c/EPA
              United States
              Environmental Protection
              Agency
              Office of Health and
              Environmental Assessment
              Washington DC 20460
   EPA/600/8-90/057A
   July 1990
   Workshop Review Draft
              Research and Development
Health Assessment
Document for
Diesel Emissions
Workshop
Review
Draft
(Do Not
Cite or Quote)
              Appendices: A Thru D
                               Notice

              This document is a preliminary draft. It has not been formally
              released by EPA and should not at this stage be construed to
              represent Agency policy. It is being circulated for comment on its
              technical accuracy and policy implications.

-------
                                         EPA/600/8-90/057A
                                         July 1990
                                         Workshop Review Draft
Health Assessment Document  for
            Diesel Emissions

        Appendices: A Thru D
 This document is an internal draft for review purposes
 only and does not constitute Agency policy. Mention of
 trade names or commercial products does not constitute
 endorsement or recommendation for use.
         U.S. Environmental Protection Agency
         Office of Reasearch and Development
     Office of Health and Environmental Assessment
      Environmental Criteria and Assessment Office
          Research Triangle Park, NC 27711

-------
                   NOTICE
This document is an internal draft for review purposes
only and does not constitute Agency policy.  Mention of
trade names or commercial products does not constitute
endorsement or recommendation for use.
                       11

-------
                        TABLE OF CONTENTS



LIST OF TABLES	ix

LIST OF FIGURES	  xii


1.    SUMMARY	1-1

     1.1.  INTRODUCTION 	1-1
     1.1  COMPOSITION OF DIESEL EXHAUST	1-4

          1.11. Introduction 	1-4
          1.2.2. Overview of Pollutants and Pollution Formation	1-4
          1.13. Emission Factors and Inventories	1-5
          1.2.4. Emission Controls-Now and Projected	1-5
          1.2.5 Conclusions	1-5

     1.3.  CONCENTRATIONS OF DIESEL-DERIVED POLLUTANTS IN
          AIR, THEIR TRANSPORT AND TRANSFORMATIONS	  1-6

          1.3.1. Fundamental Nature of Diesel Pollutants in Air	1-6
          1.3.2. Dispersion of Primary Emissions in Air	1-6
          1.3.3. Atmospheric Transformations 	1-6
          1.3.4. Mutagenicity of Ambient Air Particles	1-6

     1.4.  NONCANCER HEALTH EFFECTS OF DIESEL EXHAUST  .... 1-7

          1.4.1. Toxic Effects of Diesel Exhaust on Humans	1-7
          1.4.2. Toxic Effects of Diesel Exhaust on Animals	1-8
          1.4.3. Interactive Effects of Diesel Exhaust
                Components  	1-10
          1.4.4. Comparison of the Effects of Gasoline and
               Diesel Exhaust 	Ml

     1.5.  MUTAGENICITY OF DIESEL ENGINE EMISSIONS  	1-12
     1.6.  METABOLISM AND MECHANISM OF ACTION OF DIESEL
          EMISSIONS INDUCED CARaNOGENICITY	1-12
     1.7.  CARCINOGENICITY OF DIESEL ENGINE EMISSIONS IN
          LABORATORY ANIMALS	1-13
     1.8.  PHARMACOKINETIC CONSIDERATIONS IN THE PULMONARY
          CARCINOGENICITY OF DIESEL ENGINE EMISSIONS	1-15
                                  111

-------
                    TABLE OF CONTENTS (CONT)

     1.9.  EPIDEMIOLOGY OF DIESEL EMISSIONS
          CARCINOGENICITY  	1-16
     1.10.  QUANTITATIVE ESTIMATE OF UNIT RISK	1-20
     1.12.  REFERENCES	1-21

2.    DIESEL EMISSIONS  	2-1

     2.1.  INTRODUCTION  	2-1
          2.1.1. Diesel Engine - What it is and How it is Used	2-1

     2.2.  OVERVIEW OF POLLUTANTS AND POLLUTANT
          FORMATION  	2-1

          111. Gas Phase Emissions	2-1
          2.2.2. Carbon Formation and Emission	2-1
          2.2.3. Gas-to-Particle Conversions	2-1
          114. Mutagens 	2-2

     2.3.  EMISSION FACTORS AND INVENTORIES	2-2

          13.1. Existing Data	2-2
          13.1 Models	2-2

     14.  EMISSION CONTROLS: NOW AND EXPECTED 	2-2
          14.1. Engine Modifications	2-2
          14.1 Add-On Devices: Descriptions and Performance 	2-2
          14.3. Alternative Fuels: Performance	2-2

     15.  CONCLUSIONS	2-2
     16.  REFERENCES	2-2

3.    DIESEL DERIVED POLLUTANTS: ATMOSPHERIC CONCENTRATIONS,
     TRANSPORT, AND TRANSFORMATIONS 	3-1

     1.0.  INTRODUCTION	3-2

     10.  PRIMARY DIESEL EMISSIONS	3-4

          11.  Gaseous Emissions	3-4
               11.1. Inorganic Gases	3-4
               11.2. Organic Gases  	3-6
                                 rv

-------
                TABLE OF CONTENTS (CONT.)

     11   Paniculate Emissions	3-9
           2.2.1. Diesel Paniculate Matter 	3-9

           12.1 Paniculate Phase Matter	3-14
           113. Paniculate Phase Organic Compounds 	3-14

13.  Factors Influencing Emissions of PAH and Nitro-PAH	3-25

2.4   Gaseous/Paniculate Phase Emission Partitioning 	3-28

3.0.  ATMOSPHERIC TRANSFORMATIONS OF PRIMARY DIESEL
     EMISSIONS  	3-32

     3.1.   The Fate of Primary Diesel Emissions and the Long Range
           Transport	3-32

     3.2.   Chemical Transformations	3-35

           3.11.  Gas-phase Reactions	3-35
           3.11  Paniculate-Phase Reactions  	3-51

3.3.  Physical Removal Processes	3-59

           3.3.1.  Dry Deposition	3-59
           3.3.1  Wet Deposition  	3-61

4.0.  ATMOSPHERIC CONCENTRATIONS OF PRIMARY DIESEL
     EMISSIONS AND THEIR TRANSFORMATION PRODUCTS .. 3-65

     4.1.   Volatile Organic Compounds (VOC) Attributable to Traffic  3-66
     4.1   Potycyclic Aromatic Hydrocarbons	3-68
     4.3.   Nitroarene Concentrations in Ambient Air	3-72
     4.4.   The Need for Atmospheric Tracers of Diesel Emissions .... 3-78

5.0.  MUTAGENICTTY OF RESPIRABLE AMBIENT PARTICLES .. 3-84

     5.1.   Bioassay Directed Chemical Analysis	3-85
     5.1   Contribution of Nitroarenes to Ambient Air 	3-88

6.0.  SUMMARY	3-93

7.0.   REFERENCES	3-95

-------
                   TABLE OF CONTENTS (CONT.)

4.    NONCANCER HEALTH EFFECTS OF DIESEL EXHAUST	 4-1

     4.1.   HEALTH EFFECTS OF WHOLE EXHAUST	4-1

          4.1.1. Human Data	4-1
          4.1.2. Animal Studies	4-8

     4.2.   COMPARISON OF HEALTH EFFECTS OF FILTERED AND
          UNFELTERED EXHAUST	4-44
     4.3.   INTERACTIVE EFFECTS OF DIESEL EXHAUST
          COMPONENTS	4-47
     4.4.   COMPARISON OF THE EFFECTS OF DIESEL EXHAUST AND
          GASOLINE EXHAUST	4-50
     4.5.   SUMMARY AND DISCUSSION	4-53

          4.5.1. Toxic Effects of Diesel Exhaust on Humans	4-53
          4.5.2. Toxic Effects of Diesel Exhaust on Animals	4-54
          4.5.3. Interactive Effects	4-62
          4.5.4. Comparisons with Gasoline Exhausts	4-63

     4.6.   REFERENCES	7	4-65

5.    MUTAGENIOTY 	5-1

     5.1.   GENE MUTATIONS	5-1
     5.2   CHROMOSOME EFFECTS	5-3
     5.3   OTHER GENOTOXIC EFFECTS 	5-5
     5.4   SUMMARY	5-5
     5.5   REFERENCES	5-6

6.    METABOLISM AND MECHANISM OF ACTION IN DIESEL EMISSION-
     INDUCED CARCINOGENESIS	6-1

     6.1   METABOLISM CONSIDERATIONS	6-1

          6.1.1.  Metabolism and Disposition of B[a]P	6-2
          6.1.2.  Metabolism and Disposition of 1-Nitropyrene  	6-7
                                VI

-------
                   TABLE OF CONTENTS (CONT.)

     6.1   CARCINOGENIC MECHANISM OF PAH COMPONENTS OF
          DIESEL EXHAUST	6-9

          6.2.1.   Carcinogenic Mechanism of B[a]P	6-11
          6.2.2.   Carcinogenic Mechanism of Nitropyrenes 	6-14

     6.3.   CARCINOGENICITY OF ALDEHYDES	6-17

          6.3.1.   Metabolism and Carcinogenicity of Formaldehyde	6-17
          6.3.1   Metabolism and Carcinogenicity of Acrolein 	6-18

     6.4.   POTENTIAL INVOLVEMENT OF PULMONARY LEUKOCYTES
          IN THE DEVELOPMENT OF LUNG TUMORS	6-19
     6.5.   SUMMARY OF METABOLISM AND MECHANISM OF ACTION
          OF CARCINOGENIC COMPONENTS OF DIESEL EXHAUST . 6-23
     6.6.   REFERENCES	6-26

7.    CARONOGENICITY OF DIESEL EMISSIONS IN LABORATORY
     ANIMALS	7-1

     7.1.   INTRODUCTION	r.	7-1
     7.1   CARONOGENICITY STUDIES IN LABORATORY ANIMALS .. 7-2

          7.11.    Long-term Inhalation Studies	7-2
          7.11    Short-term Inhalation and Intratracheal
                 Instillation Studies	7-38
          7.13.    Dermal Application, Subcutaneous Injection,
                 and Intraperitoneal Injection Studies	7-44
          7.14.    Summary of Animal Carcinogenicity Studies 	7-50

     7.3.   REFERENCES	7-55

8.    PHARMACOKINETIC CONSIDERATIONS IN THE PULMONARY
     CARdNOGENICTTY OF DIESEL ENGINE EMISSIONS	8-1

     8.1.   INTRODUCTION	8-1
     8.1   REGIONAL LUNG DEPOSITION OF INHALED
          PARTICLES BY HUMANS AND ANIMALS	8-1
     8.3.   TRACHEOBRONCHIAL CLEARANCE OF
          PARTICULATE MATTER	8-2
                                vu

-------
                     TABLE OF CONTENTS (CONT.)

     8.4.   CLEARANCE FROM DEEP LUNG REGIONS	8-3

           8.4.1.   Species Variability in Pulmonary Clearance
                  Processes  	8-10
           8.4.2.   Role of Alveolar Macrophages in the Clearance
                  of Paniculate Matter	8-12
           8.4.3.   Summary: Pulmonary Clearance of Diesel Exhaust
                  Participate Matter	8-37

     8.5.   DESORPTION OF CONSTITUENTS FROM DIESEL EXHAUST
           PARTICLES	8-38

           8.5.1.   Unavailability of Agents Adsorbed to Diesel
                  Exhaust Particles	8-38
           8.5.2.   Extraction of Carcinogens from Particles by
                   Alveolar Macrophages and Other Cell Types 	8-42
           8.5.3.   Unavailability of Adsorbed Compounds as a
                  Function of Particle Clearance Rates and
                  Extraction Rates of Adsorbed Compounds	8-48
           8.5.4.   Summary: Unavailability of Particle-Adsorbed
                  Agents 	8-50

     8.6.   INHIBITION OF RESPIRATION BY HIGH CONCENTRATIONS
           OF NOXIOUS AGENTS	8-51
     8.7.   CONSIDERATIONS FOR DOSIMETOY MODELING 	8-52
     8.8.   SUMMARY 	8-53
     8.9   REFERENCES	8-55

9.    EPIDEMIOLOGY STUDIES  	9-1

     9.1.   EPEDEMIOLOGIC STUDIES OF THE CARCINOGENICITY OF
           EXPOSURE TO DIESEL EMISSIONS	9-1
     9.2   COHORT STUDIES	9-2

           9.11.   Waller (1981): Trends in lung cancer in London
                  Relation to expsoure to diesel fumes	9-2
           9.2.2.   Howe et al. (1983): Cancer mortality (1965-1977)
                  in relation to diesel fume and coal exposure in
                   a cohort of retired railroad workers 	9-4
           9.2.3.   Rushton et al. (1983): Epidemiological survey
                  of maintenance workers in the London transport
                  executive bus garages and Chiswick works	9-6
                                   vm

-------
                 TABLE OF CONTENTS (CONT.)

      9.2.4.    Wong et al. (1985): Mortality among members of a
              heavy construction operators union with potential
              exposure to diesel exhaust emissions	 9-7
      9.2.5.    Edling et al. (1987): Mortality among personnel
              exposed to diesel exhaust 	9-12
      9.2.6.    Boffetta and Stellamn (1988): Diesel exhaust
              exposure and mortality among males in the American
              Cancer Society prospective study	9-14
      9.2.7.    Garschick et aL (1988): A retrospective cohort
              study of lung cancer and diesel exhaust exposure
              in railroad workers 	9-16

9.3    CASE CONTROL STUDIES OF LUNG CANCER	9-20

      9.3.1.    Williams et al. (1977): Associations of cancer
              site with occuaption and industry from the third
              national cancer survey interview  	9-20
      9.3.2.    Hall and Wynder (1984): A case-control study of
              diesel exhaust exposure and lung cancer   	9-21
      9.3.3.    Damber and Larsson (1987): Occupation and male
              lung cancer: A case control study in northern
              Sweden	9-23
      9.3.4.    Lerchen et al. (1987): Lung cancer and occupation
              in New Mexico 	9-25
      9.3.5.    Garschick et al. (1987): A case-contol study of
              lung cancer and diesel exhaust exposure in railroad
              workers	9-27
      9.3.6.    Benhamou et aL (1988); Occupational risk factors
              of lung cancer in a French case-control study	9-30
      9.3.7.    Hayes et  al. (1988): Lung cancer in motor exhaust-related
              (MER) occupations	9-32

9.4   CASE-CONTROL  STUDIES OF BLADDER CANCER	9-34

      9.4.1.    Howe et aL (1980): Tobacco use, occupation,
              coffee, various  nutrients, and bladder cancer	9-34
      9.4.2.    Wynder et al. (1985): A case-control study of
               diesel exhaust exposure and bladder cancer	9-37
      9.4.3.     Hoar and Hoover (1985): Truck driving and
               bladder cancer mortality in rural New England  	9-38

-------
                      TABLE OF CONTENTS (CONT.)

           9.4.4.    Iscovich et al. (1987): Tobacco smoking,
                   occupational exposure and bladder cancer in
                   Argentina	9-40
           9.4.5.    Steenland and Burnett (1987): A case-control
                   study of bladder cancer using city directories
                   as a source of occupational data  	9-43

     9.5   SUMMARY  	9-45
     9.6   REFERENCES	9-51

10.   UNIT RISK ESTIMATES	10-1

     10.1.  INTRODUCTION	10-1
     10.2.  REVIEW OF PREVIOUS RISK ESTIMATES 	10-4
     10.3.  RISK CALCULATIONS BASED IN ANIMAL BIOASSAY
           DATA  	10-10

           10.3.1.   Data Available for Risk Calculations	 10-10
           10.3.2.   Calculations of Unit Risk 	10-10

     10.4.  DISCUSSION	10-20
     10.5  WEIGHT OF EVIDENCE 	10-23
     10.6.  SUMMARY  	10-27
     10.7  REFERENCES	 10-29
APPENDIX A: Emission Standards, Exhaust Components, Emissions Models ...  A-l

APPENDIX B: Animal Carcinogenicity Summary Table	  B-l

APPENDIX C: Models for Calculating Lung Burdens	  C-l

APPENDIX D: Extrapolation Dosimetry Model	  D-l

-------
                              LET OF TABLES
TABLE 2-1    Levels of Emissions from Current Diesel and
              Gasoline Engines (FTP cycle only)	3-5

TABLE 2-2    Emission Rates of Volatile Organic Compounds (VOC)
              from Diesel and Gasoline Engines 	3-7

TABLE 2-3    Paniculate Matter Emission Rates and Their
              Distribution Between Total Carbon (TC) and Organic
              Carbon (OQ for Heavy- and Light-Duty Diesel and
              Gasoline Engines	3-13

TABLE 2-4    Summary of Composition and Emission Rates (rag/km) of
              Airborne Paniculate Matter from On-Road Vehicles,
              Tuscarora Mountain Tunnel  1977 Experiment	3-15

TABLE 2-5    Gasses of Organic Compounds Identified in Particulate-
              Phase Combustion Emissions	3-17

TABLE 2-6    Polycyclic Aromatic Hydrocarbons Identified and
              Quantified in Extracts of Diesel Particles 	3-19

TABLE 2-7    Emission Rates of Particle-Bounded PAH from Heavy-
              and Light-Duty Diesel and Gasoline Engines  	3-21

TABLE 2-8    Concentrations of Nitro-PAH Identified in a
              LDD Paniculate Extract	3-24

TABLE 2-9    Factors Affecting Rate of Emission of Polycyclic
              Aromatic Hydrocarbons in /ig/mile from Diesel
              Engine Exhaust and Mutagenicity	3-27

TABLE 2-10   Vapor Pressures at 25'C for a Series of PAH	3-29

TABLE 3-1    Calculated Atmospheric Lifetimes for Gas-Phase
              Reactions of Selected Compounds Present in Automotive
              Emissions with Atmospherically Important Reactive
              Species   	3-37
                                      XI

-------
                         LIST OF TABLES (CONT.)
TABLE 3-2



TABLE 3-3


TABLE 3-4

TABLE 3-5


TABLE 4-1


TABLE 4-2


TABLE 4-3


TABLE 4-4



TABLE 5-1


TABLE 4-1


TABLE 4-2


TABLE 4-3


TABLE 4-4


TABLE 4-5
Summary of the Nitroarenes Produced from the Gas-Phase
OH Radical-Initiated and N2O5 Reactions and
Electrophilic Nitration of PAH	 3-52

Average Atmospheric Lifetimes of Particles Due to
Dry Deposition  	3-60

Examples of Dry Deposition Velocities	3-61

Mean Particle, Gas, and Overall Scavenging Ratios
for Neutral Organic Compounds	3-64

Concentrations of Individual Hydrocarbons and
Aldehydes Measured in Raleigh, NC	3-67

Particle- and Vapor-Phase PAH Concentrations for
Baltimore Harbor Tunnel Samples	3-69

Average Ambient Concentrations of PAH Measured in
Glendora, CA 	3-71

The Maximum Concentrations of Nitrofluoranthene (NF)
and Nitropyrene (NP) Isomers Observed at Three South
Coast Basin Sampling Sites	3-73

Contribution of Nitrofluoranthene (NF) Isomers to the
Direct Mutagenicity of Ambient Paniculate Extracts	3-91

Composition of the exposure atmospheres in the
EPA studies	4-11
Composition of the exposure atmospheres in the
EPA studies	
Composition of exposure atmospheres in the
Lovelace studies	
4-13
4-15
Composition of exposure atmospheres in the
Ishinishi study  	4-17

Composition of the atmosphere in the Lewis study	4-19
                                     XII

-------
                          LIST OF TABLES (CONT.)

TABLE 4-6    Composition of the exposure atmosphere in the
              Heinrich 1982 study	4-22

TABLE 4-7    Composition of the exposure atmosphere in the
              Heinrich 1986 study	4-22

TABLE 4-8    Composition of exposure atmospheres in the
              Mauderty et ah, 1987a study	4-28

TABLE 4-9    Composition of atmosphere in Campbell et al. study	4-34

TABLE 4-10   Emission rates for diesel and gasoline engines	4-51

TABLE 4-11   Exposure atmosphere for EPA chronic toxicity study	4-53

TABLE 4-12   Short-term toxicity of diesel exhaust to laboratory
              animals  	4-56

TABLE 7-1    Gas-phase components of control and diesel exhaust
              exposure atmospheres  	7-3

TABLE 7-2    Tumor and Lung paniculate matter (soot) burden in male
              and female F344 rats exposed to diesel exhaust for
              24 mo  	7-5

TABLE 7-3    Gas-phase components of control (clean air) and Diesel
              exhaust atmospheres  	7-6

TABLE 7-4    Gas-phase components of control and diesel exhaust
              atmospheres	7-8

TABLE 7-5    Tumor incidence in female F344 rats, NMRI mice, and
              male and female Syrian golden hamsters following
              long-term (120 70 140 weeks) inhalation exposure to
              total diesel exhaust (4 mg particles/m3), filtered
              exhaust, or clean air	7-10

TABLE 7-6    Gas-phase components of diesel exposure atmospheres	7-12

TABLE 7-8    Tumor incidence in male and female F344 rats following
              long-term (30 mo) inhalation exposure to diesel
              exhaust   	7-14
                                      xui

-------
                           LIST OF TABLES (CONT.)

TABLE 7-9    Tumor incidence in ICR and C57B6 mice following
              long-term (24 mo) inhalation exposure to diesel
              exhaust and in rats exposed to diesel exhaust with
              or without DIPN treatment  	7-18

TABLE 7-10   Pulmonary lesion incidence in female Wistar rats
              following long-term (24 to 30 mo) inhalation exposure
              to unfiltered diesel exhaust	7-20

TABLE 7-11   Tumor incidence in male and female F344 rats and
              Syrian golden hamsters following long-term inhalation
              exposure to filtered or unfiltered diesel exhaust
              with or without DEN pretreatment	7-23

TABLE 7-12   Analysis of fresh air control (FA), diesel exhaust
              (DE), coal dust (CD), and coal dust plus diesel exhaust
              (DECD) atmospheres	7-26

TABLE 7-13   Concentration of gas-phase components and particles
              in exposure chambers	7-29

TABLE 7-14   Lung tumor incidence in strain A mice exposed for 8
              weeks to raw or irradiated exhaust at a particle
              concentration of 6 mg/m3	7-31

TABLE 7-15   Lung tumor incidence in strain A mice exposed to diesel
              exhaust at a  particle concentration of 6 mg/m3 until
              9 mo of age	7-32

TABLE 7-16   Effects of inhalation exposure to diesel exhaust on
              lung tumor incidence in male and female SENCAR mice  	7-34

TABLE 7-17   Lung tumor incidence in strain A mice exposed to a
              particle concentration of 12 mg/m3	7-35

TABLE 7-18   Tumor incidence in male and female F344 rats
              chronically exposed to whole exhaust from LD and HD
              engines	7-37

TABLE 7-19   Tumor incidence and survival time of rats treated
              with fractions from diesel exhaust condensate
              (35 rats/groups) 	7-39
                                      XIV

-------
                           LIST OF TABLES (CONT.)

TABLE 7-20   Pulmonary adenoma response of strain A/J mice exposed
              to diesel exhaust (1500 /ig/m3 particles)) for
              three mo .......................................... 7-42

TABLE 7-21   Tumor incidence in female Wistar rats treated with
              diphenylnitrosamine (DNP) and exposed to clean air
              (CA), filtered diesel exhaust (FDE), or whole diesel
              exhaust (DE) ....................................... 7-43

TABLE 7-22   Tumorigenic effects of dermal application of acetone
              extracts of diesel exhaust .............................. 7-47

TABLE 7-23   Dermal, tumorigenic and carcinogenic effects of various
              emission extracts   .................................... 7-51

TABLE 10-1   Paniculate lung burden and lung tumors in rats
              following a 2-year exposure to titanium dioxide (TiOj)
              and diesel exhaust  .................................... 10-2

TABLE 10-2   Estimated lifetime risk of cancer from inhalation of
                     diesel paniculate matter  ........................ 10-9
TABLE 10-3  Incidence of lung tumors in Fisher 344 rats (males and
              females combined) exposed to diesel exhaust in air .......... 10-11

TABLE 10-4  Incidence of lung tumors in Fisher 344 rats (males and
              females combined) exposed to heavy duty engine
              exhaust  .......................................... 10-12

TABLE 10-5  Incidence of lung tumors in Fisher 344 rats (males
              and females combined) exposed to diesel exhaust in air
              (Brightwell et al., 1986)  .............................. 10-13

TABLE 10-6  Combined dose-response data from four studies
              conducted on Fisher 344 rats  .......................... 10-14
                                       xv

-------
                              LIST OF FIGURES

FIGURE 2-1   Typical Size Distribution of Diesel Exhaust Particles	 3-11

FIGURE 2-2   Example of Human Respiratory System Aerosol Particle
              Deposition Curve  	3-12

FIGURE 2-3   Vapor/Particle Phase PAH Distribution in Samples
              Collected in Baltimore Harbor Tunnel 	3-31

FIGURE 3-1   Diesel-Derived Pollutants:  Emission-to-Deposition
              Atmospheric Cycle	3-33

FIGURE 4-1   Mass Chromatograms of the Molecular Ion of the
              Nitrofluoranthenes (NF) and Nitropyrenes (NP) Formed
              from the Gas-phase Reaction of Fluoranthene and Pyrene
              with the OH Radicals and Present in Ambient Paniculate
              Sample Collected at Torrance, CA	3-75

FIGURE 4-2   Mass Chromatograms of the Molecular Ion of the
              Nitrofluoranthenes (NF) and Nitropyrenes (NP) Present
              in Ambient Paniculate Samples Collected in Torrance,
              CA and Claremont, CA	-.	3-77

FIGURE 5-1   Protocol for Bioassay-Directed Chemical Analysis	3-86

FIGURE 5-2   Distribution of Direct-Acting Mutagenicity (TA98,
              -S9) Between Moderately Polar and Polar Fractions of
              Extracts of Paniculate Matter in Ambient Air, Wood
              Smoke, and in the Exhaust from Heavy-Duty and Light-Duty
              Motor Vehicles  	3-89

FIGURE 6-1   Metabolic pathway for B[a]P and formation of ultimate
              carcinogenic intermediate	6-3

FIGURE 6-2   Possible metabolic pathways for in. vivo
              biotransformation of 1-NP  to 6-hydroxy-N-
              acetyH-aminopyrine	6-10

FIGURE 6-3   Proposed pathway for cellular activation of
              1-nitropyrene (1-NP)  	6-16
                                      XVI

-------
FIGURE 6-4  Electron micrograph of an aggregate of particle-filled
              AM in an alveolus on day 106 following the deposition
              of 6.8x10* microspheres  	6-24

FIGURE 8-1  Clearance of insoluble particles deposited in
              tracheobronchial and pulmonary regions	8-4

FIGURE 8-2  Short-term thoracic clearance of inhaled particles as
              determined by model  prediction and experimental
              measurement	8-5

FIGURE 8-3  Lung retention kinetics of 2.13 mm diam. polystyrene
              microspheres after instillation into rat lungs 	8-14

FIGURE 8-4  Estimated number of AM with an indicated particle load
              of microspheres over a 174 d period following the
              instillation of 1.6xl07 microspheres	8-16

FIGURE 8-5  Relationship between the slope (exponents) of later
              phase AM disappearance relative to the particle loads
              in the AM	8-17

FIGURE 8-6  Relationship of the rates (exponents) of early
              component disappearance of AM relative to particle
              loads in the AM	8-18

FIGURE 8-7  Proportions of AM with a given load of particles that
              disappeared from the total AM population by the early
              disappearance components as of a theoretical day 0
              postinstillation time and as of day 7 after the particles
              were administered  	8-19

FIGURE 8-8  Percentage distribution of the retained lung burden
              in the various particle-AM categories over the course
              of lung clearance of the 1.6xl07 microsphere initial
              burden	8-21

FIGURE 8-9  Estimated numbers of AM with a given particle burden
              over the course of clearance of 2.0x10*
              microspheres	8-23
                                        xvu

-------
FIGURE 8-10 Percentage distributions of retained lung burdens in
              the various particle-AM categories during the clearance
               of the 2.0x10* initial lung burden of
              microspheres	8-24

FIGURE 8-11 Estimated numbers of AM in the indicated particle-
              containing categories following the deposition of
              6.8x10* microspheres into the rat's lung	8-25

FIGURE 8-12 Electron micrograph of an aggregate of particle
              filled AM in an alveolus on day 106 following the
              deposition of 6.8x10* microspheres	8-26

FIGURE 8-13 Percentages of the retained lung burden contained in
              the various particle-AM categories following the
              deposition of 6.8x10* microspheres	8-27

FIGURE 8-14 Micrograph of cells lavaged from a lung 86 d after
              the deposition of 6.8x10* microspheres 	8-29

FIGURE 10-1 Lung burden (organic material, mg) in rats, under exposure scenario
              (7 h/d5 d/week) in Mauderly et al. (1987)  	 10-15

FIGURE 10-2 Cumulative lung burden (particles, mg) in rats, under exposure
              scenario (7 h/d,5d/week) in Mauderly et al. (1987)  	10-17

FIGURE 10-3 Lung burden (organics, mg) in human adults over years after
              exposure 	 10-18

FIGURE 10-4 Effect of increasing the mitotic rate of initiated cells on
              the tumor incidence rate over age when other parameters are held
              constant	10-22
                                       xvui

-------
Principal authors of this document are:

Dr. Chao Chen
Carcinogen Assessment Statistics and Epidemiology Branch
U.S. Environmental Protection Agency
Washington, D.C.

Dr. Aparna Koppikar
Carcinogen Assessment Statistics and Epidemiology Branch
U.S. Environmental Protection Agency
Washington, D.C.

Dr. Bruce E. Lehnert
Pulmonary Biology-Toxicology Section
Los Alamos National Laboratory
Los Alamos, NM

Dr. Kumar Menon
Albert Einstein Medical Center
Philadelphia, PA

Dr. Gunther Oberdoerster
Department of Biophysics
University of Rochester
Rochester, NY

Dr. Dennis Opresko
Health and Safety Research Division
Biomedical and Environmental Information Analysis
Oak Ridge National Laboratory
Oak Ridge, TN

Dr. William Pepelko
Carcinogen Assessment Toxicology Branch
U.S. Environmental Protection Agency
Washington, D.C.

Dr. Larry Valcovic
Genetic Toxicology Assessment Branch
U.S. Environmental Protection Agency
Washington, D.C.
                                       XIX

-------
Dr. Ronald K. Wolff
Lilly Research Laboratories
Greenfield Laboratories
Greenfield, IN

Dr. Robert A. Young
Health and Safety Research Division
Biomedical and Environmental Information Analysis
Oak Ridge  National Laboratory
Oak Ridge, TN

Dr. C. P. Yu
Department of Mechanical and Aerospace Engineering
Sate University of New York
Buffalo, NY
Project Managers:

Mr. William Ewald
Environmental Criteria and Assessment Office
U.S. Environmental Protection Agency
Research Triangle Park, NC

Dr. William Pepelko
Carcinogen Assessment Toxicology Branch
U.S. Environmental Protection Agency
Washington, D.C.

-------
APPENDIX A

-------
      TABLE A-l. NEW VEfflCLE STANDARDS SUMMARY
CARS1
The following itcndarda. up to 1975. apply only to gaaoline-fualed light-duty vehicle*.  Staadaxd* tor 1975 and liter apply
to both gaaoitne-futlad and diesel light-duty vehicles.
3*
Prior to
contra is
1968-69
'.370
1971
1972
1973-74
1973-76
1977*
1978-79
1980
1981
I982io.ii

l9Mio.u

198411.12
1983-8613
1987 & later13
IZST
PKm-^LUKt
7 -Mode
7 -Mode
CTS-73
7 -Mode
20-100 CIO
101-140
Over 140
7 -Mode
7 -Mode
CVS-72
CTS-72
CTS-73
CTS-73
CTS-73
CTS-73
CTS-73
CTS-73

CTS-73

CTS-73
CTS-73
CTS-73
HYTtftQ*
CARBONS
830 ppa
11 tpa
8.8 spa
410 ppa
330 ppa
273 ppa
2.2 spa
2.2 spa
3.4 spa
3.4 spa
1.3 gpa
1.3 gpa
1.3 spa
0.41 gpa
0.41 gpa
0.41 gpa
(0.37)
0.41 gpa
(0.37)
0.41 gpa
(0.37)
0.41 gpa
0.41 gpa
CAHBCTf ^X I PRS OF
MOHODCICE HllHJUiH
3.4Z 1000 ppa
80 gpa 4 gpa
87.0 gpa 3.0 gpa
2.3Z
2.0Z
1.3Z
23 spa
23 spa
39 gpa
39 spa 3.0 gpa
13 spa5 3.1 gpa
13 spa 2.0 gpa
13 spa 2.0 gpa
7.0 gpa 2.0 gpa
3.4 gpa7 1.0 gpa''9
3.4 gpa7 1.0 gpa8'8
(7.$) (l.O)8
3.4 gpa 1.0 gpa8'8
(7.8) (l.O)8
3.4 gpa 1.0 gpa8
(7.8) (l.O)8
3.4 gpa 1.0 gpa
3.4 gpa 1.0 gpa
DIZSTL EVATOAim
PARTICUUTE HYDHOCARBOITS3
.
-
-
-
2 g/test
2 g/taat
2 g/test
2.0 g/teat
6.0 g/teat
6.0 g/test
2.0 g/test
0.60 spa 2.0 s/taat
(2.6)
0.60 spa 2.0 g/test
(2.8)
0.60 gpa 2.0 g/test
(2.6)
0.60 gpa 2.0 g/test
0.20 Spa1* 2.0 g/tsst
                                                  A-l

-------
                                           AfTEHDXX A  (continued)


LIGHT-DUTY u^Hlf'liT5

1.   Standards do not apply to vahiclee wita anginas leaa than 30 CIO from 1998 through 1974.   Diaaal
     particulata standard* apply only to diesels.

2.   ^Different t*at procedures have been used siBca the early years at emission control which vary in
     stringency.  Tha appaaranca that tha atandarda were relaxed from 1971 to 1972 Is incorraet.  The 1972
     standards are actually oora strlntant bacauaa of tha greater stringency of tha 1972 procadura.

3.   Evaporative euaaions dataxminad by carbon trap Bathed through 1977. SHED procadura beginning in 1978.
     Applies only to gasoline-fueled vehiclea.

*.   Evaporative hydrocarbon  standard doaa not apply to off-road utility vahiclaa for 1971.

'.   Carbon Bonoxide stanaard for vehicles sold  in tha Stata of California is 9.0 gpm.

5.   Cars sold  in spaeifisd high altituda countias raquirad to aaab standarda at high altltuda.

7.   Carbon oonoxida standard can ba Maivad  7.0  gpa  far  1981-82 by tha CPA Adninistrator.

8.   Oxidas of  nitrogan standard can ba waivad  to 1.5  gpa for inammtiv*  taebnology or diaaal.

 9.   Ozidaa of  nitrogan standard can ba vaivad  to 2.0  gpa for Acaarican Motors Corporation.

 10.  Standarda  in paxanthaaaa apply to vahiclaa sold in specified high altituda counties.   Vehicles  eligible
      for a carbon nonoxlde weive for 7.0  gpa at law altituda are eligible for  a waiver  to  11  gpn at  high
      altituda.

 11.  Exemptions free) tha high-altitude standarda are provided for qualifying iow-perfozaance  vehicles.
 12.  Thaaa save nuaarical standarda apply to vehicles sold in high-altitude areas.   Standarda in parenthesis
      spply to heavy paaaaagar can sold in high-altitude araaa for tha 1984 nodal year only.

 13.  Thaae aana oaaarieal atandarda apply to vehicles sold in high-altitude araaa.  Exemptions froa tha high-
      altitude standarda are provided for cpiali£ying low-performance vehielaa.

 14.  EaiMions averaging nay fa* used to meat this standard.

 Additional Jteaaizementa
 No crankcasa emiaaiona are permitted  (sppliea to gaaoliae-fueled only).

 Emiasion control component maintenance is restricted to specified interval*.
  gpa - grama par mile
  CZO - cubic inch displace
  CV8-72 - constant volume sample cold start teat
  CVS-73 - constant volume aampla teat which includee cold and hot starts
  7-ooda - #137 second driving cycle test
  ppa - pasts par million

                                                      A-2

-------
   TABLE A.2. NEW VEHICLE STANDARDS SUMMARY -
LIGHT DUTY TRUCKS1
Th. following standard*, up to 1978. apply only to guoiina-fualad light-duty truck*.  Standard* for 1976 and later apply
io both gaaolina-fual.d and diaial light-duty truck*.
345
Prior to
control*

1968-69



1370
1371

:972


1973-74
1973-775
1978
1979-806
1981
1982-83 7


1984-667

1987 7

1988 & l«t.r7 (A)

(B)

TBT
^

7 -Mod*
7 -Mod.
CVS-75
7 -Mod*
30-100 CI3
101-140
Ov»r 140
7 -Mod.
7 -Mod.

CVS-72


CVS-72
CVS-73
CVS-75
CVS-75
CVS-75
CVS-75


CVS-73

CVS-73

CVS-75

CVS-75

HYDRO-
C AMOKS
SSO pp.
11 gp»
6.3 gp.
410 pp.
330 pp.
275 pp.

2.2 ipo
2.2 gpa

3 4 BM»
J . ^ •>!••

3.4 gp.
2.0 gp.
2.0 gp.
1-7 gp.
1.7 gp.
1 7 *«M
I./ gP.
(2.0)
0.80 gpa
(1.0)
0.80 gp.
(1.0)
0.80 gpa
(1.0)
0.80
(1.0)
CAUQR
HOBOXIPE
3.4Z
80 gp.
76.0 gp.
2.3Z
2.0Z
1.3Z

23 gpa
23 gpa


39 gp.

39 gpa
20 gp.
20 gp»
18 gp.
18 gpa

1> gPM
(28)
10 gp.
(14)
10 8PM
(14)
10 gP»
(14)
10 gp.
(14)
QQCXDE3 OF
HlTKLXjiH
1000 pp.
4 ItB
••"••
3.6 gp.

.


.



•

3.0 gp*
3.1 gpn
3.1 gp.
2.3 gpa
2.3 gp.

2.3 gpa
(2.3)
2.3 gpa
(2.3)
2.3 gp.
(2.3)
1.2 gp.
(1.2)
1.7 «P»9
(1.7)
DBSZL EVAPORATE
PAMieuiATE HYDEOCAMC
-
-

*
"



_ L
6 g/taat


2 c/taat
2 g/taat
2 g/t.at
6.0 g/taat
6.0 g/taat
2.0 g/t.at

0.60 gpm 2.0 g/t.at
(2.6)
0.60 gp. 2.0 g/taat
(2.6)
0.26 gpm8 2.0 g/taat
(2.6)
0.26 gpn8 2.0 g/taat
(0.26) (2.6)
0.26 gpa8 2.0 g/taat
(0.26) (2.6)
                                            A-3

-------
                                           APPENDIX A (continued)

L.I33I-DUTT TRUCKS

1.   Standard* do not apply to vehicle* with anginss  !•••  than 20 CIO Cram 1968 through 1974.   Diesel
     paniculate standard* apply only to  diaeeia.

2.   2iffarent teat procedures have been ui«d tine* the early year* of emission control which vary in
     stringency.  The appearance that the atandard* were relaxed from 1971 to 1972 in incorrect.  The 1972
     standards are actually more stringent because of the  greater stringency of the 1972 procedure.

3.   Evaporative eeuaaiona determined by carbon trap method through 1977, SHED proeedure beginning in 1970.
     Applies only to gaaoliaa-fueled vehicles.

<•.   Evaporative standard does not apply to off-road utility vehicles for 1971.

5.   Trucks  sold in  specified high  altitude  countlea required  to smet standarda at high «i»-«t»^.  (1377
 5.   Effective in 1979, light-duty truck claaaificatlon wee extended from 0-6000 pounos GVW to 0-8300 pounds
     5VWR.

     Standards in perenthesis apply  to veiucies sold in specified high  altitude counties.   Izsoptions from
     the high altitude  etandarda are provided  for up to 301 of the high-altitude prooact line  lot 1982-84;
     tzemptions  for qualifying lev-performance trucks are provided for  1983 and later.

 3.   -missions averaging  may be used to meet this standard provided  that trucks produced for sale in
     California  or deelgaated high-altitude areas may be averaged only  within each of these arias.
 ».   Standards o£ 1.2  gpm apply to LDIs up to and InflwUnf. 3,730 pound* loaded vehicle weight;  1.7 gpa
      standard applies to LOT* equal and over 3.731 Ib* loaded vehicle weight.  Eoiasions averaging may be
      used to meet this standard provided that trucks produced tor sale in California or deaignated high-
      altitude areas mar be averaged only within each of those areas.  Dteael and gasoline-fueled engine
      families stay not be averaged together.
 Additional Rtoutrtments

 No crankeaae soaaaiona are permitted from gaaoline-fueled trucka: applies to high-altitude trucks beginning
 in 1982. dissel-powered trucka beginning in 1984.

 A 00 standard of 0.30X at idle is established for 1984 end later model years: effective at high altitude
 starting with the 1988 model year.

 Full useful life ia established  for 1983 and later model yean, defined  ae  11 years/120.000 mile*.  For 1984,
 a number of useful life option*  are available; prior to 1984, useful life i* 3  yean/30.000 miles.

 emission control component maintenance is restricted to specified interval*.
  gpm -  grams per mile
  CIO -  cubic inch displacement
  CVS-72 -  constant volume aample cold start teat
  CV3-7S -  constant volume sample test which Includes  cold and hot  start*
  7-oodo -  #137 second driving cycle teat
  ppm -  part* per million
  CrVMt - gross vehicle weight rating

                                                     A-4

-------
TABLE A3. NEW VEHICLE STANDARDS SUMMARY - DIESEL HEAVY-DUTY ENGINES
              „
                          „,
                                            m,in*
                                    CAMOH
  1970-73
  197*-78
  :379-»
          ««


1994 « i.t.r
                    1.3 ,/bhp-hr
                    1.3 ,/bhp-hr
                    0.3 ,/bhp-hr
                                    *0 »/bph-hr
                                    23 ./bh|r.te
                                    23 g/bhp-hr
                                    23,/bhp-hr
                                    23
13.3 ,/bhp-hr
13.3 ,/bhp-hr
                   1.3 ,/bhp-hr     13.3
                   1.3 ,/bhp-hr     13.S
                   1.3 ./bhp-hr     13.S
                                                    OQODES OF
                               IR0BOCARBONS
                               & OXIDES OF   DIZSEL
                                                                  16 ,/bhp-hr
                              10 ,/bhp-hr
                               ,  JW .
                               5, /bhp-hr
                              10 ,/bhp-hr
                               <  ,_,_
                               3 ,/bhp-hr
10.7 ,/bbp-hr
 9.0 ,/bhp-hr
                                                   10.7 ,/bbp-hr
                                                    6.0. /bhp-hr
                   1.3 ,/bhp-hr      13.3 ,/bhp.te     3 Q ,,,^^3
                                                                              0.60. /bhp-hr
                                             ACCZL tOZ
                                             LU3  20Z

                                             ACCEL 2QZ
                                             MB  iiz
                                             PEAK  :sz

                                                  20Z
ACCEL 20Z
LOS   13Z
PEAK  30Z

ACCEL 20Z
UE   13Z
PEAK  20Z

ACCEL 201
UKJ   isz
PEAK  SOZ
                                                                                                    20Z
                                                                                               UB  13Z
                                                                                               PEAK  30Z
                                           0.10 ,/bhp-hr*    LOB   isz
                                                           PEAK  SOZ

                                           0.10 ,/bhp-hr5    ACCEL  20Z
                                                           LW   isz
                                                           PEAK  30Z
                                                    A-5

-------
                                           APPENDIX A (continued)
DIESEL HEAVY OUR ENGIKES

1.   Tsst procedure for 1970-1983  standards  is  ths 13-ooda steady-state  teat procedure,   list procedure fox
     1984 is either steady-state or the EPA transient teat procedure.  For 1983 and liter, the EPA transient
     test procedure is used.

2.   At the manufacturers' option, either the 1983 standards and test procedures nay be used (option A), or
     standards of 1.3 BC. 15.3 CO. and 10.7 HQx en the transient test procedure (option B).  Also, standards
     at 0.3 IE. 13.3 CO and 9.0 HOz are optional standards for 1984 diesals tested on the 13-oode test
     procedure.

3.   Emissions averaging »«y be used to meet this standard, but these emissions nay not be averaged with HD
     gasoiiae engine emissions.  Averaging is restricted to within useful life subclasses (see below).  Also.
     averaging is restricted rtgionaily •- the two regions are California and the other »9 states.

     Tor urean ous engines, tr.i standard is 0.10 g/bhp-hr -- partlculate averaging is not allowed with  this
     standard, but •missions from these engines may be used in NQx averaging.

 5.   Emissions averaging aay be used to aeet this standard.  However averaging is restricted to within  useful
     life  subclasses  (see below).  Also, averaging is restricted regionally—the two regions are California
     and the other *9 states.  Emissions tram  engines used in urban bueee stay not be included  in  the
     averaging program.
 Additional ReemiremeBts

 No crankcase Missions are permitted starting in 1984.   This does not apply to turboeharged engines or
 engines whose intake air is inducted solely by pumps,  blowers,  or superchargers.

 Full useful life is.established for 1983 and later defined as:

      Light heavy-duty (normally under 19.300 Ibs GVWJO  — • yrs/110.000 Biles
      Medium heavy-duty (normally 19.300-33,000 Ibs GVHR) — S yrs/183.000 miles
      He*vy heavy-duty (normally over 33.000 Uas GVHR)   — 8 yn/290.000 Biles

 For 1984  (option £ only), a number of useful life options are available: for 1984 (option A) and prior years.
 useful life is 5 yew/30,000 miles.

 Emission  control component maintenance is restricted to specified intervala.
 g/bhp-hr -  grams per brake horsepower - hoax
 ppm - parts per million
  13-mode test procedure - dlesel engine dyn••muter teat with 13 steady-state test points
  transient test procedure -   tngine dynamometer procedure with  starts,  stops, and speed/load changes
                                                     A-6

-------
   MODIFIED EPA NAAQS EXPOSURE MODEL (NEM) FOR CO

     The following description of  the modified NAAQS Exposure Model  (NEM) for
CO is based on information given by Carey (1987)  and Ingalls (1985).
     The original NEM was based on an activity model  that  simulated  a set of
age/occupation population groups called cohorts.   In  the model  these cohorts
are referred to a specific location type (neighborhood/microenvironment)  for
each hour of the day.  Each of the several specific location types in the
urban area are then assigned a particular ambient pollutant  concentration
based on fixed site monitoring data.  The model computes the hourly exposure
for each cohort and  then  sums up  these values over the desired average time to
arrive  at  average population  exposures and exposure distributions.  Monitoring
data  for a full year arc  input  into  the  model to allov  for the computation of
annual  averages.
      The NEM methodology vas  modified to allov for the  estimation of  exposures
 from mobile source pollutants and to allov for the inclusion of data  for
 short-term exposures to high  concentrations such as might occur in  on-road
 vehicles.   The NEM for CO vas used because outdoor CO vas considered to  result
 almost exclusively from mobile sources and exposure  to other non-reactive
 mobile source pollutants could also be modeled using this relationship.
      The CO monitoring data vere used to provide CO concentrations for each
 neighborhood and most of the microenvironments.   For each location, CO emis-
 sion factors (in grams/minute) vere  chosen  to best represent vehicle condi-
  tions  in  that area.  The emission factor is a fleet average emission factor,
  thus veighting emissions from  both  LDO  and  HDD vehicles.  The model  ratios the
  CO  concentrations and  appropriate emissions factors  for each location so each
                                      A-7

-------
location contains a factor expressed as ug/mVgrams/minute.  Emissions factors
in grams/minute for the pollutant of interest for each location are input to
the model.   The model multiplies the input emission factor (grams/minute) by
the second factor (ug/3/grams/minute) to obtain concentrations in each loca-
tion for the pollutant of interest.
     The modified NEH model does not account for photochemical reactions
occurring in the atmosphere.  The model also assumes  that  the pollutant of
interest has emission formation and dispersion characeristics similar  to  that
for CO.
     The model accounts  for indoor  exposure  to mobile source-generated pollu-
tants.  A scaling  factor of 0.85 was applied to  the  CO monitoring data for  a
given  location.  The scaling  factor vas based on comparisons of  indoor and
outdoor CO  levels  for homes vith no indoor CO sources.
     In the modiified NEH procedure indocr CO exposures were set at zero, and
 three  specific mobile source  microenvironments  vere added (street canyons,
 tunnels,  and parking garages).   In addition, an extrapolation procedure  vas
 developed  to estimate nationvide exposures.
      Inputs into the model included population data for urban and rural areas,
 emission factors,  and ambient pollutant concentrations specifying the 24 con-
 centration intervals or bins for vhich the cumulative person-hours of exposure
 vere to be calculated.  The output lists the total annual person-hours of
 exposure found in each  of the specified concentration intervals.  Population
 data vere derived from  U.S. Department of Commerce  information.   The  emissions
 data included FTP emission factors (constant volume test  procedure  using both
 cold and hot starts, resulting in  an average speed  of 19.7  mph)  for veekday
 urban  residential and suburban vehicle use.  For  veekday urban  commercial  and
 industrial  vehicle  use  and for street canyons,  an emission factor for an
                                      A-8

-------
average speed of 10 mph,  expressed in grams/minute, was included.  For the
tunnel microenvironment and rural areas, a 35 mph steady-state emission factor
vas used.  For LDD vehicles an idle emission factor vas also  included.  Week-
end emission factors vere based on a fraction of the weekday  values. These
emission factors vere chosen to best represent  the vehicle operating condi-
tions  for each neighborhood or microenvironment.
     The emission  factors for each model year for  each vehicle type vere  com-
bined  into a single, weighted calendar  year  emission  factor for each  vehicle
 type.   Each model  year's emission factor vas multiplied by that model year's
 fraction of  the  calender year VMT (vehicle miles driven) and the diesel sales
 fraction for  that  model year and  then summed across 20 model years to derive
 the annual  weighted emission  factor for each vehicle type.  These factors vere
 then used  to derive a composite emission factor for the entire fleet.
      The model is highly dependent on  the assumptions used to deriva the emis-
 sion factors,  particularly those for the non-FTF situations.  The modified
 model also does not take into account  projected changes in VMT.  Since the
 model vas based on 1981 monitoring data,  projected VMT values vere compared to
 1981 VMT values, and exposures predicted  by the model vere adjusted accor-
 dingly.
      The mean exposure levels predicted by  the model, adjusted to take into
 account projected increases in VMT,  are shovn  belov:

                         	 Urban           Rural            Nationwide
  1986
  1995  (high sales)
  1995  (lov sales)
2.63 ug/m3
1.27 ug/m3
1.69 ug/m3
2.38 ug/m3
1.06 ug/m3
1.27 ug/m3
2.56 ug/m3
1.22 ug/m3
1.58 ug/m3
                                       A-9

-------
     Rural exposure are smilar to urban exposures due to the greater fraction
of some HDD vehicle classes in rural areas.  Hovever, according to Carey
(1987), the values calculated for rural areas should be considered only rough
estimates.

                                  REFERENCES
Carey, P. M.  (1987)  Air  toxic emissions from motor vehicles.   Ann  Arbor,
     MI: Office of Mobile  Sources,  U.S. Environmental  Protection Agency.
     EPA Report no.  EPA-AA-TSS-PA-86-5.

Ingalls. M. N.   (1985)   Improved  mobile source exposure estimation.  Ann Arbor,
     HI: Office  of Mobile  Sources,  U.S. Environmental  Protection Agency.
     EPA Report  no.  EPA-460/3-85-002.
                                       A-10

-------
  TABLE A-4. ORGANIC COMPOUNDS FOUND IN DIESEL EXHAUST
Aceanthrone
Acenaphthenol
Acenaphchylene
Acephenan throne
Acetaldehyde
Acetylaminofluorene
Acetylene
1-Acetylnapthalene
Acrolein
Anthanthrene
Anthracene
Anthracene carboxaldehyde
9-Anthracene carboxaldehyde
Anthracene dicarboxylic
  acid anhydride
Anthracenedione
Anthracene quinone
9,10-Anthracene quinone
Anthracenone
Anthranthene
Anthroic acid
Anthrone
9-Anthrone
Anthroquinone
Benz(j)aceanthrylene
Benz(e)acephenanthrylene
 Benzacridine
 Benzaldehyde
 Benz(a)anthracene
 Benzanthracene dicarbox-
   aldehyde
 Benz(a)anthracene-7-12-
   dione
 7H-Benz(de)anthracene-7-one
 Benzenamine
 Benzene
 Benzene, l-(l,l-dimethyl-
   ethyl)4-raethyl)
 1,4-Benzenediol
 Benzene, ethenylmethyl-
 1,2-Benzene dicarboxylic
   acid, dibutyl ester
  1,2-Benzene dicarboxylic
   acid, dimethyl ester
  Benzene, dimethyl-
  Benzene, ethyl-
  Benzene, methyl-
  Benzene,  (1-methylethenyl)-
Benzene, (1-methylethyl)-
Benzene, l,l'-oxybis-
Benz(j)fluoranthene
Benzo(a)anthracene
  carboxaldehyde
Benzo(a)anthracenequinone
7,12-Benzo(a)anthracene
  quinone
Benzoanthrone
7-Benzo(de)anthrone
Benzo(dh)anthrone
Benzo(b)chrysene
Benzo(c)cinnoline
Benzo(gni)fluoranthene
Benzo(k)fluoranthene
Benzo(j)fluoranthrene
Benzo(k)fluoranthrene
Benzofluorene
Benzo(c)fluorene
Benzo(a)fluorenone
H-Benzo(a)fluorenone
Benzofuran
Benzo(e)perylene
Benzo(gni)perylene
Benzophenanthracene-
   carboxaldehyde
 Benzo(c)phenan threne
 1-Benzopyran-A(AH)-one
 2H-l-Benzopyran-2-one
 Benzo(a)pyrene
 Benzo(e)pyrene
 Benz o(e)pyrene-A,5-d ihydrod io1
 Benzo(e)pyrene-9,10-epoxide
 1,6-Benzo(a)pyrenequinone
 3,6-Benzo(a)pyrenequinone
 6H-Benzo(cd)pyrenone
 Benzo(ed)pyrenone
 Bicyclo(2.2.1)heptan-2-one
   1,7,7-trimethyl
 Binaphthalene
 Binaphthyl
 1,1-Biphenyl
 Biphenylcarboxaldehyde
 1,1'-Biphenyl-4-carbox-
   aldehyde
 Biphenyldicarbonitrile
 Biphenylene
 1,1'-Biphenyl, 2-methoxy-
                                    A-ll

-------
                  ORGANIC COMPOUNDS  FOUND IN DIESEL EXHAUST
[l,l'-Biphenyl)-2-ol
[l,l'-Biphenyl]-4-ol
1,3-Butadiene
Butanal
2-Butanone
1-Butene
2-Butene
Butyl-2-methylpropyl-
  phthalate
Butylnaphthalene
n-Butyraldehyde
o-Chlorophenol
Chrysene carboxaidehyde
Chrysene quinone
Coronene
Croconaldehyde
Cyclodexanol
2,5-Cyclohexadiene,  1,4-dione
Cyclohexane
Cyclohexane,  methy1-
Cyclohexanol
Cycloh-xanone
Cyclohexanone, raethyl-
Cyclohexene
2-Cyclohexen-l-one,
   3,5,5-trimethyl
Cyclopentacoronene
Cyclopentano(c,d)-pyrene
Cyclopen tano(c, d)pyrene-3,4-
   cis-diol
Cyclopentano(c,d)pyrene-3,4-
   oxide
Cyclopentano(c,d)-pyrene-
   3,4-trans-diol
 4H-Cyclopenta(def)phenanthren-
   4-one
 Cyclopen taphenan thren-5-one
 Cyclopen tapyrene
 Cyclopenta(cd)pyrene
 Cyclopentenodibenzopyrene
 Cyclopenteno(cfd)pyrene
 Cyclopenteno(c,d)pyrene
   anhydride
 Decalin
 Decane
 Diazomethane
 Dibenr(a,c)anthracene
 Di benz(a,j)anthracene
Dibenz(a,h)anthracene
  dicarboxaidehyde
Dibenz(c,g)carbazole
Dibenzo(def,mno)chrysene
Dibenzo(b,ej(l,4j-dioxin
1,2,3,4-Di benzofluoranthene
Dibenzofluorene
Dibenzo(a,g)fluorene
Dibenzofuran
1,2,3-Dibenzopyrene
Dibenzo(a,i)pyrene
1,2,9,10-Dibenzotetracene
Dibenzothiophene
2,6-Di-tert-butyl-4-methyl-
  phenol
1,2-Dihydroacenaphthylene
9,10-Dihydrobenzo(e)pyrene
4,5-Dihydrodihydroxy
  benzopyrene
Dihydrodihydroxy-
  fluoranthene
Dihydrodihydroxypyrene
5,6-Dihydrodiolbenzo(a)-
  anthracene
 1,6-Dihydrodiolbenzo(a)-
   pyrene
 4,5-Dihydrodiolbenzo(a)-
   pyrene
 2,3-Dihydroinden-l-one
 1,2-Dihydro t rime thy1-
   naphtnalene
 Dihydroxyanthracene
 9,10-Dihydroxyanthracene
 Dihydroxydimethyl-
   anthracene
 Dihydroxyfluorene
 Dihydroxymethoxybenzene
 Dihydroxymethylanthracene
 Dihydroxymethylfluorene
 Dihydroxymethyl
   phenanthrene
 Dime thyIan thracene
 9,10-DimethyIanthracene
 Dime thyIan thracene
   carboxaidehyde
 DimethyIanthrone
 Dimethylbenz(a)anthracene
 Dimethylbenzonaphthothiophene
                                    A-12

-------
                  ORGANIC COMPOUNDS FOUND IN DIESEL EXHAUST
Dimethylcyclopentacene-
  naphthylene
Dimethyldecalin
1,9-Dimethylfluorene
Dimethylfluorenequinone
Dimethylfluorenone
DimethyIhydroxyanthracene
Dimethylhydroxyfluorene
Dime thyIhydroxyphenan threne
Dimethylindan
DimethyInaphthalene-
  carboxaldehyde
Dimethylnaphthalene dicar-
  boxylic acid anhydride
Dimethylnaphthothiophene
Dimethylperhydrophen-
  anthrene
Dimethylphenanthrene
Dimethylphenanthrene
  carboxaldehyde
Dimethyl  phenanthrone
Dimethyltetrahydro-
  naphthalene
Dimethyltetralin
 2,7-Dinitrofluorene
 l,5-Dinitro-2-methyl-9,10-
  anthraquinone
 1,5-Dini tronaphthalene
 Dinitropyrene
 1,4-Dioxane
 Di phenylacenaph thylene
 Diphenylacetylene
 DiphenyIbenzene
 n-Docosane
 Dodecylcylohexane
 n-Eicosane
 Ethene
 Ethylfluorene
 1-Ethyl naphthalene
 p-Ethylphenol
 Ethyl tridecane
 1,l'-U,2-Ethynediyl)-
   bisbenzene
 Fluoranthene carbox-
   aldehyde
  Fluoranthenequinone
  Fluorantheno-2,3-dihydrodiol
  Fluoranthone
Fluoranthrene
9-Fluoren-a-one
9H-Fluorene
2-Fluorene carboxaldehyde
9-Fluorene carboxaldehyde
Fluorenequinone
1,4-Fluorenequinone
9H-Fluoren-9-one
9H-Fluoren-9-one,2,4,7-
  trinitro-
Formaldehyde
Formic acid
Formic acid, ethyl ester
2-Furancarboxaldehyde
2-Furanmethanol
Furan, tetrahydro-
n-Heneicosane
n-Heptadecane
1-Heptadecene
Heptane
n-Hexadecane
Hexaldehyde
Hexane
Hexanol
2-Hexanone
Hexone
Hydrocyanic acid
Hydroxyanthracene
 Hydroxyanthroic acid
 2-Hydroxybenzaldehyde
 5-Hydroxybenzo(a)anthracene
 6-Hydroxybenzo(a)pyrene
 7-Hydroxybenzo(a)pyrene
 Hydroxychrysene
 Hydroxycoronene
 Hydroxyfluorene
 9-Hydroxyfluorene
 Hydroxyfluorenone
 Hydroxynaphthalene dicar-
   boxy lie acid anhydride
 Hydroxynaphthalic acid
 Hydroxynaphthoic acid
 Hydroxynit ro fluorene
 1-Hydroxyphenanthrene
 Hydroxyphenanthrene
 Hydroxyphenanthroic  acid
 Hydroxyphthalic  acid
 Hydroxypyrene
                                     A-13

-------
                  ORGANIC COMPOUNDS  FOUND IN DIESEL EXHAUST
Hydroxytrimethylanthracene
Hydroxytrimethylphen-
  anthrene
Hydroxyxanthene
3-Hydroxyxanthen-9-one
Hydroxyxan throne
1-Indanone
Indeno(1,2,3-cd)fluoran thene
Isoamylfluorene
1,3-Isobenzofurandione
Isobutryaldehyde
Isobutylene
Isopentene
2-Isopropylnaphthalene
o-Isopropylphenol
He thoxybenazldehyde
Methoxybiphenyl
2-Methoxy,l,l'biphenyl
Hethoxyfluorene
He thoxyphenan t h rene
He thoxyxan thenone
Hethylacetylene
Me thyIan thracene
2-Me thyIan thracene
9-Metnylanthracene
Hethylanthracene
  carboxaldehyde
2-Methyl-9-anthracene
   carboxaldehyde
Hethyl-9-10-anthracenedione
Hethylanthracenequinone
Methyl-9-10-anthracenequinone
2-Methyl-9-10-anthracene-
   quinone
 Hethyl anthroic acid
 Methylanthrone
 Methyl-9-anthrone
 4-Hethylbenzaldehyde
 Methylbenz(a)anthracene
 Me thyIbenzanthrone
 Methylbenzoate
 7-Methylbenzofuran
 Methylbenzoic acid
 Me thyIbenzo(a)pyrene
 Methylbenzo(e)pyrene
 10-Methylbenzopyrene
 3-Methylbutanal
 2-Hethylbutane
Methylcarbazole
Hethyl caronene
3-Hethylcholanthrene
Methylchrysene
Methylcyclohexanone
Methyl-AH-cyclopenta(def)-
  phenan thren-4-one
Methyldecane
Methyldibenzothiophene
Methyldihydrofluoranthene
Methyldihydropyrene
Me thyIfluoranthenequinone
1-Methylfluorene
9-Methylfluorene
Hethylfluorenequinone
Methylfluorenone
2-Methylfluorenone
Hethyl-9-fluorenone
MethyIhydroxyanthracene
5-Methylhydroxybenzo-
   phenanthrene
Methylhydroxyfluorene
Methylhydroxyphenanthrene
Methylindan
Methylnaphthaldehyde
6-Me thy1-2-naph thaldehyde
Me thyInaphthalene
 Me thyInaph thalene
   carboxaldehyde
 HethyInaphthalene dicar-
   boxy lie acid anhydride
 MethyInaphthalic acid
 Hethylnaphthoic acid
 Hethylnitroanthracene
 Hethylnitropyrene
 Hethylpentadecane
 Methylperylene
 1-Methylphenanthrene
 2-Methylphenanthrene
 3-Hethylphenanthrene
 4-MethyIphenanthrene
 9-Methylphenanthrene
 Methylphenanthrene
   carboxaldehyde
 Hethylphenanthrenequinone
 Me thy1-9-10-phenanthrene-
    quinone
 Hethylphenanthroic  acid
                                    A-14

-------
                  ORGANIC COMPOUNDS FOUND IN DIESEL EXHAUST
Hethylphenanthrone
4-He thylphenyIbenzo(c)-
  cinnoline
Hethylphenylcinnoline
Methylphthalic acid
1-Methylpyrene
Methylpyrenequinone
Hethylquinoline
Methyltetralin
Hethyltriphenylene
Methylundecane
Honomethylaniline
Naphthalene
Naphthalene acetaldehyde
1-Naphthalene carboxaldehyde
2-Naphthalene carboxaldehyde
1-Naphthalene carboxylie acid
2-Naphthalene carboxylic acid
Naphthalene dicarboxaldehyde
1,8-Naphthalene dicarboxylie
   acid
Naphthalene dicarboxylie
   acid anhydride
 IH-Naphthalenequinone
 2-Naphthalenol
 Naphtho(1234tde£)chrysene
 Naphtho-2,3(b)furan-4,9-
   dione
 Naphthopyrandi one
 Naphtho(l,8-cd)pyran-l,3-
   dione
 1H,3H-Naphtho-l,8-cd-pyran-
   1,3-dione
 l-Naphtho(cd)pyrone
 C4-Naphthothi ophene
 9-Ni troanthracene
 6-Nitrobenzo(a)pyrene
 Nitrochrysene
 Nitroethane
 2-Nitrofluorene
 3-Nitro-9-fluorenone
 Nitromethane
 Nitronaphthalic acid
 Ni trophenanthrene
 Nitropyrene
  1-Nitropyrene
 Nitropyrone
 N-Ni trosomorpholine
n-Nonadecane
n-Octadecane
Octahydrophenanthrene
Octane
Oxiranemethanol
Oxirane, (phenoxymethyl)-
Oxi rane, [(propenyloxy)-
  methyl]
n-Pentadecane
n-Pentane
2-Pentanone
Pentaphene
1-Pentene
3-Penten-2-one,  4-methyl-
Perhydrophenanthrene
Perinaphthindenone
Peroxyacetyl nitrate
Peroxypropronyl nitrate
Perylene
Phenanthrene
Phenanthrene carbox-
   aldehyde
2-Phenanthrene carbox-
   aldehyde
 Phenanthrene-9-carbox-
   aldehyde
 Phenanthrene dicarboxylic
   acid anhydride
 Phenanthrenequinone
 9,10-Phenanthrenequinone
 Phenanthroic acid
 Phenanthrone
 Phenanthroquinone
 Phenol
 Phenol, 2-methyl-
 Phenol, 3-methyl-
 Phenol, 4-nethyl-
 Phenol, 2-nethyl-
    4,6-dinitro-
 Phenylbenzopyranone
 Phenylethyl ketone
 l-Phenyl-2,4-hexadiyn-l-one
 1-Phenylnaphthalene
 2-Phenylnaphthalene
 Phenylpyrocatechol
  Phenylpyrrolopyridine
  Phthalic acid
  Propanal
                                     A-15

-------
                  ORGANIC COMPOUNDS FOUND IN DIESEL EXHAUST
Propane                                   9H-Thioxanthen-9-one
Propanone                                 Thioxanthen-9-one
2-Propenoic acid,                         9-Thioxanthone
  ethyl ester                             Trihydroxyanthraquinone
2-Propenoic acid,                         Trihydroxyfluorene
  methyl ester                            Trimethylanthracene
2-Propenoic acid, 2-methyl-               Trimethylfluorene
  methyl ester                            Trinethylfluorenone
Pyrene                                    Trimethylindan
Pyrene carboxaldehyde                     Trimethylnaphthalene
1-Pyrene carboxaldehyde                   Trimethylnaphthalene
Pyrene quinone                              carboxaldehyde
3,10-Pyrene quinone                       Trimethylnaphthothiophene
Pyreno-3,4-dicarboxylic                     2,2,4-Trimethylpenta-l,3-
  anhydride                                 dioldiisobutyrate (bkg)
Pyrenone                                  Trimethylphenanthrene
Pyridine                                  2,3,5-Trimethylphenol
Pyrone                                    Trimethyltetralin
Riphenylene                               Trinaphthenebenzene
Styrene                                   Triphenylene  carboxaldehyde
n-Tetradecane                            Triphenylene  quinone
1-Tetradecene                            9H-Xanthen-9-one
1,2,3,4-Te t rahyd ronaph thalene            9-Xanthone
Tetrahydrophenanthrene                   2,6-Xylenol
Tetraoethylnaphthalene                   3,5-Xylenol
Thioxanthenone
 SOURCE:  Adapted from Opresko et al., 1984.
                                   References


 Opresko, D. M.; Holleman, J. V.; Ross, R. H.; Carroll, J. V.  (1984)  Problem
      definition study on emission by-product hazards from diesel engines for
      confined space army workplaces.  Oak Ridge, TN: Oak Ridge National Lab-
      oratory,  ORNL Report no. 6017.
                                      A-16

-------
          EPA MOBILE SOURCE EMISSIONS MODEL (MOBIL3)



     The Mobile Source Emission Model  (MOBILES)  is an integrated  set  of  FOR-

TRAN routines which are used in the analysis of  the impact of highway mobile

sources on air quality (USEPA 1984).  MOBILES is an updated modification of

the earlier MOBILE1 and MOBILE2.   A basic description of the model is given in

the User's Guide to MOBILE2 (USEPA, 1981), and the following information is

taken from that report.

     The MOBILES program calculates emission factors for the following

categories of vehicles:


         Light duty gasoline-powered vehicles (LDGV).
         Light duty gasoline-powered trucks, GW  < 6001  Ib  (LDGT1).
         Light duty gasoline-powered trucks, GW  6000-8500  Ib  (LDGT2).
         Light duty gasoline-powered trucks, GW  <8,501  Ib  (LDGT3).
         Heavy duty gasoline-powered vehicles (HDGV).
         Light duty diesel-powered vehicles  (LDDV).
         Light  duty diesel-powered trucks (LDDT).
         Heavy  duty diesel-powered vehicles  (HDDV).
         Motocycles (MC).
         All  vehicles combined.


      The MOBILES program computes emission factors for hydrocarbons  (HC),  car-

 bon monoxide (CO), and oxides of nitrogen (NO ).   Three major geographic
                                              J\

 regions are considered: (1) low altitude, non-California regions; (2) low

 altitude California regions; and (3) high altitude, non-California regions.

 The program calculates emission estimates for January  1 of any calender year

 based on data for the  twenty  most recent model years.  The basic exhaust eais-

 sion data for the model is derived from tests conducted on vehicles operated

 under standardized  conditions.   The  basic  test  conditions for LDD and  HOD

 vehicles are given  in Table 0-1.
                                      A-17

-------
                  TABLE A-5. STANDARDIZED TEST CONDITIONS

Engine off period, cold start
Engine off period, hot start
Average trip length
Average trip speed
Average % idle
Average X VMTa cold start
Average % VMT stabilized
Average % VMT hot start
LDDV & LDDT
12-36 hr
10 man
7.5 mi
19.6 mph
18*
20. 6Z
52. 1*
27. 3*
HDDV
12+ hr
20 min
6.4 mi
19.2 mph
37*
U.3Z
OX
86. 7*
            Source:  USEPA,  1981

            a Vehicle miles traveled.
The MOBILE program estimates emissions  for each calender year by weighting the

emission factors for the twenty most recent model years based on the distribu-

tion of the total vehicle miles traveled (VMT) for  those model years.  The

program uses calculation procedures and emission factors presented in  the 1977

EPA Compilation of Air Pollution Emission Factors and  subsequent revisions

(USEPA, 1977).  Emissions from light duty diesel vehicles  during a calendar

year (n) and for a pollution (p) are approximated from the equation:


                               n

                     enp  '  iu^u cipn °in


where:
       e     * Composite  emission factor  in grams per vehicle mile for
        p         calendar year (n)  and  pollutant (p).

       c.    a The  FTP emission  rate  for  pollutant (p) in grams/mile for
        p         the i-th model year at calendar year (n).

       m.    a The  fraction of total  light-duty diesel vehicle miles
                  driven by  the i-th model year vehicles.
                                     A-18

-------
Emissions from heavy duty diesel vehicles during a calendar year (n) and for a
pollutant (p)  are approximated from the equation:
                                n
                     enps  =     -   °ipn
where :
      e    » Composite emission factor in grams per vehicle  mile for
        p        calendar year (n), pollutant (p), and average speed (s).
      c.   = The emission rate (g/mi) for pollutant (p)  for  the i-th
       •ipn
                 model year vehicles in calendar year  (n)  over a trans-
                 ient urban driving schedule vith  an average speed of
                 approximately  18  mi/hr (29 km/hr).
       v.    = The  speed  correction factor  for  i-th  model year heavy
        lp         duty diesel  vehicles  for pollutant (p) at average
                  speed  (s).
 Values for c.    are derived  from testing  of the various types of vehicles vith
 FTP procedures or on-the-road sampling.   Travel weighting factors take into
 account both January 1  registration and  fleet annual mileage accumulation dis-
 tributions for the given vehicle type.  The program computes and applies
 correction factors for speed, ambient temperature, and vehicle operating mode
 for scenarios that differ from  the basic standardized  conditions.
      MOBILE3 differs from MOBILE2 (USEPA,  1984) in tvo basic vays.   Data on
 the basic  emission rates for  certain  vehicle  types and model years  has been
 updated and revised.   The MOBILE2 program  had a built-in adjustment for tam-
 pering by  misfueling.   In MOBILES  the calculation of  the basic emission rates
 is for untampered vehicles and the  effects of tampering (e.g., misfueling,
 catalyst  removal, etc.) are  included  as  offsets which can be estimated from
  the percentage of vehicles  being tampered at a given time and the effects of
  such tampering.   MOBILES also has the capability  to take into account the
  effects of anti-tampering programs.
                                       A-19

-------
                                  References
U.S. Environmental Protection Agency  (1977)  Compiliation of air pollutant
     emissions factors.  Third edition (including supplements 1-7).  Research
     Triangle Park, NC: Offfice of Air Quality Planning and Standards.

U.S. Environmental Protection Agency  (1981)  User's guide to MOBILE2
     (Mobile Source Emission Model).  Ann Arbor, MI: Office of Mobile
     Source Air Pollution Control.  EPA  rept. no. 450/3-81-006.

U.S. Environmental Protection Agency  (1984)  User's guide  to MOBILES
     (Mobile Source Emission Model).  Ann Arbor, MI: Office of Mobile
     Source Air Pollution Control.  EPA rept. no. 460/3-84-002.
                                       A-20

-------
APPENDIX B

-------
TABLE B-l. SUMMARY OF DIESEL EXHAUST, WHOLE
       ANIMAL CARCINOGEN1CITY STUDIES
Model system
Study Species Strain Sex Number
Mauderly Kai F 344/Crl M and F 365
it al.,
IVH7 366


367


364


1 Icinrich 1 lamsler Syrian M and F 96
1 1 al., Golden
90


96
96

%


96



Treatment
None

None


None


None


None

None


N< >ne
45 Dig UEN/kg,
s.c. acute*
45 Dig DEN/kg,
s c. aculcb

45 Dig DEN/kf.
s.c. acute*

Exposure protocol
Concentration
mg/m1 Duration
("lean air Sham -exposed

0.35 7 h/d.
5 dAveek to
30 mo
35 7 h/d.
5 dAveek to
30 mo
70 7 h/d,
5 dAveek to
30 in<>
Clean air Sham-exposed

40 19 h/d.
5 dAveck,
lifetime
1 illcu-d* Sham-exposed
Clean air Sham-exposed
lifetime
4.0 19 h/d,
5 dAveck,
lifetime
Filtered* 19 h/d.
5 dAveck.
lifetime

Lung toot
burden
(mg/lung)
0

0.6
(24 mo)

11.5
(24 mo)

20.8
(24 mo)

NA

NA


NA
NA

NA


NA


Results
Histology
0.9% Pulmonary
tumor formation
1 .3% pulmonary
tumor incidence

3.6% Pulmonary
tumor incidence

12.0% Pulmonary
tumor incidence

No tumor formation

No tumor formation


No tumor formation
10% Respiratory
tract tumors
Not lignificantly
different from
control
Not significantly
different from
control

-------
                                            TABLE B-l. (continued)
T
Model System
.Study Species Strain Sex Number Treatment
llcinrich Hamster Syrian M and F 96 0 25 mg B|a|P, Ilr.
i tat., golden 20 weeks'1
96 0.2S mg U|a|P. Ilr.,
20 weeks'1

96 0.2S mg B|»|P, ilr,
20 weeks'1

96 None

llcinrich Mouse NMUI F 96 None
< l al.,

96 None


64 01 mgB(a|P,
iir.b/week,
1 0 weeks
64 I) 1 mg B(a|P,
hrb/week,
1 0 weeks
64 II 1 mg B(a|P,
iir.b/week.
10 weeks
Exposure preload
Concentration
mg/m1 Duration
Clean air Sham-exposed
lifetime
Filtered* IV h/d.
Sd/wcek.
lifetime
40 IV h/d.
J d/weck.
lifetime
Clean air Sham-expuaed
lifetime
Filtered* IV h/d.
5 d/weck,
lilclime
40 IV h/d.
5 d/week.
lifetime
Clean air Sham -exposed
lifetime

Filtered* 19 h/d.
5 d/weck.
lilclime
4.0 IV h/d.
5 d/week.
lifetime
Results
Lung soot
burden
(mg/lung) 1 listology
NA 2% Respiratory tract
tumors
NA Not significantly
different from
control
NA Not significantly
different from
control
NA 2.4% Pulmonary
adenocarcinoma
NA 19% Pulmonary
adenocarcinoma
(I8/93)1
NA 17% Pulmonary
adenocarcinoma
(IJ/76)1
NA Specific data not
presented

NA Specific data not
presented

NA Specific data not
presented


-------
TABLE B-l.  (continued)
Model System Exposure protocol
Conccnlnilion
Study Specie* Strain Sex Number Treatment rag/hi1
i kinrich Mouse NMRI F 64 O.OS mg B(a|P, ilr.%eek, Clean air
1 al., 20 weeks
64 DOS mg B[a|P, ilr.'/week. Filtered*
20 weeks

64 0 05 mg B|a)P, ilr.»/Week, 40
20 weeks

03
1
Ul 64 0 05 mg DllahA Clean Air
itr^/week, 10 week*
64 iH>5mgDBahA 1 illcicd'
iir.%eek, 10 week*

64 ii 05 mg DBahA 40
iir.Vwcek. 10 weeks

96 0 005 DBahA/kg, Clean air
s c.d, acute
96 0 005 DBahA/kg, Filtered*
s c.*, acute


Duration
Sham-exposed
lifetime
19 h/d,
5 d/week.
liletime
19 h/d,
5 d/week.
lilelime


Slmni expoced
lifetime
19 h/d,
5 d/week,
lilelime
19 h/d,
5 d/week,
lifetime
Sham-exposed
6 mo
19 h/d,
5 d/week,
6 mo
Results
Lung soot
burden
(mg/lung) Histology
NA 71% Lung tumor
formation
NA 41% Lung tumor
formation

NA Not significantly
different from
control


NA No significant effects
•
NA No significant effects


NA No significant effects


NA 46% tumor formation

NA No change relative to
control


-------
TABLE B-l. (conlinueiJ)
Model System
Study Specie* Strain Sex Number
ilcinrich Mouse NMKI F 96
I al.,
I986a
96
(neonaics)
96
(neonales)
96
(neonales)
96
96
luinrich Ral SPI- Wislar F
c: al..
96
48
48
Exposure protocol
Concentration
Treatment mg/m1
0.01 mg DBahAAg i.e.*, 40
acute
0 Ol/mg DBahA, i.e.*" Clean air
acute
0 Ol/mg DBahA, i.e.* Fillercr*
acute
0 Ol/mg DBahA, I.e.*" 40
acute
None Clean air
None Filtered*
None Clean air
250 mg DPN/kg, Clean air
s c.Aveekd 25 week*
250 mg DPN/kg, Fillered*
s c /week4 25 week*

Duration
19 h/d.
S dAveck,
6 mo
Sham -exposed
lifetime
19 h/d,
5 dAvcek.
6 mo
IV h/d,
5 dAveck,
6 mo
Mum exposed
lifetime
IV h/d,
5 dAveck,
lilelime
IV h/d,
5 dAveck,
lilelime
Shu in -exposed
lifetime
IV h/d.
5 d/wcek.
lilelime
Results
Lung soot
burden
(nag/lung) Histology
NA No change relative to
control
NA 81% tumor incidence
NA Not provided
NA 63% Tumor
incidence
NA No tumor formation
NA No Tumor formation
NA 9.5% Squamous cell
lung tumor (9/95)'
NA 2.2% Squam|K>us cell
carcinoma; 2.2%
malignant tumors
NA 4.4% Squamous cell
carcinoma

-------
TABLE B-l.  (continued)
Model System Exposure protocol

Concentration
Sludy Specie* Strain Sex Number Treatment mg/m1 Duration
Ikinrich Rat SPF Wislar F 48 250 mg DPN/kg. sc./Wk4 40 19 h/d.
tl al., 25 weeks 5 d/week,
1986 lifetime
48 500 mg DPN/kg. s.c./wkd Clean air I1' h/d,
25 weeks 5 d/wcek,
lilclime
48 500 mg DPN/kg. t.cjvk* 1 illcicJ" 19 h/J.
25 weeks 5 d/wcek.
lilclime
w
vJi 48 500 mg DPN/kg. i.cVwk* 40 19 h/d,
25 weeks 5 d/week,
lilclime
i imshi Hal F344/JCL M and F 64 M; 59 1 None 0 1 Icavy-duty engine
1 1 al.. sham-exposed
1 'M 30 mo
l>4 M; 5V 1 None 046 1 Mi/day
6dAveek
.10 mo
d4 M; 59 1 None 096 1 Mi/day
6dAvcek
30 mo
«>4 M; .VJ 1 None 1.84 K.h/day
6d/wcek
30 mo
64 M; 59 1 None 3 72 I6h/day
6dA*cek
30 mo
Results
Lung soot
burden
(mg/lung) Histology
NA 46.8% Squamous cell
carcinoma

NA 16.7% Squamous cell
carcinoma; 8.3%
malignant lumors
NA 14.6% Squamous cell
carcinoma; IZ5%
malignant lumors
31.3% Squaraous
NA carcinoma; 16.7%
malignant lumors

NA 08% Incidence of
lung tumors

NA 0.8% Incidence of
lung lumors

NA No lung lumors
observed

NA 3.3% Incidence of
lung lumors

NA 6.5% Incidence of
lung tumors


-------
'FABLE B-l.  (continued)
Model System
Study Species Strain Sex Number Treatment
iHhinishi Hal F-344/JCI. M and F 64 M; 59 1 None
rl al.,
1986
64 M; 59 1 None


64 M; 59 1 None


64 M; 59 1 None


64 M; 59 1 None

1 .ai cl al . Hal F-344/SI'F 1- 24 None
I'»H6
24 None


24 None


Exposure protocol
Conoenlraliiin
mg/m1 Duration
0 Light duty engine
sham-exposed
30 mo
0.11 16 h/d
6dArcek
34) mo
041 Id h/d
6dAveek
341 mo
1.08 16 h/d
6d/wcck
I 34) mo
2.32 Id h/d
6dAveck
Clean air Sham-exposed
24 mo
49 8 h/d
7d/Week
24 mo
Filtered* 8 h/d
TdAvcek
24 mo
Results
Lung soot
burden
(mg/lung) Histology
NA 3.3% Incidence of
lung tumors

NA 2.4% Incidence of
lung tumors

NA 0.8% Incidence of
lung tumors

NA 4.1% Incidence of
lung tumors

NA 2.4% Incidence of
lung tumors
NA 0% Tumor incidence

NA 26.3% Malignant
lung tumors (42.1%
total tumors')
NA 0% Tumor incidence



-------
TABLE B-l.  (continued)
Model System
Study Specie* Strain Sex Number
l.ikcmolo Rat F-344/JCL F 110
.1 al..
1986
Panioning
among groups
not pnivided
Fan ion ing
among groups
not provided

Panioning
among groups
not provided

Mouse O57HI/6N M and F 225

225



Mouse
ICR M and F 205



205



Exposure protocol
Concentration
Treatment mg/mj
None Clean air


None 24


1 g DIPN/kg. 2 4
i.pVweck,
3 weeks'

1 g DIPN/kg. ' Clean air
i p. /Week.
3 weeks'

None Clean air

None 24




None Clean air



None 24




Duration
Sham-exposed
24 mo

4 h/d
4dAveek
24 mo
4 h/d
4dAKeek
24 mo

Sham -ex posed
24 mo


Sham-expticed
24 mo
4 h/d
4 dAveek
24 mo


Sham-exposed
24 mo


4 h/d
4dAvcek
24 mo

Results
Lung soot
burden
(mg/lung) Histology
NA No tumors delected


NA No tumors detected


NA 39% (7/18)'
Incidence of
carcinoma (18 to 24
mo)
NA 19% (4/21)'
Incidence of lung
carcinoma (18-24
mo)
NA No adenocarcinomas
observed
NA Lung adeno-
carcinoma (2/33M;
11/38 Fat I9lo
28 mo)c

NA Lung adcno-
carcinoma (2/33M;
11/38 Fat I9lo
28 mo)c
NA Lung adeno-
carcinoma (2/33M;
11/38 Fat I9lo
28 mo)c

-------
                                                TABLE B-l. (conlinued)
DO
Model System
>iudy Species Strain Sex Number
lohrel Hat Wisiar F 96
.1, 1986
96

96
l.iighlwell Kal 1344 M and F 144
(l al.,
l"86
144
144
144
llamsler Syrian M and F 104
golden
104
KM


Exposure protocol
Conccnimiion
Treatment mg/rn1 Duration
None Clean ,iir Sham exposed
None Filtered' IH h/d
SdAveck
24 lo 30 mo
None 39 18 h/d
5 d/week,
24 to 30 mo
None Clean air Shnin-ex|Nncd
24 mo
None lf> h/d
07 5 dAwxk,
24 mo
None l<> h/d
22 5 il/Week,
24 mo
None lo h/d
66 5 uTweck,
24 mo
None Clean air Sham-exposed
24 mo
4.5 mg DEN/kg, Clean air Sham-exposed
s.c., acute' 24 mo
None 0.7 16 h/d
5 d/Week,
24 mo

Lung soot
burden
(mg/Iung)
NA
NA

NA
NA
NA
NA
NA
NA
NA
NA


Results
Histology
No neoplaslic tumors
Data nol presented

15.8% Neoplaslic
luraon (I5/96)C with
1 carcinoma
I.ung tumor
Incidence al
2%M; I%F
I.ung tumor
incidence at
1% M; 0% F
I.ung tumor
incidence at
4% M; 15 % F
I.ung tumor
incidence ol
23% M, 54% F
No tumors delected
Increase in trachea!
papillomas
No tumors delected



-------
TABLE B-l.  (continued)
Model System Exposure protocol
Concentration
Mudy Species Strain Sex Number Treatment mg/hj3 Duration
Mrighlwell Hamster Syrian M and F 104 4.5 rag DEN/kg, 07 16 h/d
< I «l , Golden s.c., acute* 5 dAveek,
, ''86 24 mo

104 None 2.2 lo h/d
5 dAveek.
24 mo
104 4.5 mg DEN/kg. 22 16 h/d
* c., acute* 5 dAveek,
U 24 mo
1
vO
104 None 66 lo h/d
5 h/d
5 dAveck
lo 30 mo
Results
1 .ung soot
burden
(tug/lung) Histology
NA No tumors attributed
to diesel exhaust
exposure
No tumors attributed
NA lo diesel exhaust
exposure

NA No tumors attributed
lo diesel exhaust
exposure

No tumors attributed
NA lo diesel exhaust
exposure

NA No tumors attributed
lo diesel exhaust
exposure
NA 2.4% Carcinoma
incidence
NA 1.6% Carcinoma
incidence

NA 0% Carcinoma
incidence


-------
                                                                            TABLE B-l.  (continued)
—i
o
Model System Eipcxure protocol
Concentration
Study Species Strain Sex Number Treatment Dig/to1 Duration
lakakiel Rat F344 M inJ F 123 None I.I1 16 h/d
il., 1989 6 dAvetk
to .to mo
124 None 2 3r 16 h/J
6d/weck
to 30 mo
123 None 0 Sham-expoced

123 None 05* 16 h/J
6dA»eek
to 30 mo
125 None 1.0* 16 h/J
6 d/weck
to 30 mo
123 None 1.8* Id h/J
6 d/week
to 30 mo
124 None 3.7* 16 h/d
6d/Weck
to 30 mo

Lung tool
burden
(rag/lung)
NA


NA


NA

NA


NA


NA


NA


Results
Histology
4.1% Carcinoma
incidence

1.6% Carcinoma
incidence

0.8% Carcinoma
incidence
0.8% Carcinoma
incidence

0% Carcinoma
incidence

3.3% Carcinoma
incidence

6.5% Carcinoma
incidence

            Tillered exhaust, no p.uliclcs.
            N "oncurrenl treatment.
            'Number of animals with tumors/numbers of animals examined.
            Jrrcireatmenl.
            'Malignant splenic lymphomas also reported. Abbreviations: s.c.. suhcuuneous; itr, inlracheal; DEN, dimethylnitimamine; DliahA, dibeiu|ah|anlhracene; DPN, dipcntylnilrosaniine;
            |)|I'N, di isopropanol-niln»amiiie;                     n|.i|P, l>eiii<>|:i)pyrenc; NA, not available.
            'i ighl-July engine emissions.
            ' Icavy-duly engine emission.

-------
APPENDIX C

-------
                  DRAFT — DO NOT QUOTE OR CITE

         APPENDIX C.   MODELS  FOR CALCULATING  LUNG BURDENS

C.I.  INTRODUCTION
  The lung burden of  diesel  exhaust  particles  (DEPs)  during
exposure is determined by both the amount and site of particle
deposition in the lung and,  subsequently, by rates of
translocation and clearance  from th<=> r^pn^ition  sites.
Mathematical models have often been used to complement
experimental studies  in estimating the lung burdens of  inhaled
particles in different species under different exposure
conditions.  This section presents a mathematical model that
simulates the deposition and clearance of DEPs in the lungs of
rats and humans.
   Diesel particles  are  aggregates  formed from primary spheres of
15-30 mm in diameter.  The aggregates are irregularly shaped  and
range in size from  a  few molecular diameters to  tens  of microns.
The mass median  aerodynamic  diameter  (MMAD)  of the aggregates is
approximately 0.2 ^m.  The primary sphere consists of a
carbonaceous core (soot) on  which numerous kinds of organic
compounds are adsorbed.  The organics normally account  for 10%  to
30% of  the particle mass.  However,  the  exact  size distribution
of  DEPs and the  specific composition of  the  adsorbed  organics
depend  upon many factors,  including  engine design, fuels used,
engine  operating conditions, and the thermodynamic process of
exhaust.  The physical  and chemical  characteristics of  DEPs have
been reviewed extensively  by Amann and Siegla (1982)  and
Schuetzle  (1983).
   Four mechamisms deposit diesel particles within the
respiratory tract during  exposure:   impaction,  sedimentation,
interception, and diffusion.  The  contribution from  each
mechanism to deposition,  however,  depends upon lung  structure and !
size,  the breathing condition  of the subject,  and particle size

                                C-l                        03/27/90

-------
                  DRAFT ~ DO NOT QUOTE OR CITE
distribution.  Under normal breathing conditions, diffusion is
found to be the most dominant mechanism.  The other three
mechanisms play only a minor role.
  Once DEPs  are deposited  in the  respiratory tract, both the
carbonaceous cores and the adsorbed organics of the particles can
be removed from the deposition sites by two mechanisms:
(a)  mechanical clearance, provided by mucociliary transport in
the ciliated conducting airways as well as macrophage
phagocytosis and migration in the nonciliated airways, and (b)
clearance by dissolution.  Since the carbonaceous soot of DEPs is
insoluble, it is removed from the lung by mechanical clearance,
whereas the adsorbed organics are removed prinicpally by
dissolution.
C.2.  COMPARTMENTAL LUNG MODEL
  To  simulate  the transport and removal  of DEPs from the lungs
mathematically, we use a compartmental model consisting  of six
anatomical compartments:  the nasopharyngeal or head  (H),
tracheobronchial  (T), alveolar  (A), gastrointestinal  tract (G),
lymph node (L), and blood  (B)
compartments as shown in Figure C-l.  In this  figure, r^, r^1,  and
are, respectively, the mass  deposition rates of DEP material
components (i=l soot), 2  (slowly cleared organics),  and  3 (fast
cleared organics)  in the head,  tracheobronchial,  and  alveolar
compartments;  and A^'  represents the transport  rate of material
from compartment  X to compartment  Y.
  Let the mass fraction  of material  component i of a diesel
particle be fi.
                                C-2                        03/27/90

-------
                  DRAFT — DO NOT QUOTE  OR CITE

Then
                                                             (02)
                                                             (C-3)
where  r^' , ZT^ • and rx(a>   are,  respectively,  the total mass
deposition rates of DEPs  in the H,  T,  and A compartments,
determined from the equations:

                        ZH - C(TV) (RF) (DF) H                  (C-4)
                         T" C(TV) (RF) (DF) T                  (C-5)
                        TA - C(TV) (RF) (DF) A                  (C-6)


   In equations C-4 to C-6, C is the mass  concentration of DEPs
 in the air,  TV is the tidal volume, RF is the respiratory
 frequency,  and (DF)H,  (DF)T, and (DF)A are, respectively,  the
 deposition  fractions of DEPs in the H, T, and A compartments  over
 a  breathing cycle.  The values of  (DF)H,  (DF)T, and  (DF)A  vary
 with the particle size distribution of DEPs, lung geometry, and
 breathing conditions, as will be discussed later.
   On the basis of the clearance characteristics,  we assume that
 a  diesel particle is composed of three material components:   (a)
 carbonaceous soot, (b) adsorbed organics which are  slowly

                                C-3                        03/27/90

-------
                             H






B





(i)
^HB

(!)


. (i)
AAB

(i)
XLB

0 (
H AH(

f
0- . (
T AT(
1
"*2
f
A
i
r
I
i)
%

6
)
%

0
t
\
AAL

•»
Figure C-l.  Coropartmental Lung Model of Particle Deposition and
Clearance

cleared, and (c) adsorbed organics that are quickly cleared.  The
presence of two separate organic components in the particle model
is suggested by observations that the removal of the particle-
associated organics from the lung exhibited two distinct half-
times  (Sun et al., 1984; Bond et al., 1986a).  In a typical
diesel particle, the mass fraction of the total adsorbed organics
is about 20%, with half of this mass composed of slowly cleared
organics and the other half composed of quickly cleared organics.
  The differential equations  for m^,  the mass  of  material
component  i in  compartment X, as a function of exposure time  t
can be written  as
                                C-4
03/27/90

-------
                   DRAFT —  DO NOT QUOTE OR CITE



Head  (H)




                    dmm
                     •£- • **" -  **W - *BX21              (C-7)
Tracheobronchial  (T)


                *mu)

                -Je-  • *P + «iX"' - «v - xtf^         («>
Alveolar  (A)
                ^r • r "' - ^X" - ^'^^  - ^'^(i)          (C-9)
GI Tract (G)
                       ^  • XS'jn"' + X«X»               (C-10)
Lymph Nodes (L)
                       dt
                                 ,(a)   jfij-ti)                cr m
                                 !A  - Aj-^ m,                  (^11)
                               C-5                        03/27/90

-------
                  DRAFT — DO NOT QUOTE OR CITE

Blood (B)
  Equation C-9 nay also be written as
where
is the total clearance rate of material component i from the
alveolar compartment.
  The total mass  of  the  particle-associated organics  in
compartment X is the sum of m™  and roj3' , and the total  mass of
DEPs in compartment X is equal to
The lung burdens of carbonaceous soot and organics are defined,
respectively, as

-------
                  DRAFT ~ DO NOT QUOTE OR CITE

                   .,(2)*(3)   _(2) __ _(3)  __ _{2)   _(3)
                  "Lung   " mT  + %•   * MA  + mA
Because clearance of the particle-associated  organics  is  much
faster than diesel soot,  M^y(2)  constitutes only  a very small
fraction of total particle mass accumulated in the  lung a short
time after exposure.
  In the compartmental model described above, both compartments
B and G do not contain any excretion pathways.  The results of
 mg*} and m^     therefore, represent the  accumulative  amount of
particle mass transported into that compartment.

C.3.  SOLUTIONS TO KINETIC EQUATIONS
  The  solutions  to Equations C-7 to C-13  can be found when
 Tui} , *TI} • and *xr    are known.   For exposures to DEPs  of  constant
 concentrations under specified breathing  conditions, r^', XT*} • and
 are constant.  The values of X^y   are also  constant,  except
 those of    XJr and A^f'   which  are  found from the experimental
 data of rats to  be a function of  mA or m™    (Yu et al., 1988).
 This is attributed to the overloading  of  DEPs in the lung.
  To find  solutions for Equations C-7 to C-13 ,  m^ is first
 determined  from  Equation C-13,  which is solved numerically by a
 second  order Runge-Kutta method.   Once mjf}  is found,  the other
 kinetic equations for both  diesel soot and the particle-
 associated  organics  are obtained readily since they are linear
 differential equations. The solutions to these equations for
 constant r#i} , r^ , and i^ are as follows:
                                 C-7                        03/27/90

-------
Head  (H)
                    DRAFT  — DO NOT  QUOTE OR  CITE
                                      « \ «—, /  1 li) *\
                                    - rw /Afl )exp(- AW c)
 where A
Tracheobronchial (T)



          » <^>   ««-^ /  5 '•*' t-\ f e
          raT   - exp(-AT  t) /
                           Jo
                                                                  (C-19)
                 (to \ _„_ / 1 (i)
                                                             (ji)
                                                    *\ j*. ^  „ j    /r« in\
                                                    t) at +  mTO    (C-20)
where
GI  Tract  (G)
                         fc
                      •  I
                        Jo
Lymph Nodes  (L)
       f '
       Jo
                                                                  (C-21)
                                                                  (C-23)
          C-8
                                                                03/27/90

-------
                  DRAFT — DO NOT QUOTE OR CITE

Blood (B)

                              i)_U>  ^ i U)«.U)
                          F c
                        - J
                                                            (C-24)
                                          "BO
  In Equations C-18 to  C-24,  ^o*   represents the value of m*1*
at t = o.
  In the sections to  follow,  the methods of determining r^',  TT*} • and
or (DF)H/  (DF)T, and  (DF)A as well as the values of  \(£  in  the
compartmental lung model are  presented.

C.4.  DETERMIKATION OF  DEPOSITION FRACTIONS
  The mathematical models  for determining the deposition
fractions of DEPs in  various  regions of  the respiratory tract
have been developed by  Yu  and Xu (1986,  1987)  and are adopted in
this report.  Yu and  Xu consider DEPs  as a polydisperse aerosol
with a specified mass median  aerodynamic diameter (MMAD) and
geometrical standard  deviation  ag.  Each diesel particle is
represented by a cluster-shaped aggregate within a spherical
envelope of diameter  de.  The envelope diameter de is related to
the aerodynamic diameter of the particle by the relation
                       d         C  1/2   r  1/2
                      ±£  . 4ri/2(±£,  (J.)                  (C-25)
                       "a         C•     ^o
where  C  is the  bulk density of the particle in g/cm3,  C0 =  1
g/cm3; 0 is the packing density, which  is  the ratio of the space
actually occupied by primary particles in the envelope to  the
overall  envelope  volume;  and Cx is the  slip factor given by the
expression:
                                C-9                        03/27/90

-------
                  DRAFT — DO NOT QUOTE OR CITE
                       r 11-257 + 0.4 exp(-  0>5,5d*)3        (C-26)
                       dx                     A.
in which A « 8 x 10"6cm3 is the mean free path of air molecules at
standard conditions.  In the diesel particle model of Yu and Xu
(1986), C has a value of 1.5 g/cm3 and a 0 value of 0.3 is chosen
based upon the best experimental estimates.  As a result,
Equation C-25 gives de/da =1.35.   In  determining the deposition
fraction of DEPs, da is used for diffusion and interception
according to the particle model.

C.4.1.  Determination of (DF)H
   Particle deposition  in the  naso- or oro-pharyngeal  region is
referred to as head or extrathoracic  deposition.  The amount of
particles that enters the lung depends upon the breathing mode.
Normally, more particles are collected via the nasal route  than
the oral route because of the nasal hairs and the more  complex
air passages of the nose.   Since the  residence time of  diesel
particles in the head region during inhalation is very  small
(about 0.1 second for human adults at normal breathing),
diffusional deposition is insignificant  and the major deposition
mechanism is impaction.  The  following empirical formulas derived
by Yu et al.  (1981) for human adults  are adopted for deposition
prediction of DEPs:
                                C-10                       03/27/90

-------
                  DRAFT ~ DO  NOT QUOTE OR CITE
For mouth breathing:

                  (DF)Hi ia - 0,        for dlz 3000
                                                            (C-27)
          (DF)Hi in - -1.117 * 0.324 log(djp) ,  for d*0 > 3000
                                                            (C-29)
and for nose breathing:
                  ~ -0.014 + 0.023
                                           , for d2aQ s 337    (C-30)
(DF)
              H>
                  - -0.959 * 0.397 log(d*C>) / for d^C > 337
                                                             (C-31)
           (DF)Ht in - 0.003 * 0.033 log(dJ0) , for dlo $ 215
                                                            (C-32)
              H. .x • -0.851 f 0.399 log(djp) , for d\Q > 215     (C-33)
where  (DF)H is the mean deposition efficiency in the head, the
subscripts  in  and ex denote inspiration and expiration,
respectively,  da  is  the particle aerodynamic diameter in fim, and
Q is the air flowrate in cm3/sec.
  Formulas to  calculate deposition of diesel  particles in the
head region of children are derived from those  for  adults using
the theory  of  similarity,  which assumes that  the  air passage in
the head region is geometrically similar for  all  ages and that
the deposition process is characterized by the  Stokes number of
                                C-ll
                                                           03/27/90

-------
                   DRAFT ~ DO NOT QUOTE OR CITE
the particle.   Thus,  the set of empirical  equations from C-27
through  C-33  are transformed into the  following form.

For mouth breathing:

                  (DF)H> ifl - 0,        for d\Q <> 3000           (C-34)
                   (D&H in - -1-117 * 0.972 logK +
                       '2         a                    CC-35)
                   0.324 log d,0) , for d\Q >  3000            *•   '
                                    . 0.                      (C-36)
and for nose  breathing:

              H in " ~ 0-014 * 0.690 logk + 0.023
                                                        ,
                            for dD & 337                      (C"37)
                      H. in " -0.959 * 1.191 logK +
                    0.397 log(d*0) ,  for d\Q > 337             ^C"3 '
                       H. ex " 0-003 * 0.099  logK +
                    0.033 log(dfff) , for C&? S215              (C"39)
                      Hi €X - -0.851 * 1.197 logK*
                    0.399 log(d*0) »  for d\Q > 215              l   J
                                C-12                         03/27/90

-------
                  DRAFT — DO NOT QUOTE OR CITE
where K is the ratio of the linear dimension of the air passages
in the head region of adults to that of children, which is
assumed to be the same as the ratio of adult/child tracheal
diameters.
  For rats, the  following empirical equations  are used for
deposition prediction of DEPs in the nose:
                                 Mt 0X - 0 . 046 t-
                  0.009 log(da£>) ,  for d\Q * 13.33
                                 _ ex - -0.522 +
                  0.514 log(dlQ) ,  fordlQy 13.33
C.4.2.  Determination  of  (DF)T and (DF)A
   The deposition model adopted for DEPs is the one previously
developed  for monodisperse  and (Yu,  1978)  and polydisperse
spherical  aerosols  (Diu and Yu,  1983).   In the model,  the
branching  airways are  viewed as  a  chamber  model shaped like a
trumpet  (Figure C-2).   The  cross-sectional area of the chamber
varies with  airway  depth, x, measured from the beginning of the
trachea.   At the last  portion of the trumpet, additional cross-
sectional  area  is present to account for the alveolar volume per
unit length  of  the  airways.
   Inhaled diesel particles that escape capture in the head
during inspiration  will enter the  trachea  and subsequently the
bronchial  airways (compartment T)  and alveolar spaces
 (compartment A).
                                C-13                       03/27/90

-------
                    DRAFT — DO NOT QUOTE OR CITE'
                      Summtd tivtotor Cross Stcuend AIM
                                              Airway Lngtfts
Figure C-2.   Trumpet model of lung airway,
                                   C-14
03/27/90

-------
                  DRAFT — DO NOT QUOTE OR CITE •
  Assuming that the airways expand and contract uniformly during
breathing, the equation for the conservation of particles takes
the form:

                                                           (013)
where c is the mean particle concentration at a given x and time
t; A, and A2 are,  respectively,  the  summed cross-sectional  area
(or volume per unit length) of  the  airways and alveoli at  rest; 77
is the particle uptake efficiency per unit length  of the airway;
(3 is an expansion factor,  given by:

                            P ' 1 *                         (C-44)
and Q is the air  flow rate,  varying with  x  and t according to the
relation
 where Q0 is the air flow rate at x = 0.  In Equations C-44 and C-
 45,  Vt is the volume of new air in the lungs and Vx and Vt are,
 respectively,  the accumulated airway volume from x = 0 to x,  and
 total airway  volume at rest.
   Equation C-43 can be solved using the method of
 characteristics with appropriate initial and boundary conditions;
 the  amount of particles deposited between location x1  and x2 from
 time t, to t2 can then be found from the expression

 For  diesel particles,  rj is the sum of those due to the individual
 deposition mechanisms described above,  i.e.,

                               C-15                        03/27/90

-------
                   DRAFT — DO  NOT QUOTE OR CITE


                          DF - I foar\dxdt                    (C-46)


                        n -tix + i)s + iip + iia                  (CAT)


where  17,, r?s, r?p,  and D are,  respectively,  the deposition
efficiencies per unit  length of the airway due to impaction,
sedimentation,  interception,  and diffusion.   On  the basis of the
particle model  described above,  we obtained  the  expressions for
H!/  ns, f?P/  and  r?D  in the following  form:
            r>  tm  ~~ n*».n _ c (2/3) _ 01/3 ./i  _ £>2/3 ±. ein"l
            T] -      [                   (C-50)
and
                                 C-16                         03/27/90

-------
                  DRAFT ~ DO NOT QUOTE OR CITE
                  Tip- -i[l-0.819exp(-14.63A) -
                       0.0976 exp(-89.22A) -
             0.0325 exp(-228A) - 0.0509  exp(-125A2/3) ]
for Reynolds numbers of the flow  smaller than 2000,  and

                     T) • — A1'2  (1 - 0.444A1/2)               (C-52)


for Reynolds numbers greater  than or equal to 2000.   SC-dju/ (18n/?)
is the particle Stokes number,
                 6 - L/(BR)e -  3\iUgL/(32uR)T - djR
and  A - Z?L/(4J?2u)  in which u is the air velocity in the airway; M
is the air viscosity;  L and R are,  respectively, the length and
radius of the  airway;  us - Cad«/(18n)  is tne particle settling
velocity; and  D  =  CekT(37TMde) is the diffusion coefficient with k
denoting the Boltzmann constant and T the absolute temperature.
In the deposition  model,  it is also assumed that 77, = rj = o for
expiration, while  r70 and ns have the same expressions  for both
inspiration and  expiration.
   During the  pause, only diffusion and sedimentation are
present.  The  combined deposition efficiency in an airway, E, is
equal  to:
                       E - 1 - (1 -Es) (1 - ED) .                (C-53)
where E0 and Es are, respectively, the deposition  efficiencies
due to  the individual mechanisms of diffusion and sedimentation
over the pause period.  The expressions for E0 and Es are given
by
                                C-17                        03/27/90

-------
                  DRAFT  —  DO NOT QUOTE OR CITE
           3                       3
           E -7- ex*>(- ***D> (1 ' E -7
          • ^"^                    •••*
                                                        2 - 1
                                                             (C-54)
where TO = Dr/R2  in which r is the pause time and a,, a2,  and a3
are the first three roots of the equation:
                             J0(a) -0.                       (C-55)
in which J0 is the Bessel  function  of  the zeroth order, and:

              Es - 1.1094T,. - 0.1604T2S,  for 0  < tff s 1.        (C-56)


and:

              Es - 1 - 0.0069T51 -0.0859t~s2 - 0.0582t~s3,        /T 5T»
                            for TS> 1,                      ^   '
where TS = usr/2R.
  The values of (DF)T  and (DF)A  over  a  breathing cycle are
calculated by  superimposing DF for inspiration, deposition
efficiency E during pause, and DF for expiration  in  the
tracheobronchial airways and alveolar space.  It  is  assumed that
the breathing  cycle consists of a constant flow inspiration,  a
pause, and a constant  flow expiration, each with  a respective
duration  fraction  of 0.435, 0.05, and 0.515 of a  breathing
period.
                                C-18                        03/27/90

-------
                  DRAFT — DO NOT QUOTE OR CITE
C.4.3.  Lung Models
  Lung architecture affects particle deposition in several ways:
the linear dimension of the airway is related to the distance the
particle travels before it contacts the airway surface; the air
flow velocity by which the particles are transported is
determined by the cross-section of the airway for a given
volumetric flowrate; and flow characteristics in the airways are
influenced by the airway diameter and branching patterns.  Thus,
theoretical prediction of particle deposition depends, to a large
extent, on the lung model chosen.

C.4.3.1.  Lung Model for Rats—Morphometric data on the lung
airways of rats were reported by Schum and Yeh  (1980).  Table  C-l
shows the lung model data for Long Evans rats with a total lung
capacity of 13.784cm3.   Application of this model to Fischer rats
is  accomplished by assuming that the rat has the same  lung
structure regardless of its strain and that the total  lung
capacity is proportional to the body weight.  In addition, it  is
also  assumed that the  lung volume at rest  is about 40% of the
total lung capacity and that any linear dimension of the lung  is
proportional to the cubic root of the lung volume.

C.4.3.2.  Lung Model for Human Adults--The lung model  of mature
human adults used in the deposition calculation of DEPs  is the
symmetric lung model developed by Weibel  (1963).  In Weibel's
model, the airways  are assumed to be  a dichotomous branching
system with 24 generations.   Beginning with the 18th generation,
increasing numbers  of  alveoli  are present  on the wall  of the
airways and the last three  generations are completely
aleveolated.  Thus, the alveolar region  in this model  consists of
all the airways in  the last seven generations.
                                C-19                       03/27/90

-------
                  DRAFT — DO NOT QUOTE OR  CITE
  Table A-2 presents the  roorphometric data  of the airways of
Weibel's model adjusted to a total lung
volume of 3,000 cm3.
  C.4.3.3.  Lung  Model for Children—The  lung model for children
in the diesel study was developed by  Yu and Xu (1987)  on the
basis of available  morphometric measurements.  The model assumes
a lung structure  with dichotomous branching of airways,  and it
matches Weibel's  model for a subject  when evaluated at the age of
25 years, the age at which the lung is considered to be mature.
The number and size of airways as functions of age t (years) are
determined by the following equations:
  C.4.3.3.1.  Number of  airways  and  alveoli.  The number of
airways N;(t)  at generation i for age t is given by

                  tf4(O -  2',        t or 0  z i *20             (C-58)
           N2i(C)  - 221,
          iN22(C)  - NT(t)  -2",         fox 22i < NT(t) i 222    (C-60)
           N3(t)  - 0,
                               C-20                        03/27/90

-------
                   DRAFT  —  DO NOT QUOTE  OR  CITE
            TABLE C-1  LUNG MODEL FOR RATS AT TOTAL LUNG CAPACITY
Generation
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Iff
17
18
19
20
21
22
23
24
Number of
Airways
1
2
3
5
8
14
23
38
65
109
184
309
521
877
1,477
2.487
4,974
9.948
19.896
39.792
79.584
159.168
318.336
636.672
Length
(cm)
2.680
0.715
0.400
0.176
0.208
0.117
0.114
0.130
0.099
0.091
0.096
0.073
0.075
0.060 ~
0.055
0.035
0.029
0.025
0.022
0.020
0.019
0.018
0.017
0.017
Diameter
(cm)
0.340
0.290
0.263
0.203
0.163
0.134
0.123
0 112
0.095
0.087
0.078
0.070
0.058
0.049
0.036
0.020
0.017
0.016
0.015
0.014
0.014
0.014
0.014
0.014
Accumulative
Volume1 (Crrr i
0.243
0.338
0.403
0431
0466
0489
0.520
0.569
0615
0674
0758
0845
0.948
1.047
1 141
1.185
1.254
1.375
1.595
2.003
2.607
4.389
7.554
13.784
* Terminal bronchioles.
0 Including the attached alveolar volume
    (number of alveoli = 3 x 107 alveolar diameter = 0.0086 cm).
                                     C-21
03/27/90

-------
                     DRAFT  —  DO NOT QUOTE  OR  CITE
    TABLE C-2.  LUNG MODEL BY WEIBEL (1963) ADJUSTED TO 3.000 CM5 LUNG VOLUME
Generation
Number
0
2
2
3
4
s
6
7
8
9
10
11
12
13
14
15
161
17
18
19
20
21
22
23
Number of
Airways
i
2
4
8
16
32
64
128
256
512
1.024
2.048
4.096
8.192
16.384
32.768
65.536
131.072
262.144
524.283
1.048.576
2.097.152
4.194.304
8.388.608
Length
(cm)
10.260
4.070
1.624
0.650
1.086
0.915
0.769
0.650
0.547
0.462
0.393
0.333
0.282
0.231
0.197
0.171
0.141
0.121
0.100
0.085
0.071
0.060
0.050
0.043
Diameter
(cm)
1.539
1.043
0.710
0479
0.385
0.299
0.239
0.197
0.159
0.132
0.111
0.093
0.081
0.070
0.063
0.056
0.051
0.046
0.043
0.040
0.038
0.037
0.035
0.035
Accumulative
Volume1 (cnrf j
1906
25.63
2863
29.50
31.69
33.75
35.94
38.38
41 13
4438
48.25
53.00
59.13
66.25
77.13
90.69
109.25
139.31
190.60
288.16
512.94
925.04
1.694.16
3.000.00
* Terminal bronchioles.
6 Including the attached alveolar volume
 (number of alveoli = 3 x if/. alveolar diameter = 0.0288 cm)
                                      C-22
03/27/90

-------
                  DRAFT — DO NOT QUOTE OR CITE

               tf2l(t) - 221,
               Wj2(t) - 222,      . for Nr(t) > 221 + 222,
               N23(t) - Nf(t) -221 -222
where Nr(t)  is the total number of airways in the last three
airway generations.  The empirical equation  for Nr which best
fits the available data is

  Thus,  Nr(t) increases from approximately 1.5  million at birth
to 15 million at  8 years of age  and  remains  nearly constant
thereafter.  Equations A-59 to A-61  also imply  that in the last

             w ,H  / 2.036 xl07(l-0.926e-°-15t) ,  t*8
             N*(C}  ( 1.468 X107,              t > 8

three generations, the  airways  in the subsequent generation begin
to appear only when those  in the preceding generation

have completed development.
  The  number of alveoli as a function of age can be represented
by the  following  equation  according to the  observed data:
                 N.(t) - 2.985 X 108(1 -0.919e-°'
-------
                   DRAFT ~ DO NOT QUOTE OR CITE
model;  (c) the  growth pattern of the bronchial airways;  and (d)
variation in  alveolar size with age.  From these  data,  it is
found that the  lung volume, LV(t) at age t,  normalized to
Weibel's model  at 4,800 cm3 for an adult (25 years old),  follows
the equation


            LV(t) - 0.959 x 105(1 - 0.998e-°-002t)  (cm3).            (C-64)


   The growth patterns of the  bronchial  airways are determined by
the following equations
                   ZUt)  - Div - ttAH(C) -#(25)],                  (C-65)
                   Lj(t) - Liv - pj[#(t)  - #(25)],                  (C-66)


where  D. (t) and  Lf(t) are,  respectively,  the airway diameter and
length at generation i and age  t,  Diw and Ljw the corresponding
values for Weibel's model, a,, and p{ are coefficients given by

                 a,. - 3,26 x 10'2exp [-1.183 (i + 1)0-5]                (C-67)

                      P; •= 1.05 x 10-6 exp [10.1 (
 and  H(t)  is the body height,  which varies with age t in the  form
               Hit) - 1.82 x 102(1 - p.725e-°-14t)  (cm).              (C-69)


   For the growth  patterns  of the airways  in the  alveolar region,
 it is assumed that
                                C-24                        03/27/90

-------
                  DRAFT — DO NOT QUOTE  OR CITE
 .±L - -p- - —i. - f (t) ,    for 17 s i * 23
  £iwr    Liw   D,,,,
                                                            (C-70)
where D. is the diameter of an alveolus at age  t,  D_ = 0.0288 cm
       0                                            AH
is the alveolar diameter  for  adults in accordance with Weibel's
model, and f(t) is  a  function determined from
f(t)  -
                                16
                      (LV(t)  -
                              2-0
                        23
                       J - 17
                             4
                                                                   (C-71)
C.5.  TRANSPORT RATES
   The values of transport rates  \r?    for  rats have been
determined  from the experimental data of clearance  for  diesel
soot  (Chan  et al.,  1981; Strom et al.,  1987, 1988)  and  for the
particle associated organics (Sun et al., 1984;  Bond et al.,
1986).   These values are used in the present model  of lung burden
calculation and are listed below:
                   j - 1.05 x lO-'expUO.lU+l)'0'2] ,
                                                (C-68)
                                O25
                                               03/27/90

-------
  DRAFT ~  DO NOT QUOTE OR CITE


      iff - 1.73  U - 1,2,3)                 (072)




     iff - iff - iff - 0.00018                (073)




    Aff - Iff -Xff -*ff  -0.013              (074)




    iff ' iff - iff - iff  - 12.57              (075)




     iff - 0.693     (2 -  1,2,3)               (076)




     - 0.00068  [1 - exp(-0.046/ni'")3          (O77)
            1 J U)     / •   , ^ V
           •"" A «B     V •*• m £ I J I
            4
    i.ir  - 0.012 exp(-0.11^'76) +

0.00068 exp{~0.046^'")  (i - 1,2,3)
                C-26                         03/27/90

-------
                  DRAFT — DO NOT QUOTE OR CITE

                          l£ - 0.00018                     (OSO)
                                   - 6
                  0.012 exp(-0.11;njf- 6)  + 0.00086
   * A   - A   - 0.012
0.00068 exp(-0.046;nalt6)  * 0 . 016 .
                                    lt62                    ^   '
            13)   1 (3)   } (3)    i (3)   n ni->
               ~ "   * ^-    *  *fl " 0-012
                  0.00068 exp(-0.046m]>62) + 15.71
where  lj£ is the unit of day'1.
  Experimental  data on the  deposition and clearance of DEPs in
humans are not  available.   To estimate the lung burden of DEPs
for human exposure,  it is necessary to extrapolate the transport
rates   A^'   from  rats  to  humans.   For organics,  we assume that
the transport rates are  the same for rats and humans.   This
assumption is based upon the observation of Schanker et al.
(1986) that the lung clearance of inhaled lipophilic compounds
appears to depend only on their lipid/water partition
coefficients and is independent of species.  In contrast, the
transport rates of  diesel soot in humans should be different from
that of rats, since the  alveolar clearance rate, TA,  of insoluble
particles at low lung  burdens for human adults is approximately
seven times that of rats (Bailey et al., 1982).
  No  data are  available on the change of the  alveolar clearance
rate of insoluble particles due to excessive lung burdens among
                               C-27                        03/27/90

-------
                  DRAFT — DO NOT QUOTE OR CITE
different species.  It is seen that       for rats can be written
in the following form

                       A.^ - a exp(-Jhm/) * d       •         (C-84)
where a, b, c, and d are constants.  The right-hand side of
Equation A-84 consists of two terms, representing, respectively,
macrophage-mediated mechanical clearance and clearance by
dissolution.  The first term depends upon the lung burden,
whereas the second term does not.  To extrapolate this
relationship to humans, we assumed that the dissolution clearance
term was independent of species and that the mechanical clearance
term for humans varied in the same proportion as in rats under
the same unit surface particulate dose.  This assumption results
in the following expression for   AJ1'  in humans

                     Af} - £ exp(-b(mA/S) + d              (C-85)
where P is the ratio  of  the  alveolar clearance  rate  by mechanical
clearance in humans to that  of rats in low  lung burdens,  and  S  is
the corresponding  ratio  of the pulmonary  surface area  between
humans and rats.   Equation A-85  implies that  rats and  humans  have
the same "local" biological  response in the lung to  inhaled DEPs,
an assumption that needs to  be confirmed  by future experiments.
   From the data  of Bailey et al. (1982) ,  we obtain a value of
^AO   ~ 0.00169 day"1   for humans.   This  leads to P = 14.4.
Also, we find S-148 from the data of the  anatomical  lung model  of
Yen and Schum  (1980)  for rats and Weibel's  model for human
adults.  For humans less than 26 years old, we  assume the same
value for P, but S is computed from the data  of the  lung model
for young humans  (Yu  and Xu  1987) .  The values  of S  as a function
of age are shown in Table C-3.

                               C-28                        03/27/90

-------
                   DRAFT — DO NOT  QUOTE OR CITE
     The equations for other transport rates that have a  lung-
burden-dependent component are extrapolated from rats to  humans
in a similar manner.  The following lists the values of  XJ£
(in day"1)  for humans used in the present model calculation:
                       JL$ - 1,73 (i - 1,2,3)                  (C-86)

                                     - 0.00018                (087)
                      (2) _ i (2)   i (2)    i (2)   n
                     -     A-     *-      **     U .
              Xj[  - 0.0694 {0.012 exp [-0.11 01^/5)1/7S] *
              0.00068 exp [-0.046 (^75)1-76]}  (i-1,2,3)
                                - 0.00018
                             VAT
0.0694 {0.012 expt-O.llOfy/S)1-76]} + 0.00086
                          - 0.693     (i - 1,2,3)              (C-90)
               - 0.00068 {1 - 0.0694 exp [-0. 046 On./S) ^"l }     (C-91)
                                         - 2,3)                   (C-92)
                                          Xff                 (094)
                                  C-29                        03/27/90

-------
           DRAFT  — DO NOT QUOTE OR  CITE
TABLE C-3  RATIO OF PULMONARY SURFACE AREAS BETWEEN HUMANS
           AND RATS AS A FUNCTION OF HUMAN AGE
                 Aoe (Yean    Surface Area
0
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
4.99
17.3
27.6
36.7
44,7
51.9
58.5
64.6
70.4
76.0
81.4
86.6
91.6
96.4
101
106
110
115
119
123
128
132
136
140
144
148
                         C-30                          03/27/90

-------
   DRAFT — DO NOT QUOTE OR CITE

        1 (2)   i (2)   i (2)   i (2)
        * A   AAL  * '•AT * AAB
 0.0694 {0.012 expt-O.llOnj/A)1-76] +
0.00068 exp[-0.046(mJ,/5)1-76]} + 0.016
        .,(3) _ JO)   ,(3)   , (3)
        AA  - A;u, + AAT f A.JIB
 0.0694 {0.012 expt-O.lK^/5)1-76]  *           (C-96)
 0.00068 exp[-0.046(^/5)1-76}  +  15.71
        1(3) m i<3)    1<3)    ,(3)
        AA  • AJU,  *  AAT  *  Aj^fl
 0.0694 {0.012 exp[-0.119^/5)1'76]  +           (C-97)
 0.00068 exp [-0.046 (m^/5)1-76}  + 15.71
                   C-31                           03/27/90

-------
                  DRAFT — DO NOT QUOTE OR CITE

REFERENCES

Amann, C.A.; Siegla, D.C.  (1982)  Diesel particles — What
   are they and why.  Aerosol Sci. Tech. 1:73-101

Bailey, M.R.; Fry, F.A.; James, A.C.   (1982)   The long-term
   clearance kinetics of insoluble particles from the human
   lung.   Ann.  Occup. Hyg. 26:273-289.

Bond, J.A.; Sun, J.D.; Medinsky, M.A.; Jones, R.K.; Yeh, H.C.
   (1986a   )   Deposition, metabolism and excretion of 1-
   [uC]nirropyrene and l-[uC]nitropyrene coated  on diesel
   exhaust particles as influenced by exposure concentration.
   Toxicol. Appl. Pharmacol.  85:102-117.

Chan, T.L.; Lee, P.S.; Bering, W.E.   (1981)  Deposition and
   clearance of inhaled diesel exhaust particles in the
   respiratory tract of Fisher rats.  J. Appl. Tox. 1:77-82.

Diu, C.K.; Yu, C.P.   (1983)  Respiratory tract deposition of
   polydisperse aerosols in humans.  Am. Ind. Hyg. Assoc. J.
   44:62-65.

Schanker,  L.S.; Mitchell, E.W.; Brown, R.A.  (1986)  Species
   comparison of drug absorption from the lung after aerosol
   inhalation or intratracheal injection.  Drug Metab. Dispos.
   14(1):79-88.

Scheutzle, D.  (1983)  Sampling of vehicle emissions for chemical
   analysis and biological testing.  Environ. Health Perspect.
   47:65-80.

Schum, M.; Yeh, H.C.  (1979[?])  Theoretical  evaluation of aerosol
   deposition in anatomical models of mammalian  lung airways.
   Bull.  Math. Biol. 42:1-15.

Strom, K.A.; Chan,  T.L.;  Johnson, J.T.  (1987)  Pulmonary
   retention of inhaled submicron particles  in rats:  diesel
   exhaust exposures and lung retention model.   Research
   Publication GMR-5718, Warren, MI: General Motors Research
   Laboratories.

Strom, K.A.; Chan,  T.L.;  Johnson, J.T.  (1988)  Inhaled particles
   VI.  Dodgson, J.; McCallum, R.I.; Bailey, M.R.; Fischer,  D.R.,
   eds.  London: Pergamon Press, pp. 645-658.

Sun, J.D.; Woff, R.K.; Kanapilly, G.M.; McClellan, R.O.  (1984)
   Lung retention and metabolic fate of inhaled  benzo(a)pyrene
   associated with diesel exhaust particles.  Toxicol. Appl.
   Pharmacol. 73:48-59.

                               C-32                        03/27/90

-------
                  DRAFT — DO NOT QUOTE OR CITE


Weibel, E.R.  (1963)  Morphometry of the human lung.
   Berlin:Springer-Verlag.

Yeh, H.C.; Schum, M.   (1980)  Models of human lung airways and
   their application to inhaled particle deposition.  Bull. Math.
   Biol. 42:461-480.

Yu, C.P.  (1978)  Exact analysis of aerosol deposition during
   steady breathing.  Powder Tech. 21:55-62.

Yu, C.P.; Diu, C.K.; Soong, T.T.   (1981)  Statistical analysis of
   aerosol deposition in nose and mouth.  Am. Ind. Hyg. Assoc. J.
   42:726-733.

Yu, C.P.; Xu, G.B.   (1986)  Predictive models for deposition  of
   diesel exhaust particiates in human and rat lungs.  Aerosol
   Sci. Tech. 5:337-347.

Yu, C.P.; Xu, G.B.   (1987)  Predicted deposition  of  diesel
   particles in young humans.  J. Aerosol Sci. 18:419-429.
                                C-33                       03/27/90

-------
APPENDIX D

-------
                       D.   EXTRAPOLATION MODEL
 5    D.I. INTRODUCTION
            Calculations of unit risk estimates based on animal inhalation data were presented
      in Chapter 10.  Alternate methods of quantitation are presented in this appendix that use
      animal and human data to construct dosimetiy models for assessing carcinogenic risk in
10    humans from exposure to diesel engine emissions.
            Both chronic animal bioassays as well as human epidemiological data provide at least
      some evidence  for the carcinogenicity of diesel engine emissions.  Use of human data for
      estimating dose response has several advantages. Species extrapolation with its associated
      uncertainties regarding differences in such parameters as delivered dose, and target organ
15    sensitivity  is  avoided.  On the other  hand,  the animal  biossays  are carried  out  under
      controlled conditions allowing for greater accuracy in estimating dose. Confounding factors,
      especially those relating to life style, are also avoided. In the case of diesel engine emissions,
      the animal studies also provide unequivocal evidence for carcinogenicity, whereas the human
      evidence is less certain.   Because  data are available  for both animals and humans and
20    because use of each has advantages and disadvantages, quantitative risk estimates will be
       carried out for both.  In the present chapter,  dosimetry models for extrapolating risk from
       rats to humans will be illustrated.
             In the development of  dosimetry models, a variety of factors must be considered.
       The first of these is rate of respiratory  gas exchange. Since respiration is seldom measured
25     in chronic animal bioassays, estimates  are usually used. The concentration and properties
       of the agent and the time of day of exposure  are the primary factors inducing variability of
       respiratory exchange rates in animal exposures.  Hesseltine et  al.  (1985) showed,  for
       example, that exposure of rats at night when they are awake and active results in minute
       volumes considerably greater than  daytime values.  Of the chronic diesel exposure studies,
30     only those by Brightwell (1986) included exposure during night-time hours. Although mice
       have been shown to decrease respiration greatly during exposure to an irritant gas, rats show

       May, 1990                              D-l     DRAFT - DO NOT QUOTE OR CITE

-------
       much less inhibition, so this is not likely to be a large potential source of error in estimating
       target organ dose (Chang et ah,  1983).  Since humans are highly variable in this respect,
       because of differences in activity levels, estimated respiratory values well above resting levels
       are generally selected.
 5           A second important dosimetric variable is  particle  deposition efficiency.  Since
       transport from the nasal region and conducting airways is more rapid than desorption of
       organics from the diesel particles, this fraction probably contributes little to the effective
       dose. Deposition in alveolar regions is of primary importance because particles not only
       remain  long enough to accumulate and possibly induce particle related effects, but also
10     because they remain long enough for the carcinogenic agents present on the surface of the
       particles to desorb and become bioavailable.
             Since desorption of the organics from diesel particles is expected to be fairly complete
       in animals as well as humans, bioavailability is not expected to differ greatly either because
       of species differences in clearance rates or impaired clearance from high exposure levels (see
15     Chapter 8 for pharmacokinetic  information).  However, clearance rates are  important
       because, as mentioned above,  accumulation  of the particles themselves may have effects
       independent of the PAHs and other potentially harmful agents present on the surface of the
       particle.
             A complete model  would  also consider  the gaseous components of the diesel
20     emissions.  Some of the diesei  toxicology studies have included groups of animals exposed
       to exhaust filtered to remove the paniculate matter and have subsequently provided negative
       results.  The reverse experiment, exposure to paniculate matter alone, unfortunately has not
       been carried out because of the  technical difficulties of such an undertaking. Thus, while
       the gaseous components alone do not appear to be carcinogenic, it is uncertain whether they
25     may exert some promotional, or even inhibitory, effect on tumor induction by the paniculate
       fraction. Because of this uncertainty, there is no reasonable method for factoring effects of
       gaseous components into the dosimetry model.
             Finally, transport of particles to lung-associated lymph nodes and to lung surface area
       will be adjusted for.
30
       May, 1990                               D-2    DRAFT ~ DO NOT QUOTE OR CITE

-------
       D.2.  BASIS FOR EXTRAPOLATION MODEL
             Two different extrapolation models will be developed.  In the first one it will be
       assumed that the retention and accumulation of inhaled paniculate matter in the lungs plays
 5     a significant role in the carcinogenic  process  and that the retardation  of alveolar
       macrophage-mediated clearance of particles has an influence on effects; i.e.,  the particle
       mediated impairment of alveolar clearance, which was observed in rats will be extrapolated
       to humans. The possibility that retention and accumulation of organic material desorbing
       from particles is primarily responsible for the carcinogenic effects will also be considered.
10     In the second model, the delivered dose rate of the organic material together with the
       particles and the duration of exposure (competing risks) as well as cell proliferation and
       transition rates will be considered the most important parameters.

       D.2.1. Principle of Extrapolation Based on Deposition and Retention
. 15            Modeling
             The principles of this approach have been described by Oberdfirster (1989) and is
       based on species specific lung dosimetry as depicted in Figure D-l.  Using deposition and
       retention  data for inhaled diesel particles for the rat as described  by Yu and Yoon
20     et al (1988) a retained dose per gram lung or per unit epithelial surface area of the lung
       can be calculated for each of the exposure situations of the different rat studies.  The
       retention model by Yu  et al.  (1987) takes into account a retardation of particle clearance
       from the alveolar region at high inhaled concentrations  that is due to a particle overload
       effect. Assuming that the long-term retained dose must be the same in rat and man to
25     induce effects, a deposited dose for  the human lung can be calculated from  the retained
       dose - applying human specific retention half-times - to arrive at an inhaled dose and at an
       Equivalent Human Exposure (EHE). This EHE is different from the rat exposure, but -
       according to the assumption made about the retained dose, it results in the same effects in
       humans as in rats.
30          The following steps are involved in this model: (1) The retained doses resulting from
       chronic exposure of rats under the different exposure conditions of the reported rat studies

       May, 1990                              D-3    DRAFT - DO NOT QUOTE OR CITE

-------
O
I
§
                 Bat
                   I
              Exposure
                  I
            Inhaled Dose \ny(kg) '\
        Deposited Dose [
                                             Lung Dosimelry
                                  Breathing
                                Minute Volume
                                            Tidal Volume. Resp. Ftnlo
                                                 Resp. Pause
                                                   characteristic
                                                  Anatomy
                                                 Clearance
                                                 Retention
                                               Regional Uptake
                                              (Metabolism.  T±.)
                                       Retained (Accumulated)  Dose
                                              li'
-------
      will be calculated and plotted against the observed respective lung tumor incidences, using
      a logistic regression.  (2) Respective EHEs for human exposures can then be calculated
      according to Figure D-l.  On the basis of the logistic model, the tumor risks for humans for
      relevant low inhaled concentrations of diesel exhaust can then be estimated.
 5
      D.2.L1. Deposition and retention model for inhaled diesel exhaust
            A mathematical model for deposition and clearance of diesel exhaust particles has
      been developed by Yu and Yoon (1988).  The retention model used by them is shown in
      Figure D-2. Diesel exhaust particles, consisting of a carbon core (80 percent) and adsorbed
10     organics (Figure D-3), are cleared from the lung by mechanical processes  (mucociliary
       escalator and  alveolar macrophage  mediated)  and by dissolution processes.  Half of the
       organic material is strongly bound to the core and half of it is weakly bound. The transport
       rate A (% of particle component i  from compartment x to compartment y that is due to
       dissolution is assumed to be  constant. However, transport rates for mechanical clearance
15     from the alveolar compartment vary with the paniculate lung burden when this lung burden
       reaches a certain value, as  has been observed in the animal studies  with high  inhaled
       concentrations  of diesel exhaust.   This functional  dependence of mechanical  alveolar
       clearance  rates on paniculate lung burden was determined by the best fits of experimental
       data and model predictions for rats.  The extrapolation to humans from the results in rats
20     was accomplished under the assumption that the relative responses (impairment of clearance
       due to particle load) of the transport rates of particles to the same paniculate dose per unit
       alveolar epithelial area were independent of the species. However, human specific clearance
       rates for normal, undisturbed particle clearance from the alveolar region were used as the
       starting point
25            At  low particle burdens the alveolar  clearance  rate was  essentially the normal
        clearance rate controlled by macrophage migration to the mucociliary escalator, whereas the
        alveolar clearance rate at high paniculate burdens was principally determined by the
        transport  to the lymphatic system and eventually to the blood compartment once alveolar
        macrophage-mediated clearance ceased completely (Yu and Yoon, 1988).

        May, 1990                              D-5    DRAFT - DO NOT QUOTE OR CTTE

-------
                     (i)
                                  H

1 1
8


'Ve
x(i)
_ AAB
ALB

H
T
T
AHG .
^TC

tx(il
|AAT
)
%
rA
<>
G

A
1


r^
L
             Figure D-2. Model of panicle retention in the lung.
May, 1990
DRAFT - DO NOT QUOTE OR CITE

-------
         carbon  core
  2 organics  strongly
    bound to core

  3 organics weakly
    bound to core
             Figure D-3. Model of diesei exhausi panicle.
May, 1990
D-7    DRAFT - DO NOT QUOTE OR CITE

-------
             The transport rates used in the model by Yu and Yoon (1988) are listed in Table D-l.
       The material components i are as follows:
             1 = carbon core
             2 = strongly bound organics
 5           3 = weakly bound organics
       The conclusions reached by Yu and Yoon (1988) based on their model are given below and
       are depicted in Figures  D-4 through D-7:
         •  When exposed to  the same exposure concentration of diesel exhaust particles, the
            reduction in the clearance of the carbon core in humans due to high paniculate
10          burden is smaller than the reduction in rats; children have a larger reduction than
            adults.
         •  The organics and  carbon core burdens in various  compartments vary with age of
            humans for the same exposure period.
15
         •  The carbon core burden per unit exposure concentration in the lung reaches a steady
            state value after a long period of exposure. The steady state value increases with
            increasing exposure concentration level.
20     By using the specific exposure data of the different rat inhalation studies for calculating
       particle deposition and the estimated specific clearance rates given in Table D-l one can
       calculate the accumulated lung burden both for the carbon core and for the organics. The
       resulting lung burdens of the organics after exposure for 24 mo and the observed lung tumor
       incidences in rats (observed at 30 mo) are shown in Figure D-8. Lung tumor incidence is
25     expressed here as the ratio of the lung tumors observed in the exposed animals (P) vs. those
       found in the unexposed (control) animals [P(O)j.  These data can be described by a logit
       function with the following expression:
                                P = 100/[l+exp(3.88-ll/374 mj].        (1)
       Since the experimental data of Mauderly et al. (1987)  are most complete with regard to
30     different exposure concentrations and different time points of study termination their results
       were also plotted separately (Figure D-9), which gave the following logistic expression (based
       on exposure of 24 mo and tumor data at 30 mo):
                                P = 100/[l+exp(4.709-40.544 mj].       (2)

       May, 1990                              D-8     DRAFT - DO NOT QUOTE OR CITE

-------
           TABLE D-l.  TRANSPORT RATES USED IN RETENTION MODEL FOR
      	DIESEL EXHAUST PARTICLES	
 5    ——-————-----———-—--—--—--——-•—-—
            The values of X (J/ in the unit of day'1 that we used in the model calculation for rats are listed
      below:
10

        X^  - 1.73,    i - 1, 2, 3                  (Chan et aL, 1981; ICRP, 1979)

15

         o\    cy\    CT\    C2\
        XWB  " xra "  *-LB " IAB " a023           (Sun et aL>  1984;
                                                Bond et aL, 1986)

20

         (31    (2}    (3)    (31
        1\ /  _ i \.JJ _  i \JJ _ i ^J/ _ 19 <7           /'Qiin ^t al  IQftd*
        Arrij  s  Tf)    t P    >dP   1*~J I           ^>JU11 Cl ai.,  17OH,
                      x-z>    /u>                  Bond et aL, 1986)

25


        x*j  " 4\i2'   '" *3                   (ICRP'1979)


30

          fi. - 0.693,   i - 1, 2, 3                 (Chan et aL, 1981)
             The following values of X ^ were determined from the data on lung burden and lymph node
       burden for rats (Strom et aL, 1987|.


 40

              "  °-012 exP(-(U1 m"*) + 0-00068 exp(-0.04 mi62)  i - 1, 2, 3
 45


       May, 1990                             D-9     DRAFT ~ DO NOT QUOTE OR CITE

-------
20
                                     TABLE D-l (Continued)
              0.00068[1 - exp(-0.046 mi'62)]
        A.    = 0.00018
10
        X/B " xra " *S *a00018
15
               0.012 exp(-0.11 mi76) + 0.00086
25      A . x® + 0^88
          A     AT
                     15.71
30
35      where mAis is the paniculate burden (rag) for the alveolar compartment;  as mA - 0, A.\2 « 0.0129.
         For humans,     - 0.00169      (Bailey et aL, 1982)
40                   AU
       May, 1990                              D-10    DRAFT - DO NOT QUOTE OR CITE

-------
I
 s
D
O


1
O
c:

§

§
l-*igmc !>••!.   Variations ttf L-lciiiancc rules.

-------
                          1 il
                                                                J.U
 I
o
o
1
o
o
                                               A Wolf fetal. H985)
                                               o llcmricliclal. (1986)
                                               « Shoinelal. (1987b)
                                               u
I	....
8
                                                      I.
                                                1
2-168        JO       \Z
   I niifj Uurden per Gram  of I UNO, mn/g
                                                                        11
  Figure D-5.  Change of alveolar clearance rate with lung burden during chronic diesel
  exhaust exposure.

-------
o
 i
O
O
2

9
O


I
O
&
               U
               E
               M
               E
               c
               o
               u
               c
               o
               o
              -o
              u
               00
               c
                                                                          I O ill p. / 111
                                                                                   \ .0

                             0
0.2
 I

0.4
—I	1 —

        O.G
 (Thousands)
0.8
                                                   rxponurr Time   (uerfc)


                        Figure D-6. Calculated Jung burdens of the carbon core per unit exposure concentration

                        in human adults exposed continuously to diesel exhaust of different concentration levels

                        of 5 d/week and 8 h d.  The dashed lines indicate the corresponding burdens in the rats.

-------
8

I
O

I
O
 to
 e
 «D
 e
 3
_*
 3
 9
 •J

 3
o
"x
 h.
 3
•3
           0
0.2
0.4
      0.6        0.8         1
            (Thousonda)
Exposure Time  (week) (7d. 12M)
      Figure D-7.  Calculated lung burdens of the carbon core per unit exposure concentration
      in humans from birth to adulthood exposed for 12 h/d, 7 d/week to diesel exhaust of
      different concentration levels.

-------
             f - 100/M  <  c.ip(3.388 - 11.37* mA(ng)/E)l
<5
















o
Cft

D

§
0
O
2!
2
M
O
g
a
i

o
p«
»
0
O
C
0
o
0
i.
0.
l_
o
E

l~
o»
c
•o
0
3
§
o









i*-
Organlcs
B

.
10 -


9 -


8 -


7 -


6 -
i •








*
*


'i
4 -

3 -
2 -
M

B *
K
^
.
1 T '. K K
K
o -$
f*1 K R
r i i i i i i i i i i
0 0.04 0.03 012 0.16 0.2 0.24

lung Burden/Lung Weight (mg/g)
Figure D-8. Correlation between normalized lung tumor prevalence and calculated lung
burden of organics from different rat studies. Asterisks represent logit fit of data,
function shown on top.
              Key to individual studies:  (see Ishinishi et al., 1986):
              B = Brightwell; K = Ishinishi, M = Mauderly; S = StOber,  Z = Iwai.

-------
             P - 100/H I ixp(A./(!'» - '.0.544 mA(mr,)/r.) I












d
O\


d
}w
£
,
d
O
5«
^^
~j
0
c:
2
o
js
0
OL.

O
O
C
o
(L
0
1-
o>
c
a
TJ
V
N
O
§
O











1 3
14

13 -

12 -

11 -
10 -
9 -
8 -


7 -

6 -

5 -

4 -
3 -

2 -

.. |
l T
Organics ft
















,
M



M
*
• i 1 1 i i
0 0.02 0.04 0.06

lung Burden/Lung Weight (mg/g)
Figure D-9. Correlation between normalized lung tumor prevalence and calculated lung
top.

-------
       Figures D-10 and D-ll depict the same relationships for lung tumors vs. lung burden based
       on the carbon core:
             P = 100/[l+exp(3.37-0.0874
 5     for data from all experimental rat studies, and
             P - 100/[l+exp(4.2633-0.1504 mj](4)
       for data from Mauderfy et al. (1987) only.

             Different exposure scenarios for exposures of rats to diesel exhaust can be simulated
10     (e.g. 100 Mg/m3,  24 h/d, 7 d/week).  These will eventually result in lung burdens (mA) of
       accumulated particles or associated organics which can be predicted with the deposition and
       retention models by  applying the specific transfer rates for organics and the carbon core
       listed in Table D-l.  With Equations (1) through (4) one can then estimate a certain lung
       tumor prevalence in the rat that correlates with the respective exposure concentrations. The
15     result of one example  is shown in Figure D-12. For a given exposure concentration, the
       predicted lung tumor prevalence is different depending on whether the prediction is based
       on the accumulated organics or on the accumulated particles of the carbon core. This result
       is difficult to interpret, showing that the organics have a lower impact than the particles at
       high inhaled concentrations. Ideally, both curves should fall together, it is possible that the
20     divergence of the two  curves reflects the importance of exposure duration. The exposure
       duration is only partly considered in this model through the retention half-times for particles
       and for organics.  Since they differ, equilibrium lung burdens are reached much faster for
       organics than for the carbon core. Thus, the equilibrium in the lung of the retained dose,
       which is reached very quickly for organics, will not change  during  further continuous
25     exposure over 2 yr;  yet it is very likely that this continuing exposure after equilibrium is
       reached will have a major influence  on tumor  induction.   For particles,  inhaled at a
       concentration above a few milligrams per cubic meter, retarded clearance will become an
       increasingly important  factor for their accumulated dose over an extended exposure period.
        May, 1990                               D-17     DRAFT - DO NOT QUOTE OR CTEE

-------
100/1
                      exp(1.37 - 0.0874 mA(mg)/r,)|

















d
H*
00


d
5
3
i
8
25
3
O
a
o
a
o


o
fb
ru

0
O
C
o
9
i.
a

o
E

o>
c
3

O
N
75
§
o
z








1 .)

12


11

10

9 -



8 -

7 -

6 -


5 -

4 -

3 -

2 -
1 -
0 -
7
Carbon Core

B



*







*
*



S

ML
**
8 •
* 1
K
*
** * M
K
° T> 1 1 1 1 1 1 1 1 1 1 1 1 1 1 I
0 -1 3 12 1G 20 24 28 32

lung Burden/Lunq Weight (mq/g)

Figure D-10. Correlation between normalized lung tumor prevalence and lung burden of
diesel exhaust particles from different rat studies.  Asterisks represent logit fit of data,
equation shown on top.  See Fig. 8 for key to individual studies.

-------
       P • 100/11 4exp(«.2633 - 0.1504  «A(ing)/g)|














HH
\J
^.
VO



d
C
M
H
1
d
O
|
O
O
a
0
III
0
a?
* ^
• 8
0 °
c
I 7 -
i.
Q.
o 6 -
E
3
*"" «\ -
o»
c
3
4-
•o
0
N
c 3 "
5
o
2 2-


1 -
1

Carbon Core
ffl


















'
M
*





0 - i i i i i i i i - - -| •• — |- i 	 r • • -7 • i i
0 2 4 6 8 10 12 14 IE
1 unq But dcn/l ling tVri(|lil (niq/a)
^-^« ir-^ 4 4 ^~i • * • L * 	 I* 	 If t t fll • • f
Figure D-ll. Correlation between normalized lung tumor prevalence and lung burden of
diesel exhaust particles from study by Mauderly et al. (1987) Logit function shown on
top.

-------
d
O
O
I
O
I
              0»
              o
              «
              o
              •
             
-------
            The impact of diesel exhaust exposures on lung clearance in rats is examined first in
      the following section.  Next, based on the retention model by Yu and Yoon (1988) and the
      extrapolation model outlined in Figure D-l, separate EHEs are estimated for a retardation
      in alveolar particle clearance (overload related) on the one hand and for rumor incidences
 5    on the other hand. This extrapolation model assumes that effects on clearance retardation
      are correlated to the particle load (carbon core) in the lungs, whereas lung tumor incidences
      are predicted separately as either a function of the load of particles or of the organics.

      D.2,12.  Extrapolation of lung tumor risk based on retention of diesel exhaust particles
10              and adsorbed organics
      DJ2.L2.1. Influence of particle load on retention
             The retardation of alveolar particle clearance found in the lungs is shown in Figure
15     D-5, which was discussed above. This figure shows the ratio of the actual alveolar particle
       clearance, A A ^ over ^c normal, undisturbed alveolar particle clearance, Aj^ for different
       lung burdens both per unit lung weight (lower x-axis) and per unit pulmonary surface area
       (upper x-axis).  Figures D-13  and D-14 show the extrapolated EHE's leading to alveolar
       clearance retardation in humans based either on the particulate burden per lung weight or
20     per pulmonary surface area.  Two cases  were considered  for each  category:  either
       continuous exposure for 24 h/d and 7 d/week (environmental exposure)  or for 8 h/d and 5
       d/week (occupational exposure). The extrapolation predicts, that a continuous exposure to
       a concentration of 1 mg diesel exhaust/m3 may result in a significant impairment of particle
       clearance in humans (Figure D-14) whereas at more relevant concentrations of 100 jj/m3 and
25     below no such effect should be expected. Likewise, at occupational exposures to 100 to 350
       /xg/m3, which were estimated in the study by Garshick et al. (1987), no significant effect can
       be  predicted  either.   If this  is correct,  observed carcinogenic responses in  quoted
       epidemiological studies may not have been influenced by increased particulate accumulation
       caused by an impairment of particle clearance. Since
        May, 1990                              D-21     DRAFT - DO NOT QUOTE OR CITE

-------
d
 I
o
o
1
§
  1


0.9 ~

0.9 -
                         v.4  -
                        •M -
   10
     -2
                                    (A)  24 hrs/day and 7 days/wk
                                    (B)   8 hrs/day and 5 days/wk
 i      I      I
io-i

Soot  Concentration


•
\
(A)









\
\
(B) \
\
'.
',
\ A
\ \
1
\ \
\ \
• 1
\
\_____ \

1.0
10
                        Figure D-13.   Predicted impairment of alveolar particle clearance in humans due to
                        accumulation of diesel exhaust particles in the alveolar region (per unit surface area) for
                        different inhaled concentrations and different exposure scenarios.

-------
 I
o
o
1
*

         v.1 -
            10
              -2
d 7 days/wk
d 5 days/wk










i 1 1 i
li>~' 1.
(B)\
\
\
\
\
\
(A) \
*
I
\
\
\
\
1 1 1 I
0 10
                                      Snot Concentration tng/m^
          Figure D-14.  Predicted impairment of alveolar particle clearance in humans due to
          accumulation  of diesel exhaust particles in the alveolar region (per unit lung weight) for
          different inhaled concentrations and different exposure scenarios.

-------
       alveolar macrophages are the mediators of mechanical clearance of particles from the
       alveolar compartment an effect on clearance might better be expressed based on paniculate
       burden of alveolar macrophages.  Therefore, assuming that the frequency distribution of AM
       alveolar macrophages per pulmonary surface area is similar in rat and man (Crapo et aL,
 5     1983), the predictions in Figure  D-14, based on pulmonary surface area, might be  better
       justified than predictions based on lung weight (Figure D-13).

       D2JL12. Extrapolation of lung tumor prevalence based on accumulated doses of carbon
10              cone and organic*
             Assuming that the same accumulated doses of either the paniculate carbon core or
       the adsorbed organics per unit lung (mass or epithelial surface area) of rat and man will
       induce the same response, the data presented in Figures D-8 through D-ll, showing the
15     relationship between lung tumors and lung doses, and in Figures D-13 and D-14, showing
       clearance rates vs. paniculate lung burden, were used to extrapolate EHEs for predicting
       lung tumor frequency in  man.  Again, two exposure scenarios were simulated, one for
       continuous  24 h/d  and 7 d/week (environmental)  and one  for 8  h/d and 5 d/week
       (occupational) exposure.  Extrapolations were also based either on the experimental data
20     from all rat exposure studies or  solely on the data by Mauderly et al. (1987).  In addition,
       in analogy to Figure D-12, predictions of lung tumors were made considering either the
       accumulated particles (carbon core) alone or the organics alone. The results are shown in
       Figures D-15 through D-18.  Depending  on the  data base used, the results  predict a
       relatively low excess lung  tumor incidence of 2J to 9 percent for continuous exposure for
25     20 yr at 1 mg/m3; for occupational exposure to the same concentration and for the same
       duration, it is  between  0.7 and 23 percent   Extrapolated risks for more relevant
       concentrations of 100 pg/m3 are 10"3 to 10"* (Table D-2).
              Compared to results from  epidemiological studies by Garshick et al. (1987) these
       numbers are very low.  However, as stated before, these  model extrapolations based on
30     accumulated lung doses of either the paniculate carbon core or the adsorbed organics do
       not take into consideration the influence of a continuing exposure after steady-state values
       of lung doses of the soot particles and organics are reached. It is very likely that such
       May, 1990                               D-24    DRAFT - DO NOT QUOTE OR CITE

-------
D

fc
 8
 as
 9
o

I
g
                  u
                  c
                  •

                  o

                  V

                 
-------
3
O
o
so
                o
                c
                0
                0>
                c
                3
                       1.6
                       1.5
                       1.4
                       1.3 -
                       1.2 -
                      1.1 -
                        1  1
                          10-2
                                (A)  24 hrs/day and 7 days/wk



                                (B)   8 hrs/day and 5 days/wk
                                Soot Concentration    mg/m3
Figure D-16.  Predictions of lung tumors in humans for different exposure conditions

based on accumulation of the particulate carbon cord (CC) or of organics (O) per unit

surface area of the pulmonary region.  Extrapolated from results of all rat studies.
                                                                                            <*>
                                                                                            >*»

                                                                                            o

-------
 I
 D
 O

 I

§
g
 u
 c
 i.
 o
 o«
 c
 3


 •Q
 9
 N
 w
I
                   (A)  24 hrs/day and  7 days/wk

                   (B)   8 hrs/day and  5 daya/wk
                                                                                             10
                                   Soot Concentration  mg/m3


       Figure D-17.  Predictions of lung tumors in humans for different exposure conditions

       based on accumulation of the particulate carbon core (CC) or of organics (O) per unit

       lung weight.  Extrapolated from result of study by Mauderty et al. (1987).

-------
g
 1
o
o
               o
               c
               o
               •
               
               c
               3
N
O
 2.6
 2.5 -
 2.4 -
 2.3 -
 2.2 -
 2.1 -
  2 -
 1.9 -
 1.8 -
 1.7 -
 1.6 -
1.5 -
1.4 -
1.3 -
1.2 -
1.1 -
  1 -
                                  (A)  24 hrs/day and 7 days/wk
                                  (8)   0 hrs/day and 5 days/wk
                          10-2
o
I
o
                                                   Soot Concentration
         Figure D-18.  Predictions of lung tumors in humans for different exposure conditions
         based on accumulation of the particulate carbon core (CC) or of organics (O) per unit
         surface area of the pulmonary region. Extrapolated from result of study by Mauderly et
         al. (1987).

-------
                                         TABLE D-2
10


15



20
           EXTRAPOLATED LUNG TUMOR RISK FROM EXPOSURE TO DIESEL
         EXHAUST FOR CONTINUOUS (24 H/D, 7 D/WEEK) OR DISCONTINUOUS (8
        H/D, 5 D/WEEK) CHRONIC EXPOSURE. BASES FOR EXTRAPOLATION ARE
                              AS INDICATED IN FIGURES 14-17.
                                                     Exposure concentration (mg/m3)
                           Bases of extrapolation
                            0.01
                      1.0024  1.0
10.0
25
30
35
Rat studies,
  continuous exposure
  (Mauderty et aL, 1987)
Lung weight
                                Pulmonary
                                surface area
                                Lung weight
                         Pulmonary
                         surface area
Organic        1.002   1.0024  1.024   1284

Carbon core      1.00023 1.0024  1.024   1.57

Organic        1.00024 1.0024  1.025   1.286

Carbon Core     1.00024 1.0024  1.03    1.58

Organics        1.0008  1.009   1.091   2.49

Carbon core      1.0004  1.004   1.043   122

Organics        1.0009  1.009   1.093   ZSO

Carbon  core      1.0004  1.004   1.053   1236
 40
 45
 50
       Rat Studies
       discontinued exposure
       (Mauderly et aL, 1987)
                         Lung weight
                         Pulmonary
                         surface area
                                Lung weight
                                 Pulmonary
                                 surface area
               Organics        1.00005  1.0005  1.0055 1.056

               Carbon core     1.00006  1.0006  1.0056 1.061

               Organics        1.00006  1.0005  51.006 1.056

               Carbon core     1.00006  1.0006  1.0057 1.115

               Organics        1.0002   1.002  1.02   122

               Carbon core     1.0001   1.001  1.01   1.11

               Orgnaics        1.0002   1.002  1.02   1234

               Carbon core     1.0001   1.001  1.01   121
        May, 1990
                                       D-29    DRAFT - DO NOT QUOTE OR CITE

-------
       continuing exposure is of decisive influence on lung tumor rates and that the dose rate at
       which a carcinogenic compound is delivered to the target cell rather than the accumulated
       dose determines the chances for transformation. An extrapolation model  based on the
       delivered dose rate to target cells and the exposure time and the proliferation and transition
 5     rates of these cells will be discussed in the following section.
       D22. Risk Extrapolation Based on Competing Risks and Proliferation
              Rates of Target Cells
10           If the induction of lung rumors  after diesel exposure is due  to  an increased
       accumulation of particles (via direct mechanisms), one should expect that effective clearance
       of these particles from the lung - i.e., no impairment of lung clearance mechanisms • would
       prevent the induction of such tumors unless these tumors are caused additionally or solely
       by the adsorbed  PAH (organics,  direct carcinogens).   Indeed,  retardation of particle
15     clearance with increased accumulation of soot particles and lung tumors have been observed
       in rats  during chronic exposures  exceeding about 1  mg/m3,  whereas low  exposure
       concentrations (035 mg/m3) did not impair clearance mechanisms  and an increased tumor
       incidence was not seen (Mauderty et aL, 1987).
             There is some  direct evidence for a particle effect.  In studies with inhaled  TjO2
20     particles of low cytotoxicity, tumors  could be induced, but only at exposure concentrations
       much greater than used in any of the chronic diesel studies (Lee et al., 1986). On the other
       hand, ten consecutive doses of carbon black particles instilled intratracheally were found to
       be as effective for induction of lung tumors as the same amount of instilled diesel exhaust
       particles, indicating the possibility of a pure particle effect (Kawabata et al., 1986). This is
25     a remarkable result, but  needs confirmation by inhalation  studies to exclude a possible
       influence of repeated mucosal wounding on tumor induction (Saffiotti  et al., 1988).
             If relatively inert particles can induce lung cancer, it is uncertain how this occurs.
       One possibility is that activation, cellular injury, or death of particle-laden macrophages may
       result in the release of a variety of mediators including reactive oxygen species, chemotactic
30     factors, lysosomal hydrolases, other proteinases, prostaglandins, plasminogen activators, and

       May, 1990                              D-30    DRAFT - DO NOT QUOTE OR CITE

-------
       growth' activators that could cause, by unknown mechanisms, further lung injury, including
       tumors. This condition is exasperated by the tendency of the macrophages to aggregate,
       especially for large lung burdens, thus increasing the local concentration of the potentially
       damaging factors.
 5           Nevertheless, even if particles can be shown to induce lung tumors under certain
       conditions, it is unlikely that this is the sole mechanism, or even the predominant one for
       induction of lung tumors by diesel exhaust More likely, panicles could act as cofactors,
       either by release of a variety of factors from the macrophage or by optimizing the dose or
       dose rate of PAH to target cells by slow desorption from particles.  Thus, both particles and
10     organics may be responsible for the observed tumorigenic effect of inhaled diesel exhaust,
       with the organics acting directly and  the particles acting  indirectly as promoters, or as
       vehicles for carcinogen delivery.
              Desorption of PAH are relatively rapid compared to particle clearance and, therefore,
:       bioavailability can be expected to be high and not greatly increased by impaired clearance.
15     Although more rapid than particle clearance, PAH desorption  nevertheless occurs over a
       period of days.  This results in  a slower delivery of PAH to the target organs than, for
       example, administration of pure B[a]P, allowing for a greater fraction to be metabolized to
       an active form. Partial activation of the PAH by the macrophages should also increase the
       effectiveness of the delivered dose.  Studies showing the formation of greater amounts of
20     metabolites after exposure to B[a]P alone and studies showing the greater inducibility of lung
       tumors after B[a]P plus particles vs. B[a]P alone are consistent with these concepts.
              This concept seems to be contradicted by inhalation studies in rats with 1000-fold-
       higher concentrations  for B[a]P than in the diesel exhaust studies,  but with the same
       concentration of particles (albeit of different kind, Le. tar particles vs. diesel particles) which
25     resulted in the same tumor incidence in both studies (Heinrich et aL, 1986).  However, in
       accordance with the concept of delivered dose rate, this  result  can be explained by the
       kinetics of the  adsorbed  organics:  slow release  of adsorbed PAH  from carbonaceous
       particles on the one hand (diesel exhaust)  and fast release of adsorbed PAH from tar
       particles on the other hand (pitch pyrotysis studies).

        May, 1990                               D-31     DRAFT -  DO NOT QUOTE OR CITE

-------
             Thus, in this extrapolation model we will consider the dose rate of particles and
       adsorbed organics delivered over an extended period of time. A scheme of this concept for
       extrapolation from rat to man is shown in Figure D-19. The deposited dose rate (expressed
       per unit epithelial surface area) of the inhaled particles is essentially equivalent to the
 5     delivered dose rate of the organics to epithelial cells because a steady state between
       desorbed and deposited amounts is reached after about 3 mo of exposure (see rate
       constants, Table D-l). For estimating the deposited dose rates and inhaled dose rates, the
       rat and man specific deposition and breathing parameters are applied (Yu and Yoon et aL,
       1988).
10           As pointed out above, the duration of exposure plays an important role, too. For
       example, a very high dose rate delivered over a short time may cause fewer tumorigcnic
       effects than a lower dose rate delivered over a longer time, although the product of time
       and exposure may be the same. Thus, the product of time and exposure may not be a good
       exposure index for a carcinogenic compound. Duration of exposure and dose rate have to
15     be considered separately since tumor induction may not be adequately described in two
       dimensions. This concept of competing risks has ben discussed extensively and is accepted
       widely in radiation dosimetry (NCRP, 1980; Raabe, 1987), and it may also be applicable to
       chemical carcinogens (Krewski et al., 1981). Results from the diesel study in rats exposed
       for different time periods and at different concentrations by Mauderty et al. (1987) point in
20     the same direction.  These results show that exposure to 7 mg/m3 for 30 mo yielded a more
       than  fourfold  higher tumor  response than  exposure  to  3.5  mg/m3  for 30 mo, i.e.,
       corresponding to an increase in exposure index (times x exposure) by a factor of 2. It is
       difficult to compare the results between the different exposure groups of this study at time
       points prior to 30 mo because of the long latency period for tumor development («12 mo
25     in rats).
             Therefore, in analogy to radiation effects, separate influences of the delivered dose
       rate and of the exposure time on the risk of tumor development will be considered here for
       the carcinogenic effects of diesel exhaust  A schematic representation  of this concept is
       shown in Figure D-20. The dose rate to the epithelial cell in this figure could be converted
30     into an exposure concentration (EHE, mg soot/m3) according to Figure D-19.
       May, 1990                              D-32    DRAFT - DO NOT QUOTE OR CITE

-------
                                         DOSE Ft A TE vs. EFFECT
   Rat
    I
Exposure
                    I
              Inhaled Dose Rale [mg(dyi\
           Deposited Dose Rate
                                                  Particles
                                                                               Man
                                                                                t
                                                                        Exposure (EHE) \\\g(irP)1\
                                                                \
                                                         Inhaled Dose Rate
                                                       Deposited Dose Rate
a
o
§
O
Jtf
                                 Organics
                          Dose Rate to Epithelial Cells
                                     I
                                 Effects

  Figure D-19.  Extrapolation of results from inhalation studies from rat to man based on
  delivered dose rate to target cells.

-------
  Figure D-20.  Model of competing risks for lung tumor induction by inhalation of diesel
  exhaust considering delivered dose rate to epithelial target cells and exposure time.
May, 1990
D-34    DRAFT - DO NOT QUOTE OR CITE

-------
             In addition, the importance of the target cell proliferation rates and transition rates
       should be considered. As described in their two stage model for human carcinogenesis by
       Moolgavkar and Knudson (1981), the growth and differentiation rate of normal target cells
       into  normal cells and their  transition rate into intermediate cells can be important
 5     determinants for the transition into malignant cells (Figure D-21). Carcinogens or promoters
       could affect these different rates to increase malignant transformation.  For example, an
       increase in the normal proliferation rate of target cells or, even more, of intermediate cells
       (e.g., as  a response to increased particle load?) will increase  the probability of tumor
       induction and will shift the dose rate-response curves of Figure D-20 towards the foreground,
 10     Le. the exposure time is shortened until the effect occurs.   In contrast, if the proliferation
       rate of the intermediate cells is inhibited, then dose rate-response curves in Figure D-20
       move towards  the background, i.e., time to the same response increases.
             Available data from the rat inhalation study by Mauderly et al. (1987) allow  the
:       construction of dose rate-response and  time-response curves for the rat necessary for the
 15     construction of Figure D-20. However, for the extrapolation of the results to humans based
       on the two-stage model by Moolgavkar and Knudson (1981), knowledge of the basic  cell
       proliferation rates of the target cells (Type II cells,, their progenitor cells, bronchial epithelial
       and stem cells) and of the generated intermediate cells as well as the transition rates for rats
       and for humans, would be needed.
 20           Lacking such data - only limited data for the rat indicating a normal cell turnover
       time of alveolar epithelial cells of about 21 to 28 d could be found (Adamson, 1986; Sanders,
       1989) - an alternative approach could  be considered which is described in the following
       paragraphs. However, this approach needs to validated before it is applied to extrapolation
       from rat to man.
 25           As pointed out by Moolgavkar and Knudson (1981), one or both of two possibilities
       exist for a chronically administered carcinogenic agent to increase the incidence of cancer.
       (1) The carcinogenic agent could increase the transition rates, MI and/or M» to new constant
       levels; this would  imply  that  the risk for  exposed individuals relative to nonexposed
       individuals remains constant with time when chronically exposed to a given dose.  Only the
 30    intensity of the dose determines this risk, not the duration of exposure. (2) The carcinogenic
       May, 1990                               D-35      DRAFT - DO NOT QUOTE OR CITE

-------
                        First  Event
        Second Event
 Figure D-21. Two stage cancer model of Moolgavkar and Knudson (1980). S. = normal
 stem cell;  = intermediate cell; D = dead cell; M = malignant cell; M = rate at which
 first event occurs; n2 - rate at which second event occurs; a, = rate of division of
 intermediate cell; B2 = rate of differentiation and death of intermediate cell.
May, 1990
D-36    DRAFT - DO NOT QUOTE OR CITE

-------
      agent increases the proliferation rate of intermediate cells by a constant amount without
      effect on the transition rates; this implies that risk in exposed individuals increases with time,
      relative to  the unexposed.   This is  because  the  ratio  of intermediate  cells of
      exposed/unexposed increases with time.
 5           As will be shown later, diesel exhaust may act like an agent as described under (1)
      in the rate, i.eM no increase in relative tumor risk with exposure time at a given inhaled
      concentration. It is possible that proliferation rates of normal cells may also be affected by
      diesel particles which however, has only little effect on tumor incidence unless large changes
      in proliferation rates occur. Such large changes of normal cell proliferation that are due to
10    chronic irritation may occur particularly at high inhaled concentrations of diesel exhaust
             Differences in age, specific lung tumor incidence between unexposed rats and man
      may also  be due to differences in the transition rates, MI and p2.  Comparing these age-
      specific tumor incidences for unexposed humans (Doll, 1971) and unexposed rats (Mauderly,
       1987), an 18-fold higher tumor incidence at the end of life span was found in control rats
15     of the diesel study compared to humans (about 0.05 percent in humans vs. 0.9 percent in
       rats).  Table D-3 shows respective  data for other time points of rat and human life span.
       For rats,  respective  tumor incidence in unexposed animals were extrapolated from the
       limited data available for control rats. The total lifespan, 1, is equivalent to 80 yr in humans
       and to 30 mo in rats.
20
                                          TABLE D-3
         RATIO OF AGE SPECIFIC TUMOR INCIDENCE FOR UNEXPOSED INDIVIDUALS,
25                   RAT/MAN, FOR DIFFERENT FRACTIONS OF LIFE SPAN
          Fraction of lifespan             O5      0.625        0.75       0.875        1.0
30        Ratio rat/man	1.95     320	4.90	933	18.23

35
             However, whether and how these ratios can be used  for deriving a  factor for
       extrapolating from rat to humans is not obvious and requires some further discussion. One

       May, 1990                              D-37    DRAFT - DO NOT QUOTE OR CITE

-------
       could argue that if both age-specific incidence of lung tumors in unexposed individuals and
       lung tumor incidence in exposed individuals are due to effects on the transition rates (see
       above), then lung tumor incidence normalized for life span may have to be lowered by these
       ratios when extrapolating from rat to man.  This assumes that the transition rates /*,, and
 5     /*2 are equally affected in rat and man leading to a corresponding increase of the age-specific
       lung tumor incidence in both species after diesel exposure.
              However, because of the long latency period for lung tumor development (rat »12
       mo) relative to the exposure duration, it may not be possible to use ratios of age specific
       lung tumor incidence between  rat and man.  Tentatively, we propose therefore not to
10     include the age-related factors but rather to express the results as relative tumor prevalence
       (relative to base line tumor rate at a specific age), which may not require further application
       of an extrapolation factor.   However, an adjustment has to be made for differences in
       numbers of exposed target cells between rat and man.  If we assume that epithelial cells are
       roughly of equal size in rat and man (Crapo et al., 1983), then a ratio of epithelial cell
15     numbers for man/rat  can be derived from the size of the epithelial surface areas.  Lung
       models by Yeh and Schum (1980) and Sebum and Yeh (1979) can  be used  to calculate
       surface areas for both species. Wojciak (1988) calculated these areas, taking into account
       overlapping areas at bifurcations, and arrived at ratios for man/rat of 125 for alveolar surface
       areas and 110 for tracheobronchial surface areas. If only the numbers of Type II cells, which
20     are probably one target cell type in rats are considered, then a factor of 450 would apply
       (Crapo et al., 1983).  Since the tumor sites may be different for rats and man - involving
       bronchial, bronchiolar, and alveolar cells - we will use the factor of 125 derived above. This
       will be too low if Type II cells are considered, but may be too high  if bronchial cells are
       involved.   Information on tumor  types  from the epidemiological diesel  studies is not
25     available; if significant differences exist between tumor types in exposed rat and man, this
       model, as well as other models, needs further refinement
              Assuming that the model is applicable, the extrapolation involves the following steps
       (using results of the inhalation study by Maudcrly et al. (1987):
       May, 1990                               D-38    DRAFT - DO NOT QUOTE OR CITE

-------
       1.  Calculation of deposition rates of inhaled diesel exhaust per epithelial surface area
          (tracheobronchial and alveolar) for the different groups of the rat study.
       2.  Calculation of EHEs from the respective dose rates according to Figure D-19.
 5
       3.  Determination of dose rate - response (tumor prevalence) relationships as a function of
          exposure time for the rat study (three-dimensional description).
       4.  Determination  of EHE - tumor prevalence relationships for human  diesel exhaust
10        exposure as a function of time adjusted for difference in cell numbers between rat and
          man (i.e., multiplication of normalized tumor prevalence in the rat by 125).

       Figure D-22 shows the result of step 3 above, the normalized tumor prevalence, P/P(0), as
       a function of the  epithelial surface dose rates r* and of exposure time / for the rat exposed
15     7 h/d, 5 d/week for 30 mo.
             The respective functions for P derived from results of the rat study, separately for the
                                               I,*
       alveolar (A) and tracheobronchial (TB) regions, are as follows:
             JA(M"A) = l/[l+exp (15.06-0.0115/-452.65rA)]  (r = 0.980) , and   (5)
20
                          l/[l+exp (15.06-0.0115MO.7r-n,)]                   (6)
       (derived by fitting the tumor data of diesel exposed rats; compares well with the data point
       for controls) where r is in days; r A and r'^ = surface dose rate of organics in alveolar and
25     tracheobronchial region in jig-cm'2*!"1,
       PA and PTB = tumor prevalence in alveolar and bronchial region assuming:
                                 12 percent of inhaled diesel soot is
                                 organics, rat tidal volume = 1.55 cm3,
                                 respiratory frequency =  97 min"1,
30                               body weight = 250 g,
                                 exposure for 24 th/d, 7 d/week,
                                 particle size = 0.25 urn with GSD of 4.4.
       May, 1990                              D-39     DRAFT - DO NOT QUOTE OR CITE

-------
 il
Figure D-22. Result of three dimensional model of normalized lung tumor prevalence in
rats considering delivered dose rate (or inhaled soot concentration) and exposure time
separately.  Exposure is for 7 h/d, 5 d/week.
May, 1990
]>40    DRAFT - DO NOT QUOTE OR CTTE

-------
           The respective equations for the exposure concentrations'c are:

                                  exp (15.06-0.0115M).336c)] = P
 5    where t is in days, is c in mg/m3.  Table D-4 shows the resulting predictions for normalized
      lung tumor prevalence as a function of time and exposure concentration in the rat This
      table shows that the  normalized lung tumor prevalence after diesel exposure in rats is
      constant with time, which would indicate an effect of diesel exhaust on the transition

10
                                      TABLE EM
        EXTRAPOLATED LUNG TUMOR PREVALENCE FOR RATS CHRONICALLY
          EXPOSED TO DIESEL EXHAUST AT DIFFERENT CONCENTRATIONS.
15
20
25
30
35     rates of cells in the cancer model by Moolgavkar and Knudson (1981). We have to keep
       in mind, however, that only very limited data were available to express tumor incidence over
       time in control rats as a function of time.
       The extrapolated lung cancer incidence in humans according to step 4 is shown in Table D-
       5. This assumes lifelong exposure at the exposure conditions of the rat study. Lung tumor
40     incidence is considerably higher than predicted from results of the epidemiological studies.
Exposure concentration, mg/m3
Time/years
05
1
2
3
0
1
1
1
1
aoi
1.003
1.003
1.003
1.003
0.1
1.034
1.034
1.034
1.031
05
1.18
1.18
1.18
1.17
1
1.4
1.4
1.4
136
5
5.4
5.4
5.4
5.0
10
29
29
29
28
       May, 1990                            D-41     DRAFT - DO NOT QUOTE OR CTTE

-------
                                       TABLE D-5

       EXTRAPOLATED NORMALIZED LUNG TUMOR PREVALENCE FOR HUMANS
           CHRONICALLY EXPOSED TO DIESEL EXHAUST 7 H/D, 5 D/WEEK AT
 5                           DIFFERENT CONCENTRATIONS,
10
15
  Exposure concentration
       mg/Qi

  Normalized lung tumor
  prevalence, P/P(0)
0.01
                                         1375
0.1
            4375
0.5
          77 ?S
1.0    5.0
                    46
                 626
20
25
30
35
D.23. Extrapolation Based on Competing Risks

      Another possible approach uses the concept of competing risks and eliminates the
necessity of having to make assumptions on differences in target cell numbers and cell
proliferation rates between species. Briefly, this approach includes:
1.  including separate time functions for diesel exposed and for exposed and for unexposed
   individuals since the approach used in the present model gave unsatisfactory results for
   high exposure rates at short exposure times (most obvious for exposure time / = 0), see
   Fig. 22,

2.  describing age-specific lung tumor rates for humans according to Doll (1971) to find the
   constants for the respective equations (5) to (7) for humans at zero exposure, and

3.  calculating the dose rates (r)  of the delivered particles (organics) for humans by using
   the deposition model of Yu et al. (1987) for insertion into new equations (5) to (7).
       D3. SUMMARY

40          Two extrapolation models were discussed in this chapter. Each gave very different
       results.  In both models the final extrapolation steps for the prediction of lung tumors are
       probably too simplistic and need consideration of additional aspects.  However, lacking
       results of mechanistic studies about the carcinogenicity of diesel exhaust, it is very difficult
       May, 1990
                                    D-42    DRAFT - DO NOT QUOTE OR CITE

-------
      at this time to refine the models in a reasonable way.  Although the extrapolation models
      and the predicted lung tumor prevalences may be based on too many assumptions, they gave
      two possible important hints.  One is that accumulation of soot particles in the lung even
      during chronic occupational diesel exhaust exposure (concentration 100 to 500 Mg/m3) will
 5    not reach lung burdens that cause significant impairment of alveolar clearance (first model);
      the other is that the normalized lung tumor prevalence in diesel exhaust-exposed rats may
      not depend on exposure duration (second model), thus pointing to a specific mode of action
      if a two-stage cancer model is considered.
             The first model, based on accumulated doses of particles or organic material in the
10    lung, does not consider fully the influence of exposure time.  This is true in particular  for
      the organic compounds since they have a relatively short retention half time that is assumed
       to be the same in humans and rats. This model is probably not applicable for substances
      with short turnover times in the lung, in which case the deposition rates of those substances
       appear to be more appropriate  dose parameters.  It is probably for this reason that
15     extrapolated risks from  the first model were much lower than  those  calculated from
       epidemiological studies.  Presently it is hoped that ongoing studies with pure carbon black
       particles wfll hopefully shed light on the involvement of carbonaceous particles in lung tumor
       induction.
             In the second model, based on the interrelation of competing risks (three-dimensional
20     relationships) and on the two stage cancer model by Moolgavkar and Knudson (1981), the
       final extrapolation steps become quite uncertain as to whether and how to adjust for
       differences between rats and man in cell proliferation rates and transition rates of normal
       and intermediate  cells, for differences in target cells and in tumor sites, and for differences
       in age-specific lung tumors.  These factors could not adequately be considered because of
25     a lack of data, and the preliminary result based on an adjustment for differences in numbers
       of target cells predicts a much higher lung tumor risk than epidemiology does. Successful
       use of this model wfll require further discussion and filling of data gaps.
        May, 1990                              D-43     DRAFT - DO NOT QUOTE OR CITE

-------
        Adamson, L Y. R. (1986) Cellular kinetics of the lung.  In: Witschi, H. P.; Brain, J. D., eds. Handbook of
 5             Experimental Pathology.  VoL 75, New York: Springer-Verlag.  pp. 289-317.

        Bailey, M. R. et aL (1982)  Ann. Occnp. Hyg. 26:273-289.

        Bond, J. A; Son, J. D.; Mitchell, C R; Dutcher, J. S^ Wolff, R. 1C; Mcdellan, R. O. (1986)  Biological
10             fate of inhaled organic compounds associated with paniculate matter. In: Lee, S. D.; Schneider, T.;
               Grant, L. D.; Verker, P. J., eds. Aerosols, Lewis Publisher, pp. 570-591

        BrightweU, J.; Fouillet, X; Cassono-Zoppi, A. -L; Bernstein, D.; Gatz, R^ DuchosaL F.  (1986) Neoplastic
               and functional changes in rodents after chronic inhalation of engine exhaust emissions. In: Ishinishi,
15             N.; Koizumi. A; Mcdellan, R. O.; StOber, W., eds. Carcinogenic and Mutagenic Fffeg»« of Diesel
               Engine Exhaust Amsterdam: Elsevier. pp. 471-485.

        Chan, T.  L.; Lee, P. S.; Hering, W. E (1981)  Deposition and clearance of inhaled diesel exhaust particles
               in the respiratory tract of Fisher rats.  J. AppL ToxicoL 1:77-81
20
        Chang, J. F. C; Cross, E A.; Swenberg, J.A.; Barrow, C S. (1983)  Nasal cavity depositioiUiistopathology,
               and cell proliferation after single or repeated formaldehyde exposures in B6C3F, mice and F-344 rats.
               ToxicoL AppL PharmacoL 68:161-176.

25      Crapo, J. D. et aL (1983) Morphometric characteristics of cells in the alveolar region of mammalian lungs.
               Am. Rev. Respir. Dis. 128:542-546.

        DolLR.  (1971)  The age distribution of cancer Implications for models of carcinogenesis. J.R.Soc.
               Med. (Series A) 134:133-166.
30
        Garshick, E4 Scbenker, M. B.; Munoz,  A; et aL (1987)  A case-control study of lung cancer and diesel
               exhaust exposure in railroad workers.  Am. Rev. Respir. Dis. 135:1242-1248.

        Heinrich, U. et aL (1986)  Comparison of chronic inhalation effects in rodents after long-term exposure to
35             either coal  oven flue gas mixed with pyrolized pitch or diesel engine  exhaust.  In:  Ishinishi, N.;
               Koizumi, A; Mcdellan, R^ Stober, W., eds. Carcinogenic and Mutagenic Effects of Diesel Engine
               Exhaust Amsterdam: Elsevier, pp. 441-457.

        Hesseltine, G. R; Wolff, R. 1C; Hanson, R. L; Mcdellan, R. CX; Mauderly, J. L.  (1985) Comparison
40             of lung burdens of inhaled panicles by rats exposed during day and night J. ToxicoL Environ. Health
               16: 323-329.

        International Commission on Radiological Protection (LARQ. (1989) Monographs on the Evaluation of the
               Carcinogenic Risk of Chemicals to Humans. VoL46. Engine Exhausts and Nitroarenes, Lyon, France,
45             14-21. June 1988 (in press).

        International Agency for Research on Cancer (ICRP). (1979) Publication No. 30 Pan 1. Annals of ICRP,
               2.
        May, 1990                                   D-44     DRAFT - DO NOT QUOTE OR CTTE

-------
        Ishinishi, N. et aL (1986)  Rffecrs of diesei soot on unscheduled DNA synthesis of tracheal epithelium and
               lung tumor formation. In: Ishinishi, N.; Koizumi, A4 Mcdellan, R4 Stober, W., eds. Carcinogenic
               and Mutagenic Effects of Diesel Engine Exhaust. Amsterdam: Elsevier, pp. 213-222.

 5      Kawabata, Y. et aL (1986) Effects of diesei soot on unscheduled DNA synthesis of tracheal epithelium
               and lung tumor  formation.   In:   Ishinishi, N.; Koizumi, A.;  Mcdellan, R^ Stober, W., eds.
               Carcinogenic and Mutagenic Effects of Diesel Engine Exhaust Amsterdam: Elsevier, pp. 213-221

        Krewski, D. et aL (1981) A comparison of statistical methods for low dose extrapolation utilizing time to
10             tumor data. Fundam. AppL ToxicoL 3: 140-160.

        Lee JCP. et aL (1986)  Pulmonary response to impaired lung clearance in rats following excessive TIO, dust
               deposition.  Environ. Res. 41: 144-167.

15      Mauderty, J.L; Jones, R. 1C; Griffith, W. C; Henderson, R. E; Mcdellan, R. O. (1987)  Diesel exhaust is
               a pulmonary carcinogen in rats exposed chronically by inhalation. Fundam. AppL ToxicoL 9: 208-221.

        Moolgavkar, S.H.; Knudson, A.G. (1981)  Mutation and cancer A model for human carcinogens.  J. NatL
               Cancer Inst 66: 1037-1052.
20
        National Committee on Radiation Protection and Measurements (NCRP) (1980)  Influence of dose and its
               distribution in time on dose-response relationships for low LET radiations.  Report No. 64, NCRP
               Publication, Washington D.C

25      OberdOrster, G. (1989) Deposition and retention modeling of inhaled cadmium in rat and human lung:
               And example for extrapolation of effects and risk estimation.  In: Crapo, J.; et aL, eds. Extrapolation
               of Dosimetric Relationships for Inhaled Particles and Gases. New York: Academic Press, in press.

        Raabe, O.G. (1987) Three dimensional models of competing risk and natural life span.  Fundam. AppL
30             ToxicoL 8: 465-473.

        Saffiotti, U. et aL (1988) Multifactorial hamster respiratory carcinogenesis:  Oxide, benzo(a)pyrene and
               Ar-methyl-Af-nitrosourea  (MNU).  Presented  at:   Biology, Toxicology and  Carcinogenesis of
               Respiratory Epithelium, Albuquerque, NM, Nov. 1988.
35
        Sanders. C personal communication. (1989).

        Sebum, M.; Yen, H.G (1979) Theoretical evaluation of aerosol deposition in anatomical models of
               mammalian lung airways.  BulL Math. BioL 42; 1-15.
40
        Strom, K.  (1987) Presentation at HEI-Diesel Symposium, Seabrooke Island, S.C Feb. 1987.
        Sun, JJX; Wolff, R. 1C; Kanapilly, G. M^ Mcdellan, R. O. (1984) Lung retention and metabolic fate of
                inhaled benzo(a)pyrene associated with diesei exhaust particles.  ToxicoL AppL PbarmacoL 73: 48-59.
 45
        Wojtiak, JJ7. (1988) Theoretical and experimental analyses of aerosol deposition in the lung: implications
                for human health effects.  PfauD. Thesis, University of Rochester, Rochester, NY.
         May, 1990                                   D-45     DRAFT - DO NOT QUOTE OR CITE

-------
        Wolff, R. 1C; Henderson, R. E; Snipes, MA; Sun, J. D^ Bond, J.A.; Mitchell, C R; Manderiy, J. L;
               Mcdellan, R. O.; (1986) Long retention of diesel soot and associated organic compounds. In:
               Ishinishi, R; Koizumi, A; McQellan, R. (X; StOber, W., eds. Carcinogenic and Matagenic Effects of
               Diesel Engine Exhaust Amsterdam: Elsevier, pp. 191-211.
 5
        Wolff, RJC; Bond, J. A.; Sun, J. D^ et aL (1989) Effects of adsorption of benzo(a)pyrene onto carbon
               black particles on levels of DNA adducts of rats exposed by inhalation.  ToxicoL AppL PharmacoL
               97:289-299.

10      Yen, KG; Sebum, M. (1980) Models of human lung airways and their application to inhaled particle
               deposition. Bull Math. BioL 42:461-480.

        Yu, CP.: Yoon, KJ. (1988) Determination of lung doses of diesel exhaust paniculates. Presented at 5th
               Annual Conference of Health Effects Institute, Colorado Springs, CO.  April 1988.
15
        Yu, CP. et aL 1987.  A nonlinear model of alveolar clearance of insoluble particles from the lung,
               ToxicoL, premier issue, 97-107.
        May, 1990                                  EM6      DRAFT - DO NOT QUOTE OR OTE

-------