.09 mg/m3), a worker or
occupational non-user also would be protected against neurotoxicity from acute occupational exposure
as well as male reproductive effects from short-term and chronic occupational exposures if ambient
exposures are kept below this draft occupational exposure value. EPA has not separately calculated a
draft short-term {i.e., 15-minute) occupational exposure value because EPA did not identify hazards for
TCEP associated with this very short duration.

EPA did not identify a government-validated method for analyzing TCEP in air, but Appendix N.2
presents summary of a method described by La Guardia and Hale (2015) and Grimes et al. (2019). The
identified limit of detection (LOD) and limit of quantification (LOQ) using the method and the resulting
monitoring data from Grimes et al. (2019) are below the lowest calculated draft occupational exposure
value, indicating that monitoring below these levels may be achievable and that some workplaces may
already be achieving the draft occupational exposure value.

The Occupational Safety and Health Administration (OSHA) has not set a permissible exposure limit
(PEL) as an 8-hour TWA for TCEP (https://www.osha.eov/taws-

rees/reeulations/standardnumber/191 (	2). EPA also did not locate other exposure

limits for TCEP.

N.l Draft Occupational Exposure Value Calculations

This appendix presents the calculations used to estimate draft occupational exposure values using inputs
derived in this draft risk evaluation. Multiple values are presented below for hazard endpoints based on

Page 568 of 572

-------
14372

14373

14374

14375

14376

14377

14378

14379

14380

14381

14382

14383

14384

14385

14386

14387

14388

14389

14390

14391

14392

14393

14394

14395

14396

14397

14398

14399

14400

14401

14402

14403

14404

14405

14406

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

different exposure durations (described further in section 5.2.6). For TCEP, the most sensitive
occupational exposure value is based on cancer and the resulting 8-hour TWA is rounded to 0.09 mg/m3.

Draft Lifetime Cancer Occupational Exposure Value

The draft occupational exposure value (EV) was calculated for the occupational lifetime cancer IUR for
kidney cancer and is the concentration at which the extra cancer risk is equivalent to the benchmark
cancer risk of 1x 10~4:

Benchmark cancer ATIUR

^resting

£ *rnnre>T —	TTTTi	* T77	77777 * "

cancer	WR	ED * EF *WY ' IRworkers

lxl0—4	24^*—*78y 0.6125^

=	2	* h zsod	*	= 7.96x10 ppm

5.26x10 ~z per ppm 8-*^W)y 125—
d y	hr

/mg\ EV ppm * MW 0.00796 ppm * 285^^	mg

EVcancer 3 J MolarVolume	24 45 —^—	0.0928 m3

mol

Draft Acute Non-cancer Occupational Exposure Value

The draft acute occupational exposure value (EVaCute) was calculated as the concentration at which the
acute MOE would equal the benchmark MOE for acute occupational exposures using the following
equation:

HECacute	ATHECacute IRresting

EVar„i-p = 		:	,		*	—	*	-

Benchmark M 0 EacU£e ED IRworkers

24/l n£10rm3
4.41 ppm ~T~ O.blzb-r—	ms

¦ a ¦	hr = 0.216 ppm = 2.51 —f

• * •

30 8h	m3 —	¦-m3

d	hr

Draft Intermediate Non-cancer Exposure Value

The draft intermediate occupational exposure value (EVintermediate) was calculated as the concentration at
which the intermediate MOE would equal the benchmark MOE for intermediate occupational exposures
using the following equation:

gy		 	HECjntermefljate	^ AThec intermediate^ Unresting

intermediate Benchmark MOfjntermediate	ED*EF	IRworkers

24/i	m3

1 27 com * 30d 0.6125^-	mg

= 1^/PP"1« 4	.	= 0.0849 ppm = 0.990 ^

30 ^.22 d 1.25^

a	hr

Draft Chronic Non-cancer Exposure Value

The draft chronic occupational exposure value (EVchronic) was calculated as the concentration at which
the chronic MOE would equal the benchmark MOE for chronic occupational exposures using the
following equation:

cy		 	HECchronic	 AThec chronic ^ IRresting

^ V rhrnnir	^

cnronic Benchmark MOEchronic ED*EF*WY IRworkers

Page 569 of 572

-------
14407

14408

14409

14410

14411

14412

14413

14414

14415

14416

14417

14418

14419

14420

14421

14422

14423

14424

14425

14426

14427

14428

14429

14430

14431

14432

14433

14434

14435

14436

14437

14438

14439

14440

14441

14442

14443

14444

14445

14446

14447

14448

14449

14450

14451

14452

14453

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

24h 365d	„ , „ „_mJ

177nnm ~~T~*	*40 y*0.6125-r—

l.Z/ppm d y J	hr

	 * ¦			

30

8h 250d An „

—*	*40 y*1.25-r—

d y J	hr

0.0909 ppm = 1.06 --f

Where:

ATiur

ATHECacute

AT HECintermediate

ATHECchronic

Benchmarkcancer
Benchmark MOEaCute

= Averaging time for the cancer IUR, based on study conditions and
adjustments (24 hr/day for 365 days/yr) and averaged over a lifetime
(78 yrs) (s qq Draft Risk Evaluation for Tris(2-chloroethyl) Phosphate
(TCEP) - Supplemental Information File: Supplemental Information
on Environmental Release and Occupational Exposure Assessment
(	20230 and Section 5.2.6).

= Averaging time for the POD/HEC used for evaluating non-cancer
acute occupational risk based on study conditions and HEC
adjustments (24 hr/day) (see Section 5.2.6).

= Averaging time for the POD/HEC used for evaluating non-cancer
intermediate occupational risk based on study conditions and/or any
HEC adjustments (24 hr/day for 30 days) (see Section 5.2.6).

= Averaging time for the POD/HEC used for evaluating non-cancer
chronic occupational risk based on study conditions and/or HEC
adjustments (24 hr/day for 365 days/yr) (see Section 5.2.6) and
assuming the same number of years as the high-end working years
(WY, 40 years) for a worker.

= Benchmark for excess lifetime cancer risk, based on 1 x 10~4 extra risk

= Acute non-cancer benchmark margin of exposure, based on the total
uncertainty factor of 30 (see Section 5.2.6.1.1)

Benchmark MOEintermediate= Intermediate non-cancer benchmark margin of exposure, based on the

total uncertainty factor of 30 (see Section 5.2.6.1.2)

Benchmark MOEchronic = Chronic non-cancer benchmark margin of exposure, based on the total

uncertainty factor of 30 (see Section 5.2.6.1.2)

EVC

Existing chemical occupational exposure value (mg/m3 and ppm)
based on lifetime cancer risk at lxl0~4

E V acute
E V intermediate

E V chronic

ED

= Occupational exposure value based on acute neurotoxicity

= Occupational exposure value based on intermediate reproductive
toxicity

= Occupational exposure value based on chronic reproductive toxicity
= Exposure duration (8 hr/day) (see Table 5-5)

Page 570 of 572

-------
14454

14455

14456

14457

14458

14459

14460

14461

14462

14463

14464

14465

14466

14467

14468

14469

14470

14471

14472

14473

14474

14475

14476

14477

14478

14479

14480

14481

14482

14483

14484

14485

14486

14487

14488

PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

EF

Exposure frequency (1 day for acute, 22 days for intermediate, and
250 days/yr for chronic and lifetime) (see Section 5.1.2.1)

HEC

Human equivalent concentration for acute, intermediate, or chronic
non-cancer occupational exposure scenarios (see Table 5-49, Table
5-50, and Table 5-51)

IUR

Inhalation unit risk (per mg/m3 and per ppm) (see Table 5-52)

IR

Inhalation rate (default is 1.25 m3/hr for workers and 0.6125 m3/hr
assumed from "resting" animals from toxicity studies)

Molar Volume

24.45 L/mol, the volume of a mole of gas at 1 atm and 25 °C

MW

Molecular weight of TCEP (285 g/mole)

WY

Working years per lifetime at the 95th percentile (40 years) {Draft Risk
Evaluation for Tris(2-chloroethyl) Phosphate (TCEP) - Supplemental
Information File: Supplemental Information on Environmental Release
and Occupational Exposure Assessment (	231))

Unit conversion:

1 ppm = 11.7 mg/m3 (see equation associated with the EVcancer calculation)

N.2 Summary of Air Sampling Analytical Methods Identified

EPA conducted a search to identify relevant NIOSH, OSHA, and EPA analytical methods used to
monitor for the presence of TCEP in air (see TableApx N-l). The following sources were included for
the search:

1.	NIOSH Manual of Analytical Methods (NMAM); 5th Edition

2.	NIOSH NMAM 4th Edition

3.	OSHA Index of Sampling and Analytical Methods

4.	EPA Environmental Test Method and Monitoring Information

EPA did not identify any government-validated methods for TCEP. However, a method was described
and used by La Guardia and Hale (2015) and Grimes et al. (2019). The method and associated
LOD/LOQ are summarized in Table Apx N-l.

Page 571 of 572

-------
PUBLIC RELEASE DRAFT - DO NOT CITE OR QUOTE
December 2023

14489	TableApx N-l. Limit of Detection (LOD) and Limit of Quantification (LOQ) Summary for

14490	Identified Air Sampling Analytical Methods			

Air Sampling
Analytical
Methods

Year
Published

LOD

LOQ

Notes

Source

Full-shift personal
sampling

2019

16 ng/m3

16 ng/m3

Method reports LOD/LOQ of
overall procedure as 16 ng/m3 using
Institute of Medicine (IOM)
sampler with a glass fiber filter at a
flow rate of 2 L/min for the
inhalable fraction of particulates
and custom OVS-2 tubes at 1 L/ per
min for vapor. Samples were sent to
lab for analysis/quantification.

Methods
described in La
Guardia and

Hale (2015) and
Grimes et al.
(2019)

ppm = parts per million; ppb = parts per billion; ppt = parts per trillion

14491

Page 572 of 572

-------