united aiaies
Environmental Protection
Agency
nazaraous waste engineering
Research Laboratory
Cincinnati OH 45268
bKA/bUO/2-86/095
October 1986
Research and Development
Technical Resource
Document:
Treatment
Technologies for
Solvent Containing
Wastes
-------
EPA/600/2-86/095
October 1986
TECHNICAL RESOURCE DOCUMENT
TREATMENT TECHNOLOGIES FOR SOLVENT CONTAINING WASTES
by
Marc Breton Mark Arienti
Paul FrilUci Michael Kravett
Stephen Palmer Andrew Shayer
Clay Spears Norman Suprenant
Alliance Technologies Corporation
Bedford, MA 01730
EPA Contract Number 68-03-3243
Project Officer
Harry M. Freeman
Alternative Technologies Division
Hazardous Waste Engineering Research Laboratory
Cincinnati, OH. 45268
HAZARDOUS WASTE ENGINEERING RESEARCH LABORATORY
OFFICE OF RESEARCH & DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
NOTICE
The information in this document has been funded wholly or in part
by the United States Environmental Protection Agency under Contract No.
68-03-3243 to Alliance Technologies. It has been subject to the Agency's
peer and administrative review and has been approved for publication.
Mention of trade names or commercial products does not constitute endorse-
ment or recommenodation for use.
-------
ACKNOWLEDGEMENT
The authors would like to thank Harry M. Freeman, the Hazardous Haste
Engineering Research Laboratory Work Assignment Manager, whose assistance and
support was utilized throughout the program. The authors also extend thanks
to other members of the HWERL staff for their assistance and to the many
industrial representatives who provided design, operating, and performance
data for the waste treatment technologies.
111
-------
CONTENTS
Acknowledgement. ......... ......... iii
Figures vi
Tables
Project Summary. 1
1. Introduction 1-1
2. Background 2-1
Schedule for Land Disposal Prohibition .......... 2-1
Treatment Standards for Certain Solvents . . . . . .. . . 2-3
3. Current Industrial Sources, Generation, and Management
Activities for Solvent Wastes .... ....... 3-1
Industrial Sources of Waste Solvents ........... 3-1
Solvent Waste Generation 3-22
Waste Management Practices 3-30
4. Commercial Offsite Recycling, Treatment, and Disposal
Capacity ....... ... 4-1
Commercial Solvent Recycling Industry .... 4-1
Available Recycling Treatment and Disposal Capacity ... 4-9
5. Waste Minimization Processes and Practices ... 5-1
Source Reduction 5-1
Recycling/Reuse ..... ..... 5-11
Additional Examples of Waste Minimization Practices ... 5-31
6. Pretreatment 6-1
Liquids Containing Suspended or Floating Materials .... 6-3
Solid/Sludge Wastes Containing Liquids ....* 6-5
Inorganic Solid Wastes Containing Low Levels of Organics . 6-5
Bulky, Non-Uniform Solid Wastes „ 6-7
Low Heat Content or Low Viscosity Liquids ......... 6-8
7. Physical Treatment Processes ...... 7-1
Distillation 7-3
Evaporation Processes ...... 7-48
Steam Stripping 7-88
Air Stripping 7-134
Liquid - Liquid Extraction 7-153
Carbon Adsorption 7-184
Resin Adsorption 7-223
8. Chemical Treatment Processes 8-1
Wet Air Oxidation , 8-2
Supercritical Fluid Oxidation . 8-31
Ozonation 8-47
Other Chemical Oxidation Processes ..... . 8-63
Chlorinolysis Processes .. ..... 8—75
Chemical Dechlorination •• ..... 8—86
iv
-------
CONTENTS (continued)
9. Biological Methods 9-1
Process Description ...... 9-1
Demonstrated Performance .... 9-8
Cost of Treatment 9-10
Overall Status of Biological Treatment 9-20
10. Incineration Processes ....... 10-1
Overview ....... 10-1
Process Description ..... ..... 10-20
Hazardous Waste Incinerator Performance . 10-36
Costs of Hazardous Waste Incineration .......... 10-55
Status of Development .................. 10-67
11. Emerging Thermal Treatment Technologies 11-1
Circulating Bed Combustion 11-2
Molten Glass Incineration .... .. 11-11
Molten Salt Destruction 11-20
Pyrolysis Processes 11-28
In Situ Vitrification . 11-44
12. Use as a Fuel 12-1
Process Descriptions ...... 12-6
Demonstrated Performance .. ....... 12-13
Cost of Treatment 12-28
Status of Development 12-31
13. Land Disposal of Residuals 13-1
Solidification/Chemical Fixation .... 13-2
Macroencapsulation ........ ....... 13-7
14. Considerations for System Selection .............. 14—1
General Approach 14-3
Assessment of Alternatives ......... 14-4
Appendices
A Properties of Organic Solvents A-l
B Solvent Distillation Equipment ^ B-l
-------
FIGURES
Number Page
4.1,1 Solvent wastes recycled by commercial reproeessors
in Illinois .. ....... 4-5
6.1 Process train for treating aqueous/organic leachate 6-2
6.2 Process train for treating waste solvent ... 6-4
6.3 Alternatives for incinerating contaminated solid wastes .... 6—6
7.1.1 Basic schematic for batch and continuous fractionation
systems 7-5
7.1.2 McCabe-Thiele diagram for distillation . . . 7-6
7.1.3 Reclamation of cold cleaning solvents via small batch
stills (15 gpd) 7-38
7.1.4 Reclamation of cold cleaning solvents via medium batch
stills (55 gpd) 7-39
7.1.5 Reclmation of cold cleaning solvents via a large
continuous still (250 gpd) .. .......... 7-40
7.1.6 Reclamation of vapor degreasing solvents (chlorinated
hydrocarbons, $4.50/gallon) ................. 7-41
7.1.7 Reclamation of waste paint thinner ($2.00/gallon) ....... 7-43
7.1.8 Reclamation of waste paint thinner ($4.50/gallon) ....... 7-44
7.2.1 Cross section of agitated thin film evaporator ........ 7-49
7.2.2 Treatment train using an agitated thin film evaporator .... 7-52
7.2.3 Selection of Luwa evaporators based on waste viscosity .... 7-54
7.2.4 Required heat transfer surface area for distilling low
boiling organics and concentrating aqueous solutions .... 7-56
VI
-------
FIGURES (continued)
Number Page
7.2.5 Required heat transfer surface area for dehydrating heavy
pastes and stripping wastes to low residual organics .... 7-57
7.2.6 Heat transfer and evaporation rates in Luwa thin film
evaporators ...... . 7-58
7.2.7 Threshold costs for onsite ATFE versus feed solids
concentration 7-80
7.2.8 Threshold cost vs feed solids concentration for solvent
recovery using ATFE 7-81
7.3.1 Typical steam stripping process 7-89
7.4.1 Air stripping towers 7-137
7.4.2 Air stripping schematic 7-139
7.5.1 Schematic of extraction process 7-154
7.6.1 Carbon adsorption flow diagram 7-192
7.6.2 Carbon Bed Configurations . 7-198
7.6.3 Schematic of biological/carbon adsorption treatment train ... 7-203
7.6.4 Schematic of carbon adsorption/biological treatment train ... 7-205
7.6.5 Schematic of biophysical treatment train ........... 7-207
7.6.6 Schematic of stripping/carbon adsorption treatment train ... 7-208
7.7.1 Phenol removal and recovery system - solvent regenration
of Amberlite adsorbent 7-230
7.7.2 Performance of resin adsorption bed 7-241
8.1.1 Wet air oxidation general flow diagram 8-3
8.1.2 4.5 MGD wastewater treatment facility 8-22
8.1.3 Installed plant costs versus capacity . 8-23
8.1.4 Unit operating costs versus unit flow rate 8-25
8.2.1 Temperature-density diagram 8-33
-------
FIGURES (continued)
Number Page
8.2.2 Properties of water at 250 atm 8-35
8.2.3 Schematic flow sheet of MODAR process 8-37
8.3.1 Schematic of top view of ULTROX pilot plant by
General Electric (Ozone sparging system omitted) ...... 8-51
8.4.1 Example process flowsheet - oxidation .. .. 8-66
8.4,2 Effect of ^02 alone, UV alone, and H2C>2 plus
UV on decomposition of trichloroethylene (TCE) at 20°C,
pH 6.8 8-70
8.4.3 Rates of reaction of halogenated aliphatics at 20°C and 30PC . 8-71
8.5.1 Schematic diagram of the Hoescht AG chlorolysis process .... 8-79
8.6.1 Soil cleaning process schematic ....... 8-88
8.6.2 Probable reaction mechanism 8-90
9.1 Flowsheet and plot of oxygen demand and oxygen supply
versus tank length for (a) conventional, (b) complete-mix,
and (c) step-aeration activated-sludge processes ...... 9-2
9.2 Estimated annual operating and materials costs as a function
of wastewater treatment facility capacity . . . . « .... . 9-18
9.3 Estimated annual man-hours needed for wastewater treatment
facility operation . . y-19
10.1 Pretreatment option logical decision flow chart 10-19
10.2 Flow sheet of an incineration plant for hazardous wastes ... 10-23
10.3 Rotary kiln incinerator with liquid injection capability ... 10-29
10.4 Cross-section of a fluidized-bed furnace ...... 10-33
10.5 Purchase cost of liquid injection system (May 1982) ...... 10-62
10.6 Purchase cost of rotary kiln system (Hay 1982) 10-62
10.7 Purchase cost of hearth incinerators (mid 1982) ........ 10-63
10.8 Purchase cost of waste heat boilers (July 1982) 10-63
Vlll
-------
FIGURES (continued)
Number Page
10.9 Purchase cost of scrubbing systems receiving 500 to
550°F gas (July 1982) .................... 10-64
10.10 Purchasse costs for typical hazardous waste incinerator
scrubbing systems receiving 1800 to 2200°F gas (July 1982), . 10-64
11.1.1 CBC incineration pilot plant located at GA Technologies .... 11-3
11.1.2 Chemical reactions that occur in CBC combustion chamber .... 11-5
11.2.1 Dirt purifier and hazardous waste incinerator ......... 11-16
11*3.1 Molten salt combustion system 11-21
11.4.1 Continuous pyrolizer 11-29
11.4.2 Pyroplasma process flow diagram 11-34
11.4.3 Advanced electric reactor [Huber] ..... . 11-38
11.4.4 High temperature fluid wall process configuration for the
destruction of carbon tetrachloride ............. 11-40
11.5.1 Operating sequence of in situ vitrification . ... 11-45
11.5.2 Off-gas containment and electrode support hood 11-46
11.5.3 Cost of in situ vitrfication for TRU wastes as functions
of electrical rates and soil moisture [Fitzpatrick, 1984] . . 11-48
14.1 Solvent waste management options ............... 14-2
14.2 Simplified decision chart for aqueous and mixed aqueous/
organic solvent waste stream treatment ..... 14-9
14.3 Simplified decision chart for organic liquid solvent
waste stream treatment ..... ..... 14-10
14.4 Approximate ranges of applicability of VOC removal
techniques as a function of organic concentration
in liquid waste streams ..........«....*... 14-13
IX
-------
TABLES
Number Page
2.1 . Scheduling for Promulgation of Regulations Banning
Land Disposal of Specified Hazardous Wastes ........ 2-2
2.2 Performance Achieved by Treatment Technologies (ing/liter) . . 2-6
3.1.1 Consumption of Major Solvents (10° Ibs/year) 3-2
3.1,2 Priority Solvent Industrial End Uses 3-4
3.1.3 Top 20 Industries Generating Solvent Wastes 3-5
3.1.4 Hazardous Solvent Waste Generation in the Paint and
Coatings Manufacturing and Factory Applied
Coatings Industry .......... . 3-7
3.1.5 Waste Solvent Generation by Type of Degreasing Operation ... 3-9
3,2.1 Summary of Waste Generation and Management ..... 3-23
3.2.2 Physical Form of Waste Ignitables and Solvents Managed
at TSD Facilities (gpy) 3-26
3.2.3 Relative Rankings of Solvent Waste Constituents ....... 3-28
3.2.4 Numbers of Small Quantity Generators and Waste Quantity
by Waste Stream ........ . 3-31
3.3.1 Characteristics of RCRA Solvent Wastes by Waste
Management Practice ...... 3-32
3.3.2 Waste Recycling Reported in the Revised National Survey ... 3-34
3.3.3 Relative Ranking by Degree of Recycling ... 3-36
3,3.4 Distribution of Offsite Recycled Waste Solvents by
Previous Use (%) . . 3-37
3.3.5 Characteristics of Waste Solvents Recovered Offsite ..... 3-38
-------
TABLES (continued)
Number
3.3.6 Onsite Recycling Practices for Solvent and Ignitable Wastes . 3-40
3.3.7 Recovery Practices for RCRA Hazardous Wastes by
Residual Category . 3-41
3.3.8 Treatment of Solvent and Ignitable Wastes by Incineration . . 3-42
3.3.9 Physical Forms of Incinerated Solvents and Ignitables .... 3-44
3.3.10 Priority Solvents Incinerated in 1981 3-45
3.3.11 Most Frequently Practiced Treatment Techniques in
Tanks for Ignitable Wastes 3-46
3.3.12 Quantities of Solvent Waste Streams Currently Managed
by Land Disposal 3-48
4.1.1 Summary of Comercial Solvent Recycling Surveys . 4-3
4.1.2 Cost for Solvent Recovery at Commercial Facilities ...... 4-7
4.1.3 Waste Solvent Transportation Costs .. 4-8
4.2.1 Solvent Waste Quantity Requiring Alternative Treatment (HGY) . 4-9
4.2.2 Annual Treatment and Recovery Capacity Demand and
Availability (million gallons/year) . 4-11
5.1 Summary of Waste Reduction Cases 5-6
5.2 Summary of Documented Solvent Recycling Practices ...... 5-13
5.3 Summary of Solvents Recycled via Two Waste Exchanges ..... 5-29
5.4 Examples of Waste Reduction Techniques/Accomplishments
(All Programs are Onsite) 5-32
5.5 Cost vs Savings for Waste Reduction Programs Carried
Out by IBM Corporation at Various Locations 5-36
7.1.1 Boiling Points and Autoignition Points of Risk Solvents ... 7-9
7.1.2 Azeotrope Formation of Compounds with Water 7-12
7.1.3 Sample Data on Azeotrope Formation .............. 7-13
7.1.4 Relative Performance Ratings of Trays and Packings ...... 7-18
xi
-------
TABLES (continued)
Number Page
7.1.5 Selection Guide for Tower Internals 7-18
7.1.6 Small Still Performance Evaluation . . . 7-20
7.1.7 Summary: GCA/OSW Case Study ........... 7-23
7.1.8 Methyl Chloroform/Freon Batch Distillation System
Operating Information 7-24
7.1.9 Distillation Column Results ................. 7-24
7.1,10 Projected Economic Benefits ... 7-27
7.1.11 Unit Costs . . 7-29
7.1.12 Estimated Cost of Recovery of Freon/Methyl Chloroform .... 7-30
7.1.13 Capital Costs for Onsite Solvent Recovery Systems
(1982 Dollars) 7-33
«
7.1.14 Threshold Costs ($ per Gallon) for Onsite Recovery
Systems for Varying Solids Contents and Percent Net
Discounted Cash Flows (DCF) 7-34
7.1.15 Commercially Available Solvent Stills ... . . 7-45
7.2.1 Keys to Selecting Kontro Thin Film Evaporators 7-53
7.2.2 Typical Agitated Thin Film Evaporator Design Characteristics . 7-60
7.2.3 Process Data for Luwa Thin Film Evaporator Runs ....... 7-62
7.2.4 Percent by Weight of Principal Organic Components ...... 7-63
7.2.5 Plant A: Thin Film Evaporator Waste Compositions and
Headspace Analysis ........ ... 7-67
7.2.6 Analysis of Liquid Samples, Thin Film Evaporator, Plant B . . 7-69
7.2.7 Analysis of Product Samples, Thin Film Evaporator, Plant B . . 7-70
7.2.8 Analysis of Liquid Samples, Thin Film Evaporator, Plant C . . 7-71
7.2.9 Hourly Costs of Luwa Thin Film Evaporator 7-73
7.2.10 Processing Costs During Three Test Runs of the Luwa
Thin Film Evaporator 7-75
xii
-------
TABLES (continued)
Number Page
7.2.11 1984 Plant A Capital and Operating Cost 7-76
7.2.12 Capital Cost Recovery Components for Onsite ATFE
Recovery Systems 7-77
7.2,13 Threshold Costs for Onsite ATFE Recovery Systems for
Varying Solids Contents and Percent Net Discounted
Cash Flow (DCF) 7-78
7.2.14 Threshold Costs for Onsite Systems Versus Changes in
Capacity Factor (Recovered Solvent) 7-82
7.3.1 Process Data for the Steam Distillation Unit 7-99
7.3.2 Percent by Weight of 1,1,1-Trichloroethane in the Waste
Streams During Operation of the Steam Distillation Unit . . 7-100
7,3.3 Metal Analyses for the Steam Distillation Unit ........ 7-101
7.3.4 Plant D: Characterization of Wastes ........ 7-104
7.3.5 Plant "D" Batch Steam Stripper" Process Data 7-105
7.3.6 Industrial Steam Stripper Survey ............... 7—108
7.3.7 Full-Scale Industrial Stipper Performance Summary ...... 7-110
7.3.8 Processing Data for Bench-Scale Stream Stripping
of Stream 221 7-113
7.3,9 Summary of Removal Efficiency Data for Stream 221A ...... 7-114
7.3.10 Summary of Processing Data for Bench-Scale Steam
Stripping of Ethylene Bichloride .... 7-116
7.3.11 Test Conditions and TOG Removals for Bench-Scale Steam
Stripping With/Without Activated Carbon . . 7-118
7.3.12 Results of Bench-Scale Steam Stripping Runs Performed
With/Without Lime Pretreatment 7-119
7.3.13 Test Results From Carbon Adsorption of Off-Gas
From Steam Stripping - Run 6............... 7-121
7.3.14 Steam Stripping Costs for Wastewater Streams Containing
Contaminants of Varying Henry's Law Constant ....... 7-125
xxii.
-------
TABLES (continued)
Number Page
7.3.15 Cost Components for Onsite Steam Stripping Solvent
Recovery: (Example Case with 30% Solvent Content
and Bottoms Used as Fuel) 7-127
7.3.16 Steam Stripping Cost Estimates as a Function of
Throughput, Solvent Content and Disposal Method ..... 7-129
7.4.1 Compounds of Interest with Henry's Law Constant
Greater Than 1 x. 10~3 atm-nrVmol 7-140
7.4.2 Results of Pilot-Scale Air Stripping Study 7-146
7.4.3 Results of Bench-Scale Air Stripping Study . . . i 7-148
7,4.4 Results of Pilot-Scale Mr Stripping Treatability Study . . 7-149
7.5.1 K« Values for Aqueous/Solvent Systems 7-159
7,5.2 Experaimentally Measured Equilibrium Distribution
Coefficients for Extraction From Water Into
Undeeane, (295-300 K) and 1-Octanol . . . , .• . 7-161
7.5.3 Measured Values of Kp for Extraction of Acrolein From
Water Into Various Solvents 7-162
7.5,4 Measured Equilibrium Distribution Coefficients (Kjj)
at 30°C for Extraction of Nitrobenzene From Water
Into Various Solvents ...... . 7-164
7.5.5 Variable Measured in Spray Column Extractor Study ..... 7-167
7.5.6 Results of Spray Column Extractor Runs ........... 7-168
7.5.7 Results of RDC Extractor Runs 7-170
7.5.8 Conditions and Results for Mini-Plant Extraction Run .... 7-171
7.5.9 Results of Solvent Extraction Studies ..... 7-173
7,5.10 Estimated Costs for a Liquid-Liquid Extraction System .... 7-174
7.5.11 Preliminary Extraction for Nitrobenzene Extraction Using
Di-Isobutyl Ketone (DIBK) 7-176
7.5.12 Breakdown of Operating Costs for Extraction of Acrolein ... , 7-177
xiv
-------
TABLES (continued)
Number Page
7.5.13 Estimation of Operating Cost: Extraction of Acrolein
by MIBK 7-178
7.5.14 Advantages and Disadvantages of Extraction Types 7-180
7.6,1 Haste Characteritics That Affect Adsorption by
Activated Carbon 7-186
7.6.2 Influence of Substituent Groups on Adsorbability ....... 7-187
7.6.3 Carbon Adsorption Isotherm Data 7-189
7.6.4 Adsorption of Solvents and Other Organics by
Activated Carbon 7-190
7.6,5 Properties of Several Commercially Available Carbons ..... 7-196
7.6.6 Typical Properties of Powdered Activated Carbon
(Petroleum Base) ..... 7-197
7.6.7 Data Reported From Bench-Scale and Pilt-Scale GAC Systems . . 7-211
7.6.8 Data Reported From Full-Scale GAC Systems 7-212
7.6.9 Data Reported from Full-Scale PAG System 7-213
7.6.10 Direct Costs for Carbon Adsorption 7-215
7.6.11 Indirect Costs for Carbon Adsorption ...... . 7-216
7.6.12 Carbon Adsorption Costs 7-218
7.6.13 Removal Ratings for Organic Compounds ............ 7-220
7.7.1 Physical Properties of Adsorbents 7-225
7.7.2 Typical Physical Properties of Ambersorb XE-340 7-226
7.7.3 Chemical Compounds Treated by Resin Adsorption 7-233
7.7.4 Adsorption of Solvents by Ambersorb XE-340 ..... 7-236
7.7.5 Removal of Chlorinated Solvents by Ambersorb XE-340 ..... 7-238
7.7.6 Pilot Plant Adsorption Summary . 7-239
xv
-------
TABLES (continued)
Number Page
7.7.7 Cost of Adsorbents . . 7-242
7.7.8 Design Criteria—Trihaloraethane Removal 7-243
7.7.9 Cost Comparison—GAG vs. Resin 7-244
8.1.1 One Hour Wet Oxidation 8-11
8.1.2 Products Identified From One Hour Oxidations at 320°C .... 8-11
8.1.3 Bench-Scale Wet Mr Oxidation of Pure Compounds 8-12
8.1.4 Bench-Scale Wet Air Oxidation of Organics in Wastewaters . . . 8-14
8*1.5 Bench-Scale Wet Air Oxidation of Wastewaters 8-16
8.1.6 Pilot-Scale Wet Air Oxidation of Organic Compounds in
Industrial Waste water . . 8-18
8.1.7 Priority Pollutant Removals Using a PACT™/Wet Mr
Regeneration System for Domestic and Organic
Chemicals Waste water . . 8-21
8.1.8 WAO Costs Versus Flow 8-24
8.2.1 Dielectric Constants of Some Common Solvents 8-32
8.2.2 MODAR Treatment Costs for Organic Contaminated Aqueous Wastes 8-43
8.3.1 Relative Oxidation Power of Oxidizing Species ... 8-48
8.3.2 Design Data for a 40,000 gpd (151,400 L/day) ULTROX Plant . . 8-53
8.3.3 Ozonation Treatment of Solvents and Ignitables 8-55
8.3.4 Equipment Plus Operating and Maintenance Costs:
40,000 gpd OT/Ozone Plant 8-58
8.4.1 Relative Oxidation Power of Oxidizing Species ... 8-64
8,5.1 Typical Chlorinolysis Reactions .. ........ 8-76
8.5.2 Operating Characteristics of Chloronolysis Processes 8-77
xvi
-------
TABLES (continued)
Number Page
8.5.3 Annual Operating Cost for Processing 25,000 Metric Tons/
Year of a Mixed Vinyl Chloride Monomer and Solvent Waste
at a Chlorolysis Plant 8-83
8.6.1 Dechlorination Processes 8-89
9.1 Typical Operating Parameters for Biological Treatment
Processes « 9-9
9.2 Removal Data for Biological Processes ............ 9-11
9.3 Average Performance of Full-Scale Biological Treatment
Facilities for Solvents of Concern (mg/L) ... 9-16
9.4 Estimated Capital Cost for Wastewater Treatment Units .... 9-19
10.1 Solvent Hazardous Waste Heats of Combustion .... 10-11
10.2 Characteristic Parameters for Several Solvent
Hazardous Wastes ..... ............ 10-13
10.3 Waste Characteristic Data for Several Solvent
Hazardous Waste Streams . 10-15
10.4 Operating Parameters of Hazardous Waste Liquid
Injection Incinerators 10-26
10.5 Operating Parameters for Rotary Kilns 10-31
10.6 Operating Parameters for Fluidized-Bed Incinerators 10-35
10.7 Solvent Constituents Present in Incinerated Waste Streams . . 10-37
10.8 Incineration Facilities Tested ........ 10-38
10.9 Summary of Results of Incinerator Test Programs 10-41
10.10 PICs Found in Stack Effluents 10-44
10.11 Residuals Analysis at Four Full-Scale Incinerators 10-46
10.12 Performance Test Synopsis 10-47
10.13 Performance Test Synopsis ....... 10-49
10.14 Performance Test Synopsis ..... ........ 10-51
xvi i
-------
TABLES (continued)
Number Page
10.15 Fluidized-Bed Incinerator Performance Test Results ...... 10-53
10.16 Survey of Hazardous Waste Incinerators - Costs of
Incineration and Cost Impacting Factors 10-57
10.17 Elements of Capital Cost for Incineration Systems ...... 10-59
10.18 Summary of Cost Data Compiled by MITRE Corporation, 1981 . . . 10-65
10.19 Number of Hazardous Waste Incinerators in Service
in the U.S.A 10-69
10.20 Summary of Incineration Technologies 10-80
11.1.1 Capacities of CBC Systems 11-7
11.1.2 Summary of Bag Sample Results .... ...;.. 11-8
11.1.3 Test Results on Hazardous Wastes Circulating Bed
Incinerator Pilot Plant ........ 11-10
11.1.4 Circulating Bed Combustion Units . 11-12
11.1.5 Circulating Bed Incinerator vs. Conventional Incinerators . . 11-14
11.3.1 PCB Combustion Tests in Sodium-Potassium-Chloride-
Carbonate Melts [Edwards, 1983] 11-25
11.3.2 Summary of Pilot-Scale Results .... 11-26
11.4.1 Carbon Tetrachloride Test Results .... 11-36
11*4.2 Summary of Operating Parameters and Results for
Huber AER Research/Trial Burns 11-43
12.1 Thermal Technologies Considered Appropriate for Burning
Hazardous Waste as Fuel 12-2
12.2 Representative Heating Values of Virgin and Spent Solvents . . 12-4
12.3 Operating Specifications: Systech Cement Kiln Process .... 12-11
12.4 Operating Specifications of Commercial Aggregate Kilns
Burning Wastes 12-12
12.5 Boiler Summary 12-15
XVlll
-------
TABLES (continued)
Number
12.6
12.7
12.8
12.9
12.10
13.1
13.2
13.3
13.4
13.5
14.1
14,2
14.3
14.4
14.5
Summary of DREs for Volatile POHC
Summary of POHC DREs, Percent — Site G
Results of Engineering-Science Industrial Boiler
Calculated Destruction and Removal Efficiencies (Percent) . .
Compatibility of Selected Waste Categories with Different
Present and Projected Economic Considerations for
Encapsulated Waste Evaluated at the U.S. Army Waterways
Guideline Considerations for the Investigation of
Treatment Processes Potentially Applicable -to
Page
12-18
12-20
12-21
12-23
12-25
13-3
13-4
13-7
13-9
13-9
14-6
14-11
14-14
14-18
14-23
XIX
-------
-------
SECTION 1.0
INTRODUCTION
This Technical Resource Document provides information that can be used by
environmental regulatory agencies and others as a source of technical
information for waste management options for solvent wastes* and many other
wastes containing low molecular weight organic compounds. These options
include waste minimization, recycling, and treatment of waste streams.
Emphasis has been placed on the collection and interpretation of performance
data for proven technologies. These are:
- incineration
— use as a fuel
- distillation
- steam stripping
— biological treatment
- activated carbon adsorption
These, and other potentially viable technologies, are described in terras of
their actual performance in removing constituents of concern, their associated
process residuals and emissions, and those restrictive waste characteristics
which impact the ability of a technology to effectively treat the wastes under
consideration. Although emphasis is placed on performance data, cost and
capacity data are provided to assist the user of this document in assessing
the applicability of technologies to specific solvent wastes. References are
cited throughout to identify additional sources of background information for
the user.
*Solvent wastes are those in RCRA Codes F001 through F005 plus those U and P
type wastes containing the solvent constituents identified in the F001
through F005 Codes.
1-1
-------
This document provides, in following sections, first a review of
regulatory background (Section 2.0), and a review of the current hazardous
waste management data base regarding sources of wastes and existing management
practices (Sections 3.0 and 4.0). This is followed by information concerning
waste minimization practices (Section 5.0) and an evaluation of the full range
of treatment/recovery processes (Section 6.0 through 13.0). In order of tneir
presentation, the latter include:
6.0 Pretreatraent
7.0 Physical Treatment Processes
8,0 Chemical Treatment Processes
9.0 Biological Methods
10.0 Incineration Processes
11.0 Emerging Thermal Treatment Technologies
12.0 Use as a Fuel
13.0 Land Disposal of Residuals
These technologies are examined with emphasis placed on identifying process
design and operating factors and waste characteristics which affect treatment/
recovery of solvent wastes. Cost data are also presented to assist the user
in evaluating and selecting options. Approaches to the selection of treatment/
recovery options are reviewed in the final section of this document
(Section 14.0). Pertinent properties of the compounds addressed in this
document which impact treatment technology/waste interactions are provided in
Appendix A.
1-2
-------
SECTION 2.0
BACKGROUND
Section 3004 of the Resource Conservation and Recovery Act (RCRA), as
amended by the Hazardous and Solid Waste Amendments of 1984 (HSWA), prohibits
the continued placement of RCRA—regulated hazardous wastes in or on the land,
including placement in landfills, land treatment areas, waste piles, and
surface impoundments (with certain exceptions for surface impoundments used
for the treatment of hazardous wastes). The amendments specify dates by which
these prohibitions are to take effect for specific hazardous wastes. After
the effective date of a prohibition, wastes may only be land disposed if:
(1) they comply with treatment standards promulgated by the Agency that
minimize short term and long term threats arising from land disposal or
(2) the Agency has approved a site-specific petition demonstrating, to a
reasonable degree of certainty, that there, will be no migration from the
disposal unit for as long as the waste remains hazardous. In addition, the
statute authorizes the Agency to extend the effective dates of prohibitions
for up to 2 years nationwide if it is determined that there is insufficient
alternative treatment, recovery or disposal capacity.
2.1 SCHEDULE FOR LAND DISPOSAL PROHIBITION
The amendments call for the banning of the land disposal of solvents
within 24 months of the 8 November 1984 enactment of the amendments. EPA must
also determine whether to ban the land disposal of the "California List" and
other listed wastes according to schedule shown in Table 2.1.
-------
TABLE 2.1. SCHEDULING FOR PROMULGATION OF REGULATIONS BANNING
LAND DISPOSAL OF SPECIFIED HAZARDOUS WASTES
Waste category
Effective date*
Dioxin containing waste
Solvent containing hazardous wastes
numbered 1001, F002, F)03, F004, F005
California List
-Liquid hazardous wastes, including free liquids
associated with any solid or sludge containing:
- Free or complex cyanides at >J.,000 mg/L
- As X500 mg/L
- Cd >100 mg/L
- Cr4""5 2:500 mg/L
- Pb _>500 mg/L
- Hg >20 mg/L
- Ni 213^ mg/L
- Se _>100 mg/L
- Ti _>130 mg/L
-Liquid hazardous wastes with:
- PH <2.0 )
- PCBs >5Q ppm (
_ j
- Hazardous wastes containing haldgenated organic
compounds in total concentration >jl,QOO mg/kg
Other listed hazardous wastes (§§261.31 and 32), for
which a determination of land disposal prohibition
must be made:
— One—third of wastes
- Two-thirds of wastes
- All wastes
Hazardous wastes identified on.the basis of
characteristics under Section 3001
Hazardous wastes identified or listed after enactment
11/8/86
11/8/86
7/8/87
7/8/87
7/8/87
8/8/88
6/8/89
5/8/90
5/8/90
Within 6 months
*Not including underground injection.
2-2
-------
2.2 TREATMENT STANDARDS FOR CERTAIN SOLVENTS
The Agency has identified treatment standard concentration levels
for certain solvent wastes which must be met before the wastes can be land
disposed. These treatment standards are based on the mobility, toxicity, and
persistence of the waste (as well as the effects of solvents on liners and the
mobilization of other wastes) and the ability of treatment technologies to
remove, destroy, or immobilize the hazardous constituents in the wastes,
2.2.1 Solvents of Concern
Treatment standards were proposed in the 14 January 1986 Federal Register
for the following spent solvents and commercial chemical products,
off-specification commercial chemical products, manufacturing intermediates,
and spill residues:
F001—The following spent halogenated solvents used in degreasing:
tetrachloroethylene, trichloroethylene, methylene chloride,
1,1, l-trichloroethane, carbon tetrachloride, and chlorinated
fluorocarbons; all spent solvent mixtures/blends used in degreasing
containing, before use, a total of 10 percent or more (by volume) of one
or more of the above halogenated solvents or those solvents listed in
F002, F004 and FOOSj and still bottoms from the recovery of these spent
solvents and spent solvent mixtures.
F002—The following spent halogenated solvents: tetrachloroethylene,
methylene chloride, trichloroethylene, I,I,l-trichloroethane, chloro-
benzene, 1,l,2-trichloro-l,2,3-trifluoroethane, ortho-dichlorobenzene,
and trichlorofluoromethanej all spent solvent mixtures/blends containing,
before use, a total of 10 percent or more (by volume) of one or more of
the above halogenated solvents or those solvents listed in FOOL, F004,
, and F005; and still bottoms from the recovery of these spent solvents and
spent solvent mixtures.
F003—The following spent nonhalogenated solvents: xylene, acetone,
ethyl acetate, ethyl benzene, ethyl ether, methyl isobutyl ketone,
n-butyl alcohol, cyclohexanone, and methanol;all spent solvent
mixtures/blends containing solely the above spent nonhalogenated
solvents; and all spent solvent mixtures/blends containing, before use,
one or more of the above nonhalogenated solvents, and a total of ten
percent or more (by volume) of one or more of those solvents listed in
F001, F002, F004, and F005; and still bottoms from the recovery of these
spent solvents and spent solvent mixtures.
2-3
-------
F004—The following spent nonhalogenated solvents: cresols and cresylic
acid, and nitrobenzene; all spent solvent mixtures/blends containing,
before use, a total of 10 percent or more (by volume) of one or more of
the above nonhalogenated solvents or those solvents listed in F001, F002,
and F005; and still bottoms from the recovery of these spent solvents and
spent solvent mixtures.
F005—The following nonhalogenated solvents: toluene, methyl ethyl
ketone, carbon disulfide, isobutanol, and pyridineJ all spent solvent
mixtures/blends containing, before use, a total of 10 percent or more (by
volume) of one or more of the above nonhalogenated solvents or those
solvents listed in F001, F002, and F004; and still bottoms from the
recovery of these spent solvents and spent solvent mixtures.
The solvent waste constituents of concern and their respective
hazardous waste code numbers are:
P022-carbon disulfide
U002-acetone
U031-n~butyl alcohol
U037-ehlorobenzene
U052-cresols and cresylic acid
U057-cyclohexanone
U070—o—dichlorobenzene
U080-aethylene chloride
U112-ethyl acetate
U117-ethyl ether
U121-triehlorofluoromethane
U140-isobutanol
U154-methanol
U159-methyl ethyl ketone
U161-methyl isobutyl ketone
11169-nitrobenzene
U196-pyridine
U210-tetrachloroethylene
U211-carbon tetrachloride
U220-toluene
U226-1,1,1-trichloroethane
U228-trichloroethylene
U229-xylene
EPA has established immediate effective dates for all but: three of the
categories of solvent wastes subject to the 14 January 1986 proposed
rulemaking. These three excepted categories are solvent-water mixtures
(wastewaters) containing less than 1 percent (10,000 ppm) of total organic
constituents and less than 1 percent (10,000 ppm) of total solids, inorganic
sludges and solids containing less than 1 percent (10,000 ppm) total organic
constituents, and solvent—contaminated soils. The Agency has proposed a
2-4
-------
2 year national variance for these solvent wastes based upon a determination
that the capacities of alternative treatment technologies capable of achieving
the treatment standards for these wastes (wastewater treatment units and
incinerators), in conjunction with the capacities of alternative recovery and
disposal technologies, are insufficient to accomodate the quantities of these
solvent wastes currently managed in land disposal units. Schedules have yet
to be established for several other P and U Code solvents and other low
molecular weight organics.
2.2.2 Proposed Treatment Standards
The legislative history of the 1984 Amendments to RC1A indicates that a
waste may be restricted from land disposal not only on the basis of hazards
posed by its inherent toxicity, but also because of its ability to degrade
clay and synthetic liners and to mobilize relatively nonmobile hazardous
constituents, when co—disposed with other hazardous waste. Since solvents
exhibit these characteristics, the Agency has considered these overriding
factors in developing treatment standards for solvents.
EPA has determined that a number of technologies are applicable to the
treatment/recovery of solvent wastes, including biological degradation, steam
stripping, carbon adsorption, distillation, incineration, and fuel
substitution. The Agency is identifying acceptable technologies for each
solvent waste based upon the wastes physical form, the specific solvent
constituents they contain, and the concentrations at which such constituents
are present. Although final evaluations have not yet been completed,
preliminary results indicate that these treatment technologies do not pose
total risks to human health and the environment greater than those posed in
the direct land disposal of most categories of the solvents wastes subject to
the proposed rulemaking.
Table 2-2 shows the technology-based treatment levels and the proposed
treatment standards for each solvent constituent in waste codes F001 through
F005 wastes.
2-5
-------
TABLE 2.2. PERFORMANCE ACHIEVED BY TREATMENT TECHNOLOGIES' (mg/liter)
Treatment Technology
Constituent
Fuel
Steam Carbon Biological Treatment substitution/
stripping adsorption treatment combination incineration
Acetone
n-Butyl alcohol
Carbon disulfide
Carbon tetrachloride
Chlorobenzene
Creosols
Cyc Iqhexanone
1 ,2-Dichlorobenzene
Ethyl acetate
Ethylbenzene/ 0.200
Ethyl ether
Isobutanol
Methanol
Methylene chloride 0.109
Methyl ethyl ketone
Methyl isobutyl ketone
Nitrobenzene
Pyridine
Tetrachloroethylene
Toluene 0.036
1,1,1-Trichloroethane 0.457
1, 1 ,2-Trichloro-l ,2 ,2-
r\
trifluoroethane 0.457
Trichloroethylene 0.019
Trichlorofluoromethane 0.4572
Xylene )
0.0522
0.1002
0.010
0.292
0.62 0.1002
0.1002
0.302
0.1002
0.010
0.1002
0.0502
0.01002
0.011
0.0502
0.0102
0.027 0.010
0.5002
0.010
0.016 0.066
0.011
0.0522
0.1002
0.0102
0.010
0.0623 0.020
0.100
0.1002
0.0533 0.0102
0.1002
0.1002
0.1002
0.1002
0.0502
0.010
0.050
0.0102
3— - — f\
.„.«<.« 0.0102
0.5002
0.010
0.2303 0.010
0.010
0.0102
0.010
0.0102
0.0054 0.0102
Source: Federal Register Vol. 51, No. 9, January 14, 1986. Table 11, Page 1722.
(1) Includes activated sludge, trickling filters, and aerated lagoons.
(2) Estimated values.
(3) Activated sludge followed by granulated activated carbon adsorption.
(4) Granulated activated carbon followed by steam stripping.
2-6
-------
SECTION 3.0
CURRENT INDUSTRIAL SOURCES, GENERATION,
AND MANAGEMENT ACTIVITIES FOR SOLVENT WASTES
3.1 INDUSTRIAL SOURCES OF WASTE SOLVENTS
3.1.1 Wastes Considered in Study
The primary purpose of this document is to summarize available
information regarding treatment of priority solvents subject to the land
disposal ban effective 18 November 1986. Knowledge of current waste
management practices and solvent waste characteristics are necessary to
determine the applicability of these various technologies. Some discussion of
these factors for other solvents and low molecular weight organic compounds
has been included since it is recognized that they will often be amenable to
the same treatment techniques. Wastes containing these low molecular weight
organic, often ignitable compounds, will compete with priority solvents for
available treatment capacity as they become subject to land disposal
restrictions.
3.1.2 Sources of Solvent Wastes
Priority solvent usage rates are summarized in Table 3.1.1,
distinguishing between use as a solvent, consumptive use as a chemical
intermediate, and use in the formulation of consumer products. Solvent use is
summarized below, first in terms of the major solvent using industries, and
then by individual priority solvent. Unless otherwise indicated, demand,
export and use data were obtained from the most recent chemical profiles
issued by the Chemical Marketing Reporter.
3-1
-------
TABLE 3.1.1.- CONSUMPTION OF MAJOR SOLVENTS (106 LBS/YEAR)
Use as 8 solvent
Solvent classification
Halogenated (RCRA)
Perchloroethylene
1,1, l-Trichloroe thane
Metbylene chloride
Pluorocarbons
Trichloroethylene
Chlorobenzene
0-Dichlorobenzene
Chloroform
Carbon tetrachloride
Subtotal:
Nonhalogenated (RCRA)
XyUne
Toluene
Methyl ethyl ketone
Hethanol
Acetone
n-Butano I
Ethyl acetate
Methyl isobutyl ketone
Ethyl benzene
Isobutanol
Cresols
Cyc lohexanone
Nitrobenzene
Benzene
Subtotal:
Total RCBA solvents
Other organic solvents
tnon-RCRA)4
Total solvent usage
X halogenated (RCRA)
Z nonhalogenated (RCRA)
Coatings
4
12
7
23
425
358
308
113
185
138
90
99
26
0
0
1,742
1,765
1,653
3,418
0.7
51.0
Adhes ives
6
60
0
66
67
71
78
0
0-10
0
221
287
—
-287
23.0
77.0
Inks Degreasing
196
378
39
1113
136
5
O-IO3
870
26
65 31
22
53
22
7
0-10
8
0 0
0-30
92 162
92 1,032
813
92 1,845
0 47.2
100.0 8.8
Process
solvent
36
132
373
113
7
O-IO3
330
62
35
156
146
121
0.54
18
12-273
20
0
19
551
881
1,129
2,010
16.4
27.4
Other
3491
113»
462 .
78
107
29
54
24
292
754
208
962
48.0
30.4
Total
solvent
use
555
486
284
148
143
113
12
O-IO3
O-IO3
1,751
658
560
464
419
328
199
136
121-1383
54
46
32
5-25
19
0-30
3,076
4,827
3,361
8,180
. 21.4
37.5
Nonaolvenc
use
363
72
157
880
10
155
35
312-4123
610-6203
2,598
28,969
6,040
23
7,3352
1,515
1,284
4
O-IO3
7,146
120
88
946
4453
54,916
—
—
—
—
Export
43
42
49
32
17
0
0
13
40
236
73
—
87
112
79
14
9
13
S75-5953
_„
15
987
—
—
—
**"*
Percent used
Total domestically
demand as solvent
865
600
490
1,060
170
268
47
425
660
4,585
29,627
6,600
560
7,754
1,930
1,595
220
152
7,200
175
133
600
965
1,460
58,971
—
—
—
._
64
81
58
14
84
42
25
0-2
0-2
38
2
8.5
83
5
17
12
62
80-91
0.8
26
24
1-4
2
0-1
5.2
—
—
—
~
^Bry cleaning.
Autouative solvent chemicals - 278,
^Estimated.
Includes ethanol, isopropanol, special napthas, ethylene glycols, hexane and mineral spirits. Excludes use of ethanol
consumer applications.
*Paint stripper.
Sources; Reference Nos. I through 7.
as a solvent in
-------
Solvent consumption is shown in Table 3.1.2 by industrial end use
category. Of total priority solvent consumption, 64 percent is represented by
nonhalogenated organics. These are widely used by the paint and allied
products industry (e.g., as ingredients and wash solvents) and in cold
cleaning applications. Conversely, halogenated organics are primarily used in
vapor degreasing, cold cleaning and dry cleaning.
The top 20 industries generating priority solvent waste streams regulated
under RCRA in 1981 are shown in Table 3.1.3. These data were compiled from an
EPA nationwide survey of hazardous waste generators which is discussed in
detail in Section 3.3. Small quantity generators, which were excluded from
this survey, include many degreasing, coating, and dry cleaning operations.
Number of generators and waste quantities for these firms are also discussed
in Section 3.3 (Table 3.).
Waste Solvents from the Paint and Allied Products Industry—
The paint and allied products industry represents the largest solvent end
use in the U.S. with over 98 percent of its priority solvent consumption
accounted for by nonhalogenated solvents. Xylene, toluene, MEK, and acetone
are the most widely used materials, while chlorinated solvents (e.g.,
1,1,1,-TCE) find wide use only in adhesives (Table 3.1.1). Solvent
consumption in the industry is on the decline primarily as a result of process
and product changes in response to environmental restrictions. Solvent
consumption declined 10 percent during the 1970's and early 1980"a as the
industry shifted towards high-solids, powder, radiation-curable, and
3 9
water—based coatings. ' Solvent consumption can be expected to decrease
even more rapidly in response to land disposal restrictions. Aliphatics and
aromatics will lose market share to oxygenated solvents (e.g., ketones,
esters, alcohols, glycols) to meet emissions and disposal standards and the
3
more demanding viscosity requirements of high solids systems. Similarly, a
shift from solvent based to water borne or UV curable inks and water based
adhesives will occur, reducing overall solvent consumption in these
applications. However, halogenated solvents will gain market share in the
growing adhesives industry relative to nonhalogenated solvents, which present
handling problems as a result of their flammability.
3-3
-------
TABLE 3.1.2 PRIORITY SOLVENT INDUSTRIAL END USES
Percent of use category
by solvent type
Use category
Faint/coatings
Vapor degrees ing/cold
Process solvent
Dry cleaning
Adhesives
— Halogenated
1.3
cleaning 84,3
37.5
100.0
23.0
Industrial paint stripper 100.0
Inks
Miscellaneous
Total industrial uses
0.0
100.0
36.3
Nonh a logenated
98.7
15.7
62.5
0.0
77.0
0.0
100.0
0.0
63.7
Percent of solvent
type by use category
Halogenated
1.3
49.7
18.8
19.9
3.8
6.5
0.0
0.0
,—
Nonhalogenated
56.8
5.3
18.0
0.0
7.3
0.0
3.0
9.6
—
Percentage
of total
RCRA Industrial
solvent usage
36.7
18.9
18.4
9.5
6.0
2.4
2.0
6.1
—
Source: References Nos. 1 through 7.
-------
TABLE 3.1.3. TOP 20 INDUSTRIES GENERATING SOLVENT WASTES
I
Ul
No. of
estab.a
2145
1160
1529
4287
541
2902
4656
4151
393
337
15490
2237
2563
6506
560
1040
32
861
5392
235
57017
SIC
code
2851
2869
2821
3471
2833
3479
3662
3714
9711
3721
3079
3674
2899
7391
3411
3711
2067
2879
3679
3951
SIC description
Paints and Allied Products
Industrial Organic Chemicals
Plastics Materials
Plating and Polishing
Medicinal, Botanical Products
Metal Coating and Allied Services
Communication Equipment
Motor Vehicle Parts
National Security
Aircraft Equipment
Plastic Products, Miscellaneous
Semiconductors
Chemical Preparations
Research & Development Labs
Metal Can Fabrication
Motor Vehicle Bodies
Chewing Gum
Agricultural Chemicals
Electronic Components
Pens & Mechanical Pencils
Weighted number of
Halogenated
solvents'*
105
327
215
471
137
136
186
241
166
107
120
93
85
103'
35
57
57
59
96
66
solvent waste streams
Nonhalogenated
solvents
1,436
654
536
176
323
279
225
161
178
230
201
194
189
163
154
127
87
85
40
59
aNumber of establishments based on Dun's Marketing Services, a company of
Dun & Bradstreet Corp., 1983 Standard Industrial Classification Statistics.
''Information on generators taken from 1981 data (National Survey of Generators).
Source: Engineering Science Reference No. 8.
-------
Data on specific waste solvent streams from the paint and allied products
industry has been presented in the literature. * * These indicate that
solvent concentration varies from over 90 percent to trace levels depending on
waste source and treatment methods applied. Solvent-bearing coating and ink
wastes originate primarily from cleaning operations on tanks and equipment
following color or other process changes. These wastes typically consist of a
blend of solvents with solids concentrations of up to 10 percent before
recovery or disposal is required.
Off-spec materials constitute a minor source of waste since much of this
product can be reused in lower grade products. Solvent concentrations reflect
13
those found in saleable products (e.g., up to 42 percent in paints).
A 1977 EPA study of the paint and coatings manufacturing and factory
applied coatings industries estimated that approximately 95 percent of the
hazardous solvent generated is contained in spent cleaning solutions as
summarized in Table 3.1.4. These data suggest that overall, 1 percent of
solvents consumed by the paint and allied products industry ultimately results
in waste material requiring disposal which is, on average, 21 percent solvent
by weight. However, solvent recovery has increased from approximately
35 percent in 1977 to 66 percent in the early 1980's (see Section 3.2).
Thus, on average, solvent wastes currently land disposed will tend to be less
concentrated than these data suggest. Instead they will tend to have higher
concentrations of non—solvent materials such as resins, extenders, driers,
anti-skinning agents) and contaminants; e.g., dirt, oil, grease.
Recovered solvents often find re-use as wash solvents as opposed to use in
product formulations unless they are recycled in a manner (e.g.,
fractionation) which enables them to meet more strict product specifications.
Waste Solvents from Degreasing Operations—
Decreasing is usually employed to remove oils, greases, waxes,
lubricants, tars, water and other foreign substances from metal surfaces and
other materials. Solvent cleaning operations are carried out in three basic
types of equipment; cold cleaners, open-top vapor degreasers, and conveyorized
degreasers. Cold cleaners account for all the nonhalogenated solvents used in
3-6
-------
TABLE 3.1.4 HAZARDOUS SOLVENT WASTE GENERATION IN THE PAINT AND COATINGS
MANUFACTURING AND FACTORY APPLIED COATINGS INDUSTRY
Solvent waste
generation8
(tons/year)
Percent of
waste which
is solvent
Percent of
total solvent
constituents
in waste
Cleaning wastes
Off-specification
material
Spills
Total
80,261
5,051
5,438
90,750
22.5
5.9
2.0
21.0
95.2
4.2
0.6
100.0
aWeighted by a factor of 1.14 to reflect estimates of solvent consumption
provided in Table 3.1.1.
Source: US EPA 1977. Reference No. 10.
3-7
-------
2
degreasing and about one third of the halogenated solvents. The remaining
A
two-thirds is used in 21,000 vapor and 3000 conveyorized degreasers in a
3
ratio of approximately two to one. 1,1,1-trichlorethane is the most widely
used degreasing solvent followed by fluorocarbons, trichloroefchylene and
perchlorothylene (Table 3.1.1).
Cold cleaning solvents are recycled or disposed when they become
contaminated with approximately 10 percent foreign matter by weight or when
cleaning efficiency is otherwise inhibited due to a reduction in solvent
14
power. Vapor decreasing solvents are recycled or disposed when they
f '
become 15 to 30 percent contaminated, when the boiling point reaches 5 to 1U°F
12
above the pure solvent boiling point, or when solids settling and buildup
14
impairs heat transfer. Fabric scouring solvents are typically recycled on
a continuous basis through distillation and decanting. Degreasing wastes are
generally amenable to recovery unless contaminants include resinous or
polymerizable compounds which will foul heat transfer surfaces. ;
Alternatively, wastes may not be recycled for reuse if stringent product
purity specifications cannot be met; e.g., using recovered solvent in
applications such as electronic circuitry or printed circuit boards.
Table 3.1.5 summarizes waste generation rates as a fraction of solvent
2
consumption for various cleaning apparatus. These data are from 1979
operations, which was prior to the upgrading of emissions control equipment in
many degreasers. Thus, it is likely that the fraction of solvent which
becomes waste is higher than the values presented here. However, these data
can be applied to current solvent usage rates to estimate th€s quantity of
solvent and potentially ignitable waste which originates from degreasing and
paint stripping operations. Using the solvent usage rates provided in
Table 3.1.1 and their distribution between types of degreasing
1 2
applications, * the data suggest an annual generation of 388 million pounds
of spent cold cleaning solvent waste (80 percent halogenated), 87 million
pounds fabric scouring (77 percent halogenated) and 109 million pounds of
halogenated vapor degreasing waste. If, on average, 75 percent of the
material is recovered, this results in 146 million pounds of oily still
bottoms (55 percent oil).
3-8
-------
TABLE 3.1.5. WASTE SOLVENT GENERATION BY TYPE
OF DECREASING OPERATION8
Total solvent consumption,
that becomes waste solvent, '
Degreasing operation
Range
aSource: Reference No. 2.
Assumed to be a conveyerized cold cleaner.
Average
Cold cleaners:
Manufacturing (44 percent)
Maintenance (56 percent)
Open top vapor degreasers
Conveyorized vapor degreasers
Fabric scourers**
40 to 60
50 to 75
20 to 25
10 to 20
40 to 60
50.0
62.5
22.5
15.0
50
3-9
-------
Total solvent consumption in metal degreasing is expected to grow less
than 1 percent per year because of improved solvent conservation through
3
equipment modification and recycling. Also, industry surveys report a
shift towards the use of acid-alkali detergent cleaners for degreasing,
-------
3«t*3 Halogenated Organic Solvent Usage in Industry
Perchloroethylene (PERC)—
PERC is the most widely used halogenated solvent with a. total annual
industrial solvent consumption of 555 million gallons. PERC solvents are
primarily used in the dry cleaning (56 percent) and textile processing
industries (24 percent) (e.g., fabric scouring, carrier solvent for fabric
finishes and water repellants and sizing and desizing operations) because of
its high cleaning power, low toxicity, nonflammability, and stability. The
EPA estimates that 70 percent of all dry cleaners used 349 million gallons of
4
PERG in 1983 which represents a majority of the industry's solvent usage.
Other solvent applications include use in metal degreasing (10 percent) and
coatings/adhesives (1 percent). PERC is used in about 15 percent of the vapor
12
degreasers nationwide, but is not more widely applied due to its high
boiling point relative to other degreasing solvents. It is used when high
temperatures are required to remove high melting waxes and grease, when long
• c".
12
2
cleaning cycles are required, when higher cleaning efficiency is required,
or when water is present on part surfaces,
Nonsolvent uses of PERC include use as a chemical intermediate
(42 percent), primarily in the production of fluorocarbon F-113, and export
(5 percent). Zero growth is expected for PERC with greater recycling in dry
cleaning and metal cleaning offsetting increases in demand for F-113 in the
electronics industry.
1,1,1-Trichloroethane (1,1,1-TCE)—
Demand for 1,1,1-TCE was 600 million Ibs/yr of which 7 percent was
exported and 81 percent was consumed domestically as a solvent. The primary
solvent uses of 1,1,1 TCE are cold cleaning (41 percent), vapor degreasing
(22 percent), adhesive formulations (10 percent), electronics equipment
cleaning and process solvent (6 percent), and coatings (2 percent), as well as
nonsolvent uses in adhesive applications (7 percent). It is the most widely
applied degreasing solvent currently in use. Growth is forecast at 2 percent
per year which should outperform most other chlorinated solvents. It replaced
3-11
-------
trichloroethylene in many metal cleaning applications during the 1970*s due to
its similar characteristics (nonflammable, suitable evaporation rate, medium
solvency), lower toxicity, and lower temperature of parts on removal from
12
degreasers. 1,1,1-TCE is available in uninhibited and inhibited grades
using such stabilizers as nitromethane, methylpyrole, 1,4-dioxane, butyiene
3
oxide, 1,3-dioxolane and secbutyl alcohol. Improperly stabilized,
7 19
1,1,1-TCE can decompose in the presence of aluminum, zinc, or magnesium.
/ !
Methylene Chloride (MC)—
MC is a nonflammable, noncorrosive, high solvency petrochemical which is
used as a solvent in nearly 60 percent of its applications. It is most widely
used as a solvent in paint removal for both industrial and consumer
applications (23 percent) chemical processing (e.g., plastics and1fiber
processing), as an extraction solvent in food processing and for oil dewaxing
(20 percent) metal degreasing (8 percent, roughly 80 percent in cold cleaning
operations), and electronics (7 percent). Nonsolvent uses include aerosols
(20 percent), urethane foam blowing agents (5 percent) , a vapor pressure
depressant in consumer products and manufacture of polycarbonate insecticides
and herbicides. An additional 10 percent of MC is exported. MC is used
in the urethane foam industry to clean foam heads and lines between production
11
runs.
MC has been declining in domestic consumption since the late 1970's due
to regulatory pressures. It is expected to continue to decline by 1 to
2 percent per year despite advantages over other solvents in several of its
applications. In degreasing, it may be used to remove polymer residue (high
solvency) or for cleaning heat sensitive parts (low boiling point). However,
its low boiling point and consequent low volume of condensate generated may
preclude its use where high cleaning efficiency is required. Low operational
costs compared to PERC and higher stability in the presence of water compared
to 1,1,1-TCE maintain its competitiveness in the degreasing market.
3-12
-------
Fluorocarbons—
Fluorocarbons include a variety of fluorinated aliphatics, among them
F-ll (trichlorofluoromethane), F-12 (dichlorodifluoromethane) and
F-113 (1,1,2-trichloro-trifluoromethane) which are RCRA solvents. Solvent
uses account for only 14 percent of demand and consist primarily of F-113 used
in specialized solvent cleaning applications in the electronic, aerospace, and
3
optical industries. Similar to methylene chloride, fluorocarbons are
suitable for removing polymer residue and for cleaning heat sensitive parts.
Their higher cost is offset by higher vapor densities which result in less
diffusion of vapors from the degreasers and larger volumes of condensate which
make them more suitable for cleaning.
Proprietary mixtures of F-113 with solvents such as acetone, ethanol,
methylene chloride, and chloroform have found uses for cleaning fluxes from
printed circuit boards, electrical relays, missile guidance systems,
electrical meters, hearing aids, and semiconductors. F-113 is preferred
relative to other solvents as a result of its minimal attack on paints,
3
gaskets and wire insulation. A small amount of F-113 is also used in the
4 16
dry cleaning industry (less than 1 percent of dry cleaners). *
The major use for F-12 is in air conditioning and refrigeration which,
combined with F-22, accounts for 39 percent of flourocar&on use. F-ll is
primarily used as a foam blowing agent (17 percent of total use). Other
nonsolvent uses include resin intermediates and aerosol propellants.
Fluorocarbon demand is expected to increase 4 to 5 percent per year with the
highest growth experienced by F-113 in electrical and electronic applications
as a substitute for chlorinated solvents.
Trichloroethylene (TCE)—
TCE is used primarily as a decreasing solvent which account for
80 percent of total demand. Other solvent uses include paints, coatings and
general solvent applications (4 percent) while 10 percent is exported and
6 percent is consumed as a chemical intermediate (pesticides and herbicides).
High purity, low residue grades are used for cleaning critical electronic
components, chemical synthesis, and extraction processes for wax, oils and
3-13
-------
resins. Stabilized grades are used for vapor degreasing and other cleaning
operations and have been developed to remain effective through repeated
distillations and degreasing of aluminum. Demand is expected to decline 2
to 4 percent per year as a result of solvent recycling practices and
regulatory pressures. However, TCE has superior dissolving properties and is
technically preferred to tetrachloroethylene in vapor degreasing because of
its lower boiling point.
Cblorobenzene—
Chlorobenzene is used as a solvent in pesticide formulation and toluene
»
5
di~isocyanate processing and, to a lesser extent, as a degreasing agent,
dye assist agent, and synthetic rubber solvent for dipping applications."
These represent 42 percent of current demand. Nonsolvent uses include
production of nitrochlorobenzenes (32 percent) and diphenyl oxide and
phenylphenols (15 percent). Demand is expected to decline 3 percent per year
or more, primarily due to reduced consumption in pesticides.
o-Dichlorobenzene—
o-Dichlorobenzene is used as a solvent in 25 percent of its applications,
primarily in the phosgenation of mono and diamines to toluene di-isocyonate
(15 percent) and in cold cleaning operations (10 percent). The latter include
paint removing, engine cleaning, and de-inking. It's primary nonsolvent
use is in the synthesis of pesticides (65 percent) including 3,4-dianiline in
dyestuff manufacture (5 percent) and as a heat transfer agent. Demand is
expected to decline 1 percent per year.
Chloroform— '
The vast majority of chloroform is used as a chemical intermediate in the
production of Fluorocarbon F-22 (93 percent). Exports account for 3 percent
of demand and the remaining quantity (4 percent) is divided between the
following miscellaneous uses: manufacture of polytetrafluoroethylene,
preparation of dyes and pesticides, extractant in the production of penicillin
vitamins and flavours, a general solvent for adhesives, resins,
3-14
-------
Pharmaceuticals and pesticides, as a solvent for removing fat from waste
1 5 19
products, and as a dry cleaning spot remover, " * One source reported
that chloroform was widely used as a solvent in the lacquer industry but did
20
not substantiate this claim. Total use as a solvent is probably leas than
10 million pounds annually.
Carbon Tetrachloride—
Carbon tetrachloride1s use as a solvent is negligible. It is an ideal
extraction solvent (low surface tension, high density, and good solvency) for
oils, waxes and fats, particularly those derived from animal and vegetable
sources, and has found solvent use in degreasing, shoe and furniture polishes,
3
paints, lacquers, printing inks, floor wastes, stains, and as a grain
fumigant. However, regulatory pressures have since reduced its uses to
that of a chemical intermediate, primarily in the production of
Fluorocarbons F-ll and F-12 (91 percent). The bulk of the remaining carbon
tetrachloride produced in the U.S. is exported (6 percent) or used in other
nonsoivent chemical intermediate applications(3 percent). Similar to
chloroform, total use as a solvent probably does not exceed 10 million pounds
annually.
3.1.4 Non-halogenated Organic Solvent Usage in Industry
Xylene—
Xylenes used for solvent applications are mixed xylenes which have been
depleted in ortho- and para-xylene. These represent only 1.9 percent of total
3
xylene demand in the U.S. Xylene is less expensive than most solvents and
evaporates rapidly. Similar to other aromaties, it is able to solubilize
resins and lacquers and is consequently widely employed in the paint and
coatings industry, e.g., wash solvent, stripping and general solvent use in
formulations. This accounts for approximately 58 percent of mixed xylene
use. Other solvent uses include adhesives (9 percent), process solvent
(9 percent), agricultural sprays (9 percent) cold cleaning and fabric
2
scouring (4 percent) and miscellaneous applications (2 percent) such as
printing inks. Total mixed xylene demand is expected to decline by more
than 2 percent per year.
3-15
-------
Nonsolvent uses of xylenes include production of dimethyl terephthalate
and terephthalic acid using p-xylene (92.6 percent), phthalic anhydride with
o-xylene (2.5 percent), gasoline (2,8 percent), isophthalie acid and, to a
lesser extent, isophthalonitrile with m-xylene (0.2 percent). Growth for
nonsolvent uses should bs 3 to 4 percent per year.
Toluene—
Chemical uses of toluene are estimated to require 6600 million pounds in
1986, of which 8.5 percent is used in solvent applications. Of this,
64 percent is consumed in paints and coatings and 36 percent in formulating
adhesives, inks (rotogravure) and Pharmaceuticals with minor uses such as
chemical processing and cold cleaning (e.g. varnish remover). Toluene is
relatively fast drying and inexpensive. It is used along with other aromatics
as a wash solvent and for thinning resins which are difficult to solubilize by
aliphatics (e.g. short-oil alkyds, vinyl alkyds, phenolics, chlorinated rubber
coatings, lacquers). * Use of toluene as a solvent has decreased in
recent years from estimates of 625 million pounds in 1980 to 5t»0 million
20
pounds in 1986. As a result of continued regulatory pressures it will
probably continue to decline at roughly 2 percent per year. However, other
chemical uses have been increasing at nearly 5 percent per year since 1981.
These nonsolvent uses include the production of benzene (71 percent), toluene
di-isocyanate (9 percent), benzoic acid (2 percent), benzyl chloride
(1 percent), and other chemicals (0.75 percent).
Methyl Ethyl Ketone (MEK)—
The majority (83 percent) of MEK is used as a solvent or exported
(13 percent). Primary solvent uses include vinyl, nitrocellulose, acrylic and
other coatings (55 percent), a solvent in adhesives (14 percent); magnetic
tapes ( 6 percent) printing inks (4 percent), and lube oil dewaxing
(4 percent). Nonsolvent uses are minor and consist of applications as a
chemical intermediate (4 percent). Other possible uses include vegetable oil
11 2
extraction, azeotropic distillation in refineries, and cold cleaning.
3-16
-------
MEK is rarely, if ever, used in paints since it is relatively expensive.
However, its rapid drying capability and strong solvency for lacquer binders
(i.e., nitrocellulose, acrylics, vinyls, etc.) make it the preferred solvent
lii
6
for thinning of epoxy and PVC ' coatings and occasional use as a
stripper.
Growth in MEK demand is expected to be 3 percent per year since its high
solvency has allowed it to maintain its coatings market share. However,
increased recycling in the magnetic tape and printing industries will reduce
demand in these sectors.
3
Methanol —
Low price, high purity and excellent dissolving properties are
responsible for making methanol one of the most widely used solvents.
However, only 5 percent of methanol consumption consists of use as an
industrial solvent. Of this, methanol is most widely used as a process and
wash solvent in the production of paints and coatings (27 percent), for
cleaning and as a component of paint remover (13 percent), as a chemical
process solvent in the preparation of Pharmaceuticals (35 percent), and as a
solvent for inks, coated fabrics, and other applications (25 percent). Total
solvent use should expand at a rate of 2.2 percent annually, with growth in
most applications somewhat offset by stable demand in the paint and coatings
industry.
Nonsolvent uses constitute the majority of methanol consumption
(95 percent). These include production of formaldehyde (34 percent), acetic
acid (8 percent), chloromethanes (7 percent), methylamines (6 percent),
windshield cleaner and deicer (4 percent), methyl methacrylate (5 percent),
methyl tertiary butyl ether (5 percent), and other uses (26 percent).
Methanol demand is expected to be high in nonsolvent uses (5.5 percent
annually) led by growth in acetic acid, MTBE, and several of the other
chemicals. Demand for formaldehyde, the largest methanol derivative, is also
expected to be strong due to growth in the building and construction industry.
3-17
-------
Acetone—
Approximately 17 percent of acetone is used in solvent applications.
These include coatings (10 percent) and miscellaneous uses (7 percent) such as
a spinning solvent in the manufacture of cellulose acetate fiber
(5 percent), a solvent in the manufacture of inks, smokeless powder,
3'
cements and artificial leather, dewaxing of lubricating oils, and cold
2
cleaning. In the coatings industry it is used in varnishes, lacquers,
thinners, and as a wash solvent. Most acetone is used in nonsolvent
applications including the manufacture of methacrylates (33 percent),
methyl isobutyl ketone (10 percent), bisphenol A (9 percent), aldol
derivatives (7 percent), vitamins, Pharmaceuticals and cosmetics (6 percent),
and methyl isobutyl carbinol (2 percent). Despite its preferred use in high
solids coatings, acetone use in coatings will probably decline slightly as the
industry shifts toward water-based and powder coatings. Other solvent uses
3
may increase 2 percent per year and nonsolvent uses are expected to
/
increase 3 percent per year.
n-Butanol—
Approximately 12.5 percent of n-Butanol is used in solvent applications
with 9 percent in paints, coatings and adhesives and 3.5 percent as a process
1 2
solvent and cold cleaner* Process solvents are used in the production
3 22
of Pharmaceuticals, waxes, resins, dyes, and photographic chemicals.
Butanol is not a solvent for nitrocellulose but its presence in the solvent
mixture of a nitrocellulose lacquer or enamel, or in alkyd/amino enamels,
improves flow and leveling during coating application. Other coatings,
such as alcohol soluble shellacs may use butanol as a thinning agent.
Nonsolvent uses include the production of butyl aerylates and tnethacrylate
(30 percent), glycol ethers (23 percent), butyl acetate (12.5 percent),
plasticizers (8 percent), amino resins (5 percent), amines (1 percent), and
exports (7 percent).
Demand for n-butanol as a solvent is expected to remain stable but
overall growth is projected at 3 to 4 percent per year. n-Butanol 'based
water-borne coatings and adhesives are favored over competing products due to
their solubility characteristics. This compensates for the shift to more
active solvents in high solids coatings.
3-18
-------
Methyl Isobutyl Ketone (MIBK)—
MIBK is similar to MSK in that the majority of the compound
(80-91 percent) is used in solvent applications, primarily in the coatings
industry. The largest single use is as a solvent in nitrocellulose lacquers
(30 percent) with another 36 percent used for various coatings, adhesives, and
inks. Its strong solvency, low density, and high electrical resistivity mate
it a good solvent for high solids coatings such as acrylics, polyesters,
alkyds, and acrylic/urethanes. Another significant use (8 percent) is as a
rare metal extractant and the remainder is exported (9 percent) or applied
in miscellaneous uses such as a solvent in pesticides, Pharmaceuticals
(tetracyclene antibiotics purification), and use as a denaturant. MIBK is
frequently blended with MEK to achieve balanced active solvent characteristics
in high solids lacquers and for use as a solvent in low-viscosity vinyl resin
solutions.
As with MEK, MIBK is expected to increase in demand by 3 percent per
year, as shifts to high solids coatings encourage the increased use of pure
ketone formulations at the expense of lower solvency blends.
Ethyl Acetate—
Ethyl acetate is widely used as a solvent, particularly in protective
coatings such as nitocellulose lacquers. Excluding exports (36 percent),
solvent use accounted for 98 percent of ethyl acetate demand with the
remainder being used in chemical synthesis. Coatings accounted for 66 percent
of domestic solvent usage, plastics solvents (acrylic and cellulosic polymers)
accounted for 13 percent, and use in inks, adhesives, and other applications
made up the remaining 21 percent of solvent use. Demand for ethyl acetate has
been stalled due to increased production of water-based coatings and the
substitution of ketones in place of the less active esters in high-solids
coatings. However, since it is a moderate cost, low toxicity solvent which
serves mature markets, demand is still expected to increase from I to
2.5 percent per year.
3-19
-------
Isobutanol—
Isobutanol can be readily interchanged with n-butanol in solvent
applications. Solvent end uses consumed approximately 25 percent of
isobutanol demand with 14 percent being used in coatings and 11 percent as a
3
process solvent. Coatings which use isobutanol include nitrocellulose
lacquers, latexes, acetate exters, urea-formaldehydes, butyl acetates and
alcohol soluable shellacs. Demand in these applications is expected to
mirror that of n-butanol. Non-solvent uses include production of isbbutyl
amine herbicides (21 percent), lube oil additives (14 percent), isobutyi
acetate production (14 percent), acrylates (4 percent), exports (50 percent)
3
and miscellaneous uses (17 percent). Demand for these diverse non-solvent
uses should parallel economic growth.
Oresols—
Cresols are used in the production of magnetic wire as an enamel solvent
(18 percent), as an ore flotation agent (6 percent), as an intermediate in the
production of phosphate esters (16 percent), resins (16 percent), and
antioxidants (16 percent). Miscellaneous chemical intermediate applications
(13 percent) include perfumes, herbicides, disinfectants, and use a;S a
22 . '
textile scouring agent. Six percent of eresol demand is made up of use as
a cleaning compound which, together with wire enamel solvents, represents the
bulk of eresol use as a solvent. Exports account for the remaining 10 percent
of U.S. consumption. Growth of cresols is expected to be 2 percent per year,
primarily in the specialty chemicals areas.
Cyclohexanone —
Cyclohexanone is primarily used in nonsolvent applications with the vast
majority consumed as an intermediate for adipic acid (50 percent) and
caprolactum (45 percent) which, in turn, are nylon intermediates.
Gyclohexanone has not captured a significant share of the solvent market due
to economics (high boiling, slow evaporation, relatively expensive) but it is
an excellent solvent for vinyl resins, cellulose esters, ethyl cellulose,
polystyrene, and acrylic resins. It is reportedly used as a solvent for
3-20
-------
protective coatings, adhesives, inks, cleaning products, magnetic tapes, and
pesticides. Its cost is offset by its ability to tolerate large amounts of
inexpensive hydrocarbon solvents in the formulation of low cost thinners.
Long term demand as a solvent is not expected to increase significantly and
cyclohexanone demand for nylon should increase only 1 to 2 percent per year.
Nitrobenzene—
Nitrobenzene is used primarily in the production of aniline
(97.5 percent) which is expected to exhibit a growth rate of 6 percent per
year. Other nonsolvent uses include production of n-aeetyl p~amino phenol.
Use as a solvent, which is probably 2 percent or less of total demand,
includes the refining of certain lubricating oils and in various
Friedel-Crafts reactions. Other reported solvent applications include TNT
production, cellulose acetate manufacturing, metal and shoe polish, dyestuff,
22
rubber chemicals and photographic chemicals.
Benzene
Benzene is used as a solvent only in rare cases when it cannot be
replaced by less hazardous substitutes; e.g., other aromatics like toluene,
xylene.
One source reports use of benzene in dyes, coatings, and photographic
22
materials but it is probably employed as a chemical raw material rather
than as a solvent. A 1972 survey of the paint and coatings industry
2
.reported less than 9 million pounds per year of benzene consumption*
Another source reported 15 million pounds consumed in cold cleaning and
another 77 million pounds used in fabric scouring in 1974, representing
4
0.8 percent of total benzene consumption. It is likely that these solvent
uses have declined radically in the intervening years due to regulatory
pressures and awareness of benzene's toxicity.
Nonsolvent uses of benzene include production of ethylbenzene/styrene
(52 percent), cumene/phenol (22 percent), eyclohexane (15 percent),
nitrobenzene/aniline (4.5 percent), detergent alkylate (2.5 percent),
chlorobenzenes, maleic anhydride, and miscellaneous chemicals (3 percent), and
exports (I percent). Demand is expected to increase 2.3 percent per year.
3-21
-------
3.2 SOLVENT WASTE GENERATION
The most recent evaluation of solvent waste generation, management
practices and waste physical forms has been compiled by the U.S. EPA in the
form of a background document to support 40 CFR Part 268 land disposal
23
restrictions. This evaluation was based on an updated version ,of the 1981
National Survey of Hazardous Waste Generators and Treatment, Storage and
Disposal (TSD) Facilities Regulated Under RCRA which was performed by
24
WESTAT. In cases where this data was not available, the results of an
o
earlier study performed by Engineering Science , formed the basis of this
subsection. Specifically, data pertaining to the relative utilization of most
management practices by waste category (e.g., ignitables, halogenated
solvents, nonhalogenated solvents) and waste physical form (e.g., solid,
liquid, sludge) were based on the earlier survey.
The National Survey data was compared with other national, regional and
state surveys to develop an overview of solvent waste generation. The results
are summarized in Table 3.2.1. The primary difference between waste
quantities reported in these surveys is inclusion or exclusion of large
quantity, dilute wastewater streams which contain the constituents of
concern. For example, the EPA reports that three facilities accounted for
23
94 percent of solvent wastes treated in surface impoundments. GCA
concluded that little confidence can be placed in the reported generation and
management figures unless the wastes are analyzed on the basis of physical
form or, preferably, waste constituent concentrations. This approach was
adopted when possible.
Highlights of the combined analysis of these surveys are summarized below:
* 3 to 5 billion gallons of waste solvents and 1 to 3 billion gallons
of ignitable wastes are generated annually, the majority of which is
aqueous waste which is treated and discharged by methods which do
not constitute land disposal.8,25,23
* Approximately 1.5 billion gallons of nonaqueous waste solvents and
ignitables are managed at TSD facilities" which represent only
19 to 38 percent of the total wastes generated.
3-22
-------
TABLE 3.2.1. SUMMARY OF WASTE GENE1ATION AND MANAGEMENT (MILLIONS OF GALLONS)
Haste activity
Generation
Management Practices
Recycling (onsite)
(offsite)
Total:
Deep well injection
Incineration
Landfill
Surface impoundment
(includes all TSD)
Haste pile
Land application
Land disposal (excludes
deep-well injection and
TS impoundments)
Treatment
(excludes incineration)
Storage
Total TSD
Ignitables
603
—
1,400
231
37
63
26
301
63
14
26
122
56
16
— —
—
—
29
21
1,240
Halogenated
solvent
2,572
765
— -
121
110
59
52
179
162
290
62
—
. — .
—
—
__
38
—
588
Nonhalogenated
solvent
998
3,812
— ™"
315
275
51
45
366
320
12
133
—
—
— -
—
__
15
—
835
Total
solvent
• 3,570
4,577
3,200
436
385
110
97
545
482
302
291
195
32
32
1,169
0,7
0.001
53
39
1,423
Total
4,173
—
4,600
673
422
173
123
846
545
317
317
317
88
48
—
—
__
82
60
1,179
1,408
2,663
Source
(Reference No.)
8
25
23
8
26
8
26
8
26
8
27
8
27
23, 27
23
23
23
8
27
8
8
8
Total TSD (excluding
inorganic liquid) 1,157 116 261 377 1,534
Actual data reported before
extrapolation 447 85 173 258 705
-------
* 50 percent of large quantity generators (>^»000 kg/month) nationwide
(7,180 facilities) produce solvent-containing wastes; 43 percent of
generators (6,117 facilities) produce ignitable wastes. **
* At least 0.55 billion gallons were recycled, of which approximately
80 percent were recycled onsite. However, recycling,
particularly onsite activities, is not well defined and, therefore,
may differ significantly from this figure.
» Approximately 60 percent of nonaqueous solvents are currently
recycled.°» 2-J t
* 0.12 to over 0.2 billion gallons of solvents and ignitables are
recycled offsite with approximately 15 percent of this quantity used
as fuel.
• 0.4 billion gallons of waste are land disposed (1.5 billion
including storage and treatment impoundments") with the following
physical profile:
Inorganic liquids: 266 million gallons (<1% TOG, 17, TS;
includes soils)
Organic liquids: 36 million gallons C>1% TOC, <1% TS)
Organic sludges: 11 million gallons (>1% TOC, >l% TS) .
- Other/unknown: 46 million gallons
- Solids: 3.4 million gallons (no free .liquid)
* Land disposed solvent wastes are currently managed by:°
— Deep well injection: 79 percent
- Landfill: 15 percent
— Surface impoundment: 3 percent
These results should be interpreted with caution. The large volume of
dilute aqueous wastes which comprise the vast majority of the waste reported
above, distorts the perceived disposition of actual solvent constituents. The
National Survey questionnaire was structured in a manner which limits the
reliability of the data; e.g., generation estimates include possible double
counting and preferentially weight halogenated waste quantities relative to
3-24
-------
nonhalogenated solvent quantities. The waste quantity data itself is not
normally distributed which, combined with erroneous reporting of non-RCRA
wastes and differences between the current and 1981 definitions of RCRA wastes
further limit its reliability. Finally, waste generation and management
practices have changed since 1981 due to regulatory changes, heightened
awareness of hazardous waste issues, and economic factors. A more detailed
discussion of the survey structure and deficiencies can be found in the
8,24
references.
If the inorganic liquid streams are extracted from the National Survey
data, a more accurate picture of waste quantity emerges. Physical
characteristics are summerized by waste category in Table 3.2.2. As shown,
aqueous wastes accounted for only 6.8 percent of known ignitable waste By
volume, whereas it accounted for 80 percent of halogenated and 69 percent of
nonhalogenated solvent wastes. When weighed by these factors, ignitable
wastes account for 63 percent of the nonaqueous waste, while nonhalogenated
solvents account for 25 percent and halogenated solvents 12 percent
(33 percent of solvents).
This distribution agrees well with that reported in other
2 *$8 29 ^0 31
surveys. ' ' * ' Ignitable waste quantities ranged from 55 to
58 percent of total solvent/ignitable' waste generation while halogenated
compounds accounted for 12 to 50 percent of solvent wastes with an average of
30 percent.
o
Total solvent waste generation estimates provided by ES (3,570 MGY)
25
and the Congressional Budget Office (4,577 MGY) appear to adequately
represent the universe of solvent waste generation. More highly concentrated
organic and solid wastes handled through recycle, use as a fuel, land disposal
(excluding deep-well injection) and incineration represent 22 percent and
17 percent, respectively, of solvent waste generation identified in these
surveys.
A third survey which covered a majority, but not all, hazardous waste
generating industries, identified 3.0 billion gallons of solvent waste of
31
which 15 percent was handled in nonwastewater systems. Thus, a total
national solvent waste generation of 3.5 to 4.6 billion gallons with 15 to
20 percent concentrated solvents is probably in the correct range.
3-25
-------
TABLE 3.2.2. PHYSICAL FORM OF WASTE IGNITABLES AND SOLVENTS MANAGED AT TSD FACILITIES (GPY)
Physical form
Inorganic Inorganic Organic Organic
Work category Solido sludges liquids liquids sludges Miscellaneous9 Unknown Total
Ignltables
Totals 11,930,825 2,680,816 26,646,844 161,754,587 7,829,073 181,351,105 88,169,305 480,362,555
Percentage 2.48 0.56 5,55 33.67 1.63 37.75 18.35 —
Halogenated solvents
Totals 1,022,484 18,756,917 314,939,925 33,020,129 23,806,641 474,418 40,946,534 432,967,048
Percentage 0.24 4.53 72.74 7.63 5.50 0.11 9.46 —
Nonhalogenated solvents
Totals 474,862 157,585 371,126,235 137,403,686 31,269,647 188,044 11,871,670 552,491,729
Percentage 0.09 0.03 67.17 24.87 5.66 0.03 2.15 —
All IRD!tables & solvents
Totals 13,428,171 21,595,318 712,713,004 332,178,392 62,905,361 182,013,567 140,987,509 1,465,821,322
Percentage 0.92 1.47 48.62 22.66 4.29 12.42 9.62 —
aHiscellaneoua probably consists mostly ol organic, sludge, and solid wastes.
Source; Engineering Science, Reference ti.
-------
Although individual constituent codes (i.e., U and P RCRA codes) and
mixed codes accounted for roughly 30 percent of the waste streams reported in
the National Survey, they contributed only 5.2 percent of the nonaqueous waste
a
volume (2 percent of total volume). D001 represented 98 percent of the
nonaqueous ignitable waste volume. Nonaqueous halogenated solvents were
dominated by F002 (81 percent) followed by F001 (8 percent), and miscellaneous
codes (11 percent). Nonhalogenated nonaqueous solvent waste volume was
dominated by F003 (68 percent) followed by F005 (20 percent), F004
(0.1 percent) and miscellaneous (12 percent). In decreasing order of
nonaqueous waste volume for all solvents/ignitables we have D001 (441 MGY),
F003 (117 MGY), F002 (69 MGY), F005 (35 MGY), F001 (7 MGY), F004 (0.2 MGY) and
miscellaneous codes (37 MGY).
Individual constituent (i.e., U and P codes) were frequently reported in
the National Survey for wastes which were not off-specification commercial
products, spills, or contaminated containers. Thus, they can be used to
estimate the relative frequency of wastes generated which contain these
compounds. Solvent constituents are ranked in Table 3.2.3 in decreasing order
of frequency as they occurred in the National Survey. Their reported waste
volumes are also provided for nonaqueous wastes. However, due to the small
sample and skewness of the data from which these figures are derived, the
frequency of reporting these constituents is probably more indicative of their
relative importance than their reported waste volumes.
Table 3.2.3 shows a comparison between use of priority solvents
(Section 3.1) and frequency of reporting these constituents in the National
Survey. With the exception of perchloroethylene, agreement between these data
is generally quite good. This is most likely due to the fact that 55 percent
of PERC usage consists of dry cleaning solvent. These firms recycle wastes as
part of their process and small firms would not generate enough waste to be
considered large-quantity generators (Section 3.1). Excluding dry cleaning
use, PERC would drop down to the ninth most-commonly used solvent which would
put it more in line with the National Survey waste generation estimate.
Methylene chloride ranked seventh in both studies, but was second highest
in waste quantity reported in the National Survey. Methylene chloride is
widely used as a paint stripper (40 percent), which would be high in solids
3-27
-------
TABLE 3.2.3. RELATIVE RANKINGS OF SOLVENT WASTE CONSTITUENTS
w
00
Waste constituent
Toluene
Xylene
Acetone
Methyl ethyl ketone
1,1, 1-fr ichloroe thane
Methanol
Methylene chloride
n-Butanol
Ethyl acetate
Trichloroethene
Methyl isobutyl ketone
Chloroform
Tetrachloroethylene
Isobutyl alcohol
Benzene
Cresols
Nitrobenzene
Cyclohexanone
Fluorocarbons
1 , 2-Mehlorobenzene
Compounds Cor which use as
Tetrahydrofuran
Pyridine
Ethyl ethers
Acetonitrile
Carbon disulfide
Frequency of
occurrence
in National
Survey''
1,304
1,074
970
970
880
768
641
4.51
364
352
348
322
248
221
210
NA
HA
108
89
65
a solvent was
163
138
137
126
21
Waste quantity
reported in
National Survey in
(No. of streams)
1,439,163 (175)
714,063 (109)
9,587,153 (131)
404,750 (86)
1,645,496 (117)
410,116 (115)
6,481,266 (122)
54,324 (13)
188,070 (33)
41,755 (42)
26,402 (18)
85,488 (50)
111,515 (74)
35,057 (8)
384,947 (27)
HA
HA
-
-
5,367 (34) "
not determined:
93,156 (57)
-
-
618,369 (18)
292,575 (22)
Ranking by
frequency in
National Survey*
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16*
17d
18
19
20
Ranking
by use as
a solvent'*
2
1
7
5
4
6
7
8
11
10
12
22
3
15
18
16
17
19
9
20
Ranking by
quantity
reported in
Hew Jersey6
1
5
2
7
10
4
3
-
_
9
6
-
11
-
—
-
_
-
8
-
aSource: Engineering Science analysis of the National Survey TSD Questionnaire, Reference 8.
^Source: Section 3,1.
cSource: Survey of Waste Management Practices in New Jersey, Reference 30.
"Estimated ranking.
-------
content and, therefore, not economically recoverable relative to other
solvents. Thus, a significant fraction of this material is probably land
disposed, which would preferentially weight this material in a survey which
was not specifically designed to measure recycling.
30
A 1985 survey of hazardous waste in New Jersey determined spent
solvent waste quantities by constituent type for offsite wastes containing
greater than 10 percent solvent. These are also presented in Table 3.2.3,
along with the National Survey and solvent use data. The New Jersey survey
generally supports the above observations. Methylene chloride was again
ranked high in terms of waste volume. Acetone was also ranked high relative
to its reported use as a solvent. This is possibly due to the fact that
acetone may find wider use as a wash solvent than was indicated in Section 3.1.
Impact of Regulatory and Other Changes on Waste Generation Estimates—
Since 1981, the definitions of small quantity generators and FOOi through
F005 wastes have been and are being revised to include more waste under RCKA.
Spent solvents which resulted from the use of solvent mixtures were not
regulated until April 30, 1985. At this time, the EPA promulgated new
regulations covering solvent blends which originally contained 10 percent or
more of one or more listed solvents (FOOi, F002, F004, and F005) or was a
blend of F003 solvents. On June 30, 1985, the EPA proposed adding
1,1,2-trichloroethane to the F002 listing and benzene, 2-ethoxyethanol and
33
2-nitropropane to F005. Finally, on August 1, 1985, the EPA proposed
lowering the small-quantity generator exclusion limit from 1,000 kg to 100
kg/month. Thus, given these new definitions, the ES estimates would
understate solvent waste generation.
The impact of these regulatory changes on spent solvent generation
regulated under RCRA is unclear because it is not known to what extent
generators made use of the FOOi through F005 solvent mixture loophole. This
loophole is potentially significant since a large percentage of solvents are
used in blends as opposed to technical grade or pure forms* As an indication
of this, over 18 times as much wastes being shipped to Illinois recyclers
consisted of hazardous solvent blends relative to single component waste
3-29
-------
streams. However, available waste characterization data suggests that
many firms reported any waste containing hazardous solvents as RCRA wastes,
regardless of the strict definition as stated in the RCHA Part 261
. . 31
provisions.
The impact on projected waste quantities from changes in waste code
definitions and small quantity generator requirements can be estimated. A
1985 report on small quantity generators (see Table 3.2.4) concluded that
85,923 MTY of spent solvents and 1,863 MTY of solvent still bottoms were
generated by 33,475 and 738 firms, respectively, that produced between 100 and
1,000 kg/month of hazardous waste. Another 17,844 MTY of ignitable wastes
were generated by this group. Total quantities of spent benzene,
1,1,2—trichloroethane, 2—nitropropane, and.2—ethoxyethanol solvents from large
quantity generators were estimated to be 22,940 MTY and their recovery still
33
bottoms were estimated to be 8,770 MTY. The combined additional organic
residues from these sources (137,340 MTY) would add approximately 38 MGY
(1.6 to 2.5 percent) to the estimated 1,534 to 2,380 MGY of concentrated
wastes currently recycled or land disposed.
3.3 WASTE MANAGEMENT PRACTICES
Most of the following discussion relies on National Survey data which was
a
available from ES. This is compared with more recent survey data and other
literature sources to describe current management practices. Characteristics
of RCRA priority solvent wastes are summarized in Table 3.3.1 by management
practice. These data are discussed in more detail in the following
subsections which are devoted to specific waste management methods.
3.3.1 Recycling (Includes Use as a Fuel)
This subsection summarizes quantities of recycled solvent and ignitable
wastes by waste code and constituent type, waste characteristics of recycled
streams and end uses of recycled solvents.
3-30
-------
TABLE 3.2.4. NUMBERS OF SMALL QUANTITY GENERATORS8
AND WASTE QUANTITY BY WASTE STREAMb
Waste type No. of generators Waste quantity (MTY)
Solvent wastes
Spent solvents
Solvent still bottoms
Ignitable wastes
Ignitable paint waste
Formaldehyde wastec
Other ignitable wastes
33,475
738
3,122
2,014
2,873
85,923
1,863
4,872
5,396
7,576
a
'Generators of 100 kg to 1,000 kg of hazardous waste/month.
^Source: Reference 27.
cPotentially ignitable.
3-31
-------
TABLE 3.3.1. CHARACTERISTICS OF WSA SOLVEHT HASTES BY HASTE MANAGEMENT PRACTICE
Total priority solvent concentration
Waste
management
practice
Boiler
Incineration
Landfill
Recovery and
Reclamation
Underground
Injection
W
1 On-site
KJ Treatment
'Tank
Surface
Impoundment
Waste water
Discharge
All Processes1
Minimum
0.40
0.01
0.0013
0.01
7.4 E"6
3.4 E~4
1.6 E"4
3.0 E-4
7.4 E~6
Maximum
80.50
80.00
35.50
98.00
60.00
5.50
4.00
10.78
98.00
Mean
37.41
27.57
7.42
40.04
5.63
0.71
0.30
1.06
14.36
Volume
weighted
mean
5.64
11.44,
10.62
8.57
0.75
0.52
0.30
0,56
1.26
Percentage
Greater
than 10%
solvent
9.44
18.72
36.69
12.22
0.20
0.0
0,0
1.00
1,72
of total waste volume
1% - 10%
solvent
1.58
30.14
31.15
7.99
7.85
8.68
2.21
4.53
5.43
Less
than 1%
solvent
88.97
51.14
32.16
79.79
91.95
91.32
97.79
94.47
92.85
Total
waste
quant ity
MT/YR
231,053
124,802
109,304
1,190,515
1,500,974
2,771,207
4,888,157
6,540,467
17,360,911
Total
number
of wastes
61
157
91
136
61
120
101
221
964
Average
waste stream
volume
MT/YR
3,788
795
1,201
8,754
24,606
23,093
48,398
29,595
18,009
^-Includes 16 Streams (4,432 MT/YR) Dicharged'to Unidentified Private Treatment Works.
Source! Adapted from Industry Studies Data Base. Reference So. 31.
-------
Total quantity of solvent recycling reported in the National Survey is
difficult to verify primarily due to lack of data on onsite activities. Where
these data exist, interpretation of the results is hindered by the ambiguities
surrounding the definitions of RCRA wastes. Whether an onsite waste is
considered a hazardous waste prior to recycling, or whether the recycling
process is considered to be an integral part of the process from which the
waste was derived, is subject to interpretation. For example, solvent
refining, polymerization processes and vegetable oil manufacturers annually
recycle billions of gallons of solvent internally in their processes which are
3
not considered to be hazardous wastes.
ff *} £L
Since the majority of waste solvents are recycled onsite (78 * to
3
92 percent), uncertainties in onsite recycling estimates translate into
large errors in total recycling quantity estimates. In contrast, offsite
recycling is less subject to the ambiguities which surround onsite activities
and has been studied more extensively.
Recycled solvent and ignitable quantities, as reported in the revised
National Survey, are summarized in Table 3.3.2. Recycling estimates for
ignitable, halogenated solvents and nonhalogenated solvents were 63, 62 and
320 MGY, respectively. Of this, 22 percent (123 MGY) was recycled offsite and
30 percent was halogenated solvents. Although no comparative data on onsite
recycling activities was available, offsite recycling has been evaluated by a
number of other sources, all of which involved surveying a distinct population
and extrapolating to national totals. These estimates resulted in higher
8
projected recycled solvent quantities, ranging from 149 to over 240 MGY
[References 9,11,18,25,28 and 30], Thus, it seems probable that the National
Survey underestimated recycled waste quantities.
In general, the National Survey showed good agreement with other waste
surveys in terms of percentage of solvent recycled and the relative quantities
of halogenated versus nonhalogenated solvent recycled. A review of the data
[References 3,8,26,31 and 35] show roughly one fourth of all solvent wastes
and 60 to 65 percent of nonaqueous wastes are currently recycled. Of this,
nearly 30 percent are halogenated solvents (References 3,8,10,11,18,26 and 28).
3-33
-------
TABLE 3.3.2. WASTE RECYCLING REPORTED IN THE REVISED NATIONAL SURVEY
Waste
category
Igni tables
Halogenated
solvents
Nonhalogenated
solvents
All solvents/
ignitables
Waste
code
D001
F001
F002
Other
Total:
F003
F004
F005
Other
Total
Volume
recycled
offsite
(1,000 gal/yr)
26, 140
32,155
7,029
12,784
51,968
16,713
97
13,237
14,964
45,011
123,119
Volume
recycled
onsite
(1,000 gal/yr)
36,664
10,011
95,014
5,200
110,225
182,994
0
83, 804
8,387
275,185
422,074
Percent
recycled
onsite
58.3
23.7
93.1
28.9
68.0
91.6
0
86.4
35.9
85.9
77.4
Total
recycled
waste
(1,000 gal/yr)
62,804
42,166
102,043
17,984
162,193
199,707
97
97,041
23,351
320,196
545,193
Source: DPEA Analysis of the National Survey, Reference 26.
3-34
-------
A breakdown of individual solvent types which are most frequently
recycled is shown in Table 3.3.3, ranked in decreasing order of volume
o
recycled. As shown, the surveys of solvent recyclers by the NASR and the
11 8 •
state of Illinois show good agreement. The National Survey, which
reported the bulk of its solvent as RGRA F001 through F005, does not
correspond as well with these other surveys or with solvent generation
rankings (Table 3.2.3). Thus, it is judged to be a less accurate indicator of
relative volumes of specific solvents recycled.
Table 3.3.4 summarizes the distribution of waste solvents recycled
offsite by the previous use of the solvent. Wastes will be recycled primarily
on the basis of net costs relative to other waste management alternatives.
Although halogenated solvents have generally higher replacement costs relative
to nonhalogenated solvents, the latter were recycled as frequently on a
percentage basis. This results from the fact that nonhalogenated solvent
O -I Q 01
wastes tend to be higher in solvent concentration, 2.5 to 5.0 times
larger in average waste stream volume and are more amenable for use as a fuel
substitute. The National Survey data also show that off-spec commercial
products are frequently recycled, despite small average stream volume, and low
Q
volume streams are more commonly recycled in offsite facilities. These
reflect the economics associated with ease of recycling and economies of
scale, respectively.
Characteristics of recycled solvent and ignitable.wastes have been
summarized in Table 3.3.1. Wastes used as fuel or otherwise recovered tend to
have high solvent concentrations, as would be expected. Nearly all wastes
shipped to offsite recovery facilities are highly concentrated with solvents
except for streams with other recoverable constituents (e.g., solvent
contaminated oil). Recycling of wastewater and solids currently contributes
1 8 31
little to total recycled solvent quantity. '
Table 3.3.5 summarizes characteristics of solvent wastes recycled offsite
by previous use of the solvent. Vapor degreasing wastes tend to be low volume
streams with high oil, low solids and high solvent contents. Conversely, dry
cleaning solvent wastes are high in solids and water content.
3-35
-------
TABLE 3.3.3. RELATIVE RANKING BY OF RECYCLING
Illinois National
NASR^- reprocessors^ Survey^
Xylene
Toluene
Acetone
Methyl ethyl ketone
1,1, 1-trichloroe thane
Methylene chloride
Methanol
Perehloroethylene
Trichloroethylene
Methyl isobutyl ketone
Isobutanol
Freon
N-butyl alcohol
Cyclohexanone
Ethyl acetate
Ethyl benzene
Dichlorobenzene
Chlorobemzene
Carbon sulfide
Ethyl ether
Cresols
Nitrobenzene
Pyridine
1
2
3
4-5
4-5
6
7-8
7-8
9-11
9-11
9-11
12
13-14
13-14
15-17
15-17
15-17
18-23
18-23
18-23
18-23
18-23
18-23
1
2
4
3
5
7
11
9
6
10
13
8
12
-
_
—
_
_
_
-
-
_
*"**
4
6
3
1
2
5
9
8
7
10
17-18
13
17-18
_
—
—
_
. -
-
-
_
_
«M»
•^•Source: Written communication between Emery Hukill,
President of NASR and GCA, May 4, 1981, Reference 11.
^Source: Survey of Illinois Solvent Reprocessors.
1983,., GCA, Reference 11.
^Source: Frequency reported in the National Survey for
recycling in 1981. Engineering Science,
Reference 8.
3-36
-------
TABLE 3.3.4. DISTRIBUTION OF OFFSITE RECYCLED WASTE SOLVENTS BY PREVIOUS USE (%)
CO
w
-4
Previous use
Hash Solvents
(fetal Cleaning
S Vapor Degress ing
Faint Solvent Watte
Dry Cleaning
Ink Solvent
Off-spec Products
Kelt Exchange
Othera
Xylol
44.19
0.00
7.06
0.00
0.13
0.00
0.00
48.66
MEK
63.85
0.00
6.92
0.00
0.76
0.77
0.00
27.70
.TCE
3.18
64.98
0.00
0.00
0.00
0.00
0.10
31.74
1,1,1-TCS
1.57
70.23
0.00
0.93
0.00
0.00
4.00
27.27
Toluol
40.31
0.00
14.99
0.00
0.48
0.89
0.00
43.33
Acetone
39.56
0.00
12.33
0.06
0.90
1.43
0.00
45.72
Hethlene
Chloride
1.75
11.17
10.14
0.31
0.00
0.00
0.33
76.29
PCB
1.48
8.59
5.30
15.61
0.00
0.00
0.00
69.02
MIBK Hethanol
72.32
0.00
4.27
0.00
0.00
0.00
0.00
23.42
50.64
1.45
1.18
2.60
0.00
0.00
5.86
38.26
Isobutanol
0.00
0,00
0.00
0.00
0.00
0.00
0.00
100.00
Butanol
96.33
0.00
0.00
0.00
0.00
0.00
0.00
3.67
Miscellaneous
priority
Freoo solvents
48.11
2.76
0.00
4.31
0.00
0.00
0.00
44.83
47.42
4.24
11.43
0.00
0.00
0.00
0.00
36.92
Potentially
igai table
solvents
38.95
1.97
13.40
0.15
0.16
0.35
1.29
43.72
•Includes wastes categorized as thinner*, blending solvents, and unspecified waste solvents.
Source: Adapted from GCA. Reference No.11
-------
TABLE 3.3.5. CHARACTERISTICS OF WASTE SOLVENTS RECOVERED OFFSITE
w
00
Number of
waste
Previous use streams
Wash Solvents
Vapor Degreasing
& Metal Cleaning
Paint Solvent Waste
Dry Cleaning
Ink Solvent Waste
Off- spec Product
Heat Exchanges Haste
Other3
150
112
41
4
2
2
5
195
Average
waste
stream
volume
(MT/YR)
134
31
123
71
52
102
33
127
Average waste composition (%)
Halogenated Nonhalogenated
priority priority
solvents solvents
3
85
49
53
0
0
6
16
.66
.28
.58
.85
.0
.0
.13
.51
55
1
6
1
72
62
5
35
.79
.43
.28
.75
.39
.27
.99
.76
Potentially
ignitable
solvent
21.57
4.75
32.22
4.79
12.69
18.68
81.56
18.00
Water
6.53
0.06
1.04
11.13
7.54
0.0
1.15
6.59
Oil
1.82
8,17
0.43
0.0
0.0
0.0
1.41
5.75
Solids
10.61
0.31
10.45
28.49
7.39
19.05
3.77
17.39
Solvent
81.02
91.46
88.08
60.39
85.08
80.95
93.68
70.27
alncludes wastes categorized as thinners, blending solvents, or unspecified waste solvents. .
-------
Relative to solvent wastes managed through other treatment processes,
31
recycled wastes frequently contain metals; e.g., Cr, Mi, Hg and Pb.
However, metals are found most often in halogenated wastes and rarely in
31
wastes which are used as fuel supplements. The latter also tend to have
fewer solvent constituents relative to waste managed in other processes.
The primary end uses of recycled solvents and ignitables include use as a
fuel, direct reuse as a feedstock and recovery for reuse. Onsite recycling
practices, as summerized in the National Survey, are presented in
Table 3.3.6. As shown, most solvent wastes are recycled as a reclaiment
whereas ignitables are predominately used as fuel. Only 1 percent of recycled
halogenated wastes are used as fuel versus 11 percent for nonhalogenated
solvents. An independent survey of wastes burned as fuel performed by WESTAT
in 1983 showed a similar distribution between these waste categories.
Nearly three-fourths of waste derived fuel was ignitables, one fourth was
nonhalogenated solvents and less than 2 percent was halogenated solvents.
31
Finally, the ISDB provided a further breakdown of recycling practices
(excluding use as a fuel) for solvent wastes by residual category
(Table 3.3.7). Major residual categories included distillation residues
(51 percent) and light ends (38 percent). The majority of these residuals
(78 percent) were recovered for sales, whereas reuse in a process accounted
for the remaining 22 percent of known reuse methods. Use as a fuel was not
identified by residual category. If this method of reuse is included, the
distribution between handling methods becomes 66 percent reclaimed (i.e., for
sales), 15 percent used as fuel, and 19 percent reused as a feedstock.
Use as a fuel accounted for only 0.2 percent of halogenated and 17 percent of
nonhalogenated solvent recovery.
3.3.2 Incineration
Quantities of solvents and ignitables incinerated are summarized in
a
Table 3.3.8 by management method as provided by the National Survey. These
represent maximum values, since the total quantity of wastes which reported
3-39
-------
TABLE 3.3.6. ONSITE RECYCLING PRACTICES FOR SOLVENT AND IGNITABLE WASTES3
Recycling method
Recycled or reclaimed
as a feedstock
Used as fuel or fuel
supplement
Recycled in a manner
constituting disposal
Recycled as a reclaiment
Other recycle
Percentage
Ualogenated
Igni tables
22
49
1
24
4
of onsite quantity recycled
Nonhalogenated
solvents
1
1
0
96
2
Solvents
5
11
6
83
1
All solvent
ignitable
wastes
10
24
0
64
2
aNote: Total presented is a maximum value that includes double-counting of
a small number of streams. Refer to Table 3.2.1 for, actual totals.
Source: Engineering Science, Reference 8.
3-40
-------
TABLE 3.3.7. RECOVERY PRACTICES FOR RCRA HAZARDOUS WASTES BY RESIDUAL CATEGORY (MT/YR)
Residual category
Spent solvents
Distillation residues
Miscellaneous heavy ends
Condensible light ends
Noncondensible light ends
Off-specification products
Total Residuals
Percent of known
recovery /reuse method
Reuse in
Unspecified in same
recovery process
16,962
62,222
15,591
8,825
0
307
103,907
~
37,398
210
0
515
0
7,619
45,742
4.2
Reuse in
in different
process
1,161
23,122
41,274
6,166
119,545
4,318
195,586
18.2
Percent of
Total residual volume Percent of
Recovery recovery undergoing all solvent
for sales and reuse recovery recovery
137 55,658 67.0
513,636 599,190 78.9
2,784 59,649 50.5
228,214 243,720 86.6
90,000 209,545 61.4
500 12,744 4.5
835,271 1,180,506
77.6 ~ ~
4.7
50.8
5.1
20.6
17.8
1.1
-
—
Source: Adapted from ISDB, Reference No. 31.
-------
TABLE 3.3.8. TREATMENT OF SOLVENT AND IGNITABLE WASTES BY INCINERATION1
Treatment
process
Waste category
Ignitables Halogenated Nonhalogenated ; Total
Incineration only 40,140,456 46,205,850
Tank and incineration 47,140,000 7,910,000
Incineration and other
Tank and surface
impoundment and
incineration
Surface impoundment
and incineration
Total:
3,171
6,134,147
408,049
93,825,823 54,115,850
34,149,611
69,841,463
1,486,251
100,488
120,495,917
124,891,463
1,489,422
6,134,147
308,537
105,577,813 253,519,486
Source: Engineering Science, Reference 8.
3-42
-------
incineration along with other processes was included in this total. Relative
to other wastes, a disproportionately high fraction of halogenated wastes are
. . , 8,31
incinerated.
Physical profiles of incinerated wastes were also provided.in the
National Survey as summarized in Table 3.3.9. The major waste categories
incinerated were organic liquids (59 percent),, followed by inorganic liquids
(37 percent). Very little solid or sludge (1.8 percent) was reportedly
incinerated, possibly as a result of coding interpretations which categorized
Q
still bottoms and other slurries as organic liquids. Typical incinerated
wastes containing priority solvents tend to have low volume, high solvent
content (Table 3.3.1), low frequency of occurrence of heavy metals and a low
number of organic solvents per stream relative to solvents handled in other
31
waste management methods.
Finally, waste constituents incinerated have been reported by the Mitre
37
Corporation as the result of a 1983 industry wide survey. These data are
presented in Table 3.3.10 in decreasing order by volume. These compounds
include all the priority solvents affected by the land disposal ban.
Halogenated compounds accounted for only 14 percent of the volume which
suggests that they are present in lower concentrations or lower total waste
31
volume. Both of these possibilities are supported by other data.
3.3.3 Treatment Methods
Treatment methods summarized by ES were only provided for ignitable
wastes, since solvent treatment data was judged to be insufficiently validated
Q
to provide useable information. Table 3.3.11 summarizes the most commonly
used treatment methods in tanks for ignitables by frequency of use. Physical
separation methods accounted for 64 percent of treatment methods with chemical
treatment practiced on 28 percent of the waste streams. Miscellaneous
treatment processes accounted for the remaining 8 percent.
Land Disposal—
Solvent and ignitable wastes managed by land disposal are summarized in
1 H
Table 3.3.12 by waste category and management practice. Excluding two
large waste streams which are managed through deep-well injection yields the
3-43
-------
TABLE 3.3.9. PHYSICAL FORMS OF INCINERATED SOLVENTS AND IGNITABLES (GALLONS/YEAR)
Physical
4?- solvents
Inorganic Inorganic Organic Organic
Waste type Solids sludges liquids liquids sludges Miscellaneous Unknown Total
Ignltables 3,005,701 — 12,893,269 24,515,214 1,654,428 2,659,632 49,097,579 93,825,823
Halogenated ~ — 39,878,053 13,491,806 590,976 — 155,015 54,115,850
Nonhalogenated 210 — 21,793,585 82,087,571 347,497 746 1,348,210 105,577,819
solvents
Total! 3,005,911 — 74,564,907 120,094,591 2,592,901 2,660,378 50,600,804 253,519,492
Percent of known 1.5 0.0 36.7 59.2 1.3 1.3 0.0 100.0
waste forms
Source: Engineering Science, Reference 8.
-------
TABLE 3.3.10. PRIORITY SOLVENTS INCINERATED IN 1981
Constituent
Number of
waste streams
containing
constituent
Amount of
constituent
incinerated
(million gallons)
Methanol
n— Butyl alcohol
Toluene
Acetone
Xylene
Methyl ethyl ketone
Ethyl acetate
Tetrachloroethylene
Ch lorobenzene
Cresols
Methylene chloride
1 , 1 , 1-Tr ichloroethane
Methyl isobutyl ketone
Isobutanol
Pyridine
Tr ich loroethy lene
Carbon tetrachloride
Tr ich loro f luorome thane
Cy c lohexanone
Ethylbenzene
Ethyl ether
1 , 2-D ich lorobenzene
Carbon disulfide
Nitrobenzene
1,1,2-Trichloro - 1,2,2-Trif luoroethane
95
9
10.3
80
78
54
24
19
21
8
26
23
10
9
9
15
9
2
4
6
2
2
3
3
2
44.4
32.3
17.3
17.2
15.2
14.0
9.62
6.74
6.18
5.92
5.84
5.05
4.68
4.50
4.38
3.69
0.547
0.401
0.344
0.264
0.248
0.240
0.128
0.0057
0.0008
Source: Mitre Corporation, Reference No. 37.
3-45
-------
TABLE 3.3.11. MOST FREQUENTLY PRACTICED TREATMENT TECHNIQUES
IN TANKS FOR IGNITABLE WASTES1
Technique Frequency Percent
Physical methods
1. Decanting 261 . 21.8
2. Sedimentation 180 15.0
3. Blending 172 14.4
4. Filtration 56 4.7
5. Clarification 27 2.2
6. Flotation 22 1.8
7. Solvent recovery 21 1.7
8. Other physical separation 21 1.7
9. Other physical removal 13 1.1
Subtotal: 773 64.4
Chemical methods
1. Chemical oxidation 161 13.4
2. Neutralization 68 5.7
3. Chemical fixation 64 5.3
4. Chemical reduction 34 2.8
5. Activated carbon 13 1.1
Subtotal: 340 28.3
Note= 7.3 percent of waste streams reported miscellaneous
treatment methods.
Source: Engineering Science, Reference 8.
3-46
-------
following distribution: deep-well injection - 43 percent, landfill -
42 percent, surface impoundment - 9 percent, and other - 7 percent.
Physical profiles of land disposed solvent and ignitable wastes were also
Q
provided in the National Survey. The majority of solids and sludges are
landfilled (75 percent) with the remainder disposed in surface impoundments
(18 percent) and land applied (7 percent). Inorganic liquids account for only
0.4 percent of landfilled waste, thus the distribution between ignitables
(25 percent), halogenated (52 percent), and nonhalogenated solvents
(22 percent) provided in Table 3.3.12 is representative of actual constituent
quantities disposed* In general, implementation of the solvent land disposal
ban will have the most severe impact on generators of spent halogenated
31
solvents. This data is supported by results obtained from the ISDB which
showed 84 percent of landfilled waste solvent volume containing halogenated
30
solvents (versus 70 percent in the National Survey).
As shown in Table 3.3.1, landfilled wastes tend to have higher solvent
concentration (10,6 percent) and lower waste stream volume relative to solvent
wastes handled in other waste management practices. They also tend to be more
31
frequently contaminated with heavy metals, particularly Cr and Ni. Wastes
handled in tanks and impoundments tend to be very low in solvent concentration
and contain multicomponent mixtures. These wastes have little potential for
solvent recovery.
3-47
-------
TABLE 3.3.12. QUANTITIES OF SOLVENT WASTE STREAMS CURRENTLY
MANAGED BY LAND DISPOSAL3 (MGY)
Manageme nt
practice
Deep well injection
Surface impoundment
( disposal only)
Landfill
Land application
Surface impoundment
Waste category
Igni tables Halogenated Nonhalogenated
14.2 290b 12.3
5.3 6.1 0.9
15,1 31.3 13.3
3.2 - -
5.3 •
Total
316. 5b
12.3
59.7
3.2
5.3
and land application
Surface impoundment
and landfill
Total:
Total after removal
of two wastewater
streams currently
deep-well injected
43.1
43.1
0.8
328.2b
73.2
0.9
27.4
27.4
1.7
398.7b
143.7
aNote: These data gave been recently revised by the EPA (see Table 3.2.1)i
Therefore values presented here should only be used to provide an approximate
distributuion of wastes between management practices.
^Estimates include two wastewater streams totalling 255 million gallons,
Source: Engineering Science, Reference 8.
3-48
-------
REFERENCES
1. Chemical Marketing Reporter. Chemical Profiles. Schnell Publishing
Company. 1982 through 1986.
2. Hoogheem, T.J. et al, Monsanto Research Corp. Source Assessment; Solvent
Evaporation-Degreasing Operations. EPA-600/2-79-019£, U.S. EPA 1ERL,
Cincinnati, OH, August 1979.
3. Horsak, R.D., et al., Pace Company Consultants and Engineers, Inc.
Solvent Recovery in the United States: 1980-1990. Houston, TX.
Prepared for Harding Lawson Associates, January 1983.
4. U.S. EPA. Source Assessment Document for Perchloroethylene.
PB85-233418, 1985.
5. Mansville Chemical Products Corporation, Cortland, N.Y. Chemical Products
Synopsis. 1982.
6. Kirk-Othmer Encyclopedia of Chemical Technology. John Wiley & Sons,
N.Y., N.Y. Third Edition. 1978. :
7. Goodwin, Don R. Organic Solvent Cleaners — Background Information for
Proposed Standards. U.S. EPA OAQPS EPA-450/2-78-0045a - October 1979.
8. Engineering-Science. Supplemental Report on the Technical Assessment of
Treatment Alternatives for Waste Solvents. Washington, D.C.: U.S.
Environmental Protection Agency. 1985.
9. Minnesota Waste Management Board - Hazardous Waste Management Report.
MWMB Crystal, MN. December, 1983.
10. U.S. EPA. Industrial Assessment of Hazardous Waste Practices: Paint and
Allied Products Industry, Contract Solvent Reclaiming Operations and
Factory Application of Coatings. EPA-SW~119c, U.S. EPA/OSW, 1976.
11. Hobbs, B. and R.R. Hall, GCA Technology Division. Study of Solvent
Reprocessors. Bedford, MA. 1PA Contract No. 68-01-5960 (Draft), U.S.
EPA Office of Chemical Control, January 1982.
12. Cheng, S.C., et al,, Monsanto Research Corp. Alternative Treatment of
Organic Solvents and Sludges for Metal Finishing Operations,
EPA-600/2-83-094, U.S. EPA IERL, Cincinnati, OH, September 1983.
13. Higgins, T.E. CH2M/Hill, Reston, VA. Industrial Processes to Reduce
Generation of Hazardous Waste at DOD Facilities Phase 2 Report Evaluation
of 18 Case Studies. July 15, 1985.
14. Nelson, W.L., Naval Energy and Environmental Support Activity (NEESA).
NEESA In-House Solvent Reclamation. Port Hueneme, CA. NEESA 20.3-012,
October 1984.
3-49
-------
15. Fisher, William and Cindi Busier. International Fabricate Institute.
Source Reduction and Small Generator Considerations in the Dry Cleaning
Industry. Presented at the Massachusetts Hazardous Waste Source
Reduction Conference. June 14, 1984. Sponsored by MA Bureau of Solid
Waste. Boston, MA.
16. Ehrenfeld, J. and Jeffrey Bass, Arthur D. Little, Inc. Evaluation of
Remedial Action Unit Operations at Hazardous Waste Disposal Sites.
Cambridge, MA, Noyes Publication.
17. Ruder, Eric, et al., ABT Associates. National Samll Quantity Generator
Survey. EPA-530/SW-85/004, U.S. EPA, Washington, D.C., February 1985.
18. Radimsky, Jan, et al. Recycling and/or Treatment Capacity for Hazardous
Wastes Containing Halogenated Organic Compounds. State of California,
Department of Health, September 1984.
19. International Agency for Research on CAncer World Health Organization
IARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals
to Humans. Some Halogenated Compounds. Volume 20 October, 1979.
20. Sittig, Marshal. Handbook of Toxic and Hazardous Chemicals. Noyes
Publications. Park Ridge, N.J. 1981.
21. Stanford Research Institute. SRI Chemical Profiles. 1985. Stanford, CA.
22. Verschueren, Kursel. Handbook of Environmental Data on Organic
Chemicals. Van Nostrand Reinhold Company. N.Y., N.Y. Second Edition
1983.
23. Federal Register. Proposed Rule: Land Disposal Restrictions. Vol 51 #9
(51FR1602). 14 January 1986.
24. Deitz, S., et al., Westat, Inc. National Survey of Hazardous Waste
Generators and Treatment, Storage, and Disposal Facilities Regulated
Under RCRA in 1981. Rockville, MD. U.S. EPA/OSW, April 1984.
25. U.S. Congressional Budget Office. Hazardous Waste Management - Recent
Changes and Policy Alternatives. CBO Congress of the United States.
May 1985.
26. DPRA Inc. Analysis of the RIA Mail Survey (National Survey) and Duns
Marketing Data. Data Request #M850415W. 10 June 1985.
27. U.S. EPA Office of Air Quality Planning and Standards, Assessment of
Incineration as a Treatment Method for Liquid Organic Hazardous Wastes.
Background Report III. Assessment of the Commercial Hazardous Waste
Incineration Market. U.S. EPA March 1985.
28. New England Congressional Institute. Hazardous Waste Generation and
Management in New England. Washington, D.C., February 1986.
3-50
-------
29. Kohl, J., P. Moses, and B. friplett, North Carolina State University,
Industrial Extension Service. Managing and Recycling Solvents: North
Carolina Practices, Facilities, and Regulations, 1984.
30. White, R.E., T. Busmann, J.J. Cudahy, et al., IT Enviroscience, Inc. New
Jersey Industrial Waste Study (Waste Projection and Treatment).
Knoxville, TN. EPA-600/6-85/003, May 1985.
31. Science Applications International Corp. Industries Studies Data Base,
1986.
32. Federal Register. 40 CFR Part 261 (50FR533152) 31 December 1985.
33. Federal Register. 40 CFR Part 261 (50FR30908) 30 July 1985.
34. Federal Register. 40 CFR Part 261 (50FR31278) 1 August 1985.
35. Battelle Columbus Labs. Hazardous Waste Management Needs Assessment.
Ohio. DE84901754/WEP, June 1984.
36. Kerwin, John. Westat Inc. Data Base for the Survey of Handlers and
Burners of Used or Waste Oil and Waste-Derived Fuel Material. Prepared
for the U.S. EPA, Office of Solid Waste, Economics and Policy Analysis
Branch 1985.
37. Mitre Corporation. Composition of Hazardous Waste Streams Currently
Incinerated. Prepared for U.S. EPA, Office of Solid Waste. 1983.
3-51
-------
-------
SECTION 4.0
COMMERCIAL OFFSITE RECYCLING, TREATMENT, AND DISPOSAL CAPACITY
This section summarizes commercial recycling, treatment, and disposal
capacity available for handling solvent wastes which are scheduled to be
banned from land disposal. The commercial solvent recycling industry is
discussed to provide insight into its processing capabilities and current
practices. This is followed by a summary of available data on recycling,
incineration, and waste fuel burning capacity. These represent the bulk of
available land disposal alternatives for nonaqueous wastes.
4.1 COMMERCIAL SOLVENT RECYCLING INDUSTRY
The current commercial solvent recycling industry consists of at least
135 firms with 243 facilities nationwide. Of these, 13 percent accept only
halogenated solvents and 6 percent accept only nonhalogenated solvents. The
latter tend to be waste oil dealers and fuel blenders, while the former are
generally chemical distributers, equipment vendors (e.g., degreasers), or
small firms which only process expensive solvents due to economic
considerations. Overall, approximately 25 to 30 percent of the solvents
handled are halogenated (see Section 3.3.1). The solvent reclaiming industry
is represented by the National Association of Solvent Recyclers (NASR) which
is comprised of 43 solvent reclaiming firms, accounting for approximately
2
70 percent of offsite recycling capacity.
Large firms tend to serve regional markets and are capable of producing
high-purity solvents through the use of distillation, fractionation, and other
specialized recovery techniques. Their market edge is enhanced by laboratory
facilities, skilled operators, and a good understanding of regulations. These
firms are often involved in other hazardous waste operations (e.g.,
transportation, treatment, storage, or disposal facilities) and represent
3
approximately 3 percent of the solvent reclaiming facilities.
4-1
-------
Small firms, representing approximately 70 percent of the solvent
3
reclaiming facilities, have limited recycling capabilities. Roughly half
of these facilities operate small distillation units while the remainder rely
on filtration, precipitation and decanting to recover or blend solvents for
3
use as fuel supplements. Medium-sized firms exhibit capabilities ranging
between those of the other size classes, but they lack the distribution
networks of the large firms.
Several sources of information regarding the solvent recycling industry
have been identified which provide information on the types of wastes handled,
treatment processes employed, costs, capacity utilization, and residual
handling practices.
GCA and Metcalf & Eddy (M&E) surveyed and profiled 22 recycling firms in
4.
1985 for the EPA's Office of Research and Development. An earlier survey
was performed by the NASR in 1982, which resulted in 31 telephone responses
2
from its member firms. Finally, the State of California Department of
Public Health performed a statewide survey of solvent reclaiming operations in
1984, identifying 11 facilities. Data generated from these surveys on
processing capabilities and residual handling practices are summarized in
Table 4.1.1. The GCA and NASR surveys targeted larger facilities which
represent the bulk of solvent management capacity, particularly for wastes
which are less easily recycled.
Several conclusions from these surveys can be drawn:
• The use of some form of distillation is universal among larger
firms. The only facilities without some type of evaporative or
distillation technology were a blending facility and a facility
using destructive wet air oxidation;
• A majority of the facilities (70 percent) use bottom residuals
(particularly nonhalogenated) as fuel and restrict processing to
maintain a pumpable liquid;
• Recovery ranged from 74 to 98 percent at the four sites which
reported data; and
• Drying recovered solvents is fairly common although this practice,
like decanting and filtration, was rarely reported in the NASR
survey.
4-2
-------
TABLE 4.1.1. SUMMARY OF COMMERCIAL SOLVENT*RECYCLING SURVEYS
Process type
Simple distillation
Fractional ion
(including
packed column)
Thin-film evaporation
Steam stripping
Drying
Solvent extraction
Use as fuel
Landfill
Incineration
Total facilities
in survey
Number
California
survey*
11
3
5
5
9
0
5
6
0
11
of facilities using process
GCA/M&E
survey2
22
14
25
6
2
4
14
HA
NA
22
NASR
survey^
11
5
13
7
6
2
29
23
10
39
Total
facilities
using process
44
22
43
18
17
6
48
29
10
72
Percent of
facilities
using
process
61
31
60
25
24
8
67
58
20
_
Comments
Three flash units.
Seven vacuum units.
Some equipment with
surge columns.
Of nine specified,
five were vacuum.
Seven molecular sieve,
five calcium chloride,
and one each for caustic
extraction, ionic resin,
and drum dryer.
Of 30 which -specified
use: 67% cement kiln,
16Z boilers, 17% steel
furnaces.
^Source: State of California Survey, Reference 5.
2Source: GCA/Metcalf & Eddy Survey, Reference 4.
Source: Engineering Science/HASR Survey, Reference 2.
4-3
-------
Wastes exhibiting certain characteristics are sometimes not accepted by
recyclers» including:
• Low flash point materials - wastes with a flash point of less than
100°F.
* High solids content wastes - solids contents of 30 to 50 percent
(60 percent for steam injection) are limiting values reported by
reclaimers. Others specify minimum recoverable solvent levels of
50 to 60 percent."'' However, one facility was identified which
ground high solid wastes to £800 ym, and suspended this waste in a
solvent/oil blend for use as a fuel.'*
* Heavy metals and cyanides were also restricted at some facilities,
but concentration limits were not provided.
Other restrictions adopted by same solvent reclaimers include minimum waste
volumes and requirements that the recovered, and sometimes, the residual
bojttoms product are returned to the supplier.
Wastes accepted for recycling tend to be highly concentrated with
solvents. Analysis of waste characterization data provided by Illinois
reprocessors showed two-thirds of the total waste quantity recycled contained
o
over 70 percent solvent. Less than 3 percent contained 10 percent or less
solvent and 41 percent contained over 90 percent solvent (see
g
Figure 4.1.1). Wastes with very low solvent content (less than 30 percent)
tended to be wastewater washes and line rinses. Wastes with solvent contents
ranging from 30 to 60 percent were dominated by oily wastes such as degreasing
solvent distillation residues. Highly concentrated wastes were contaminated
solvents which were not recovered to a significant extent by the generators.
Solvent recovery rates reported in the literature average around
75 percent * * with values ranging from less than 60 percent to
9
99 percent. Safety Kleen, which recovered 23.6 million gallons of spent
parts cleaning solvent (D001) in 1982, reported average recovery rates of
92 percent with an additional 7 percent being blended as fuel. However, spent
9
chlorinated immersion cleaning solvent recovery averaged only 74 percent.
The extent of solvent recovery is determined either by equipment
processing capability or constraints imposed by the end-use of the bottoms
product. The NASR survey showed that at least 62 percent of solvent bottoms
2
from commercial recovery facilities are used as supplemental fuel. These
4-4
-------
100
TOTAL SOLVENT CONCENTRATION, percent by WEIGHT
Figure 4.1.1.
Solvent wastes recycled by commercial
reprocessors in Illinois.
Source: Reference No. 8.
4-5
-------
materials are deliberately kept in a pumpable state for ease in handling, as
are wastes which are treated through liquid injection incineration.
Restrictions on wastes accepted for use as a fuel blend include maximum solids
content (30 to 50 percent maximum) and chlorine content (0 to 10 percent after
blending with typical limits of 3 to 5 percent). One case was identified
where chlorine contents of up to 35 percent were accepted for a kiln producing
4
low alkali cement.
Solvent bottoms destined for rotary kiln incineration or landfills, in
particular halogenated solvents, are processed to the extent allowable in the
facility's recovery equipment. Simple coil-still distillation units are least
able to process high solids wastes, whereas steam injection stills and
thin-film evaporators can operate at higher solids levels (see Section 7.0).
Certain distillation recovery units with lined boilers have been reported to
reduce solvent concentrations in the bottoms product to less than
1 percent. One facility used a double-drum dryer to reduce nonflammable
4
solvent bottoms 'to 99.9 percent solid. In the literature, facilities were
reported to add nonvolatile liquid additives to distillation uiaits to keep
bottoms in a fluid state as solvent is recovered.
Other uses for chlorinated bottoms products that have been reported
ide use as an asphalt c
(concrete block extender).
include use as an asphalt extender and incorporation into a building product
8
Commercial Recycling Costs—
Commercial solvent recycling costs are summarized in Table 4.1.2. Toll
arrangement costs typically vary from $0.20 to $1.70/gallon. Sell-back prices
are 70 to 90 percent of virgin product prices, whereas open-market sales range
from 50 to 90 percent. These are lower due to increased uncertainty regarding
the presence of undesirable contaminants. Disposal costs in cement kilns and
light aggregate manufacturers range from revenue of $0.05 to a cost of
$0.35/gallon ' ' with an average cost of approximately $0.20/gallon.
Incineration costs are highly dependent on physical form of the waste, but
typically range from $0.30 to $1.25/gallon for liquids and $1.30 to $4.20 for
solids. More details on costs are provided in Sections 7.0 through 12.0
for specific technologies.
4-6
-------
TABLE 4.1.2. COST FOR SOLVENT RECOVERY AT COMMERCIAL FACILITIES
*>
I
Source
(Ref. No.)
3
5
13
6
8
7
5
11
4
12
Year
1983
1984
1983
1976
1982
1982
1984
1984
1985
loll Sell-back
Percent arrangement arrangement Open market
solid (I/gallon) (% of virgin price) (% of virgin price)
5 0.60 80 - 90 50 - 90
10 0.80
15 1.10
20 1.70
0.75
0.15 - 1.50
0.20 - 10.00 80-90 50 - 90
65
0.50 - 1.00
(0.70 average)
0.75 - 1.60
0.10 - 1.55
0.40 - 1.55 80 - 90
70 - 80
-------
Costs for offsite reprocessing of solvents vary as a function of the
waste volume, percent recycleable material (including solvent, metals, oil,
etc.), disposal method required for the residuals, transportation mode and
distance, and the selling/purchasing arrangement. In "toll" arrangements,
solvents are typically segregated,, recycled in a batch mode, and returned to
the generator for a fee based on the recycler's internal costs and profit
margin. Sell-back and open-market arrangements involve purchase of solvents
from generators at a nominal fee and sale of reprocessed solvent. Sales price
is some fraction of the market value of virgin material which is dependent on
product purity and other factors. Toll arrangements are common among
generators using solvent blends in applications which can tolerate some degree
of contamination. Conversely, open-market arrangements are used by firms such
as electronics manufacturers which prefer to purchase virgin product.
Transportation costs vary most directly with distance, but also with mode
of transport (e.g., ship, rail, truck), containment method (e.g., tank, drum,
bulk), special shipping requirements (e.g., ignitable wastes, manifests), and
waste volume. Typical one-way transportation costs by truck are summarized in
Table 4.1.3.
TABLE 4.1.3.
WASTE SOLVENT TRANSPORTATION COSTS
(COSTS = ^/GALLON/MILE)
Source
(Ref. No.)
3
13
11
Year
1983
1983
1984
Drum Bulk Not specified
0.09 - 0.13 0.07 - 0.10 ~
0.06 - 0.010
0.06 - O.U8
4-8
-------
4.2 AVAILABLE BICYCLING T1EATMENT AND DISPOSAL CAPACITY
In order to establish effective dates for enactment of the land disposal
ban, the EPA was required to assess available alternative recycling, treatment
and disposal capacity. The results of this analysis were presented with the
January 14, 1986 proposed rule for the land disposal restrictions as
summarized below.
The EPA assumed 1,202 million gallons/year (MGY) of solvent wastes
currently land disposed would be subject to the land disposal provisions.
This excludes wastes which are currently deep-well injected (exempt from
regulation until August 8, 1988), but includes wastes which are treated or
stored in impoundments. SPA evaluated the economic feasibility of
impoundments meeting the regulatory requirements which permit them to continue
operating under the ban (RCRA Section 3005(j)(ll)(A) and (B)). Based on this
analysis, EPA determined that 80 percent of the volume currently treated will
meet the exemption requirements leaving 185 MGY requiring alternative
treatment. The remaining waste which will require alternative treatment or
recycling capacity is 20.6 MGY of organic liquid, 10.4 MGY organic sludges and
solids, and 6.7 MGY of inorganic sludges and solids as summarized in
Table 4.2.1.
TABLE 4.2.1. SOLVENT WASTE QUANTITY REQUIRING ALTERNATIVE TREATMENT (MGY)
Still bottoms
Currently Small from recovery
land quantity of banned
Solvent waste type disposed generators organic liquid Total
Inorganic aqueous wastes
Inorganic sludges and solids
Organic liquids
Organic sludges and solids
Total :
185.0
6.7
14.6
7.3
213.6
-
-
6.0
1.9
7.9
185.0
6.7
20.6
1.2 10.4
1.2 222.7
Source: U.S. EPA, Reference 14.
4-9
-------
The EPA assumed all aqueous wastes would undergo some form of wastewater
treatment. Fifty percent of halogenated and one-third of nonhalogenated
liquids would undergo some form of distillation generating 1.2 MGY of still
bottoms (14 percent of original waste volume). These bottoms and the
remaining waste would undergo incineration. Use as a fuel was not considered
as an available alternative disposal practice due to uncertainty regarding the
impact of waste-derived fuel regulations which are currently being developed.
Also, due to lack.of capacity data, other treatment options were not
considered.
EPA assumed all wastes would require offsite recycling), treatment or
disposal due to lack of information on available capacity at solvent waste
generators. Unused offsite distillation capacity was taken from the NASR
2
survey (224 MGY) and tank capacity for treating wastewater was derived from
the National Survey ( 112 MGY). Incineration capacity was estimated to be
25.6 MGY, based on design capacity data and an assumed current utilization
14
rate of 80 percent.
Capacity requirements and unused capacity data are summarized in
Table 4.2.2. As shown, there is not sufficient commercial capacity to
incinerate inorganic sludges and solids or to treat aqueous wastes. Thus, EPA
proposed a 2-year extension of the land -disposal ban for these materials to
ensure that sufficient capacity will be available.
It is possible that the quantity of recoverable solvent in wastes which
are currently land disposed is less than that estimated by EPA. EPA assumed
42 percent of currently land-disposed organic liquids would be recoverable,
yielding a bottoms product of only 14 percent of the original waste volume.
Analysis of waste characterization data shows only 37 percent of
landfilled waste with a total solvent concentration exceeding 10 percent by
weight. Concentrations ranged from near 0 to 36 percent total solvents with a
volume weighted mean of 11 percent. Thus, even if all the solvent could be
recovered, the bottoms product would be nearly 90 percent of the original
waste volume. If these data are assumed to accurately represent
landfilled solvent wastes, the resulting incineration demand would become
28.7 MGY, which exceeds available capacity by more than 3 MGY. Other
4-10
-------
TABLE 4.2.2. ANNUAL TREATMENT AM) RECOVERY CAPACITY DEMAND AND AVAILABILITY
(MILLION GALLONS/YEAR)
Inorganic wastes
Inorganic wastes:
Solvent-water mixtures
Sludges and solids
Organic wastes:
Halogenated liquids
Nonhalogenated liquids
Halogenated sludges and solids
Nonhalogenated sludges and solids
Total capacity demand
Available capacity
Capacity shortfall
Wastewater Incineration Incineration
treatment Distillation of organics of inorganics
185
6.7
5.4 5.5
3.2 6.5
7.2
3.2
185 8.6 22.4 6.7
112 225 25.6 0
73 0 0 6.7
Source: U.S. EPA, Reference 14.
-------
5 12
studies ' show higher fractions of recoverable solvent in land disposed
waste but, with current recovery practices, the bottoms product would still
represent closer to 50 percent of the original waste volume.
Additional factors which should be considered include the geographical
distribution of available capacity and the physical form of the waste. WESTAT
indicated that four out of the ten EPA regions have severely restricted
commercial treatment capacity. Also, much of the required capacity is
going to be needed to incinerate solids and sludges. This will put further
strains on available local capacity., since these types of equipment (e.g.,
rotary kilns, fixed hearth) represent a modest fraction (20 percent) of total
hazardous waste incinerators in the country.
Capacity for liquid wastes is expected to be less of a problem since more
wastes are ammenable to recovery and alternative treatment practices.
However, overall capacity utilization at commercial facilities may exceed the
80 percent figure used in EPA's determination of available capacity, and
demand could increase in the short-term if boiler regulations force burners of
waste-derived fuel to find alternative disposal methods.
Finally, it is unclear how waste solvent land disposal quantities have
changed since 1981. Overall, landfilled waste quantity has increased
40 percent from 1981 to 1984 at eight facilities which, in 1981, accounted for
80 percent of the landfill market. However, this growth rate has been
outpaced by resource recovery and incineration which, combined with low
1 3
growth in solvent demand, ' suggests that land disposed solvent quantities
have increased modestly at best.
Despite the above considerations, a capacity shortage is not likely to
result for two primary reasons. First, the quantity of waste requiring
alternative means of disposal is likely to decline in response to the land
disposal ban. Increased waste minimization and use of technologies yielding
higher solvent recovery rates will be implemented as the next lowest cost
alternative to land disposal. These practices will essentially be forced,
since there is currently no economical, commercially demonstrated, means to
solidify or encapsulate wastes containing large quantities of solvent which do
not result in significant releases through leachate generation. Secondly,
4-12
-------
required incineration capacity is also likely to be lower since other
treatment and disposal alternatives will be utilized; e.g., use as fuel, wet
air oxidation, etc. These factors are discussed below.
Increased product and process substitution and other waste minimization
activities are likely to reduce solvent demand and waste generation. Waste
segregation may result in lower volumes .of waste solvents, particularly for
aqueous wastes. Availability of small-scale, high-recovery package
distillation units should significantly reduce solvent waste generation from
small-quantity generators (21 percent of nonaqueous solvent disposal capacity
requirements). Finally, use of sludge driers or addition of oil or other
nonvolatile liquids to still bottoms will enable recyclers to maximize solvent
recovery and minimize the volume of waste requiring disposal. These methods
are already in use, primarily for low Btu materials. They should become much
more widely applied for all solvents when land disposal of the bottoms product
is no longer an acceptable option.
Required commercial incineration capacity will also be lower as a result
of other, more economical disposal options and available onsite incineration
19 2
capacity. Data show that 90 percent to 95 percent of incinerated
hazardous solvent wastes are handled onsite in units with low average capacity
19
utilization (20 to 60 percent). Since onsite incinerators tend to be
operated by large firms, it is highly likely that a significant fraction of
the organic liquids currently landfilled will be incinerated onsite when the
cost advantage of landfilling is removed.
It is also likely that much of the estimated 186 MGY of hazardous waste
20
currently used as fuel will continue to be disposed in this manner due to
the economic benefits resulting from utilization of its heating value. The
distribution of this waste will shift towards more efficient, high temperature
thermal processes (DRE's _>99.99 percent) as a result of the impending EPA
regulations. However, several types of high temperature industrial processing
units have demonstrated acceptable destruction capabilities (see Section 12)
and will be available for use. Due to reduced cost advantages of offsite
versus onsite use as a fuel, increased onsite recycling of these solvent
wastes is likely to occur to reduce the total volume requiring alternative
disposal.
4-13
-------
Use of halogenated wastes as cement kiln fuel blends is gaining increased
popularity* In California, the quantity of halogenated wastes used as cement
kiln fuel exceeds the quantity incinerated by a factor of 4.5. One source
estimated that the wet kiln capacity in the United States is sufficient to
destroy nearly four times the annual quantity of chlorinated hazardous waste
12 5 3
produced. Cement kilns can use from 0.7 percent to 1.25 percent of
their design feed as chlorine waste. Upon implementation of the land disposal
ban, it is likely that solvent recovery operations will adopt methods to
recover higher percentages of halogenated solvents. This will make these
wastes more economically blended with spent nonhalogenated solvents and oils
to attain the low chlorine content required for supplemental fuels.
Blast furnaces have also demonstrated acceptable DREs for halogenated
wastes and are currently in use for destroying spent solvent wastes. For
example, nine solvent reprocessors produce 30 MGY of recycled solvent waste
bottoms (Cadence Product 312) which are sold to steel mill blast furnaces.
The chlorine generated from burning the waste is said to prevent alkalis from
building up on the furnace walls. Company officials say demand for the
product is such that they could turn most of the solvent waste in the country
21
into fuel for blast furnaces.
Industries and thermal processes considered to be capable of yielding
99.99 percent DREs for nonhalogenated waste solvents include melting furnaces
in the glass industry, blast and open-hearth furnaces in the iron and steel
industry, rotary kilns in lime production, and reverbatory furnaces in the
copper industry (see Section 12.0). Finally, processes such as wet air
oxidation, super-critical fluids and others (see Sections 7.0 through 12.0)
are gaining acceptance as cost-effective alternative treatment methods.
The EPA Office of Solid Waste is currently performing an analysis of
recycling, treatment and disposal capacity utilization to support development
of the land disposal ban provisions. The capacity determination is based on
survey data being collected by the Office of Policy Assessment for both onsite
and offsite TSD facilities and solvent reclaimers. Preliminary screening
results, available in June 1986, are expected to form the basis for a detailed
questionnaire. These data will be available by December 1986 and should
provide valuable insight into both onsite and offsite recycling, treatment,
22
and disposal capacity utilization.
4-14
-------
SECTION 4.0 REFERENCES
1. Environmental Information, LTD. Industrial and Hazardous Waste
Management Finns. E.I. Minneapolis, MN, 1986.
2. Engineering-Science. Supplemental Report on the Technical Assessment of
Treatment Alternatives for Waste Solvents. Washington, B.C.: U.S.
Environmental Protection Agency. 1985.
3. Horsak, R.D., et al., Face Company Consultants and Engineers, Inc.
Solvent Recovery in the United States: 1980-1990. Houston, TX.
Prepared for Harding Lawson Associates. January 1983.
4. GCA and Metcalf & Eddy. Survey of Solvent Recycling Industry Performed
as part of an assignment to identify firms for performance evaluations.
Work conducted under contract with U.S. EPA ORD (HWERL) for Ron Turner,
1985. Unpublished material.
5. Radimsky, J., et al. Recycling and/or Treatment Capacity for Hazardous
Wastes Containing Halogenated Organic Compounds. State of California,
Department of Health. September 1984.
6. U.S. EPA. Industrial Assessment of Hazardous Waste Practices: Paint and
Allied Products Industry, Contract Solvent Reclaiming Operations and
Factory Application of Coatings. EPA-SW-119c, U.S. EPA/OSW. 1976.
7. Michigan Department of Commerce. Hazardous Waste Management in the Great
Lakes: Opportunities for Economic Development and Resource Recovery.
September 1982.
8. Hobbs, B., and R.R. Hall, GCA/Technology Division, Inc. Study of Solvent
Reprocessors. Bedford, MA. EPA Contract No. 68-01-5960 (Draft), U.S.
EPA Office of Chemical Control. January 1982.
9. Coates, R.H., Safety Kleen Corporation, Elgin, IL. Solvent Recovery
Protects the Environment and Conserves Resources. Industrial Wastes.
July-August 1983.
10. Conversation between Mike Mabry, Marketing Representative for Recyclene
Products and Marc Breton, GCA Technology Division, Inc. January 22, 1986.
11. IGF, Inc. Survey of Selected Firms in the Commercial Hazardous Waste
Management Industry: 1984 Update, Final Report. Washington, D.C.:
U.S. Environmental Protection Agency. 1985.
4-15
-------
12. Dames and Moore, Inc. Economic Impact of Proposed Regulation R81-25:
Prohibition of Chlorinated Solvents in Sanitary Landfills. Park
Ridge, IL. ILENR/RE 83/08. February 1983.
13. Minnesota Waste Management Board - Hazardous Waste Management Report.
MWMB Crystal, MN. December, 1983.
14. Federal Register. Proposed Rule: Land Disposal Restrictions. Vol. 51,
No. 9 (51FR1602). 14 January 1986.
15. Deitz, S.» et al., Westat, Inc. National Survey of Hazardous Waste
Generators and Treatment, Storage, and Disposal Facilities Regulated
Under RCRA in 1981. Roekville, MD. U.S. EPA/OSW. April 1984.
16. Science Applications International Corp. Industries Studies Data Base.
1986.
17. Mitre Corporation. Composition of Hazardous Waste Streams Currently
Incinerated. Prepared for U.S. EPA, Office of Solid Waste. 1983.
18. Chemical Marketing Reporter. Chemical Profiles. Schnell Publishing
Company. 1982 through 1986.
19. U.S. EPA Office of Air Quality Planning and Standards, Assessment of
Incineration as a Treatment Method for Liquid Organic Hazardous Wastes.
Background Report III. Assessment of the Commercial Hazardous Waste
Incineration Market. U.S. EPA. March 1985.
20. Kerwin, J., Westat Inc. Data Base for the Survey of Handlers and Burners
of Used or Waste Oil and Waste - Derived Fuel Material. Prepared for the
U.S. EPA, Office of Solid Waste, Economics and Policy Analysis Branch.
1985.
21, Environmental Reporter - Current Developments. Recycled Wastes Burned
for Material Value Should Not Fall Under RCRA, Court is Told. Bureau of
National Affairs, Inc., Washington, D.C. p. 973. October 4, 1985.
22. Conversation between Jim Craig, U.S, EPA Economic Analysis Branch and
Marc Breton, GCA Technology Division, Inc. February 3, 1.986.
4-16
-------
SECTION 5.0
WASTE MINIMIZATION PROCESSES AND PRACTICES
Waste minimization, as defined here, consists of two distinct aspects of
hazardous waste management: source reduction and recycling/reuse. Source
reduction refers to preventive measures taken to reduce the volume or toxicity
of hazardous waste generated at a facility; recycling/reuse refers to
procedures and processes aimed at the recovery of generated waste or its
reuse, e.g., as a fuel. The two approaches will be discussed separately in
this section, using case studies to illustrate the potential of these
activities for the control of hazardous solvent waste. However, as will
become apparent from subsequent discussions, both source reduction and
recycling/reuse are practices that often are carried out simultaneously by a
facility, as management undertakes multifaceted programs to achieve waste
minimization.
5.1 SOURCE REDUCTION
Source reduction is defined as any onsite activity which reduces the
volume and/or hazard of waste generated at a facility. Source reduction
represents a preventive approach to hazardous waste management, since the
reduction of waste volume or hazard reduces problems associated with waste
handling, treatment, disposal, or liability. Source reduction practices may
impact all aspects of industrial processes generating hazardous wastes, from
raw materials to equipment, to products. A primary motivation for plants to
implement certain source reduction practices is the potential economic benefit
they may accrue. These economic benefits increase as restrictions on waste
management practices become more stringent.
Waste source reduction practices will vary widely from plant to plant,
reflecting the variability of industrial processes producing wastes, and of
the characteristics of the wastes themselves. In general, source reduction
practices may be classified as follows:
5-1
-------
• Raw material substitution;
• Product reformulation;
• Process redesign/modernization; and
* Waste segregation.
A brief description of each type of practice is presented below.
5.1.1 Raw Material Substitution
Raw material substitution involves the replacement of one feedstock,
catalyst, or other material involved in production for another. The
substitute is less hazardous or results in lower hazardous waste generation,
while serving a similar function in the production process and satisfying the
specifications of the end-product. The ideal raw material substitution
would be the replacement of a hazardous material with a nonhazardous material,
while achieving equivalent product quality. The experience described in the
literature, however, indicates that either some compromise in product quality,
or some alteration in process equipment, is often required. Examples of raw
material substitution include the use of water-based material in place of
solvent-based materials (e.g., using an alkaline cleaner, instead of a solvent
cleaner) and the use of prepared materials which eliminate the need for a
hazardous material (e.g., using precoated metal parts thereby eliminating the
need for solvent-based surface preparations). Aside from recycling, raw
material substitution appears to be the most common source reduction practice
employed in industry.
5*1.2 Product Reformulation
Another method that is employed to reduce the volume or toxicity of
wastes produced by a plant is to change the product in some manner. This may
involve the lowering or altering of certain product specifications
(e.g., purity), changing the chemical composition, or changing the physical
5-2
-------
state. An example of product reformulation is shifting paint and coatings
compositions to higher solids content resulting in a. corresponding decrease in
solvent content of rinses and off-specification product wastes. Product
reformulation is considered relatively common in industry, particularly among
2
manufacturers of specialty chemicals.
5.1.3 Process Redesign/Equipment Modification
Process redesign includes the alteration of the existing process design
to include new unit operations, the implementation of new technologies to
replace older operations, changes in operating conditions employed in
processing, or changes in operating practices affecting the process;
e.g., housekeeping or maintenance. Process redesign can, therefore, vary
widely in terms of the effect upon production, product quality, and operating
expenses. Redesigning is often necessary when increasingly stringent
environmental protection standards are to be met. Many processes which
utilize solvents and produce solvent wastes appear to have been designed in an
3
era when pollution control was not a priority.'
Equipment modification or modernization appears to be a prevalent method
for achieving source reduction, despite the potentially high initial costs
involved. New or better equipment may achieve the goals of source reduction
in three ways. First, it may allow for the elimination of a hazardous
material by performing mechanically an equivalent operation to a chemical
process. An example of this may be the replacement of solvent cleaning of
surfaces with a mechanical cleaning system. Second, new equipment may allow
for the replacement of a hazardous material by a less hazardous one. For
example, the use of high pressure alkali cleaning can replace solvent cleaning
of equipment in the paint industry. Third, new or better process equipment
may simply provide better environmental control. An example of this would be
the installation of air pollution equipment such as a wet scrubber to collect
organic vapors.
Improving process controls is considered a particularly important aspect
of equipment modification. Process controls may be less costly and more
technically feasible to implement than replacement or modification of
5-3
-------
large-scale equipment. Process controls include manual, automatic, and
computer-controlled systems. Examples of the use of improved process controls
in industry to reduce waste generation include the increased usage of
computerized controls for paint formulation and batch dyeing operations in the
textile industry. These operations often process a wide variety of raw
3
materials. Improved process control minimizes the potential for generating
off-specification products and excess formulations which may otherwise be
disposed.
The manner in which a process is operated may also be changed to effect
waste reduction. This may be accomplished through the use of different
temperatures or flow rates, by reducing the frequency of process startups or
shutdowns, or by changing maintenance schedules. Improved housekeeping
practices are commonly employed to achieve source reductions. These practices
Include minimizing equipment cleaning and maintenance, shutting down ancillary
equipment when not in use, replacing gaskets, tightening valves, and other
measures. Another manner in which process changes can effect source
reductions is through increased management attention to pollution control and
waste generation. For example, many companies offer employee incentive
programs for identifying cost-cutting measures, some of which involve source
reduction of solvent wastes.
5.1.4 Waste Segregation
Waste segregation entails special storage or handling procedures to avoid
the mixing of different waste streams. The segregation of wastes allows for
certain streams to be treated, recovered or reused, or disposed of in a more
environmentally and perhaps economically sound manner. Segregation is
particularly desirable in eliminating the mixing of toxic waste streams with
nontoxic streams, which results in a larger volume of waste requiring
management. Waste segregation most often will require implementation of new
equipment to collect the separated streams. The technical and economic
feasibility of waste segregation, therefore, may be somewhat limited. An
example of waste segregation is the installation of settling systems to
produce slurried or sludge wastes.
5-4
-------
5.1.5 Case Summaries - Source Reduction
Numerous examples of source reduction practices have been documented in
the literature. The industries involved in these practices are extremely
diverse, ranging from large chemical manufacturing plants to small sized
printing operations. Reduction of solvent hazardous wastes has been
documented more often than any other type of waste. This is probably
attributable to the multiple non-consumptive uses which solvents serve in
industry and the favorable economics of recovery. A number of the documented
cases of source reduction for solvents are discussed below and summarized in
Table 5,1. The case studies are grouped in accordance with the source
reduction practice which resulted in the most significant waste minimization.
However, as noted in Table 5.1 and in the subsequent discussions, use of
multiple minimization practices is not uncommon.
Raw Material Substitution—
Case No. 1; Rexham Corporation,Greensboro, N.C.—Rexham Corporation is
a printer of product labels. Their primary waste stream is derived from
an alcohoi/acetate-based ink used in the flexographic printing process.
The company substituted a water borne ink for several of their
applications. The substitution reduced both waste solvent volume and
solvent air emissions, with only a small effect upon their overall
operations. The costs saved were not estimated.'*
Case No. 2; Riker Laboratories, Northridge, CA—Riker Laboratories is a
coater of medicine tablets. They had previously used a solvent-based
coating solution. Because new air pollution standards would have
required implementation of equipment totaling $180,000 to control solvent
air emissions, the company replaced the solvent coatings with a new
water-based material. This also required new process equipment and
operating procedures. The substitution resulted in a reduction of
24 tons of waste/year, and a raw materials cost savings of
i!5,000/year.1
5-5
-------
IABI.I 5.1. SUMMARY 01 WASri REDUCTION CASES
Case
number Reference
Company
Industry/process
type
Source reduction
procedure
Recycling
procedure
Cost
savings
Comments
Rexhaa Corporation, Printing and coating
Greensboro, NC
Raw material substitution
Distillation
Raw material costs
decreased by 16X
(i5,000/yr)
Waste disposal costs
decreased by 742
Payback of 1 yr
Ui
Riker Laboratories,
Northrldge, CA
Scovill,
Clinton, NC
Coater of medicine
tablets
Manufacturer of
electrical appliances
Raw material substitution
Equipment modification
Raw material substitution
Emerson Electric Co.,
Murphy, NC
Manufacturer of
stationary power tools,
metal finishing and
painting
Equipment replacement
Raw material substitution
Managment Improvement
3M Microelectronics, Manufacturer of flexible Equipment modernization
Columbia, MO circuits Raw material substitution
Materials costs -
$15,000/yr
Reuse of solvents Materials costs-
Jl2,QOO/yr from
the substitution,
t5,320/yr from
recycling
Disposal costs-
t3,040/vr
Materials costs-
|600,000/yr
Disposal costs-
JlO,OQO/yr
Labor costs-
savings of 40%
Managment system
saved Jl47,QQO/yr
Total savings of
il5,000/yr
Saved $180,000 in
pollution control
equipment that
would have been
required
(continued)
-------
TABLE 5.1. (continued)
Case
number
6
7
8
9
10
11
12
Reference Company
4 Tor ring ton Company,
Halhalla, SC
4 Kemp Furniture
Industries,
Goldsboro, NC
4 ITT Telecom,
Raleigh, NC
1 ICI Americas, Inc.,
Goldsboro, NC
1 Rexham Corporation,
Matthews, NC
I DeSoto Corporation,
Greensboro, NC
1 Daly-Herring,
Kingston, NC
Indust ry/ process
type
Manufacturer of
automobile bearings
Manufacturer of
furniture products,
coating operations
Manufacturer of
telecommunications
equipment, printed
circuit boards
Research on agricultural
chemicals
Printer and coater
Manufacturer of trade
sales paints
Manufacturer of
pesticides
Source reduction
procedure
Equipment replacement
RAM material substitution
Equipment modernization
Raw material substitution
Process modification
Improved housekeeping
Waste segregation
Haste segregation
Some recycled for
processing, others sold
Recycling
procedure
Reuse of solvents
in equipment
cleaning
Reuse of solvent
for paint
formulations
Materials eosts-
J2,000/yr
Cost
savings Comments
Expected payback
period of 1 year
Material cost
savings of 23/5/yr
Payback period of
1 year
Total cost savings
of 437, 000 /year
Material reuse was
almost 100% for
solvents
Haste reduction of
98%
for reuse
Disposal costs-
*9,000/yr
-------
Case No. 3; Scovi.ll, Clinton, N.C.—Scovill manufactures small
electrical appliances. The plant had previously used
1,1,1-trichloroethane (TCE) as a solvent degreaser. A water-based
compound was substituted for' the TCE for most of their applications,
providing a cost savings of $12,QOO/year for raw materials. The company
also began substituting waste solvents for virgin solvents in metal
cleaning operations. The reuse of solvents resulted in an additional
$5»320/year saving in material costs. Total costs of disposal that were
saved as a result of these substitutions was $3,040/year, resulting in a
net cost savings of $20,360/year.^
Other Cases—A large metal-working facility eliminated the use of cutting
oils by replacing them with synthetic or water soluble machining
lubricants and coolants. No information detailing cost savings was
provided for this case.
Process Redesign/Equipment Modification—
Case No. 4; Emerson Electric Company, Murphy, N.G.—Emerson Electric
Company manufactures stationary power tools, which require both metal
finishing and painting. Emerson substituted a water-based anodic
electrostatic immersion paint system for an existing organic solvent
paint system. This has resulted in increased productivity and product
quality. The water-based system allows for recovery and reuse of paint.
The resultant cost savings have been $600,000/year for paint, and nearly
$10,000/year for disposal. Labor costs were decreased by 40 percent. In
addition, air emissions and workplace exposure to solvents has been
eliminated.l
Case No. 5; 3M/Microelectronic8 Division, Columbia.MO—The 3M's
Microelectronics Division makes flexible electronic circuits from copper
sheeting. Metal cleaning operations were previously performed using a
chemical spray system. A new process has been installed which cleans the
sheets mechanically instead of chemically, through use of a rotating
brush system. The process resulted in a reduction of 40,000 pounds of
waste/year. Raw materials, disposal, and labor costs were reduced by
ilS.OOO/year.1
Case No. 6; Torrington Company, Walhall, S.C.—Torrington Company
manufactures bearings for the automobile industry. The bearings have a
layer of oil coating on the metal surface to protect it from the stamping
press. Later, heat treatment with oils is carried out. Torrington had
previously used a 1,1,1-trichloroethane vapor degreaser to remove the
stamping and quenching oils. The degreaser has been replaced with a new
parts washer which uses hot water and a low cost alkaline cleaner. The
installed parts washer cost was $40,000 and the company expected a
payback period of -one year. The new equipment reportedly reduced raw
material costs significantly, and is considered to offer an improvement
in worker safety.*
5-8
-------
Case So. 7; Kemp Furniture Industries, Goldsboro, M.C.—Kemp Furniture
Industries manufactures furniture products from particle board, press
board, and plywood. Major operations involving solvents at the plant
include the printing of wood grains and surface coating applications. A
change in the spray guns used at their 25 coating booth from conventional
air guns to air-assisted (airless) spray guns resulted in a materials
savings of 23 percent. The payback period was estimated at one year.^
Case No. 8: ITTTelecon, Raleigh, N.C.—ITT Telecon manufactures
telecommunication equipment using printed circuit boards (PCB). The
plant had been using 1,1,1-trichloroethane as a developing agent, and
methylene chloride as a stripping agent. ITT changed their process to
permit use of an aqueous "photo resist" formulation. This allowed them
to eliminate certain systems including distillation equipment, various
emission controls, and vapor recovery units. The wastes generated were
nonhazardous and could be disposed of in POTW sewers. Cost savings
information was not provided for this case.
Case No. 9: ICI Americas, Inc., Goldsboro, N.C.—ICI Americas, Inc. is
involved in research on agricultural chemicals. They generate small
quantities of a variety of wastes, including spent solvents.
Housekeeping practices were implemented at the facility including
separation of container facilities, identification of chlorinated and
nonchlorinated solvents, and general maintenance. The company reports a
70 percent reduction in waste generation, and estimates a total cost
savings of $37,000/year (based on 1984 dollars).^
Other Cases: Centrifuging—Spent solvent usage from degreasing may be
reduced by centrifuging oily parts before degreasing. This practice is
used by many companies. In some cases, centrifuge processes can be
combined with a detergent spray to eliminate solvent degreasing
altogether.^
Waste Segregation—
Case No.10: Rexham Corporation, Matthews, N.C.—Rexham Corporation
operates a coating facility in which a very large number of different
chemical formulations are employed. A primary means of waste reduction
implemented at the plant was the segregation of spent toluene solvent.
This permitted the reuse, recovery, and sale of toluene, methyl ethyl
ketone and incinerable solvents. Segregation allowed for almost
100 percent reuse of liquid toluene. Information detailing raw material
cost savings was not provided.
Case No. 11; DeSoto Corporation, Greensboro,N.C.—DeSoto Corporation
manufactures trade sales paints. Waste mineral spirits are now collected
in drums and set aside for reuse as a solvent. A waste reduction of
98 percent (from 25,000 gallons in 1981 to 400 gallons in 1982) was
realized.*•
5-9
-------
Case No. 12; Daly-Herring, Kinston, N.C.—Daly-Herring is a pesticide
manufacturer, producing dusts containing a variety of organic chemicals
(mainly pesticides) which are all controlled by a single baghouse. The
company installed separate baghouses to control their two production
lines to permit economic recycling of some of the segregated materials.
The savings accrued by this practice is estimated to be $2,000/year for
materials conserved and $9,000/year for reduced disposal costs. *•
5.1.6 Source Reduction Summary
Source reduction involves a wide variety of practices, some of which may
be applicable at virtually any plant generating solvent wastes. Because the
potential application of these practices is so diverse, there are little
documented data which indicate the significance of waste source reduction on
nationwide industrial waste generation patterns. The EPA and other State
agencies believe that some form of source reduction is applicable to most
industrial plants generating hazardous wastes and will result in a significant
reduction in waste generation as more companies implement waste minimization
programs.
Review of documented case studies on source reduction indicates that
these practices have been applied in more instances to solvent wastes than any
other waste type. Additionally, it appears that source reduction practices
are used more frequently for chlorinated solvents, especially
1,1,1—trichloroethane and methylene chloride. With respect to cost savings,
the data appear to indicate that source reduction of large generating sources
may yield annual savings of tens of thousands of dollars. Savings in the
hundreds of thousands of dollars may even be possible if source reduction
practices allow for the elimination of unit operations such as air pollution
controls.
Regulatory trends appear to be moving towards the promotion of source
reduction at sites generating hazardous wastes. The EPA has recently proposed
requirements that generators certify institution of hazardous waste reduction
programs (51 FR 10177, March 24, 1986). This would involve the institution at
generator sites of programs to reduce the volume or toxicity of hazardous
wastes to a degree determined by the generator to be economically
practicable. Generators must also certify that their current method of
management is the most practicable method available to minimize present and
5-10
-------
future threats to human health and the environment. Three States currently
have established source reduction/pollution prevention programs: North
Carolina, Minnesota, and Massachusetts. In addition, Tennessee has
established a "pilot program", and Kentucky, California, Maryland, and
Washington have programs currently in development. These programs vary from
state to state but, in general, include information exchange, technical
assistance, and economic incentives to companies to encourage development of
their programs.
5.2 RECYCLING/REUSE
According to EPA guidance issued on January 4, 1985, "recycling" was
defined as practices in which wastes are: (1) reclaimed, or (2) reused. A
reclaimed waste is one which is processed or treated through some means to
purify it for subsequent reuse, or to recover specific constituents for
reuse. Reused wastes are those which serve directly as feedstocks, without
any treatment. Recycling of wastes may be done by either the original
generator or other firms. This section discusses the various advantages and
disadvantages of the available technologies for spent solvent waste recycling,
and the potential consequences of their application to specific waste types.
Of the major classifications of hazardous wastes, solvents are recycled
most often. Based on information provided by the National Association of
Solvent Recyclers, significant volumes of the following 11 solvents have been
recycled:
xylene methanol
toluene perchloroethylene
acetone trichloroethylene
methyl ethyl ketone methyl isobutyl ketone
1,1,l-trichloroethane isobutanol
methylene chloride
Volumes of various solvent waste types recycled have been discussed in
Section 3.3. Roughly 80 percent of solvent recycling occurs in onsite
facilities.
5-11
-------
5.2.1 Recycling Technologies
The most common solvent waste recycling technologies are distillation,
settling, decanting and filtration, and solvent extraction, with distillation
the most prevalent. A compilation of documented solvent recycling practices
is shown in Table 5.2.
Solvent recovery technology commonly includes three types of operating
systems: distillation; solvent extraction; and adsorption. The key to the
effective performance of any solvent recovery system lies in the
characteristics of the solvent itself. Solvents which are more volatile, for
example, will be readily recovered through distillation. A wide variety of
treatment systems exists for recovery of solvents, most of which are discussed
in detail in other chapters of this report. A brief summation of several
important solvent recovery technologies is provided below. Further detail is
provided in Section 7.
Distillation—
Distillation is a liquid separation process which takes advantage of
differences in the relative volatilities of constituents present in a process
stream. The products obtained from this process can possess relatively high
levels of purity. The most common distillation processes are:
• Batch Distillation—Batch distillation is the simplest available
distillation system, consisting of an evaporator followed by a
condenser. This process is generally effective and economical for
solvents with low solid content (<5 percent) and low viscosity.^
Between 50 and 95 percent of the solvent can be recovered by batch
distillation.*> The still bottom products from batch distillation
of halogenated solvents can contain 20 to 40 percent halogenated
solvents.8
Batch distillation is widely used onsite by operations such as dry
cleaners, print shops, and metal degreasing operations that do not
have high solvent volume demands . Studies show that for streams
containing 3,000 to 6,000 ppm chlorinated hydrocarbon, removal
efficiencies of up to 90 percent have been obtained.^
Distillation is not recommended for certain spent solvents (e.g.,
mixtures of toluene and methyl ethyl ketone from plastic and paint
manufacturers) which contain resinous materials which polymerize on
still walls or coil surfaces and reduce heat-exchange
5-12
-------
TABLE 5.2. SUMMARY OF DOCUMENTED SOLVENT RECYCLING PRACTICES'
Production process/ SIC
industry type code
Type of solvent
recovery
Location
Waste recycled/
recovered
Industrial
precedent
Manufacture wood
office furniture
2521 Still used to recycle acetone
Added to reclaimed solvent and
this reconstituted mix is used
as a thinner
Onsite Spent lacquer thinner Bowling Co., Ht. Olive,
N.C.
Ul
I
u>
Polyester resin 2824
and fiber plants
Cellulosics plant 2823
Industrial organic 286
chemicals
Distillation (tolling
arrangement)
Bench scale distillation
process to separate
solvent from oil
Fractional distillation
Offsite Dowtherm (solvent)
Celanese Fiber Operations,
Charlotte, N.C.
Onsite Freon THC® (solvent) Celanese Fiber Operations,
Charlotte, N.C.
Halogenated still bottoms Not documented
Printing industry 27
Dry cleaning opera- 9711
tions at a paint
shop
Printing and 2751
publishing
Manufacture coatings 2851
(including water and
solvent based paints)
Collection in a still,
followed by distillation
(still bottoms incinerated)
Nonfractionating batch
still
Collected in a still and
reclaimed (reused for cleanup
with virgin materials)
Spent solvent continuously
pumped into still. Solvent is
distilled (primarily reused
for equipment cleanup)
Onsite Xylene contaminated with Lenoir Mirror, Co.,
paint Lenoir, N.C.
Not given Waste solvents (including: Norfolk, NSY, Norfolk, VA
mineral spirits, ketones,
and epoxy thinners
containing paint pigments)
Not given Alcohol/acetate mixture Rexham Corp., Greensboro,
N.C.
Onsite Spent solvent (primarily Southern Coatings, Sumter,
toluene and xylene) S.C.
-------
TABLE 5.2 (Continued)
Ul
I
Production process/
industry type
SIC
code
Type of solvent
recovery
Waste recycled/
Location recovered
Industrial
precedent
Manufacture solvent 2893
based inks for retro-
gravure printing
Maintaining aircraft, 9711
helicopters and
missiles
Still used to collect solvent
(reclaimed solvent used
twice for cleaning equipment
before redistillation)
Batch atmospheric pressure
stills (single-stage batch
still; water separator A
electrically powered steam
generator)
Production of electric 3825 Closed-loop continuous distil-
meters
ODD base
Paint thinning
lation units (from vapor
degreasing operations)
Distillation (incineration
if contaminated)
2851 Single stage distillation to
separate volatile solvent
from paint thinners
Photographic equipment 3861 Fractional distillation
Machinery 35 Fractional distillation
Thiele-Engdahl,
Winston-Salem, N.C.
Warner Robins AFB,
Hacon, GA
Onsite Spent isopropyl acetate
Onsite Trichloroethane, Freon-
113, isopropanol (a
new still has been
installed to reclaim
PD-680 dry cleaning
solvent, silicon damping
fluid, paint thinners,
and collanol 25R fluid)
Onsite Dow stabilized perchloro- westinghouse Electric
ethylene and Freon TMS Meter Plant,
degreaser Raleigh, NC
Onsite Freon Type 2
Not given Paint thinners
DOO Base
DOO Base
Not given Solvents, halogenated Not documented
Not given Solvents, halogenated Not documented
-------
TABLE 5.2 (Continued)
Production process/ SIC
industry type code
Type of solvent
recovery
Location
Waste recycled/
recovered
Industrial
precedent
V/i
t->.
Ul
Pesticide 4 agricultural 2879 Fractional distillation Not given
chemicals
Plastics & synthetics 282 Fractional distillation Not given
Stone, glass, clay
Industrial organic
chemicals
Pharmaceuticals
Paint & allied
products
Industrial organic
chemical 4 chemical
preparation industry
Electronics &
machinery industry
Petroleum refining
industry
Transportation equip-
ment industry
32 Fractional distillation Not given
286 Fractional distillation Not given
283 Fractional distillation Not given
285 Fractional distillation Not given
286 Distillation (often Offsite
tolling arrangement)
36 Distillation (often Offsite
tolling arrangement)
2911 Distillation {often Offsite
tolling arrangement)
40 Distillation (often Offsite
tolling arrangement)
Solvents, halogenated Not documented
Solvents, halogenated Not documented
and nonhalogenated
Solvents, nonhalogenated Not documented
Solvents, nonhalogenated Not documented
Solvents, nonhalogenated Not documented
Not documented
Solvents containing
heavy metals
Solvents
Solvents
Not documented
Not documented
Solvents, nonhalogenated Not documented
Solvents, nonhalogenated Not documented
-------
TABLE 5.2 (Continued)
Production process/
industry type
Chemical preparation
industry
Electronics industry
Machinery industry
Parts cleaner service
(cleaning equipment
with solvents)
Manufacture wood office
furniture
SIC
code
2899
362
35
2842
2521
Type of solvent
recovery
Distillation (often
tolling arrangement)
Distillation (often
tolling arrangement)
Distillation (often
tolling arrangement)
Settling followed by distil-
lation for solvents - bottom
oils blended for fuel
Heat recovery
Location
Off site
Off site
Off site
Off site
(for gene-
rators)
Onsite
Waste recycled/
recovered
Solvents, nonhalo-
genated
Solvents, nonhalo-
genated
Solvents, nonhalo-
genated
Solvents
Spent lacquer thinner
used for heat recovery
Industrial
precedent
Not documented
Not documented
Not documented
Not documented
Bowling Co.,
Nt. Olive, NC
Wood and Furniture
industry
2521
Heat recovery
Cellulose acetate fiber 2823
plants (spinning
process)
Concentrate HeO/acetone and
send to off-site kiln for
heat recovery
Onsite Heat recovery using spent Burlington Furniture Co.,
solvents (toluene, Lexington, N.C.
xylene, acetone, ethanol,
butanol, isopropyl
alcohol, neptha, methyl
ethyl ketone X esters)
Offsite Heat recovery using Celanese Fiber Operations,
concentrated HsQ/aeetone Charlotte, N.C.
mix
-------
TABLE 5.2 (Continued)
Ul
I
Production process/
industry type
Chemical & allied
products (cosmetics)
Chemical distri-
butor & recycler
High tech printing
and coating
Hultidi visional
manufacturing
Wood & furniture
industry
Paint, coatings
& ink industry
Industrial organic
SIC
code
2844
5161
27
3861
25
2851
2893
286
Type of solvent
recovery
201 hydro-alcoholics used
in fuel (modified steam &
hot water boilers)
Residue from recycling
is blended and used
as a fuel
Heat recovery
Fuel supplement in
existing boiler
Heat recovery
Heat recovery
Use in cement kilns
Location
Onsite
Off site
(for gene-
rators)
Onsite
Onsite
Onsite
Onsite
Of f si te
Waste recycled/
recovered
Heat recovery using
hydroalcoholics
Residue is blended with
other hydrocarbons
and sold as fuel for
cement kilns
Methyl ethyl ketone vapors
burned for heat recovery
Solvent laden air (pre-
dominantly heptane)
Spent solvents
Spent solvents
Solvents
Industrial
precedent
Coty Div., Pfizer Inc.,
Stanford, H.C.
Hukill Chemical Corp.,
Bedford, OH
Rexham Corp. ,
Matthews. N.C.
3H, Minneapolis, HN
(headquarters)
Not documented
Not documented
Not documented
chemical & chemical
preparation industry
Electronics and 36
machining industry
Petroleum refining 2911
for heat recovery
Use in cement kilns
for heat recovery
Heat recovery
Offsite Solvents
Offsite & Solvents, non-
onsite halogenated
Not documented
Not documented
-------
TABLE 5.2 (Continued)
I
I-1
oo
Production process/ SIC
industry type code
Transportation 40
equipment
Electronics 36
High tech printing 27
& coating
Recondition used tanks 9711
and other armored
vehicles (removal of
paint, grease, oil,
rust, and oxides)
Printing process, paper 2653
& allied products
Multidivisional 3861
manufacturing
Tape manufacturer 2295
Multidivisional 3861
manufacturing
Type of solvent
recovery
Heat recovery
Heat recovery
Solvent vapors recovered
& sold to coating industry
Vapor degreasing units
equipped with distillation
solvent recovery system
Recovery of vaporized
solvent
Condensation of vapors from
drying oven exhaust
carbon adsorption systems -
after adsorption & desorption
processes the solvent is
decanted from water and reused
Carbon adsorption
Location
Off site &
Onsite
Off site &
Onsite
Onsite
Onsite
Onsite
Onsite
Onsite
Onsite
Waste recycled/
recovered
Solvents, non-
halogenated
Solvents, non-
halogenated
Solvent vapors
(toluene)
Trichloroethylene
Vaporized solvent
Hydrocarbons
Toluene
Solvents
Industrial
precedent
Not documented
Not documented
Rexham Corp.,
Matthews, N.C.
Anniston Army Depot
RJR Archer, Inc.,
Winston-Salem, N.C.
3H, Minneapolis, HN
(headquarters)
Shuford Hills,
Hickory, N.C.
3H, Minneapolis, HN
(headquarters)
-------
TABLE 5.2 (Continued)
Production process/ SIC
industry type code
Industrial orpnic 286
chemical
Chemical preparation 2899
industry
Oil refining industry 2911
Electronics industry 362
JL Plastics industry 30
v£>
Nultidi visional 3861
manufacturing
(film developing)
Industrial organic 286
chemicals
Plastics industry 30
Chemical preparation 2899
industry
Oil refining industry 2911
Type of solvent
recovery
Carbon adsorption
Carbon adsorption
Carbon adsorption
Carbon adsorption
Carbon adsorption .
Gravity separation
Steam stripping
Steam stripping
Steam stripping
Steam stripping
Location
Onsite A
Off site
Onsite A
Off site
Onsite A
Off site
Onsite A
Off site
Onsite A
Off site
Onsite
Onsite A
Off site
Onsite A
Off site
Onsite A
Off site
Onsite A
Off site
Waste recycled/
recovered
Solvents, non-
halogenated
Solvents, non-
halogenated
Solvents, non-
halogenated
Solvents, non-
halogenated
Solvents, non-
halogenated
Solvents
Solvents, nonhalo-
generated A halo-
genated
Solvents, nonhalo-
genated and
halogenated
Solvents, nonhalo-
genated
Solvents, nonhalo-
genated
Industrial
precedent
Not documented
Not documented
Not documented
Not documented
Not documented
3N, Minneapolis, m
(headquarters)
Not documented
Not documented
Not documented
Not documented
-------
TABLE 5.2 (Continued)
I
Ni
O
Production process/
industry type
Electronics industry
Aerospace industry
SIC
code
362
9661
Type of solvent
recovery
Steam stripping
Steam stripping
Location
Onsite &
Off site
Onsite 4
Off site
Waste recycled/
recovered
Solvents, nonhalogenated
and halogenated
Solvents, halogenated
Industrial
precedent
Not documented
Not documented
Hazardous waste
management, facility
Industrial organic
chemicals
Plastics industry
286
Solvent extraction
followed by a flash
evaporator and calcium
chloride adsorption bed
Distillation to remove
oil; followed by
solvent extraction to
reclaim solvent
Solvent extraction
30 Solvent extraction
Chemical preparation 2899 Solvent extraction
industry
Oil refining industry 2911 Solvent extraction
Onsite Waste solvent containing
typically 851 methylene
chloride and 151
isopropyl alcohol to
obtain saleable quality
methylene chloride
(98-991 pure)
Onsi te Waste material of oi1,
freon and other solvents;
freon is extracted after
oil has been removed
Onsite & Solvents, nonhalogenated
Offsite and halogenated
Onsite & Solvents, nonhalogenated
Offsite and halogenated
Onsite & Solvents, nonhalogenated
Offsite and halogenated
Onsite & Solvents, nonhalogenated
Offsite and halogenated
Si 1resin Chemical
Corp.,
Lowell, Mass.
Not documented
Not documented
Not documented
Not documented
Not documented
-------
TABLE 5.2 (Continued)
Production process/ SIC
industry type code
Type of solvent
recovery
Location
Waste recycled/
recovered
Industrial
precedent
Ul
I
N>
Electronics industry 362 Solvent extraction
Aerospace industry 9661 Solvent extraction
Onsite & Solvents, nonhalogenated Not documented
Offsite and halogenated
Naval Air Rework
Facility
9111
Manufacture trade sales 5198
paints
High tech printing & 21
coating
Produces coatings for 3296
fiberglass or
2221
Magnetic tape facility 3679
Onsite &
Offsite
Settling and straining to remove Onsite
dirt oil 4 water-based cooling
emulsion separated by a centrifuge
Waste reuse-wash solvent from each Onsite
batch is collected in druns, waste
solvent from previous cleanup is
used in the manufacturing process
Cleanup toluene is segregated by
ink type and reused as a thinner
Solvent recycling
Nitrogen-based solvent recycling
system (with this system higher
concentration levels can be
obtained safely for solvent
recovery)
Onsite
Offsite
Solvents, nonhaloge-
nated and halogenated
Water-based cooling
emulsion (e.g.,
Irimsol, Simcool)
Waste solvent
Not documented
Pensacola Naval Air
Rework Facility,
Pensacola, FL
Oesoto,
Greensboro, N.C.
Spent toluene (from press Rexham Corp.,
and roller cleanup) Matthews, N.C.
Spent acetone (from
cleanup process)
Onsite Tetrahydrofuran
American Colors,
Charlotte, N.C.
Airco Industrial Gases,
Hurray Hill, N.J.
(facility located in CA)
-------
TAELE 5.2 (Continued)
I
fo
ho
Production process/
industry type
Cellulose acetate
fiber plants
(spinning process)
Flushing refrigerant
through shipboard
SIC
code
2823
9711
Type of solvent
recovery Location
Recovery and recycle Onsite
of acetone
"Flushing rig" made of spare Onsite
parts used to clean impurities
Waste recycled/
recovered
Acetone
Refrigerant
Industrial
precedent
Celanese Fiber
Operations,
Charlotte, N.C.
Charleston HSY,
Charleston, H.C.
refrigeration units
Solvent Recyder
Chemical distributor
and recycler
Manufacture coatings
for furniture
industry
out of refrigerant and the
refrigerant is recirculated
through the system
7399 Use still bottoms to make
a lon-grade paint
(residue is reused either
as a fuel supplement or as
a source of pigments 4 resins
for the paint industry)
5161 Batch tolling or purchase
waste solvents outright
2851 Solvent recycling
Solvent recycling
Offsite Still bottoms
(for fran paint
generators) Romulus, HI
Offsite
(for
generators)
Offsite
Onsite
Accepts wide range
of solvents
Low boiling point
solvents (including
ketones, esters,
aliphatics & aromatics)
Hydrocarbon thinners
& solvents,
fluorocarbons,
freons & esters
Chemical Recovery
Systems,
Hukill Chemical Corp.,
Bedford, OH
Lilly, High Point, H.C.
Oil & Solvent Processing
Co., Azusa, CA
-------
TABLE 5.2 (Continued)
Production process/ SIC
industry type code
Printing 27
Pesticide manufac- 2879
turing (using
organic solvent
media)
Ul
I
u> Industrial organic 286
chemicals
Chemical prepara- 2899
tion industry
Electronics & 36
machinery
industry
Petroleum refining 2911
industry
Transportation equip- 40
ment industry
Type of solvent
recovery
Flexoprinter reclaims useable
solvents from waste inks
Solvent recycling
Solvent recycling
Solvent recovery for reuse
for equipment cleaning
Distillation, evaporation,
filtration
Distillation, evaporation,
filtration
Distillation, evaporation
filtration
Distillation, evaporation
filtration
Distillation, evaporation
filtration
Location
Onsite
Onsite
Onsite
Onsite
Onsite
Onsite
Onsite
Onsite
Onsite
Waste recycled/
recovered
Solvents
Solvents
Solvents
Solvents
Solvents from degreasing
and dry cleaning
Solvents from degreasing
and dry cleaning
Solvents from degreasing
and dry cleaning
Solvents, nonhalogenated
Solvents, nonhalogenated
Industrial
precedent
Rexham Corp. ,
Greensboro, N.C.
Ramie Chemical Co.,
Palo Alto, CA
Solvent Services,
San Jose, CA
Not documented
Not documented
Not documented
Not documented
Not documented
Not documented
^Source: Versar, 1985. Reference 6.
-------
* Separators—There are two types of separators used to recover spent
solvents. Scraped-surface separators are well suited for solvent
streams with a high concentration of suspended solids and sludges
and can be effective in separating spent solvent fractions by
density. 10 The other type of separator is the thin (falling.) film
separator. Thin film separators are best suited for reclaiming low
boiling point (150 to 200°C) solvents and are not applicable to
spent solvents containing suspended or dissolved resinous
materials. Both types of separators can be used independently or as
a pretreatment step prior to distillation columns.°»10
• Distillation (Fractionation) Columns—Distillation columns are
employed when a high purity product (typically solvent contents
greater than 90 percent) is required. The feedstock stream has to
be a free-flowing fluid with negligible solids content.
Pretreatment is usually required prior to this process.
Two types of distillation columns exist, plate columns and packed
columns. By installing intermediate drawoff levels on plate
columns, a number of mixed solvents can be recycled at one time if
they have boiling ranges separated by approximately 20 to 30°C. *•'•'
Packed columns usually result in a higher quality solvent but are
more susceptible to fouling. The choice between these two types of
distillation columns is usually based on economics."
Solvent Extraction—
In solvent extraction, a solute is transferred from one liquid phase to a
second immiscible phase. The operation can be conducted as a batch process or
by contact of the solvent with the feed in staged or continuous contact
equipment. Solvent extraction is most suitable for recovering organic
9
solvents from aqueous waste streams. The costs for solvent extraction will
be highly dependent on the value of the recovered organics for concentrated
solvent wastes.
Adsorption—
Adsorption is a well established separation process. For the purpose of
recovering spent solvent, activated carbon and resin systems should be
considered. The major distinction between the two systems is the type of
adsorbent regeneration process employed. Resin regeneration is generally
easier and can be achieved using rinse waters and solvents, whereas high
temperature furnaces are usually required to drive off the sorbate from the
5-24
-------
carbon surface. Both carbon and resin systems require the feedstock stream to
be a single liquid phase. The solid content in feedstocks should be less than
50 ppm and may, in some cases, have to be below 10 ppm to prevent bed
. . 9
clogging.
Many solvents, especially low boiling point chlorinated solvents are
readily reclaimed using an activated carbon adsorption system. However, the
major application of adsorption systems to date has been confined to dilute
9
aqueous streams.
Purification--
When high purity of recovered products is required, further purification
processes become necessary to increase the solvent content of the final
product. Purification processes may include fractionation columns, further
decanting with additional cooling to increase separation or drying through a
desiccant or salt bed. Some solvents must be treated with additives to
13
restore their buffering capacity.
5.2.2 Selection of Recycling Technology
The economic benefits of recycling a wastestream or mixture of waste
streams are dependent upon the physical and chemical characteristics of the
waste stream and the quantity of waste to be recycled.
The physical and chemical characteristics of the waste determine the
technical constraints of the process including raw materials, handling and
pretreatment, and operating parameters of the system such as temperatures and
pressures. In general, physical form determines whether or not a process can
be used whereas chemical characteristics affect ease of separation and
selection of optimal processing conditions.
The types of constituents in a waste will influence its reactivity,
volatility, and adsorbability. The most significant constituents to consider
are halogens (especially chlorine), metals, other solids and moisture which
present particular limitations on certain technologies. Most recycling
technologies have practical limits on their capability of handling high
solids, viscous and polymerizable materials. Pretreatment through the use of
5-25
-------
physical separation processes (e.g., filtering, decanting, settling, skimming,
centrifuge) is common in recycling applications. Similarly, post treatment
processes may be required to bring recycled materials up to process
specifications. Typical post treatment methods include decanting, dehydrating,
and fractionation.
The quantity of waste to be recycled is also a significant factor in the
selection of an appropriate recycling technology. Waste quantity will
determine the size of equipment, volume of raw materials to be used in
recycling (e.g., carbon for carbon adsorption), and pollution control and
disposal requirements. Certain technologies may be preferable for small
quantity processing but, in general, the larger the quantity of waste to be
recycled, the more economically attractive recycling becomes.
Economic considerations play a major role in determining the
recyclability of a solvent hazardous waste. The primary economic
considerations are the capital and operating costs of the recycle system,
disposal costs and value of recovered solvent. Among the recovery
technologies discussed earlier, solvent extraction processes have the highest
capital and operating costs, due to their high demand for extraction agents
and associated equipment, while adsorption systems are probably the least
expensive. Disposal costs and value of recovered solvent become increasingly
important as waste volume increases. If recovered products are used onsite,
their value is reflected in the reduced demand for virgin raw materials. If
sold for offsite use, their value is dependent on the market price of virgin
solvent and the degree of purity. Typical prices of recovered solvent range
from 50 to 90 percent of the price of virgin materials (see Table 4,1,2).
Another option to be considered in selecting a recycling technology is
whether the operation should be conducted onsite, or at an offsite facility
such as a commercial recycler. The choice between onsite and offsite recovery
is dependent on many factors including recycling costs, volume of spent
solvent generated, availability of equipment, personnel and markets, facility
size, technical capability of personnel, and use of recovered solvent.
Transportation cost must also be considered for offsite recycling. The
cost for transportation is a function of the distance from the generating
facility to the recycling facility, the volume of the waste being transported,
5-26
-------
and the transportation method used. Generators of low volumes of wastes can
significantly reduce their handling and transportation cost by participating
in cooperative storage arrangements with other small quantity generators of
similar wastes. A good example is the Neighborhood Cleaners Association (NCA)
collective recycling program. Under NGA1 s new recycling program, a
contractor has been hired to collect and recycle perchloroethylene solvent
wastes which were previously landfilled by 1,400 dry cleaning facilities.
Part of this program's success results from the chemical compatibility of the
waste streams collected from the dry cleaners.
In addition to purely economic factors, the size of a facility and its
technical expertise may also influence the decision to recycle solvents onsite
or offsite. Large facilities usually have the advantage of a strong technical
staff to manage onsite recovery. However, onsite recovery has been found to
be a competitive option to offsite recovery for both small and large
generators.
Ultimate selection of a recycling technology will be highly site
specific. A selection methodology has been presented in Section 14.U
outlining economic and other considerations. Details on costs and
capabilities of specific recycling technologies can be found in Section 7.0.
5.2.3 Reuse
Certain solvent hazardous wastes generated by an industrial process may
be directly used for a different purpose in another process. Examples of
direct reuse of spent solvents include wastes used in precleaning and wastes
fired as fuel in high temperature industrial processes. Reused wastes have
the advantage of much lower costs relative to use of virgin materials. In
general, reusable wastes are produced by large manufacturing operations, or
those which require high purity, and are consumed by smaller, often batch
processors, which do not necessarily require high levels of purity in
feedstocks. Reused wastes are usually only slightly contaminated, but several
instances have been documented which where involve reuse of wastes with only
60 to 70 percent solvent content.
5-27
-------
Three primary factors should be considered when evaluating reuse as a
potential waste management option. First, the ability to reuse a solvent
waste depends upon its chemical composition and effect of the various waste
contaminants on the reuse practice. Second, the economic value of the reused
waste must justify the expense incurred in changing a process to accommodate
it. Third, the availability of the waste must be considered. A processor
using a secondary material must be sure that the material will be available to
satisfy his demand.
Reuse of wastes may be accomplished either by the generator itself, or
through sales to a different processor. Marketing of wastes for reuse is
often facilitated through use of waste exchanges. Waste exchanges are
institutions which serve as brokers of wastes or clearing houses for
information on wastes available for reuse. In some waste exchanges, potential
buyers of wastes are brought into contact with generators, while other waste
exchanges accept or purchase wastes from a generator for sales to other
users. Waste exchanges are considered by EPA to be of great potential value
in future waste management because through waste exchanges, recycling
. . ,12
practices may be increased.
A wide variety of wastes have been recycled via the waste exchange
system. A listing of solvent hazardous wastes available through two waste
exchanges is presented in Table 5.3. Solvent wastes are among those which
have been sought most highly, due to their versatile reuse potential.
In general, the "exchangeability" of a waste is enhanced by higher
concentration and purity, quantity, availability, and higher offsetting
disposal costs. Some of the limitations to waste exchangeability are the high
costs and other difficulties associated with transportation and handling,
costs of purification or pretreatment required and in certain cases, the
effect on process or product confidentiality. In general, waste exchange
involves transfer of products from large, continuous processors to small,
batch processors, manufacturing products from basic chemical manufacture to
chemical formulators, or products from high purity processors
(e.g., pharmaceutical manufacturers) to low purity processors (e.g., paint
manufacturers).
5-28
-------
TABLE 5.3. SUMMARY OF SOLVENTS RECYCLED VIA TWO WASTE EXCHANGES
Type
of
wastes
Solvents
Org antes/solvents
Carbon tetrachloride
Ethanol
Lacquer solvent
Mixed solvents
Paint & ink wash
solvents
Paint solvents
Phenol
Poiydimethylsiloxane
Trichioroe thane
Trichloroethylene
Acetone0
Solvents0
Triehloroethy lenec
Trichioroe thanec
Solvents
Ethylene glycol
Solvents
Mixed ethylene
glycols
Trichioroe thane
Paint thinner
Trichioroe thane
Alcohols
Solvents
Waste
exchange3
IME
NEIWE
NEIWE
NEIWE
NEIWE
NEIWE
NEIWE
NEIWE
NEIWE
NEIWE
NEIWE
NEIWE
NEIWE
NEIWE
NEIWE
NEIWE
NEIWE
NEIWE
NEIWE
NEIWE
NEIWE
NEIWE
NEIWE
NEIWE
NEIWE
Time
period
1985
1983
1983
1983
1983
1983
1983
1983
1983
1983
1983
1983
6/81-12/81
6/81-12/81
6/81-12/81
6/81-12/81
6/81-12/81
6/81-12/81
6/81-12/81
6/81-12/81
6/81-12/81
2/82-2/83
2/82-2/83
2/82-2/83
2/82-2/83
Quantity
(tons) (3)
639.3
0.9
3.8
Unknown
75.0
8.3
27.5
0.3
2.7
3.4
4.2
0.6
10
3
3
55
5
142
16
3
42
2
Unknown.
210
Distance
hauled
(miles)
—
400
450
200
200
50
175
5
175
375
60
75
125
75
25
250
75
150
Unknown
Unknown
175
75
Unknown.
Unknown
Estimated
$ value
937,960
396"
400
875
16,000
2,000
20,000
250
3,000
1,500
2,000
208
2,040
None
1,100
10,000
3,000
39,000
Unknown
2,992
14,000
385
Unknown.
110,000
aIME = Industrial Material Exchange.
NEIWE = Northeast Industrial Waste Exchange.
^Unit cost estimate obtained from Chemical Marketing Reporter,
May 28, 1984 issue.
C0ne-time only transaction.
Source: Versar 1985. Reference 6.
5-29
-------
Waste exchanges are operated by both private firms and public
organizations. Several waste exchanges are listed below:
* California Waste Exchange (California);
* Canadian Waste Materials Exchange (Ontario);
* Chemical Recycle Information Program (Texas);
* Colorado Waste Exchange (Colorado);
* Georgia Waste Exchange (Georgia);
* Great Lakes Regional Waste Exchange (Michigan);
* Industrial Materials Exchange Service (Illinois);
* Industrial Waste Information Exchange (New Jersey);
» Inter-Mountain Waste Exchange (Utah);
» Louisville Area Waste Exchange (Kentucky);
* Midwest Industrial Waste Exchange (Missouri);
* Montana Industrial Waste Exchange (Montana);
* Northeast Industrial Waste Exchange (New York);
* Piedmont Waste Exchange (North Carolina);
* Southern Waste Information Exchange (Florida);
* Techrad (Oklahoma);
* Tennessee Waste Exchange (Tennessee);
* Virginia Waste Exchange (Virginia);
» Western Waste Exchange (Arizona); and
» World Association for Safe Transfer and Exchange (Connecticut).
The following is a list of the private material exchanges currently in
business:
• Zero Waste Systems, Inc. (California);
* ICM Chemical Corporation (Florida);
» Environmental Clearinghouse Organization - ECHO (Illinois);
• Americal Chemical Exchange - ACE (Illinois);
* Peck Environmental Laboratory, Inc. (Maine);
• New England Materials Exchange (New Hampshire);
• Alkem, Inc. (New Jersey);
• Enkarn Research Corporation (New York);
» Ohio Resource Exchange - ORE (Ohio); and
* Union Carbide Corporation (in-house operation only; West Virginia).
5-30
-------
5.3 ADDITIONAL EXAMPLES OF WASTE MINIMIZATION PRACTICES
There is a growing incentive for companies to undertake waste
minimization programs as a consequence of increasing waste disposal costs and
liability. Besides protecting human health and the environment by drastically
lowering the amount of waste generated, waste minimization programs can, in
many cases, provide substantial economic benefits. Table 5.4 provides
information concerning onsite waste reduction techniques employed at three
major hazardous waste generators. The accomplishments, as measured by the
percent reduction of wastes (mainly solvents), are impressive. Waste
reductions and resulting dollar savings at a fourth company are shown in
Table 5.5.
Finally, seven example case summaries describing waste recycling/reuse
are provided below. Data describing net financial benefits is reported in
most of these examples.
Case 1: 3MElectronics Products Division, Columbia,MO—3M is involved in
film developing, in which a wastewater contaminated with
1, 1,1-trichloroethane is produced. 3M installed a gravity decanter
system to separate and subsequently recycle TCE. The process
resulted in an annual savings of $12,000 (based on first year
results), and also reduced their contamination of process water. *•
Case 2: Celanese Fibers, Charlotte, ,NC—Celanese Fibers manufactures
cellulosic fiber products. Solvents are used in a variety of
operations, ranging from extractions to coatings. Solvent wastes
are recycled both in—house and offsite. Laboratory solvents are
recycled in-house in a bench-scale batch distillation process which
separates Freon from finishing oil. An annual savings of $1,&>4 was
reported. In another application, contaminated Dowtherm, which is
used as a heat transfer medium, is sent to a distiller for
regeneration. An annual savings of $163,765 was reported, which
includes savings for materials and disposal.**
Case 3: Westinghouse Electrical Corporation Meter Plant, Raleigh, NC—
Westinghouse Meter produces electrical meters for power plants.
Three large vapor degreasers produce contaminated perchloroethylene
solvents. These degreasers are equipped with continuous
distillation to recycle clean solvents. The company estimates that
the payback period for such units is 5 years.*
5-31
-------
TABLE 5.4 EXAMPLES OF WASTE REDUCTION TECHNIQUES /ACCOMPLISHMENTS
(ALL PROGRAMS ARE ONSITE)
Wanra
«BCG
reduction
method
Hewlett Packard
Solvent conservation
practices
Haste solvent recovery and
oil recycling
Conversion to water based
paints
Conversion f roa water mil
paint booths to dry filters
Install sensors on photoresist
1 bottle
N> Eliminate phenolic stripper
Haste
EPA
number
F002
F003
F006
F001
D001
DOOl
F003
F003
reduced
Type
Halogenated solvents
Alcohols
Ketones
Halogenated solvents
Cutting fluids
Haste paints
Paint sludge
Solvents
Phenolic stripper
Percent
reduced
. 20
50
30
100
20
100
?__ _,
ear
program
started
1983
1982
198S
1985
1985
1983
Description
Installed covers on sinks, increased freeboard.
Used distillation and centrifuge to reclaim
materials for onsite use.
Hater based paints have lower solvent content,
are less hazardoua, and create less air pollution.
Bet paint booths create a liquid hazardous wastej
dry filters are a solid waste and less volume.
Sensor allows complete use of photoresist in
bottle.
Conversion to dry etch process.
JH
Vaate recovery DQOl
Process modification D001
Process nodifIcation D001
Process nodlflcatlon D001
Haste recovery DOOl
Solid
Solid
liquid
Liquid
Solid
77
100
100
50
1983 Haste solvent recovered from distillation tower,
still bottoms and skimming lro» wastewster holding
tanks is used as boiler fuel.
Core configuration was modified to remove resin
deposit and a still was purchased to recycle
•ethylene chloride used in cleanup.
1984 Solvent was eliminated from the process.
1984 Cleanup procedure for new latex adhesive
compounding equipment developed,
1983 Recycle heptane instead o£ scrapping it.
-------
TABLE 5.4 (continued)
reduction
method
3H (continued)
Process modification
Product foraatloa
Waste recovery
Recovery /reuse
Product modification
1 Process modification
U>
W
Process modification
Recovery /reuse
Production reformulation
Recovery/reuse
Product reformulation
Product reformulation
Process aodificstion
Waste
EPA
number
D001
D001
D001
B001
D001
D001
D001
D001
D001
D001
D001
D001
D001
reduced
Type
Solid
Solid
Solid
Solid
Solid
Solid
Solid
Solid
Solid/Air
Solid
Solid
Solid
Solid
Percent
reduced
50
100
100
30
60
ao
100
20
100
100
100
100
50
program
started
1983
1982
1983
198S
1985
1985
1985
1985
1985
1985
1985
1985
1985
Description
Process modifications preclude the need for one of
two kettle cleanings, and subsequent discharge.
A new starter solution containing no propanol was
substituted for 3H's which contains propanol.
Solvent Is recovered from wet scrap via stean
stripping and reused rendering scrap nonhazardous .
Sludge recrystalllzed to recover reusable product.
Seduction of defects in product resulted in less
rejection.
Teflon coated rolls eliminated scratches making
cleanup easier and less product scrapped .
Superior inline blender provided more stabilized
adhesive viscosity reducing scrapped adhesive.
Education program resulted ID fewer destroyed
druas containing nonpumpable waste.
Reformulated from a solvent based adhesive to a
noDsolvent based adhesive. *
Excess solvent reclaimed.
Reformulated to nonhazardous.
Substitution of water base adhesive for solvent
base.
Confidential.
(continued)
-------
TABLE 5.4 (continued)
ui
u>
Waste
reduction
method
3M (continued)
Product reformulation
Product reformulation
Product reformulation
Product reuse
Process modification
Process modification and
Waste
EPA
number
D001
B001
D001
D001
D001
B001
reduced
type
Solid
Solid
Solid
Solid
Solid/Air
Solid
Percent
reduced
100
60
100
100
10
100
progran
started
1985
1985
1985
1985
1985
1985
Description
Change in process equipment resulted in rapid
parts changing and cleaning.
Solvent replaced with higher flash point solvents.
Solventless adhesive replaced solvent based
adhesive.
Waste previously incinerated now sold to customer
for reuse.
Haste reduction by Improved sampling techniques.
Process and equipment Improvement. Improved
redesign
Product reformulation
Process modification
Process modification
D001
D001
D001
Solid/Air
Solid
Solid
50
100
100
charging accuracy.
1985 3H product replaced vendor product allowing
reduction in solvent usage.
1985 Process changed from water addition to heating
resulting In less product loss.
1985 Resin solutions that have a final viscosity under
or over spec, are now used in making other
products.
-------
TABLE 5.4 (continued)
Una*-*
waste
reduction
method
East oan Chealcal
Abatement
Recycling
Recycling
Recycling
Recovery
tn
I
fji Minimization
Ut
Minimization
Haste reduced
„_.„„„„„„ Yfifir
— - - -— 16&r
EPA Percent program
number Type reduced started Description
D001 Liquid 100 1985 Substituted Inflammable for flammable solvent.
D001 Spent catalyst 100 1986 Catalyse reclaimed for reuse.
D001 Liquid 99 1981 Recovery of solvent by distillation.
FOOS
F005
F003 Liquid 99 1983 Recovery of solvents by distillation.
FOOS
D001 Liquid 99 1981 Recovery of solvents by distillation.
F003
FOOS
FOQ1 Liquid and Solid 99 1981 Incineration In rotary kiln.
F002
FOOS
K009
K010
D001
FOOl Liquid and" Solid 99 1981 Incineration In rotary kilo and liquid
F002 lacloeratloa.
POOS
FOOS
DOOl
Source: References 14 and 15.
-------
TABLE 5.5. COST VS SAVINGS FOR WASTE REDUCTION PROGRAMS CARRIED OUT
BY IBM CORPORATION AT VARIOUS LOCATIONS
WASTE REDUCTION
Location
Austin, TX
Boulder, CO
East Fishkill, NY
Oswego, NY
Poughkeepsie, NY
year
1982
1983
1984
1982
1983
1984
1982
1983
1984
1982
1983
1984
1983
1984
103lbs
7,873
15,652
23,663
11,748
11,870
47,974
43,737
67,738
786
728
644
1,030
990
Cost of
program
$24,441
49,550
96,000
492,327
1,028,350
0
0
0
13,710
0
0
9,065
4,500
Savings
$1,360,007
3,022,170
7,853,154
3,094,077
4,028,511
14,495,197
13,714,699
39,157,600
7,910,000
11,320,000
11,320,000
3,251,978
2,114,152
ONSITE WASTE RECYCLING (103lbs)
Location
Austin, TX
Boulder, CO
East Fishkill, NY
Oswego, NY
Poughkeepsie
1984
9,177
2,153
33,521
311
495
1985
19,668
274
41,068
108
675
Source: References 14 and 15,
5-36
-------
Case 4: Southern Coatings, Sumter, SC—Southern coatings manufactures a wide
variety of coatings including solvent-based paints. Equipment
cleaning operations were producing a large volume of solvent
wastes. The company installed a distillation unit to recover
100 gallons/hour solvent containing primarily toluene and xylene
(also containing mineral spirits and VHP). The company achieved a
payback period of nine months. They estimate a raw materials cost
savings of $0.67/gallon solvent,
Case 5: Rexham Corporation, Greensboro, NC—Rexham Corporation does printing
and coating for a variety of applicators. Spent alcohol/acetate
solvents are produced during these operations. The company purchased
a batch distillation unit. They achieved a waste reduction of
60 percent in reclaiming the solvent mixture for use in reprocessing
and in cleanup operations.
Case 6: American Enka, (location unknown) American Enka is a nylon yarn
production and research facility. They have found it is
economically profitable and environmentally sound to recycle their
waste isopropyl alcohol solvent in-house rather than having it
recycled by an outside firm. They purchased a used distillation
unit and, with appropriate modifications, American Enka is now
saving $90,000/year. They have also been able to reuse the still
bottom residues as an asphalt emulsifier in another produce line.*-
Case 7: DeSoto Chemical, Greensboro, NC~-Wash solvents from manufacturing
operations are collected and stored. The solvents are then reused
during the next batch operation."
5-37
-------
REFERENCES
1. Huisingh, D,f et al. "Proven Profit from Pollution Prevention."
Conference draft. The Institute for Local Self-Reliance,
Washington, DC. July 1985.
2. Garner, P. Telecon with M. Kravett, GCA Technology Division, Inc.
W.R. Grace and Company. Lexington, MA. June 1986.
3. Roeck, D.R., et al. GCA Technology Division, Inc. Hazardous Waste
Generation and Source Reduction in Massachusetts. Bedford, MA. Contract
No. 84-198, MA Dept. of Env. Mgt., Bureau of Solid Waste Disposal,
June 1985 (Draft).
4. Kohl, Jerome, P. Moses, and B. Triplett. Managing and Recycling
Solvents: North Carolina Practices, Facilities and Regulation.
Industrial Extension Services, School of Engineering, North Carolina
State University, Raleigh, NC. December 1984.
5. Minnesota Waste Management Board. Hazardous Waste Management Report.
1983.
6. Versar, Inc. National Profiles for Recycling/A Preliminary Assessment,
Draft. EPA Contract No. 68-01-7053, U.S. EPA Waste Treatment Branch.
July 1985.
7. Hobbs, B., and R.R. Hall, GCA Technology Division, Inc. Study of Solvent
Reprocessors. Bedford, MA. EPA Contract No. 68-01-5960 (Draft),
U.S. EPA Office of Chemical Control, January 1982.
8. Radminsky, Jan, et al. Department of Health Services, Alternative
Technology and Policy Development Section. Alternative Technology for
Recycling and Treatment of Hazardous Wastes, Second Biennial Report.
California, July 1984.
9. Spivey, J.J., et al. Research Triangle Institute. Preliminary
Assessment of Hazardous Waste Pretreatment as an Air Pollution Control
Technique. U.E. EPA/IERL, March 15, 1984.
10. Engineering—Science. Supplemental Report on the Technical Assessment of
Treatment Alternatives for Waste Solvents. Washington, DC. U.S.
Environmental Protection Agency. 1985.
5-38
i
-------
11. Noll, K.E., et al., Illinois Institute of Technology. Recovery, Reuse
and Recycle of Industrial Waste. Chicago, IL. EPA-600/2-83-114,
U.S. EPA/ORD, Cincinnati, OH. November 1983.
12. Tucker, S.P., and G.A. Carson, NIOSH. Deactivation of Hazardous Chemical
Wastes. Cincinnati, OH. Environmental Sci. Technol., 19(3):215-220.
1985.
13. Tierney, D.R., and T.W. Hughes, Monsanto Research Corp. Source
Assessment: Reclaiming of Waste Solvents, State of the Art.
EPA-60Q/2-78-004f, 66 pp. April 1978.
14. Center for Environmental Management, Tufts University, Medford, MA.
"Waste Reduction: The Ongoing Saga" Conference, League of Women Voters
of Massachusetts and U.S. EPA. Woods Hole, MA. June 4-6, 1986.
15. Center for Environmental Management, Tufts University, Medford, MA.
"Waste Reduction: The Untold Story" Conference, League of Women Voters
of Massachusetts and U.S. EPA. Woods Hole, MA. June 19-21, 19S5.
5-39
-------
SECTION 6.0
MISTREATMENT
Pretreatment, in general, includes any process which precedes a final
treatment/recovery process. The purpose of the pretreatment step is to remove
constituents which would interfere with a subsequent process step or to
transform the waste in some other way that will make subsequent processing
simpler. Certain technologies, such as gravity settling, filtration and
flotation, are commonly labeled as "pretreatment" technologies. However, in
actuality, almost any technology, including those discussed in subsequent
sections, could serve as a pretreatment measure. For example, pretreatment of
a waste solvent may involve batch evaporation to remove dissolved solids.
Following this pretreatment measure, the recovered solvent (volatile fraction)
may then be fractionated to separate one or several reusable components.
Another example of the use of a "final" treatment technology as a
"pretreatment" measure involves the biological treatment of an aqueous
leachate stream prior to carbon adsorption as shown in Figure 6.1. In this
particular case, the purpose of carbon adsorption is to remove toxic
chlorinated organic compounds, however in addition to these compounds the
waste stream contains a large fraction of easily biodegraded organic
compounds. Biological degradation of these compounds prior to adsorption
reduced the organic load to the adsorbers, thus lowering the carbon usage rate
and cost, and produces an effluent of equal quality.
Therefore, pretreatment of a waste stream will be highly waste stream
specific: various combinations of processes may be used depending on the
characteristics of the waste and the requirements of the final treatment
process. In general, pretreatment measures may be broken down into several
6-1
-------
AQUEOUS LEACHATE CONTAINING SOLVENT
I
GRAVITY
SETTLING
i
pH ADJUSTMENT
(NEUTRALIZATION)
HEAVY ORGANIC PHASE
(TO INCINERATION)
GRAVITY SETTLING
(SEDIMENTATION)
i
BIOLOGICAL
TREATMENT
I
FILTRATION
i
CARBON
ADSORPTION
\
TREATED WATER
TO SEWER
SOLIDS (DISPOSAL)
-*^ BIOLOGICAL SOLIDS
SPENT CARBON TO REGENERATION
Figure 6*1. Process train for treating aqueous/organic leachate.
6-2
-------
categories based on the type of waste stream or constituent for which they are
most applicable. Depending upon the categorization scheme used, these
categories could consist of wastes with the following characteristics:
1. Liquid wastes containing suspended or floating materials
2. Solid/sludge wastes containing liquids
3. Inorganic solid wastes containing low levels of organics
4. Bulky, non-uniform solid wastes
5. Low heat content or high viscosity liquids
Certain technologies or techniques, including sedimentation, flotation,
filtration, mixing, blending, crushing, and screening, are used to pretreat
wastes with these characteristics prior to final treatment. The theoretical
principles of these technologies are discussed in many texts on unit
2345
operations. ' ' ' Therefore, one should consult these references if
further detail is desired. The application of the technologies to the five
waste types are discussed below.
6.1 LIQUIDS CONTAINING SUSPENDED OR FLOATING MATERIALS
This may be one of the most common situations where pretreatment is
required. Many treatment processes require that suspended solids, oils and
greases be removed prior to feeding into the unit. For example, granular
carbon adsorption generally requires that the influent contain less than
50 ppm suspended solids and less than 10 ppm of oil and grease. ' Common
methods of removing these materials include screening, sedimentation, and
flotation. Screening, sedimentation and filtration will remove suspended
materials that are more dense than the liquid phase, while flotation will
remove those that are less dense such as oil. In addition, dissolved air
flotation (DAF) may be used to remove solids with specific gravities close to
that of water, and emulsified materials. Settling, or decantation, is also a
commonly used pretreatment to separate water from contaminated solvents prior
to evaporation or distillation as shown in Figure 6.2. In this case, gravity
6-3
-------
MIXED SOLVENT STREAM WITH SOLIDS
i
SCREENING
GRAVITY SETTLING
(SEDIMENTATION)
FILTRATION
EVAPORATION
(THIN FILM, SCRAPED SURFACE)
FRACTIONATION
RECOVERED SOLVENTS
LARGE SOLIDS
SOLIDS
WATER
SOLIDS
STILL BOTTOMS
Figure 6.2. Process train for treating waste solvent.
6-4
-------
settling will result in a bottom layer consisting of organic solvent and an
aqueous layer above. The organic solvent layer is then distilled and the
aqueous layer is sent to the wastewater treatment plant to remove residual
organics using biological treatment or carbon adsorption.
6.2 SOLID/SLUDGE WASTES CONTAINING LIQUIDS
Frequently sludges or other solids are contaminated with free liquids,
primarily water. To reduce the volume of waste that must be treated, the
waste can be dewatered. Methods for accomplishing this include vacuum
filtration, centrifugation, and use of a filter press. The most common
application of these processes is to wastewater treatment sludges. Solids are
settled out of a wastewater stream in a sedimentation tank; the sludge formed
at the bottom will generally have a solids content of about five percent.
This sludge can then be dewatered using a vacuum filter centrifuge or a filter
press. These dewatering units will generally produce a sludge with ten to 40
percent solids depending on the specific process used and the type of sludge.
The liquid stream from the dewatering process will probably be recycled
through the wastewater treatment process while the thickened sludge can then
be incinerated or otherwise disposed. The dewatering of a sludge from five
percent solids to 30 percent solids will not only reduce the volume
significantly (from 40,000 Ibs. to 6,600 Ibs. for 1 ton of solids), but if the
sludge is incinerated much less auxiliary fuel will be required to evaporate
the entrained water.
6.3 INORGANIC SOLID WASTES CONTAINING LOW LEVELS OF ORGANICS
Solids are sometimes contaminated by low levels of organic compounds. In
most cases, wastes of this form are hazardous only due to the organic
contamination. Nevertheless, a common method of treatment is to incinerate
the entire waste matrix. This is extremely expensive because a large amount
of energy is required to heat up the non-combustible matter. It is much more
desirable to extract the contaminant from the solid portion of the waste, and
to incinerate only this portion as shown in the right—hand side of
Figure 6.3. Extraction of contaminants from solids such as soil is similar in
6-5
-------
AUXILIARY
?U!L ~~i
CONTAMINATED SOLID WASTES
SIZE REDUCTION
(SHREDDING/
PULVERIZING)
i
SOLIDS
INCINERATION
EXTRACTION
(SUPERCRITICAL
OR OTHER)
MIXING/BLENDING
WITH FUEL
LIQUID
INCINERATION
-CLEAN SOLID
CLEAN SOLID
Figure 6.3. Alternatives for incinerating contaminated solid wastes.
6-6
-------
principle to liquid/liquid extraction (which is discussed later) except that
the selection of the extracting solvent may be more critical. Since it is
desirable to return the soil to its place of origin, the solvent must be
nonhazardous. The EPA Mobile Soils Washing System (MSWS) uses water or water
Q
with nontoxic and/or biodegradable additives as the solvent. Another
variation to the extraction process is to use supercritical fluids to extract
the contaminants. A supercritical fluid is a substance that has been heated
above its critical temperature and compressed beyond its critical pressure.
Under these conditions fluids generally have greatly increased solute
capacities. In addition to water, whose application as a supercritical fluid
is discussed later, fluids that have potential in this application include
carbon dioxide and ethylene. Ihe use of supercritical ethylene to extract
trichlorophenol from soil has been studied at the University of Illinois.
Supercritical carbon dioxide has been utilized to remove compounds that have
been adsorbed to activated carbon (as an alternative to thermal
9
regeneration). The use of supercritical fluids to extract contaminants
from soil and other solids may become a common pretreatment operation in the
future.
6.4 BULKY, NON-UNIFORM SOLID WASTES
Contaminated solids can exist in a variety of shapes and forms. Many
treatment processes, however, require that solids be of a uniform shape and
size in order to be fed to the process. For example, the High Temperature
Fluid Wall process requires that solids be free-flowing, non-agglomerating and
smaller than 100 mesh (0.0059 inches) in size. Fluidized incinerators
generally require that solids be less than 1 inch in diameter. In order to
treat large, non-uniform solids with these processes, some type of •
pretreatment, as indicated in the left-hand side of Figure 6.3, that reduces
the size of the solids must be employed. Equipment to reduce size includes
shredders, hammermills, and crushers.
6-7
-------
6.5 LOW HEAT CONTENT OR HIGH VISCOSITY LIQUIDS
Final treatment of many solvent wastes involves destruction by
incineration. Sometimes, when the heating value of the waste solvent is too
low, it is possible to blend it with another waste or with fuel to produce a
mixture with a heating value high enough to support combustion ( > 8000
Btu/lb). Mixing or blending of a waste with another liquid is also sometimes
employed to reduce the viscosity of the waste so that it can be atomized.
Liquid injection incinerators generally require that the kinematic viscosity
of the waste be less than 150 SSU (Standard Saybolt Units) for proper
atomization. Preheating or blending of a high viscosity waste with, for
example, a number 2 fuel oil (maximum viscosity of 38 SSU at 100°F), are
options that could be considered to achieve the viscosity needed for good
atomization.
Pretreatment requirements for specific technologies are discussed in
following sections.
6-8
-------
REFERENCES
1. Ying Wel-ehi, et al. Biological Treatment of a Landfill Leachate in
Sequencing Batch Reactors. Environmental Progress. 5(1) February 1986.
2. Weber, W.J. Physicochemical Processes for Water Quality Control. John
Wiley and Sons. 1972.
3. Metcalf and Eddy, Inc. Wastewater Engineering: Treatment/Disposal/Reuse,
Second Edition. McGraw-Hill. 1979.
4. Arthur D. Little. Physical, Chemical, and Biological Techniques for
Industrial Wastes. Volume 1. Prepared for USEPA. 1977.
5. McCabe, W.L. and J.C. Smith. Unit Operations of Chemical Engineering.
Third Edition. McGraw-Hill. 1976.
6. Troxler, W.L., C.S. Parmele and G.A. Barton. Survey of Industrial
Applications of Aqueous Phase Activated Carbon Adsorption for Control of
Pollutant Compounds from Manufacture of Organic Compounds. Prepared by
IT Enviroscience, Inc. for Office of-Research and Development. U.S. EPA,
EPA-600/2-83-034.
7. Becker, David L. and Stephen C, Wilson. The Use of Activated Carbon for
the Treatment of Pesticides and Pesticidal Waters. Chapter 5 in Carbon
Adsorption Handbook, edited by P.N. Chereminisoff and F. Ellerbusch,
Ann Arbor Science Publishers, Ann Arbor, MI, 1978.
8. IT Corporation. Interim Summary Report on Evaluation of Soils Washing
and Incineration as On-Site Treatment Systems for Dioxin-Contaminated
Materials. EPA Contract No. 68-03-3069. 1985.
9. Eckert, C.A., et al. Supercritical Fluid Processing. Environ. Sci.
Technol. 20(4) 1986.
10. GCA Technology Division, Inc. Industrial Waste Management Alternatives
and Their Associated Technologies/Processes. Volume IV. Prepared for
Illinois Environmental Protection A~ency, Springfield, Illinois.
February, 1981.
6-9
-------
SECTION 7.0
PHYSICAL TREATMENT PROCESSES
The treatment processes discussed in this section are based on physical
methods of separation and generally do not result in destruction of the
contaminants in the waste feed stream. Processes discussed are:
7.1 Distillation
7.2 Evaporation
7.3 Steam Stripping
7.4 Air Stripping
7.5 Liquid-Liquid Extraction
7.6 Carbon Adsorption
7.7 Resin Adsorption
All of these processes are used to some extent for the treatment of
wastes, but differ in their applicability to various types of wastes and their
need for pretreatment and post—treatment procedures. Evaporation and
distillation, for example, are applicable to high organic content wastes Dut
usually generate a large volume residual that contains appreciable organic
contamination. Incineration is often the principal means of handling this
type of residual. The next three processes, air and steam stripping and
liquid-liquid extraction, are applicable to a broad concentration range of
contaminants, usually in aqueous media. Air stripping is generally applied to
low levels of volatile contamination, steam stripping is suitable for somewhat
higher levels of contamination, and liquid extraction possible for both low
and high contaminant levels. However, some level of residual contamination
requiring additional treatment can often be expected from these technologies.
Post-treatment options could be biological treatment or possibly adsorption
processes. Carbon and resin adsorption, the last two technologies discussed
in the section, are commonly used to remove trace amounts of contaminant to
achieve low organic concentration levels. However, in certain cases,
7-1
-------
particularly when contaminant recovery is advantageous, resin adsorption may
be considered for treatment of contaminant levels greater than 0.5 percent, a
level above which carbon adsorption may become too costly to be practical.
The merits of these processes, including technical performance and costs, must
be assessed to select the optimum process or processes. Processes based on
chemical and biological treatment and other technologies discussed in
following sections must also be considered in arriving at the optimum
selection of treatment system components.
The physical treatment processes (and the other treatment processes
discussed in following sections) are considered within the framework of four
major areas; i.e., (1) Process Description including pre-treatment and
post-treatment requirements, (2) Demonstrated Performance in Field and
Laboratory, (3) Cost of Treatment, and (4) Overall Status of the Technology.
Following discussions of various types of treatment processes applicable to
solvent and other low molecular weight organics (Sections 7,,0 through 13.0), a
review section (Section 14.0) is provided that addresses possible approaches
to identifying potential treatment processes or combination of processes that
is likely to meet treatment requirements in the most cost effective manner.
7-2
-------
7.1 DISTILLATION
7.1.1 Process Description
Distillation can be broadly described as a separation technique that
operates on the principle of differential volatility. More volatile
constituents can be enriched or separated by heating and volatilizing from
less volatile constituents. Most distillation systems fall into one of four
general categories.
1. Batch distillation;
2. Continuous distillation;
3. Batch fractionation; and
4. Continuous fractionation.
In batch distillation, the system is charged with a given quantity of
waste solvent and heated indirectly with steam or oil, with coils or the
vessel wall acting as the heat transfer surface. Heating continues until a
predetermined fraction of the volatile components is removed, as indicated by
the time/temperature profile of the charge within the still. At this point
heating is discontinued and the bottom product or less volatile residues
removed. The disposal of these residues will be limited by their
characteristics. In many cases, it will be expeditious to maintain a residue
that can be handled readily and that can be incinerated, i.e., that is
pumpable and has Btu value. In any event, it must be recognized that complete
removal of volatiles will not be feasible. Attempts to reduce solvent content
to low levels can result in compound destruction and equipment fouling.
Solvent removal efficiencies must balance the benefits of recovery against the
costs of equipment maintenance and the cost and manner of residual disposal.
Continuous distillation functions much the same as batch distillation,
except the waste solvent is charged continuously, resulting in steady-state
operation. Removal of the bottoms product can take place continuously or it'
7-3
-------
may be batch removed at specific intervals. Continuous bottoms removal
generally implies that the stream is pumpable. In response to economic
incentives to recycle and minimize wastes, several small stills have been
introduced to the solvent recovery market which permit close to complete
recovery as a result of design features such as removable liners for bottoms
disposal. Many can be operated either continuously or by batch mode and under
vacuum to effect even greater recovery.
The separation of solvent mixtures requires multiple distillations or
fractionation since adequate separation of fluids with similar boiling ranges
is not achieved in a single stage. Fractionation column designs include the
use of multiple trays or packing to maximize surface area so that rising
vapors are intimately contacted with falling condensate (reflux). These
columns can be operated as batch or continuous units. Figure 7.1.1 shows
basic process schematics for batch and continuous fractionation systems.
Fractional distillation is a very well developed industrial technology
that has been studied extensively for many years, particularly by the
petrochemical industry. Vapor-liquid equilibrium data provide the basis for
evaluating the feasibility of fractionation. A MeCabe-Thiele diagram, which
graphically illustrates the fractionation process, is shown in
2
Figure 7.1.2. The figure shows the minimum number of theoretical column
stages required to effect separation; i.e., 100 percent efficiency at total
reflux. Detailed discussions of fractional distillation theory and practice,
including the use of well-developed models for predicting separation, can be
found in chemical engineering texts and in Perry's Chemical Engineers'
Handbook.
In continuous fractionation, feed is constantly charged to the column at
a point which provides the specified top and bottoms product. The section of
the tower above the feed point is the rectifying or enriching section, and the
section below the feed point is the stripping section. A reboiler is
connected to the bottom of the fractionation tower to provide the reboil heat
needed for added reflux and better fractionation of complex mixtures.
Batch fractionation differs in that the charge is introduced at the bottom of
the tower. Consequently, it is possible to obtain a distillate of high
purity, but recovery of the less volatile component(s) must proceed in a
step-wise manner. As the more volatile solvent is taken off, the reflux and
7-4
-------
COLUMN
CONDENSER
REFLUX
OVERHEAD
PRODUCT
BOTTOM PRODUCT
1
"S.
a
S
ly
FEED
O
%
1
%
k
>o
\
§
S
«j
i
%
8
ii
U»
!<>
1
X
s^
*• ^COHOCNStH
d 2
1 I COifl
ACCUMULATOR L WATER
j ,. f „
. ^ |
g DISTILLATE
I
** '\J REFLUX
V/ 011 HP F
FEED
PLATE
VAfOR aeanit eo
fl_T{ ^ — 1 j * STEAM
^_ S- »»^n
BOTTOMS PRODUCT
BATCH FRACTIONAT10N
CONTINUOUS FRACTIONATION
Figure 7.1.1. Basic schematic for batch and continuous fractionation systems.
-------
cc
o
0.
u_
o
tr
u.
LU
_i
o
VAPOR
COMPOSITION
LINER
IDEAL
EQUILIBIUUM
STAGES
LIQUID
COMPOSITION
LINE
BOTTOMS
FEED
DISTILLATE
MOLE FRACTION OF A IN LIQUID
Figure 7.1.2.
McCabe-Thiele diagram for distillation.
(Total reflux =• no products taken off.)
7-6
-------
thus, energy consumption, must be increased to maintain overhead product
purity. At some point, the overhead quality decreases to the point where it
must be removed and stored separately until the less volatile component(s) is
enriched enough to generate a useful product. However, batch fractionation
does permit handling of wastes with higher dissolved solids content which
would foul the stripping zone surfaces in a continuous feed column.
7.1.1.1 Restrictive Waste Characteristics—
Batch Distillation
Batch distillation equipment has undergone considerable development in
recent years, particularly that of interest to the small quantity generator.
Almost any solvent waste can be processed in commercially available equipment,
including high solids, ignitable and potentially explosive mixtures. However,
there are some restrictions which affect safety and product purity that should
be considered.
Typically, the boiler temperature is set at 20°F to 30°F higher than the
boiling point of the solvent to ensure vaporization. Highly contaminated
solvents (i.e., high solids) And mixtures which form high boiling azeotropes
would require higher temperatures to effect separation. However, the
temperature of the still must remain below (20 to 30°F) the autoignition point
of the solvent. Thus, at least 40°C between autoignition point and boiling
point is required for distillation of high purity mixtures. Autoignition
temperatures for most solvents can be found in the National Fire
Protection Association's "Fire Protection Guide for Hazardous Materials".
Heptane is generally considered to have the lowest autoignition point (399°F)
4
of commonly recovered solvents. For this reason (to minimize risk of
explosion), many manufacturers design units which are limited to a maximum
temperature of 365°F.
Many small stills are explosion proof, but the potential for explosion
should be evaluated, especially for solvent mixtures. Autoignition
temperatures of ignitable wastes may pose particular handling problems in
distillation units since some of the particular solvents addressed in this
7-7
-------
document have even lower autoignition points Chan does heptane. Table 7.1.1
gives examples of some of these. It should be noted that there are several
other constituents not addressed by this document that exhibit even lower
autoignition points (see Reference 3).
Most halogenated solvents are nonflammable but are susceptible to thermal
decay. Thermal decay occurs at low temperatures relative to autoignition
points. For low autoignition point and low decomposition temperature
solvents, vacuum operation is an option that is available in many of the
currently marketed batch stills. Additional expenses, capital and operating,
are at least partially offset by the reduced boiling temperature and higher
potential recovery rates.
Solids content, which is a critical limitation for continuous distilla-
tion and fractionation processes is less critical for batch distillation or
continuous feed with batch bottoms removal. In particular, jacketed and
immersion heated boilers with provisions for easy bottoms removal are capable
of achieving high recovery rates. Residual solvents remain in the bottoms
produced from these operations, but potential recoveries are usually greater
than 90 percent regardless of initial solid constituent concentration. In
many cases, the distillation occurs over a long enough period of time such
that potentially reduced mass diffusion and heat transfer do not greatly
reduce the overall effectiveness of the recovery process.
ContinuousDistillation
Continuous distillation is subject to the same constraints that apply to
batch distillation in terms of operating temperature. However, the
distinction between batch and continuous bottoms removal greatly effects the
applicability and achievable performance for certain waste types.
Continuous feed, batch bottoms removal units are essentially the same as
batch units in terms of processing capabilities. Continuous feed units have
greater capital costs since they do require some additional control features
(i.e., level control), but are more automated thereby requiring less labor.
These units are preferred for high throughput applications where labor costs
become a higher fraction of total costs than capital equipment coats.
7-8
-------
TABLE 7.1.1. BOILING POINTS AND AUTOIGNITION POINTS
OF RISK SOLVENTS
Solvent
acrolein
ethanol
ethyl ether
1,3-dioxane
heptane
Autolgnltion Polnt(°F)
428
347
356
356
399
Boiling Point(°F)
125
70
95
214
209
Sources Reference 3.
7-9
-------
The continuous feed, continuous bottoms removal units are different in
that the achievable recovery is controlled by the ultimate disposition of the
bottoms. Continuous bottoms removal implies the bottoms remain fluid. In
cases where bottoms become highly viscous, ultimate recovery is limited by
equipment processing constraints. These materials are more optimally
processed in thin-film evaporators which are capable of achieving high
recovery rates and throughputs in a shorter period of time (see Section 7.2).
Scraped-surface stills also increase the ultimate recovery but by a smaller
amount. Thus, continuous bottoms recovery stills are best suited to
recovering spent solvents which produce residual bottoms that are non-fouling
and that remain pumpable after separation.
Batch Fractionation
Fractionation is a multi-stage process used for separating solvent
mixtures when the value of the pure component products justifies the
additional expenses associated with separation. Pure components can be sold
for 80 to 90 percent of the virgin price whereas solvent blends typically sell
for only 50 to 60 percent of virgin prices. As with batch distillation,
batch fractionation can handle a higher solids content waste form since these
materials do not come into contact with the packing or trays. The quantity
and nature of the solids in the waste may become a limitation depending upon
the design of the heat transfer unit. Excessive fouling may interfere with
heat transfer requiring higher energy costs, reduced throughput, and
additional labor. Agitated units are available to reduce the potential
problems due to fouling.
Continuous Fractionation
As opposed to batch fractionation, continuous fractionation is reserved
for materials which are essentially void of dissolved and suspended solids.
The feed enters at a mid-point in the column where it comes into intimate
contact with tray or packing surfaces. Labor costs associated with cleaning
these units justify pretreatment in either a distillation or evaporation
unit. Also, the bottoms product will have to be treated to remove the
nonvolatile constituents.
7-10
-------
Azeotrope Formation
Two limiting characteristics of a waste stream effect the application of
both batch and continuous fractionation. As described for batch distillation,
temperature limits must be considered for constituents with low autoignition
and thermal decomposition limits. A second potential separation problem
arises with the formation of azeotropes when, at a certain limiting solvent
concentration, two or more constituents have the same relative volatility,
making further enrichment impossible.
Table 7.1.2 provides a list of water/solvent azeotropes with azeotropic
and pure component boiling points and overhead product composition. All of
the systems shown form low boiling azeotropes with water, with nonmiscible
compounds separating into two phases as indicated. Table 7.1.3 provides
further data on multicomponent organic systems that form tertiary and
quaternary azeotropes with water. Additional azeotrope data can be found in
the literature. Minimum boiling azeotropes represent the limiting
concentration obtainable in an overhead product unless the azeotrope is
disrupted through the addition of another constituent, e.g., extractive
distillation. This is discussed further in Section 7.2. In certain cases ,
azeotropes may not occur, but relative composition differences between the
liquid and vapor phase might be so small as to render the separation
economically unfeasible. The vapor—liquid equilibrium (VLE)data must be
evaluated for separational feasibility. The McCabe-Thiele diagram
(Figure 7.1.2) is based on VLE data. It is evident that if the vapor and
liquid composition lines are close enough, many equilibrium stages might be
required.
7.1.1.2 Operating Parameters and Design Criteria—
Distillation and fractionation processes are based on the evaporation and
condensing of constituents. Operating parameters of critical importance for
all units are process temperature and process pressure. Higher operating
pressures are routinely used for low boiling constituents to avoid the use of
energy intensive refrigeration to achieve condensation. This is particularly
true for fractionation processes which requires greater reboiler and condenser
7-11
-------
TABLE 7.1.2. AZEOTROPE FORMATION OF COMPOUNDS WITH WATER
Solvents of Concern
1, l,2-Trichloro-l,2 ,2-tri£luoroethane
1-Butanoi (1)
Carbon disulfide
Carbon tetrachloride
Chlorobenzene
Cyclohexanone (I)
Ethyl acetate (I)
Ethyl ether (I)
Ethylbenzene
Isobutyl alcohol (I,T)
Methsnol (I)
Hethyl ethyl Icetone
-------
TABLE 7.1.3. SAMPLE DATA ON AZEOTROPE FORMATION
Binaries
A
Methylene Chloride
Methanol
Berehloroethylene
Ethanol
Acetone
Carbon Tetrachloride
Tertiaries
A B
Hater Ethanol
Ethanol
Ethanol
Ethanol
Methanol Acetone
Quarte rnar ie s
A B C
Water Ethanol Benzene
B
Methanol
Ethanol '
Dichlorobenzene
1, 1, 1-Trichloroethane
Acetone
Methyl Acetate
Ethyl Acetate
MEK
Ethanol
Methyl Acetate
Ethyl Acetate
Toluene
Heptane
Benzene
Methyl Acetate
Heptane
Ethyl Alcohol
MEK
Ethyl Acetate
C bp°C
Ethyl Acetate 70.3
Benzene 74. 1
MEK 73.2
Toluene 74.4
Hexane 47
D bp°C
Cyclohexane 62.2
Water Butyl Alcohol Butyl Acetate Butyl Ether 90.6
bp°C
37.8
39.9
171.3
56
55.5
54
62.3
64.5
76.7
56.9
71.8
76.7
70.9
67.9
56
55.8
65
50
73.7
74.8
Pe rcent
A B
9 8.4
7.4 18.5
11 14
12 37
14.6 30.8
Itereent
A B
7.1 17.4
30 ' 13
%A
92.7
95
48
21.7
12
19.5
44
70
37
.3
31
68
49
31.7
50
89.5
84
84.3
81.6
57
C
82.6
74.1
75
51
59. 6
C D
21.5 54
51 6
Source: Reference No. 5
7-13
-------
duties as a result of occurring reflux. Distillation units, particularly
small batch units capable of processing up to 60 gallons per hour, are often
equipped with vacuum capability. High temperatures combined with low pressure
makes high recoveries possible, even for high boiling constituents.
Other parameters that must be considered include batch time, viscosity
(flow and mass transfer), reflux ratio, and the location of feed introduction
to the column. Batch time is chosen based on economics, desired recovery, and
restrictive waste characteristics. Certain units may be susceptible to
fouling problems or viscosity problems. High viscosity wastes are best
treated in agitated thin-film evaporators (see Section 7.2).
Reflux rate and feed tray location are process variables strictly
applicable to fractionation, with feed tray designation, applicable
specifically to continuous distillation. Reflux rate or ratio is set based on
the economic evaluation of product purity versus utility costs. High reflux
ratios produce higher purity products but are more energy intensive. In
addition, the high internal flow rates often establish the need for either
larger units, or lower throughputs. Optimal feed location in a column will be
at the point of intersection of the rectifying and stripping operating
lines.
In reference to these process considerations and other design criteria,
Custom Organics, Inc. compared batch and continuous fractionation based on
experience in the production of specialty organic pi
of continuous fractionation were stated as follows:
9
experience in the production of specialty organic products. The advantages
1. A continuous fractionating column has both a rectifying section in
which the low boiling point component is being enriched and a
stripping section in which the higher boiling component is being
enriched. The available trays are more effective in this
configuration than as a single rectifying section.
2. Increased separation efficiency for continuous fractionation is best
achieved through increasing the number of trays. In a batch still
the most effective way of improving the separation is to increase
the reflux ratio which reduces production capacity and energy
efficiency.
7-14
-------
3. The rate of separation of components is greater in a continuous
still than in a batch still.
4. Liquid holdup volume is not a factor in the design of a continuous
fractionating process because the temperature/composition profile is
constant* In batch columns, the holdup volume critically affects
the column efficiency.
5. Constant operating conditions enable predicted performance of
continuous fractionation units to be determined with some degree of
accuracy. Conversely, mathematical models for batch units are
complex and less exact.
6. Optimization of a continuous distillation column involves a fairly
simple balance between product quality and increased operating costs
associated with a higher reflux ratio. A batch column requires a
trade-off for optimum operation between a constant reflux ratio and
a constant overhead product composition.
7. Because the composition of all streams is fixed in continuous
fractionation, it is easier to plan and manage process and quality
control.
8. The flexibility of changing the feed and take off trays with
continuous column make continuous fractionation a very powerful
separation technique. Any tray in the column can be used for feed
or product withdrawal, and it is not unusual to have three product
streams withdrawn simultaneously. In contrast, the only product
from a batch unit is distillate. It is more difficult to seperate
multicomponent vapor streams due to continually changing
concentrations.
9. Short product residence time makes continuous fractionation very
much superior to batch for temperature sensitive materials.
10. The time required for handling or housekeeping operations is minimal
for continuous relative to batch fractionation.
However, continuous fractionation has several disadvantages. It requires
considerably more skill to operate than batch systems. Batch distillation is
also capable of handling wastes with suspended or dissolved solids without
prior evaporative treatment. In addition, it not always possible to remove
products as desired from a multicomponent waste in a continuous column. Batch
processing offers the capability of recovering different products at different
times when potentially more than one column may be required for a continuous
process. Another potential limitation of continuous fractionation is the
7-15
-------
effect of inconsistent feed composition. The continuous column is sensitive
to change, and a change in feed composition could upset steady-state operation
and adversely impact product quality.
As mentioned previously, fractionating towers can be either a packed
tower or tray (plate) tower. The choice between use of plate or packed tower
should be based on a cost analysis. In many cases, however, the decision can
be made by a qualitative analysis of the advantage and disadvantages
eliminating a detailed cost comparison. The following summarizes the
advantages and disadvantages of plate versus packed towers as presented by
Beters and Timmerhaus in Plant Design and Economics for Chemical Engineers.
* Stage efficiencies for packed towers must be based on experimental
tests with each type of packing.
• The design of plate columns is more reliable as a result of liquid
dispersion difficulties in packed columns, especially for the case
at a low liquid to vapor flow ratio.
* Plate towers can be designed to handle wide ranges of liquid rates
without flooding.
* Plates are more accessible for cleaning.
* Plate towers are preferred if interstage cooling is required.
Cooling coils can be installed on the plates or the liquid-delivery
line from plate to plate can be -passed through an external cooler.
* Distillation packing may be crushed by thermal expansion or
contraction brought on by large temperature changes.
* Random-packed towers are seldom designed with diameters larger than
4 ft, and diameters of commercial plate towers are seldom less than
2 ft.
• Packed towers are usually preferred if the liquids have a large
tendency to foam.
* The amount of liquid holdup is considerably less in packed towers.
* The pressure drop through packed towers may be less than the
pressure drop through plate towers designed for the same duty. This
advantage, plus the 'fact that the packing serves to lessen the
possibility of tower-wall collapse, makes packed towers particularly
desirable for vacuum operation.
7-16
-------
In a report provided for EPA, the Water General Corporation summarized
>erformance of pack:
Tables 7.1.4 and 7.1.5.
the performance of packing and different tray types as presented in
11
7.1.1.3 Treatment Combinations—
For distillation, pretreatment options consist of filtration,
centrifugation and other physical means (see Section 6) to separate solids
from the liquid stream and decanting to separate gross sediment and immiscible
fluids.
Post treatment methods used for overhead product include further
refinement through separation of mixtures into its pure components. This is
performed to enhance the value of the recovered solvent or to meet product
purity specifications. Fractionation is usually performed on spent solvents
which have already been separated from nonvolatile constituents. It is also a
typical regeneration technique for solvent used in solvent extraction. In
cases where the distillate product consists of two phases, decanting is a
typical post-treatment procedure. If water is soluble in the solvent to an
extent which exceeds product purity specifications (e.g., methylene chloride),
the solvent stream will undergo some form of drying to remove the water,
Coawionly employed drying methods include molecular sieve, calcium chloride,
ionic resin adsorption, and caustic extraction.
Post treatment options for bottoms products depend on the physical form
of the material. Approximately two-thirds of recycled solvent bottoms
generated by commercial reeyclers are used as is or blended with higher Btu
products for use as a fuel (Section 3.2). For solvent blends containing
expensive halogenated constituents, there are cases where liquids are added
prior to distillation to promote maximum removal of the halogenated
constituents by keeping the bottoms fluid. Other treatment options consist of
further solid-liquid separation for organic liquids and sludges, organic
removal or extraction processes for aqueous wastes, and thermal destruction
techniques.
Solid-liquid separations can be-accomplished by physical means such as
centrifugation, filtration, decanting and extraction. Dilute aqueous wastes
can be treated through air or steam stripping, carbon or resin adsorption, or
biological treatment as described in other sections.
7-17
-------
TABLE 7.1.4. RELATIVE PERFORMANCE RATINGS OF TRAYS AND PACKINGS*
Trays
Parameter
Vapor capacity
Liquid capacity
Efficiency
Flexibility
Pressure drop
Cost
bubble cap
3
4
3
5
3
3
sieve
4
4
4
3
4
5
valve
4
4
4
5
4
4
Packings
high void
5
5
5
2
5
1
normal
2
3
2
2
2
3
*5 = excellent; 4 = very good; 3 = good; 2 » fair; 1 = poor.
Source: Reference 11.
TABLE 7.1.5. SELECTION GUIDE FOR TOWER INTERNALS*
Trays
Parameter bubble cap
Pressure, low
moderate
high
High turndown ratio
Low liquid flow rates
Foaming systems
Internal tower cooling
Suspended solids
Dirty or
polymerized solutions
Multiple feeds or
sidestreams
High liquid flow rates
Small diameter columns
Column diameter 1 to 3 m
Larger diameter columns
Corrosive fluids
Viscous fluids
Low pressure drop
Expanded column capacity
Low cost
Reliability of design
1
2
2
3
3
1
3
1
1
3
1
1
2
1
1
1
0
0
1
2
sieve or valve
2
3
3
2
1
2
2
2
2
3
2
1
3
3
2
2
1
2
2
3
Packings
random
2
2
2
1
I
3
1
1
1
1
3
3
2
2
3
3
2
2
2
2
stacked
3
1
0
2
2
0
0
0
0
0
0
2
2
1
1
0
3
3
1
1
*0 «= do not use; 1 = evaluate carefully; 2 - usually applicable;
3 = best selection.
Source: Reference 11.
7-18
-------
7.1.2 Demonstrated Performance
A lumber of studies have been conducted demonstrating the feasibility of
solvent recovery by distillation processes. Although residual solvent
concentrations in treated wastes were always high enough to require additional
treatment, all these studies demonstrated significant economic benefits
similar to those shown previously in Section 5. The cost benefits of the
studies discussed below will be discussed in detail in Section 7.1.3.
3M Company: Packaged Batch Distillation Units, Reference 12
Small batch distillation units for recovery of solvents from low volume
12
wet scrap were evaluated by the 3M Company in 1984. Many of the wastes
were described as nonpumpable in nature, which limited the options for
recovering the solvents to small batch stills. Units built by
Zerpa Industries and the Finish Engineering Company were evaluated for three
different applications, as shown in Table 7.1.6, Table 7.1.6 also summarizes
the achieved recovery rates and stream compositions.
Application 1:
Application No. 1 involved the recovery of 1,1,1-trichloroethane
(1,1,1-TCE) from degreasing scrap. The scrap contained 84.1 percent
1,1,1-TCE, 6.4 percent oil and grease, 6.6 percent total solids, and
2.9 percent unidentified inhibitors which act as stabilizers for 1,1,1-TCE.
This scrap responded well to distillation resulting in an overhead rich
in low boilers (160 to 178°F boiling point range of 1,1,1-TCE and inhibitors)
and bottoms product rich in high boilers (300°F for these oils) and
nonvolatiles (solids). Batch distillation of 2 gallons per hour of scrap at
290°F resulted in 99 percent solvent recovery. The overhead was clear, and
contained 95 percent 1,1,1-TCE, 0.02 percent oil and grease, 0.02 percent
total solids, and a 4.96 percent unidentified component which was believed to
consist of the inhibitor system. Twelve percent of the feed weight was
recovered as a bottoms product. This consisted of a solid cake residue
containing freely drainable oil and 8.2 percent by weight 1,1,1-TCE,
7-19
-------
TABLE 7.1.6. SMALL STILL PERFORMANCE EVALUATION
Application 1; Recovery; 99Z. Separation of Low and High Boilers
Stream Composition (I)
1,1,1-trichloroethaneOil/grease
Solid
Inhibitors
Comments
waste feed
distillate
bottoms
84.1
95
8.2
6.4
0.02
NK
6.6
0.02
HE
2.9
4.96
NR
clear
HE; Hot Reported
Application 2; Recovery: 76%. Processing High Viscosity and High Solids Scrap
Stream Composition (Z)
Solvent
Solid
Comments
waste feed
distillate
bottoms
79.0*
99.9
b
21.0
0.002
HE
clear
elastic cake
•mostly heptane/remainder naphthones and other paraffins (IP 190-219°F)
"subjected to EPA leach test: leachate contained 43ppm heptane
Application 3:
Stream
waste feed
Recovery: 97 .4%.
MEK/ Toluene
48.0
( mmluu™_-_^.™
f M™^ ^..«. *
Processing Solvent Contaminated
Composition (%)
Other Solvents
13
— — QQ QQ-.— — _ — .— — ~™"i
_,___.,/, _____ \ a
Refuse
Solid
39
Om
Q&
ComffleaCB
—
"subjected to EPA leach test: leachate contained 2ppm MEK,
1.7ppm toluene, 900ppm unidentified
Source: Reference 12.
7-20
-------
The economics of using a wet scrap processing are attractive, with
investment payback period estimated to be 4 months, after considering credit
for reclaimed solvent and reductions in waste transportation and disposal
costs.
Application 2:
Application No. 2 involved processing a difficult-to-treat, highly
viscous (10,000 cp) adhesive scrap containing heptane and other solvents
(79 percent) and high solids content (21 percent). Previous attempts were
made within 3M to recover solvent from this scrap through spray drying and
thin film evaporation. Spray drying failed when the processed scrap became
tacky and plugged the dryer. Thin film evaporation was successful, but was
not economical due to the low volume of waste generated (44,400 gal.yr).
Distillation of this scrap produced an overhead rich in low boilers
including heptane, naphthenes and other parafins (190 through 219°F boiling
point range) and a bottoms product consisting mainly of nonvolatiles
(solids). Batch distillation of 2.7 gal/hr of scrap at 350°F resulted in
76 percent solvent recovery. The overhead was clear and contained
99.9 percent solvent and 0.002 percent total nonvolatiles. Forty percent of
the feed weight was recovered as bottoms product. This product resembled a
nonflowable elastic cake and contained 55 percent solvent. The bottoms
product was subjected to the standard EPA leach test. The leachate contained
43 ppm heptane and 1 ppm of another material that was unidentified*
The economics of using a wet scrap processing still were judged to be
favorable with the investment payback period estimated to be 17 months.
Application 3:
Application No. 3 involved processing solvent contaminated industrial
refuse which primarily consisted of low-density rags and filter cartridges.
This scrap required wet scrap processing units with larger distillation
chambers relative to previous tests. The nature of this scrap (high solids
content and low packing density) was expected to result in poor heat transfer,
long processing times, and a wet bottoms product but these problems did not
materialize.
7-21
-------
The scrap tested contained 61 percent solvent and 39 percent solids. MEK
and toluene accounted for 48 percent of the feed weight. Distillation of this
scrap produced an overhead rich in low boilers and a bottoms product
consisting of nonvolatiles (solids). Batch distillation of 3.4 gal/hr of
scrap at 350°F resulted in 97.4 percent solvent recovery. The overhead was
clear and contained 99.99 percent solvent and 0.01 percent solids. Forty-one
percent of the waste was recovered as bottoms product, which was a hard
conglomerate of dried rags and filter cartridges. Pour percent of this
bottoms product's weight was solvent. The bottoms product was subjected to
the standard EPA leach test. The leachate contained 2 ppm MEK, 1.7 ppm
toluene, and 900 ppm unidentified material.
The economics of using a wet scrap processing still are promising . The
estimated investment payback period was 6 months.
GCA: Package Batch Distillation Unit
GCA, under contract to EPA, evaluated the performance of a Zerpa RX-35
Recyclene still used for the reclamation of spent solvent used as a printed
13
circuit board developer. The application involved the recovery of
1,1,1-trichloroethane (ltl,l-TCE) contaminated with undeveloped photoresist (a
100 percent organic thermoplastic polymer consisting of acrylic monomers and
polymers and small quantities of chlorine substituted nitro heterecycles).
The spent developing solution, concentrated through a previous batch
distillation process using a Dupont-Riston SRS-120 unit, consisted of
88 percent 1,1,1-TCE, 2 percent solids, and 10 percent breakdown products and
additives. A 25 gallon charge distilled at 350°F for 2 hours in the Zerpa
still resulted in 99.9 percent solvent recovery. The overhead was
contaminated with less than 0.001 percent total solids and was suitable for
reuse in the developer. The bottoms consisted of a dry solid residue
containing 7.5 percent 1,1,1-TCE and 1 percent breadkown products and
additives (see Table 7.1.7). The unit was able to achieve a 6 month payback.
7-22
-------
TABLE 7.1.7. SUMMARY: GCA/OSW CASE STUDY
Recovery: 99%. Recycling Photoresist Solvent
Composition (wt %)
Stream
waste feed
distillate
bottoms
1, 1, 1-triehloroethane
88.0
8.0
7.5
solid
2.0
0.001
91.5
other
10.0
12.0
10.1
Source: Reference 13.
GCA: Batch Fractionation Process
GCA conducted a waste minimization study for EPA in 1986 involving a
batch fractionation process used for solvent recovery at a printed circuit
board and semiconductor manufacturing firm. The solvent was a methyl
chloroform and Freon TF (trichlorotriflouroethane) mixture. The Freon and
methyl chloroform are used to preclean substrate semiconductor chips and
develop photoresist, respectively, producing a waste which typically contains
up to 10 percent methyl chloroform and 90 percent Freon.
The waste is distilled in a stainless steel mesh packed column. The tower
is 28 feet tall and 12 inches in diameter. Waste solvent is automatically fed
from a 1,500 gallon holding tank to a 1,450 gallon batch pot using level
controllers in the batch pot. The pot is equipped with a reboiler which heats
the solvent mixture to 133°F. Other process parameters are reported in
Table 7.1.8.
As the distillation proceeds, the concentration of methyl chloroform in
the batch tank increases from an initial concentration of 10 percent in the
waste feed to 80 percent after 11 to 13 days. Upon reaching 80 percent, the
still bottoms are removed for further processing. Analytical results from
samples obtained during the site visit are provided in Table 7.1.9. These
data were collected after 6 days of processing which is the reason for the
relatively low bottoms concentration.
7-23
-------
TABLE 7.1.8. METHYL CHLOROFORM/FREON BATCH DISTILLATION SYSTEM
OPERATING INFORMATION
Normal Operating Parameters
1. Solvent
Recovery Rate
Reboiler Temperature
Distillate Temperature
2. Cooling Water
Flow
Temperature
3. Steam
Flow
4. Electrical Requirements
33 gal/hr
133°F
75 °F
72 gal/min
47°F
470 Ibs/hr
30 hp
Source: Reference 13.
TABLE 7.1.9. DISTILLATION COLUMN RESULTS
Parameter
Volatile Organics (% w/w)
- Freon TF
- Methyl Chloroform
- Other
Solids (mg/kg)
Influent
96,1
3.9
0.1
1.2
Recovered
product
99.1
0.9
0.1
0.06
Still
bottoms
51.5
47.5
0.1
27
Virgin
Freon TF
99.99
0.01
0.01
0.13
Source: Reference 13.
7-24
-------
The column is operated continuously and is capable of handling up to
1,200 gallons per hour. Approximately 40 million pounds of solvent were
processed in 1984, with estimated savings in excess of $8 million.
CH2M Hill: Batch Packaged Stills14.
CH2M Hill evaluated 18 case studies of waste recycling and minimization
for the DOD Environmental Leadership Project and the U.S. Army Corps of
Engineers, some of which provided economic justification for various solvent
distillation processes. A study of a solvent recycle program at Robins AFB is
described below.
Robins AFB operates a batch, atmospheric pressure still ($48,000 purchase
price) manufactured by Finish Engineering Corporation which was used to
reclaim trichloroethane, Freon-113 (trichlorotriflouroethane), and
isopropanol. A facility engineer estimated that in 1984 the recycling of
these three chemicals saved the base $118,000 in virgin material and hazardous
waste disposal costs. Recovery costs were $13 per drum, whereas disposal of
the chemicals and repurchase of new materials would have cost from $250 to
$500 per drum.
The organic fluid recovery system consists of a single—stage batch still,
a solvent/water separator, and an electrically powered steam generator. The
still can operate up to a boiler temperature of 300°F and can reclaim organic
fluids at a rate of up to 55 gallons per hour. Freon and isopropanol were
processed at a rate of approximately 50 gallons per hour and trichloroethane
was processed at a rate of 35 to 40 gallons per hour. Recovery efficiency for
isopropanol and Freon-113 is approximately 95 percent. The recovery
efficiency for trichloroethane is only 70 percent since the used material
contains nonvolatile waxes, dirt, and greases that are removed from metal
parts during degreasing operations.
The recovered Freon does not meet Type I military specifications;
however, it does meet Type II military specifications and is consequently used
for initial cleaning. Virgin solvent is required for final cleaning
operations that require Type I Freon. Approximately 175 drums per year of
trichloroethane were being reclaimed. Laboratory tests indicate that the
recovered trichloroethane meets military specifications and that reclaimed
7-25
-------
isoproponal is 99.8 percent pure. A 5 micron filter, installed at the
condenser outlet to remove fine metal particles carried over with the vapor,
was found to enhance recovered product purity.
Robins AFB recently purchased a second still from Finish Engineering for
$97,000 to supplement their existing unit. Operating under vacuum, this new
system will have the capability of distilling organic fluids that have
atmospheric boiling points of up to 500°F while maintaining a 300°F limit in
the boiler pot. This new still is to be used to recover materials, such as
Stoddard dry cleaning solvent and silicone damping fluid, that cannot be
reclaimed with the existing still. The new still will also be used to reclaim
materials, such as paint thinners (e.g., MEK and toluene) and Coolanol 25R
fluid, that were not being recovered because of inadequate capacity. The
total potential savings in material costs and disposal costs for the new still
is expected to be $315,000 per year.
7.1.3 Cost of Treatment
Cost data, developed for the three 3M batch distillation unit
applications, the GCA batch distillation and fractionation process studies
discussed above in Section 7.1.2,, will be presented here followed by
discussion of three recent studies not previously presented. The last three
studies are a CH2M Hill study of cost savings derived from a solvent
recovery program at the Norfolk Naval Shipyard, a Pace Company examination of
the costs of large scale recovery units , and a study by the U.S. Navy of
the costs of small scale package batch distillation units. Manufacturer
profiles of the distillation units evaluation by the Navy are presented in
Appendix B.
7.1.3.1 Cost Estimates for Small Packaged Batch Distillation Units—
The economic benefits to be derived from solvent recovery operations by
the 3M Company in 1984 using small packaged batch distillation units has been
described in Reference 12. For the three applications described earlier in
Section 7.1.2, projected savings are as estimated in Table 7.1.10. As shown,
cost savings based on test results, will be significant for the three
applications.
7-26
-------
TABLE 7.1.10. PROJECTED ECONOMIC BENEFITS
•-J
1
to
-J
Application
1. Separation of Low and High
Boilers
2. Processing High Viscosity and
High Solid Content Scrap
3. Processing Solvent Contaminated
Industrial Refuse
4. Alternative to Utilizing Vendor
Solvent Recovery Services
Volume
to be
processed
(drums
/yr)a
96
808
1,800
1,350
Projectedb
savings
(t/yr)
23,000
53,700
137,000
108,000
Estimated
wet scrap
processing
operating
cost
U/yr)
1,400
25,600
31,300
20,300
Net
savings
(*/yr)
21,600
28,100
105,700
87,700
Total
estimated
i capital
costc
-------
3M's evaluation also included a fourth application shown in the table,
involving the processing of waste solvent currently shipped offsite to solvent
recovery services. The plant generates 74,250 gallons of pumpable waste
annually. The waste contains heptane, toulene, MEK, alcohol, and
1,1,1-trichloroethane with an average solids content of 10 percent. The
utilization of in-house wet scrap processing technology would elminate the
current reclaimer and transportation costs, the associated liability risk, and
raw materials-purchase cost. Payback period for an in-house system was
estimated to be 8 months as shown in Table 7.1.10.
7.1.3.2 Cost Estimates for a Packaged Batch Distillation Unit—
GCA evaluated the performance of a Zerpa RX-35 recyclene unit used in
combination with a DuPont Riston-SRS-120 still used to recover solvent wastes
13
for a printed circuit board manufacturer. Previously, the concentrated
spent developer solution was sent offsite for disposal and raw material was
then purchased for make-up. Use of the Zerpa RX-35 unit resulted in the
recycling of 10,605 gallons of 1,1,1-trichloroethane (annual basis) which
resulted in a payback period of 6 months. Projected net savings were
$49,100/year, after deduction of operating costs of $8,300/year. This
reclamation effort represents a significant reduction in waste volume
(99+ percent), regulatory compliance costs, and possible future liabilities.
7.1.3.3 Cost Estimates for the Batch Fractionation Process—
As part of the Reference 13 study, for solvent recovery operations
conducted at a printed circuit board manufacturing firm, GCA developed a cost
analysis. A comparison was made between the cost of onsite recovery of Freon
TF versus the costs of offsite recovery. Factors and cost estimates are
presented in Tables 7.1.11 and 7.1.12 respectively. They are based on the
quantities of waste solvent and still bottoms that were generated in 1984.
Most of the cost factors (unit costs) in Table 7.1.11 are self-explanatory,
however those pertaining to offsite residue disposal (or management) were
generated by GCA survey and assumption. Based on discussion with several
waste disposal firms, a per gallon cost was derived for the offsite recovery
of spent solvent in the hypothetical case that onsite distillation was not
practiced. In the former case, it was assumed that the waste mangement
7-28
-------
TABLE 7.1.11. UNIT COSTS
Electricity
Steam
Cooling water
Operating Labor
Engineering
Other Capital Costs
Annual!zed Capital
Methyl Chloroform
Freon 113
Offsite Recovery of Spent Solvent
(Residue Disposal)
Offsite Recovery of Still Bottom
(Residue Disposal)
- $0.05/kw-hr
- $0. 96/1,000 Ib
- $0.25/1,000 gal
- $15/hr
- 10% of equipment cost
- 10% of equipment cost
- Based on 10 years and 10% interest
- $4.50/gallon
- $11.92/gallon
- $0.25/gallon credit
- No charge
Source: References 10, 13, 14, 15.
7-29
-------
TABLE 7.1.12. ESTIMATED COST OP RECOVERY OP FREON/METHYL CHLOROFORM
CO
o
Ons ite rec
Cost item Quantity
Capital Cost
Equipment 1
Engineering
Other
TOTAL CAPITAL
Annual O&M
Electricity 1.5 kw
Steam 470 Ibs/hr
Cooling Water 4,302 gal/hr
Labor 2,800 hrs/yr
- Maintenance - —
- Residue Disposal 119,500 gallon
TOTAL O&M
Annual Cost
Annualized Capitol
Annual O&M
Solvent Cost 23,900 gallon
TOTAL COST
sovery
Cost
130,000
13,000
13,000
156,000
600
3,800
9,100
42,000
15,600
0
71,100
12,600
71,100
190,900
274,600
Offsite recovery
Quantity Cost
0 0
0
295,800 gallon (74,000)
(74,000)
(74,000)
176,300 gallon 1,408,000
1,334,000
Source: Reference 13.
-------
facility would accept the still bottoms, containing 5-10 percent nonvolatiles,
for no charge, but no credit. In the latter case, where the solvent would
contain less than 1 percent nonvolatiies, the waste management facility would
give the generating facility a 25 cent per gallon credit for each of the
solvent types. Also, it was assumed that the industrial facility would pay
two-thirds of the current price for virgin solvent.
The equipment cost is based on the use of a. 1,200 gallon per day APV
batch distillation system. The $130,000 FOB price includes a 28-foot
distillation column with metal mesh packing, a 1,450 gallon batch pot, a U
bundle reboiler, U bundle condenser, shell and tube vent condenser, bottom and
top product pumps, bottom and top product shell and tube coolers, and
instrumentation for automatic operation as provided by APV Crepaco (1986).
The unit would be preassembled and include all valves and piping.
O&M costs for this system are based on operating this unit 24 hours per
day, 350 days per year. This unit is operated virtually continuously as long
as spent solvent is available for input. Since the unit is equipped with
instrumentation to allow for automatic operation, it need not be monitored
full time. Of 24 hours, 8 hours of operating labor was assumed to De required,
Onsite recovery of Freon results in a cost savings of approximately
$1 million per year as compared to offsite recovery.
7.1.3.4 Study of the Norfolk Navy Shipyard Operation—
At the Norfolk NSY, approximately 15 gallons per day of numerous waste
solvents including mineral spirits, ketones, and epoxy thinners containing
paint pigments are generated in the paint shop during cleaning operations.
Historically, the waste solvents have been disposed of at a cost of $7.80 per
gallon. A nonfractionating batch still, Model LS-15V, manufactured by Finish
Engineering (see Appendix B) was installed at Norfolk. This unit is equipped
with a vacuum system which enables the still to recover solvents with boiling
points of up to 500°F. Norfolk reported the cost of the unit to be $9,000
($5,000 without the vacuum system). They reported recovering more than
50 percent of their waste solvents at a cost of about $0.05 per gallon.
7-31
-------
7.1.3.5 Cost Estimates For Moderate To Large-Scale Recovery Units—
In their 1983 report entitled, "Solvent Recovery in the United States,
1980-1990" , the Pace Company generated comprehensive cost data for a
variety of solvent recovery technologies (base case flow rate of 200 gallons
per hour). Costs were based on the threshold cost of recycled solvent; i.e.,
the costs required to provide various rates of return for installation and
operation of a newly constructed recovery system. Costs for solvent recovery
facilities located within an existing industrial plant were based on the
following anticipated capital investment expenditures: process equipment
(vessel, condenser, etc.), tankage, engineering, electrical, instrumentation
and contingency. The majority of distillation systems assessed in the study
are sold by vendors as modular packages including the vessels, an overhead
condenser, decanter or receiver, and various pumps and motors. Tankage
requirements will increase with the number of waste types to be processed and
the processing volume. Pace assumed that cone bottom tanks would be purchased
since they allow for greater ease of cleaning settled solids.
Engineering, electrical, and instrumentaion costs associated with
designing and constructing a facility were assumed to be 20 percent of the
process equipment costs. A contingency cost of 15 percent was included to
cover any unanticipated costs. Capital costs are summarized in Table 7.1.13.
Operating costs include utility, maintenance (negligible), disposal and
labor costs. Pace assumed that utilities such as steam could be easily
obtained from existing units. Energy consumption costs for various solvents
average 4 cents per gallon of recovered solvent at an electricity cost of
4 cents/kWh. It should be noted, however, that fractional distillation can be
considerably more energy intensive with high reflux.
Disposal costs were proportional to throughput and increased with solids
concentration in the feed. Pace assumed the bottoms product would be
45.5 percent solids and cost 54 cents per gallon for disposal. For most
solvent wastes not already being recovered, such a high recovery may not be
realistic. Depending upon expected recoveries relative to what is assumed by
Pace, threshold costs will vary. Adjusted values can be derived by comparing
the Pace estimated recovery to specific expected recoveries in addition to
adjusting disposal costs. The Pace estimated disposal costs may be inaccurate
7-32
-------
TABLE 7.1.13. CAPITAL COSTS FOR ONSITE SOLVENT RECOVERY SYSTEMS (1982 Dollars)
Coil still Scraped surface still
Nominal Feedrate 50 100 200 7.5 25 50 100 200 300
Process Equiment 37,500 45,000 50,900 19,400 24,400 30,400 37,900 47,700 62,600
Tanks 18.500 18.500 37.000 18.500 18.500 18.500 18.500 37.000 37.000
Sub-Total (1) 56,000 63,500 87,900 37,900 42,900 48,900 56,400 84,700 99,600
Engineering, Elec-
trical, Instrumen- 11,200 125700 17,600 7,600 8,600 9S800 11S300 16,900 19S900
tation (20% of (1)) _____
Sub-Total (2) 67,200 76,200 105,500 45,500 51,500 58,700 67,700 101,600 119,500
i
w •
w
Contingency
15% of (2) 10.100 11.400 15.800 6.800 7.700 8.800 10.200 15.300 17.900
Total 77,300 87,600 121,300 52,300 59,200 67,500 77,900 116,900 137,400
Source: The Pace Company. Reference 5.
-------
depending upon the waste type or disposal method. Threshold costs may be
lower for solvent waste used as a fuel or considerably higher for halogenated
solvent residuals which must be incinerated.
Finally, labor requirements were assumed to be two employees per hour
onstream at a cost of $14.42 per hour, including overhead. These labor
requirements are high, especially for units which are often automated and
operated with existing labor levels. The Pace base case was assumed to apply
for the 200 gph flow rate with labor being proportional to the flow rate for
the smaller units. Combined with the above assumptions for capital and
operating costs and financial parameters, the threshold costs for onsite
recovery are summarized in Table 7.1.14 for varying solids contents, 50 and
200 gph feed rates, and 15 and 30 percent net discounted cash flow (DCF).
TABLE 7.1.14. THRESHOLD COSTS3 ($ PER GALLON) FOR ONSITE RECOVERY
SYSTEMS FOR VARYING SOLIDS CONTENTS AND PERCENT NET
DISCOUNTED CASH FLOWS (DCF)
Technology
Coil Still
Scraped Surface Still
Nominal
(gph)
50
200
50
200
50
200
50
200
DCF
15
30
15
30
5
0.62
0.40
0.87
0.49
0.66
0.39
0.80
0.49
Percent
10
0.80
0.53
1.08
0.64
0.76
0.53
1.00
0.63
Solids
15
1.06
0.71
1.39
0.84
1.01
0.71
1.30
0.83
20
1.39
0.97
1.80
1.13
1.33
0.96
1.69
1.11
aCosts in 1982 dollars.
Source; Reference 5.
For most onsite facilities, threshold cost will be relatively insensitive
to variations in equipment capital costs due to the effects of labor on
overall recovered solvent cost. Pace concluded that a threshold change of
only +5 percent may result from capital cost fluctuations of 20 percent.
7-34
-------
However, cost will be sensitive to changes in on-stream time (capacity
factor). If the solvent recovery equipment were operated less than 40 hours
per week, threshold costs could increase dramatically due to increased fixed
cost per gallon recovered.
7.1.3.6 Cost Estimates for Small Packaged Stills —
The Naval Energy and Environmental Support Activity (NEESA) has assessed
the costs of small to moderate sized batch stills and a larger continuous
still, with results reported in Reference 15. The major cost parameters
considered in evaluating the potential economics of a solvent reclamation
program were as follows:
D - Cost of distillation unit
E - Recovery efficiency of still (decimal fraction)
I - Cost of installation ($)
M - Cost of additional labor (varies with size and operational
requirements of still; assume $0 for under 3,500 gallons per year,
$2,500 between 3,500 and 13,000 gallons per year, and $7,500 for over
13,000 gallons per year)
S - Cost of solvent ($/gallon)
U - Cost of utilities in reclaiming solvent (^/gallon)
V - Volume of waste solvent generated by activity or shop (gallons)
W - Waste disposal costs for solvent and still bottoms (^/gallon)
(A/P:i:n) - Appropriate capital recovery factor to evaluate payback
period.
These parameters may be combined to form an equation representing the
parameter interactions at specific payback periods as follows:
V[ES+(1-E)W] - (A/P:i:n) (D+I) - UV - M » 0
7-35
-------
Using this equation it is possible to calculate break-even volumes of solvents
for various applications and payback periods. Minimum estimated quantities of
solvent required to break even in 3 years (relative to land disposal), are as
follows:
Mineral Spirits: 1,500 gallons per year
- Degreasers: 500 gallons per year
Waste Paint Thinner (Epoxy): 700 gallons per year
The following sections examine the economics of in-house solvent reclamation
for the three major types of solvent.
Cold Cleaning Solvents—
The mineral spirits type solvents used for cold cleaning of metal parts
tend to have relatively low cost, low volatility, and high boiling points.
These factors make cost-effective recovery high volume dependent because of
the more expensive reclamation equipment required and the low value of the
solvent. Payback periods were calculated using the cost equation and the
assumptions listed below.
Assumptions:
Cost of Still (D): $ 8,600 for under 3,500 gallons per year
$17,700 for under 3,500-13,000 gallons per year
$45,000 for 13,000-60,000 gallons per year
Installation Cost (I): 1.5 D
Recovery Efficiency (E): 0.95
Additional Manpower (M): See original assumptions
Solvent Cost (S): $1.75 per gallon
Utility Costs (U): $0.05 per gallon
Waste Disposal Costs (W): $1.00 per gallon, $2.00 per gallon, $3.00 per
gallon
Discount Factor (i): 10%
7-36
-------
Due to the high sensitivity to waste disposal costs for the small (15 gpd) and
medium (55 gpd) size distillation units, economic analyses were conducted
using three disposal costs. For the large distillation unit (250 gpd) where
the waste disposal cost is less influential, the economic analysis was
conducted at the median disposal cost of $2.00 per gallon. NEESA originally
assumed a $0.25 per gallon utility cost. Case studies conducted by GCA and
others report utility costs to be on the order of $0.05 per gallon. Working
with the basis provided by NEESA, payback period curves were generated for the
reduced utility costs in Figures 7.1.3 to 7.1.5.
Vapor Degreasing Solvents
Since solvents used for vapor degreasing are volatile and expensive, they
are ideally,suited for cost—effective reclamation. These low boiling solvents
require the least expensive reclamation stills. Cost assumptions in
developing Figure 7.1.6 are:
Cost of Still (D): $ 4,600 for under 3,500 gallons per year
$12,000 for 3,500 to 13,000 gallons per year
$30,500 for 13,000 to 60,000 gallons per year
Solvent Cost (S): $4.50 per gallon
Waste Disposal Cost (w): $2.00 per gallon
Other costs are the same as for cold cleaning solvents.
Paint Thinners—
The composition of waste paints and waste paint thinners varies markedly
depending upon the manufacturer and application. Since the primary purpose of
reclaiming paint thinners is for reuse in cleaning paint spray equipment, the
exact composition is not important so long as the reclaimed solvent is
compatible with the paint to be cleaned from the equipment. Since most paint
thinners are mixtures, vacuum type distillation units are required since one
or more components of the thinner usually have high boiling points. The
assumptions used in developing the payback curves are those given for
7-37
-------
5,000 r-
4,000
3,000
o.
3
o
I
J 2,000
1,000
W-Jl.OO/gallon
W-S2.00/gaIlon
W^S^OO/gallon
I
4 6
Payback Period (yean)
10
Figure 7.1.3. Reclamation of cold cleaning solvents via
small batch stills (15 gpd).
Source: Reference 15.
7-38
-------
8,000
o
e
o
a
at
c
o
"5
o>
%*•
Ul
6,000
4,000
2,000
\
\
\
4 6
PAYBACK PERIOD (years)
8
10
Figure 7.1.4. Reclamation of cold cleaning solvents via
medium batch stills (55 gpd).
Source; Reference 15.
7-39
-------
CCS
*
0
Q.
25,000 r
20,000
15,000
~ 10,000
Iti
2
5,000
= $2. OO/ gallon
8
10
PAYBACK PERIOD (years)
Figure 7.1.5. Reclamation of cold cleaning solvents
via a large continuous still (250 gpd),
Source: Reference 15,
7-40
-------
15.000
10,000
CO
e
a
CO
c
5,000
es
tu
C.
STILL (250 gpd)
JATCH STILL (55 gpd)
SMALLBATCH STILL (15
6
PAYBACK PERIOD (years)
8
10
Figure 7.1.6. Reclamation of vapor degreasing solvents
(chlorinated hydrocarbons, $4.50/gallon).
Source: Reference 15,
7-41
-------
reclaiming cold cleaning solvents except that solvent costs range from $2.00
to $4.50 per gallon. Payback curves are given in Figures 7.1.7 and 7.1.8 for
two solvent costs. Utility costs were not adjusted for these curves.
7.1.4 Overall Status
7.1.4.1 Availability/Application—
Distillation and fractionation are two of the most established unit
operations. Many firms provide design services and manufacture equipment.
The Chemical Engineers Equipment Buyers' Guide provides a comprehensive list
of such firms. In December 1985, the Naval Energy and Environmental Support
Activity (NEESA), presented the results of a package distillation equipment
manufacturer's survey. The document provides detailed price, option
(including safety features), and specification information for products from
17 different manufacturers of small batch and continuous units (< 60 gallons
per hour). Table 7.1.15 lists some of the general specifications of the
stills that were considered. Appendix B provides more detailed information
including capital purchase cost information.
7.1.4.2 Environmental Impact—
The environmental impact of distillation processes for solvent recovery
should be minimal. Apart from questions related to the disposal of still
bottoms, other emissions, including air emissions from the condenser, do not
appear to be significant.
7.1.4.3 Advantages and Limitations—
Distillation would appear to be a key technology for the recovery of
solvent wastes. Based on well understood principles, its implementation is
relatively straightforward and its economic benefits can be appreciable.
7-42
-------
20,000
a
0)
>*
bt
Q)
a
o>
%**
Ul
2
13
_j
o
15,000
10,000
5.00O
cj SHALL BATCH STILL (15 gpd)
4 6
PAYBACK PERIOD (years)
8
10
Figure 7.1.7. Reclamation of waste paint thinner ($2.00/gallon),
Source: Reference 15.
7-43
-------
15.000
10,000
cs
-------
TABLE 7.1.15. COMMERCIALLY AVAILABLE SOLVENT STILLSa
Manufacturer
Alt. Resource Mgrat
Baron-Blakeslee
Branson
Brighton
Cardinal
DCI
Disti
Finish Engineering
Hoyt
Lenape
Phillips
Progressive Recovery
Raraco
Recyclene
Unique Industries
Venus
Hestinghouse
Solvent types
All
Halogenated
Halogenated
All
Halogenated
All
All
All
Halogenated
Halogenated
All
Ha iogenated
All
.Halogenated
.All
Halogenated
Maximum Explosion Vacuum
b.p.(F) proof Available
500 X X
350
350
500 X X
350 X
500 X X
500 X X
500 X X
All 350 X
350
350
500 X X
350
420 X
350
210 X
350
Water
Separator Heating
X Electric/Steam
X Electric/Steam
X Electric/Steam
X Hot Oil/Steam
Electric
X Dir. Stream Inj.
X Hot Oil/Steam-.
Electric/Steam
Hot Oil
X Electric
X Slectric/Steaw/Gas
Hot Oil/Steam
Electric/Steam
Hot Oil
X Electric/Steaa/Gas
Electric
X Electric
Cooling
Water
Water/Refrig
Water/Refrig
Water
Refrig
Water
Water
Water
Water
Water/Refrig
Water
Water
Water/Air
Water
Water/Refrig
Water/Refrig
Water/Refrig
Source: Reference 4.
aCapacity less than 60 gallons per hour.
-------
REFERENCES
1. Allen, C., et al. Field Evaluations of Hazardous Waste Pretreatment as
an Air Pollution Control Technique. Report prepared for U.S. EPA, ORB,
Cincinnati, Ohio under Contract No. 68-02-3992. April 1985.
2. Berry, R.H., et al. Chemical Engineers' Handbook, Sixth Edition.
McGraw-Hill, 1984.
3. National Fire Protection Association, Fire Protection Guide for Hazardous
Materials, 9th edition, 1986.
4. Naval Energy and Enviromnetal Support Activity, Assessment of Solvent
Distillation Equipment, NEESA 20.3 - 012, December 1985.
5. Horsak, R.D., et al., Pace Company Consultants and Engineers, Inc.
Solvent Recovery in the United States: 1980 - 1990. Houston^ TX.
Prepared for Harding Lawson Associates, January 1983.
6. Dean, John A., Lange's Handbook of Chemistry, McGraw-Hill, 1979.
7. Neast, R.C., CRC Handbook of Chemistry and Physics. CRC Press, 1978.
8. Flick, Earnest, ed. Industrial Solvents Handbook, Third Edition. Noyes
Publications, Park Ridge, NJ, 1985.
9. Gavlin, A. Benagali and W. Lagdon. Case Studies of Solvent Recovery Via
Continuous Processing. AIChE Symposium Series, 76(192): 46-50, 1980.
Paper presented at September Symposium of AIChE Annual Meeting, New York,
NY, November 1977.
10. Peters, M.S. and K..D. Timmerhaus, Plant Design and Economics for Chemical
Engineers, McGraw-Hill, 1980.
11. Shulka, H.M., and R.E. Hicks, Water General Corp. Process Design Manual
for Stripping of Organics. EPA-600/2-84-139, U.S. EPA/IERL, Cincinnati,
Ohio, August 1984.
12. Yeshe, P. Low-Volume, Wet-Scrap Processing. Chem. Eng.
Progress, 80(9): 33-36, September 1984.
7-46
-------
13. GCA, Waste Minimization Case Studies for EPA OSW, Contract 68-03-3243,
1986.
14. Higgens, I.E. CH2M Hill, Reston, Va. Industrial Processes to Reduce
Generation of Hazardous Haute at DOD Facilities Phase 2 Report Evaluation
of 18 Case Studies, July 15, 1985.
15. Nelson, W.L., Naval Energy and Environmental Support Activity (NEESA),
In-House Solvent Reclamation, Port Hueneme, CA. NEESA 20.3-012,
October 1984.
7-47
-------
7.2 EVAPORATION PROCESSES
Available evaporation equipment designs include simple stills, flash
evaporators, forced circulation evaporators, and falling film and agitated
thin film evaporators. All designs are capable of concentrating nonvolatile
components of waste mixtures. However, in cases where fouling, foaming, high
viscosity, thermal degradation or other factors present potential operational
problems, agitated thin film evaporators provide the most versatile
service. They also represent the most effective, high volume evaporation
equipment which is capable of reducing viscous wastes to low residual
2
organics. This is a direct result of high mass transfer rates achieved
through turbulence. For these reasons, a majority of large commercial solvent
recycling companies use agitated thin film evaporators (ATFEs) as indicated by
several industry surveys.
The emphasis in this section will be on the agitated thin film designs
because of their widespread use and applicability in reclaiming solvent wastes
which are too viscous or otherwise too difficult to recover in conventional
distillation equipment. Extensive reference to conventional equipment was
provided in Section 7.1.
7*2.1 Process Description
Liquid waste is fed to the top of ATFEs where longitudinal blades mounted
on a motor driven rotor centrifugally force the waste against the heat
transfer surface; i.e., the inside wall of .the cylindrical vessel. This
surface is enclosed in a heating jacket which usually employs steam or hot oil
as the heating medium (temperatures up to 650°F).
The agitation and liquid film are maintained by the blades as they move
along the heat transfer surface. The blade tips typically travel at 30 to
40 feet per second at clearances of 0.007 to 0.10 inches which creates high
turbulence (see Figure 7.2.1). This facilitates efficient heat and mass
transfer, shortens required waste residence time and creates a degree of
mixing which maintains solids or heavy molecular weight solutes in a
manageable suspension without fouling the heat transfer surface. Mass
diffusivities in ATFEs can be increased by 1,000 to 10,000 times over
7-48
-------
Turbulent Liquid Film
Heated Wall
Figure 7.2.1. Cross section of agitated thin film evaporator.
Source: The LUWA Corporation
Bulletin EV-24, 1982.
Reference 3.
7-49
-------
2
nonagitated designs. To further promote separation, the unit is usually
operated under vacuum conditions which permits lower temperature processing of
thermally unstable mixtures.
Ultimate recovery in all evaporation/distillation units is limited by
vapor-liquid equilibrium constraints as discussed in detail below. However,
for viscous wastes, economical recovery is further limited by waste viscosity
as a result of decreasing diffusivity of volatile compounds through the waste
as viscosity increases. As diffusivity decreases, resistance to overall mass
transfers into the gas phase increases. However, this effect is less
pronounced in ATFEs relative to other evaporator designs due to high
turbulence and large exposed waste surface area.
Finally, thermodynamic properties of the waste and ATFE operating
pressure set a limit on ultimate recovery imposed by vapor-liquid equilibrium
constraints. Material will boil when its vapor pressure reaches the operating
pressure of the ATFE. Waste vapor pressure for miscible fluids is equal to
the sum of the partial pressure of each volatile species. Partial pressure of
each component is, in turn, equal to its molar concentration multiplied by the
pure component vapor pressure and a constant which is dependent on the
ideality of the solution. Thus, operating pressure and partial pressure
determine the minimum attainable (i.e., equilibrium) concentration of each
volatile species in solution. High separation efficiency will be associated
with low system pressure, high pure component vapor pressure, high activity
coefficient, high Henry's Law constant, and low solubility (decreases with
temperature). Theoretically, very high separations can be achieved for highly
volatile compounds in systems of low liquid phase miscibility. Ultimate
recovery will depend on the extent to which equilibrium is achieved which will
be limited by diffusive resistance to mass transfer and residence time in the
system.
Residual solvent concentrations below 1,000 ppm have been routinely
achieved and a concentration below 100 ppm can be achieved if conditions are
optimal. Except in unusual circumstances (e.g., immiscible fluids), the sole
use of ATFE cannot be expected to yield residual solvent concentrations in the
low ppni range.
7-50
-------
7.2.1.1 Pretreatment and Post-Treatment Requirements—
A schematic of an ATFE and associated pretreatment and post-treatment
options is shown in Figure 7.2.2. As shown, the pretreatment techniques most
commonly applied to wastes undergoing ATFE are a previous solvent recovery
process, oil or suspended solids removal or a dissolved solids concentration
process. The most cost-effective application of a ATFE is in treating viscous
wastes which are generally not amenable to treatment using other, less
expensive evaporative processes. In many cases, the source of these wastes
will be bottoms products from conventional evaporation/distillation
processes. These processes have been described previously (Section 7.1).
Constraints on the acceptable viscosities of wastes differ between
manufacturers and specific unit types. In general, specially designed ATFEs
can process wastes with viscosities up to 1,000,000 cps as indicated in
Table 7.2.1 and Figure 7.2.3 which presents equipment selection guides for
1 4
evaporator products based on waste viscosity. *
Post-treatment methods are also identified in Figure 7.2.2. These
basically include further refinement of the overhead product, through water
removal or separation of solvent mixtures, and further solvent recovery or
disposal of bottom products.
The recovered solvent may be used as is or further purified through
decanting, dehydration or fractionation. In cases where the waste feed is a
mixture of solvents, separation by fractional distillation is sometimes
justified by the increased value of the separated components.
Further recovery of bottoms from ATFE treatment of organic wastes is
generally not considered practically achievable in liquid handling equipment.
However, in some cases a drum dryer, centrifuge or other solids handling
equipment might be employed depending on the nature of the waste. The residue
is often solidified and landfilled, incinerated, or burned as fuel if the Btu
value, chlorine content, ash content and viscosity are within required
specifications.
Further treatment of aqueous ATFE bottoms will generally be required to
remove remaining volatiles and other contaminants. Candidate technologies
include steam and air stripping, carbon adsorption, or biological treatment if
toxic contaminant concentrations are low.
7-51
-------
FEED
PRETREATMENT
OPTIONS
AGITATED
THIN FILM EVAPORATION
POST-TREATMENT
OPTIONS
SPENT SOLVENT
WASTE FEED
-4
I
NS
CONVENTIONAL
DISTILLATION OR
EVAPORATION
DECANTING OR OTHER
LIQUID/SOLID
SEPARATIONS (e.g.
SKIMMING, CENTRIFUGING)
AGITATED
THIN FILM
EVAPORATOR
PRODUCT
SOLVENT
- DECANTING
- DEWATERING
- FRACTIONATION
- REUSE AS IS
BOTTOMS
FUEL BLENDING
INCINERATION
SOLIDIFICATION
& DISPOSAL
DRYING
PHYSICAL SEPARATION
OF LIQUIDS AND SOLIDS
(e.g., CENTRIFUGING,
SETTLING)
Figure 7.2,2. Treatment train using an agitated thin film evaporator.
-------
-xl
m
w
Liquid
Property
Horizontal Vertical
Agitated Agitated
Thin Film Thin Film
Evaporator Evaporator
Low viscosity ++ ++
10,000 cps or less
Medium viscosity ++ +
up to 50,000 cps
High viscosity X X
up to 1,000,000 cps
Slurry + +•
High
Viscosity
ATFE Conventional
(Film Truder) Evaporator
_
X
•n- X
X
High vacuum evaporation ++ ++ ++ ++•
High concentration ++ - - -
Residence
time control ++ +
-
Note: ++ - Particularly suitable
+ = Suitable
- - Usable in special cases
X = Unusable
Source: Reference 4.
-------
Evaporator Type
Natural Circulation. 1
Falling Film
Circulating
Agitated Thin Film
Agitated Thin Film
(reinforced)
High Viscosity ATF
(film truder)
Applicable Viscosity Range (cps)
10-3
10-2
T ySSS^JrSSfjFSf-
Y/s
10-1
/yyyy/
1
1
\/ //////
K//
10
yyyyyy
\ffS
10*
i
y/yyj
103
/yyyy
\//s
10*
/ s s/^,
/ S / / /
105
yyyy y
106
*"yy yy
Source: Reference 1.
Figure 7.2.3. Selection of Luwa Evaporators based on waste viscosity.
-------
7.2.1.2 Operating Parameters and Design Criteria—
ATFEs are suited to treatment of concentrated, nonvolatile organic wastes
contaminated with water, and aqueous wastes with volatile organic
concentrations above 5 percent which are not amenable to treatment using less
expensive conventional evaporation/distillation technologies. Typically,
waste viscosity (feed or bottoms) is the restrictive waste characteristic
which results in adoption of ATFE as the preferred technology. Consequently,
pretreatment requirements are less stringent and may be reduced to gross solid
removal or waste concentration through decanting.
Design and operation of an evaporator depends on the waste
characteristics and desired recovery efficiency. Ultimate recovery may be
restricted due to limiting waste characteristics or deliberately restricted
based on the competing economics of processing, residual post-treatment and
raw solvent purchase costs. Recovery will be a function of operating
temperature, pressure, heat transfer and residence time* For recovery to low
residual organics, particularly in viscous wastes, mass transfer resistance
• 2
presents the biggest obstacle to recovery efficiency.
Operating system temperature and pressure are limited by waste type and
equipment design. Temperature must be higher (0 to 30°F) than the boiling
point of the material, which is to be recovered as the overhead product, and
sufficiently high to maintain a minimum waste viscosity. Maximum temperature
(less than 650°F) may be limited by explosion limits or by the decomposition
temperature of the recoverable materials, which is of particular concern for
some halogenated organics (Section 7.1). Operation at low pressure reduces
the temperature required to reach the boiling point. It also enables higher
recovery rates to be achieved, as discussed previously. The lower limit of
pressure is restricted by cost and equipment design. Typical system pressures
range from atmospheric to 2 mm Hg.
For a given flow and desired recovery, an ATFE system has to be designed
to produce a specific evaporation rate. Evaporation rate depends on
temperature and pressure as discussed above, heat transfer surface area, waste
type and heat transfer coefficient. Figure 7.2.4 and 7.2.5 show the
relationship between heat transfer surface area, waste type and evaporation
rate for a Cherry-Burrell ATFE (Turba-film processor). Figure 7.2,6 shows
the same relationship, based on unit area of heat transfer surface, for a LUWA
3
ATFE. As shown, high heat transfer surface area is required to evaporate
7-55
-------
A
-------
(0
oc
o
D.
UJ
oc
til
s
LLJ
H
6000
5000
4000
3000
2000
1000
DEHYDRATING HEAVY PASTES
STRIPPING TO LOW RESIDUAL ORGANICS
20
40
60
80 100
120 140
TURBA-FILM PROCESSOR HEAT TRANSFER
AREA (FT2)
Figure 7.2.5.
Required heat transfer surface area for dehydrating heavy
pastes and stripping wastes to low residual organics*.
*Source: Cherry—Burrell, Reference 5.
7-57
-------
Heating Medium: dry and saturated steam
A Concentration of aqueous solutions
B Dehydration of organics
C Distillation of organics
D Stripping of low boilers from organics
E Reboiler service
F Solvent reclaiming
G Deodorization
Heating Medium: Dowtherm or hot oil
H Distillation of high-boiling organics
I Stripping of high boilers
J Reboiler service
Figure 7.2.6.
Heat transfer and evaporation rates in
Luwa Thin Film Evaporators.
7-58
-------
solvents from aqueous wastes and highly viscous materials; e.g. waxes and
pastes. Aqueous wastes, because of their low viscosity, are often handled in
conventional evaporation equipment.
In summary, increased evaporation rate can be attained by increasing heat
transfer surface area, operating at the high temperature limit, or operating
at the low pressure limit. However, if low concentration of VOCs in the
bottoms product is desired, mass transfer limitations will shift optimal
design toward high transfer area, increased turbulence and increased residence
time.
Typical ATFE system operating data are summarized in Table 7.2.2. As
shown, heat transfer surface area, steam consumption and cooling water
requirement vary directly with solvent recovery rate. However, electricity
requirement drops per unit of solvent quantity recovered thus providing a
slight drop in unit operating costs as capacity increases. Throughput of
ATFEs is generally high. If waste solvents are generated in low quantities,
package distillation units capable of handling high solids wastes may be more
economical (see Section 7.1).
Commercially available evaporator equipment design parameters are
2
summarized as follows. Evaporators range from 1 to over 400 square feet of
heat transfer surface with liquid throughput ranging as high as
2
250 Ib/hr/ft . Overhead to bottoms splits for lightly contaminated fluids
can be as high as 100 to 1 with controlled residence times of up to
100 seconds. Blade tip speeds of nonscraping designs average 30 to 40 ft/sec
while scraping blades average 5 to 10 ft/sec. Some units come equipped with
variable clearance while scraping blades are typically spring mounted or
maintain contact with the wall as a result of centrifugal force. Operating
temperatures range up to 650°F and pressure ranges from 2 mm Hg to
atmospheric. Finally, configurations of commercially available equipment
include vertical or horizontal shells, cylindrical or tapered design and
cocurrent, countercurrent or separated vapor/liquid flow.
7.2.2 Demonstrated Performance
Actual performance data from commercially operated units, conducted in
accordance with EPA Quality Assurance/Quality Control requirements, are
limited to EPA sponsored studies as reported by GCA (1986) and the Research
7-59
-------
TABLE 7.2.2. TYPICAL AGITATED THIN FILM EVAPORATOR DESIGN CHARACTERISTICS
Solvent
Recovery
Cgal/hr)
40
85
130
240
500
Heat
Transfer
Surface
Area
4.2
8.8
13.4
25
51.2
Heating
Btu/hr
(1000)
79
168
251
474
989
Utilities3
Cooling
Water
(gpra)
5
11
16
30
63
Electricity
(KW)
1.5
1.5
3.5
3.5
4
oyscetn
Dimensions
L x W x H,
(ft)
4 x 6 x 10
4 x 6 x 11.5
4 x 6 x 13
5.x 8 x 11.5
5 x 8 x 14
aBased on an average latent heat of vaporization of 175 Btu/lb and preheating
feed by 200°F.
Sources Reference 6.
7-60
-------
2
Triangle Institute (1984). Other data generated by GCA and Metcalf & Eddy
under studies currently being performed for the U.S. EPA were not available at
the time this document was prepared.
As part of GCA and RTI's studies, a cost assessment was made for each of
the four ATFEs which were sampled. None of the units studied demonstrated a
capability of reducing volatile concentration in the bottoms product to low
ppm levels. The following discussion summarizes the performance data for
these units.
7.2.2.1 Thin Film Evaporator and Milsolvrator —
GCA performed an evaluation of an ATFE in 1986 as part of an EPA
sponsored study to obtain background information on alternative technologies
to support the RCRA land disposal ban. An oil heated, batch LIMA ATFE was
evaluated at Miisolv Corporation along with a still bottoms processing unit
called a Milsolvrator. The latter consisted of large, jacket heated, inclined
cylinder equipped with an internal auger to convey solids. The equipment
operating and design features and test results are summarized below.
The LUWA ATFE contains an internal stainless steel rotary blade which
spins at approximately 385 rpra with a 0.20 mm clearance between the blade tip
and internal evaporator wall. The heat transfer surface area is 43 square
feet and is heated by a hot oil jacket. The unit is operated under vacuum to
enhance recovery, and is equipped with an optional recycle loop to recircuiate
bottoms if required. Process data during the three test runs are summarized
in Table 7.2.3.
Miisolv operates the unit to achieve as high a reclaimed solvent yield as
possible when processing halogenated solvents. However, a small percentage of
nonhalogenated solvents are allowed to remain in the bottoms to ease handling
and improve the fuel value. All bottoms are shipped offsite to a fuel
blending facility or are shipped directly to a cement kiln for use as a
supplemental fuel at an average disposal cost of $0.26 per gallon.
Three days of sampling were performed on the ATFE during which time one
halogenated waste (F001) and two nonhalogenated wastes (F003 and F005) were
processed. Table 7.2.4 summarizes the principal organic components of the
waste streams. As shown, the halogenated waste was recovered to the greatest
extent. However, the ATFE was only able to reduce the bottoms concentration
7-61
-------
TABLE 7,2.3. PROCESS DATA FOR LUWA THIN FILM EVAPORATOR RUNS
Run Number
Date of Test
Total Gallons
Processed
Time to Process
Batch (hours)
Average Feed
Rate (gal/min)
% Yield as Distillate
Average Temperature:
Hot Oil Jacket in
to the LUWA (°F)
Average Temperature :
Hot Oil Jacket
exiting the LUWA (°F)
Average Coolant
Temperature (°F)
Average Distillate
Temperature (°F)
Average Vacuum Surge
Tank Temperature (°F)
Average Vacuum (in Hg)
1
12/4/85
340
4.25
1.3
69.8
311
304
49
48
52
5.6
2
12/3/85
2835
20.6
2.3
37.6
318
307
49
45
51
7.1
3
12/5/85
3910
17.0
3.8
45.8
327
317
51
52
52
25.5
Source: Reference 7.
7-62
-------
TABLE 7.2.4. PERCENT BY WEIGHT OF PRINCIPAL ORGANIC COMPONENTS
Run 1
Waste Feed
Distillate
Still Bottoms
Run 2
Waste Feed
Distillate
Still Bottoms
Run 3
Waste Feed
Distillate
Still Bottoms
Waste
Ojuantity
(§al) Trichloroethylene
342 98. 14a
239 99. 9a
103 74. 7
Waste
Quantity Ethyl n-propyl
(gal) Ethanol n-propanol acetate acetate
2,835 24 7.7 24 9.8
1,066 30 4.9 39 10
1,769 20 14 8.3 11
Waste
Quantity
(gal) 2-Butanone Toluene/Heptane
3,910 1.7 51
1,791 1 46
2,119 2.8 54
Solids
1.86
0.008
15.2
Solids
11.8
0.035
36.5
Solids
7.44
0.033
23.7
aBased on solids content. GC analysis and density measurement yielded
greater than 100% trichloroethylene based on an average of 3 samples.
Source: Reference 7.
7-63
-------
of trichloroethylene to 75 percent with a waste volume reduction of 70 percent.
The nonhalogenated wastes experienced a volume reduction of only 42 percent
with 55 percent solvent remaining in the waste. Operational difficulties
during the first run required solvent feed rates at approximately one-third of
design level which may have had a negative impact on recovery rate.
GCA also collected samples from the processing of solvent contaminated
waste in the Milsolvrator. However, these data were not yet available at the
time this document was prepared. The Milsolvrator represents a unique design
in high solids processing equipment although company officials still consider
it to be in the design stage. Since units like these are likely to become
more widely applied treatment methods in response to the land disposal ban, a
description of the unit is provided below.
Waste is fed to the Milsolvrator manually, whereby an operator scoops the
waste from a drum and places it in a grinder. Shredded waste is introduced
into the elevated end of a long (20 ft) horizontal cylinder. A screw auger
transports the waste through the cylinder. The clearance between the screw
auger and the outside wall is an eighth of an inch. The cylinder is heated by
a hot oil jacket which is capable of achieving an internal temperature of
550°F. The unit operates under 15 in. Hg vacuum. As the waste is heated
during its path down the cylinder, solvent vapors are released, collected t>y a
manifold system, then passed through a mist entrainer and condenser and
collected in a receiving tank.
The solids portion of the waste traverses the entire length of the
cylinder and is discharged to a receiving drum. This drum is attached
directly to the Milsolvrator with a rubber o-ring seal in order to maintain
the unit under vacuum. The residual solids discharged from the unit have
passed EP-Toxicity tests.
2
7.2.2.2 Thin Film Evaporator - Plant A —
The LUWA thin film evaporator at Plant A processes organic wastes,
including sludges, from the furniture, chemical, dry cleaning and paint
industries. The sludges include dirt and grease, paint films, particulates
and insoluble organic materials. Since the still bottoms are used as a fuel
substitute, the ATFE is operated to maintain pumpable bottoms containing less
than 1 percent chlorides and an energy content greater than 28,000 kJ/L
7-64
-------
(100,000 Btu/gal). Distillation bottoms are shipped offsite and utilized as
fuel in cement and expanded aggregate kilns. Process economics require that
60 percent or more of the waste be recoverable. On average, 70 to 95 percent
of the material is stripped off as overhead product which may be further
refined in a fractionation column depending on the product's intended end
2
use.
The LUWA Thin Film Evaporator System at Plant A consists of a 4.0 m
2
(43 ft ) heat transfer surface area LUWA Evaporator complete with an
entrainraent separator, condenser, feed pump, bottoms pump, distillation pump,
vacuum system, and instrumentation. Evaporator feed rate and system pressure
(vacuum) are determined based on the material being processed. A typical feed
rate is 23 L/min (6 gpm), but may range from 4 to 45 L/min (1 to 12 gpm).
Steam, controlled at about 30°C (54°F) above the boiling point of the
distillate, is typically used as the heating medium. The cooling water for
the overhead condenser is generally maintained in the range of 10 to 16°C
(50 to 60°F), with a flow rate of about 1,500 L/min (400 gpm).
RTI reports that the thin film evaporator at Plant A can be used in the
following applications:
1. Removal of VOCs from organic streams which may contain viscous high
molecular weight organics or solids.
2. Removal of VOCs from sludge such as insoluble organics and
particulate solids.
3. Concentration of aqueous sludges such as insoluble organics and
particulate solids.
4. Removal of VOCs from aqueous streams where the VOC volatility is
higher that that of water.
5. Removal of water from streams containing relatively high
concentrations of organics of lower volatility than water (water
removed as overhead product).
During RTI's field visit to Plant A, a waste consisting of mixed xylenes
with a small amount of solids and approximately 5 percent VOCs was being
processed through the thin film evaporator. The VOCs included several
7-65
-------
solvents of concern as listed in Table 7.2.5. Since this material contained
few solids, approximately 95 percent of the feed was being taken overhead,
leaving bottoms which were acceptable for fuel.
The data obtained from samples of the feed, bottoms, and the overhead
product (shortly after process startup) are presented in Table 7.2.5. Removal
efficiency estimates in excess of 99 percent were reported for the three
chlorinated solvents, based on headspace analysis and the overall recovery
rate of 95 percent. Toluene removal efficiency was reported as 95.4 percent.
Further details of the estimation procedure used were not available. Recovery
rates would be less than these values since appreciable, but unknown,
quantities of these solvents were reportedly lost through the vacuum pump vent.
7.2.2.3 Thin Film Evaporator - Plant B2—
Plant B reclaims contaminated solvents and other chemicals through
evaporation and distillation. About 10 percent of the incoming chemicals are
contaminated products with the remainder being classified as hazardous waste.
Approximately 85 percent of the reclaimed chemicals are returned to the
generator with the remainder marketed to other suitable end users,
Processing equipment include two Votator ATFEs, two distillation
reboilers, eight fractionation columns, one caustic drying tower, support
facilities and waste transport and storage equipment. Plant B processes
wastes from the chemical, paint, ink, recording tape, adhesive film,
automotive, airlines, shipping, electronic, iron and steel, fiberglass and
pharmaceutical industries. Volatile organics recovered include alcohols,
ketones, esters, glycols, ethers, freons, and.specialty solvents. Plant B is
able to recover VOCs from still bottoms, coating residues, obsolete paints and
inks using thin film evaporators.
At the time of RTI's field visit, a batch consisting of isopropyl
alcohoKIPA), xylene, and other VOCs was being processed through the larger
ATF1. The reclaimable overhead product was being further purified through
fractionation while the bottoms product was maintained at a concentration
suitable for offsite fuel use. RTI reports a typical ratio of bottoms product
to feed of 1:5. Normally, two passes through the ATFE are used to process
this material, with the more volatile IPA removed on the first pass and the
less volatile xylene removed on the second pass. However, during this
7-66
-------
TABLE 7.2.5. PLANT A: THIN FILM EVAPOBATOR WASTE COMPOSITIONS AND HEADS?ACE ANALYSIS
Methylene chloride
Chloroform
1,1,1-trichloroethane
Toluene
Mixture of high boiling
hydrocarbons
freon TF
Feed
Liquid
composition
(vol. %)
2.0
1.5
0.7
1.3
94.4
ND
Product
Headspace
analysis"
(mg/L)
1.7
5.1
0.11
0.01
ND
0.06
Liquid
composition
(vol. %)
0.9
ND
ND
1.6
93.9
1.8
Headspace
analysis
(mg/L)
0.97
0.14
0.14
0.04
ND
1.5
Bottoms3
Headspace
analysis**
(ug/L)
0.03
0.01
0.01
0.03
ND
0.24
ND = Not Detected.
aBottoms solid upon cooling and no solids analysis was performed.
bat 25°C.
Source: Reference 2.
-------
sampling occasion, both IPA and xylene were being recovered on the same pass.
Processing conditions were 70°C (158°F) operating temperature and 22 inches
of mercury vacuum pressure. Flow rates were not reported. Samples were
collected from the feed stream, bottoms product, overhead product, and vacuum
pump discharge (gas). A summary of the liquid sample analysis is presented in
Tables 7.2.6 and 7.2.7, representing two sampling occasions. The data
indicate that the overhead product is more concentrated in volatile species,
as expected. However, since no data were presented on the relative volumes of
feed, overhead and bottoms streams, removal efficiency cannot be calculated.
2
7.2.2.4 Thin Film Evaporator - Plant C —
Plant C uses thin film evaporation for the reclamation of organic
solvents for recycle or resale as well as for producing specialty solvent
blends. The solvent recovery processes include a LUWA ATFE, an SRS Riston
Batch Distillation unit and support facilities. The wastes processed by
Plant C are from the chemical, plastics, paint, adhesive film, electronics,
and photographic industries and include waste chlorinated solvents, freons and
ketones. Plant C has no capability for operating the LUWA evaporator under
reduced pressure, which precludes processing of high boiling compounds such as
naphtha and xylene as overhead products.
The standard recovery procedure is to process each batch of chemicals
through the LUWA ATFE during which 70 to 95 percent of the material is
stripped off as overhead product.
2 2
The ATFE system consists of a 1.0 m (10.8 ft ) heat transfer surface
area evaporator, entrainment separator, condenser, product tank, feed
recirculating tank, and pumps. Steam is used as the heating medium with the
pressure being set between 207 kPa (30 psig) and 552 kPa (80 psig) depending
upon the solvent being processed. The evaporator bottoms are recirculated
through the unit until a predetermined VOC removal is reached. RTI reports
that the thin film evaporator at Plant C is used in similar applications as
reported by Plant A.
During the sampling period, a batch of acetone contaminated with several '
solvents of concern (see Table 7.2.8) was being processed through the LUWA
evaporator. The reclaimed product was being stripped off as overhead and
pumped into a product receiver. The bottoms from the evaporator was being
7-68
-------
TABLE 7.2.6. ANALYSIS OF LIQUID SAMPLES, THIN FILM EVAPORATOR, PLANT B
Is op ropy 1
alcohol
Freon TF
Toluene
Ethyl
benzene
Xylenes
Feed No.
Liquid
composition
(vol. %)
38.2
0.6
0.34
11.4
49.2
1
Headspace
analysis'*
(mg/L)
0.75
38.
0.58
5.5
22.
Feed No.
Liquid
composition
(vol. %)
42.9
0.5
0.3
10.4
45.7
2
Headspace
analysis"
(mg/L)
0.69
29.
0.48
10.4
23.
Bottoms3
Headspace
analysis
(mg/L)
1.6
5.3
0.32
9.0
39.0
^Bottoms solid upon cooling and no solids analysis was performed.
bat 25°C.
Source: Reference 2.
7-69
-------
TABLE 7.2.7. ANALYSIS OF PRODUCT SAMPLES, THIN FILM EVAPORATOR, PLANT B
Product Sample No. 1
Isopropyl alcohol
Freon TF
Toluene
Ethyl benzene
Xylenes
aat 25 °C.
Source: Reference 2.
Liquid
composition
(vol. %)
53.8
0.7
0.4
8.4
34.0
Headspace
analysis3
(mg/L)
1.1
62.
0.94
5.3
19.
Product Sample No. 2
Liquid
composition
(vol. %)
60.3
0.6
0.4
7.0
27.4
Headspace
analysisa
(mg/L)
1.1
51
0.71
4.8
17.
7-70
-------
TABLE 7.2.8. ANALYSIS OF LIQUID SAMPLES, fHIN FILM EVAPORATOR, PLANT C
Acetone
Freon TF
1,1, 1-Trichloroethane
Trlchloroethylene
Toluene
Ethyl Benzene
Xylene
Tetrachloroethylene
Fi
Liquid
composition1
(vol. %)
74.3
0.1
1.5
0.2
0.5
ND
5.9
0.6
ted
Headspace
1 analysis
(mg/L)
378.0
2.0
17.9
0.1
0.3
0.1
2.1
2.4
Produc
Liquid
composition*
(vol. %)
82.2
<0.1
2.2
0.3
0.9
0.3
2.0
0.5
t Bottoms
Headspace Liquid Headspace
analysis composition3 analysis
(mg/L) (vol. %) (mg/L)
383.0 60.6 308.0
2.0 0.1 1.5
19.1 0.9 9.2
0.1 <0.1 0.1
0.2 0.3 0.4
<0.1 <0.1 <0.1
0.2 13.6 5.0
1.6 0.9 4.1
ND - not detected.
Approximately 17 percent of the waste was high boiling organics and resins.
Source: Reference 2.
-------
pimped back to the feed tank for recirculation through the evaporator.
Grab samples of the feed, overhead and bottoms were collected and analyzed
with results as summarized in Table 7.2.8.
The following process parameters were provided by Plant C during the
field test:
1. Feed rate (recirculated)
2. Feed temperature
3. System pressure
4. Vapor temperature
5. Steam pressure
6. Jacket (upper) temperature
7. Jacket (lower) temperature
8, Condenser water inlet temperature
9. Condenser water outlet temperature
10. Distillate rate
11. Bottoms rate (feed-distillate)
12. LUWA drive motor, amps
1,635 L/hr (432 gal/hr)
38°C (100°F)
760 torr
57°C (135°F)
310 kPa (30 psig)
132°C (270°F)
107°C (225°F)
20°C (68°F)
25°C (77°F)
344 L/hr (91 gal/hr)
1,292 L/hr (341 gal/hr)
1.1
AB shown in Table 7.2.8, acetone and other low boiling point compounds
were concentrated in the distillate and xylene and other high boilers were
enriched in the bottoms. VOCs in the bottoms at the end of the run were not
removed to a higher degree because of the requirements to maintain the resins
in solution. However, the overall volume of waste was reduced by about
70 percent.
7.2.3 Cost of Treatment
7 2
Cost estimates were obtained by GCA and RTI during their case
studies to assess ATFEs discussed above. These estimates and those developed
Q
by the Pace Company are discussed below.
7.2.3.1 Milsolv Corporation Case Study —
Operating and capital cost data provided by Milsolv enabled GCA to
estimate costs of solvent recovery for the three test runs. Table 7.2.9
itemizes operating, maintenance and capital depreciation costs incurred per
hour of operation. Assuming an annual production of 940,000 gallons of waste
solvent and a capital recovery factor of 17.5 percent, total cost per gallon
7-72
-------
TABLE 7.2.9. HOURLY COSTS OF LUWA THIN FILM EVAPORATOR
Cost
Expenses (it/hour)
Fuel 8.00
Auxiliary Chemicals 0.72
Electricity 2.60
Laboratory 9.37
Operating Labor 15.76
Maintenance Labor 14.58
Spare Parts 12.69
(Repair and Maintenance)
Regulatory Compliance 6.30
Insurance 0.69
Capital Depreciation 7.27
Total 77.98
In addition to hourly cost, Milsol pays $0.26/gallon to dispose of still
bottoms to an incinerator.
Source: Reference 7.
7-73
-------
of recovered solvent was calculated as summarized in Table 7.2.10. As will be
noted in a discussion below, the costs are in reasonably good agreement with
those derived by Reference 8.
7.2.3.2 RTI Case Studies2—
RTI investigators, during their study of ATFIs at three plants, also
obtained cost data of varying degrees of completeness. Capital and operating
data provided to RTI by Plant A personnel (see Section 7*2.2.2) are summarized
in Table 7.2.11. This data is also in good agreement with the results of the
Reference 8 study, as discussed below.
Q
7.2.3.3 Cost Estimates Prepared by Pace —
In 1983, the Pace Company generated comprehensive cost data for a variety
of solvent recovery technologies, including thin film evaporation.
Pace defined a threshold cost of recycled solvent as the cost required to
provide various rates of return for a newly constructed recovery system based
on a nominal feed rate. Capital cost was divided into three cost components;
process equipment, siting and permitting, and engineering and contingency.
The methodology adopted here has been described previously in Section 7.1 and
uses identical assumptions.
Bottom products were assumed to be 45.5 percent non-solvent (e.g. solids,
oil, grease) and cost 54 cents per gallon for disposal. Optimal recovery
efficiency and disposal costs will depend on the nature of the waste and
ultimate disposal process. These figures represent a compromise between
incineration, use as a fuel, and additional solvent recovery costs. If actual
disposal costs deviate significantly from this figure (e.g., use as a fuel is
typically 20 to 30^/gal and incineration is $2 to $3/gal) overall treatment
cost estimates should be adjusted accordingly.
Labor requirements were assumed to be two employees per hour on-stream at
a cost of $14.42 per hour, including overhead. All other operating costs were
assumed to be the same as previously presented for distillation. Capital cost
estimates are summarized in Table 7*2.12 for ATFEs with three different
nominal feed rates. The resulting threshold costs for onsite ATFE recovery
are summarized in Table 7.2.13 for varying solids contents, nominal feed rate
and percent after-tax discounted cash flow (DCF). Graphical representations
7-74
-------
TABLE 7.2.10. PROCESSING COSTS DURING THREE TEST RUNS OF THE LUWA THIN FILM EVAPORATOR
Run #
date
1
12/4/85
2
12/3/85
3
12/5/85
Ul
Gallons
of waste Time of
product processing
processed (hours)
342 4.25
2,835 20.6
3,910 17.0
Gallons
of
recovered
solvent
239
1,066
1,791
Gallons
of Operating3
still costs
bottoms (fc77.98/hr)e
103 331.42
1,769 1,606.39
2,119 1,325.66
Disposal
cost of
bottoms"
26.78
459.94
550.94
Cost per**
gallon ($)
Totalc recovered
cost ($) solvent
389 1.63
2,216 2.08
2,000 1.12
aOperating Costs = $77.98 x Time of Processing.
^Disposal Costs of Bottoms = $0.26 x Gallons of Still Bottoms.
cTotal Cost = Operating Costs + Disposal Cost and Capital Depreciation Cost.
^Cost per Gallon Recovered Solvent = Total Cost - Gallons of Recovered Solvent.
eSee Table 7.2.7 for $77.98/hr figure.
Source: Reference 7.
-------
TABLE 7.2.11. 1984 PLANT A CAPITAL AND OPERATING COST
REPLACEMENT CAPITAL COSTS3
LUWA Evaporator Model LN-0500b
Main Auxiliaries (Condenser, Pumps, Vacuum
System, Controls)
Piping, Fittings (materials only)
Structural Frame
Installation (Foundation, Erection, Piping,
Wiring, Insulation)
Total Installed Cost
$120,000
36,000
12,000
6,000
120,000
$294,000
ANNUAL OPERATING COSTS
Feed Rate
Overhead Product, Percent of Feed
Operating Labor
Maintenance Labor
Maintenance Materials
Laboratory (1.2 Analysts)
Fuel (Steam System, 19 L/hr [5 gal/hrj)
Electrical (45 hp)
Cooling Water 1,500 L/min (400 gpm)
Overhead
Evaporator Bottoms Disposal ($0.05/L l$.22/gal])
Schedule Production
Utilization (88 percent of schedule, 24 hr/day)
23 L/min (6 gpm)
85 percent
$10/hr, $66,OOQ/yr
$25,000/yr
$9,000/yr
$24,000/yr
$8/hr, $60,000/yr
$1.70/hr, ill,000/yr
$4/hr, $26,000/yr
$100,000/yr
$78,000/yr
310 days/yr
273 days/yr
a!984 Cost (per Ray Danaher, LUWA Corporation, July 23, 1984),
°5.00 square meters (53.8 square feet).
Source: Reference 2.
7-76
-------
TABLE 7.2.12. CAPITAL COST RECOVERY COMPONENTS FOR
ONSITE AIFE RECOVERY SYSTEMS
Thin film evaporator
Nominal feed rate
100 350 500
Process Equipment 126,000 212,900 264,900
Tanks 18.500 37.000 74.OOP
Subtotal(l) 144,500 249,900 338,900
Engineering, Electrical,
Instrumentation (20% of (D) 28.900 50.000 67.800
Subtotal (2) 173,400 299,900 406,700
Contingency 151 of (2) 26.000 45,000 61.000
TOTAL 199,400 344,900 467,700
Source: Reference 8.
7-77
-------
TABLE 7.2.13. THRESHOLD COSTS FOR OKSITE A1FE RECOVERY SYSTEMS
FOR VARYING SOLIDS CONTENTS AND PERCENT NET
DISCOUNTED CASH FLOW (DCF)a
Nominal
flow rate
Technology (gph) DCFb 5
Thin Film Evaporator 100
350
500
100
350
500
100
350
500
15 0.83
0.39
0.35
30 1.15
0.55
0.50
50 1.59
0.77
0.71
Solids (percent)
10 15 20
1.02
0.52
0.48
1.39
0.70
0.654
1.90
0.96
0.89
1.29
0.70
0.65
1.72
0.91
0.85
2.32
1.21
1.13
1.67
0.96
0.89
2.19
1.21
1.14
2.92
1.57
1.48
«1982 dollars.
"Percent net discounted cash flow.
Source; Reference 8.
7-78
-------
of these results (Figures 7.2.7 and 7.2.8), show how threshold costs become
increasingly independent of solids content and flow as nominal flow rate
increases. This reflects decreasing unit capital and labor costs.
For most onsite facilities, threshold cost will be relatively insensitive
to variations in equipment capital costs due to the effects of labor on
overall recovered solvent cost. Pace concluded that a threshold change of
only plus or minus 5 percent may result from capital cost fluctuations of 20
per cent » However, cost will be sensitive to changes in on-stream time
(capacity factor). This is illustrated in Table 7.2.14 for the case of
constant labor (2 employees) and increasing labor with run time. Cost of
recovery is highly sensitive to labor costs, particularly for small units, and
equipment utilization as shown.
If the solvent recovery equipment were operated less than 40 hours per
week, threshold costs could increase dramatically due to increased fixed cost
per gallon recovered. For example, cost per gallon for a 100 gph unit jumps
from $1.02 to $1.38 if the unit is operated only 50 percent of the time ($2.10
at 25 percent utilization). Assuming a minimum of two people are required to
operate a ATFE system, significant cost savings could not be expected by using
a 50 gph system since labor costs will compensate for much of the reduced
capital cost. Instead, low utilization rates may result in more economical
processing offsite or through some combination of onsite distillation and
offsite recovery of still bottoms. Offsite recovery is discussed in
Section 3.4. In general, offsite facilities incur additional fixed costs
(e.g., transportation, permitting, siting, boiler and lab facilities) relative
to onsite recovery systems. However, these costs are offset by their high
volume capability.
7.2.3.4 Cost Comparisons Between Pace and Site Visit Reports—
A rough comparison of the Pace cost estimates can be made with the case
studies presented earlier that were performed by GCA and RTI. Using cost data
supplied by Milsolv to GCA personnel (Tables 7.2.9 and 7.2.10), ATFE
processing costs can be compared with estimates derived from Figure 7.2.7.
With an average feed rate of 169 gph at 9.2 percent solids Pace estimates
processing costs of approximately $0.85 per gallon recovered compared to $l-$2
as reported by Milsolv. The major difference in the cost figures is accounted
7-79
-------
5.00
4.00
I
CO
o
(0
O
U
I-
III
-
O <
(0 0
2 8 2.00
K
III
O
u 1.00
m
BASIS: 15% DCF
I
i
100
200 300
NOMINAL FEEDRATE (GPH)
400
500
Figure 7.2.7. Threshold costs for onsiteATFE versus feed solids concentration.
Source: Pace, Reference 8.
-------
(0
o
o
Ul _
Ul
K
IU
O
u
III
E
5.00
4.00
3.00
2.00
1.00
1
2
3
4
5
6
100 GPH,
100 GPH,
100 GPH,
350 GPH,
350 GPH,
350 GPH,
15% DCF
30% DCF
50% DCF
15% DCF
30% DCF
50% DCF
10 15
SOLIDS
CONCENTRATION (%)
20
25
Figure 7.2.8. Threshold cost vs. feed solids concentration for solvent recovery using ATFE.
Source: Pace, Reference 8.
-------
TABLE 7.2.14. THRESHOLD COSTS FOR ONSITE SYSTEMS VERSUS
CHANGES IN CAPACITY FACTOR (Recovered Solvent)3
BASIS: 2,080 hours base nominal run time
10% solids
15% discounted cash flow (after taxes)
Nominal
feed rate
(gph)
100
500
Number of
persons
(labor)
2
3
4
5
2
3
4
5
Run
time
(hours)
2,000
3,000
4,000
6,000
2,000
3,000
4,000
6,000
Constant
labor
($/gal)
1.02
0.77
0.63
0.49
0.48
0.39
0.35
0.30
Variable
labor
($/gal)
1.02
0.91
0.84
0.70
0.48
0.42
0.39
0.34
a!982 dollars.
Source: Reference 8.
7-82
-------
for by a higher projected recovery rate by Pace (80% overhead) versus that
reported by Milsolv (441 overhead). Milsolv generated bottoms which are
blended as fuel at $0.26/gal while Pace's estimates provide far more complete
recovery and higher bottoms disposal cost ($0.54/gal). If costs were based on
feed rate as opposed to overhead rate, Pace's estimate becomes $0.68/gal
versus $0.66 based on Milsolv1s figures.
Substituting figures provided through RTI's interviews with Plant A
2
personnel (Table 7.2.11) , processing costs were approximately 20^/gal of
recovered product based on the capital depreciation schedule and land use
g
costs suggested by Pace for commercial facilities. Pace's figures result
in an estimated 40^/gal when the ATFE is assumed to account for 50 percent of
the facility's fixed costs. This figure is higher primarily because of the
increased disposal cost (54^/lb of bottoms versus 22^ reported by Plant A) and
labor costs assumed in the Pace analysis. Overall costs from both estimates
are low due to high utilization (6,473 hours per year) and high recovery rate
(87.5 percent). Energy consumption (4^/gal) and LUWA purchase and
installation cost ($294,000) were in good agreement.
Specific operating and capital costs were not discussed for Plant B
primarily because of the diversity of process systems at this facility. In
addition, processing costs depend upon the particular feed stream being
processed and the intended use of the product. However, Plant B personnel did
state that it would not be economical to recover waste streams with less than
6 to 8 percent reclaimable organics. They provided additional processing
information as follows:
• Processing costs are approximately $l/gal when organic is stripped
as the overhead product.
* Processing costs are approximately $1.50/gal when water is stripped
overhead with the organic becoming the bottoms product.
* The installed cost of a new 5.76 m2 (62 ft2) Votator thin film
evaporator system would be about $300,000.
• The cost of shipping ATFE bottoms to a cement kiln are approximately
20 to 30 cents per gallon.
7-83
-------
Thus, capital equipment and processing costs (tl.OO/gai) are in general
agreement with estimates provided by Pace for solvent reprocessing facilities
(*0.61/gal).8
Cost data from Plant C was not detailed enough to make any meaningful
comparison with other data.
7.2.4 OverallStatus of Process
7.2.4.1 Availability—
ATFEs are widely used in a number of industries including the solvent
recovery industry due to their unique ability to process viscous wastes
relative to other evaporation/distillation technologies. Evaporators and
accessory equipment can be obtained from a number of manufacturers in various
sizes and configurations. Ten firms are identified as suppliers of the ATFE
in the 1986 edition of McGraw-Hill's Chemical Engineering Equipment Buyers'
9
Guide. Major producers include Blaw-Knox (Buffalo, NY), Cherry Burrell
(Louisville, KY), LUWA (Charlotte, NC), Kontro (Orange, MA), Pfaudler
2
(Rochester, NY) and Artisan Industries (Waltham, MA).
7.2.4.2 Application—
Evaporators can be used to recover solvents and other volatile organics
from both organic and aqueous waste streams provided the treated waste does
not exceed viscosity limits imposed by the system design (see Table 7.2.1)'
and operating temperature. Excessive solids content will increase
viscosity and foul heat transfer surfaces. Therefore, some pretreatment for
gross solids removal may be required.
In practice, recovery to low residual organics is limited by viscosity
due to increased resistance to mass transfer. This resistence is partially
offset in an ATFE due to the high turbulence generated in the vessel.
Recovery is also limited by operating pressure since this pressure determines
the equilibrium concentration of solvent remaining in the waste. Finally,
recovery of organics from waste streams may not be economical unless the
2
recoverable organic content is greater than 6 to 8 percent. A rough cost
analysis based on solvent purchase and disposal costs supports this figure.
7-84
-------
However, wastes with recoverable solvent concentrations of as little as 3 to
5 percent may be profitably recovered when processed in existing, high volume
onsite facilities which are underutilized.
ATFEs are likely to find more widespread use relative to conventional
distillation equipment. Land disposal restrictions and limitations on halogen
content in supplemental fuels will compel recyclers to pursue higher recovery
rates when processing waste halogenated solvents.
7.2.4.3 Environmental Impacts—
In selecting evaporators as a treatment technology it should be
recognized that, except in isolated cases, further treatment of the bottoms
stream will be required to meet EPA land disposal or NPDES discharge
requirements. Air emissions from the overhead condenser have been identified
2
as a potentially significant source of VOC emissions by RTI. VOC
concentration averaged 41.1 mg/l in Plant A and 34.4 mg/1 in Plant B at the
vacuum pump outlet. However, no estimates of total release of emissions
factors were provided. Emission rate would be greatest during process
start—up. They would increase if air was leaking into the system,
noncondensible gases were being formed or if the condenser became overloaded.
Vacuum pump emissions controls should be examined as a potential additional
cost since treatment requirements (e.g., carbon adsorbers) may be necessary to
avoid adverse environmental impacts.
7.2.4.4 Advantages and Limitations—
Evaporation, as a means of recovering useful solvents and other low
boiling organic materials, is a common unit operation used by a variety of
industries. It also finds application in removing water from viscous,
non-volatile fluids. The ATFE units most significant advantage compared to
other recovery processes is its ability to handle viscous liquids. However,
its cost must be compared with that of less expensive, conventional recovery
technologies (e.g., distillation) and their associated residual treatment
costs; e.g. thermal destruction, solidification, and land disposal. The cost
of the entire treatment train will ultimately dictate selection of the optimal
recovery technique.
7-85
-------
Limitations arise from the need to follow evaporation processes with
other treatment technologies in order to eliminate bottoms stream and,
possibly, vapor phase residuals.
7-86
-------
REFERENCES
1. Kappenberger, P. F., P. G. Bhandarkar. LUWA Ltd., Process Engineering
Division. Zurich, Switz. Thin Film Technology in Environmental
Protection. Chemical Age of India V. 36(1). January 1985.
2. Allen, C. C., et al. Research Triangle Institute. Field Evaluation of
Hazardous Waste Pretreatment as an Air Pollution Control Technique.
U.S. EPA/ORD, Cincinnati, OH. January 1986.
3. Luwa Corporation, Luwa Thin-Film Evaporator Technology,
Bulletin EV-24, 1982.
4. Kontro Company, Bulletin 7510.
5. Cherry-Burrel, ANCO/Votator Division, Turba-film Evaporator Bulletin.
6. Pfaudler Co., Recover Wash Solvent with the Pfaudler Solvent Recovery
System, Data Sheet 146, Supplement 1.
7. Roeck, D.t et al. GCA Technology Division, Inc. Sampling Data Collected
at Milsotv Corporation, Milwaukee, WI. under Contract No. 68-03-3243 with
the U.S. EPA Office of Solid Waste. December 1986.
8. Horsak, R. D., et al. Pace Company Consultants and Engineers, Inc.
Solvent Recovery in the United States: 1980 - 1990. Houston, TX.
Prepared for Harding Lawson Associates. January 1983.
9. Chemical Engineering Equipment Buyers' Guide. McGraw Hill. 1986.
. McGr
7-87
-------
7.3 STEM STRIPPING
7.3.1 Process Description
Steam stripping is used in both industrial chemical production and waste
treatment to remove gases or volatile organic chemicals from waste streams by
injection of steam into a tray or packed distillation column. This unit
operation is most effectively applied in aqueous solutions for the removal of
volatile components that are immiscible in water. It can also be used for
stripping organic solutions when water forms low boiling azeotropes and does
not adversely affect overhead or bottoms quality. The presence of water must
either be acceptable or economically separated to achieve product purity
specifications.
Steam stripping is commonly employed to separate halogenated and certain
aromatic compounds from water, but is less effectively used to recover
miscible organics such as ketones or alcohols. It is preferable to
conventional distillation processes for recovering high yields of contaminated
wastes which would otherwise foul heat transfer surfaces. It is also more
economical and effective at recovering wastes with high concentration of
volatiles and wastes with low volatility when compared to air stripping.
Figure 7.3.1 illustrates a typical steam stripping process. Waste enters
near the top of the column and then flows by gravity countercurrent to steam.
As the waste passes down through the column it contacts vapors rising from the
bottom of the column that contain progressively less volatile organic
compounds* The concentration of volatile compounds in the waste reaches a
minimum at the bottom of the column where it is discharged. The overhead
vapor is condensed as it exits the column and the condensate is then decanted
to achieve solvent/water separation. Reflux may or may not be used, depending
upon the desired composition of the overhead stream.
2
The common uses for steam distillation can be summarized as follows :
1. To separate relatively small amounts of a volatile impurity from a
large amount of material.
2. To separate appreciable quantities of low solubility, higher-boiling
materials from nonvolatile wastes when the materials to be separated
form low boiling azeotropes with water.
7-88
-------
REFLUX (DECANTED AQUEOUS PHASE)
WASTE FEED
FROM
PROCESS
ED
WASTE
STORAGE/
PRETREATMENT
TANK
PREHEATE
R
OVERHEAD
VAPOR
VO
INDIRECT
HEATING
t
NON-CONTACT
COOLING WATER
COUNTERFLOW
STEAM
STRIPPING
COLUMN
-RECOVERED ORGANIC
PHASE TO STORAGE
OR POST-TREATMENT
I I
TREATED AQUEOUS
WASTE DISCHARGE
TO REBOILER, STORAGE
OR POST-TREATMENT
•DIRECT STEAM
ADDITION
Figure 7.3.1. Typical Steam Stripping Process.
-------
3. To recover material which is thermally unstable or reacts with other
waste components at the boiling temperature.
4. To recover material which cannot be distilled by indirect heating
even under low pressure, because of the high boiling temperature.
5. To recover material in instances where direct-fired heaters cannot
• be used because of ignition or explosion hazards.
Theoretical Considerations—
The residual streams from steam stripping of aqueous wastes typically
consist of decanted overhead products and treated waste stream bottoms. The
stripped waste is sewered or undergoes additional treatment as necessary to
further reduce contaminant levels; e.g., carbon adsorption. Depending on
required purity, decanted solvent is either used directly or further purified;
e.g., drying, fractionation. The overhead aqueous phase is typically returned
to the stripping column if even a slight solubility exists between water and
the organic components.
The use of steam in distillation permits a more complete separation of
immiscible liquids at lower temperatures for the same conditions of total
pressure or vacuum. The essential feature of an immiscible system is that
each liquid phase exerts its own total vapor pressure, regardless of the
quantity of the other liquid phase. At constant system pressure, the presence
of steam reduces the total vapor pressure from the liquids which is required
to induce boiling, thereby lowering system temperature requirements. This
permits separation of compounds which could not be accomplished through
conventional distillation due to polymerization (e.g., pyridine, cresols,
monomers) or thermal decomposition (e.g., halogenated solvents) of waste
constituents.
With the exception of certain acids, most organic compounds produce
minimum boiling azeotropes with water. This phenomena is characteristic of
mixtures with dissimilar molecular species with activity coefficients greater
than unity* As can be seen in Appendix A, most solvents and ignitables fall
into this category.
A minimum boiling azeotrope forms at a temperature below that of the
boiling point of the pure compounds. Unless this azeotrope is shifted to'more
7-90
-------
favorable equilibrium conditions through lowering operating pressure or
addition of a chemical complexor (entrainer), it will act as the limiting
concentration which can be achieved in the overhead product. Kith some
organic mixtures, water can act as an entrainer to preferentially distill
compounds by creating a low boiling azeotrope.
As compounds become more dissimilar, they tend to approach liquid
immiscibility (e.g., chlorinated solvents or aromatics in water). Their
equilibrium vapor concentration will be essentially constant over an
increasing range in liquid concentrations and only begin to deviate from this
level at very high or very low liquid concentration. Generally, the azeotrope
will occur within the liquid immiscibility composition range forming a
heterogeneous overhead product which can easily be separated into two phases.
Examples clarifying this and other concepts are provided below.
Water and 1,1,2-trichloroethane, two slightly miscible liquids, boil at
86°C (1 atm) to form a heterogeneous azeotrope consisting of 83.6 percent
1,1,2-trichloroethylene. The normal boiling point of 1,1,2-trichloroethylene
(113.7°C), is substantially higher than the azeotropic boiling point, thus
heating costs are reduced in the presence of steam. The overhead product
readily separates into two phases, the upper layer consisting of 99.55 percent
water and the lower layer consisting of 99.95 percent 1,1,2-trichloroethylene
3
(specific gravity = 1.443) . Thus, steam distillation is readily applied to
this compound, particularly if it must be removed from polymerizable,
nonvolatile material which can foul distillation equipment at the normal
boiling point.
Steam stripping is also effective in instances where water acts as an
entrainer to shift the vapor-liquid equilibrium toward more desirable
conditions. For example, mixtures can shift toward higher concentrations of
the less ionic component in the overhead product in the presence of steam.
Benzene-alcohol overhead products resulting from steam stripping will be more
highly concentrated with benzene relative to normal distillation and will
separate into two phases upon condensing (e.g., benzene and isopropanol/
3
water), thereby further concentrating the components. A summary of
azeotrope data for solvents and ignitables was presented previously in
Table 7.1.1. Additional information regarding azeotrope theory and azeotropic
4 5
and extractive distillation can be found in standard engineering texts. '
7-91
-------
In the absence of azeotrope data for wastes, the principal indices used
to predict steam stripping feasibility are boiling point and Henry's Law
constant. Compounds with boiling points less than 150°C (i.e., volatile
compounds) have good steam stripping potential, as do compounds with Henry's
» n £"
Law constants greater than 10 atm-m /mole. The constant expresses
the equilibrium distribution of a compound between air and liquid for dilute
solutions. It is roughly proportional to the product of vapor pressure and
the reciprocal of solubility, thus taking into consideration the miscibility
of the compound in the liquid phase. Therefore, increasing value of the
Henry's Law constant also correlates with increasing favorability of
volatilization through the use of steam stripping.
It should be noted, however, that a study performed by the U.S.
EPA/OAQPS suggested that the use of Henry's Law constants given in the
literature for some chemicals could result in underestimating the required
contact time and overestimating the removal efficiency of steam stripping. As
part of the study, Henry's Law constants were calculated from headspace and
liquid composition sampling data. These calculated constants were
substantially less than their corresponding literature values, but did provide
good correlation with test data. It is recommended that vapor-liquid
equilibrium data be established through headspace analysis and activity
coefficient models for more complex solutions. Alternatively, relative
volatilities in non-ideal situations can be modeled through the use of
partition coefficients and critical constants.
7.3.1.1 Pretreatment and Post-Treatment Requirements—
Certain waste characteristics will determine the viability of steam
stripping as a waste treatment technique. Restrictive waste characteristics
include:11
• High solubility of the organic compound in water; usually more than
1,000 ppm;
• Organic compounds with high boiling points (more than 150°C);
7-92
-------
• VOC concentration in excess of 10 percent; above this concentration,
distillation may be more cost effective; and
* Suspended solids concentration in excess of 2 percent or materials
that tend to polymerize at operating temperatures; these can cause
fouling of packing material and eventual plugging of equipment.
Pretreatment requirements for wastes therefore consist of reducing high
concentrations of volatiles, solids, and polymerizable organics. Highly
concentrated volatile wastes are more economically pretreated through
conventional evaporation/distillation technologies. The diluted bottoms
product can then be treated via steam stripping. Alternatively, chemical
oxidation pretreatment can convert polymerizable and miscible wastes to more
inert and less soluble forms.
Solids, metals, oil and grease concentrations can be reduced through a
variety of pretreatment techniques as discussed in Section 6.0. These include
precipitation, coagulation, floculation, centrifugation, membrane separation
processes, flotation and other chemical/physical separation processes. For
example, fouling of packing material with oxidized iron and manganese can be
reduced through pretreatment via lime precipitation. Membrane separation
processes are effective in removing high molecular weight compounds* Since
these compounds are typically nonvolatile, and thus not ammenable to steam
stripping, pretreatment methods using membrane separation techniques
compliment steam stripping removal efficiency while reducing column fouling.
High solids removal in pretreatment steps also enables higher solvent recovery
rates since often times maintaining a pumpable bottoms product is the limiting
factor in solvent recovery."
Post-treatment is generally required of both the overhead and bottoms
stream, although data show that steam stripping may be capable of reducing
solvent, concentration in wastewater bottoms streams to levels which meet the
land disposal ban treatment standards. However, due to economic
considerations, conventional wastewater treatment methods (e.g., adsorption,
air stripping, biological or chemical treatment) are more commonly employed to
remove residual organic levels from aqueous streams.
7-93
-------
Concentrated organic bottoms such as oily wastes and low volatility,
organic1* sludges must be separated from the condensed steam through decanting,
eentrifugation, and other physical separation techniques. Overhead products
undergo liquid-liquid separation, typically through decanting followed by
dehydrating of the recovered solvent. Depending on its solvent levels, the
decanted aqueous stream is reprocessed through the stripper or treated via
other wastewater treatment processes. Alternatively, it could be used along
with a portion of the stripped wastewater bottoms to generate steam if the
boilers are properly equipped to accommodate the presence of volatile
10
components.
7.3.1.2 Design Characteristics and Operating Parameters—
Stripping towers operate in a batch or continuous mode. Generally, batch
stripping is of less commercial interest and is reserved for low volume
processing or for staged stripping of streams with multiple volatile
components which have different boiling points. Continuous operation more
effectively separates components of comparable volatility, provides higher
purity of separated products and uses less stripping medium for the same
degree of separation, particularly when stripping to low levels of organics.
Three modes of flow are possible: cocurrent, countercurrent, and
crossflow. Cocurrent flow, being least efficient, is not usually used, while
crossflow operation is often preferred to counterflow since it provides
greater transfer efficiency over a wider operating range.
A tower can be operated isothermally or adiabatically. Steam stripping
is typically performed isothermally; i.e., temperature is constant along the
length of the tower. The feed is usually preheated to the boiling point
before entering the tower to minimize steam requirements and, consequently,
treated waste volume.
Reflux involves condensing a portion of the vapors from the top product
and returning it to the tower. This can enhance separation by increasing the
concentration of the stripped organics in the vapor stream. This occurs
because condensation of vapor in the column is required to heat the refluxed
7-94
-------
liquid to its bubble point* This increases the concentration of strippable
*.
components in the liquid stream which, in turn, will increase their
equilibrium vapor concentrations. This effect will be most important as the
solution components become more miscible in one another.
Similarly, for miscible fluids, introducing the feed at a lower tray in
addition to refluxing can increase the concentration of organics beyond that
obtainable by reflux alone. Addition of reflux shifts the distribution in the
column from rectifying to stripping zones. A column designed with variable
feed plate location can accommodate this shift, as well as changes in waste
feed to permit column operation at maximum efficiency.
The optimal size of the rectifying zone depends on the waste being
treated. If feed enters at the top of the column (i.e., no rectifying zone),
the limiting overhead concentration is given by the vapor equilibrium with the
feed. As the rectifying zone is increased, the overhead concentration is
similarly increased and approaches the azeotropic concentration limits.
Finally, stripping can be carried out in two types of towers. Tray
towers provide staged contact between the liquid and vapor streams.
Alternatively, packed towers can be used which allow for continuous contact
between the two phases. Packed towers are less expensive, have low liquid
hold—up, low pressure drop, and are preferred for low pressure operation and
g
treatment of corrosive, foaming or viscous liquids. However, tray columns
have been more widely used in the past and consequently are more predictable
in their performance. Tray columns are more flexible in that they operate
efficiently over a wide range in flow rates and can readily be adapted to
Q
process multiple feeds or sidestreams. They are also more easily cleaned
and are, therefore, preferred for processing wastes with high concentrations
of metals, solids, or polymerizable materials. Selection of packed versus
tray towers has been discussed in more detail in Section 7.1.
Steam stripper design will ultimately be dependent on the waste
characteristics, throughput, and desired residual characteristics. Thus,
tower height, diameter, packing material and bed volume (or type and number of
trays), materials of construction, and use of ancillary equipment (e.g.,
reflux, heat exchangers) are highly specific to the waste being treated. For
example, a survey of commercial steam strippers currently in use to treat
pollutants revealed tower diameter ranges of 1.0 to 9.5 feet, column height
Q
ranges of 10 to 180 feet, and throughputs ranging as high as 500,000 GPD.
7-95
-------
In typical applications of stripping volatiles from aqueous wastes, steam
requirements range from 10 to 30 mole percent of the feed at system pressure
Q
of 1 atra and 100°C. Steam consumption is directly related to the
equilibrium vapor pressure of the material being stripped and the resistance
to diffusion of the compound through the waste. The latter determines the
extent to which equilibrium conditions are approached and becomes increasingly
important as the concentration of volatile species is diminished. Equipment
manufacturers provide steam consumption data for stripping organic streams
(e.g., degreasing solvents) which are appropriate if significant solvent
quantities remain in the bottoms product. For example, one manufacturer
reported 1,236 lbs/hr of steam required to steam distill spent mineral spirits
in a 100 gph capacity still. Processing xylene on the same unit would require
829 lbs/hr whereas toluene steam requirements would be only 419 lbs/hr.
Steam strippers are generally custom designed for specific applications.
Conversely, steam injection stills include modified pot stills and low volume
packaged units. These have throughput ranges of 10 gallons/hour or less and
are available in portable skid-mounted designs. However, despite this
versatility jin design, in practice steam strippers are limited to the extent
that they can recover solvent wastes.
For compounds which form minimum boiling azeotropes, purity of the
bottoms product is limited by the equilibrium liquid phase mole fraction
which, in turn, is limited by economic considerations. Commercially available
distillation systems have lower operating pressure limits due to the increased
costs associated with maintaining a vacuum and condensing overhead product at
temperatures below the range of air or water cooling.
A given pressure (typically atmospheric) defines the temperature required
to boil a mixture. As the liquid mole fraction of the stripped component
decreases in the bottoms product, the temperature required to continue
stripping approaches the boiling point of the remaining liquid. If thermal
decomposition or polymerization can occur, temperature as well as pressure
limitations may restrict attainable degree of separation. Otherwise, economic
factors associated with capital costs of the column can make post-treatment
alternatives (e.g., activated carbon) more attractive as methods to separate
residual contaminant levels from the bottoms stream.
7-96
-------
As the initial liquid concentration approaches that of the equilibrium
vapor concentration, the utility of steam distillation as a concentration
technique is reduced. Conversely,, compounds which form azeotropes at weight
fractions significantly above the weight fraction in the untreated waste are
good candidates for concentration through steam stripping, particularly if the
compound is volatile. More soluble compounds will only be ammenable to steam
stripping if they are highly volatile.
The reader is referred to design methodologies and cost estimation
procedures which have been developed in the literature for more information on
12459
steam stripping design, optimization, and cost effectiveness. ' » » »
7.3.2 Demonstrate d Berformance
The results of a number of different case studies, both pilot-scale and
full—scale, are presented in this section. Each subsection addresses the
results of a specific study. The first two cases involve full-scale batch
steam distillation of organic wastes at solvent recovery facilities. These
are followed by summaries of performance data characterizing operation of
13 industrial scale, continuous flow steam strippers treating solvent
contaminated wastewater streams. Finally, results from two laboratory scale
steam stripping studies are summarized. These studies were conducted to
design full—scale units for treating petrochemical processing wastes and
ground water which was heavily contaminated with solvents.
Steam Injection Still—
GCA conducted a performance evaluation of a steam injection still
operated by Environmental Processing Services (EPS) of Dayton, Ohio in
12
October, 1985. The roughing still employed by EPS to treat halogenated
solvent wastes is a direct steam injection batch still which is operated at
atmospheric pressure. Initially, three drums of waste solvent are pumped into
the still with another drum of solvent waste added approximately once every
half hour. As the run progresses, temperatures in the still increase from
just over 100°F and approach the boiling point of water (212°F). Attainment
of this temperature indicates that most of the solvent has been removed.
After ten drums are processed, still bottoms are discharged to a holding tank
from which they are shipped offsite for use as a supplemental fuel in a cement
kiln. _
-------
Steam supplied by an existing onsite boiler is injected directly into the
solvent waste. As the solvent is heated by the steam, water and solvent
vaporize, pass through a condenser, and are discharged to a coalescer* The
decanted water layer is discharged to the sewer and recovered solvent is sent
through a polishing still for final treatment.
A mixed stream of spent 1,1,1-trichloroethane was processed during the
sampling period. Processing data are summarized in Table 7.3,1. A total of
ten drums were processed in 264 minutes, yielding an average feed rate to the
batch unit of 2 GPU* A recovery rate of over 90 percent was reported for
1,1,1-trichloroethane. Processing of this waste stream was limited to the
extent required to generate a bottoms product which was acceptable as a. cement
kiln fuel substitute; e.g., Cl content of roughly 3 percent.
GCA collected samples of the 1,1,1-triehioroethane waste feed, still
bottoms, solvent overhead product (from the polishing still), and the sewered
water phase which was separated in the coalescer. These streams were sampled
for 1,1,1-triehloroethane and metals as summarized in Tables 7.3.2 and 7.3.3,
respectively. As shown, 1,1,1-trichloroethane concentration in the bottoms
product was significantly reduced with overall solvent recovery exceeding
90 percent. Solvent concentration in the sewered aqueous stream was 6 percent
which indicates that this stream should have been subjected to additional
treatment.
Some metals were concentrated in the bottoms product from 4 to 7 times
greater than levels found in the waste feed. In particular, lead was present
in high enough concentration (14 ppm) in the still bottoms so that it may fail
the EPA toxicity test. The presence of metals in the overhead products
indicate some carry-over of these products with the distillate. Zinc, copper,
lead and chromium were present in these streams in significant
concentrations.
The still bottoms were also analyzed for fuel value. The overall heat
content was 14,240 Btu/lb with a viscosity of 38 SSU (at 100°F), ash content
of 0.05 percent, sulfur content of 0.56 percent, and nitrogen content of
0.04 percent. GCA noted that removal of the lower phase (85 percent water)
could significantly increase the fuel value of this residual stream.
7-98
-------
TABLE 7.3.1 PROCESS DATA FOR THE STEAM DISTILLATION UNIT
Run Number
Date of Test
Waste Description
EPA Waste Code
Number of drums processed
during test (55 gals/each)
Length of test
period (minutes)
Process Rate (gal/min)
Temperature (°P)
4
10/29/85
i,1,I Trichlorethane waste solvent coming
from several generators.
F001
10
264
2.1a
212b
a!0 drums were processed in 264 minutes. The feed rate is not constant.
Initially 3 drums are pumped into the unit in about 15 minutes. About once
every hour another drum is added (takes about 3 minutes to pump entire
contents of a drum) to the still after sufficient solvent has been distilled.
°0ver the course of the run the temperature gradually increases until the
boiling point of water is reached.
Source: GCA. Reference No. 12.
7-99
-------
TABLE 7.3.2 PERCENT BY WEIGHT OF 1,1,1-TRICHLOROETHANE IN THE WASTE STREAMS
DURING OPERATION OF THE STEAM DISTILLATION UNIT
Upper phase
Single phase
Lower phase
Waste Feed
Still Bottoms
Solvent Product
Roughing Still Water
5.0a
2.8b
92.0
4.4C
89.0
5.5
aPrimarily water (49.6%) and contaminants such as solids and oils.
^Primarily oil and Solids (0.14% water)
cPrimarily water (85.7%)
Source: GCA. Reference No. 12.
7-100
-------
TABLE 7.3.3 METAL ANALYSES FOR THE STEAM DISTILLATION UNIT (pg/g)
Waste Feed
Metal
Ag
As
Be
Cd
Cr
f Gu
o
H- Ni
Pb
Sb
Se
Tl
Zn
1
<0.08
<0.32
0.0080
0.057
0.607
3.5
0.348
7.7
<0.40
<0.32
<0.24
15.8
2
<0.08
<0.33
0.12
0.074
0.885
4.3
0.442
3.7
0.82
<0.33
<0.24
20.2
3
<0.08
<0.32
0.008
0.065
0.454
4.0
0.333
4.8
<0.40
<0.32
<0.24
18.2
Still Bottoms
<0.09
<0.36
0.060
0.404
4.57
17
1.93
14
0.98
<0.36
<0.27
132
Roughing Still Water Solvent Product
1
0.02
<0.04
<0.0005
<0.002
<0.003
0.31
<0.006
0.12
<0.05
<0.04
<0.03
0.315
2
<0.01
<0.04
<0.0005
0.002
<0.003
0.35
<0.006
.011
<0.05
<0.04
<0.03
0.320
3
0.02
<0.04
< 0.0005
0.003
<0.003
0.32
<0.006
<0.12
<0.05
<0.04
<0.03
0.294
0.004
<0.013
<0.0003
< 0.0006
0.016
0.014
0.006
0.006
<0.016
<0.013
<0.010
0.004
Source: GCA. Reference No.12.
-------
Finally, the extent of volatile release from the process was determined
by taking OVA measurements (readings were in methane equivalents) during the
264-minute run. Sampling locations included directly above the coalescer and
near the two exits in the room where the unit was located. The highest
recorded emissions were 30 ppm above the coalescer, but levels dropped to less
than 20 ppm at the room exits (this probably includes compounds other than
1,1,1-triehloroethane).
The data indicate that significant recovery is achievable in
steam-injected stills. Mixed chlorinated waste, containing approximately
30 percent solvent contaminated with oil, water and solids, was processed to
yield over 90 percent solvent recovery while generating a fuel bottoms
product. However, a metals analysis on the bottoms showed levels of toxic
compounds (lead, chromium) which were comparable to levels found in waste
oil. These could represent a health hazard if burned in boilers which are not
equipped with particulate emission controls.
Full-Scale Steam Distillation (Stripping)—
Research Triangle Institute (RTI) and Associated Technologies, Inc.,
under contract with the EPA, collected field data at a hazardous waste
facility, "Plant D", to evaluate the performance of a batch steam stripping
13
unit used to recover solvents from waste.
The contaminated organics processed by Plant D are generated mostly by
the chemicals, paint, pharmaceutical, plastics, and heavy manufacturing
industries. The types of chemicals recovered include ketones, aromatic
hydrocarbons, chlorinated solvents, Freons, and petroleum naphthas.
Generally, 50 to 70 percent of the waste consists of recoverable solvent which
is returned to the generator or marketed to suitable end users. Aqueous
residues from the stripping process are either sewered or solidified by mixing
with sorbents and landfilled. Organic residues are typically incinerated.
The steam stripping system, consists of a 250-gallon batch stripping
vessel equipped with a steam sparger, overhead vapor condenser, distillate
receiver and decanter, a miscible solvent tank, product storage tanks, a
residue tank, and associated pumps and support facilities.
7-102
-------
A typical 250-gallon batch can be processed in one hour. Most solvents
handled by the facility are relatively immiscible with water and decant
readily. The aqueous phase from the decanter is collected in a miscible
solvent tank (MST) to be reprocessed through the stripper for recovery of
residual solvent. In many cases, the stripped aqueous phase is then suitable
for discharge to the municipal wastewater treatment system.
Four batches of wastes were evaluated: an aqueous xylene waste
(Batch 1), a chlorinated organic-oil mixture (Batch 2), a chlorinated
organic-water mixture (Batch 3), and a mixture of solvents and water
(Batch 4), Waste characterization data for these batches is summarized in
Table 7.3.4. Processing data is summarized in Table 7.3.5. Steam stripping
of all batches was discontinued when the vapor temperature reached 211°F. to
avoid excessive water in the overhead product.
Batches 1, 3, and 4 were two-phase liquid systems with the organic phase
ranging from 3 to 19 percent by weight. After steam stripping, the bottoms
were reduced to a single-phase aqueous product with volatile organic
constituent (VOC) concentrations ranging from 391 to 12,031 ppm. VOC removal
efficiencies ranged from 93.7 to 99.8 percent.
Batch 2 was a solvent-oil mixture with an initial VOC content of
74 percent by volume. This was stripped to 0.41 percent VOC (99.5 percent
removal efficiency) yielding an oil-water bottoms product.
Steam requirements varied considerably depending on the batch volume,
latent heat, heat of vaporization, and heat transfer efficiency. Direct steam
injection was used to heat the waste to its boiling point. This results in
maximum steam consumption and water concentration in both the overhead and
bottoms products. ~fhe ratio of steam consumption to recovered VOCs was lowest
for the organic waste (Batch 2) and highest for the least concentrated stream
(Batch 4) as shown in Table 7.3.5.
RTI also presented data for headspace and liquid VOC concentrations as a
function of stripping time. However, these data are incomplete for the three
multiphase batches, since the organic phases were not fully characterized.
Thus, strictly speaking, stripping rate constants can only be meaningfully
discussed for Batch 2.
7-103
-------
TABLE 7.3.4. PLANT D: CHARACTERIZATION OF WASTES
Number of phases
Total solids (ng/L)
Water (weight Z)
Oil (weight %)
VOC (weight %)
Density (g/cm )
VOC Analysis (influent/bottoms)
Aqueous phase (ing/L)
Acetone
Isopropanol
Methyl ethyl ketone
1, 1, 1-lrichloroethane
Tetrachloroethene
Ethyl benzene
Toluene
Xylene
Organic phase (weight %)
Xylene
Methyl ethyl ketone
1,1, 1-Ir ichloroethane
Xylene and toluene
Miscellaneous organics
Batch 1
2
880
81.3
Negligible
18.7
1.0a, 0.866b
39/<6
960/<6
1,040/34
170/20
290/<20
360/100
86/42
2,000/270
70.0/0
-
-
-
30.0/0
Batch 2 Batch 3
1 2
2,800 130
Negligible 82
19.9 Negligible
80.1 18.0
1.2d 1.0e
290/<6
37/<6
320/<7
180,000/12,000
-
44/12
_
_
_
75.000/ 7C
660, 000/4, 100C <100/0
_
>0 >0/0
Batch 4
2
130
97
Negligible
3.1
1.0
6.500/ 6
951-
112/-
2,200/230
551-
-
170f/35
900 £/ 120
-
_
-
< 100/0
>0/0
aAqueous phase.
"Organic phase.
WL
"Estimated from pure components.
eEstimated as density of water.
^Values reported here, are concentrations reported after 10 minutes of stripping.
Actual measured values were: toluene - 86 rag/L; to xylene - 4 mg/L. These were
low due to incomplete mixing when the feed sample was collected.
Source: Adapted from RTI - Reference No. 13.
7-104
-------
TABU; 7.3.5. PLANT "D" BATCH STEAM STRIPPER: PROCESS DATA3
Initial waste volume (L)
Total VOC constituents (L)
Stripping and heating time (min. )
Stripping time (min. )
Steam rate (L/min)'5
Ratio of steam consumption to
recovered VOCs
Overhead product (L)
Organic phase
Aqueous phase
Total VOC constituents1*
Stripper residual (L)
Oil content
Water content
VOC content
Percent VOC removed
Batch 1
1,260
272
125
86
5.9
2.72
581
333
248
271.2
1,420
Negligible
1,420
0.82
99.7
Batch 2
897
664
103
87
4.69
0.73
1,060
660
400
662.7
320
233
87
1.3
99.8
Batch 3
564
69
57
57
2.82
2.33
225
45
180
64.6
545
Negligible
539
4.4
93.7
Batch 4
360
12.8
50
33
2.64
10.4
143
3.1
140
12.7
349
Negligible
344
0.14
98.7
aSource: Adapted from RTI. Reference No. 13.
''Calculated based on mass balance.
7-105
-------
For Batch 2, stripping rate showed a negative, linear correlation between
stripping time and the logorithm of waste concentration. Methyl ethyl ketone
(MEK) had a stripping rate constant which was 2.7 times that of
1,1,1-trichloroethane (1,1,1-TCE), suggesting that it has an activity
coefficient in the organic mixture which is significantly higher than that of
the more concentrated 1,1,1-TCE. This conclusion is supported by vapor-liquid
equilibrium data. A laboratory headspace analysis of the still contents
showed an equilibrium vapor-liquid partition coefficient for MEK which was
2.1 times that of 1,1,1-TCE (at 25°C), while condenser vent samples taken just
prior to stripping showed a ratio of 4.3 to 1 in the vapor phase..
Correlations between stripping time and the logarithm of waste
concentration were also nearly linear for Batches 3 and 4, since these had a
fairly high proportion of VOCs in the aqueous phase. However, stripping rate
constants presented by RTI do not correlate well with aqueous phase
theraodynamie property data (e.g., Henry's Law constant, vapor pressure,
solubility, activity coefficient), due to the competing mass transfer
mechanisms of liquid—liquid diffusion and volatilization from the organic
phase. Stripping rate constants also could not be compared with relative
volatility data provided by headspace analysis, since the organic phase was
not fully characterized (see Table 7.3.4) and, therefore, initial constituent
concentrations are unknown. Batch 2, which had a large fraction of VOCs in
the organic phase, showed poor correlation between the aqueous phase
concentration data (Table 7.3.4) and stripping rate constants developed by RTI,
RTI also collected air emissions data from the condenser vent and
developed emissions factors based on the results. Batch 1 showed a total VOC
_c
emission factor of 5,7 x 10 g/g VOC recovered and Batch 2 showed a value
— S
of 2.7 x 10 g/g VOC recovered.
Several conclusions can be drawn from RTI's analysis:
* As recoverable solvent concentration in the waste decreases, steam
consumption and cost will increase dramatically per unit of
recovered solvent;
* Headspace analyses provide the most reliable estimate of relative
volatility between compounds in a given waste;
7-106
-------
* Stripping rate of a given compound in a single-phase waste shows a
linear relationship between time and the logarithm of
concentration. This relationship can be used to estimate stripping
times required to achieve desired treatment levels. For example,
RTI predicted that 40 minutes would be required to reduce I,I,1-TCE
levels in Batch 3 by an order of magnitude with the processing
conditions summarized in Table 7,3.5. However, this correlation was
not adequately demonstrated at low concentrations such as those
specified in the solvent treatment standards;
* Use of Henry's Law constant as provided in the literature
significantly overestimate the stripping rate as determined through
sampling;
* Batch steam stripping' of the four waste streams was not economically
or technologically capable of reducing solvent concentrations to the
EPA-specified treatment standards. All overhead and bottoms
products required additional treatment to meet the standard.
Further stripping would have resulted in overhead product which
contained excessive amounts of water and, therefore, negated the
beneficial effect of treatment; and
* Recovery could have been improved by indirect heating of the waste
prior to steam stripping and segregation of overhead product during
the course of each batch run.
Industrial Steam Stripper Survey (OSW)—
The EPA Office of Solid Waste compiled operating and cost data from
18 Supplemental 308 Questionnaires for 36 full-scale industrial steam
strippers. Influent and effluent solvent concentration data were collected
for four steam strippers and one steam stripper/carbon adsorption combination
system. Data for compounds which had influent concentrations exceeding 2 ppm
are summarized in Table 7.3.6.
All waste streams shown in Table 7.3.6 were contaminated wastewaters.
However, waste constituents other than priority pollutants and processing
conditions were not provided in the reference. Thus, sampling results cannot
be interpreted in relation to these variables.
All solvent compounds, with the exception of nitrobenzene, were stripped
to levels of 2 ppm or less. Nitrobenzene was treated at two facilities
(Facility Nos. 246A and 297) in wastewater streams which also contained
benzene. With influent concentrations ranging from 87 to 1,966 ppm,
7-107
-------
TABLE 7.3.6. INDUSTRIAL STEAM STRIPPER SURVEY
O
00
Plant
284
246A
225
248
297
Number of
Compound name observations8
1 ,2-Dichloroethane
1,1, 2-Tr ichloroethane
Chloroethane
Chloroform
1, 1-Dichloroethene
Trans-1 , 2-Dichloroethene
Methylene chloride
Trichloroethylene
Benzene
Nitrobenzene
Methylene chloride
Toluene*1
Benzene
Nitrobenzene
15/15
8/15
15/15
15/15
15/15
15/15
5/5
9/9
14/13
15/15
7/7
1/1
10/10
10/10
Number of
nondetectabli
concentratiot
0/7
0/15
0/15
0/13
0/14
0/13
0/5
0/7
0/6
0/0
0/4
0/1
0/10
0/0
Concentration8
t Mean
is (tag/la)
9,615/0.056
6.811/0.010
20.39/0.050
399.3/0.010
4.358/0.010
13.68/0.014
5.132/0.010
3.049/0.019
819.9/0.045
391.0/251.3
1.973/0.308
8.650/0.010
92.16/0.010
190.4/11.79
Minimum
(mg/L)
2,340/0.010
0.220/0.010
0.690/0.050
7.330/0.010
0.200/0.010
4.860/0.010
2.400/0.010
0.210/0.010
0.239/0.010
91.20/94.23
1.290/0.010
8.650/0.010
34.69/0.010
87.00/4.600
Maximum
(mg/L)
23,476/0.374
14.50/0.010
42.00/0.050
1,088/0.016
10.80/0.013
43.00/0.057
12.10/0.010
10.30/0.085
2,008/0.171
1,966/619.6
5.100/0.985
8.650/0.010
147.2/0.010
330.0/21.99
iean removal
efficiency
'(minimum)
99.99+
99.85
99.75
99.99+
99.77
99.90
99.81
99.38
99.99
35.73
84.39
99.88
99.99
93.81
aNuniber of observations and concentration data given in the form of influent/effluent.
"Data includes steam stripping and carbon adsorption.
Source: U.S. EPA, Reference No. 7.
-------
nitrobenzene was stripped to effluent concentrations ranging from 4.6 to
620 ppm. Removal efficiency at the two facilities averaged only 36 and
94 percent. In contrast, benzene was consistently stripped from similar inlet
concentrations to levels below 2 ppm with removal efficiencies in excess of
99.99 percent. Benzene has a vapor pressure and Henry's Law constant which is
over two orders of magnitude higher than nitrobenzene. This, combined with
possible low temperature operation, account for the marked differences in
removal rates achieved for these two compounds.
Industrial Steam Stripper Survey—
As part of a program to develop pollution control standards for the
pesticide chemicals industry, the USEPA Effluent Guidelines Division evaluated
achievable performance of full-scale steam strippers used in the
14
industry. In addition, the EPA summarized available steam stripper
performance data from the organic chemicals industry which were used to strip
pesticide chemical compounds. Table 7.3.7 summarizes these data for solvent
compounds which were present in the influent wastewater stream at levels
exceeding 2 ppm. Eight full-scale systems were identified which stripped
pesticide wastes. Of these, concentration data were provided for only three
units as summarized in Table 7.3.7.
Plant 1 used a steam stripper to recycle methylene chloride. The
stripper operated at 960 gph feed with a nominal steam flow of 23 percent of
the feed rate. The column contained 15-feet of packing using l-inch
polypropylene saddles. Removal efficiency was 99.9 percent which was
sufficient to reduce the effluent concentration to less than 0.01 ppm.
Plant 2 operates a 24—tray steam stripper to remove chloroform and hexane
from pesticide wastewater. A feed rate of 2,700 gph is preheated and stripped
to less than 5 mg/L at a removal rate of over 93 percent. Overhead product is
incinerated onsite. No breakdown between chloroform and hexane concentration
was provided.
Plant 3 operates a vacuum stripper to reduce toluene levels and
wastewater temperature as a pretreatment step for resin adsorption. Toluene
had to be reduced to a level which would eliminate agglomeration in the resin
7-109
-------
TABLE 7,3.7. FULL-SCALE INDUSTRIAL STRIPPER PERFORMANCE SUMMARY
Plants using
steam stripping Stripped compound
Pesticide industry
Plant 1 Methylene chloride
Plant 2 Chloroform and hexane
Plant 3 Toluene
Organic chemicals industry
Plant 4 Benzene
»- Plant 5 Methylene chloride
o
Toluene
Plant 6a Methylene chloride
Carbon tetrachloride
Chloroform
Plant 6b Methylene chloride
Chloroform
1 , 2-Dichloroethane
Carbon tetrachloride
Benzene
Toluene
»
Plant 7 Methylene chloride
Chloroform
1 , 2-Dichloroethane
Concentration (ppm)
Influent
<159
70.0
721
<15.4
<3.02
178
1,430
<665
<8.81
4.73
<18.6
<36.2
<9.7
24.1
22.3
34
4,509
9,030
Effluent
<0.01
<5.0
43.4
<0.230
<0.0141
<52.8
<0.0153
< 0.0549
1.15
< 0.0021
<1.9
4.36
<0.030
<0.042
<0.091
<0.01
<0.01
<0.01
Percent
removal
99.9
>92.6
94.0
98.5
99.5
>70.3
>99.99
>99.99
<86.9
>99.95
89.8
88.0
99.7
>99.8
>99.6
>99.97
>99.99
>99.99
Sampling
period
(days)
3
-
>4
>30
>40
7
7
1
Source: U.S. EPA Effluent Guidelines Division, Reference No. 14.
-------
system regenerate. Vacuum operation reduces stream temperature, thereby
improving resin adsorptivity. Thus, this stripping system was not designed to
achieve maximum removal efficiency. Instead, the goal was optimal cost
effectiveness when applied in conjunction with adsorptive polishing which
resulted in a design stripper removal efficiency of 98.3 percent (actual
removal was 94 percent),
Other industrial steam strippers used in pesticide wastewater treatment
applications were identified, but not sampled. Solvent wastes treated include
xylene, isobutyl alcohol, methanol, toluene, 1,2-dichloroethane, ethylene
dichloride, and methylene chloride at inlet wastewater flow rates ranging from
33,000 to 90,000 GPD.
The USEPA also evaluated steam strippers used in the organic chemicals
industry. Inlet and effluent solvent concentrations' and removal efficiencies
for five strippers are summarized in Table 7.3.7. As shown, all compounds
were removed to levels below 2 ppm except for toluene (<52,8 ppm, Plant 5) and
1,2-dichloroethane (4.36 ppm, Plant 6B). As stated previously, Plant 5 is
operated only as a pretreatment method which explains why toluene was not
stripped to a greater extent. Conversely, Plant 7 is operated to achieve
maximum removal efficiency and was capable of achieving removal rates of
99.99 percent for all compounds.
The data which, in general, represented results of long-term testing,
demonstrate the difficulty of predicting removal efficiency on the basis of
single thermodynatnic property correlations. For example, Henry's Law
constants for methylene chloride and chloroform are nearly the same, but
methylene chloride appears.to be much more easily stripped (Plant 6A and 6B).
Similarly, toluene has a higher Henry's Law constant than methylene chloride,
but was not as easily stripped (Plant 5). This effect may be explained by the
fact that methylene chloride has a higher vapor pressure than either of the
other compounds. Vapor pressure may be a better indicator of attainable
removal efficiency for stripping towers equipped with rectifying sections or
which otherwise yield concentrated overhead products. In these instances,
Henry's Law assumption of a dilute aqueous solution does not apply.
7-1H
-------
Bench-Scale Steam Strippers—
GSRI, under contract with the USEPA Office of Research and Development,
performed laboratory scale investigations of steam stripping petrochemical
processing wastewater streams containing biorefractory halogenated organics.
Detailed analyses of individual contaminant concentrations were performed for
two separate streams.
The bench-scale steam stripper consisted of a 5.08 cm diameter column
packed with polypropylene Pall rings to a bed depth of 367 cm. The column
design was counterflow with provisions for reflux of decanted aqueous phase
overhead. Overhead arid bottoms product samples were composited throughout the
run, whereas the feed stream analysis for individual constituents was based on
a single sample collected at the end of the trial. A mass balance analysis
indicated that significant quantities of VOCs were emitted from the overhead
product. Thus, percent removal data were determined on the basis of bottoms
and feed concentrations. Since some variability in feed composition was
evident (see TOC data in Table 7.3.8), the calculated removal efficiencies may
deviate somewhat from actual conditions. Additional detail on experimental
design and procedures may be found in the reference.
Processing and individual compound removal efficiency data for steam
stripping of stream 221A are summarized in Tables 7.3.8 and 7.3.9,
respectively. The primary VOCs in stream 221A were ethylene dichloride and
1,2-dichloroethylene followed by dichloromethane, chloral and
1,1,1,2—tetrachloroethane (Table 7.3.9). No correlation between thermodynamic
property data and removal efficiency was observed although chloral showed
consistently low removal efficiency. This is probably a result of its high
solubility and only moderate volatility which result in a low partial pressure
relative to other VOCs in the waste.
In the cases with no reflux, removal of VOCs correlated positively with
overhead volume, as would be expected. Addition of reflux, if anything,
impaired removal efficiency due to the experimental set-up. Reflux,
consisting of the decanted overhead aqueous phase, was introduced at the
7-112
-------
TABLE 7.3.8. PROCESSING DATA FOR BENCH-SCALE STEAM STRIPPING OF STREAM 221
Test8
Volume (tuL/min)
Overhead (total)
Bottoms
Net overhead as a
percentage of feed (%)
Steam (mL/min)
Ratio of reflux to
net overhead
Stripper column TOO (iig/L)
Bottoms
Overhead
Feed
Total TOC removal (%)
Stripper column VOC (mg/L)
Bottoms
Overhead
Feed
Total VOC removal (%)
Selected VOC removal (Z)b
1
5.76
250
2.3
44.98
0
256
10, 446
645
60.3
809
11,639 ,
-
85.2
86.7
2
7.05
281.25
2.82
53.1
0
292
10,462
668
50.8
651
11,850
-
86.6
87.3
3
12.75
302.5
5.10
50.8
0
293
4,766
645
45.0
507
11,240
-
88.8
90.8
4
13.86
305
2.3
59.7
1.411
241
4,519
785
62.5
853.
9,621
-
81.0
81.4
5 V<
13.5
275
2.5
59.7
0.8511
243
9,806
636
58.0
1,111
16,297
_
77.7
78.0
alume Weighed
Average
10.58
282. 75
3.07
53.66
0.35
265.3
7,516
675.8
60.7
734
12,010
5,480
84.9
-
aProcessinjj conditions: column feed = 250 mL/min for all tests. Pressure ° 1 atm,
temperature of overhead and bottoms ranged from 102 to 104°C.
"This data includes only those VOCs which were quantified for all five test runs to
permit comparison between different processing conditions.
Source: Adapted from GSRI, Reference No. 15.
7-113
-------
TABLE 7.3.9. SUMMARY OF REMOVAL EFFICIENCY DATA FOR STREAM 221A
Compound
Vinylidene chloride
Dichloromethane
1, 2-Dichloroethylene
Chloroform
1, 1, l-Trichloroethaneb
Ethylene dichloride
Tr ich lor oethy lene
Chloral
lf 1, 2-Trichloroethane
terchloroethylene
1, 1, 1, 2-Tetraehloroethane
1,1,2, 2-Te trachloroe thane
Feed
composition
(n»g/L)
61.5
800.9
1,583.3
140.3
50.9
1,593.0
0
693.2
14.1
14.9
512.8
14.9
Removal efficiency
Minimum
46.7
54.9
76.4
49.3
10.1
69.9
NA
18.3
98.7
54.3
99.7
NA
a
Maximum Average
100
87.4
100
100
10.1
97.4
100
75.2
100
100
100
100
82.2
76.7
85.9
89.9
10.1
91.3
NA
53.1
99.8
77.2
99.9
NA
aBased on bottoms and feed concentration and flow data.
one sampling data point available.
Source: Adapted from GSRI, Reference No. 15.
7-114
-------
column midpoint, whereas waste feed was introduced at the top of the column.
Thus, the overhead product composition is essentially limited by the
equilibrium vapor concentration of the feed and, therefore, independent of
reflux. However, if the reflux was more concentrated in VOCs than the
wastewater at the column midpoint, removal efficiency would decrease with
addition of reflux, particularly for water soluble compounds which would be
preferentially returned to the column.
In summary, recovery of VOCs averaged 85 percent with steam flows of 18
to 24 percent of the feed. Recovery was limited due to column design as
indicated by the insensitivity of recovery rates to steam consumption and
reflux. With the exception of chloral (low removal efficiency), the relative
removal efficiencies of different compounds could not be differentiated due to
uncertainties in the feed composition and fugitive emissions.
GSRI also performed sampling on a second waste (Stream 110) which was
from an oxyclor operation manufacturing ethylene dichloride. The VOCs
(1,999 mg/L) consisted almost entirely of ethylene dichloride (95.7 percent).
Table 7.3.10 summarizes flow rates and ethylene dichloride concentrations for
feed, overhead, and bottoms streams along with steam and reflux flow
measurements. Only runs for which all data were collected were included in
Table 7.3.10.
Overall material balances for this trial were good, but VOC losses from
the overhead product occurred. As a result, removal efficiencies presented in
Table 7.3.10 are based on bottoms and feed composition and flow data. Removal
efficiency correlated positively with overhead to feed ratio (i.e., increased
steam as a percentage of feed rate) and negatively with increased feed rate
(i.e., lower contact time). An "overhead to feed ratio of 5 to 6 percent
(steam to feed ratio of 21.5 percent) or more was necessary to achieve greater
than 99 percent removal efficiency at feed rates up to 400 mL/min. Further
increases in steam or reflux showed little improvement suggesting that
equilibrium and mass transfer limitations restricted further volatile removal.
7-115
-------
TABLE 7.3.10. SUMMARY OF PROCESSING DATA FOR BENCH-SCALE STEAM
STRIPPING OF ETHYLENE DICHLORIDE
Run
Nos.
1
12
28
31
32
36
37
39
41
Flow rates (mL/min)
Feed
395
245
390
260
290
350
225
240
250
Overhead
11.4
55.0
21.7
15.7
94
18.3
12.0
14.1
42.3
Bottoms
410
290
460
300
350
400
279
309
285
Reflux
0
0
0
0
0
0
0
0
33.3
Steam
69
87
84 .
56
65
80
55
56
77
Ethylene dichloride (mg/L)
Feed
1,716
10,734
5,640
5,W9
5,762
1,463
1,247
1,426
1,457
Overhead
8,835
16,278
19,629
16,409
18,100
6,587
6,838
6,030
4,925
Bottoms
65
0.3
43
3.7
5.7
6.0
9.2
7.0
6.1
Removal
efficiency
96
99
99
99
99
99
99
99
99
.1
.9+
.1
.9+
.9
.5
.1
.4
.5
Steam/Feed
ratio (%)
17.5
35.5
21.5
21.5
22.4
22.9
24.4
23.3
30.8
Source; Adapted from GSRI, Reference No. 15.
-------
Except for a single optimal operating condition (Run No. 12; low feed
rate, high steam addition), a residual solvent concentration below 2 ppm was
not achieved. However, with high steam flow and addition of a rectifying
section, it is likely that this level could be reached while still maintaining
a low volume, separable overhead product.
Other parameters recorded during the trials include: ethylene dichloride
range of 1,247 to 10,734 mg/L in the feed stream with 96.1 to 99.9+ percent
removal; chemical oxygen demand reduction of 66.5 percent from 464 mg/L; total
organic carbon reduced by 83.4 percent from 790 mg/L; and total oxygen demand
reduced by 95.7 percent from 5,190 mg/L.
Bench-Scale Ground Water Stripping Feasibility Study—
Several wastewater treatment technologies, including steam and air
stripping, were evaluated in bench-scale treatability studies to select the
best combination of unit processes to treat contaminated ground water. A
continuous—flow packed column was used for the testing with air and steam
flowing countercurrent to the ground water. The column was 3-1/8 inches in
diameter and 48 inches long with 26 inches of packing using 7 mm glass Raschig
16
rings.
Test conditions and TOO removals for the steam stripping study are given
in Table 7.3.11. Runs were performed with and without lime pretreatment and
activated carbon post-treatment, as noted in the table. The stripping removal
rates for specific volatile and extractable organics are detailed in
Table 7.3.12. All volatile organics were present in the raw water above 2 ppm
and were included in the analysis.
At steam to feed ratios ranging from 31 to 56 percent by weight and air
to feed ratios of 6 to 56 percent (i.e., Runs 1 through 5), total VOC removal
efficiency consistently exceeded 99 percent. On average, total VOCs in these
runs were reduced by 99.87 percent from 771 to 1 mg/L. The lowest removal
rate reported for an individual compound was 98.6 percent for
1,1,1-trichioroethane in Run 4. Extractable organics were reduced by
94 percent from 2.6 to 0.2 mg/L. The data in Table 7.3.11 suggest that
removal of organics was essentially complete at these experimental conditions
(Runs I through 5) since removal rate showed little response to increased air,
steam, or wastewater flow rate.
7-317
-------
TABLE 7.3.11. TEST CONDITIONS AND TOG REMOVALS FOR BENCH-SCALE
STEAM STRIPPING WITH/WITHOUT ACTIVATED CARBON
Hater flow
(L/min)
Sample3
Run 1
Run 2
Run 3
Run 4
Run 5
Run 6
Run 7
Before
steam
0.13
0.13
0.045
0.045
0.045
0.08
0.35
After
steam
0.17
0.17
0.07
0.07
0.07
0.095
0.37
Air
L/8
o.ii
0.11
0.11
0.23
0.35
0.47
0.47
flow
Cfitt
0.25
0.25
0.25
0.50
0.75
1.0
1.0
Final
t empe rature
(°c)
Water
85
86
83
79
74
69
46
Air
35
35
83
79
73
42
22
TOC (mg/L)b
Air and Steam stripping
steam plus activated
stripping carbonc
1,500
1,700
900
900
800
1,615
3,633
-
870
400
500
400
-
-
alnitial TOC concentration was 4,000 rag/L.
Test conditions for Run 1-5: air temperature ** 24°C
water temperature = 23°C
pH » 6.0
Test conditions for Run 6 & 7: air temperature - 22°C
water temperature = 22°C
pH = 6.0
Raw sample was treated with lime to pH 20.0 for metals removal; then
the pH was adjusted to 6.5 with sulfuric acid.
TOC measurement is for diluted effluent (increased flow due to
condensate).
C50,000 mg carbon/liter. Use of, activated carbon alone reduced
TOC to 2,240 mg/L. '
Source: Adapted from Reference No. 16.
7-118
-------
TABLE 7.3.12. RESULTS OF BENCH-SCALE STEAM STRIPPING RUNS PERFORMED WITH/WITHOUT LIME PRETREATMENT
Concentrations in ground water Concentration for ground water with
(no pretreatment) - ug/L lime pretreatwent - gg/L
Feed to Stripper
Compound Raw water Run i Run 2 Run 3 lun 4 Run 5 Stripper effluent
Volatile organic compounds
Tetrahydrofuran 22,000 a a
1,1,1-Trichloroethane 150,000 a a
Benzene 68,750 a a
Trichloroethylene 338,000 a a
Methyl isobutyl ketone 76,400 a a
Xylcne a a a
Toluene 92,000 126 a
Ethylbenzene 23,500 a a
Extractable organic compounds
1,4— Dichlorobenzene 35 - 17
l,2-0ichlorobenzene 5 - a
Haphthalene 51 - <1
2-Chlorophenol 540 - 40
2-Hitrophenol 15-6
Phenol 370 - 20
2,4-Dimethylphenol 20 - a
o-Creaol 80 - 25
at-Cresol 220 - a
Benzoic acid 1,230 - a
Pentachlorophenol 40 - 40
a a a a a
150 2,135 a 123,000 123,000
1a 51,000 18,250
l,640b
a 330,000 100,000
a a a 78,000 58,600
a a a 517 80
a a 53 98,400 51,600
a 992 a 1,636 1,636
- - 930 930
260 260
- 99
- 75 75
- - "ISO 150
- - 70 70
- 80 80
200 200
- - - a a
- - 90 90
- 40 40
Run 6
a
2,830
25
213
1,300
a
975
a
a
I1
-------
During these steam stripping runs the packing materials became heavily
coated with oxidized iron and manganese. It was decided to treat the raw
water with lime to increase the pH to 10.0 to precipitate heavy metals. The
pH was then adjusted to 6.5 with sulfuric acid prior to steam stripping. Lime
treatment was effective in reducing fouling of the packing by reducing metal
concentration in the feed from 442 mg/L to less than 2 ppm. However, lime
treatment did not significantly reduce the concentrations of volatile organics
as shown in Table 7.3.12.
Runs 6 and 7 were performed with pretreated wastewater at steam-to-feed
ratios of 19 and 6 percent and air-to-feed ratios of 42 to 10 percent by
weight, respectively. Total VOC removal efficiency was 98.5 percent for Run 6
and only 41.2 percent for Run 7.
Comparison of these results with those obtained for Runs 1 through 5
suggest that steam stripping is significantly more effective than air
stripping when equal quantities of stripping agent are applied. Furthermore,
stripping efficiency is enhanced as the ratio of stripping medium to feed and
contact time increase, as expected.
The last phase of the steam stripping study involved evaluating the
effectiveness of activated carbon adsorption in treating atmospheric emissions
from the stripper. During Run 6, portions of the off-gas were pulled through
an activated carbon bed and the adsorbed volatile organics were analyzed by
gas chromatograph purge-and-trap procedures. The results, which clearly
indicate the effectiveness of this treatment method, are summarized in
Table 7.3.13. The volatile organic load applied to the carbon and the
corresponding amount of volatile organics removed per weight of carbon are
shown. VOC removal ranged from 100 percent at a carbon loading of 1.9 ug/mg
to 95 percent at 5.7 yg/mg loading.
7.3.3 Cost of Steam Stripping
Steam stripping costs are highly site specific. Large-scale units used
for stripping wastewater streams are custom designed for specific
applications. As a result, equipment manufacturers are reluctant to supply
cost estimates without detailed waste characteristic and volume data for
specific applications.
7-120
-------
TABLE 7.3.13. TEST RESULTS FROM CARBON ADSORPTION OF OFF-GAS
FROM STEAM STRIPPING - RUN 6
Volatile organic compound
1,1, 1-Trichloroethane
Benzene
Trichloroethylene
Methyl isobutyl ketone
Toluene
Ethyl benzene
Total volatile organics
Volatile Volatile organic
organic loading applied
stripped to carbon3
Ug/min yg/mg
9,570
9,570
9,570
1,457
1,457
7,980
7,980
4,564
4,564
4,564
4,035
4,035
4,035
131
131
131
27,737
27, 737
0.8
1.6
2.4
0.07
0.21
0.13
0.40
0.52
1.05
1.57
0.31
0.62
0.94
0.06
0.13
0.17
1.89
5.69
Volatile
organic adsorbed
Ug/ag
0.8
1.6
2.2
0.07
0.20
0.13
0.37
0.52
1.05
1.51
0.31
0.62
0.94
0.06
0.13
0.17
1.89
5.39
aCarbon used was GAC-40; Carborundum, Niagra Falls, N.Y.
Source: Reference No. 16.
7-121
-------
Cost of these units is basically a function of throughput, initial VOC
concentration and relative volatility of the compounds which are to be
stripped. Flow rate determines the diameter of the column (without reflux),
initial concentration determines the required removal efficiency to meet
effluent treatment standards, and relative volatility determines the ease with
which VOCs can be stripped. Depending on relative volatility, a tradeoff
between column height, reflux, steam rate, operating pressure, and
post-treatment costs can be established.
Treatment cost data presented in the literature were generally not useful
for predicting total waste treatment costs for a range in waste
13
characteristics. RTI presented cost data for the four batches of solvent
waste discussed in Section 7.3.4. However, these data were for specific
wastes treated in an offsite facility and, therefore, were not indicative of
9
general onsite processing costs. Water General presented a detailed design
and treatment cost modeling approach. However, this methodology does not
include an evaluation of post-treatment costs or cost reductions achieved
through solvent recovery. For dilute wastewaters, these costs can effectively
be ignored since they are relatively small and offset one another. However,
these costs can represent significant fractions of total waste processing
costs for more concentrated wastes, and therefore must be taken into
consideration.
Two cost analyses are presented below. The first provides capital and
operating cost equations for steam stripping of dilute wastewater streams. It
is based on a review of actual onsite steam stripping installations performed
a
by JRB Associates. The second analysis, performed by GCA, is appropriate
for developing cost estimates for more concentrated solvent wastes. This
analysis takes into consideration three residual disposal options (wastewater
treatment, use as a fuel, incineration) and discusses the impact of various
waste characteristics and cost centers on overall processing costs.
Steam Stripping Costs for Wastewater Streams—
JRB analyzed cost and design data for 15 industrial steam strippers used
to recover secondary materials or organic priority pollutants from wastewaters
Q
that flowed into secondary biological treatment systems. Steam strippers
used to recover or recycle primary products/raw materials were excluded from
7-122
-------
this analysis since they differ in design from units used to treat
wastewater. In addition, tray towers were chosen for use in the analysis
instead of packed towers since data on these units were more readily available.
Capital and operation and maintenance (O&M) costs were normalized to 1980
dollars using the appropriate Engineering News Record indices. Where
installation costs were not provided, they were assumed to be 50 percent of
capital costs. Capital costs include: stripping columns, feed tanks, feed
preheaters, condensers, decanters, organic phase pumps, bottom pumps, and
existing equipment modifications. O&M costs include: operation and
maintenance labor, maintenance materials, steam, and electricity.
An analysis was performed to determine a mathematical relationship
between capital and O&M costs and significant steam stripper design parameters
such as contaminant volatility, wastewater flows, column diameter, and column
height. The results of this analysis showed that capital costs were best
related to the diameter (D, in inches) and height (H, in feet) of the column,
while O&M costs were best related to the diameter and wastewater flow (Q, in
million gallons/day) as follows:
Capital cost (in million dollars) = 0.246 - 2.88 x 10~4 (D)
+ 1.546 x 10~6 (D2 H)
O&M cost (in million dollars) = 3.68 x 10~3 (D) + 0.809 (Q) - 0.023
Overall, predicted capital costs were within a factor of 3 of reported
costs, O&M cost estimates were within a factor of 5, and cost per gallon of
treated waste was within a factor of 3.7 of actual values. With an annual
capital recovery factor of 0.177, capital costs accounted for an average of
26 percent of total cost per gallon of treated waste. Excluding a single
facility which had a low capacity utilization (largest diameter tower but
lowest flow rate), cost per gallon of treated waste averaged 0.9 cents per
gallon with a range of 0.14 to 20.4 cents per gallon. Costs for four packed
towers, excluded from the analysis, averaged 1.12 cents per gallon. Since
flow rates to these units were only one-fourth of that in the average tray
column, the cost difference may be attributable to economies of scale.
7-123
-------
The above data are applicable to steam stripping costs for continuous
flow columns treating dilute (i.e., less than 1 percent solvent) organic
contaminated aqueous wastes. They do not include waste pretreatment and
bottoms post-treatment costs or net cost benefits derived from solvent
recovery. Overhead products will consist of a solvent-water mixture which
will require further treatment (e.g., distillation, additional stripping) to
recover a valuable product. This additional processing is likely to result in
a small net cost advantage relative to raw material purchasing, and thus its
effect can be ignored. Bottoms post-treatment will be required if solvent
concentrations continue to exceed disposal limitations or if the waste is
still considered to be hazardous due to the presence of other nonvolatile
contaminants. Post-treatment by activated carbon, biological treatment, or
other methods will add roughly 2 cents per gallon to waste treatment costs.
JRB also attempted to determine cost variability as a function of
contaminant volatility. JRB used the design methodology provided by Water
Q
General Corporation to determine variability in column height and
therefore, capital cost, which is required to strip compounds with different
Henry's Law constants. Water General's methodology involves calculation of a
stripping factor which is proportional to Henry's Law constant, the
vapor/liquid flow ratio, and the reciprocal of tower operating pressure.
JRB assumed a steam-to-liquid feed ratio of 10 percent and atmospheric
column operating pressure. Costs were based on stripping to a maximum
residual VOC concentration of 1 ppm. Minimum column diameter was set at
l»0-feet and minimum height at 10 feet to reflect wastewater processing
equipment currently in use.
Table 7.3.14 summarizes the resulting cost data based on the above
assumptions. As shown, cost per gallon of waste treated shows little
variability between compounds with different volatility. However, if JRB'a
column size constraints were removed, the data would show lower treatment
costs for wastes with highly volatile constituents and low flow rates. Also,
operating costs would constitute a higher fraction of total costs since
optimal operating conditions would, in some cases, be represented by higher
steam rates instead of increases in column height.
7-124
-------
TABLE 7.3.14.
STEAM STRIPPING COSTS FOR WASTEWATER STREAMS CONTAINING
CONTAMINANTS OP VARYING HENRY'S LAW CONSTANT
NJ
Ul
Henry ' s Law
constant range
Greater than 10" *
10~2 to 10"3
Less than 10"^
Flow
Rate
Example compound (MGD)
1,1,1-Trichloroethane 1.0
0.10
0.01
Acrylonitrile 1.0
0.10
0.01
Nitrobenzene 1.0
0.10
0.01
Height
(ft)
10,0
10.0
19.6
14.2
18.8
30.9
36.1
42.9
65.2
Diameter
(ft)
6.28
2.11
1.00
6.28
2.11
1.00
6.28
2.11
1.00
Capitol
Cost
($MM)
0.312
0.249
0.247
0.349
0.257
0.249
0.540
0.281
0.257
O&M
Cost
($MM)
1.06
0.151
0.029
1.06
0.151
0.029
1.06
0.151
0.029
Unit cost
U/gal)a
0.36
0.63
2.30
0.36
0.63
2.30
0.37
0.64
2.40
aAssuming 312 operating days/year and an annual capital recovery factor of 0.177.
Source: Adapted from JIB. Reference No. 8.
-------
Steam Stripping Costs for Nonaqueous Wastes—
GCA performed a cost analysis for steam stripping of nonaqueous wastes
using three nominal flow rates (10, 50, and 500 gpm) and three disposal
methods (wastewater treatment, use as a fuel, incineration). The cost of
stripping concentrated wastes (i.e., greater than 1 percent solvent) is highly
dependent on method of disposal. Other major cost variables considered
included capital, installation, maintenance, labor, overhead and utility costs
and value of recovered solvent.
Capital costs for process equipment, tanks, engineering, electrical and
instrumentation were taken from the literature. A contingency of
15 percent and an annualized cost of 17.7 percent of the total were assumed as
summarized in Table 7.3.15. Maintenance costs for stream strippers were based
9
on an EPA estimate of 4.13 percent of annualized capital cost. Labor costs
were assumed to be $14.42/hour including overhead. Labor usage was
assumed to range linerly from 0.5 to 3.0 operators for the flow rates under
consideration with a base case operating time of 2,080 hours/year.
Utility costs were assumed to average $0.04/gallon of recovered solvent.
9
This value is based on the cost of steam ($3.00/million Btu) , electricity
($0.04/KWH)17, and cooling water ($0.25/lrOOO gallons)17 which is
necessary to separate and condense solvents from a solvent-water mixture.
This value will vary depending on the solvent which is to be recovered. For
example, at 1 percent solvent concentration, utility costs will range from
approximately 2 cents (e.g., acetone) to 7 cents (e.g., nitrobenzene) per
gallon of recovered solvent, depending primarily on the solvent boiling point
in the mixture. Utility costs for concentrated solvent waste depend on both
the boiling temperature and heat of vaporization of the solvent, and range
from roughly 3 to 6 cents per gallon of recovered solvent. However, since
utility costs generally represent a small fraction of total treatment costs,
an average value of 4 cents per gallon was assumed.
Three methods of bottoms disposal were used in this analysis. Wastewater
treatment technologies such as adsorption and biological treatment were
assumed to cost 2 cents per gallon. Organic bottoms which could be used as a
fuel substitute were assumed to cost $0.20/gallon and bottoms which required
incineration (e.g., liquid injection) were assumed to cost $2.00/gallon.
7-126
-------
TABLE 7.3.15. COST COMPONENTS FOR ONSITE STEAM STRIPPING SOLVENT RECOVERY:
(EXAMPLE CASE WITH 30% SOLVENT CONTENT AND
BOTTOMS USED AS FUEL)
Nominal feed rate (gpm)
Capital costs
Process equipment
Tanks
Subtotal (1)
Engineering, electrical,
instrumentation (20% of (1))
Subtotal (2)
Contingency (15% of (2»
Subtotal (3)
10
8,000
18,500
26, 500
5,300
31,800
4,800
36,600
50
39,000
18,500
57,500
11,500
69,000
10,400
79,400
500
72,000
74,000
146, 000
29,200
175,200
26, 300
201,500
Annualized capital cost (17.7% of
Subtotal (3))
Operating and maintenance costsa
Maintenance (4.13% of annualized
capital cost)
Labor ($30,000/man-year
including overhead)
Utility costs (4^/gallon of
recovered solvent)
Solvent recovery benefit cost
($2/gallon of recovered solvent)
Disposal cost ($0.20/gallon)
Net O&M cost
Total capital and O&M cost
Cost/gallon of waste treated
Threshold cost of recovered solvent
(I/gallon)
6,478
268
15,000
237
14,054
580
22,800
1,186
4.24
1.83
35,666
1,473
90,000
11,856
(11,850) (59,300) (592,800)
3,391 16.953 169,530
7.046 (17.781) (319.941)
13.524 (3.727) (284,275)
0.65 (0.04) (0.27)
1.00
aNumbers in parenthesis represent revenues.
7-127
-------
Finally, recovered solvent was assumed to have a value of $2.00/gaiion
for the purposes of calculating overall unit cost of waste treated. Although
many solvents have higher purchase prices, this value was used since many
recovered materials will be in the form of less valuable mixtures or require
additional treatment thereby increasing treatment costs.
Table 7.3.15 provides an example cost analysis for treating an organic
waste containing 30 percent solvent and using the residual bottoms product as
a fuel. Table 7.3.16 summarizes the results of the cost analysis for wastes
ranging from 1 to 70 percent solvent for the three bottoms disposal
scenarios. Cost figures are presented on the basis of cost per gallon of
waste treated and threshold cost per gallon of solvent recovered.
As shown in Table 7.3.16, costs increase dramatically with increasing
bottoms disposal cost, particularly for high flow rate units. For example, at
10 gpm and 10 percent solvent content, disposal costs for wastewater treatment
account for only 4 percent of total treatment costs. This fraction jumps to
16 and 66 percent for use as a fuel and incineration, respectively. At a feed
rate of 500 gpm, disposal costs became even more significant accounting for
24, 61, and 93 percent of total costs as more expensive disposal methods are
used. Although these costs are lower for wastes with higher percentages of
recoverable solvent, disposal costs will remain a significant cost component
for high-volume units with nonaqueous bottoms products.
Similarly, as capacity and percent recoverable solvent increase, total
treatment cost becomes increasingly sensitive to the value of recovered
solvent. For example, at 10 gpm with 10 percent solvent content in the waste,
total value of recovered solvent is 16 percent of processing costs (assuming
$2.00 value per gallon of recovered solvent and bottoms used as fuel). This
percentage jumps to 113 percent as solvent content in the feed increases to
70 percent, and increases further to 554 percent of processing costs when
capacity is increased to 500 gpm. Thus, unit value of recovered solvent has
an increasingly significant impact on processing economics as throughput and
percent recoverable solvent increase.
Utility costs show a similar relationship to throughput and recoverable
solvent content. For wastes with less than 10 percent solvent content,
utility costs are only a few percent or less of total costs. However, at" high
7-128
-------
TABLE 7.3.16. STEAM STRIPPING COST ESTIMATES AS A FUNCTION OF
THROUGHPUT, SOLVENT CONTENT AND DISPOSAL METHOD
•xl
t-«
NJ
VO
Bottoms
disposal
method
Wastewater
treatment
Use as fuel
Incineration
Solvent
content in
waste (%)
1
5
10
10
30
50
70
10
30
50
70
Cost ($)/gal of waste treated8
10 GPM
1.05
0.98
0.89
1.05
0.65
0.24
(0.17)
2.00b
2.00b
1.36
0.61
50 GPM
0.36
0.29
0.21
0.3"
(0.04)
(0.45)
(0.85)
2.00b
1.43
0.68
(0.07)
500 GPM
0.12
0.06
(0.03)
0.13
(0.27)
(0.68)
(1.09)
1.93
1.19
0.44
(0.31)
'Cost ($)/gal of solvent recovered*
10 GPM
108.00
21.79
11.01
13.11
4.24
2.46
1.70
b
b
4.83
2.88
50 GPM
38.58
7.95
4.08
5.90
1.83
1.02
0.67
b
6.98
3.39
1.85
500 GPM
14.59
3.15
1.68
3.39
1.00
0.52
0.31
22.36
6.15
2.89
1.49
aNumbers in parenthesis represent revenues.
Cost of treatment via steam stripping exceeds incineration cost of raw waste
(|2.00/gallon), therefore, incineration represents the lower cost alternative.
-------
flow rate and high solvent content, these costs become significant. For
example, at 500 gpm and 70 percent solvent in the feed, utilities account for
11 percent of total processing costs when bottoms are used as fuel
(2.6 percent when bottoms are incinerated).
In contrast to disposal and utility costs, capital and labor represent
the primary costs of processing low volume, dilute wastes. Capacity
utilization is more critical for low volume units since capital costs are
distributed over the total volume of waste processed. For example, increasing
processing time from 8 to 16 hours per day reduces processing costs by
16 cents per gallon for 10 gpm units versus only 1 cent per gallon for 500 gpm
strippers.
7.3.4 Overall Status of Process
7.3.4.1 Availability--
Steam stripping is a commonly applied waste treatment technology for the
separation of low solubility solvents from water or low volatility organics
(e.g., oil). The SPA has identified 27 industrial steam stripping wastewater
treatment units, 11 units used by the pesticides industry, and 8 steam
strippers (packed towers) used by a pharmaceutical manufacturer. GCA*s
analysis of the commercial solvent recycling industry showed 25 percent of the
reclaimers using steam distillation (Table 4.1.1).
7.3.4.2 Application—
Steam stripping is commonly used as a pretreatment method, particularly
when applied to concentrated solvent wastes. Although in many cases it can be
used to reduce solvent concentrations to levels which permit direct discharge,
it is often more cost-effective to use other treatment methods for final
bottoms processing. Steam stripping of wastewater streams is typically
followed by biological treatment or adsorption systems for final effluent
polishing. Stripped organic waste bottoms can be decanted and used as fuel
provided that chlorine content has been sufficiently reduced in the stripper.
7-130
-------
7.3.4.3 Environmental Impact—
Post-treatment of both the overhead and bottoms streams is usually
required, although in certain instances the solvent concentration in the
bottoms stream may be below proposed standards. Overhead products undergo
liquid-liquid separation, typically through decanting. However, the organic
solvent may require drying and the aqueous phase may require further treatment
to remove dissolved organics. Air emissions from the column vent can be
13
significant, and should, at least, be monitored.
7.3.4.4 Advantages and Limitations—
Steam stripping is preferrable to other physical separation technologies
in the following instances:
• For treating wastes which contain high solids or polymerizable
materials which would otherwise foul heat transfer surfaces;
• For treating wastes which contain constituents that form low boiling
azeotropes with water, particularly those which require low
processing temperatures due to thermal degradation; and
• For treating wastes to low residual solvent content, particularly
when the bottoms product would be rendered unpumpable in the absence
of water.
Steam stripping is not well suited to treating wastes in which either the
overhead or bottoms are difficult to separate from water. Thus, it is better
utilized for separating solvents which decant readily and have low
solubilities in water (e.g., halogenated organics) and less applicable to
treating water soluble wastes; e.g., alcohols.
7-131
-------
REFERENCES
1. Allen, C. C., and B. L. Blaney. Techniques for Treating Hazardous Waste
to Remove Volatile Organic Constituents. Research Triangle Institute
performed for U.S. EPA HWERL. EPA-600/D-85-127, PB85-218782/REB. March
1985.
2. EHerbe, R. W., Steam Distillation/Stripping. The Rust Engineering
Company. In: Handbook of Separation Techniques for Chemical Engineers.
McGraw-Hill Book Company, New York, N.Y. 1979.
3, Weast, R. C., Handbook of Chemistry and Physics. 65th Edition, CRC
Press, Cleveland, OH. 1984-1985.
4. Perry, R. H., Chemical Engineers' Handbook, 6th Edition, McGraw-Hill
Book Company, New York, N.Y. 1984.
5. Kirk-Othmer Encyclopedia of Chemical Technology. John Wiley & Sons,
New York, N.Y., 3rd Edition. 1978.
6. U.S. EPA Office of Solid Waste. Background Document for Solvents to
Support 40 CFR Part 268 Land Disposal Restrictions. Volume II. Analysis
of Treatment and Recycling Technologies for Solvents and Determination of
Best Available Demonstrated Technology. U.S. EPA Public Docket. January
1986.
7. USEPA Office of Solid Waste. Analysis of Organic Chemicals, Plastics and
Synthetic Fibers (OCPSF) Industries Data Base. U.S. EPA Public Docket.
January 1986.
8. JRB Associates and Science Applications International Corporation.
Costing Documentation and Notice of New Information Report. Draft Report
prepared for U.S. EPA. June 1985.
9. Shukla, H. M., and R. E. Hicks. Water General Corporation, Waltham, MA.
Process Design Manual for Stripping of Organics. Prepared for U.S. EPA
IERL. EPA-600/2-84-139. August 1984.
10. Michigan Department of Commerce. Hazardous Waste Management in the Great
Lakes: Opportunities for Economic Development and Resource Recovery.
September 1982.
7-132
-------
11. M. Arienti, et al. GCA Technology Division, Inc. Technical Assessment
of Treatment Alternatives for Wastes Containing Halogenated Organics*
Draft Report prepared for U.S. EPA Office of Solid Waste under EPA
Contract No. 68-01-6871. October 1984.
12. O'Brien, S., et al. GCA Technology Division, Inc. Performance
Evaluation of Existing Treatment Systems: Site-Specific Sampling Report
for Environmental Processing Services, Dayton, OH. Draft Report prepared
for U.S. EPA HWERL under EPA Contract No. 68-03-3243. July 1986.
13. Allen, C. C., Research Triangle Institute, and Simpson, S., and 6. Brant
of Associated Technologies, Inc. Field Evaluation of Hazardous Waste
Pretreatment as an Air Pollution Control Technique. Prepared for U.S.
EPA HWERL under EPA Contract No. 68-02-3992. January 1986.
14. Jett, G. M., Development Document for Expanded Best Practicable Control
Technology, Best Conventional Control Technology, Best Available Control
Technology in the Pesticides Chemicals Industry. Effluents Guidelines
Division, U.S. EPA. EPA-440/l-82-079b. November 1982.
15. Coco, J. H., et al. Gulf South Research Institute, New Orleans, LA.
Development of Treatment and Control Technology for Refractory
Petrochemical Wastes. U.S. EPA, Ada, OK. EPA-600/2-79-080. April 1979.
16. Stover, E. L.t and D. F. Kincannon. Contaminated Ground Water
Treatability - A Case Study. Journal of The American Water Works
Association. June 1983.
17. Horsak, R. D., et al. Pace Company Consultants and Engineers, Inc.
Solvent Recovery in the United States: 1980-1990. Houston, TX.
Prepared for Harding Lawson Associates. January 1983.
7-133
-------
7.4 AIR STRIPPING
7.4.1 Process Description
Air stripping is one of several processes available that effectively
removes certain volatile organic compounds (VOCs) from aqueous media. Air
stripping operations employ gas-liquid contacting systems to enhance the
transfer of VOCs from the liquid phase to the gas phase. Various types of
aeration devices have been designed to optimize transfer and they fall into
two general categories; (l) injection of water into air, and (2) injection of
air into water. In both systems, mechanical energy creates air-water
interfaces across which mass transfer of the contaminant can occur. Examples
of water-into-air systems are spray systems into the open air, spray towers,
and tray and packed towers. Air—into-water systems typically used are
diffused or mechanical aeration.
Each of these units has certain process and economic advantages.
However, tray and packed tower aeration will be the focus of this study since
they are commonly used and design criteria for tray and packed towers are
available to meet desired contaminant removal efficiencies. Other systems
have less structured design requirements and are used primarily to supply
oxygen for aerobic biotreatments.
Diffused Aeration—In a diffused aeration system, air is injected into the
water through a sparging device or through porous diffusers which produce a
multitude of fine bubbles. As the bubbles rise, mass transfer occurs across
the water—air interface until the bubble either leaves the water column or
becomes saturated with the contaminant. The rate of mass transfer and its
extent can be enhanced by increasing the depth of the tank, improving bubble
dispersion, decreasing the bubble size, and increasing the volumetric air to
water ratio. Increasing the depth of the tank may not increase the mass
transfer if the air bubble reaches saturation before it exits the liquid
surface.
Generally, diffused aeration systems are operated as continuously* stirred
tank reactors, and are thus inherently less efficient that a counter-flow
process. The system is widely used for gas adsorption, principally adsorption
7-134
-------
of oxygen for biological oxidation. Diffused aeration can also be used for
removing highly volatile compounds and is the basis of the widely used
purge-and-trap analytical method for concentrating organic compounds. In
water treatment applications, removal of up to 90 percent has been reported
for several VOCs. Removal of VOCs during aeration of activated sludge process
has also been reported.
Mechanical Aeration—Air can also be introduced into water by mechanical
mixing. In this system the contents of the tank or impoundment are
circulated providing continuous contact between the atmospheric air and
water. Principal design parameters include mixing intensity, the type of
mixer, the depth of the tank or impoundment, and the residence time if the
system is continuous. The system exhibits relatively low rates of mass
transfer, but has the advantage of mechanical simplicity and negligible liquid
head loss. The system has been used for removal of ammonia and VOCs with
moderate success, but is normally used to supply oxygen for biological
activity.
Sprays and Spray Towers—High rates of mass transfer can be achieved by
pumping water through spray nozzles that break the liquid stream into fine
droplets. Sprays can be used to inject water into the open air or into a
tower. Principal design parameters for sprays and spray towers include
selection of the type of nozzle, nozzle size, allowable nozzle pressure drop,
and the process configuration. Because of back mixing in the tower and in the
air, true countercurrent transfer does not occur and mass transfer rates are
usually modest. As a consequence, removal efficiencies greater than
90 percent may not be economically feasible.
Tray and Packed Towers—The design principles for air stripping in packed and
tray towers have been extensively examined in the chemical engineering
literature over the past 30 to 40 years. Most chemical engineering
applications generally involve design of systems to treat concentrated
solutions. However, the general design procedures developed in the chemical
processing industry have recently been extended to cover the full range of
concentrations, including the case of dilute solutions typically encountered
in water treatment applications.
7-135
-------
Tray or packed tower air stripping systems similar to those depicted in
Figure 7.4.1. are more effective for the removal of organics from water than
the other systems described above. The stripping medium, air, is introduced
near the base of the tower, while the liquid feed is introduced near the top.
As the air rises and comes into contact with the falling water, the volatile
components are transferred from the organic rich water to the air. The
organic laden air is carried out the top as overhead, and the stripped water
exits the bottom of the tower.
In tray towers, the tower is filled with regularly spaced trays or plates
allowing for staged contact between the two phases. The vapor passes through
openings in each tray and contacts the liquid flowing across the tray. A
quantity of liquid is retained on each tray by a weir. To reach the next
stage, the liquid flows over the weir through a downcomer which provides
sufficient volume and enough residence time for the liquid to be freed of
entrained vapor before entering the next tray. The overall effect is a
multiple countercurrent contact of air and water, although each tray is
characterized by a crossflow of the two. Each tray of the tower is considered
a stage.
Packed towers are simple when compared with tray towers. A typical tower
consists of a cylindrical shell containing a support tray for the packing
material and a liquid distributor designed to provide effective irrigation of
the packing. The water is distributed over, and flows down through, the
packed bed, exposing a large surface area for transfer into the air which
enters at the bottom of the tower. Commercially available packing materials
come in a variety of shapes and sizes. Host packings are made of either
ceramic, metal, or plastic. Depending on the types and size of packing they
may be either randomly dumped or carefully stacked in the column.
7.4.1.1 Pretreatment Requirements—
Pretreatment requirements for air stripping operations are usually
minimal and basically related to contact equipment limitations for solids.
Solid removal by filtration, sedimentation, or other means may be required to
avoid plugging of spray nozzles and fouling of packing and tray towers. A
limiting suspended solids content of 2 percent is called out in Reference 2.
7-136
-------
FEED
•sj
OVERHEAD
VAPOR
WATER
PACKED TOWER
OVERHEAD
VAPOR
FEED
PACKING
SUPPORT
GRID
AIR OR
STEAM
CLEANED
WATER
AIR OR
STEAM
TRAY TOWER
Figure 7,4,1. Air stripping towers.
-------
The removal of oil and grease and dissolved metals which may oxidize and
precipitate could also be necessary to avoid column plugging. A schematic of
an air stripping system, with a variety of pretreatment and post-treatment
options, is shown in Figure 7.4.2.
7.4.1.2 Operating Parameters--
Organic contaminants found in water and wastewater exhibit a wide range
of water solubilities and polarities. Each compound has a different tendency
to pass from the aqueous phase into the air. For aqueous mixtures containing
low concentrations of VOC, the distribution of a VOC between air and water
under equilibrium conditions can often be expressed by Henry's Law. It states
that at equilibrium the partial pressure of a contaminant (y) in air is
proportional to its concentration in water as shown below. -
Py ~ HyXy
wheres Py = Partial pressure of compound y (atm)
Hy = Henry's Law constant (atm-m^/mole)
Xy » Molar concentration (mole/m^)
-3
Henry's Law is valid for mole fractions less than 10 , although the
exact limit of this law depends upon the molecular interactions of the
compounds in water. In general, the larger the Henry's Law constant the more
easily a compound can be stripped from water by aeration methods.
As a general rule air stripping appears practical for compounds having a
~"3 3
Henry's Law constant of 10 atm-m /mole or higher. Many solvents and
other low molecular weight organic compounds have Henry's Law constants in
this range. Henry's Law constants, as estimated by EPA for all organic
compounds of interest, can be found in Appendix A. Table 7.4.1 lists all
those compounds of interest whose estimated Henry's Law constants are in
excess of 10 .
The air stripping rate (rate of mass transfer) can be increased by
increasing the magnitude of either the overall mass transfer coefficient or
the gas-liquid interfacial area. Both parameters are influenced by the
7-138
-------
AIR
WASTEWATER
Co
vo
LIME
i
PH
ADJUSTMENT
10
LIME AND
METAL HYDROXIDES
CARBON
ADSORBER
WATER
RECHARGE
PUMP
CARBON
ADSORBER
(POLISHING STEP)
AIR
CLEANED
WATER
Figure 7.4.2. Air stripping schematic.
-------
TABLE 7.4.1. COMPOUNDS OF INTEREST WITH HENRY'S LAW CONSTANT
GREATER THAN 1 X 10~3 atm-m3/Biol
COMPOUNDS HENRY'S LAW CONSTANT
Priority Solvents
Carbon Disulfide 1.6 x 10~2
Carbon Tetrachloride 2.1 x 10~2
Chlorobenzene 3.9 x 10~3
1,2 Dichlorobenzene 1.9 x 10~3
Ethylbenzene 8.7 x 10~3
Methylene Chloride 3.2 x 10~3
Tetrachlorethylene 2.9 x 10""2
Toluene 6.7 x 10~3
1,1,1 Trichloroethane 3.0 x 10" 2
Trichloroethylene 1.0 x 10~2
Triehloromonofluoromethane 5.8 x 10~2
1,1,2 Trichloro -1,2,2-Trifluoroethane 4.8 x 10~l
Xylene 3.0 x 10~3
Other Organics
Benzene
Chloroform
Cyclohexane
Dichloro Difluoro Methane
1, 1-Dichloroethylene
1 , 2-Dichloroethylene
1 , 2-Dichloropropane
1,3 Dichloropropanes
Ethylene Dichloride
Furan
Hexac hloroe thane
2-Nitropropane
Chloroacetaldehyde
Cumene
Me thac ry loni tri le
1-Methylbutadiene
Methyl Bromide
Methyl Chloride
5.5 x 10~3
3.4 x 10~3
1.8 x 10~l
4.0 x 10""1
1.9 x 10~!
5.3 x 10~3
2.3 x 10~3
1.3 x 10"3
4.3 x 10~3
5.7 x 10~3
2.5 x 10~3
I.2 x lO"1
1.0 x 10~3
1.5 x 10~2
3.9 x 10"1
4.2 x 10""2
5.3 x 10~3
4.0 x 10~2
Source: References 2-4.
7-140
-------
specific engineering design, and the type of aeration system. It is important
to note, however, that if mass transfer is controlled by the system's liquid
phase resistance, its magnitude is relatively insensitive to changes in gas
flow rate. The mass transfer of most compounds amenable to air stripping is
usually liquid phase controlled stripping rate mass transfer can be improved
most readily by increasing the interfacial area using packed or tray tower air
stripping systems.
The optimum design will be that which will achieve effluent concentration
goals with the lowest total cost (capital plus operating costs), and is best
determined by evaluating a range of values for key parameters. For aqueous
wastes containing multiple contaminants, final design criteria will be based
on the compound whose effluent standard is most difficult to achieve.
In a preliminary design procedure the following information will
generally be known:
* waste stream (liquid) flow rate
* compound(s) to be treated
* desired removal efficiency
Using this information along with thermodynamic data, etc., a design
determination of the following can be made for tray towers, using a standard
chemical engineering text such as Berry (Reference 1).
1. The number of stages theoretically necessary for the required
separation,
2. The stage efficiency of the trays relating the theoretical plate to
the actual trays,
3. The diameter of the tower necessary to avoid flooding or excessive
ent ra inment, and
4. The pressure drop across the tower.
For packed towers, efficiency is defined as the ability of a system to
achieve effective mass transfer between the gas phase and liquid phase. It is
inversely related to the height of packing that is equivalent to one transfer
7-141
-------
unit (HTU). Apart from geometrical consideration related to the shape and
arrangement of the packing, the efficiency is affected by viscosity, liquid
flow, and Henry's Law constant in more or less the same manner that tray
towers are affected.
The design of a packed tower consists of the following steps:
1. Selection of the packing type and size,
2. Calculation of the total height of packing required,
3. Estimation of tower diameter to avoid flooding, and
4. Estimation of the pressure drop.
This procedure involves incorporating several equations and correlations in an
iterative fashion to arrive at the optimum system size.
The choice between tray and packed towers is usually made on the basis of
least cost of removal. However, there are distinct advantages and
disadvantages associated with each type of system. Peters and Timmerhaus
present the following comparative merits of both systems:
1. If the operation involves liquids that contain dispersed or
suspended solids, use of a tray tower is preferred because the trays
are more accessible to cleaning.
2. Random packed towers are seldom designed with diameters larger than
4 feet, and tray towers have diameters that are seldom less than
2 feet.
3. Packed towers are cheaper and easier to construct than tray towers
if highly corrosive fluids must be handled. It is easier and
cheaper to replace packing periodically than trays.
4. Packed towers are usually preferred for liquids that have a tendency
to foam.
5. Liquid holdup is considerably less in packed towers.
6, The pressure drop through packed towers may be less than the
pressure drop through tray towers designed for the same duty. This,
plus the fact that the packing serves to lessen the possibility of
tower wall collapse, makes packed towers particularly desirable for
vacuum operations.
7. Tray towers can operate efficiently over a wider range of liquid
flow rates that can packed towers.
7-142
-------
Other considerations, while not directly related to performance, may have a
bearing on the selection. These include the following:
8. Design information for tray towers is generally more readily
available and more reliable than that for packed towers.
9. Because of liquid dispersion difficulties in packed towers, the
design of tray towers requires less of a safety margin when the
liquid to gas ratio is low.
10. Reliable design data for packed towers must often be obtained from
experiment.
11. The total weight of a dry tray tower is usually less than that of a
packed tower designed for the same duty. However, if liquid holdup
during operation is taken into account, then both types of towers
have about the same weight.
7.4.1.3 Post-Treatment Requirements—
Post-treatment requirements should be assessed for the treated aqueous
stream to ensure that effluent standards have been achieved. The air stream
containing the stripped organics should also be examined to ensure that
emissions to air are at acceptable levels. Depending upon the concentration
of the contaminants in the feed stream and the required air flow rate, the use
of a secondary treatment technology (e.g., carbon adsorption, condensation, or
direct fume incineration) may be needed to remove the volatile organics from
the stripping stream. Collected material may then be recovered or incinerated.
7.4.1.4 Treatment Combinations—
In a wastewater treatment train, stripping is typically the first process
step that separates dissolved substances. It follows clarification or
filtration steps that are used for removal of suspended solids, and may
preceed polishing steps such as biological systems or carbon or resin
adsorption. Figure 7.4.2 shows an air stripping treatment train with a
variety of pre- and post-treatment options.
The presence of metals such as iron and manganese in the wastewater can
cause the precipitation of iron and manganese oxide on the packing material
thus reducing the column's operating efficiency. As shown in Figure 7.4.2, a
method of pretreating the wastewater involves raising the pH of the feed water
7-143
-------
to about 10. The metal salts and excess lime used to convert the soluble
metals to an insoluble form can then be removed from suspension by gravity
settling or filtration. After the metals are removed, adjustment of the
water's pH to about 6 or 7 is necessary to reduce the wastewater's corrosive
characteristics. The metal containing sludge produced during this
precipitation must be removed and treated. Determining the quantity of sludge
and its settling, thickening, and dewatering characteristics is necessary for
proper design of the treatment train.
Once the water has passed through the stripping column, it may enter into
a polishing step to remove residual organics* The most common polishing step
is one in which the water is passed through an activated carbon column where
adsorption of the remaining contaminants on to the carbon prepares the water
for discharge or reuse.
On the air side of the system a preheater may be put on line to warm the
air when the air temperature drops below desired levels. Also, depending on
the wastes that the system is designed to remove, preheating the air stream
(or, more commonly, the aqueous waste stream) may be necessary to affect
removal of some of the less volatile compounds in the liquid stream.
As the air enters the column, water will also evaporate from the liquid
stream and thus cool the column. To prevent this, an air saturator or
humidifier can be added to saturate the air so that the column does not cool
due to evaporative losses of water.
As noted, the air leaving the top of the tower containing organics is
often allowed to vent to the atmosphere, with the dilute concentrations of
organics usually treated by this process posing little threat to air quality.
However, if this process were used on a more concentrated waste, some
treatment of the exiting air might be required. In this case the air can be
vented through a fume incinerator or through a carbon adsorber for removal of
the volatiles and then vented to the air.
Apart from the treatment combinations normally used when air stripping
represents a primary treatment process, air stripping also occurs in many
biological treatment systems. Information available in EPA1s Industry Studies
Data Base indicates that the total concentration of solvents entering surface
impoundments average about 3000 rag/liter and most of these are aerated to
2
assist biological treatment. An appreciable but unknown quantity of these
materials are lost through stripping.
7-144
-------
7.4.2 Demonstrated Performance
Few data are available on full scale air stripping operations treating
industrial effluents. According to information available from EPA's Office of
Water there are very few air stripping units dedicated to VOC removal from
industrial wastewater1s. A 1980 questionnaire sent out to 981 facilities
revealed that only five facilities claimed to use air stripping equipment in
their wastewater treatment systems. The information and data on these
facilities have not been made public knowledge. By far the predominant
application of air stripping technology for removal of VOCs is found in the
removal of trichloroethane (TCE), and trihalomethanes (THMs) from contaminated
ground waters, usually at concentration levels below 1 ppm (see Reference 2
for a discussion of some of the studies). However, some data were found for
wastewaters with higher inlet solvent concentrations and are presented below.
7.4.2.1 Pilot Study at IBM's Facility—
A pilot scale air stripping study was done at IBM's 100,000 gallon per
day (gpd) zero discharge industrial wastewater treatment facility (IWTF) in
Yorktown Heights, NY. The purpose of the study was to determine the
effectiveness of a prototype packed air stripping column for removal of larger
than trace quantities (1 to 50 mg/1) of volatile organics. The column was
constructed of 6 inch diameter pipe packed with 1.5 inch Pall rings to a depth
of 6 feet during the study. The stripping column was operated at the facility
for several days to determine the effect of liquid flowrate on removal of
various solvents. The data in Table 7.4.2 were generated relative to the
solvents of interest.
The data show that increasing the liquid flowrate through the column,
while holding the gas rate constant, decreases the removal efficiency.
Conversely, decreasing the liquid rate, thus increasing the gas/liquid rate,
enhances the performance of the stripping column.
7.4.2.2 Bench Scale Mixed Organics Air Stripping Study (Mumford, 1982)—
A bench scale air stripping treatment study was carried out at the
Department of Civil Engineering and Environmental Engineering at the
University of Iowa. The stripping apparatus consisted of a 4.0 foot high,
7-145
-------
TABLE 7.4.2. 1ESDLTS OF PILOT-SCALE AIR STRIPPING STUDY
I
I—«
4S
Q, gpm/ft2
G/La
Acetone
Tetrachloroethylene
Toluene
Trichloroethylene
4.1 5.8- 8.2 13.3 15.3
184 130 83 45 39
Inlet % Inlet % Inlet % Inlet % Inlet % Inlet
ppm Removal ppm Removal ppm Removal ppm Removal ppm Removal ppm
14 20 41 27 18 8 6 16 6 18 20
44 29 24 26 86 5 15 32
*»•* mmmm _« — «»*«* -.«- *™ — *• — *» — «• 1 1
_» — . «-— — — *•» — — — » — — — — -"* *7
_ _ ._ __ __ _. 3
__ __ ___ *»
<£.
0.9 98 4 89 3 88 1 86 2 67 2
7 90 2 87 2 92 1 99 2
23
24
%
Removal
-29
24
83
80
79
81
79
88
aGas/Liquid Ratio, By Volume
Source: Reference 6.
-------
3.75 inch ID plexiglas column. The column was packed with 1.5 feet of
1/4 inch Berl saddles. The system was operated at gas/liquid ratios ranging
from 25 to 300. Table 7.4.3 presents the influent concentration and percent
removal of selected compounds relevant to this study. In general, the study
showed that removal efficiencies, for a constant gas to liquid ratio, were
highest for compounds with high Henry's Law constants. The data also showed
that as gas to liquid ratio increases, the percent removal also increases. At
even higher ratios, it is anticipated that the percent removal will level off
due to the influence of liquid phase mass transfer resistance. Column
flooding will also occur when the gas to liquid ratio exceeds design criteria.
7.4.2.3 Pilot Scale Air Stripping Treatability Study—
A pilot scale air stripping system was constructed to investigate the
feasibility of air stripping volatiles from a contaminated ground water
g
supply. The system consisted of a 3-1/8 inch x 48 inch glass stripping
column that was packed to a depth of 26 inches with 7 mm glass Rashig rings*
Six runs were made at different air and water feed rates and Table 7.4.4
presents the data generated. This study showed surprisingly poor removal
efficiencies for TCE in comparison to other published reports. The authors
noted that the formation of iron and manganese oxide on the packing material
affected the performance of the system; although it does not appear that this
would affect the volatility of TGE relative to the other compounds studied.
7.4.3 Cost of Treatment
As mentioned earlier, the optimum design for an air stripping system will
be that which involves the least cost for the desired removal efficiency. The
total cost is divided into two categories: (1) capital costs, and
(2) operating costs. Capital costs can be estimated using methods described
by Peters and Timmerhaus. The capital cost is broken down into three
parts: (1) tower shell costs, including heads, skirts, and nozzles,
(2) internal element costs, including packing, supports, and distributor
plates, and (3) auxiliaries costs, including platforms, ladders and
handrails. Using the tables and curves presented in Reference 5, one can
arrive at a capital cost estimate for a variety of column and packing
7-147
-------
TABLE 7.4.3. RESULTS OF BENCH-SCALE AIR STRIPPING STUDY
6/L Ratio8 25 50 100 200
L(tn3m2-hr) 10.3 11.3 7.81 3.98
Inlet % Inlet X Inlet % Inlet %
Solvent ppm Removal . ppm Removal ppm Removal ppm Removal
Carbon 5.4 69 84.2 82 67.4 89 15.3 87
Tetrachloride
Toluene 3.4 74 5.3 77 37.1 93 2.7 96
Chlorobenzene 13.3 65 9.2 72 7.8 77 3.6 97
1,2-Dichloro-
benzene 14.2 60 27.8 70 24.0 74 11.0 94
Nitrobenzene 4.2 5 111 11 115 16 3.9 28
aGas/Liquid Ratio, by Volume.
S ource: Re fe re nee 7.
7-148
-------
TABLE 7.4.4. RESULTS OF PILOT-SCALE AIR STRIPPING TREATABILITY STUDY
Gas to Liquid Ratio 53 26 54 107 396 9
L, liters/m 0.75 0.75 0.37 0.37 0.1 0.75
Raw % Removal
Compound ppm Run 1 Run 2 Run 3 Run 4 Run 5 Run 6
1,1,1-Trichloroethane 150 65 56 60 74 95 56
Trichloroethylene 338 0 9 9 — 44 9
(TCE)
Methyl Isobutyl 76 41 21 68 . 43 75 21
Ketone
Toluene 92 67 75 79 81 92 51
Ethylbenzene 24 99 99 98 98 99 96
Benzene 69 48 — — — — 0
Tetrahydrofuran 22 ND — — — — ND
ND ~ Not Detected.
Source: Reference 8.
7-149
-------
materials. Cost estimates can vary greatly depending on the diameter and
height of tower needed, the depth and type of packing, and the material of
construction. The operating cost for an air stripping system is also related
to the complexity of the pre- and post-treatment options necessary to operate
the system as well as the operating parameters of the column itself.
Operating costs that are specifically associated with the stripping operation
are incurred from the electricity required to run the liquid feed pump and the
air blower.
Some data are available from a field operation in Wisconsin where in 1984
an air stripping column was constructed to remove VOCs from contaminated
Q
ground water. . The system treats an average of 2 million gallons of ground
water a day with about a 98 percent VOC removal efficiency. Data covering the
first 5 months of operation show the cost of treatment to be about 5.11 cents
per 1,000 gallons of water treated. The tower is reported to have cost
roughly 104,000 dollars to install. Operating costs were about 33,000 dollars
per year, including 14,000, 16,000, and 3,300 dollars per year for capital,
operating and maintenance costs, respectively.
7.4.4 Overall Status of Process
7.4,4.1 Availability™
The use of air stripping as a primary treatment process for the removal
of volatile organics from water streams is relatively new and is generally
used only for dilute concentrations. However, equipment suitable for almost
any application is available from a number of vendors specializing in the
design and sale of vapor-liquid contacting equipment. The Chemical
Engineering Buyers* Guide , for example, identifies about 100 manufacturers
who custom design tower equipment for contacting operations including air
stripping and similar applications. Another 50 or so firms are involved in
supplying aeration equipment for vapor-liquid contacting. Some bench or pilot
scale activity will undoubtedly be required to establish design requirements
for equipment for specific waste stream applications.
7-150
-------
7.4.4.2 Application—
As noted in Reference 2, air stripping may not achieve very low
concentrations of volatile solvents present above 5 mg/liter, even at high
air-to-water ratios. It is suggested in Reference 2 that air stripping may be
useful in treating water containing volatiles at low concentrations; or in
treating higher concentrations of volatiles prior to further treatment, such
as biological treatment. Steam stripping should probably be considered a more
effective process for higher concentrations of solvents, particularly those
that are less volatile.
7.4.4.3 Environmental Impact—
Consideration of the environmental impact of air stripping processes
should focus on assessing the concentration levels of the air/water primary
process streams. Additional treatment may be indicated. Residuals from
secondary treatment (e.g. activated carbon adsorption) may pose additional
treatment or disposal requirements, as would residuals from pretreatment
operations to remove solids and metals from waste stream feedstocks.
7.4.4.4 Advantages and Limitations—
The principal advantage of air stripping is its simplicity and inherently
low cost. However, increases in processing time and the inclusion of design
features aimed at improving mass transfer efficiency may add appreciably to
the cost. The needs for offgas treatment could have a significant influence
on column design and on overall process economic viability. Air stripping is
most cost effective as a final polishing treatment for aqueous wastes with low
(e.g., less than 5 ppm) VOC concentrations. At higher inlet concentration, it
may be used as a pretreatment process followed by biological treatment or
carbon/resin adsorption.
7-151
-------
REFERENCES
1. Berry, R.H. et al. Chemical Engineers' Handbook, 6th Edition,
McGraw-Hill Book Company, New York, NY, 1984.
2. U.S. EPA. Background Document for Solvents to Support 40 CFR Part 268
Laws Disposal Restrictions. OSW, Washington, DC. January, 1986.
3. U.S. EPA. Physical-Chemical Properties and Categorization of RCRA Water
According to Volatility. EPA-450/3-85-007, February, 1985.
4. Lyman, W.J., et al. Handbook of Chemical Property Estimation Methods:
Environmental Behavior of Organic Compounds. McGraw-Hill Book Company,
New York, NY, 1982.
5. Peters, M. S. and K. D. Timmerhaus. Plant Design and Economics for
Chemical Engineers, McGraw-Hill Book Company, 1980.
6. Riznychok, W., J. A. Mueller, and J. J. Giunta. "Mr Stripping of
Volatile Organic Contaminants From Sanitary and Industrial Effluents."
Proceedings: 15th Mid-Atlantic Industrial Waste Conference - TOKIC and
Hazardous Wastes, 1983.
7. Mumford, R. L. and J. L. Schnoor. "Air Stripping of Volatile Organics in
Water." Proceedings: American Water Works Association Conference, 601,
1982.
8. Stover, E. L. and D. F. Kincannon. Contaminated Groundwater Treatability
- A Case Study." Journal AWWA, 292, June, 1983.
9. Environmental Reporter. Current Developments, July 12, 1985.
10. Chemical Engineering Equipment Buyers' Guide, McGraw-Hill, New York, NY,
1986.
7-152
-------
7.5 LIQUID - LIQUID EXTRACTION
Liquid - liquid extraction is the separation of constituents of a liquid
solution by transfer to a second liquid, immiscible in the first liquid, but
for which the constituents have a preferential affinity. Although not a
commonly used treatment technology, liquid extraction has potential for
removal of many organic constituents from effluent waste streams. Liquid
extraction can be attractive in cases where the solutes in question are toxic
or non-biodegradable, where the solutes are present at high enough
concentration levels to provide economic recovery value, when steam stripping
would be rendered less effective by low solute volatility or formation of
azeotropes, or when high concentrations increase activated carbon adsorption
costs to excessive levels. Historically, its application to wastewater
treatment generally has been limited to removal of phenol and phenolic
compounds present at high concentration (5% or greater). Steam stripping is
not very effective on phenolic compounds because of low Henry1s Law constants
and activated carbon adsorption is not feasible at these high concentrations.
7.5.1 Process Description
The liquid-liquid extraction process is shown in Figure 7.5.1. The
process includes typically the following basic steps;
1. extraction of organic pollutants from wastewater,
2. recovery of solute from the solvent phase, or extract,
3. removal of solvent from treated wastewater, or raffinate.
The first step, extraction brings two liquid phases (feed and solvent)
into intimate contact to allow transfer of solute from the feed to the
solvent. Any method by which single or multistage mass-transfer processes can
be conducted can conceivably be used to conduct liquid extractions. For
example, an extractor unit can be a mixer-settler device in which feed and
solvent are mixed by agitation, then allowed to settle and separate into two
7-153
-------
SOLVENT aid SOLUTE
WASTE WATER
I
)_*
Ul
TREATED WATER
Figure 7.5.1. .Schematic of extraction process,
-------
liquid streams; or it can be a column in which two liquids are brought into
contact by counter-current flow caused by density differences. The process
yields two streams, the cleaned stream or raffinate and the extract or
solute-laden solvent stream.
The second step, solvent regeneration, can be accomplished by a second
extraction or distillation. For example, a second extraction, with caustic,
is sometimes used to extract phenol from light oil, which is used as the
primary solvent in dephenolizing coke plant wastewaters. However,
distillation is much more common. Potential difficulties with distillation
might arise if azeotropes are present, or if the relative volatilities of the
solvent and the extracted compound are close enough to hinder separation.
The third step, removal of solvent in the treated wastewater or raffinate
is necessary when solvent concentrations are great enough to create solvent
losses that would add significantly to the process cost or have a detrimental
environmental impact. This can be accomplished by a number of technology
options. When treating large quantities of dilute wastes, an additional
extraction step usually cannot compete on an economic basis with other
technologies such as stripping, biological or adsorption post-treatments.
7.5.1.1 Pretreatment Requirements for Different Waste Forms and
Characteristics—
Pretreatment is necessary to remove material which will interfere with
the mass transfer of the organic contaminant into the solvent extract and
which will require higher solvent/aqueous phase ratios to obtain desired
levels of extraction. Thus, any emulsion or organic phase droplets should be
removed by any one of several pretreatment options. Solids, to the extent
that they retain sorbed contaminants or hamper column performance, should be
removed. In certain cases dissolved solids can also affect partitioning of
the solute(s), and removal or addition of material may be desirable to enhance
the separation. Similarly, changes in temperature may also modify
partitioning behavior. Distribution data, if not available in the literature,
will generally have to be developed in the laboratory, although an estimation
method based on vapor liquid data for binary systems can be used to estimate
the distribution of an organic compound (at low concentrations) between water
and an organic solvent. (See Section 15 of the Sixth Edition of the Chemical
Engineers' Handbook by Iterry, et al.)
7-155
-------
7.5.1.2 Operating Parameters—
Liquid extractions may be carried out in various ways. In the simplest
case the solvent is added to a liquid mixture, causing a second liquid phase
to form. It may be desirable to add a salt to an aqueous phase to enhance the
activity of a component, causing it to transfer into a nonaqueous phase in
which the salt is insoluble. It may also be desirable to adjust the pH of an
aqueous phase containing organic acidic or basic solutes to depress their
ionization and cause them to concentrate in the nonaqueous solvent phase. It
is often helpful to change the temperature of the phases in contact to give
the most favorable equilibrium at each step of the extraction.
Theoretically, any aqueous organic waste can be treated by extraction.
However, determining potential feasibility requires a series of analyses to
assess overall system practicality. Much depends on how residual solvent is
to be removed from the treated water stream, how the solvent is to be
regenerated, and what restrictions exist for each unit operation.
In general, extraction is best suited for waste streams of consistent
composition to assure satisfactory performance. In cases where performance is
less important, acceptable ranges in waste characteristics become a little
broader, as in the case where extraction is to be used as a pretreatment. For
example, when several waste streams are to be combined for final treatment, a
single waste stream with higher constituent concentration may be extracted to
reduce the load on the final treatment process.
As noted in Perry's Chemical Engineers' Handbook, the removal mechanisms
in extraction are primarily physical, since the solutes being transferred are
ordinarily recovered without chemical change. On the other hand, the physical
equilibrium relationship on which such operations are based depend mainly on
the chemical characteristics of the solutes and solvents. Thus, use of a
solvent that chemically resembles one component of a mixture more than the
other components will lead to concentration of that like component in the
solvent phase, with the exclusion from that phase of the dissimilar components.
The choice of solvent is a key factor in evaluating the utility of liquid
extraction as a means of removing hazardous organic compounds from aqueous
waste streams. Perry, et al. lists characteristics which must be assessed
in selecting a solvent. These are:
7-156
-------
* Selectivity-the ability of a solvent to extract the organic
contaminant preferentially from the aqueous phase. It is a
numerical measure that is equal to the ratio of the distribution
constants of contaminant and water in the solvent. As such, it is
analogous to relative volatility as used in distillation. Poor
selectivity (ratios near unity) means large solvent feed ratios and
a large number of extraction stages will be needed for good
separation.
* Recoverability-the solvent must be recoverable from both extract and
raffinate. Since distillation is the usual recovery.method,
relative volatilities of all components should be favorable and low
latent heats for volatile solvents are desirable.
* Distribution Coefficient-the distribution coefficient of the
contaminant should be large in order to achieve selectivity and
reduce equipment size and costs.
* Contaminant Solubility-the solubility of the extracted contaminant
in the solvent should be high in order to reduce solvent
requirement s.
* Solvent Solubility-the solubility of the solvent in the aqueous
phase should be low. This will generally increase selectivity, the
range of waste stream concentrations that can be handled, and reduce
costs of solvent recovery or makeup.
* Density-a difference in density is essential since the flow rates
and separation of the two phases is directly affected.
• Interfacial Tension-the interfacial tension should be large to
assist in the coalescence of dispersed phase droplets.
* Other-other desirable solvent properties are low corrosivity, low
viscosity for higher mass transfer rates, nonflammability, low
toxicity, and low cost.
Binary critical solution temperatures of solute components with
prospective solvents is suggested as a guide to solvent selection. The
solvent having the lower critical solution temperature with the solute
compound will be more selective in an extraction from the aqueous phase.
A significant quantity of data have been collected for the distribution
of pollutants in water and various extractive solvents. These values, called
equilibrium distribution coefficients (Krj)» generally express the
equilibrium concentration of the solute as the ratio of the weight percent in
solvent relative to water. It can also be expressed as mole or volume (Ky)
fraction ratios.
7-157
-------
X
K = Os
*D X
oa
where: X08 the weight fraction of organic solute in the solvent
phase and
Xoa is the weight fraction of organic solute in the aqueous
phase, both at equilibrium.
KD values for the octanol/water system for solvents and other low
molecular weight organic compounds of interest are provided in Appendix A.
These data represent just a small fraction of the data available in the
literature for ternary systems consisting of water and two organic compounds.
For example, a number of references are provided in Perry along with
distribution coefficients for over 200 selected water/organic solute/organic
solvent systems which include many of the organic compounds of concern.
Values of K-. or TL. that are specific to the compounds of concern are
provided in a number of recent publications. These data, of use in assessing
the potential of liquid-liquid extraction as a treatment technology, are
provided in Tables 7.5.1 through 7.5.4. The K., data in Table 7.5.1 were
v 2
reported in the AIChE Symposium Series (1981) by S. T. Hwang. The basic
references used are provided in the table. Tables 7.5.2 through 7.5.4 report
KL data from a two-part University of California, Berkeley report published
by the EPA (see Reference 8).
Higher values of K mean that less solvent is required to extract a given
amount of solute from the wastewater and thereby usually leads to less
expensive extraction processes. The ratio of the distribution coefficients of
solvent systems for extraction of a specific solvent from water is a measure
of the relative amounts of solvent that must be employed to achieve a given
level of extraction. However, distribution coefficients are only one of the
many solvent properties that must be considered.
The Chemical Engineers' Handbook, Kirk Othmer (Reference 4) and other
background materials present calculation and design methods that can be used
to assess the applicability of liquid-liquid extraction to specific waste
*
streams. The techniques generally involve the use of equilibrium distribution
data to develop equilibrium and operating line curves which can be used to
provide graphical calculations of the number of theoretical stages required to
7-158
-------
TABLE 7.5.1. Ky VALUES FOR AQUEOUS/SOLVENT SYSTEMS
•vj
I
Solute
acrolein
benzene
carbon tetrachloride
ch lore-benzene
1 , 2-d ic h loroe thane
1 , 1 , l-t r ich loroethane
hexachloroe thane
1 , 1-dichloroethane
1,1,2-trichloroethane
chloroethane
bis(2-chloroethyl)
ether
chloroform
1,2-dichlorobenzene
1,1-dichloroethylene
l,2-trans-
-------
TABLE 7.S.I (continued)
I
I—•
O\
O
Solute
acrolein
benzene
benzidine
carbon tetraehloride
chlorobenzene
1,2-dichloroethane
1 , 1, 1-trichloroethane
hcxachloroethane
1,1-dichloroe thane
1 , 1 ,2-trichlaroethane
chloroethane
bis(2-chloroethyl)
ether
chloroform
1, i-dichloroe thy lene
1 ,2-trans-dichloro-
ethylene
1 , 2-d ichloropropaite
1 , 3-d ichloropropy lene
ethylbenzene
metbylene chloride
methyl chloride
methyl bromide
bromoform
trichlorofluoro-
tnethane
dlchlorodif luoto-
methane
nitrobenzene
toluene
t richloroe thy lene
CuiDene
1*
263*
85*
1,460*
1,080*
72*
196*
2.4xl04
66*
108*
33*
34*
56*
1,900*
1,300*
170*
168*
2,480*
46*
10*
509*
208*
701*
5,127*
266*
1,300*
375*
Hetityl
oxide
3*
130*
66*
770*
1,700*
34*
212*
8,183*
163*
256*
67*
114*
134*
990*
738*
369*
555*
995*
69*
20*
1,090*
682*
925*
4,180*
309*
606*
1,074*
HER
4*
114*
78*
690*
3,480*
30*
221*
6,600*
200*
309*
79*
148*
162*
870*
665*
440*
703*
840*
106*
24*
1 , 300*
880*
1,000*
4,600*
' 349*
527*
1,320*
Ethyl
acetate
3*
109*
70*
668*
2,360*
28*
208*
6,100*
186*
278* •
75*
129*
150*
840*
640*
404*
632*
797*
89*
22*
1,220*
778*
975*
23,000*
298*
506*
1,220*
Ethyl
ether
2*
174*
131*
1,200*
1,000*
40*
239*
IxlO4*
127*
174*
61*
63*
103*
1,360*
982*
298*
333*
1,570*
45*
16*
846*
366*
1,200*
2.1xl04*
176*
919*
771*
Solvent
Ethyl
benzene
2*
218*
86*
1,266*
1,450*
55*
22*
2.8xl04*
93*
165*
40*
64*
79*
2,127*
903*
234*
•315*
2,940*
59*
14*
700*
418*
718*
4,573*
309*
1,290*
595*
n-hexanol
3*
45*
16*
236*
1,180*
14*
96*
1,184*
130*
202*
50*
92*
109*
359*
294*
250*
492*
230*
58*
20*
1,049*
696*
454*
269*
270*
170*
840*
Ethylene
DichlorM*
1*
473*
65*
2,329*
3,700*
142*
212*
9xl04*
55*
115*
25*
38*
48.3*
2,770*
1,768*
158.8*
180*
6,070*
134*
9*
434*
238*
503*
2,578*
353*
2,674*
344*
Toluene
2*
271*
98*
1,570*
1,660*
70*
248*
3. 7x10**
101*
184*
43*
72*
86*
2,400*
1,100*
259*
246.8*
3,700*
68*
15*
751*
462*
773*
4,500*
363*
1,690*
. 650*
Xylene
3*
360*
136*
2,069*
2,500*
92*
388*
4.7xl04*
163*
290*
70*
110*
138*
3,800*
1,488*
407*
548*
4,700*
103*
24*
1,249*
734*
1,249*
7,380*
550*
2,080*
1,030*
n~hexane
1*
116*
101*
866*
610*
24*
177*
8,600*
73*
95*
39
28
60*
1,330*
562*
170*
178
1,400
30*
U*
550*
188*
894*
14,880*
80*
708**
436*
*Estimated from solubility parameters.
**Estiinated using the method in Eeference 4.
Other reference numbers in ().
aK_ * K Specific Gravity ptSolvent m Grama Solute/1,000 ml Solvent
v 1) Specific Gravity of Water Grams Solute/1,000 ml Water
-------
TABLE 7.5.2.
EXPERIMENTALLY MEASURED EQUILIBRIUM DISTRIBUTION
COEFFICIENTS FOR EXTRACTION FROM WATER INTO
UNBEGANE, (295-300 K) AND 1-OCTANOL
Solute
Undecane
1-octanol
Benzene
Toluene
Ethylbenzene
Chlorobenzene
1,2-Dichlorobenzene
Chloroform
Carbon tetrachloride
1,1, 1-trichloroe thane
1,1,2, 2-tetrachloroethane
1 , 2-dichloropropane
1 , 2-tr ans-dichloroethy lene
Tr ichloroe thy lene
Tetrachloroethylene
Bromoform
214
740
2,500
845
3,030
75
738
414
110
74
119
354
2,700
127
160
196
1,710
838
3,040
96
650
357
297
«, —
318
409
««_
Source: References 3 and 8,
7-161
-------
TABLE 7.5.3. MEASURED VALUES OF KD FOR EXTRACTION OF ACROLEIN
FROM WATER INTO VARIOUS SOLVENTS
Solvent
Measured
Alcohols
2-ethylhexanol
2-ethyl-l,3-hexanediol
Carboxylie Acids
2-ethylhexanoic acid
Naphthenic acids
Alkane
Undecane
Chlorinated Organics
Methylene chloride
Tetrachloroethylene
1,2-dichloroethane
1,1,2-trichloroethane
1,1,2,2-tetrachloroethane
Aromatics
Toluene
Styrene
Nitrobenzene
Olefins and Terpenes
1,5-cyclooctadiene
Turpentine
Esters
n-hexyl acetate
n-butyl acetate
n-octyl acetate
Ether
di-n-butyl ether
Diisopropyl ether
1.25
1.41, 1.59
0.45, 0.59
0.62, 0.62
0.47, 0.41
6.6
0.54
3.75
3.45
4.6
2.05, 2.25
2.46, 2.53
2.68, 2.58
0.54
0.39, 0.43
1.99, 2.17
2.38, 2.75
1.71
1.06, 1.09
1.7
(continued)
7-162
-------
TABLE 7.5.3 (continued)
Solvent Measured K
•D
Ketones
Isobutyl heptyl ketone (IBHK) 0.93
Methyl isobutyl ketone (MIBL) 4.90, 4.91, 4.84
Mesityl oxide 3.01, 5.02
Diisobutyl ketone (DISK) 1.57, 1.63
5»5-dimethyl-l,l,3-cyclohexanodione;
-saturated in 2-ethylhexanol 5.78, 5.9, 3.15
-saturated in DISK 2.66, 2.15
Furan
2-methyl furan (99% pure) 2.66, 2.23
2-methyl furan, with up to 10%
methyl tetrahydrofuran 8.4, 17.4, 9.29
Phosphoryl Compounds
Tributyl phosphate (TRP) 1.97, 1.99
fricresyl phosphate (TCP) 1.77, 1.68
Di-2-ethylhexyl phosphoric acid 1.72, 1.04
22.1% w/w trioctyl phosphine oxide (TOPO) in DIBK 2.06, 1.63
Others
Ethyl propionate 1.5
Isobutyl isobutyrate 1.98
Nitropropane 3.6
Source: Reference 8
7-163
-------
TABLE 7.5.4. MEASURED EQUILIBRIUM DISTRIBUTION COEFFICIENTS (KD)
AT 30°C FOR EXTRACTION OF NITROBENZENE FROM WATER
INTO VARIOUS SOLVENTS
Solvent
Toluene
DIBK
Solvent/ Feed
feed concentrations,
(v/v) ppm (w/w)
Undecane
0.25
0.25
0.25
1,206
1,206
1,206
36
39
37
TBP
(tri butyl phosphate)
MIBK
0.03
0.03
0.03
0.10
0.10
0.10
0.15
0.10
0.10
0.10
0.10
0.10
0.05
0.05
0.10
965
1,206
1,206
503
1,206
1,206
1,206
603
603
1,206
1,206
1,206
1,206
1,206
1,206
352
353
353
277
278
301
270
269
297
277
289
310
244
217
204
7-164
-------
achieve desired extraction levels. The general method is analogous to the use
of McCabe-Thiele diagrams to assess distillation performance. Formulas, such
as the Kremser equations, are also available that quantitatively express the
effect of flow variations on exit concentration levels. The use of such
techniques (in conjunction with laboratory data) will provide the basis for
determining equipment size and post-treatment requirements and, therefore, the
costs and applicability of the liquid extraction process.
7.5.1.3 Bost-Treatment Requirements—
The post-treatment requirements of a liquid-liquid extraction process
will be determined by many of the system component properties discussed above,
e.g., solvent solubility in the aqueous phase will determine the need for
further treatment to eliminate solvent discharge with the treated waste
stream; and the relative volatilities of the solute and solvent will affect
the ease of their separation following extraction. These operations could be
considered to be process steps as opposed to post-treatment steps, and careful
selection of solvent and proper design can minimize the cost and difficulty of
such processes. Given the necessity for post-treatment, those technologies
most commonly used for raffinate are established technologies such as
steam/air stripping, carbon adsorption, and biological treatment.
Distillation will probably be used to separate solvent and solute. These
.technologies are discussed in some detail in other sections of the report.
7.5.2 Demonstrated Performance
The use of liquid-liquid extraction for the treatment of aqueous organic
waste streams has been limited. Application of the technology has been
primarily for the treatment of phenol contaminated waste streams from the
petroleum and coal processing industries. Because of difficulties involved in
removal of phenol from these specific waste streams by steam stripping and
adsorption, liquid-liquid extraction has proven to be particularly well suited.
Since the scope of this document does not include phenols, actual
performance data in the field are limited to a series of pilot runs. Most of
the research was conducted under EPA auspices to assess the extractability of
priority pollutants from industrial waste streams. Results of these studies
are summarized below.
7-165
-------
7.5.2.1 Results of Study #1; Reference 10, Earhart, S.P., et al. (1976)—
Solvent extraction was explored in this EPA sponsored program as a method
of treating wastewaters from petroleum refineries and petrochemical plants.
Waste constituents included solvents covered in this report. Results were
obtained from the use of both spray columns and rotating disc contactors (RDC).
7.5.2.1.1 Results FromSpray Column Extractor—The primary objective of
experiments conducted in the spray column was to prove the overall process
feasibility of volatile liquid extraction. Several types of organic chemicals
were extracted. Most of these solutes are known to be present in the
wastewater from chemical processing plants, but they were also chosen to
determine if any class of organic chemicals might cause unexpected problems
such as an irreversible reaction with isobutylene and the other extraction
solvents tested. Host of the solutes studied were present in synthetically
prepared water solutions, but industrial samples of lube oil refining
wastewater and cresylic acid recovery wastewater were also tested.
The data, which were measured at steady state in each run included the
solvent and water flow rates, the temperatures of the two streams leaving the
extractor, and the concentrations of each solute in the feed and product
water. Other variables measured included, component viscosity and density,
phase velocity, interfacial area per column volume, percent of flooding, and
dispersing phase rise (or fall) time. Typical variables recorded during the
study are shown in Table 7.5.5. Details can be found in the reference for
specific runs. Results for several solvent/contaminant combinations were
reported, as summarized in Table 7.5.6. As shown in the table, average
effluent product water contaminant concentrations were 10 mg/liter or more.
7.5.2.1.2 Results from RDC extractor—Steady state data obtained in the study
included the solvent and water flow rates, the diameters of the discs and
stator holes, the compartment heights and column height, the rotational speed
of the discs, the temperatures of the two streams leaving the RDC, and the
concentrations of each solute in the feed and product water. In some
experiments the solvent hold up, and the solute concentrations in the loaded
solvent, were also measured.
7-166
-------
TABLE 7.5.5. VARIABLE MEASURED IN SPiAY COLUMN EXTRACTOR STUDY
Spray Column Run #1
Dispersed Phase
Continuous Phase
Priority Constituents
Vd = 82.43 ft/hr,
Pd = 0.5908 gm/cc,
pc = 0.9976 gm/cc,
dn =.01865 inch,
FF = 15.:
Temperature = 22.7°C
= Isobutylene,
= Prepared waste water
~ o—cresol
Vc - 19.38 ft/hr, FS/FW = 2.519
Cd ~ 0.180 cp,
Cc - 0.942 cp,
a - 17.11 ft2/ft3
RT « 6.78 seconds
Solute KD
1. Phenol 0.70
2. 0-cresol 4.80
Feed
water
(ppm)
20,000
10,000
Product
water
(ppm)
4,180
492
Percent
removal
79.1
95.1
Pd
PC
p
a
FF
RT
= dispersed phase velocity
= continuous phase velocity
= solvent mass flow rate/water mass flow rate
= density of dispersal phase
- density of continuous phase
= viscosity of dispersal phase
= viscosity of continuous phase
= drop diameter
= interfacial area per column volume
— percentage of flooding
— dispersed phase rise (or fall) time
= equilibrium distribution coefficient
7-167
-------
TABLE 7.5.6. RESULTS OF SPRAY COLUMN EXTRACTOR RUNS
r
CO
Wastewater/solvent
contaminant
Kuaber
of
KJJ runs
F8/FW
range*
Feed
waterb
(ppm)
Product
Haterb
(ppm)
Removal
(percent)
Removal
range
(percent)
Prepared Wastewater/Isobutylene
Phenol 0,7
Acetone 0.63
Benzene 407
m-butanol 0.76
n-butyl acetate 168
o-cresol 4,8
ethylene dichlorlde 70
methyl ethyl ketone 2.49
Prepared Waatewater/n-butane
Ethylene dichlorlde 44 3 0,189 to 0.475 2,890
Cresylic Acid Recovery Wastewater/Iaobutylene
4
6
2
4
6
4
4
4
1.381 to 2.519 18,150
0.686 to 2.189 2,050
0.686 to 1.833 295
0.686 to 2.624 5,204
0.686 to 2,184 2,125
1.381 to 2.519 3,550
0.198 to 2.189 2,980
1.330 to 2.189 3,390
3,260
1,025
13
1,420
150
165
510
1,090
1,190
82
50
97
63
93
95
83
68
59
76 to 82
28 to 74
93 to 98
35 to 88
76 to 98
95 to 97
62 to 93
62 to 78
43 to 75
fhenol
o-eresol
m p-cresol
xylenola
lube Oil Refining
Phenol
Acetone
Benzene
o-cresol
0
4
2
7
.7
.8
.7
.0
1
1
1
1
1.785
1.785
1.785
1.785
579
327
291
227
163
31
25
10
72
90
91
96
_
—
—
_
Wastewater/Isobutylene
0
0
407
4
methy ethyl ketone 2
.7
.63
.8
.49
3
2
2
2
2
1.429
1.429
1.429
1.429
1.429
to 2
to 2
to 2
to 2
to 2
.535
.535
.535
.535
.535
21,150
37
170
232
232
8,080
19
21
2040
39
62
49
88
91
84
41
41
80
84
76
to 80
to 57
to 96
to 98
to 95
a *B^FW ** solvent mass flow rate/water mass flow rate.
" Average values.
Source: Earhart, J.P., et al. (1976), Reference 10.
-------
During the second portion of the experimental program, several prepared
aqueous solutions and industrial wastewaters were treated by extraction in the
RDC as described in Reference 10. Experiments included runs using volatile
solvents, less volatile polar solvents, and mixtures of volatile and polar
solvents. In these experiments the solvent was always the dispersed phase,
and the solvent to water flow ratio (F /F ) was set at much lower values
(i.e., ratio ranged from 0.1 to 0.3) in order to demonstrate solvent
extraction under conditions which would be most likely to lead to favorable
process economics.
In addition to the choices of solvent and the value of F /F , the
s w*
independent variables which could be varied on the RDC included the water flow
rate, the disc rotational speed, the disc diameter, the stator hole diameter,
and the compartment height. The principal measured responses were the solvent
hold up, the concentrations of each solute in the treated water, and in most
runs, the concentration of each solute in the loaded solvent. The temperature
(ambient 21°C) was measured but not controlled. Results for several runs are
summarized in Table 7.5.7. Full details concerning all data collected during
the test study are available in Earhart, et al. (1976). The removal
efficiencies are generally similar to those obtained in the spray column
studies.
7.5.2.2 Results of Study #2: Ricker and King, 1980, Reference 11—
Using the same rotating disc contactor pilot equipment for runs conducted
by Earhart, et al. (1976), a solvent extraction study of wastewaters from
acetic acid manufacturers was performed by University of California, Berkeley
staff members and was published in 1980 as EPA Report 600/2-80-064. The test
conditions studied and the results of a test run are shown in Table 7.5.8.
Waste constituents studied included those shown in the table plus a variety of
other low molecular weight organic acids, alcohols, aldehydes, and ketones.
Solvents included 2-ethyl hexanol, n-amyl alcohol, eyelohexanone, 2-heptanone,
and diisobutylketone. Removal efficiencies for all materials tested fell
within the range shown in Table 7.5.8. Contaminant concentrations in the
treated raffinates (water phases) were generally high enough to require some
form of post treatment.
7-169
-------
TABLE 7.5.7. RESULTS OF RDC EXTRACTOR RUNS
Sastewatet/solvent
contaminant
KD
VF»
range8
Feed
waterb
(ppra)
Product
waterb
(pwn)
Removal
(percent)
Removal
rangec
(percent)
Prepared Wostewater/Isobutylene
Phenol
Acetone
Benzene
n-butyl acetate
o— creBol
methyl ethyl ketone
0.70
0.63
407
168
4,8
2.49
0.0996
0.0996
0.0996
0.0996 to 0.3538
0.0996
0.0996 to 0.3538
605
29.9
68.5
4,050
72.8
1,450
522
28,2
60.4
125
17
660
13.7
5.7
11.8
97
76.6
54
—
—
—
93 to 99
— -
1 to 70
Prepared Wastewater/n-butyl Acetate
Phenol
Acetone
Benzene
o— cresol
methyl ethyl ketone
57
1.05
61.5
206
4.56
0.0974 Co 0.1196
0.0974
0.0974
0.0974
0.0974 to 0.1196
9,720
38
169
2,107
1,215
690
34.3
31
25
720
93
10
82
99
37
83 to 98
— —
—
_
36. to 42
Lube Oil Refining Hastewater/Isobutylene
Phenol
n-butyl acetate
o— cresol
Methyl ethyl ketone
0.7
168
4.8
2.49
Lube Oil Refining Wastewater/n-butyl
Phenol
o-cresol
Hethyl etnyl ketone
57
206
4.56
0.1019
0. 1019
0.1019
0.1019
Acetate
0.1010 & 0.3042
0.1010 & 0.3042
0.1010 & 0.3042
268
5,960
21
4,190
8,751
892
12,216
209
13
215
2,745
90
5.5
4,020
23
99.8
88
35
99
99.4
67
17 & 26
99.8 & 99.9
84 & 96
33 & 36
98.8 & 98.8
99.3 & 99.5
52 & 82
Ethylene Quench Hater/Isobutylene
Phenol
Benzene
Toluene
Xylenes
0.7
487
1,690
-—
0.1010
0.1010
0.1010
0.1010
66.9
71.1
40.5
40.3
63.1
2.9
2.3
1
5.7
95.9
94.3
97
__
~
—
—
Ethylene Quench Water/Isobutane
Phenol
Benzene
Toluene
Xylenes
0.2
338
1,460
—
0.0973
0.0973
0.0973
0.0973
68.2
81.2
43.8
33.6
66.0
2.4
1.6
1
3.2
97.0
96.3
93
__
__
—
—
Syrene Wastewater/Isobutylene
Benzene
Ethyl benzene
Styrene
407
—
0.1072
0.1072
0.1072
290
120
15
10
4
1
96.6
96.7
93
__
—
* Fa/Fv ~ solvent oass flow rate/water mass flow rate
" Average values for runs of 2 or more.
cUsc of & denotes only two values.
Source: Earhart, et al. (1976), Reference 10.
7-170
-------
TABLE 7.5.8. CONDITIONS AND RESULTS FOR MINI-PLANT EXTRACTION RON
Solvent * n-butyl acetate" (dispersed phase)
Water flow rate - 9.46 L/h - 9.48 kg/h
Solvent flow rate =3.56 L/h "3.18 kg/h
Solvent to water ratio =0.33 kg/kg =0.38 L/L
Density of solvent - 0.876 kg/L (25°C)
Shaft rotation speed = 1,200 rpm
Estimated droplet size = 0.75 mm diameter
Column pressure (top) = 390 kPa
Average column temperature - 20.5°C
Rotor disc diameter = 3.81 em
Stator hole diameter = 5.72 cm
Component
acetic acid
acetaldehyde
2-butanol
methyl isobutyl ketone
acetone
n-butyl acetate
Feed
water
(ppm)
850
300
180
130
0
0
Steady state
raffinate
(ppm)
785
250
60
4
1,910
7,320
Percent
removal
8
17
67
97
—
— "•
Source: Reference 11.
7-171
-------
7,5.2.3 Results of Other Studies: EPA summaries of solvent extraction data—
12 13
Two earlier studies ' were conducted by EPA summarizing available
solvent extraction data as shown in Table 7.5.9. These and other related data
from the same studies can also be found in Reference 14, the U.S. EPA
Treatability Manual.
7.5.3 Costof Treatment
As noted in EPA's Treatability Manual it is quite difficult to predict
costs of solvent extraction because of the wide variety of systems, feed
streams, and equipment that may be involved. However, EPA, in Volume IV of
the Treatability Manual does present some cost data based on a waste phenol
feed of 45,000 Ibs/hr containing 1.5 percent phenol by weight and a similar
toluene solvent feed rate. (K_ * 2 for the phenol/toluene/water
distribution coefficient.) The unit is a rotating disc unit containing an
equivalent of about five theoretical stages to produce a wastewater discharge
containing 75 ppm phenol. Using the equations provided in Kirk Othmer
(Volume 9, Liquid-Liquid extraction), approximately five additional
theoretical stages would be required to achieve a discharge level of 21 ppm
which is slightly above the design residual phenol level of the entering
solvent stream.
The costs are presented in Table 7.5.10 with capital costs modified by
ENR index adjustment from 3119 to 4230 (May 1986) and minor changes made in
the costs of toluene and power to represent May 1986 values. The annual costs
of $917,000/year represent costs in excess of $21 per 1000 gallons of treated
wastewater, up from the value of $17 per 1000 gallons provided in Reference 14
(1980 dollars). Assuming capital costs are related to size (number of stages)
through a 0.7 exponential factor as indicated in Reference 14, the capital
costs shown in Table 7.5,10 would increase from $1.5 to over $2.4 million to
achieve a discharge level of 21 ppm phenol in the effluent wastewater (almost
$3.0 million if the phenol content of the toluene solvent is assumed to be
zero and wastewater phenol discharge level is at 1 ppm). Total annual
operating costs would be well in excess of $30/1000 gallons.
7-172
-------
TABLE 7.5.9. RESULTS OF SOLVENT EXTRACTION STUDIES
Chemical
Description of study
Study Study Influent
type8 type" cone.
Results of study
Conaenta
Acrolein
Acrylonitrile
Chlorobenzene
R
R
R
U
0
D
Extractable w/xylene.
Solvent recovery by
azeotropic distillation.
Extractable w/ethyl ether.
600 ppa . 3 ppa effluent cone, using
bls-chloroethyl ether R
Chloroethane
1,1-Dichloroethane R
Bexachloroetoane
1,1,2,2-f etraehloro- R
ethane
1,1,1-lrlehloroethane R
1,1,2-Triehloroethane R
Dichloroethylene
Etnylene Bichloride P, C
0
I*
I*
49 ppa
23-1,804
ppm i
2.76-
3.76 L/mln
chloroform solvent.
Extractable w/ethyl ether
and benzene.
Extractable w/alcohols
and aromatics.
Extractable w/alcohols,
aromatics and ethers.
Extractable w/aromatics,
alcohols and ethers.
Extractable v/arotnatics,
alcohols and ethers.
Extractable w/alcohols
and aromatics.
Extractable w/aromatics,
methanol and ethers.
Kerosene effluent cone.
2 tspffl} ^T.0*" i2 €*•"»**
luent cone. !•*• ppm.
A 5.5:1 water to solvent
ratio gave 94-96Z redue-
tion. C^0~''12 P»*~
affin solvent at 5sl to
16.5:1 water to solvent
ratio showed 94-991
reduction.
Solvent extraction
w/kerosene S CJ.Q~
C^2 hydrocarbon
at 7:1 solvent to
wastewater ratio.
Wastewater contained 14
other halocarbone
including 30-350 ppm
1,1,2-trichloroethane
and 5-197 ppa 1,1,2,2-
tetrachloroethane. A
532 L/mln extractor
w/1,000 ppm influent
eetimated to have a
capital cost of
1315,000 and total
annual cost of 4143,000
including credit for
recovered EDO.
8 Describes the scale of the referenced study:
B - Batch Flow P - Pilot Scale
C - Continuous Flow E - Literature Review
L - Laboratory .Scale.
Source: Reference 12
*Reference 13
Describes the type of wastewater used in the
referenced study:
X - Industrial Wastewater
HI - Unknown
7-173
-------
TABLE 7.5.10. ESTIMATED COSTS FOR A LIQUID-LIQUID EXTRACTION SYSTEM
Operating Characteristics
Value
Water/phenol/feed
Toluene feed
Discharge water
Extraction column
Loss of toluene/cycle
Electrical requirements
(column only)
Operation
45,000 Ib/hr containing 1.5X phenol (by wt);
temperature is UO°F.
45,000 Ib/hr (containing 20ppm phenol from
steam stripper reecycle).
Contains 75 ppm phenol.
Rotating disc type; 6 ft diameter, 60 ft highi
made from carbon steel; contains 50 compartments
and equivalent of about 5 theoretical stages.
(Equilibrium distribution coefficient of phenol
between toluene and water is about 2.)
Approximately O.lZ/cycle.
One 10 hp electric motor.
330 d/yr; 24 hr/d.
Fixed capital costs baaed on an ENR index of 4,230 (Kay 1986) are estimated to
be $1,500,000 (up from an estimate cost of $1,100,000 provided by reference 14
in 1980). Estimation of annual operating cost is presented below.
Cost item
Direct operating cost
Labor
Operating
Maintenance
Chemicals - Toluene
Materials
Steam
Power
Annual
quantity
12,000 man hr
15,000 gal
33 106 Ib
150,000 kWh
Cost per unit
quantity
$16/hr
*1.35
$5/1, 000 Ib
10.05/kWb.
Annual cost8 i
192,000
16,000
208,000
20,300
16,700
165,000
7,500
Total
Total indirect
Operating Cost
Total annual
operating cost*
917,500
&Excludes annual credit for phenol recovery.
Source; Reference 14 (modified to represent May 1986 dollars).
7-174
-------
Conceptual designs and economic analyses were carried out for several
cases in the Reference 8 study, including extraction of nitrobenzene with
diisobutyl ketone (DIBF - a low boiling solvent); and extraction of acrolein
by methyl isobutyl ketone (MIBK), n—butyl acetate, toluene, and
1,1,2,2-tetrachloroethane (all high boiling solvents). The costs derived
represent 1982 dollars. The costs, while lower, appear to be in reasonable
agreement with the costs of Reference 14, given the differences in base year
and operating volumes and concentrations.
Table 7.5.11 presents a breakdown of cost components for extraction of
3
1,000 ppm of nitrobenzene from 11.4 m /h (50 gpm) of water at 30°C using
DIBK. Two cases are considered, with and without vacuum steam stripping to
recover residual DIBK from the effluent water. There is clearly a substantial
incentive to recover DIBK.
Table 7.5.12 summarizes the cost components for extraction of 200 ppm of
acrolein from 30,000 Ib/hr of water by means of four different high boiling
solvents. Details of these analyses are given in Table 7.5.13 for the MIBK
solvent systems. Procedures are identical for the other three solvents. The
costs range from $8.55 to $13.48 per 1,000 gallons of water, with 1,1,2,2-
tetrachloroethane being least expensive because of low steam cost, and toluene
the most expensive because of high capital cost. No cost penalty was imposed
on the chlorinated hydrocarbon for possible extra processing to ensure full
removal from the treated water.
A similar set of analyses was also completed for the extraction of
200 ppm acrylonitrile using methylene chloride and tetrachloroethane as the
solvents. Costs per 1000 gallons of water treated were $9.27 and $4.10,
respectively. The difference is due to lower steam costs associated with the
use of the higher boiling solvent which is not taken overhead during solvent
regeneration. Again no penalty was imposed for recovery of solvents from the
water phase.
The methods used for the cost calculations for extractions of
2-chlorophenol and nitrobenzene are somewhat different from those used for the
cost calculations for the extraction of acrolein and acrylonitrile. These, as
noted in Reference 8, cover a range of conditions which might occur in
different plant locations.
7-175
-------
TABLE 7.5.11. PRELIMINARY EXTRACTION FOR NITROBENZENE EXTRACTION USING
DI-ISOBUTYL KETONE (DIBK)
Direct Costs:
Regenerator
Extract-Solvent
Heat Exchanger
Extractor
Reboiler
Regenerator Condenser
Vacuum Stripper
Stripper Condenser
Total Direct Cost
Indirect Costs:
(38% of Direct Costs)
Fixed Capital Investment (FCI)
Fixed Charges:
Depreciation (8% FCI)
Maintenance (6% FCI)
Insurance and Taxes (32 FCI)
Total Fixed Charges (172 FCI)
Direct Operating Costs:
Make-Up Solvent
Utilities: 100* Steam
400fl Steam
Cooling Water
Labor, Supervision, Lab Charges
Total Direct Operating
Annual Operating Cost:
Operating Costs (per 1,000 gallons of
Items
Labor—dependent
Capital-dependent
Solvent
Utilities
Total Operating Cost
Case I
No Solvent
Recover?
4 50,700 (41. OX)
27,200 (22.0%)
22,100 (17.9%)
17,500 (14.1%)
6,200 (5.0%)
4123,700 (100%)
4 47,000
4170,700
4 13,700/yr
10,200
5,100
4 29,000/yr
4 71,400/yr
32,500
1,000
41,300
4l46,200/yr
4l75,200/yr
wastewater)
Case X
No Solvent
Recovery
41.65
1.16
2.86
1.34
37.01
Case II
Including Steam
Stripping
4 50,700 (30.5%)
27,200 (16.4%)
22,100 (13.3%)
17,500 (10.5%)
6,200 (3.7%)
33,000 (19.9%)
9,400 .(5. 7?)_..
4166,100 (100%)
4 63,100
4229,200
4 18,300/yr
13,800
6,900
4 39,000/yr
4 9,100/yr
3,100
32,500
2,300
41,300
4 88,300/yr
4l27,300/yr
Case II
Including Steam
Stripping
tl.65
1.55
0.37
1.52
45.09
Sources Reference 8
7-176
-------
TABLE 7,5.12. BREAKDOWN OF OPERATING COSTS FOR EXTRACTION OF ACROLEIN
Solvent
Labor
Capital
Solvent loss
Utilities
Credit for
Acrolein
TOTAL
MIBK
2.477
3.054
1.453
4.109
(-0.511)
10.58
Costs - $/100(
Butyl Acetate
2.477
3.223
1.446
3.747
(-0.511)
10.38
) Gallons of Water
Tetrachloroethane
2.477
3.756
0.746
2.079
(-0.511)
8.55
Toluene
2.477
6.347
0.558
4.605
(0.511)
13.481
Source: Reference 8.
7-177
-------
TABLE 7,5.13. ESTIMATION OF OPERATING COSTS EXTRACTION OF ACROLEIN BY MIBK
(30,000 Ib/hr WATER, 200 ppm ACROLEIN)
Item and Basis
Dollars
per year
Dollars per
1000 gallon water
Operating Labor Charges:
1/3 operator/shift at 125,000
dollars/operator/shift
Labor Dependent Charges:
45,000
1.501
Fringe Benefits 22% of operating labor
Supervision, Clerical, 18% of operating labor
Operating Supplies, 10% of operating labor
Laboratory, 151 of operating labor
Fixed Charges:
Maintenance, 6% of DFC
Depreciation, 10% of DFC
Insurance, 1% of DFC
Local Taxes, 2% of DFC
Factory Expenses, 5% of DFC
Make-Up Solvent Costs:
Loss in Regenerator
Loss in Stripper
Utilties:
150 psig steam, $ 8.90
1000 Ib
Cooling water, and refrigeration
Credit for by Product Acrolein
(1/2 x market price)
Total Operating Cost Without Interest
Interest, 10% of DFC
Total Operating Cost With Interest
9,900
8,100
4,500
6,750
22,890
38,150
3,820
7,640
19,080
19,380
24,190
84,810
38,390
(-15,320)
317,260
38,150
355,410
0.331
0.270
0.151
0.225
*
0.764
1.273
0.127
0.255
0.636
0.646
0.807
2.829
1.280
(-0.512)
10.580
1.272
11.854
7-178
-------
7.5.4 Overall Status of Process
7.5.4.1 Availability—
Although liquid-liquid extraction processes have not been extensively
applied to the treatment of waste streams, they are extensively used within
the chemical process industry to affect separations and recoveries.
Table 7,5.14 lists a number of processing equipment units which can be used
for liquid-liquid extractions. Advantages and disadvantages of each type are
listed in the table, and these are discussed in References 1, 9, 14 and other
standard texts dealing with separation processes. There are a number of
commercial suppliers of liquid-liquid extraction equipment and accessory
equipment such as that needed for the regeneration of solvent and removal of
solvent from the wastewater effluent.
7.5.4.2 Application—
Integration of equipment .into an overall system for successful treatment
of waste streams will require considerable analysis of the waste stream of
interest and the candidate processes. Liquid-liquid extractions are most
useful when separations involve materials that are not easily separated by
distillation or other treatment processes. Generally, liquid—liquid
extractions of water streams are conducted to remove materials which have high
water solubility and therefore almost invariably a low Henry's Law Constant.
Air or stream stripping do not appear to be viable options for wastes of this
type. Liquid-liquid extractions may be particularly applicable when the
relative volatilities of solute/solvent compounds make separation by
distillation difficult and high concentration levels make carbon adsorption
une c onoraica1.
Options with regard to choice of solvent and design parameters are many
and varied. Although design and operation of a liquid-liquid extraction
system to achieve acceptable effluent levels is theoretically possible,
existing experimental and field data indicate that most units, as now designed
and operated, fall short of this goal.
7-179
-------
lABtE 7,5.14. ADVANTAGES AND DISADVANTAGES OF EXTRACTION TYPES
Class of equipment
Advantages
Di sadvantage s
Mixer-Settlers
oo
o
Gravity Columns
Spray Column
Packed Column
Tray Column
Mechanically
Agitated Columns
Agitated Column
Pulsed Column
Centrifugal
Extractors
Reliable scaleup
Good contacting
Handles wide flow ratio
Low headroom
Many stages available
Low capital cost
Low operating cost
Simple construction
Handles wide flow ratio
(tray column)
Handles suspended solids
(spray column)
Good dispersion
Reasonable cost
Many stages possible
Relatively easy scaleup
Handles systems of high
interfacial tension
Handles low density difference
and high interfacial tension
between phases
Low holdup
Low space requirements
Small inventory of solvent
Handles stable emulsions
Large holdup
High power costs
High capital costs
Large floor space
Interstage pumping may be required
Extensive backmixing (spray, packed column)
Limited throughput with snail density difference
Cannot handle high flow rate (packed column)
High headroom
Low efficiency (spray column)
Difficult scaleup
Internals subject to fouling (packed column)
Limited throughput with small density difference
Cannot handle emulsifying systems
Cannot handle high flow ratio
High capital cost
High operating cost
High maintenance cost
Limited number of stages in single unit
Subject to fouling
Sources Reference 10, 12-14
-------
7.5.4.3 Environmental Impact—
Properly designed and operated, the liquid-liquid extraction process does
not appear to pose significant problems, primarily because both process exit
streams contain potential contaminants that must be addressed as part of the
process. The solvent will contain solute (contaminant in the feed) that must
be removed if the solvent is to function adequately in recycle. The treated
waste stream (assuming all significant traces of contaminant have been
transferred to the solvent) could in practice contain dissolved solvent which
may or may not be significant and warrant additional treatment. Because the
potential conditions are recognized and must be dealt with by system
designers, the environmental impacts of a viable liquid-liquid extraction
system should be minimal.
7.5.4.4 Advantages and Limitations—
Some potential advantages of liquid-liquid extraction processes are:
• Recovery of costly materials can be accomplished usually without
threat of thermal decomposition or chemical interaction.
• Recovery (separation) of materials which have similar relative
volatilities or adsorption isotherms can generally be accomplished
by liquid-liquid extraction.
Some potential limitations of liquid-liquid extraction are:
• Some residues will generally be present in both the raffinate and
extract streams, thus, some provision must be made for their removal
and subsequent disposal.
• Economics may not be favorable due to deviations from ideal behavior
which alter or limit the extent of removal under acceptable design
conditions.
7-181
-------
REFERENCES
1. Perry, R. H., et al. Editors; Chemical Engineers' Handbook, Sixth
edition, McGraw Hill Book Company, NY. NY. 1984.
2. Hwang, S. T., Treatability of Toxic Waste Water Pollutants by Solvent
Extraction. AIChE Symposium Series, 1981.
3. "Extraction of Chemical Pollutants from Industrial Wastewaters with
Volatile Solvents," U.S. EPA, Ada, Oklahoma, EPA-600/2-76-220, December
1976
4. Leo, A., C. Hansch, and D. Elkins, "Partition Coefficients and Their
Uses," Chemical Reviews, 71, 525(1971).
5. Murray, W. J., L. H. Hall, and L. B. Kier, "Molecular connectivity Ills
Relationship to Partition Coefficients," J. of Phar. Sci., 64, 1978(1975),
6. Tute, M.S., Principles and Practice of Hansch Analysis, Advances in Drug
Research, £, 1(1971).
7. Earhart, J. P., K. W. Won, H. Y. Wong, J. M. Prausnitz, and C. J. King,
"Recovery of Organic Pollutants via Solvent Extraction," Chem. Eng.
Prog., May, jtf (1977)
8. King, C. J., D. K. Joshi, and J. J. Senetar, University of California,
Berkeley, Department of Chemical Engineering. Equilibrium Distribution
Coefficients for Extraction of Organic Priority Pollutants From
Water-II. EPA-600/2-84-060b, February 1984.
9. Kirk-Othmer. Encyiopedia of Chemical Technology, Third Edition,
Volume 9. A Wiley-Intraseience Publication. 1978.
10. Earhart, J. P., et al. University of California, Berkeley Department of
Chemical Engineering. Extraction of Chemical Pollutants from Industrial
Wastewaters with Volatile Solvents. EPA 600/2-76-220. PB226241.
December, 1976
11. Ricker, N. L. and C. J. King. University of California, Berkeley
Department of Chemical Engineering. Solvent Extraction of Wastewaters
from Acetic Acid Manufacture. EPA-600/2-80-064. April 1980.
7-182
-------
12. Dryden, F. E., J. H. Mayes, R. J. Planchet, and C. H. Woodard. Priority
Pollutant Treatability Review. EPA Contract No. 68-03-2579, U.S.
Environmental Protection Agency, Cincinnati, Ohio, 1978.
13. Coco, J. H., et al. Development of Treatment and Control Technology for
Refractory Petrochemical Wastes. EPA-600/2-79-080, U.S. Environmental
Protection Agency, Ada, Oklahoma, 1979. 236 pp.
14. U.S. EPA Treatability Manual, Volume III, EPA-600/2-82-001a. September
1981.
7-183
-------
7.6 CARBON ADSORPTION
Adsorption is a widely-used process for the removal of organic
contaminants from gas or liquid waste streams. Activated carbon is the most
commonly used adsorbent. Largely nonpolar, it is a particularly effective
adsorbent for the removal of hydrophobic, high molecular weight organic
compounds from aqueous streams. However, it is also a good adsorbent for many
of the solvent and other low molecular weight organic compounds considered in
this document. Activated carbon adsorption must be considered a
potentially viable treatment technology for many solvent bearing waste
streams, either as a primary treatment for moderately high (up to 0.5 percent)
concentrations of organic compounds in an aqueous stream or as a secondary
1 2
polishing type treatment for much lower levels of contamination. ' The
cost effectiveness of adsorption is dependent on flow rates and concentrations
of the organic contaminants and on the adsorptive capacity of the carbon for
the contaminants. Adsorption should be cost effective for concentrations of
organic compounds up to about 1,000 mg/L, and could be cost effective for
concentrations up to 5,000 mg/L. Most likely for concentrations above
3
5,000 mg/L, another unit process may be more effective.
Activated carbon is available as a powder (PAC) or in the form of
granules (GAG). GAG is most commonly used because its larger size is most
amenable to handling in the equipment used to achieve contact and
4
regeneration. Both types of carbon adsorbent have large surface areas, far
in excess of their nominal external surface areas. Surface areas, resulting
from a network of internal pores 20 to 100 angstroms in diameter, are of the
order of 500 to 1,500 square meters per gram. Porosities can be as large as
80 percent. The characteristics of the micropore structure are largely
dependent on the activation process which is a controlled process of
dehydration, carbonization, and oxidation of raw materials including coal,
wood, peat, shell, bone, and petroleum based residues. The capacity of an
activated carbon for a contaminant is a function of the surface area and the
surface binding process and can approach one gram per gram of carbon.
Adsorbent binding forces result from the interaction of the surface
molecules with the field of force of the surface atoms. The attractive
forces, in the case of activated carbon, are generally weaker and less
7-184
-------
specific than those of chemical bonds and, hence, the term physical adsorption
is used to describe the binding mechanism. In agreement with the nature of
the forces, the effective range of their forces is small, and the adsorbed
material is generally present only as a monolayer upon the adsorbent surface.
The process is considered analagous to condensation of gas molecules, or to
crystallization from a liquid. The process is reversible, and molecules held
at the surface will subsequently return to the fluid stream with the length of
time elapsing between adsorption and desorption dependent upon the intensity
of the surface forces. Adsorption is a direct result of this time lag.
Water solubility and carbon affinity are two properties that, in general,
correlate with the adsorption of hazardous contaminants onto activated
carbon. Generally, less soluble organic materials are better adsorbed.
Several factors are associated with decreased water solubility of organics
and, as a result, correlate with increased adsorption: high molecular weight,
low polarity, low ionic character, low pH for organic acids or high pH for
inorganic bases, and aromatic structures. As a rule of thumb, molecules of
higher molecular weights are attracted more strongly to activated carbon than
are molecules of lower molecular weights. Strongly ionized or highly polar
compounds are more water soluble and thus usually poorly adsorbed.
Compounds with solubilities of less than 0.I g/mL in water and molecular
weights between 100 to 1,000 are considered moderately to highly adsorbable.
Several other aspects of molecular structure also affect adsorbability.
In general, branch-chain compounds are more adsorbable than straight-chain
compounds. Increasing hydrocarbon unsaturation also tends to decrease
solubility and increase carbon adsorption. Thus, unsaturated organics such as
ethylenes tend to more readily adsorb on carbon than saturated compounds, such
as ethanes. Table 7.6.1 identifies the specific waste characteristics that
affect adsorption; Table 7.6.2 summarizes the influence of substituent
chemical groups on adsorbability.
The adsorption of organic compounds by adsorbents is usually determined
in the laboratory through adsorption isotherm tests. These tests measure, at
a given temperature, the amount of substance adsorbed and its concentration in
the surrounding solution at equilibrium. Isotherms provide information on the
relative affinity of an organic compound for the adsorbent and the adsorption
7-185
-------
TABLE 7.6.1. WASTE CHARACTERISTICS THAT AFFECT ADSORPTION
BY ACTIVATED CARBON
A. General
1. Polar, low-molecular weight solvents and ignitables with high
degrees of solubility are poorly adsorbed.
2, Conversely, nonpolar, high-molecular weight solvents and ignitables
with limited solubility tend to be preferentially adsorbed.
B. Molecular Structure
1. Branched—chain compounds are more adsorbable than straight—chain
compounds.
2. Type and location of substituent groups also affect the degree to
which a compound may be adsorbed from solution. Table 7.6.2 gives
some general guidelines as to how substituent groups affect
adsorbability.
C. Effect of pH
1. The affect of pH on carbon solute equilibrium varies significantly
from compound to compound. Adsorption isotherms for some solvents
and ignitables are affected dramatically, whereas others show no
significant change as a function of pU.
2. Dissolved organics generally adsorb most efficiently at that pH
which imparts the least polarity to the molecule. For example, a
weak solvent can be expected to adsorb best at low pH value.
D. Temperature Effects
1. Adsorption reactions are generally exothermic^ therefore lower
temperatures favor adsorption. Although this makes physical sense,
little information has been found that documents significant shifts
in adsorbability within the range of temperatures normally
encountered in waste stream applications.
E. Physical Form
1. Carbon adsorption is suitable for aqueous wastes, nonaqueous liquids
and gases.
2. The oil and grease concentration should be less tnan 10 mg/L.
3. Suspended solids concentrations higher than about 10-70 mg/L will
cause clogging of the bed.
7-186
-------
TABLE 7.6.2. INFLUENCE OF SUBSTITUENT GROUPS ON ABSORBABILITY
Substituent
Nature of influence
Hydroxyl
Ami no
Carbonyl
Double bonds
Halogens
Sulfonic
Nitro
Aromatic rings
Generally reduces adsorbability; extent of decrease
depends on structure of host molecule.
Effect similar to that of hydroxyl but somewhat
greater; many amino acids are not adsorbed to any
appreciable extent.
Effect varies according to host molecule; glyoxylic is
more adsorbable than acetic but similar increase does
not occur when introduced into higher fatty acids.
Variable effect as with carbonyl.
Variable effect.
Usually decreases adsorbability.
Often increases adsorbability.
Greatly increase adsorbability.
Source: Reference 6
7-187
-------
capacity. Thus, isotherm tests can be useful in making qualitative
evaluations of different carbons for adsorption of specific components from a
given waste stream.
Isotherms data are frequently evaluated using the Freundlich Equation,
which describes the adsorbability characteristics of a constituent for a given
carbon. This equation can be expressed as follows:
m r
where: x = mass of adsorbate, mg
m =» mass of dry adsorbent, g
k = constant, adsorbability indicator
Cf * solution concentration at equilibrium, fflg/L
1/n m constant, adsorption intensity
Values of k and 1/n for a compound are found by a plot of experimentally
determined carbon adsorption data in which values of x/M are plotted against
C on log-log paper.
The k intercept is an indicator of adsorption capacity, and the slope 1/n
an indication of the change in capacity with concentration. A high k value
and a low 1/n value would be representative of a high capacity adsorption
system that is not strongly dependent upon contaminant concentration.
The adsorption data are useful in estimating the relative effectiveness
of an adsorbent for organic compounds. However, care must be exercised in
assessing performance when the waste stream contains a large number of
competing contaminants. Although it is possible to develop equilibrium
equations that apply to multi-component systems as noted in standard texts on
adsorption and Perry's Chemical Engineers' Handbook, most users will rely on
laboratory scale carbon adsorption/ isotherm tests to assess performance and
design an appropriate system for a specific waste stream.
The constants k and 1/n are summarized in Table 7.6.3 for compounds of
interest. These have been ranked in decreasing order of their k value as
determined by isotherm tests using Filtrasorb 300 activated carbon.
Table 7.6.4 shows similar data for activated carbon adsorption using 5 grams
of Westvaco WVG carbon per liter of solution.
7-188
-------
TABLE 7.6.3. CARBON ADSORPTION ISOTHERM DATA
Sources: References 7 and 8.
*Data for Adsorbent Filtersorb 300
l/n
Solvents of Concern
1, 2-dichlorobenzene
chlorobenzene
p-xylene
nit robenzene
ethylbenzene
tetrachloroethylene
methyl ethyl ketone
trichloroethylene
toluene
carbon tetraehloride
eye lohexanone
trie hlorofluorome thane
n-butyl alcohol (butanol)
1,1, 1-trichloroethane
ethyl acetate
methylene chloride
Other Compounds
acrolein
benzene
bromoform
1, 1, 2,2-tetrachloroethane
1, 3-dichloropropene
1, 2-dichloropropane
1, 1, 2-trichloroethane
1, 1-dichloroethene
1, 2-dichlorome thane
chloroform
ethyltdene diehloride
( 1, 1-dichloroethane)
hexachloroethane
129
91
85
68
53
50.8
36
28.0
26.1
11.1
6.2
5.6
5.5
2.48
2.4
1.3
1.2
1.79
19.6
10.6
8.21
5.86
5.81
4.91
3.6
1.0
1.79
96.5
0.43
0.99
0.19
0.43
0.79
0.56
0.211
0.62
0.44
0.83
0.75
0.24
0.50
0.34
0.78
1.16
0.65
0.53
0.52
0.37
0.46
0.60
0.60
0.54
0.8
1.6
0.53
0.38
7-189
-------
TABLE 7.6.4. ADSORPTION OF SOLVENTS AND OTHER ORGANICS BY ACTIVATED CARBON
I
I—"
o
Compound
Solvents of Concern
Methanol
Butanol
Isobutanol
Pyridine
Toluene
Ethyl benzene
Nitrobenzene
Ethyl acetate
Acetone
Methyl ethyl ketone
Methyl idobutyl ketone
Cyclohexanone
Other Solvents
Benzene
Aniline
Propylene glycol
Ethylene dichloride
Igni tables
Allyl alcohol
Formaldehyde
Acetaldehyde
Acrolein
Paraldehyde
Ethylenediamine
Ethyl acrylate
Acrylic acid
Molecular
weight
32.0
74.1
74.1
79.1
92.1
106.2
123.1
88.1
58.1
72.1
100.2
98.2
78.1
93.1
76.1
99.0
58.1
30.0
44.1
56.1
132.2
60.1
100.1
72.1
Aqueous
solubility
weight %
.*
7.7
8.5
a
0.047
0.02
0.19
8.7
a
26.8
1.9
•2.5
0.07
3.4
a
0.81
a
a
a
20.6
10.5
a
2.0
a
Concentration
(mg/liter)
Initial
1,000
1,000
1,000
1,000
317
115
1,023
1,000
1,000
1,000
1,000
1,000
416
1,000
1,000
1,000
1,010
1,000
1,000
1,000
1,000
1,000
1,015
1,000
Final
964
466
581
527
66
18
44
495
782
532
152
332
21
251
884
189
789
908
881
694
261
706
226
355
Adsorbability
g of compound/
g of carbon
0.007
0.107
0.084
0.095
0.050
0.019
0.196
0.100
0.043
0.094
0.169
0.134
0.080
0.150
0.024
0.163
0.024
0.018
0.022
0.061
0.148
0.062
0.157
0.129
Reduction
(%)
3.6
53.4
41.9
47.3
79.2
84.3
95.6
50.5
21.8
46.8
84.8
66.8
95.0
74.9
11.6
81.1
21.9
9.2
11.9
30.6
73.9
29.4
77.7
64.5
Source: Reference 9.
Test Conditions: 5 g of Westvaco WVG Carbon Per Liter of Solution.
-------
7.6,1 Process Description
A schematic of a carbon adsorption system utilizing a prefilter and a
multiple hearth furnace regeneration system is shown in Figure 7.6.1. In this
system, adsorbed material is driven from the carbon surface by thermal forces;
however, other methods (e.g., extraction or steam stripping) can be used to
drive off adsorbed material held largely by physical rather than chemical
forces. Regeneration is usually complete, although some loss of effective
2
surface area over time (3 to 8 percent per cycle) can result from build up
of hard to remove adsorbent, attrition, and other mechanisms. Collection or
destruction of the desorbed material will also be necessary for these
regeneration processes.
Carbon adsorption is applicable to single-phase aqueous solutions
containing low concentration of organic contaminant (up to 0.5 weight
3 5
percent) and inorganic contaminants (up to 0.1 weight percent). It is
also applicable to some organic liquid solutions (e.g., those consisting of a
poorly adsorbed solvent and a readily adsorbed solute), although it is less
likely that the selectivity will approach that for adsorption from a water
stream.
Carbon adsorption may be used as a pretreatment process for conventional
biological treatment, but is more frequently used as a polishing step for
biological treatment effluent to remove compounds that are resistant to
biodegradation. In this capacity, it is generally used for high volume waste
streams which contain dilute organic constituents.
7.6.1.1 Pretreatment Requirements—
Pretreatment of the feed to carbon adsorption columns is often required
to improve performance and/or prevent operational problems. As discussed in
Reference 10, there are four primary areas where pretreatment requirements for
different waste forms and characteristics may be required. These include;
1. Equilization of flow and concentrations of primary waste
constituents.
2. Filtration.
7-191
-------
Incoming
Process Waste
Waste Storage
Tank
Multt-Media
Praflltcr
Split |
Volatilization |
I
• Fugitive
• Emissions
I
Granular Activated
Carbon Columns
Discharged
EHkiant
Muftlple Hearth
Furnace/Carbon
Regeneration
S ^
/ Regenerated
( Carbon
I Storage
\ Ta*
1 \^^-^
• Furnace
1 Emlsakma
I
Figure 7.6.1. Carbon adsorption flow diagram.
SourceJ Reference 4.
-------
3. Adjustment of pH.
4. Adjustment of temperature.
Equalization of flow and concentrations of primary waste constituents—It
is generally assumed that both the flow to the GAG columns and the
concentration of the primary waste constituent, namely a solvent or ignitable
in the feed, are constant. Such is not generally the case, and since
variations in either flow or concentration can have a detrimental impact on
system performance, it is necessary to make provisions to equalize flow and
minimize concentration surges.
Flow equalization is accomplished by employing a surge tank of sufficient
capacity to accommodate flow variations. The result is a constant flow rate
to the GAG columns. Concentration equalization can be handled in the same
manner as flow equalization by employing surge tanks. However, provisions
must be made for mixing tank contents prior to discharging to the GAG
columns. Mixing prevents concentration surges which can lead to premature
column leakage and breakthrough or conversely, low concentration swings
resulting in premature regeneration of an underloaded GAG column.
Filtration-~It is a general requirement for GAG processes that the feed
of the column be low in suspended solids. In treating solvent and ignitable
waste streams, it has been suggested that solidly jgreater than 50 mg/L in
concentration will interfere with column operation. It is difficult,
however, to set an upper limit on the absolute level of suspended solids that
is acceptable. This is because the physical nature of the solids is as
important as their concentration. For example, finely divided, silty solids
tend to pass through the bed, but coarse material of varying particle size
could rapidly form a mat on top of the bed, constricting flow.
In addition to solids removal, many waste contaminants can interfere with
carbon adsorption of solvent and ignitable waste streams. For example, if
calcium or magnesium are present in concentrations greater than 500 mg/Lj
these constituents may precipitate out and plug or foul the column. _pil
and grease in excess of 10 mg/L can interfere in column operation. Lead
and mercury are also of concern because they may compete for adsorption sites
7-193
-------
and are difficult to remove from the carbon during the regeneration
cycle. The presence of many other compounds can influence adsorption as
they compete for available adsorption sites on the carbon surface.
For efficient use of GAG for treating solvent waste streams, removal of
suspended solids and other waste contaminants noted above must be achieved by
pretreatment with, for example, multi-media pressure filters. Such filters
are very compatible with fixed bed adsorption processes and can be readily
integrated into a. total design. Other possibilities include membrane
filtration when a highly clarified feed is desired; ultrafiltration if high
molecular weight contaminants were encountered (over 1,000); and reverse
osmosis could be used to concentrate a feed containing numerous dissolved
species both organic and inorganic. Obviously other pretreatments will be
needed to remove dissolved solids such as calcium and magnesium.
Adjustment of pH--GAC adsorption systems are sensitive to changes in pH.
If the contaminants to be removed are either weakly acidic or weakly basic,
then the pH of the feed will effect their adsorption. Weakly acidic organics
such as cresols or cyclohexanone are most readily adsorbed in the nonionized
state and consequently a low pH (acid) favors adsorption. Weakly basic
compounds such as aniline or dimethylamine are also most readily adsorbed in
their nonionized state and, therefore, adsorption is favored by high pH
(alkaline). The adsorption of neutral organic compounds is unaffected by pH.
The control of the feed pH should perhaps be considered a subcategory of
the previously discussed concentration/equalization requirement. It can be
readily controlled by applying pH measurement and feedback control for acid or
base addition to the equalization system at the surge tank to achieve the
desired pH feed to the GAC adsorption columns.
Adjustment of Temperature—The adjustment of temperature is rarely
required in GAC adsorption for solvent and ignitable processes. High feed
temperature could lead to increased ?OG emissions to air in an open gravity
feed system and is unfavorable for adsorption and retention of the volatile
constituents. If the possibility for temperature surges exists, temperature
moderation through flow equalization should be considered.
7-194
-------
7.6.1.2 Operating Parameters—
Process design activities must take into account a number of equipment
design parameters to develop a system which is optimal for the characteristics
of the waste or wastes to be treated. The design parameters will be
considered in terms of both the adsorption system and the regeneration or
desorption system.
Adsorption System—Isotherms, determined in a laboratory, measure the
affinity of activated carbon for the "target" adsorbates in the process
liquid. This provides data for determining the amount of carbon which will be
required to treat the full scale process stream. Carbon requirements will be
based on a limiting constituent for which attainment of effluent limitations
is the most difficult. However, adsorption isotherms can vary widely for
different carbons, and isotherm data cannot be used interchangeably.
Table 7.6.5 gives properties of some commercially available granulated
activated carbons. Properties of a typical powdered activated carbon are
shown in Table 7.6.6. Adsorption properties of the two types of carbon are
generally comparable. The principal difference is in the particle size; the
fine size of the PAC makes it unsuitable for use in the contacting and
regeneration equipment used for GAG applications.
A typical continuous adsorption system consists of multiple columns
filled with activated carbon and arranged in either parallel or series. Total
carbon depth of the system must accommodate the "adsorption wavefront";
i.e., the carbon depth must'be sufficient to purify a solution to required
specifications after equilibrium has been established. Bed depths of
8-40 feet are common. Minimum recommended height-to-diameter ratio of a
column is 2:1. Ratios greater than 2:1 will improve removal efficiency but
result in increased pressure drop for the same flow rate. Optimum flow rate
must be determined in the laboratory for the specific design and carbon used.
For most applications, 0.5 to 5 gpm per square foot of carbon is common.
Various configurations are available for GAG adsorption applications.
Based on influent characteristics, flow rate, size and type of carbon,
effluent criteria and economics, each design offers uniqueness in its mode of
operation. Figure 7.6.2 illustrates several arrangements typically used for
GAG adsorption systems. There are two basic modes of operation for column
7-195
-------
TABLE 7.6.5. PROPERTIES OF SEVERAL COMMERCIALLY AVAILABLE CARBONS
PHYSICAL PIOPERTIES
Surface area, m^/g (BET)
Apparent density, g/cm-*
Density, backwashed and drained, lb/g3
Seal density, g/cur
Particle density, g/cm3
Effective aize, mo
Uniformity coefficient
fore volume, enr/g
Mean particle diameter, mm
SPECIFICATIONS
Sieve size (U.S. std. series)3
Larger than No. 8 (max. X)
Larger than No. 12 (max. %)
Smaller than No. 30 (max. Z)
Smaller Chan No. 40 (max. %)
Iodine Mo.
Abrasion Ho. minimus
Ash {%)
Moisture as packed (max. %)
ICI
America
Hydrodarco
(lignite)
600 - 650
0.43
22
2.0
1.4 - 1.5
0.8 - 0.9
1.7
0.95
1.6
8
—
5
—
650
b
b
b
Calgon
Filtrasorb
300
(bituminous)
950 - 1050
0.48
26
2.1
1.3 - 1.4
0.8 - 0.9
1.9 or less
0.85
1.5 - 1.7
8
—
5
—
900
70
8
2
Westvaco
Nuehar
WV-L
(bituminous)
1000
0.48
26
2.1
1.4
0.85 - 1.05
1.8 or less
0.85
1.5 - 1.7
8
- —
5
—
950
70
7.5
2
Witco
517
(12x30)
(bituminous)
1050
0.48
30
2.1
0.92
0.89
1.44
0.60
1.2
—
5
5
—
1000
85
0.5
1
*0thcr sizes of carbon are available on request from the manufacturers.
"No available data from the manufacturer.
—— NoC applicable Co this'size carbon.
TYPICAL PROPERTIES OF 8 X 30-MESH CARBOHS
Total surface area, m^/g
Iodine number, min
Bulk density, lb/ft3 bacfcwashed and drained
Particle density wetted in water, g/cm3
Pore volume, cm-Vg
Effective size, mm
Uniformity coefficient
Mean particle dia. , mm
Pittsburgh abrasion number
Moisture as packed, max.
Molasses KS (Relative efficiency)
Ash
Mean-pore radius
Lignite
carbon
600 - 650
500
22
1.3 - 1.4
1.0
0.75 - 0.90
1.9 or less
1.5
50 - 60
92
100 - 120
12 - 18%
33 A
Bituminous
coal carbon
950 - 1,050
950
26
1,3 - 1.4
0.85
0.8 - 0.9
1.9 or less
1.6
70 - 80
2%
40 - 60
5-8%
14 A
Source: Reference 9
7-196
-------
TABLE 7.6.6. TYPICAL PROPERTIES OF POWDERED ACTIVATED CARBON (PETROLEUM BASE)
Surface Area m2/g(BET) 2,300-2,600
Iodine No. 2,700 - 3,300
Methylene Blue Adsorption (mg/g) 400 - 600
Phenol No. 10 - 12
Total Organic Carbon Index (TOCI) 400 - 800
Pore Distribution (Radius Angstrom) 15 - 60
Average Pore Size (Radius Angstrom) 20-30
Cumulative Pore Volume (cm3/g) 0.1 - 0.4
Bulk Density (g/cm3) 0.27-0.32
Particle Size Passes: 100 mesh (wt%) 97 - 100
200 mesh (wt%) 93 - 98
325 mesh (wt%) 85 - 95
Ash (wt%) 1.5
Water Solubles (wt%) 1.0
pH of Carbon 8-9
Source: Reference 6.
7-197
-------
in
out
r
UPFLOW IN SERIES
"
in'
UPFLOW IN PARALLEL
in
i
DOWNFLOW IN SERIES
1
out
1
out
-*»-
DOWNFLOW IN PARALLEL
out
MOVING
BED
Figure 7.6.2. CARBON BED CONFIGURATIONS
Sources Reference 12.
7-198
-------
adsorbers; namely, fixed beds and moving or pulsed beds. In the fixed bed
modet the entire bed is removed from service when the carbon is to be
reactivated. In the moving or pulsed bed, only the exhausted (inlet) portion
of the total bed is removed as an increment of new adsorbent is added
simultaneously.
Arrangement in series permits the first column to become saturated with
impurities while a solution of required purity is obtained from the final
column. At this point, the first column is emptied and refilled with fresh
carbon or regenerated. Fluid flow is redirected so that the column is now
placed in the downstream position, affording a form of countercurrent use of
the carbon.
The adsorption beds of both series and parallel design can be operated in
either an upflow or downflow direction. A downflow mode of operation must be
used where the GAG is relied upon to perform the dual role of adsorption and
filtration. Although lower capital costs can be realized by eliminating
pretreatment filters, more efficient and frequent backwashing of the adsorbers
is required. Application rates of 2-10 gallons per minute per square foot
2 2
(gpm/ft ) are employed, and backwash rates 6"f 12-20 gpm/ft are required
to achieve bed expansions of 20—50 percent. The use of a supplemental air
scour can be used to increase efficiency of the backwashing.
In the upflow-expanded mode, while prefiltration is normally required to
prevent blinding the beds with solids, smaller particle sizes of GAG can be
employed to increase adsorption rate and decrease adsorber size. Application
rates can be increased in the upflow-expanded mode even to the extent that the
adsorbent may be in an expanded condition.
The design arrangements offer the following advantages and limitations
noted in Reference 12.
Method Comments
Adsorbers in Parallel - For high volume applications
- Can handle higher than average suspended
solids (<65—70 ppm) if downflow
- Relatively low capital costs
— Effluents from several columns blended,
therefore, less suitable where effluent
limitations are low
7-199
-------
Method Comments
Adsorbers in Series - Large volume systems
— Easy to monitor breakthrough at tap
between units
- Effluent concentrations relatively low
- Can handle higher than average suspended
solids (<65-70 ppm) if downflow
- Capital costs higher than for parallel
systems
Moving Bed - Countercurrent carbon use (most
efficient use of carbon)
- Suspended solids must be low (<10 ppm)
- Best for smaller volume systems
- Capital and operating costs relatively
high
- Can use such beds in parallel or series
Upflow-expanded - Can handle high suspended solids (they
are allowed to pass through)
- High flows in bed (~1S gpm/ft^)
The above systems are not generally used with the much finer powdered
activated carbons. The PAC systems now used involve mixing the PAC with the
waste stream to form a slurry which usually can be separated later by methods
such as filtration or sedimentation. PAC is generally used with biological
13
treatment to enhance organic removal by biological processes.
Regeneration—The success of an economical adsorption system usually
depends on the regenerability of the adsorbent. The exception is where there
are very long adsorption or loading cycles due to very low concentrations of
solvent constituents in the inlet feed; this type of system usually operates
on a "throw away" basis". If very large quantities of adsorbent are involved,
then regeneration and reuse are required. The regeneration techniques
employed in industry are thermal regeneration, steam regeneration, and acid or
base regeneration . Solvent washing or biological treatment are other
methods that are occassionally used for regeneration. The moat commonly
applied regeneration technique for GAC systems is thermal.
Thermal regeneration involves high temperatures and a controlled gaseous
atmosphere. The regeneration of the spent carbon can be considered to take
place in three distinct phases. First, wet carbon is dried at a temperature
7-200
-------
range of approximately 100-150°C. Water and some low boiling organics will be
removed during this process but higher boiling solvents such as nitrobenzene
(b.p. 210.8°C), cresols (b.p. 201.9°C), and 1,2-dichlorobenzene (b.p. 173°C)
will remain. Next, the temperature is raised to 250 to 750°C where these more
tightly bonded solvents and higher boiling organics are removed by
vaporization. An inert gas atmosphere can be employed to minimize oxidation.
Finally, the temperature is raised to 800 to 975°G where residues and tars
that may have accumulated are reacted and driven off the carbon surface.
Steam is sometimes used to assist removal. Even with careful control, GAC
losses are reported to be 3 to 8 percent per cycle due to both oxidation and
mechanical attrition. Regeneration furnaces have been designed to conduct all
three steps of drying, vaporization under inert gases, and regeneration
separately in different zones. Multiple hearth furnaces and fluidized-bed
furnaces are two types of thermal regenerators commonly found in commercial
use.
Steam regeneration can be used to displace the liquid in the adsorber
bed, heat up the adsorbent and, finally, strip off the solvents and ignitables
from the GAC. The solvents and the other low molecular weight organics of
concern are volatile enough to permit regeneration with steam., Average
pressures of one to three atmospheres are utilized with steam flow rates of
3
1/2 to 4 Ibs/min/ft . The amount of steam required depends upon the size of
the carbon bed. The majority of steam used in regeneration is used to heat
the carbon bed to the necessary temperature for vaporization to occur. The
heat capacities of the adsorbed constituents and their heats of vaporization
do not represent a large fraction of the total steam requirement. Thus units
for steam usage should be in Ib steam/Ib carbon or equivalent.
As discussed in the pretreatment section, the adsorption of weak organic
acids and bases from aqueous solutions is very dependent upon pH. Therefore,
if the adsorbed organic is acidic, regeneration with a basic solution is
feasible. Conversely, basic constituents can be regenerated with an acidic
solution. Acid or base regeneration is not as widely used as other
regeneration techniques, but nonetheless, some solvents such as cresols and
ethylene diamine have been successfully recovered commercially, by base and
acid regeneration, respectively.
7-201
-------
7.6.1.3 Post-Treatment Requirements—
Air and water discharges from carbon adsorption systems employing carbon
regeneration can be relatively innocuous. Under proper design and operating
conditions, the treated water will generally be suitable for discharge to
surface waters. Other aqueous streams such as backwash, carbon wash and
transport waters are recycled or sent to a settling basin. Emissions will
result from thermal reactivation, but when afterburners and scrubbers are
used, the controlled emissions are essentially non-polluting. In some
installations, particulates must be removed from the air stream (e.g., via a
cyclone and baghouse) resulting in a solid waste.
7.6.1.4 Treatment Combinations—
The high cost associated with the treatment of moderate to high total
organic carbon (TOG) wastes and the ineffectiveness of carbon as an adsorbent
for many low molecular weight water soluble organic compounds has impacted the
use of carbon adsorption as a waste treatment technology. Except when used
alone as a polishing step for low levels of adsorbable materials in aqueous
streams, carbon adsorption is usually employed in a "treatment train" with
other treatment processes to achieve maximum efficiency at reduced cost.
The following treatment train combinations illustrate the use of
activated carbon systems with other technologies to optimize performance. The
process trains shown do not represent the only possible configurations. They
do, however, represent the treatment technologies which, when combined with
activated carbon, are expected to have broadest range of applicability and
effectiveness. These also have been demonstrated to some degree for treatment
of hazardous aqueous solvent and other organic containing waste streams. The
examples have been taken from material presented in Reference 14 and in other
material cited herein.
Treatment Train One
Figure 7.6.3 illustrates a sequence of solids removal and biological
treatment followed by granular carbon adsorption for polishing of the liquid
effluent. This train is applicable to treatment of wastewaters high in TOG
and low in toxic (to a biomass) organics.
7-202
-------
pH adjustment
I
NJ
O
X' V
chemicals
COAGULATION
Influent A
backwash
GRANULAR
ACTIVATED
CARBON
effluent
SETTLING
sludge
FILTRATION
off gases
BIOLOGICAL
V sludge
Ywaste sludge
Figure 7.6.3. Schematic of biological/carbon adsorption treatment train.
Source? Reference 14.
-------
Biological treatment such as activated sludge, rotating biological
contactors, or anaerobic filters are used to reduce BOD as well as
biodegradable toxic organics. This reduces the organics load to subsequent
treatment (adsorption) processes. To prevent rapid head losses caused by
accumulation of solids in the adsorption columns, clarification and
multi-media filtration are provided following biological treatment to reduce
suspended solids to 25-50 mg/L. Granular carbon adsorption is then used to
polish liquid effluents to remove nonbiodegradable and toxic organics.
This process train is expected to be highly cost effective. Its success,
however, is dependent on biological system performance. Moreover, the
presence of high concentration of volatile organic constituents may create a
potential air contamination problem during biodegradation. Three by-product
wastes are produced: chemical sludge, biological sludge, and spent carbon.
Spent carbon can be regenerated but the sludges must be subjected to further
treatment prior to disposal.
Because the process is intended to handle multi-component waste streams,
pollutant recovery for reuse is unlikely. The only potential for such
recovery is during carbon regeneration if materials can be desorbed by steam
or solvent washing. This would be reasonable only if a small number of
separable compounds were sorbed on the carbon.
Treatment Train 2
The flowsheet depicted in Figure 7.6.4 employs the same unit processes as
in Figure 7.6.3, but granular carbon is positioned ahead of the biological
treatment system. This process train, which also is applicable to high TOG
wastewaters, is used when highly adsorbable waste stream components may be
toxic to biological cultures. The rationale is to utilize the activated
carbon to protect the biological system from toxicity problems. Therefore,
the carbon is allowed to pass relatively high concentrations of biologically
nontoxic organics and selectively adsorb only those constituents which impact
negatively on the subsequent biological process. The train is appropriate
only for wastes with appropriate chemical configuration and adsorption
characteristics and may be impractical for many wastes.
7-204
-------
O
Ul
pH
\'
chemicals
COAGULATION
SETTLING
influent A
backwash I
I
ffluent F 1
1
L I
FILTRATION
(optional)
sludge
backwash
SETTLING
FILTRATION
V
1
off gases
\ I sludge
BIOLPQICAL
waste sludge
\f
GRANULAR
ACTIVATED
CARBON
Figure 7.6.4. Schematic of carbon adsorption/biological treatment train.
Source: Reference 14.
-------
In this configuration, the chemical coagulation step (including settling
and filtration) plays a role both in soluble inorganics removal and in
particulate removal to minimize head losses in GAC columns. As with the
treatment train above there is little potential for recovery of pollutants.
Treatment train 3
The third process train, illustrated in Figure 7.6.5, utilizes
biophysical treatment which is a combination of biological and powdered
activated carbon treatments (PACT ) conducted simultaneously. This
approach is simpler than the previously described sequential carbon-biological
treatments and has the potential of achieving comparable effluent quality.
Potential advantages include the use of less costly powdered carbon and
minimization of the physical facilities required. Spent carbon-biological
sludge can be regenerated or dewatered and disposed directly. However, if the
latter approach is considered, it is necessary to include additional costs for
raw carbon purchase and disposal of toxics-laden carbon when making economic
comparisons. Further information concerning the use of PAC can be found in
other references, including References 14 and 15.
Treatment Train 4
A processing system consisting of a stripping unit and a carbon
adsorption system is illustrated in Figure 7.6._6_^ This configuration will be
applicable primarily to organic wastewaters, although provision for chemical
coagulation and removal of inorganics is provided. This treatment train is
suited to situations involving volatile and nonbiodegradable toxic organics.
It is especially pertinent if a single or limited number of volatile compounds
are present which can be recovered from the overhead condensate stream (if
steam stripping is used).
Aside from pH adjustment prior to stripping, little pretreatment is
necessary, other than filtration to remove solids to prevent the build-up in
the column stripping unit.
7-206
-------
influent , \
O
-4
pH adjustment
chemicals
COAGULATION
I
( Jaackwaah __
I
effluent
•\
( --- 1
FILTRATION
(optional)
powdered activated
carbon
SETTLING
V
sludge
Aoff gases
BIOLOGICAL
SETTLING
Y_A
, sludge
waste sludge
Figure 7.6.5. Schematife of biophysical treatment train.
Source: Reference 14.
-------
I
NJ
O
oo
pH adjustment
Influent y
FILTRATION
(optional)
A overhead condensate
^1
LTJ
backwaajil
bottoms
\'
AIR/STEAM
STRIPPING
chemicals
COAGULATION
SETTLING
FILTRATION
->
sludge \ /
backwash
GRANULAR
ACTIVATED
CARBON
effluent
Figure 7,6.6. Schematic of stripping/carbon adsorption treatment traia.
Source: Reference 14.
-------
In addition to carbon treatment effluent, this process train generates
three waste streams: overhead condensate, chemical sludge, and spent carbon.
Assuming that carbon will be regenerated, either onsite or by a commercial
service, the two remaining streams require additional treatment and/or
disposal. Preferably, the organic phase of the overhead condensate can be
recovered and reused, with the water phase returned to the treatment system.
However, if this is not possible, incineration is the best method for
condensate disposal. Chemical sludge should be dewatered and disposed by a
method commensurate with the materials contained in the sludge.
7*6.2 Demonstrated Performance
Information gathered from activated-carbon manufacturers and industry
indicates that many granular-activated carbon systems are being used for the
treatment of hazardous aqueous solvent and ignitable bearing wastes and
wastewaters. One IP A study , performed in 1982, found that over 100 GAG
systems were being used nationally to treat industrial wastewaters. Another
13
report documented the use of PAC at seven and four facilities in the
United States and Japan, respectively.
Despite the large number of units in use, data for full scale
applications are incomplete for two primary reasons. For many applications,
essential operating parameters or pollutant removal characteristics have not
been generated or are considered to be proprietary information. The available
information does, however, contain material gathered from a wide variety of
data sources, including carbon manufacturers, industrial users of activated
carbon, the available literature, and EPA files.
A major shortcoming of the available data base dealing with the removal
of solvents and other low molecular weight organics from aqueous waste streams
by activated carbon, is the sparsity of performance data for the higher
concentration (>0.1 percent) levels. In addition, most of the data found in
the literature do not consider the removal of individual compounds from
concentrated waste streams, although data for BOD, COD, TOG, and other
parameters are fairly common. Data for individual compounds provided in EPA's
7-209
-------
background* document for solvents (Reference 5) are, with few exceptions, for
treatment of influent concentrations at the part per billion level. All data
presented, however, do indicate that levels acceptable for direct discharge
can be reached for essentially all solvents of concern. Thus, the utility of
adsorption as a treatment process hinges on the economics of the specific
situation.
Data taken from References 6 and 8 have been summarized in Tables 7.6.7
and 7.6.8. These data provide results of bench/pilot and full scale GAC
systems, respectively. Because of the sparsity of information concerning
system design and operating conditions, including carbon loading, no attempt
has been made to include such information in the tables. However, more
details can be found in the cited references, e.g., Reference 8.
Data are similarly sparse for systems using PAC, although PAG and GAC
should exhibit little, if any, difference in adsorption performance. As noted
previously, the most significant difference between the two sorbents is in
their particle size. The fine particle size of PAC is not suitable for use in
contacting equipment normally used for GAC systems.
The addition of powdered activated carbon to the aeration basin of a
biological activated sludge system combines physical adsorption with
biological oxidation and assimilation. 'It has been shown to be particularly
effective in treating waste streams which are variable in concentration and
composition. The characteristic advantages of the addition of PAC over
conventional activated sludge are:
1. Higher BOD and COD removal.
2. Stability of operation with variability in influent concentration
and composition.
3. Enhanced removal of toxic substances and priority pollutants. *•->
A performance summary for a full scale study is shown in Table 7.6,9. In
this study PAC was added to the sludge and the mixture was then fed to three
activated sludge units operating in parallel. This was followed by two
clarifiers also operating in parallel. The sludge from the clarifiers was
sent to a multiple hearth furnace for carbon reactivation. The influent
7-210
-------
TABLE 7.6.7. DATA REPORTED FROM BENCH SCALE AND PILOT PLANT GAC SYSTEMS
Concentration
(mg/L)
Constituent
Acetone
Chloroform
1 , 1-dichloroe thane
1 , 2-dichloroe thane
1 , 2-dichlorpropane
Ethanol
1,1,1, trichloroe thane
Influent
60
34
11
92
12
59
40
26
91
64
2,000
1,000
450
510
1,120
1,220
16
28
7
1,440
5
12
Effluent
6
0
ND
ND
ND
4
0
ND
ND
9
12
190
0
ND
150
330
0
ND
ND
140
ND
ND
nemo va i
90
100
>99
>99
>99
93
100
>99
>99
86
99.4
81
100
>99
87
73
100
>99
>99
90
>99
>99
Reference
6
6
8
8
8
6
6
8
8
8
6
6
6
8
8
8
6
8
8
6
8
8
7-211
-------
TABLE 7.6.8. DATA REPORTED FROM FULL SCALE GAG SYSTEMS
Concentration
(mg/L)
Constituent
t
f
Acetone
Benzene
Carbon tetrachloride
Chloroform
Cresol
1, 1-dichloroethylene
Methylene chloride
Tetrachloroethylene
Toluene
Trichloroethylene
Trichloroethane
Influent
10
10
590
590
4.3
190.
70
16.5
5,000
130
108
38
32
100
—
2,500
—
2,500
12
2,000
2,500
Effluent
1
1
210
210
1.8
136
18
0.6
350
ND
28
3
—
32
—
630
—
630
5
4.4
1.9
Removal
90
90
64
64
58
29
74
96.2
93
99
73
92
78
68
99.7
75
99
75
58
99
99
Reference
6
6
6
6
6
6
8
6
6
8
6
8
8
8
6
6
6
6
8
8
8
7-212
-------
TABLE 7.6.9. DATA REPORTED FROM FULL SCALE PAG SYSTEM
Concentration
(mg/L)
Constituent
Benzene
Carbon tetrachloride
Chlordbenzene
Chloroform
1 , 2-dichlorobenzene
Ethyl benzene
Methyl chloride
Nitrobenzene
Tetrachloroethylene
Toluene
1,1, 1-trichloroe thane
Trichloroethylene
Tr iehlorofluorome thane
BOD
COD
Influent
0.105
0.094
1.720
0.201
0.259
0.041
1.770
0.454
0.024
0.519
0.013
0.041
0.155
169
596
Effluent
0.0009
0.0014
0.030
0.021
0.120
0.0017
ND
ND
0.0017
0.0047
0.0006
0.0019
0.0030
5.3
10.9
rercenc
removal
99
99
98
89
54
96
>99
>99
93
99
95
95
98
97
98
Source: Reference 6, IT Enviroscience
System Operating Parameters:
Flow rate: 151,372 m3/d
Carbon Dosage: 134 ppm
Aerator Detention Time: 8.4 hrs
Sludge Age: 38 days
Regeneration Technique: Multiple Hearth
7-213
-------
concentrations were low because this study was from a contaminated municipal
wastewater treatment system; however, the results are promising, showing the
high removal possible with the combination of PAC and activated sludge.
7*6«3 Cost of Carbon Absorption
The cost of carbon adsorption treatment can be described in terms of
capital investment, and operation and maintenance costs. Capital costs
consist of direct and indirect expenses. For the small scale system, direct
capital investment costs include the purchase of a waste storage tank, a
prefilter, carbon columns, a waste feed pump, piping and installation. For
the large scale system, direct capital investment costs include the purchase
of the components in the small scale system plus storage tanks for spent and
regenerated carbon, a multiple hearth furnace and automatic controls.
A model has been developed by IGF, Inc. (Reference 17) for calculating
carbon adsorption costs. Table 7.6.10 contains the equations used in this
model to calculate direct capital costs as a function of carbon consumption
rate and storage volume. Indirect capital costs include the costs of
engineering, construction, contractor's fee, startup expenses, spare parts
inventory, interest during construction, contingency and working capital.
These costs are expressed as percentages of either direct capital costs or the
sum of direct and indirect capital costs as summarized in Table 7.6.11.
Direct and indirect capital costs are assumed to be incurred in year zero.
Operation and maintenance costs also consist of direct and indirect
costs. Direct operation and maintenance costs (in 1984 dollars) include the
operating labor and electricity and carbon consumption. Table 7.6.10 also
contains the equations used in the model to calculate direct operation and
maintenance costs. As with the capital costs, the model considers operation
and maintenance costs for carbon consumption rates less and greater than
400 Ibs/day. For large scale systems, the operation and maintenance costs
also include the natural gas consumption necessary for the furnace. Indirect
operation and maintenance costs include the costs for insurance and the system
overhead.
7-214
-------
TABLE 7.6.10. DIRECT COSTS FOR CARBON ADSORPTION8
Carbon Consumption Direct Capital Direct Operation and
Rate Costs Maintenance Cost^
(Ibs/day) (*) (t/yr)
less than 400 l,256(c)-603 + I40(s)'54 29(c)-6 + 350(c)(cp) +
6l9(e)'168(h) + 5(c)(p)
greater than 400 I4,23l(c)*522 + 140(s)'54 58(c)«657 + 35(c)(cp) +
25,012*383(e)(p)
1.49 I06(c)(f)
where: c ** carbon consumption rate in pounds per day
s = storage volume in. gallons
cp - carbon price in dollars per pound (|>0.8/lb)
h =" hourly wage rate in dollars per hour C$l4.56/hr)
p = power price in. dollars per kilowatt-hour (t0.05/KWh)
f = fuel price (natural gas) in dollars per Btu ($6xlO~6/Btu)
aCost estimates were developed for three model treatment systems (three
small scale and three large scale systems). The cost estimates for these
systems were then used to develop a cost equation In the form of a power
curve.
"The power requirement Is derived from the equipment specifications.
SOURCE: Reference 17
7-215
-------
TABLE 7.6.11. INDIRECT COSTS FOR CARBON ADSORPTION
Item
Indirect Capital Costs
Engineering and
Supervision
Construction and
Field Expenses
Contractors Fee
Startup Expenses
Spare Parts
Inventory
Interest During
Construction
Contingency
Working Capital
Indirect Operation and
Maintenance Costs
Insurance, Taxes,
General
Administration
System Overhead
Percent
of direct
capital
costs
12
10
7
5
2
10
0
0
0
0
Percent of the
sum of direct
and indirect
capital costs
0
0
0
0
0
0
15
18
5
5
Percent
of total
annual costa
0
0
0
0
0
0
0
0
0
10
aThe total annual cost is defined as the sum of the total capital cost
multiplied by the capital recovery factor and the total operation and
maintenance costs.
7-216
-------
Based on the RCRA Risk-Cost Analysis Model, Table 7.6.12 shows carbon
adsorption costs for 100, 400, 1,000 and 2,500 gal/hr processes.
7.6.4 Overall Status of Process
7.6.4.1 Availability—
Activated carbon adsorption is a widely used technology for treating
waste streams containing organic compounds, including many hazardous organic
solvents and other low molecular weight organic compounds. Its ability to
treat solvents and other organics has been demonstrated at bench, pilot, and
full scale levels by many firms. Manufacturers of activated carbon produce a
carbon to fit just about any service need. Companies who use these activated
carbon systems, both GAG and PAG, are numerous and may be found in several
literature sources (see References 6, 8, 14, and 18). Equipment designers and
suppliers can be found in the Chemical Engineering Equipment Buyers* Guide
published by McGraw-Hill, New York, NY.
7.6.4.2 Application—
Activated carbon adsorption systems are widely used in industry to
process chemical product streams as well as waste streams. The technology has
proven to be effective as a pretreatment for aqueous wastes prior to their
introduction into biological treatment systems. Concurrent treatment of waste
streams with PAG and biological treatment has also proven to be effective.
However, the most common application of carbon adsorption systems would appear
to be as a polishing step for low concentration level effluents from other
treatment technologies. The use of carbon adsorption systems for treatment of
wastes containing 0.5 percent or greater organic concentration levels is not
considered to be cost effective. Other technologies should be considered at
these concentrations.
Removal efficiencies which permit direct discharge can usually be met by
GAG systems for most organic solvents and other low molecular weight
organics. However, performance will depend upon the specifics of waste stream
contamination, including the need for pretreatment, post-treatment, and other
aspects of system operation.
7-217
-------
TABLE 7.6.12. CARBON ADSORPTION COSTS3
Quantity processed
(gal/hr)
100 400 1,000 2,500
Capital Expenditures
Capital Cost Including Installation"
($1,000) 59 561 904 1,462
Annual Operation and Maintenance ($l,000)c
Energy
Labor
Carbon
Other
Capital Recovery
Total Annual Cost
Cost per 1,000 gald
2
23
7e
1
10
42e
210e
11
35
27
5
99
177
221
27
53
67
10
160
317
159
68
80
168
18
259
593
119
aCosts are based on the RCRA RISK-COST ANALYSIS MODEL.17
^Capital costs for the 100 gal/hr system include waste storage tank,
prefilter, carbon columns, waste feed pump, piping and installation; the
other flow levels (400, 1,000, 2,500) include these units plus storage
tanks for spent and regenerated carbon, a multiple hearth furnace and
automatic controls.
GThese costs are based on the following data:
carbon price = $0.8/lb
hourly wage rage = fcl4.56/hr
power price = $0.05/kwh
fuel price (natural gas) = $6 x 10~*VBtu
capital recovery factor = 0.177
"Unit costs are based on 2000 hours of operation per year.
eModified to reflect a direct relationship between carbon requirement and
quantity processed.
*Note: 1984 dollars, prices are similar to 1986 values.
7-218
-------
Although the adsorbability of a contaminant/carbon combination does not
provide all the information needed to assess the potential applicability of
activated carbon to a specific waste stream, it is the prime determinant.
Table 7.6.13 illustrates a treatability rating system for priority pollutants
based on carbon adsorptability. The table shows that polar, low molecular
weight solvent and ignitables with high solubilities (e.g., methyiene
chloride) are poorly adsorbed on carbon. Conversely, nonpolar, high molecular
weight solvents and ignitables with low solubilities (generally less than
0,1 g/mL; e.g., chlorobenzene) were found to be preferentially adsorbed. Data
such as that shown in Table 7.6.13 obtained from isotherms, manufacturers'
literature, and existing data make it possible to predict performance with
some degree of accuracy for compounds of interest provided the characteristics
of the stream can be determined and variations are not extreme.
7.6.4.3 Environmental Impacts—
Environmental impacts can result from emissions during the regeneration
of carbon. However, there will be no serious environmental impacts if the
exit gases are treated by a control system, e.g., an afterburner and/or
scrubber, and in some cases, a particulate filter. Where the carbon is
chemically regenerated (acid, base, or solvent), the regeneration stream will
require future treatment, e.g. incineration or distillation to remove the
organic contaminants.
The recovery or reuse of desorbed solutes from the adsorption process is
an area where opportunities could exist for both cost savings and reduction of
environmental impacts. Disposal of desorbed solutes as waste materials can be
costly and also result in an environmental hazard. Therefore,recycling of
solute following desorption and recovery should be considered and practiced if
possible.
7-219
-------
TABLE 7.6.13. REMOVAL EATINGS FOR ORGANIC COMPOUNDS
Compound
PRIORITY SOLVEHTS
acetone
butanol
carbon disulfide
carbon tetrachloride
chlorobenzene
creaolc
cyclohexanone
1 ,2-dichlorobenzene
ethyl acetate
ethyl ether
ethyl benzene
iaobutyl alcohol
oethanol
OTHER SOLVEHTS
aeeconitxile
aniline
benzene
bis(chloromethyl)ether
bromofortn
chlorofont
cyclohexane
Br dichlorobenzene
dichlorodifluoromethane
1 , 1-dichloroethy lene
1 , 2-d ichloroethy lene
1,2-dichloropropane
1 , 3-dichloropropane
1 ,4-d ichloro-Z-butene
1,4-dioxane
IGHITABtES
aero Ic in
acrylic acid
allyl alcohol
allyl chloride
2,2-bioxirane
chlor*cet«ldehyde
chloroiaethyl methyl ether
chloroprene
cunene
ditaethylamine
dipropylaraine
epichlorohydrin
ethanol
ethyl acrylsee
Gthylenediamine
ethylenemine
Removal rating
P
G
G
G
E
E
G
E
F
P
G
P
P
F
E
F
G
E
P
F
E
F
P
F
G
G
F
P
P
F
P
P
F
F
F
F
E
P
F
P
P
F
P
P
Compound
Methyl ethyl ketone
methyl isobutyl ketone
methylene chloride
nitrobenzene
pyridine
tetrachloroethylene
toluene
1,1,1-triehloroethane
trichloroethene
trichloromonof luororae thane
1,1,2-trichloro-
1 ,2 ,2-trif luoroethane
xylene
2-ethoxy ethanol
ethyl carbonate
ethylene dichloride
ethylidene dichloride
furan
furfural
hexachloroe thane
2-nitropropane
2-picoline
propylene glycol
1,1,1, 2-tet rachloroethane
I, 1 ,2 ,2-tetrachloroethane
tetrahydro furan
1,1, 2-trichloroethane
ethylraethacrylate
formaldehyde
glycidylaldehyde
me thac ry lonit rile
2-raethylaziridine
methyl bromide
1-uethyl butadiene
methyl chloride
methyl chlorocarbonate
methyl iscyanate
methyl methacrylate
oxirane
paraldehyde
propylamine
thionjethanol
Removal rating
G
F
P
E
E
E
G
G
F
G
G
E
P
G
G
F
F
G
E
F
E
F
E
E
F
G
P
P
F
P
P
P
P
P
P
P
P
G
F
P
P
Source: Reference 3.
Key: E • Excellent Removal - Adsorbs at levels 100 mg/g carbon at Cf » 10 mg/L
Adsorbs at levels 100 mg/g carbon at Cf 1.0 mg/L
G • Qpod Removal - Adsorbs at levels 100 ag/g carbon at Cf » 10 mg/L
Adsorbs at levels 100 rag/g carbon at Cf 1.0 mg/t.
F " Fair Removal - Adsorbs at levels 100'mg/g carbon at Cf = 10 tng/L
Adsorbs at levels 100 mg/g carbon at Cf 1.0 mg/L
P • Poor Removal — Essentially no removal-
Cf " Final concentration of pollutant at equilibrium
7-220
-------
REFERENCES
1. Rizzo, J. L. Calgon Carbon Corporation, Letter to Paul Frillici, GCA.
May 5, 1986.
2. Berkowitz, J. B. et. al. Physical, Chemical and Biological Treatment
Techniques for Industrial Waste. Noyes Data Corporation; Park Ridge, New
Jersey. 1978
3. Rizzo, J.L. Calgon Carbon Corporation. Letter to Paul Frillici, GCA.
June 11, 1986
4. ICF Inc. Survey of Selected Firms in the Commercial Hazardous Waste
Management Industry: 1984 update. Final report to U.S. EPA,
Section II. OSW Washington, DC. 1985.
5. U.S. EPA Background Document for Solvents to Support 40 CFR Part 268,
Land Disposal Restrictions, Volume II, January 1986.
6. IT Enviroscience, Incorporated. Survey of Industrial Applications of
Aqueous-Phase Activated-Carbon Adsorption for Control of Pollutant
Compounds from Manufacture of Organic Compounds, Prepared for U.S. EPA
IERL. EPA-600/2-83-034. PB-83-200-188. April 1983
7. Dobbs, R.A. and Jim Cohen, Carbon Adsorption Isotherms for Toxic
Organics. EPA-600/8-80-023. April 1980.
8. U.S. EPA. Treatability Manual, Volume III. EPA-600/2-82-001a, U.S. EPA
ORD, Washington, DC. 1981.
9. U.S. EPA. Activated Carbon Treatment of Industrial Wastewater-Selected
Papers. EPA-600/2-79-177. Robert S. Kerr Environmental Research
Laboratory. August 1979.
10. Slejko, F. L., Applied Adsorption Technology, Chemical Industry Series
Volume 19. Marcel Dekker, Inc. NY, NY. December 1985.
11. Perrich, J. R. ed., Activated Carbon Adsorption for Wastewater
Treatment. CRC Press Inc., Boca Raton, Florida. 1982.
12. Lyman, W. J. Carbon Adsorption, In: Unit Operations for Treatment of
Hazardous Industrial Wastes. Pollution Technology Review No. 47, Noyes
Data Corporation, Park Ridge, New Jersey. 1978.
7-221
-------
13. Heidi, J. A. Zimpro Inc. PAC Process. Engineering and Management.
June 1982.
14. Touhill, Shuekrow and Associates, Inc. Concentration Technologies for
Hazardous Aqueous Waste Treatment. Pittsburg, PA. EPA-6002-81-019.
15. Button, D. G. "Removal of Priority Pollutants by the DuPont PACT1*1
Process." Proceedings of the 7th Annual Industrial Pollution Conference,
Philadelphia, PA, June 5-7, 1979.
16. U.S. EPA. Development Document for Effluent Limitation Guidelines and
Standards for Petroleum Refining Point Source Category.
EPA-440/1-82-014. October 1982.
17. ICF, Inc. RCRA Risk-Cost Analysis Model, Phase III U.S. EPA, OSW.
March 1984.
18. Radian Corporation. Full-Scale Carbon Adsorption Applications Study.
EPA-600/2-85-012. May 1984.
7-222
-------
7.7 RESIN ADSORPTION
Eesin adsorption is an alternative treatment technology for the removal
of organic contaminants from aqueous waste streams. The underlying principle
of operation is similar to that for carbon adsorption; organic molecules
contacting the resin surface are held on the surface by physical forces and
subsequently removed during the resin regeneration cycle. Resin adsorbents
can be made from a variety of monomeric compounds which differ in their
polarity and thus, their affinity for different types of compounds. The
choice of resin type, in combination with modifications in pore structure, can
lead to an adsorbent tailored specifically for effective removal of special
classes of compounds, e.g., low molecular weight, polar organics. For
example, hydrophobic resins such as those prepared from styrene - divinyl
benzene monomers, are most effective for nonpolar organics and bonding is
largely the result of Van der Waal's forces; acrylic based resins on the other
hand are more polar and dipole-dipole interactions may play the major role in
the binding of polar molecules to the resin surface. The general concept is
that like molecules attract. Polar resins will attract polar organics;
nonpolar compounds will be attracted by the more hydrophobic or nonpolar
1
resins.
A significant aspect of resin adsorption is that the attractive forces
are usually weaker than those encountered in granulated activated carbon (GAG)
adsorption . Regeneration can be accomplished by simple, nondestructive means
such as solvent washing, thus providing the potential for solute recovery.
Thermal regeneration (generally not possible with resin adsorbents because of
their temperature sensitivity) is usually required for carbon adsorbents
eliminating the possibility of solute recovery. The resins differ in many
respects from activated carbon adsorbents. In addition to differences in the
ease and usual methods of regeneration associated with the chemical nature of
the two adsorbents, there are significant differences in shape, size, porosity
and surface area. Resin adsorbents are generally spherical in shape rather
than granular, and are smaller in size and lower in porosity and surface area
than GAG adsorbents. Surface areas for resins are generally in the range of
2 21
100-700 m /g, as opposed to 800-1,200 m /g for activated carbon.
7-223
-------
Adsorptive capacities are thus less for the resin adsorbents, although the
chemical nature and the pore structure of the resin can be tailored to enhance
the selectivity of the resin and, therefore, its adsorption capacity for
specific organic components.
Properties of several typical resin adsorbents are shown in Table 7.7.1.
A more detailed list of physical properties is provided in Table 7.7.2 for a
specialty resin provided for experimental study by Rohm and Haas Company,
Ambersorb XE-340. The resin was designed initially by Rohm and Haas for the
selective adsorption of chloroform. Selectivity was based on controlled pore
structure adjustment. Macropores are adjusted such that low molecular weight
organics (e.g., chloroform), can rapidly diffuse into the interior of the
resin (micropores) for adsorption, but high molecular weight compounds and
collodial or bacterial matter are inhibited. The size of the pores protects
the active sites from exposure to other materials that would normally be
competitive for adsorption and would, thus, reduce the selectivity of
2
adsorption and possibly complicate solute recovery. Selectivity based on
chemical structures and pore dimensions is not totally exclusive, however.
Ambersorb XE-340, for example, although originally tailored for the adsorption
of chloroform, is also a good adsorbent for most other low molecular weight
2
organics, e.g., carbon tetrachloride and trichloroethylene.
Other notable properties of resin adsorbents include their nondusting
characteristics, their low ash content, and their resistance to bacterial
growth. The last characteristic is primarily a result of the fine pore
3
structure which inhibits bacterial intrusion.
Another significant difference between resin and carbon adsorbents is
their cost. Resin adsorbents are much more expensive. They generally will
not be competitive with carbon for the treatment of waste streams containing a
number of contaminants with no recovery value. However, resin adsorption
should be considered if material recovery is practical, selectivity is
possible, and for cases where carbon regeneration is not effective. Like
carbon adsorption systems, resin adsorption can produce an effluent with low
levels of contaminant concentrations, particularly in cases where contaminants
are well characterized and few in number. Resin adsorption combined with
carbon adsorption may be effective for certain waste streams containing a
number of contaminants.
7-224
-------
TABLE 7.7.1. PHYSICAL PROPERTIES OF ADSORBENTS
Manufacturer Adsorbent
Rohm and Haas Amberlite XAD-2
Amberlite XAD-4
Amberlite XAD-7
Amberlite XAD-8
Ambersorb XE-347
Ambersorb XE-348
Mitsubishi Diaion HP-10
Diaion HP-20
Diaion HP-30
Chemical
nature
Polystyrene
Polystyrene
Acrylic Ester
Acrylic Ester
Polymer Carbon
Polymer Carbon
Polystyrene
Polystyrene
Polystyrene
Pore
volume
0.68
0.96
0.97
0.82
0.41
0.58
0.64
1.16
0.87
Surface
area
(m2/g)
300
725
450
160
350
500
500
720
570
Pore
diameter
average
(I)
100
50
85
150
200, I5a
200, 15a
b
70
b
Surface
polarity
Low
Low
Intermediate
Intermediate
Low
Intermediate
Low
Low
Low
aAverage pore diameter of the macropores and micropores, respectively.
"Average pore diameter not available.
Source: Reference 2.
-------
TABLE 7.7.2. TYPICAL PHYSICAL PROPERTIES OF AMBERSORB XE-340
Appearance
Black, spherical, nondusting
polymer carbon
Total surface area (N2, BET method), m2/g
Bulk density, lb/ft3
Bulk density, g/cm3
Particle density, g/eir* (Hg displacement)
Skeletal density, g/cm^ (He displacement)
Pore volume, cm^/g
Particle size (U.S. Sieve Series)
Crush strength, kg/particle
Ash content, percent
Macropore diameter, A
Hicropore diameter, A
Surface polarity
400
37
0.60
0.92
1.34
0.34
20-50
3.0
0.5
100-350
6-40
Very low
Source: leference 2.
7-226
-------
7.7.1 Process Description
Resin adsorption systems are designed and operated in similar fashion to
GAG systems. A principal difference will be in the regeneration step;
regeneration of the resin is usually performed in situ with aqueous solutions
or solvents. Solute recovery from the regeneration liquor will also be
required, with distillation the most likely method.
7.7.1.1 Pretreatment Requirements—
Polymeric adsorbents require pretreatment of feed streams to remove
suspended solids, oils and greases, and to adjust pH and temperatures, as
appropriate. Suspended solids in the influent should be less than 50 mg/L
and, in the case of oil and grease, less than 10 mg/L to prevent clogging of
the resin bed. The control of pH may be necessary to prevent resin attack
and to enhance adsorbability. Low temperature will also generally enhance
adsorption. Resin adsorbents, although generally resistant to chemical attack
because of their cross-linked structure, should not be brought into contact
with compounds such as chemical oxidants and functional reagents which may
degrade the resin or poison adsorption sites. High levels of dissolved
solids, particularly inorganic salts, do not compete with organics for
adsorption sites, and their presence may in some instances increase the
adsorption of organics.
Pretreatment options are similar to those proposed previously for carbon
adsorption systems. For example, filtration or coagulation/sedimentation type
separations can be used for suspended solids, and flotation/extraction
procedures can be used for removal of oils and greases. Each pretreatment
option will result in a residual which may or may not require additional
processing prior to disposal.
There are no definite limitations on the upper or lower contaminant
concentration levels that can be treated. An upper limit of 8 percent (for
phenol) is suggested in Reference 1, however, to maintain cycle time and
regeneration frequency within reasonable limits. As with carbon adsorption,
the efficiency of resin adsorption (weight of adsorbed material per weight of
adsorbent) is greater at high concentrations.
7-227
-------
7.7.1.2 Operating Parameters—
The design of a resin adsorption system requires the development of basic
information such as feed stream flow rate, contaminant concentration, and
adsorbent type and capacity. Other information such as flow rate variations,
suspended solid level, pH, and temperature will be required to ensure that
adequate pretreatment precautions and operating practices are followed.
The choice of adsorbent type can be guided by the concept that attractive
forces will be greatest for similar molecules. The similarity concept is also
useful in identifying regeneration solvents. The solubility of the adsorbate
in the regeneration solvent is quite important. The solvent not only must be
capable of overcoming the attractive forces of adsorbate/adsorbent but must
also remove the adsorbate in the smallest possible volume.
Although the relative strengths of the attractive forces between solute,
solvent, and resin can be predicted through the use of solubility
4
parameters, there is no practical method for determining the actual
capacity of an adsorbent for contaminants, particularly those existing in
complex waste streams. It is, therefore, necessary to carry out experimental
studies to determine working capacities for candidate adsorbents. In most
cases it will be possible to design systems to achieve EPA treatment standard
levels, although it may not be possible for complex mixtures without use of
multiple adsorbents (or technologies). Costs may also be prohibitive, and
activated carbon may often be a more attractive adsorbent, particularly where
solute recovery is not desirable or practical.
Assuming a resin adsorbent can be found that can achieve required
treatment levels, additional tests will be required to identify and select a
regeneration process. The selection of a regeneration solvent can be guided
by use of solubility parameters. However, other factors such as cost of
solvent regeneration and adsorbate recovery must be considered. Distillation
appears to be the most likely solvent and solute recovery technology assuming
a solvent/solute match can be found that is amenable to such a separation
process.
Design of a resin adsorption process operation would include the
following steps as a general procedure: 1) determine wastewater effluent
7-228
-------
purity desired, 2) select adsorbent and determine adsorption capacity,
3) select regeneration process based on bench or pilot scale tests, 4) size
adsorbent bed, 5) check loading run length and determine if it is compatible
with the regeneration time cycle, 6) repeat 4 and 5 until loading and
regeneration cycles are compatible, 7) determine bed dimensions by hydraulic
considerations, 8) design and size pumps, storage tanks, pretreatment
equipment and auxiliary equipment.
As noted in Reference 1, a system for treating low volume waste streams
will commonly consist of two beds. One bed will be on stream while the second
is being regenerated as shown in Figure 7.7.1,
The adsorption bed is usually fed downflow at flow rates in the range of
0.25 to 2 gpm per cubic foot of resin; this is equivalent to 2-16 bed
volumes/hr, and thus contact times are in the range of 3-30 minutes. Linear
2
flow rates are in the range of 1-10 gpm/ft . Adsorption is stopped when the
bed is fully loaded and/or the concentration in the effluent rises above a
certain level. As noted in the previous section on carbon adsorption, a
contact time of 30 minutes may not be adequate for attainment of treatment
standards. EPA has suggested that limited contact times may play an important
role in reducing column loadings in the field to values less than those
predicted from isotherm testing. Reference was made to a study which
attributed carbon contact times of greater than 230 minutes to applications
which requires high degree of pollutant removal. Although rapid adsorption
kinetics are attributed to resin adsorbents, caution should be exercised in
assessing the contacting time requirements and design and operating features
needed to meet EPA treatment standard concentration levels.
Regeneration of the resin bed is performed in situ with basic, acidic,
and salt solutions or recoverable nonaqueous solvents being most commonly
used, Basic solutions may be used for the removal of weakly acidic solutes
and acidic solutions for the removal of weakly basic solutes; hot water or
steam could be used for volatile solutes; and methanol and acetone are often
used for the removal of nonionic organic solutes. A prerinse and/or a
postrinse with water will be required in some cases to remove certain
contaminants such as salts. As a rule, about three bed volumes of regenerant
will be required for resin regeneration; as little as one-and-a-half bed
volumes may suffice in certain applications.
7-229
-------
PHENOLIC
WASTE
fOlYMEBie
AOSORIER
«t
TREATED
DlCTIUATIOn
COLUMN ml
JOLVENTATtTERffHEPfOL
t
SEPARATaR
PHEHOIAVATER AZEDTROPt
IRiqrtd l» A^ttrkm! Ctlnl)
BiSTIUAftOK*
coiuMn a
PHEKOL/WATER
I
McevtxtB
fHHBLtllXl
70 X
Column #3 utilized when highly pure phenol is required.
Mmrul B*ianee. Ibt/hr
1 ©
264
Water
Acetone
Tom
21.736
22,000
@
1480
1480
©
*
4
©
264
9
273
(D * ©
< 10 pom
23,207
4
^ 23.211
Figure 7.7.1. Phenol removal and recovery system - solvent
regeneration of Amberlite adsorbent.
Source: Reference 1.
7-230
-------
The use of steam as the regenerating agent should be considered; steam
regeneration for volatile organics may provide some cost benefits in that it
can reduce the need for subsequent treatment to separate the waste solvent
from the dissolved organics. However, the condensed steam may also require
additional treatment prior to discharge to also eliminate dissolved organics.
When using steam regeneration for polymeric adsorbents, one must consider
the upper temperature limit of the resin in choosing the steam pressure. The
styrene based polymeric adsorbents are usually stable to 200°C; acrylic based
resins up to 150°C. Since the adsorbed solvent and other organic constituents
can cause the adsorbent resin matrix to swell and weaken, removal of these
constituents by steaming could result in disruption and breakup of the resin
matrix. Therefore, adsorbent stability is of concern when using steam
regeneration and should be studied using multi-cycling tests to confirm the
integrity of the adsorbent before proceeding with design of the regeneration
system.
Steam requirements are normally significantly lower for the polymeric
adsorbents than those for granular activated carbon to achieve a certain
desorption level of a given constituent. The reason for this is that the
attractive forces binding the solvent or other organic constituent to the
adsorbent are much lower for the polymeric adsorbent.
7.7.1.3 Post-Treatment Requirement—
Assuming effluent goals are realized, the post-treatment requirements are
restricted to treatment .of the regeneration effluent. Other possible waste
streams requiring further processing could include the washing effluents (if
required for the prerinse and/or postrinse of the resin), the regeneration
solvent containing the organics removed from the feed stream, and the
condensed regeneration steam containing dissolved organics. Requirements will
depend upon the process scheme used.
7.7.1.4 Treatment Combinations—
Resin adsorption will normally be given consideration in applications for
which carbon adsorption would be considered as a potentially viable treatment
7-231
-------
alternative. However, it will not generally be economically competitive with
carbon adsorption. In certain situations a combination of resin and carbon
adsorption could be used to advantage. For example it may be attractive as a
polishing step to remove specific contaminants (particularly if the
contaminants have recovery value) passing a carbon adsorption bed, e.g., polar,
low molecular weight compounds.
7.7.2 DemonstratedPerformance
Resin adsorption technology is not as established as activated carbon
adsorption is for full scale treatment of waste streams containing organic
solvents and other low molecular weight organic contaminants. Studies have
been conducted to determine the performance of resins as adsorbents for
several types of organic chemical compounds. The results of one such study
(Reference 7) are as follows:
1. Alcohols — Polymeric resins have provided complete removal of
several'alcohols at lOOyg/L concentrations.
2. Aliphatics - The adsorption of several chemical groups by polymeric
resins was, in order of decreasing adsorbability, aldehydes and
ketones, alcohols, chlorinated aromatics, aromatics, amines, and
chlorinated alkenes and alkanes. Adsorption of aliphatics ranged
from 25-100 percent. All but the chlorinated alkanes and
chlorinated alkenes were removed more readily by resin than by
activated carbon.
3. Amines - Complete adsorption of amines at 100 yg/L concentrations
was reported. Resin adsorbents were more effective than activated
carbon adsorbents.
4. Aromatics - A polymeric resin, Amberlite XAD-2, was found to adsorb
aromatics more effectively than did activated carbon.
These results were the basis for Table 7.7.3 which identifies those solvent
and ignitable organic compounds that have been readily addressed by polymeric
resins. These results are not all inclusive and should only be used as a
basis for further study.
7-232
-------
TABLE 7.7.3. CHEMICAL COMPOUNDS TREATED BY RESIN ADSORPTION
Priority solvents
Other solvents
Ignitables
Aldehydes
and Ketones
Alcohols
r
N»
to
Chlorinated
Aromatics
Aromatics
Esters
Acetone
Cyclohexanone
Methyl ethyl ketone
Methyl isobutyl ketone
1-Butanol
Cresols
Isobutyl alcohol
Methanol
Chlorobenzene
1,2-Dichlorobenzene
Ethylbenzene
Nitrobenzene
Pyridine
Toluene
Xylene
Ethyl acetate
Ethyl ether
2-Ethoxy ethanol
Propylene glycol
o-Dichlorobenzene
Benzene
Cyclohexane
1,4-Dioxane
2-Picoline
Bis(chloromethy1)ether
Ethyl carbamate
Furan
Furfural
Tetrahydrofuran
Chloroacetaldehyde
Ethanol
Formaldehyde
Glycidylaldehyde
Paraldehyde
Allyl alcohol
Thio methanol
Cumene
Acroleln
Acrylic acid
2-2'-Bioxirane
Chloronethyl methyl ether
Epichlorohydrin
Ethyl aerylate
Ethyl methacrylate
Methyl chlorocarbonate
Methyl isocyanate
Methyl methacrylate
Oxirane
(continued)
-------
TABLE 7.7.3 (continued)
Priority solvents
Other solvents
Ignitables
Amines
Chlorinated
Alkanes and
Alkenes
r
Miscellaneous
Methylene chloride
Tetrachloroethylene
1,1,1-Triehloroethane
1,1,2-Trichloro-
1,2,2-tri-fluoroethane
Trichloroethene
Trichloromonofluoromethane
Carbon disulfide
Carbon tetrachloride
Acetonitrlle
Aniline
2-Nitropropane
Dichlorodifluoromethane
1,1-Dichloroethylene
1,2-Dichloroethylene
1,2-Dichloropropane
1,3-Dichloropropene
1,4-Dichloro-2-butene
Ethylene dichloride
Ethylidene dichloride
Hexachloroethane
1,1,1,2-Tetrachloroethane
1,1,2,2-Tetraehloroethane
1,1,2-Trichloroethane
Bromoform
Chloroform
Dimethylamine
Dipropylamine
Ethylene diamine
Ethylenimine
Methacrylonitrile
2-Methyl aziridine
n-Propylamine
Chloroprene
Allyl chloride
Methyl bromide
1-Methylbutadiene
Methyl chloride
Source: References 7 through 10.
-------
A U.S. EPA study examined the adsorption capacity of Atnbersorb XE-340 and
compared these data to that of granular activated carbon adsorption.
Table 7.7.4 shows the adsorption capacity of Ambersorb XE-340. The capacities
of the resin adsorbent were generally higher than those shown by the carbon
adsorbent for these constituents under similar test conditions.
Table 7.7.5 shows the adsorption capacity of Ambersorb XE-340 for several
solvents and concentration ranges. The results of a pilot study investigating
treatment methods for ground water contaminated by chlorinated solvents,
namely 1,2-dichloroethylene, trichloroethylene, and tetrachloroethylene are
shown in Table 7.7.6. A direct comparison can be made between
Ambersorb XE-340 and two carbons from ICI and Calgon. Figure 7.7.2,
graphically illustrates the removal of the organics from the influent in the
pilot plant investigation to yield a low concentration effluent.
7.7.3 Cost ofResin Adsorption
Eesin adsorbents are quite expensive (Table 7.7.7), The cost exceeds
that of granular activated carbon (GAG) ($0.80 to $1.00 per pound). However,
the economics of using resins or polymeric adsorbents may in certain cases be
more favorable than those for granular activated carbon.
Thermal regeneration costs for GAG adsorption systems are quite high and
carbon losses are of the order of 3 to 9 percent per regeneration. Even
though macroreticular (resin) adsorbents cost more per pound, they are
relatively cheaper to regenerate and regeneration does not result in any
appreciable adsorbent loss. Thus, smaller beds and more frequent
regenerations may be economically viable with resin adsorbents.
Design criteria for a one million gallon per day treatment plant are
shown in Table 7.7.8. Assuming influent concentrations of 300-1,000 ppb, the
operation is designed to remove greater than 90 percent of the incoming
contaminant. A comparable GAG system is analyzed simultaneously for
comparison. The capital and operating costs for each system are given in
Table 7.7.9. It can be seen that both the capital investment and the
operating costs are lower when the more expensive (by volume) adsorbent is
used. This comes about primarily because fewer and smaller contactors are
utilized and expensive thermal regeneration furnaces are not required.
7-235
-------
1ABLE 7.7.4. ADSORPTION OF SOLVENTS BY AMBERSORB XE-340
•-j
N3
10
OH
Average
concentration, Bed depth,
yg/L m (ft)
Triehloroethylene
Tetrachloroethylene
1,1, 1-Trichloroethane
Carbon Tetraehloride
215
210
210
177
4
3
41
51
65
70
94
1,400
3
2
5
33
237
23
1
19
19
0.3 (1)
0.6 (2)
1.2 (4)
0.8 (2.5)
0.8 (2,5)
0.2 (0.8)
0.3 (1)
0.6 (2)
1.2 (4)
0.3 (1)
0.8 (2.5)
0.8 (2.5)
0.8 (2.5)
0.8 (2.5)
1.2 (4)
0.8 (2.5)
0.2 (0.8)
0.2 (0.8)
0.8 (2.5)
0.8 (2.5)
0.8 (2.5)
Empty bed
contact time,
minutes
2
4
7.5
9
8.5
5
2
4
7.5
2
5
9
8.5
9
7.5
9
5
5
9
5
10
Loading to 0.1 yg/L
breakthrough,
m3/m3a
83,700b
78,600b
53,300b
> 20, 160
> 123, 340
>117tOOO
>99,900b
78,600b
>53,300b
106,000b
112,900
17,920
> 123, 340
>20,160
39,300b
56,000
82,600
100,800
20,160
7,560
15,120
(continued)
-------
TABLE 7.7.4 (continued)
Average
concentration, Bed
i
Nl
w
Cis-1 ,2-Dichloroethylene
1, 1-Dichloroethylene
Wg/L
40
38
40
40
25
22
16
6
2
122
4
0
0
1
0
0
0
0
0
0
0
0
m
.3
.6
.2
.3
.8
.8
.8
.8
.8
.2
.2
depth,
(ft)
(1)
(2)
(4)
(1)
(2.
(2.
(2.
(2.
(2.
(0.
(0.
5)
5)
5)
5)
5)
8)
8)
Empty bed
contact time,
minutes
2
4
7.5
2
6
6
6
9
8.5
5
5
Loading to 0.1 yg/L
breakthrough ,
m3/ '
37
39
19
36
14
7
11
20
mJa
,200b
,500b
,700b
,400b
,400
,200
,500
,160
59,000 but 123,340
80
110
,600
,800
am sater/nr* carbon.
bBreakthrough defined by shape of wavefront curve; generally 20 to 25 yg/1 of contaminant in
adsorbent effluent.
Source: U.S. EPA. Treatment of Volatile Organic Compounds in Drinking Water, EPA-600/8-83-019.
May 1983.
-------
TABLE 7.7.5. REMOVAL OF CHLORINATED SOLVENTS BY AMBERSORB XE-340
Compound
Carbon Tetraehloride
Chloroform
1 , 1-Dichloroethy lene
1 » 2-Dichloroethylene
1, 2-Dichloroethylene
1, 2-Dichloroethylene
Tetrachloroethylene
Tetrachloroetliylene
1, 1, 1-Trichloroe thane
1,1, 1-Trichloroethane
Trlchloroethylene
Irichloroethylene
Trlchloroethylene
Concentration
range
(yg/L)
nf
64
41
5
<1
3
60
597
172
11
1
1
120
- 87a
- 95
- 74
- 28
- 4
- 9
- 205
- 2,500
- 286
- 214
- 2
- 10
- 276
Flow rate
(gpm/ft3)
1.5
1.5
1.5
1.2
0.9
0.8
1.5
0.88
1.5
0.9
1.5
0.9
0.8
EBClk
(mln)
5
5
5
6.2
8.5
9
5
8.5
5
8.5
5
8.5
9
BVC to 10% TOC
leakage (mg/L)
14,976
6,048
>42,000 0.5
4,645
>58,108
20,160 0.5
>42,000 0.5
17,788
>42,000 0.5
>58,100
>42,000 0.5
>58,100
>20,100 0.5
Not found.
* Empty Bed Contact Time.
CBV •* Bed Volume (m^ water/m3 carbon).
Source: Eeference 10.
7-238
-------
TABLE 7.7.6. FILOT PLANT ADSORPTION SUMMARY
NS
35,300
17,300
38,400
>49,000
3,000
17,000
>21,000
27,000
105,000
121,000
38,400
46,000
46,000
(4)
(17)
(6)
(4)
(15)
(9)
(4)
(17)
(6)
(15)
(12)
(5)
(17)
(7)
(7)
(13)
(17)
Days
40
67
92
13
37
92
45
100
130
8
44
55
35
137
158
50
60
60
(continued)
-------
TABLE 7.7.6 (continued)
N3
Column
4
4
4
5
5
5
5
5
5
5
5
5
5
5
5
Adsorbent
Ambersorb
Ambersorb
Ambersorb
Ambersorb
Ambersorb
Ambersorb
Ambersorb
Ambersorb
Ambersorb
Ambersorb
Ambersorb
Ambersorb
Ambersorb
Ambersorb
Ambersorb
XE-340
XE-340
XE-340
XE-340
XE-340
XE-340
XE-340
XE-340
XE-340
This
XE-340
XE-340
XE-340
XE-340
XE-340
XE-340
Cycle
number
1
1
1
1
1
1
2
2
2
Column was
3
3
3
4
4
4
Flow rate
4
4
4
2
2
2
2
2
2
gpm/ft3
gpm/ft3
gpm/ft3
gpm/ft3
gpm/ft3
gpm/ft3
Steam
gpm/ft3
gpm/ft3
gpm/ft3
Compound
1, 2-Dichloroethylene
Trichloroethylene
Tetrachloroethylene
1, 2-Dichloroethylene
Trichloroethylene
Tetrachloroethylene
Regeneration
1 , 2-Dichloroethylene
Trichloroethylene
Tetrachloroethylene
Bed volumes
to 10% leakage
(leakage ppb)
si.
69,
>77,
40,
80,
83,
23,
23,
23,
000
000
000
000
000
000
000
000
000
(7)
(13)
(7)
(4)
(20)
(6)
(6)
(17)
(9)
Days
40
90
100
104
208
216
60
60
60
stopped prematurely and then Steam Regenerated
4
4
4
4
4
4
gpm/ft3
gpm/ft3
gpm/ft3
f"! *_ _ —
gpm/ft3
gpm/ft3
gpm/ftj
1, 2-Dichloroethylene
Trichloroethylene
Tetrachloroethylene
1 , 2-Dichloroethylene
Trichloroethylene
Tetrachloroethylene
15,
69,
77,
28,
64,
73,
000
000
000
000
000
000
(4)
(17)
(6)
(5)
(16)
(11)
20
90
100
36
83
95
-------
2.5 LITERS OF AMBERSORB XE-340
O INFLUENT
D EFFLUENT
3-CHl-atO-D O-D—D-r-O
70
BED VOLUMES (THOUSANDS)
Figure 7.7.2. Performance of resin adsorption bed.
80
90
100
-------
fABLE 7.7.7. COST OF ADSORBENTS8»b
Adsorbent
Amber lite XAD-2
Amberlite XAD-4
Amberlite XAD-7
Amberlite XAD-8
Chemical nature
Polystyrene
Polystyrene
Acrylic ester
Acrylic ester
Cost fc/ft3
282.95
355.05
223.25
337.25
aPersonal communication with Rohm and Haas Company,
Fluid Process Chemicals Department, Philadelphia, PA,
April 3, 1986.
bAt a Bulk Density of 37 lbs/ft3, costs are roughly
$6 to $10 per pound.
7-242
-------
TABLE 7.7.8. DESIGN CRITERIA—TRIHALOMETHANE REMOVAL
Parameter
Adsorbent
Ambersorb XE-340
Granular activated
carbon
Density
Nominal Flowrate
Contactors
On-stream Time
Regeneration
Type
Time/Contactor
Absorbent
Lifetime
37 lb/ft3
6.0 gpm/ft3
1.25-min EBCTa
58 ft3 each
2 on-stream
1 regeneration/standby
3.3 days
in-place steam
8 hr
5 yr
(fouling limited)
25 lb/ft3
1.0 gpm/ft3
7.48-min EBCTa
348 ft3 each
2 on-stream
1 regeneration
1 standby
20 days
thermal reactivation
11 days
8 months
@ 8% loss/cycle
Design Basis: 1.0 mgd average flow, 1.43 ragd peak flow.
aEmpty bed contact time.
Source: References 11 and 12.
7-243
-------
TABLE 7.7.9. COST COMPARISON—GAG VS. RESINa
1.0 mgd Plant
Capital Cost
Contactor, Pumps,
Regeneration Facilities
Plus 25% for Engineering
Contingencies
Adsorbent Cost
Total
Resin
$350,000
45,000
$7.00/lb
395,000
Granular Activated
Carbon
4950,000
* 20,000
@ $0.55/lb
4970,000
Operating Costs
Adsorber Power
Regeneration Fuel
Solvent Regeneration
Adsorbent Makeup
*/yr
7,100
3,000
6,188
9,000
(5
^/IjOOO gal
1.945
0.822
1.695
2.466
yr)
$/yr <£/l,000 gal
3,550 0.973
4,203 1.152
_ —
18,, 000 4.932
(8% loss /cycle)
Subtotal *25,288 6.928
Capital Related Costs
(exclude adsorbent):
Depreciation 9%
Maintenance 3%
Property Overhead 2%
Quality Control
Total
49,000 13.420
9,000 2.460
22.
§83,288 1,000 gal
$25,753 7.057
133,000 36.438
9,000 2.460
46.
$167,753 1,000 gal
aNo specific GAG or resin product. Values taken at average costs.
Source: References 11 and 12,
7-244
-------
The resin system looks very promising because of the many assumptions
made concerning design and performance, e.g., high capacity, rapid kinetics,
and a 5 year resin lifetime. The assumptions have not yet been demonstrated.
Moreover, the design is for a waste influent loading (1 ppm) that is extremely
low for an industrial waste stream. Costs, already high relative to many
other technologies, will increase drastically as influent loadings (and system
size) increase.
However, the example does indicate that resin adsorption may be more
economical than carbon adsorption. Similar reasoning has been applied in
Reference 1 where costs have been estimated for resin adsorption applied to
three different waste streams. Costs ranged from $38.60 per 1,000 gallons for
a phenol recovery system (at 5 percent phenol in waste) to $0.83 per
1,000 gallons for a chlorinated pesticide removal system. In the latter case,
the cost of a GAG treatment system was estimated at $1.33 per 1,000 gallons.
The cost data are outdated (from the 1970s); costs in 1986 dollars would be
about 50 percent greater, based on changes in the chemical engineering plant
cost index.
The high costs of resin adsorption for the treatment of moderate to high
concentration contaminant levels can only be justified in situations where
cost benefit is realized from product recovery. In the case of the phenol
recovery system used in the example above, credit from the sale of phenol
exceeded total annual operating costs, therefore justifying use of the process
on an economics basis.
7.7.4 Overall Status
7.7.4.1 Availability—
Resin adsorption technology parallels that for carbon adsorption.
Equipment requirements are similar and available from a number of
manufacturers serving the chemical process industries. However, there appears
to be some question about the commercial availability of many of the resin
adsorbents for which data are reported in the literature. Ambersorb XE-340,
for example, manufactured by Rohm and Haas and the subject of numerous
technical studies, is not available in commercial quantities. The
availability of some other resin adsorbents may also be questionable.
7-245
-------
7.7.4.2 Application—
Because of their expense, resins are not commonly used full-scale to
remove organics from wastewaters. There is also little publicly available
information on current or proposed industrial applications. Information of a
general nature does report that resins are being used for color removal from
dyestuff and paper mill waste streams, for phenol removal, and for polishing
of high purity waters.
The following applicants have been identified as being particularly
attractive for resin adsorption technology.
• Treatment of highly colored wastes where color is associated with
organic compounds
• Material recovery where solvents of commercial value are present in
high enough concentration to warrant material recovery since it is
relatively easy to recover solutes from resin adsorbents
* Where selective adsorption is an advantage and resins can be
tailored to meet selectivity needs
* Where low leakage rates are required; resins exhibit low leakage
apparently as a result of rapid adsorption kinetics
* Where carbon regenerations is not practical, e.g., in cases when
thermal regeneration is not safe
* Where the waste stream contains high levels of inorganic dissolved
solids which drastically lowers carbon activity; resins activity can
usually be retained ,although prerinses may be required
7.7.4.3 Environmental Impacts—
The only major environmental impacts resulting from resin adsorption
systems are associated with the disposal of the regeneration solution and the
extracted solutes when they can not be recycled. Distillation to recover
solvent and incineration of the separated solute are likely treatment/disposal
options. Air emissions would have to be considered as a result of these
treatment processes.
7-246
-------
7.7.4.4 Advantages and Limitations—
As noted, resin adsorption appears to offer advantages in certain
situations; e.g., for treatment of highly colored wastes, for material
recovery, where low leakage is required, and in instances where carbon
adsorption is not practical. The advantages of resin adsorption are a result
of their potential for selectively, rapid adsorption kinetics, and ease of
chemical regeneration.
Major limitations of resin adsorbents result from: 1) the generally
lower surface area and usually lower adsorption capacities than those found in
activated carbon; 2) their susceptibility to fouling due to poisoning by
materials that are not removed by the regenerant; and 3) their relatively high
cost. The high cost of the resin may be balanced by its ease of regeneration
and their predicted long lifetimes in situations where carbon must be
thermally regenerated and carbon losses become appreciable (up to 10 percent).
7-247
-------
References
1. Lyman. W.J., Resin Adsorption in: Unit Operations for Treatment of
Hazardous Wastes, Pollution Technologies Review No. 47. Noyes Data
Corporation, Park Ridge, New Jersey, 1978.
2. Rohm and Haas Company, Fluid Process Chemicals Department,
Amber-Hi-Lites, Winter 1980 (Technical Bulletin).
3. Neely, J.W. and E.G. Isacoff, Carbonaceous Adsorbent For the Treatment of
Ground and Surface Waters, Marcel Dekker, Inc, New York, N.Y., 1982.
4. Mark, H., et al. Encyclopedia of Polymer Science and Technology,
Cohesive-Energy Density. Vol. 3, p. 833. John Wiley & Sons, Inc., 1970.
5. Slejko, F.L., Applied Adsorption Technology, Chemical Industry Series,
Marcel Dekker, Inc, New York, N.Y. 1985.
6. U.S. EPA, Background Documents for Solvents to Support 40 CFR Part 268
Land Disposal Restrictions, Volume II, January 1986.
7. Chriswell, C.D. et al., "Comparison of Macroreticular Resin and Activated
Carbon as Sorbents", J. American Water Works.
8. Treatment of Volatile Organic Compounds in Drinking Water, Environmental
Research Laboratory, EPA 600/8-83-019, May 1983.
9. Dyksen, J.E. and A.F. Hess, Alternatives for Controlling Organics in
Ground Water Supplies, J. American Water Works Association, August 1982.
10. Symons, J.M., J.K. Carswell, J. DeMarco, and O.T. Love, Jr., Removal of
Organic Contaminants from Drinking Water Using Techniques Other Than GAG
Alone, A Progress Report, Cincinnati, 1979.
11. U.S. EPA. Synthetic Resin Adsorbents in Treatment of Industrial Waste
Streams, EPA 600/2-84-105, May 1982.
12. McGuire, M.J. and Sublet, I.A., Activated Carbon Adsorption of Organics
from the Aqueous Phase, Volume 2; Economic Analysis Employing Ambersorb
XE-340 Carbonaceous Adsorbent in Trace Organic Removal from Drinking
Water, Ann Arbor Science 1980.
7-248
-------
SECTION 8.0
CHEMICAL TREATMENT PROCESSES,
The chemical treatment methods discussed in this section include some
processes which could equally well be classified as thermal processes
(i.e., wet air and supercritical water oxidation) since the general result of
these high temperature processes is the conversion of the organic contaminants
to fundamental products of oxidation such as carbon dioxide and water. Other
technologies like the experimental UV/ozonation process and other oxidation
processes do not achieve total destruction and must be considered as a
pretreatment step for a second treatment technology, usually a biotreatment
process. Chlorinolysis, another process discussed in this section, is unique
in that it utilizes chlorinated organic wastes to produce a product, namely
carbon tetrachloride. The processes addressed in this section are:
8.1 Wet Air Oxidation
8.2 Supercritical Water Oxidation
8.3 Ozonation
8.4 Other Chemical Oxidation Processes
8.5 Chlorinolysis
8.6 Dechlorination Processes
Discussions of these chemical treatment processes are provided using the same
format as was used for the discussions of physical treatment processes in the
previous section.
8-1
-------
8.1 WET AIR OXIDATION
Wet air oxidation (WAO) is the oxidation of dissolved or suspended
contaminants in aqueous waste streams at elevated temperatures and pressures.
It is generally considered applicable for the treatment of certain
organic-containing media that are too toxic to treat biologically and yet too
1 2
dilute to incinerate economically. ' A leading manufacturer of
commercially available WAO equipment reports that WAO takes place at
temperatures of 175 to 320°C (347 to 608°F) and pressures of 2,169 to 20,708
kPa (300 to 3000 psig). Although the process is operated at subcritical
conditions (i.e., below 374°G and 218 atmospheres), the high temperatures and
the high solubility of oxygen in the aqueous phase greatly enhances the
reaction rates over those experienced at lower temperatures and pressures. In
practice, the three variables of pressure, temperature and time are controlled
to achieve the desired reductions in contaminant levels.
In addition to serving as the source of oxygen for the process, the
aqueous phase also moderates the reaction rates by providing a medium for heat
transfer and heat dissipation through vaporization. Generally, pressures are
maintained above the vapor pressure of water to limit water evaporation rates,
thus limiting the heat requirement for the process. The reactions proceed
without the need for auxiliary fuel at feed chemical oxygen demand (COD)
3
concentrations of 20 to 30 grams per liter. The extent of contaminant
destruction will depend upon the wastes to be oxidized and the reaction
conditions. Typically, 80 percent of the organic contaminants will be
oxidized to C0~ and ELO. Residual organics will generally be low
molecular weight, biodegradable compounds such as acetic acid and formic acid.
8,1.1 ProcessDescription
4
A schematic of a continuous WAO system is shown in Figure 8.1.1. The
Zimmerman WAO System, as shown in the figure, has been developed by Zimpro,
Inc., Rothschild, Wisconsin. It represents an established technology for the
treatment of municipal sludges and certain industrial wastes. Full scale
treatment of solvent wastes has not yet been demonstrated. However a 10 gpm
8-2
-------
O3
OJ
OXIDIZABLE
WASTE
FEED
PUMP
PROCESS
HEAT
EXCHANGER
REACTOR
PCV
WASTEWATER
AIR COMPRESSOR
Figure 8.1.1. Wet air oxidation general flow diagram
Source; Reference 4
-------
pilot unit has been used to treat an alkaline solvent waste and a solvent
still bottom waste, as well as other organic wastes, at a commercial waste
o A_Q
treatment facility in California. ' As will be noted later, the
effectiveness of WAO as an alternative to land disposal for certain solvent
containing waste streams will depend upon a number of factors including the
molecular structure and concentration of the contaminants and the processing
..... 1,3,6-12
conditions.
In the WAO process, the waste stream containing oxidizable contaminants
is pumped to the reactor using a positive displacement, high pressure pump.
The feed stream is preheated by heat exchange with the hot, treated effluent
stream. Air (or pure oxygen) is injected following the high pressure pump,
and as shown in the figure, usually directly into the reactor vessel. Steam
is added as required to increase the temperature within the reactor to a level
necessary to support the oxidation reactions in the unit. As oxidation
proceeds, heat of combustion is liberated. At feed COD concentrations of
roughly 2 percent the heat of combustion will generally be sufficient to bring
about a temperature rise and some vaporization of volatile components.
Depending upon the temperature of the effluent following heat exchange with
the feed stream, energy recovery may be possible or final cooling may be
required. Following energy removal, the'oxidized effluent, consisting mainly
of water, carbon dioxide, and nitrogen, is reduced in pressure through a
specially designed automatic control valve. The effluent liquor is either
suitable for final discharge (contaminant reduction achieves treatment
standards) or is now readily biodegradable and can be piped to a biotreatment
unit for further reduction of contamination levels. Similarly, noncondensiole
gases can either be released to the atmosphere or passed through a secondary
control device (e.g., carbon adsorption unit) if additional treatment is
o
required to reduce air contaminant emissions to acceptable levels.
The pressure vessel is sized to accommodate a fixed waste flow and
residence time. Based on the characteristics of the waste, a combination of
time, temperature, pressure, and possibly catalyst can be utilized to bring
about the destruction of many solvent contaminants.
8-4
-------
The continuous reactor can reportedly take two forms: a tower reactor
or a reactor consisting of a cascade of completely stirred tank reactors
13
(CSTRs). The bubble tower reactor is by far the more common. It is a
vertical reactor, as shown in Figure 8.1.1, in which air is passed through the
feed. The reactor is sized, based on feed rate, to provide the holding time
required for the reactions to proceed to design levels. The stirred tank
cascade reactor consists of a series of horizontal reactor chambers contained
within a horizontal cylinder. The wastewater cascades from one chamber to the
next, and then is released for discharge or post-treatment. Air is generally
injected into each of the CSTRs.
8.1.1.1 Pretreatment Requirements for Different Waste Forms and
Characteristics-
Very little discussion is found in the literature concerning the physical
form of wastes treatable by WAO. However, WAO equipment and designs have been
used successfully to treat a number of municipal and industrial sludges.
According to a representative of the leading manufacturer of WAO systems,
wastes containing up to 15 percent COD (roughly equivalent to 7 to 8 percent
14
organics) are now being treated successfully in commercial equipment.
Treatment of solid bearing wastes is dependent upon selection of suitable
pump designs and control devices. WAO units used for activated carbon
1 A
regeneration now operate at the 5 to 6 percent solids range. Treatment of
higher solid levels is not precluded by fundamental process or design
limitations. Column design must also be consistent with the need to avoid
settling within the column under operating flow conditions. Thus,
pretreatment to remove high density solids (e.g., metals by precipitation) and
accomplish size reduction (e.g. filtration, gravity setting) would be required
for some slurries. It should be noted that the WAO unit operated by Casmalia
Resources in California does not accept slurries or sludges for treatment.
This may be a result of design factors precluding their introduction into the
15
system.
Several bench scale studies have been conducted to determine the
susceptibility of specific compounds to wet air oxidations. Results of these
studies and other studies have been summarized in the literature. » » »
8-5
-------
The results indicate that the following types of compounds can be destroyed in
wet air oxidation units.
• Aliphatic compounds, including those with multiple halogen atoms.
Depending upon the severity of treatment, some residual oxygenated
compounds such as low molecular weight alcohols, aldehydes, ketones,
and carboxylic acids might be present, but these are readily
biotreatable.
• Aromatic hydrocarbons, such as toluene and pyrene are easily
oxidized.
• Halogenated aromatics can be oxidized provided there is at least one
nonhalogen functional group present on the ring; the group should be
an electron donating constituent such as an hydroxyl, amino, or
methyl group.
• Halogenated aromatics, such as 1,2-dichlorobenzene, PCBs, and TCDDs,
are resistant to oxidation under conventional conditions although
these compounds are destroyed to a greater extent as conditions are
made more severe or catalysts are employed. However, Casmalia
Resources does not accept chlorinated aromatics. *••*
• Casmalia Resources also does not accept for WAO treatment wastes
containing highly volatile organics like Freon which would enter the
unit in the gas phase, and tin, which is corrosive to heat exchanger
surf ace s.^
Batch process results obtained in the laboratory are applicable to continuous
process design for pure compounds and complex sludges, i.e., specific compound
destruction is similar and predictable for pure compounds and those compounds
contained in complex industrial wastes. '
8.1.1.2 Operating Parameters—
Although operation of a WAO system is possible, by definition, under all
aubcritical conditions; i.e., below 374°C and 218 atm (3220 psig),
commercially available equipment is designed to operate at temperatures
ranging from 175 to 320°C and at pressures of 300 to 3000 psig.
Of all variables affecting WAO, temperature has the greatest effect on
reaction rates. In most cases, about 150°C (300°F) is the lower limit for
appreciable reaction. About 250°C (482°F) is needed for 80 percent reduction
of COD, and at least 300°C (572°F) is needed for 95 percent reduction of COD
within practical reaction times. Destruction rates for specific constituents
2
may be greater or less than that shown for COD reductions.
8-6
-------
Initial reaction rates and rates druing the first 30 minutes are
relatively fast. After about 60 minutes, rates become so slow that generally
2
little increase in percent oxidation is gained in extended reaction.
An increase in reaction temperature will lead to increased oxidation Out
generally will require an increase in system pressure to maintain the liquid
phase and promote wet oxidation. A drawback to increasing the temperature and
pressure of the reaction is the greater stress placed on the equipment and its
components, e.g., the increased potential for corrosion problems. Corrosion
is controlled by the use of corrosion resistant materials such as titanium.
As noted by Zimmerman, et al., the object of WAO is to intimately mix the
right portion of air with the feed, so that under the required pressure,
combustion will occur at a speed and temperature which will effectively reduce
the organic waste to desired levels. Pressures should be maintained at a
level that will provide an oxygen rich liquid phase so that oxidation is
maintained. Charts and curves are provided in this reference to aid in
the determination of waste heating value, stoichiometric oxygen requirement,
and the distribution of water between the liquid and vapor phases at given
temperatures and pressures.
Previous experience with the design of wet oxidation systems has shown
that batch results are applicable to continuous process design when the oxygen
transfer efficiency is 90 percent (11 percent excess air) or less. A model
was developed to gain insight into the key system parameters using a common
industrial waste stream and fixed temperature, residence time, and COD
reduction. The model was used to estimate costs for the system. Its
value, as a predictive tool, along with that of supplementary kinetic
18
studies of batch wet oxidation, is limited by the sparsity of experimental
data concerning reaction products and their phase distributions at the
elevated temperatures and pressures encountered during MAO.
8.1.1.3 Post-Treatment Requirements—
The use of WAO to meet EPA treatment levels (see Table 2.2) for solvent
o
wastes has not yet been demonstrated. As will be noted later, WAO has been
used under certain conditions to achieve destruction levels that meet the EPA
8-7
-------
proposed treatment standards. However, for the most part, this level of
performance has been achieved for specific compounds oxidized in batch
reactors under conditions that are more rigorous than those normally used in
commercial systems.
Destruction levels will vary for different compounds in complex waste
mixtures and there is evidence that certain of the low molecular weight WAO
breakdown products (e.g., tnethanol, acetone, acetaldehyde, formic acid, etc.)
are resistant to further oxidation. Thus, under typical WAO operating
conditions it is likely that both contaminant residuals and low molecular
weight process by-product residuals may be present. While it is entirely
possible that imposition of,more stringent operating conditions will serve-to
reduce these residuals to acceptable levels, the manufacturers and users of
commercial WAO systems stress that the major applications involve the
pretreatment of waste, usually for subsequent biological treatment.
Even under conditions that are favorable for wet oxidation, it is also
likely that certain contaminants, particularly some of the more volatile
components, will partition between the vapor phase and the liquid phase. The
partitioning will be a function of operating conditions and the contaminant
partial pressure. The Henry's Law constant at the temperature of operation
will fix the distribution, however, Henry's Law constant is not generally
known under most conditions of tfAO system operation. Although a method of
estimation has been proposed by researchers at Michigan Technological
1 Q
University, empirical tests will be necessary to establish vapor and
liquid phase residuals and some post-treatment of both streams may Be
necessary. Existing post-treatment methods for the liquid generally involves
bacteriological treatment. Although the results of post-treatment schemes for
vapors from the WAO system have not been found in the literature, a two—stage
water scrubber/activated carbon adsorption system has been used to treat WAO
3
vapor emissions. Presumably carbon adsorption or scrubbing systems could
be routinely employed if necessary.
8-8
-------
8.1.1.4 Treatment System Combinations-
Most of the commercial WAO systems in operation today are employed as
pretreatment devices to enhance the biotreatability of municipal and
industrial wastes. Wet air oxidation is also used as a means of regenerating
spent activated carbon used as an adsorbent. In the latter case the WAO
regenerates the activated carbon through oxidation of the organics adsorbed on
19
the carbon surfaces.
The application of WAO to industrial organic wastes has generally been
limited to treating specific, homogeneous waste streams, including soda
pulping liquors at pulp mills and n-nitrosodimethylamine and acrylonitrile
wastes. However, WAO has been used since 1983 to treat varied waste streams
at the Casmalia Class I disposal site, located near Santa Maria, California.
Phenolic, solvent still bottoms, and other organic wastes have all been
treated at Casmalia, in certain instances in conjunction with a powdered
activated carbon treatment system and a two stage scrubber-carbon adsorption
20
system for vapor treatment.
Treatment of specific waste streams to meet EPA solvent treatment
standards by a WAO system is not precluded, as evidenced by some of the
performance data shown below for removal of specific contaminants. However,
in most instances reaction conditions would have to be tailored to the waste
stream and pollutant. Generally an increase in the pressure/temperature
conditions normally employed by the users of WAO systems would be required.
Equipment problems associated with the more stringent operating conditions
would have to be considered.
8.1.2 Demonstrated Performance of WAO Systems
o
As noted by EPA, full scale use of WAO technology is well demonstrated
for the treatment of municipal sludge but full scale treatment of solvent
wastes is not demonstrated. However, data showing the WAO destruction of
specific organic compounds including several solvents of concern to EPA, have
been provided in the literature. These data are largely the result of bench
scale testing, but do include results of pilot-scale and full-scale
performance tests. The data indicate that WAO can be effective in treating
specific organic contaminants, including many solvents and various
8-9
-------
industrial wastewaters. However, not all specific contaminants are destroyed
to the extent that direct discharge would be allowable under the proposed EPA
treatment standards for solvents of concern. Chlorinated organics appear to
be the most difficult compounds to destroy. Residuals in both the gas and
liquid phase would also have to be considered on a case by case basis if WAO
technology is to be used for the treatment of specific solvent containing
waste streams.
8.1,2.1 Bench-Scale Studies—
Bench scale studies of the destruction of specific organic substances by
1 Q I fi
wet oxidation have been conducted at Zimpro, Inc. ' * Some of these data
are shown in Tables 8.1.1 through 8.1.3. The tables include destruction data
for organic compounds other than the solvents and ignitable organics of
interest to this study in order to illustrate the effect of operating
variables, catalysts, and chemical structure on the effectiveness of wet air
oxidation.
The data shown in Table 8.1.1 were taken from Reference 16. The results
were obtained using single pure compounds in distilled water as the material
undergoing treatment. Concentrations of starting material at 5 to 12 g/L were
used for the study. Batch oxidations were carried out in a titanium 3.8 liter
(1—gallon) magna—drive stirred autoclave. Temperature regulation was achieved
with a controller-recorder that regulated the external heating elements.
Generally, one liter of material was charged into the autoclave, which was
sealed and pressurized with compressed air. After heating to the desired
temperature for one hour, the contents were cooled by an internal coil
carrying cold water. The resulting gases were analyzed to determine oxygen
use and the contents were removed for further testing. The test and
analytical procedures used are identified in the reference, although no data
are provided with regard to the excess air used and the pressures realized
during the 1 hour test period.
As shown in Table 8.1.1, all 10 compounds (only two of which, acrolein
and acrylonitrile are considered as solvents/ignitables) were destroyed to an
appreciable extent at 320°C and at 275°C, although a soluble copper catalyst
was required for extensive destruction of 2-chlorophenol and pentachlorophenol
at 275°G. These two chlorinated aromatics were the most difficult to destroy
8-10
-------
TABLE 8.1.1. ONE HOUR WET OXIDATIONS
Compound
Starting
Concentration
(g/D
Percent starting material destroyed
320°C 275°C 275°C/Cu++
Acenapthene
Acrolein
Acrylonitrile
2-chlorophenoi
2,4-dimethylphenol
2-4-dinitrotoluene
1, 2-diphenylhydrazine
4-nitrophenol
Pentachlorophenol
Phenol
7.0
8.41
8.06
12.41
8.22
10.0
5.0
10.0
5.0
10.0
99.96
99.96a
99.91
99.86
99.99
99.8
99.98
99.96
99.88
99.97
99.9
99.05
99.00
94.96
99.99
99.74
99.98
99.60
81.96
99.77
__
—
99.50
99.88
—
•
—
—
97.30
'"" "
aThe concentration remaining was less than the detection limit of 3 mg/L.
TABLE 8.1.2. PRODUCTS IDENTIFIED FROM ONE HOUR OXIDATIONS AT 320°C
Starting
compounds
Formic acid
(mg/L)
Acetic acid
(mg/L)
Acenapthene
Acrolein
Acrylonitrile
2-chlorophenol
2,4-dimethylphenol
2-4-dinitrotoluene
1,2-diphenylhydrazine
4-nitrophenol
Pentachlorophenol
Phenol
6
81
48
52
75
134
37
793
0
13
443
862
921
442
1,527
213
526
500
101
1,034
8-11
-------
TABLE 8.1.3. BENCH-SCALE WET AIR OXIDATION OF PURE COMPOUNDS
Wet oxidation
conditions
Compound °C /minutes
Arochlor 1254
Carbon Tetrachloride
Chlorobenzene
Chloroform
1-Chloronaphthalene
Dibutylphthalate
2,4-Bieb.loroaniline
1, 2-Dichlorobenzene
1, 2-Diehloroethane
Formic Acid
Hexachlorocyclopentadiene
Isophorone
Kepone
Malathion
Toluene
Nitrobenzene
N-Nitrosodimethylamine
Pyrene
Pyridine
2,4, 6-Trichloroaniline
320/120
275/60
a275/60
275/60
a275/60
275/60
a275/60
a320/60
275/60
300/60
300/60
275/60
a280/60
250/60
275/60
a320/120
275/60
275/60
a320/120
a320/120
Starting Final
concentration concentraion
(mg/L) (mg/L)
20,000
4,330
5,535
4,450
5,970
5,320
259
6,530
6,280
25,000
10,000
4,650
1,000
11,800
4,330
5,125
5,030
500
3,910
10,000
7,400
12
1,550
3
5
26
0.5
2,017
13
410
15
29
690
18
12
255
22
0.26
570
2.5
Percent
destroyed
63.0
99.7
72.0
99.9
99.92
99.5
99.8
69.1
99.8
98.3
99.9
99.4
31.0
99. »5
99.7
95.0
99.5
99.95
85.4
99.97
aCatalyzed.
8-12
-------
and unlike the other compounds continued to show a marked increase in
destruction efficiency with temperature through the 275°C to 320°C range.
Presumably destruction efficiencies would be somewhat higher at more elevated
temperatures.
Although no attempt was made to measure vapor phase residuals,
Reference 16 does present data for liquid phase residuals. While the details
of analysis are not specified and it is not entirely certain that other
residual compounds are not formed, only formic acid and acetic acid were
identified in the amounts shown in Table 8.1.2. Although these residuals
represent as much as 20 weight percent of the original charge of the specific
test compound, the two low molecular weight acids formed are readily
biodegradable by conventional treatment methods. Thus, it was concluded that
wet oxidation of the waste constituents followed by biotreatment would yield
an effluent suitable for discharge to a publicly owned treatment plant.
Additional specific organic compound data taken from References 1 and 8
are shown in Table 8.1.3. As stated in Reference 1, bench-scale studies were
performed using batch autoclaves having a total volume of 500 to 750 mL, and
constructed of 316L stainless steel, nickel, or titanium. The autoclaves were
charged with 100 to 300 mL of the sample to be oxidized, sealed, charged with
air or oxygen sufficient to satisfy the sample oxygen demand, and then
subjected to a controlled time/temperature cycle. Continuous agitation was
provided during the treatment period.
As shown in Table 8.1.3 the chlorinated aromatic compounds (e.g., kepone,
Arochlor 1254, 1,2-dichlorobenzene, and chlorobenzene) were resistant to
degradation, whereas the lower molecular weight chlorinated aliphatics (e.g.,
chloroform, carbon tetrachloride, and 1,2-dichloroethane), including several
solvents of concern, were relatively easy to destroy. Reference 8 presents
data showing destruction efficiencies in excess of 99 percent for Arochlor
1254 and other chlorinated aromatics using a proprietary catalyst system. No
data were presented for liquid or vapor phase residuals.
Bench-scale wet oxidation studies were also performed on industrial
wastewaters in which the destruction of specific organic contaminants was
monitored. The results, as shown in Table 8.1.4, indicate that all compounds,
with the exception of the one halogenated aromatic organic tested,
1,2-dichlorobenzene, undergo significant oxidation. However, the residual
8-13
-------
TABLE 8.1.4. BENCH-SCALE WET AIR OXIDATION OF ORGAN1CS IN WASTEWATERS
Compound
Dimethylaniline
Toluene
Acetonitrile
Propionitrile
2 , 4-Dichlorophenol
D ipr opy 1 f ormamide
N-ni trosod ime thy lamina
Tr ich lor oethy lene
1 , 2-Dichlorobenzene
Wet air
oxidation
conditions
°C/min
280/60
320/120
3275/60
»2 75/60
320/60
250/60
320/60
320/60
320/120
Feed
concentration
(mg/L)
1,300
5.0
1,040
391
500
219
510
500
540
Product
concentration
(mg/L)
1.6
0.5
17
7
2
1
1
1.7
150
Removal
(%)
99.9
90.+
98.4
98.2
99.6
99.5
99.8
99.7
72.2
aCatalyzed.
8-14
-------
concentration of one solvent of concern, trichloroethylene, is higher than the
proposed EPA treatment standard (0.1 rag/liter) and thus would require
additional treatment before discharge. The residual concentration of
150 mg/liter found for 1,2-dichlorobenzene is well above the proposed EPA
treatment standard of 2 mg/liter. This is consistent with the pure compound
WAO results presented in Table 8.1.3.
Results of the WAO of several solvents in wastewaters have also been
reported in Reference 1, and are shown in Table 8.1.5. The compounds measured
were either present in the wastewater, or in the case of the chlorinated
compound tested in study 3, added as spikes to industrial wastewater. Again,
the data show the difficulty in oxidizing the halogenated aromatic compound,
chlorobenzene. The low molecular weight alcohols also exhibit resistance to
WAO.
Results of wet oxidation tests involving the use of a patented catalyst
consisting of bromide, nitrate, and manganese ions in acidic solution have
21
also been reported in the literature. Batch tests were conducted using a
1-liter titanium autoclave. Compounds studied include solvents of concern
such as nitrobenzene and xylene as well as other low molecular weight
compounds such as ethylene dibromide, acetonitrile, hexachlorobutadiene, and
trichloropropane. Reaction temperatures used were somewhat lower than those
used in the Zimpro studies. Destruction efficiencies were also low, e.g.,
54 percent of xylene destroyed at 165°C and up to 31 percent of nitrobenzene
destroyed after I hour at 200°C. Some reaction product data were also
presented and discussed. It was noted that HC1 is a byproduct of the wet air
oxidation of chlorinated compounds and thus some provisions must be made for
removal of this compound from the reaction products.
8.1.2.2 Pilot-Scale Studies—
The results of several pilot scale studies have been reported in the
literature. The flow rates of systems used in these studies ranged from
2.5 to 28.9 gallons per hour (0.23 to 2.6 cubic meters per day). In a pilot
8-15
-------
TABLE 8.1.5. BENCH-SCALE WET AIR OXIDATION OF WASTEWATERS
Study
1
1
1
1
1
1
2
2
2
3
3
3
3
Compound
Acetone
Ethanol
Methyl Ethyl Ketone
Methanol
2-Propanol
Toluene
Ethanol
Methyl Ethyl Ketone
2-Propanol
Chloroform
Chlorobenzene
Trichloroethylene
Dichloromethane
Feed
1,680
1,530
276
3,230
2,230
80
2,800
8,200
9,900
270
792
300
252
Concentration (mg/L)
Oxidized Oxidized
280°C/60 min 320°C/60 min
10 10
60 10
1 3
1,380 290
40 30
1 1
1,200
1
170
1
61
2
1
Percent
removal
at 280°C
99.4
96.1
99.6
57.3
98.2
98.8
57.1
99.9
98.2
99.6
92.3
99.3
99.6
8-16
-------
scale test of coke plant wastewater, removals of phenols and cresol were
measured at 99.8 and 99.9+ percent, respectively. The test was carried out at
279°C, 1558 psig, and a flow rate of 6.3 gallons per hour. Residence time was
69 minutes and a catalyst was used to assist the reaction.
Pilot-scale tests were also conducted with a wastewater containing many
specific organic solvents of concern along with pesticide and herbicide
wastes. The results of this pilot-scale WAO are presented in Table 8.1.6.
Destruction efficiencies were generally high, even for 1,2-diehlorobenzene, a
compound that was not readily oxidized in the bench—scale tests.
8.1.2.3 Full-Seaie Studies—
Several full—scale studies have been conducted at the Casmalia Resources
facility in Santa Barbara County, California using a skid mounted WAO systems,
capable of 10 gallon per minute flow rate for waste materials with a COD of
40 g/liter. Tests have been conducted on wastewater containing phenolies,
organic sulfur, cyanides, nonhalogenated pesticides, solvent still bottoms,
and general organics. In the case of the general organic wastewater, COD was
reduced 96.7 percent to a level of 2.5 g/liter. During this test the wet
oxidation unit was operated at 277°C (531°F), 1550 psig, and a residence time
2
of 120 minutes. For the solvent stxll bottoms, the unit was operated at an
average reactor temperature of 268°C (514°F), a reactor pressure of 1550 psig,
and a nominal residence time of 118 minutes. COD, BOD, and TOG reductions of
2
95.3, 93.8, and 96.1 percent, respectively, were measured. Although no
specific organic compound destruction efficiencies were reported for solvents
of concern, phenol destruction efficiencies of 99.8 percent were reported in a
test of a spent caustic waste from a petroleum refinery.
8.1.2.4 Studies of Treatment Systems Using WAO—
The use of pilot-scale and full-scale treatment systems combining
TM
PACT (powdered activated carbon addition to the reaction basin of an
activated sludge process) with wet air oxidation regeneration have been
19 22
reported in the literature. * Reference 22 reports destruction
efficiencies of greater than 99 percent for several priority pollutant
solvents present in a domestic and organic chemical wastewater not treatable
by conventional biological treatment systems.
8-17
-------
TABLE 8.1.6. PILOT-SCALE WET AIR OXIDATION OF ORGANIC
COMPOUNDS IN INDUSTRIAL WASTEWATER
Wet Air Oxidation Conditions:
Temperature, °C 314
Flow, gph 2.6
Residence Time, min 128
Pressure, psig 1,943
COD
Feed, g/liter 77.5
Effluent, g/liter 9.4
% Removal 87.9
1,2-dichlorobenzene
Feed, rag/liter 2,213
Effluent, mg/liter 29
% Removal 98.7
Methylene Chloride
Feed, mg/1 i ter 60
Effluent, mg/liter 0.01
% Removal 99.9+
Perchloroe thylene
Feed, mg/liter 4,000
Effluent, mg/liter 0.9
% Removal 99.9+
Freon TF
Feed, mg/liter 3,000
Effluent, mg/liter 2
% Removal 99.9+
Xylene
Feed, mg/liter 8,385
Effluent, mg/liter 20
% Removal 99.8+
Toluene
Feed, mg/liter 30
Effluent, mg/liter 0.5
% Removal 98.3+
(continued)
8-18
-------
TABLE 8.1.6 (continued)
Wet Air Oxidation Conditions:
Temperature, °C 314
Flow, gph 2.6
Residence Time, min 128
Pressure, psig 1,943
Phenols
Feed, mg/liter 1,556
Effluent, mg/liter 2.1
% Removal 99.9
Isopropyl Alcohol
Feed, mg/liter 1,700
Effluent, mg/liter 400
% Removal 76.5
Methyl Ethyl Ketone
Feed, mg/liter 6,000
Effluent, mg/liter 1.0
% Removal 99.9+
S.I. Conversion
m3/d = gph x 0.0908
kPa = psig x 6.89
Source: Reference 1.
8-19
-------
As noted in Reference 22, the PACT /Wet Air Regeneration pilot system
was operated at aeration detention times of 2.5 to 5.5 hours and at wet air
regeneration temperatures of 230 C.
Initial testing of conventional biological treatment systems and sludge
digestion confirmed the inability of a pure biological system to treat the
test wastewater. The BOD,, to COD ratio, an indicator of biotreatability,
averaged 0.35 (a ratio of 0.5 to 0.6 for domestic wastes is common) and over
72 EPA priority pollutants were detected in the raw wastewater. An activated
sludge pilot plant operating at a 9 hour hydraulic retention time provided
marginal organic removal (56% GOD reduction) and was unable to produce an
effluent with less than the required 25 mg/L suspended solids. Variations in
waste strength prevented accurate system control and dictated a need for waste
equalization. Attempts to anaerobically digest the waste activated sludge
* T*!yl
were not successful due to the presence of toxic components. The PACT
effluent was analyzed for priority pollutants. The results (Table 8.1.7) show
T*M
high priority pollutant removal through the PACT /Wet Air Regeneration
22
treatment system. Many of the treatment levels achieved are below the EPA
treatment standards for specific solvents.
Reference 19 presents results obtained during treatment of RCRA
wastewater and CERCLA ground water at the Bofors-Nobel facility in Muskegon,
Michigan. Cleanup at the site was conducted in accordance with the system
schematic shown in Figure 8.1.2. Two WAO units are used, one dedicated solely
to detoxification, the other used primarily as a. carbon regeneration unit with
occasional use as an additional detoxification unit. Although no data are
provided for specific organic solvent components of the waste, an average
efficiency of 99.8 percent is stated for toxics in the feed.
8.1.3 Cost of Treatment
8.1.3.1 Wet Air Oxidation Costs—
Treatment costs for wet air oxidation systems will be affected by a
number of parameters including the amount of oxidation occurring, the
hydraulic flow, the design operating conditions necessary to meet the
treatment objectives, and the materials of construction. These factors
account for the band of capital costs shown in Figure 8.1.3. The figure was
8-20
-------
TABLE 8.1.7. PRIORITY POLLUTANT REMOVALS USING A PACT™/WET AIR
REGENERATION SYSTEM FOR DOMESTIC AND ORGANIC
CHEMICALS WASTEWATER3
Parameter
Benzene
Ghlorobenzene
1,2,4-Trtchlorobenzene
1,1, 1-Trichloroethane
Chloroform13
2-Chlorophenol
1, 2-Dichlorobenzene
1, 3— Dichlorobenzene
Ethylbenzene
Methylchloridec
D ic h lorob romorae t hane
Nitrobenzene
Tetrachloroethylene
Toluene
Influent
(yg/L)
907
597
62
7
87
98
113
67
26
11
2.3
ND
1
1,195
Effluent
-------
HIGH STRENGTH
TOXIC WASTES
ACID CAUSTIC
00
10
NJ
DILUTE
PROCESS WASTE,
LANDFILL
LEACHATE
ASH TO LANDFILL
VIRGIN
CARBON
MAKEUP ,
PHOSPH.
ACID POLYMER
NEUTRALIZATION
MUSKEGON
COUNTY
WWTP
Figure 8.1.2. 4.5 MGD wastewater treatment facility
Source: Reference 19
-------
2
_j
_j
€0
o
JS
CO
10
20 30 40 50 60
WET OXIDATION PLANT CAPACITY, gpm
70
Figure 8.1,3. Installed plant costs versus capacity.
8-23
-------
taken from Reference 2 and updated to reflect changes in the 1982 to 1986
Chemical Engineering (CE) plant cost index* The costs do not include any
costs associated with pretreatment of the feed or post-treatment of the vapor
phase component of the treated liquor. However, post-treatment costs were
included in another capital cost estimate of $2.45 million (adjusted to 1986
using the CE plant cost index) for a 20 gpm plant. This estimate is within
the capital cost band shown in Figure 8.1.3.
Operating costs for the wet oxidation unit are shown in Figure 8.1.4.
These data were also derived from data given in Reference 2 with adjustment
made for the costs of labor and cooling water. As noted in Reference 2, power
accounts for the largest element of cost. This power cost is primarily the
result of air compressor operation. Additional power for supplying energy for
the oxidation of very dilute wastewaters would be at most 500 Btu/gallon. The
associated costs for this energy would be less than one (1) cent/gallon.
Total costs, capital plus operating, on a per unit of feed basis,
requires assumptions on life cycle, depreciation, taxes, and current interest
rates for the capital cost. One avenue for financing that has been used
commercially, common lease terms, are 5 years and 20 percent value at end of
term.* Table 8.1.8 illustrates the effect on total costs per unit of feed.
TABLE 8.1.8. WAO COSTS VERSUS FLOW
Hydraulic
flow (gpm)
2.0
10
20
40
Cost
operating
23
6
3
2-3
elements per gallon,
capital
31
7
5
4-5
cents
total
54
13
8
6-8
At Casmalia Resources, the prices, (April, 1985) for treatment of wastes
are computed based on the oxygen demand of the material. Prices range from a
minimum of $120 per ton to a maximum of $700 per ton versus $15 per ton for
the land disposal of low risk wastes.
*Assume lease charges of $17/1000 per month based on total installed cost.
8-24
-------
4.5
4,5
4.0
1986 cost: Power-$0.05/KWH; C.W.-$0.25/1000 gallon;
maintenance-1% capitol cost; labor-$30,000/yr/operator.
4.0
3.5
3.5
3.0
3.0
CO
I-
V)
o
o
CD
tc
UJ
Q.
O
o
i
2.5
2.0
1.5
1.0
.. . POWER
2.5
2.0
1.5
i.O
0.5
0.5
I I
— MAINT.
LABOR
"- C.W.
10 20 30 40 50
WAO UNIT CAPACITY (US SPM)
60 70
Figure 8.1.4. Unit operating costs versus unit flow rate.
8-25
-------
8.1.3.2 Comparison of WAO Costs with Other Alternative Treatment Costs—
A cost comparison between a 20 gpm WAO system and a comparable incinerator
was presented in Reference 4 for a wastewater containing 7 percent COD. It
was concluded that, although the installed capital cost for WAO was 50 percent
higher than that for incineration, operating costs were appreciably less
($132,000 annual operating cost for WAO versus'$463,500 for incineration in
1979 dollars) despite a charge for scrubbing of the WAO off gases and an
operating surcharge for BOD discharges to an average municipal wastewater
treatment plant. It was concluded that total operating costs including
amortization favor WAO when the fuel value of the waste organics is low (less
than approximately 50 g/liter Chemical Oxygen Demand).
Other sources of cost data, including comparative costs, are References 3
and 17. Reference 3 states that WAO is generally less expensive than
incineration when the COD concentration ranges between 10 to 150 g/liter.
Rough cost estimates of from about 10 to 50 cents per gallon were proposed
depending upon type of waste, concentration, and amount to be treated. For
comparison, landfilling costs of 12 to 25 cents per gallon for drummed wastes
were provided. Reference 17 provides cost data for a WAO system designed to
treat a 7 percent COD waste at a 10 gallon per minute treatment ra'te. Net
operating costs of $90,780 per year (December 1980) were estimated, a value
roughly equivalent to 3 cents per gallon, assuming a zero rate of return on
investment. This relatively low operating cost was compared to a landfilling
cost of roughly $1 per gallon for barrelled waste and $0.55 to $0.75 per
gallon for bulk waste. Although WAO costs were roughly two orders of
magnitude greater than typical costs for secondary biological municipal
wastewater treatment, the cost of $0.07 per pound of COD removed was suggested
as comparable to the typical municipal charge to industry of $0.05 to $0.10
per pound of COD removed.
Another source of cost data, Reference 21, provides data showing that
costs are a strong function of the contaminant type, its concentration, and
the amount of waste to be treated. Costs ranged from $0.12 per pound of
pentachlorophenol to $1.04 per pound of hexachlorobutadiene treated.
8-26
-------
8.1.4 Overall Status of WAO Process
8.1.4,1 Availability and Application of WAO Systems—
The WAO process is available commercially, and reportedly well over
150 units are now operating in the field treating municipal and various
14
industrial sludges. The process is used predominately as a pretreatment
step to enhance biodegradability. Only a few units are now being used to
treat industrial solvent/ignitable wastes. These include the 10 gallon per
minute unit at Casmalia Resources in California and other units operating at
Bofors-Nobel in Muskegon, Michigan and Northern Petrochemical in Morris,
Illinois.
The oxidation of specific contaminants in waste streams by the wet
oxidation process is not highly predictable. Equipment manufacturers rely
largely on the result of bench-scale results to tailor the design of
full-scale WAO continuous units for specific wastes. Full-scale data confirm
the results of WAO performance data obtained in bench and pilot-scale
studies.
8.1.4.2 Energy and Environmental Impacts—
As noted, the process is thermally self-sustaining when the amount of
oxygen uptake is in the 15-20 g/liter range. Below this range, some energy
input will be required to initiate and sustain reaction. However, the energy
requirement will be appreciably less than that required for incineration.
The environmental impacts of WAO will hinge upon the residuals remaining
after treatment. Wet scrubbing and carbon adsorption cleanup systems have
been used to treat the HC1 formed as a product of chlorinated organic
oxidation and to remove volatile organics from the waste off gases. Residuals
in the liquid phase may also require post treatment if, for example,
100 percent conversion to CO- and H~0 is not realized when treating
hydrocarbon contaminants. The available data do suggest that some form of
post treatment of both liquid and vapor phases will be required to meet
acceptable discharge levels.
8-27
-------
8.1.4.3 Advantages and Limitations—
There are several advantages associated with the use of WAO as noted by
the developer and stated in Reference 2.
1. The process is thermally self-sustaining when the amount of oxygen
uptake is in the 15-20 grams/liter range.
2. The process is well suited for wastes that are too dilute to
incinerate economically, yet too toxic to treat biologically.
3. Condensed phase processing requires less equipment volume than gas
phase processing.
4. The products of WAO stay in the liquid phase. Offgases from a WAO
system are free of NOX, SO2, and particulate. Mater scrubbing
and, if need be, carbon adsorption or fume incineration are used to
reduce hydrocarbon emissions or odors.
5. WAO also has application for inorganic compounds combined with
organics. The oxidation cleans up the mixture for further removal
of the inorganics. WAO can detoxify most of the EPA priority
pollutants. Toxic removal parameters are in the order, of
99+ percent using short-term, acute, static toxicity measurements.
Limitations of the WAO process relate to the sensitivity of destruction
efficiency associated with the chemical nature of the contaminant, the
possible influence of metals and other contaminants on performance, the
unfavorable economics associated with low and high concentration levels, and
the presence of residuals in both the vapor and liquid phases which may
require additional treatment. Costly materials of construction and design
features may also be required for certain wastes including many of the solvent
wastes which will form corrosive reaction products or require extreme
temperature/pressure conditions to achieve destruction to acceptable treatment
standard levels. In particular, chlorinated aromatic compounds are more
resistant to degradation and can result in the production of HC1 byproduct.
8-28
-------
REFERENCES
1. Dietrich, M.J., T.L. Randall, and P.J. Canney. Wet Air Oxidation of
Hazardous Organics in Wastewater, Environmental Progress, Vol. 4, No. 3,
August 1985.
2. Freeman, H. Innovative Thermal Hazardous Treatment Processes, U.S. EPA,
Hazardous Waste Engineering Research Laboratory, Cincinnati, Ohio, 1985.
3. California Air Resources Board. Air Pollution Impacts of Hazardous Waste
Incineration: A California Perspective, December 1983.
4. Wilhelrai, A.R., and P.V. Knopp. Wet Air Oxidation - An Alternative to
Incineration, Chemical Engineering Progress, August 1979.
5. Zimmerman, F.J., and D.G. Diddams, The Zimmerman Process and its
Applications in the Pulp and Paper Industry, TAPPI Vol. 43, No. 8,
August I960.
6. Copa, William, James Heimbuch, and Phillip Schaeffer. Full Scale
Demonstration of Wet Air Oxidation as a Hazardous Waste Treatment
Technology. In: Incineration and Treatment of Hazardous Waste,
Proceedings of the Ninth Annual Research Symposium, U.S. EPA
600/9-84-015, July 1984.
7. Copa, William, Marvin J. Dietrich, Patrick J. Cannery, and
Tipton L. Randall. Demonstration of Wet Air Oxidation of Hazardous
Waste. In Proceedings of Tenth Annual Research Symposium, U.S. EPA
600/9-84-022, September 1984.
8. U.S. Environmental Protection Agency, Background Document for Solvents to
Support 40 CFR Part 268, Land Disposal Restrictions, Volume II,
January 1986.
9. Reible, Danny D., and David M. Wetzel. Louisiana State University, A
Literature Survey of Three Selected Hazardous Waste Destruction
Techniques In Proceedings of Ninth Annual Symposium on Land Disposal of
Hazardous Waste. May 2-4, 1983.
10. Randall, T.R. Wet Oxidation of Toxic and Hazardous Compounds.
Zimpro, Inc. Technical Bulletin 1-610, 1981.
8-29
-------
11, Canney, P.J., and P.T. Schaeffer. Detoxification of Hazardous Industrial
Wastewaters by Wet Air Oxidation. Presented at 1983 National AIChE
Meeting, Houston, TX, March 27-31, 1983.
12. Baillod, C. Robert, and Bonnie M. Faith. Wet Oxidation and Ozonation of
Specific Organic Pollutants, U.S. EPA 600/52-83-060, October 1983.
13. Baillod, C.R., and R.A. Lamporter. Applications of Wet Oxidation to
Industrial Waste Treatment. Presented at 1984 AIChE National Meeting,
Philadelphia, PA, August 19-22, 1984.
14. Telephone Conversation with A. Wilhelmi on April 3, 1986.
15. Metcalf & Eddy, Inc. Hazardous Waste Treatment Storage and Disposal
Facility - Site Evaluation Report, Casmalia Resources, Casmalia,
California, Publication NS J-1074, April 8, 1985.
16. Randall, Tipton L., and Paul V. Knopp. Detoxification of Specific
Organic Substances by Wet Air Oxidation, Journal WPCF, Vol. 52, No. 8,
August 1980.
17. Baillod, C.R., R.A. Lamporter, and B.A. Barna. Wet Oxidation for
Industrial Waste Treatment, Chemical Engineering Progress, March 1985.
18. Baillod, C.R., B.M. Faith, and D. Masi. Fate of Specific Pollutants
During Wet Oxidation and Ozonation, Environmental Progress, August 1982.
19. Meidl, J.A., and A.R. Wilhelmi, PACT™/Wet Oxidation: Economical
Solutions to Solving Toxic Waste Treatment Problems. Paper presented at
Indiana Water Pollution Control Association Annual Meeting,
August 20, 1985.
20. California Department of Health Services, Alternative Technology for
Recycling and Treatment of Hazardous Wastes, Second Biennial Report,
July 1984.
21. Miller, R.A., and M.D. Swietoniewski. IT Enviroscience The Destruction
of Various Organic Substances by a Catalyzed Wet Oxidation Process, Work
Done Under U.S. EPA Contract No. 68-03-2568, 1982.
22. Randall, T.L. Wet Oxidation of PACTR Process Carbon Loaded with Toxic
Compounds. Paper presented at 38th Industrial Waste Conference, Purdue
University, West Lafayette, Indiana, May 10-12, 1983.
23. Wilhelmi, A.R., and P.V. Knopp. Wet Air Oxidation - An Alternative to
Incineration, Chemical Engineering Progress, August 1979.
8-30
-------
8.2 SUPERCRITICAL FLUID OXIDATION
Supercritical fluid oxidation is a technology that has been proposed for
the destruction of organic contaminants in wastewaters. It is basically an
oxidation process conducted in a water medium at temperatures and pressures
that are supercritical for water; i.e., above 374°C (705°F) and
218 atmospheres. In the supercritical region, water exhibits properties that
are far different from liquid water under normal conditions; oxygen and
organic compounds become totally miscible with the supercritical water (SCW)
and iaorganic compounds, such as salts, become very sparingly soluble. When
these materials are combined in the SCW process, organics are oxidized and any
inorganic salts present in the feed or formed during the oxidation are
precipitated from the SCW.
The oxidation reactions proceed rapidly and completely. Reaction times
are less than I minute, as compared to reaction times of about 60 minutes used
in the subcritical wet air oxidation (WAO) process. Moreover, the reaction is
essentially complete. Carbon and hydrogen atoms within the organic
contaminants are reacted to form CO- and H^O (residuals such as the low
molecular weight organic acids and alcohols found in the treated WAO effluent
are not found in the SCW process effluent). Heteroatoms (e.g., chlorine and
sulfur) are oxidized to their corresponding acidic anion groupings. These
anions, and those occurring naturally in the feed, can be neutralized by
cation addition to the feed, and the total inorganic content of the waste,
save that soluble in the SCW, can be precipitated and recovered by mechanical
separators operating at SCW conditions.
8.2.1 Process Description
In the supercritical region, water exhibits properties that are far
different from liquid water at normal ambient conditions. The density,
dielectric constant, hydrogen bonding, and certain other physical properties
change significantly with the result that SCW behaves very much like a
moderately polar organic liquid. Thus, solvents such as n-heptane and
benzene, for example, become miscible with SCW in all proportions. On the
8-31
-------
other hand, the solubility of salts such as sodium chloride (NaGl) is as low
as 100 ppm and that of calcium chloride (CaCl~) as low as 10 pptn. This is
the reverse of the solubilities in water that are found under ambient
conditions where the solubilities of NaCl and CaCl, are about 37 weight
percent and up to 70 percent, respectively (Josephson, 1982).
The solubility characteristics of SCW are strongly dependent upon
2
density. A temperature-density diagram is shown in Figure 8.2.1. The
critical point which is located on the dome of the vapor-liquid saturation
curve is at 374°C and 0.3 gram/cubic centimeter. The supercritical region is
that above 374° and the 218 atmosphere isobar. Near the critical point
(e.g., between 300° and 450°) the density varies greatly with relatively small
changes in temperature at constant pressure.
Insight into the structure of the fluid in this region has been obtained
from measurements of the static dielectric constant, values of which are shown
in Figure 8.2,1. * The dielectric constants of some common solvents are
given for comparison in Table 8.2.1.
TABLE 8.2.1. DIELECTRIC CONSTANTS OF SOME COMMON SOLVENTS
Carbon dioxide
n-Hexane
Benzene
Ethyl ether
Ethyl acetate
Benzyl alcohol
Ammonia
Isopropanol
Acetone
Ethanol
Methanol
Ethylene glycol
Formic acid
1.60
1.89
2.28
. 4.34
6.02
13.1
16.9
18.3
20.7
24.3
32.6
37.
58.
Source: Reference 5.
The dielectric constant is a measure of the degree of molecular
association. While dielectric constant is not the sole determinant of
solubility, the solvent power of water for organics is consistent with
variations in the dielectric constant. According to Figure 8.2.1, as
8-32
-------
600 -
500 -
O
o: 40°
of 30°
200 -
100 -
0.05
0.2
DENSITY (g/cnr5)
0.5 !.0
Figure 8.2.1.
Temperature-Density Diagram
Source: Reference 2
8-33
-------
temperature rises along the saturated liquid-vapor curve the dielectric
constant (normally at about 80 due largely to strong hydrogen bonding)
decreases rapidly despite only small changes in density. The large decreases
in the dielectric reflect the strong dependence of hydrogen bonding forces on
distance, with small decreases in density leading to large decreases in
3
dielectric constant. At 130°C (d= 0.9 g/cm ), the dielectric constant is
3
about 50, which is near that of formic acid; at 260°C (d=0.8 g/cm ) the
dielectric constant is 25 similar to that of ethanol. At the critical point
Che dielectric constant is 5, and little, if any, residual hydrogen bonding is
present. The major contribution to the dielectric constant is due to
dipole-dipole interactions, which gradually decrease with density.
Depending upon the pressure and temperature, the dielectric constant can
be varied to achieve values similar to those of moderately polar to nonpolar
organic solvents. Solubility behavior parallels the changes in dielectric and
at some points supercritical conditions are reached and the components are
miseible in all proportions.
The solubilities of inorganic salts in water exhibit different behavior
fron that shown by the organic compounds. At 250 atmospheres, the
solubilities of salts reach a maximum at 350-400°C. Beyond the maximum, the
solubilities drop very rapidly with increasing temperature. For example, NaCl
solubility is above 40 weight percent at 300°C and 100 ppm at 450°C; CaCl2
has a maximum solubility of 70 percent at subcritical temperatures which drop
to 10 ppm at 500°C.2
The properties of water, as a function of temperature, are summarized in
Figure 8.2.2. The figure shows that water goes through a complete reversal in
solubility behavior between 300—500°C. Above 450°C, inorganic salts are
2
practically insoluble, and organic substances are completely miscible.
Given the complete miscibility of oxygen and organic contaminants in the
supercritical fluid and the high temperature of operation, oxidation reactions
proceed rapidly and completely. In the MODAR process described below,
organics, air and water wastes are brought together at 250 atmospheres and at
temperatures above 400 G. The heat of oxidation is released within the
fluid and results generally in a rise in temperature to 600-650°C.
8-34
-------
(oiling
Point
*f 77 212
1.00
Critical
(00
800
1000
*C 25 100
koriul
200
100 r
SO •
20 r
Figure 8.2.2.
Properties of water at 250 a tin
Source Reference 2
8-35
-------
The products of supercritical water reforming are subjected to oxidation
while under these homogenous (i.e., single phase) supercritical conditions.
The residence time required for oxidation is very short, which greatly reduces
the volume of the oxidizer vessel.
When toxic or hazardous organic chemicals are subjected to SOW oxidation,
carbon is converted to C02 and hydrogen to H?0. The chlorine atoms from
chlorinated organics are liberated as chloride ions. Similarly, nitrogen
compounds will produce nitrogen gas, sulfur is converted to sulfates,
phosphorus to phosphates, etc. Upon addition of appropriate cations (e.g.,
Na+, Mg-n-, Ga-n-), inorganic salts are formed.
The heat of oxidation is sufficient to bring the supercritical stream to
temperatures in excess of 550 C. At these conditions, inorganic salts have
extremely low solubilities in water. Inorganic salts are precipitated out and
readily separated from the supercritical fluid phase. After removal of
inorganics, the resulting fluid is a highly purified stream of water at high
temperature and high pressure. The fluid is used as a source of
high-temperature process heat by generating steam.
A schematic flow sheet for the MODAR process as applied to liquid wastes
is presented in Figure 8.2.3. This figure and subsequent discussion was
provided by MODAR, Inc., (Reference 6).
The process consists of the following steps:
A. Feed
1. Organic waste materials in an aqueous medium are pumped from
atmospheric pressure to the pressure in the reaction vessel.
2. Oxygen, stored as a liquid, is pumped to the pressure of the
reaction vessel and then vaporized.
3, Feed to the process is controlled to an upper limit heating value of
1800 BTU/lb by adding dilution water or blending higher heating
value waste material with lower heating value waste material prior
to feeding to the reactor.
8-36
-------
L
--""-
----
•\
^
e
Aqueous
Waste
/ 1
c/
Auxiliary
loustie
1 Oxygen
,. r~p -
I/
— .
r-T
k
I
Ste«n
?n^ «.., » 1
1 |
1 1 ' G"s
Oxldiier ]
Cas-Li
-------
4. When the aqueous waste has a heating value below 1800 BTU/lb, fuel
may be added in order to utilize a cold feed to the oxidizer.
5. Optionally for wastes with heating value below 1800 BTU/lb, a
combination of preheat by exchange with process effluent and fuel
additon, or preheat alone may be used.
6. When organic wastes contain heteroatoms which produce mineral acids,
and it is desired to neutralize these acids and form appropriate
salts, caustic is injected as part of the feed stream.
7. A recycle stream of a portion of the supercritical process effluent
is mixed with the feed streams to raise the combined fluids to a
high enough temperature to ensure that the oxidation reaction goes
rapidly to completion.
B. Reaction and Salt Separation
1. Because the water is supercritical, the oxidant is completely
miscible with the solution; i.e., the mixture is a single,
homogenous phase. Organics are oxidized in a controlled but rapid
reaction. Since the oxidizer operates adiabatically, the heat
released by the readily oxidized components is sufficient to raise
the fluid phase to temperatures at which all organics are oxidized
rapidly.
2. Since the salts have very low solubility in SCW they separate from
the other homogenous fluids and fall to the bottom of the separation
vessel where they are removed.
3. The gaseous products of reaction along with the supercritical water
leave the reactor at the top. A portion of the supercritical fluid
is recycled to the SCW oxidizer by a high temperature, high pressure
pump. This operation provides for sufficient heating of the feed to
bring the oxidizer influent to optimum reactor conditions.
8-38
-------
4. The remaining reactor effluent (other than that recycled) consisting
of superheated SOW and carbon dioxide is cooled in order to
discharge C0? and water at atmospheric conditions.
C. Cooling and Heat Recovery
1. Most of the heat contained in the effluent is used to generate steam
for use outside the MODAR Process.
2. The heat remaining in the effluent stream is used for lower level
heating requirements and is also dissipated.
D. Pressure Letdown
1. The cooled effluent from the process separates into a liquid water
phase and a gaseous phase containing primarily carbon dioxide along
with oxygen which is in excess of the stoichiometric requirements,
2. The separation is carried out in multiple stages in order to
minimize erosion of valves as well as to optimize equilibria,
3, Salts are removed from the separator as a cool brine through
multiple letdown stages and are either dried (and water recovered)
or discharged as a brine depending upon client requirements.
8.2.1.1 Pretreatment Requirements--
Very little information exists in the literature to assess pretreatment
requirements for the process and its feed streams. The process reportedly can
handle slurries, thus, filtration or some other solids removal process may not
be required or even desirable if the contaminant is partitioned in the feed
between the aqueous phase and the suspended solids. Similarly, the need to
remove inorganic constituents may not exist since these constituents will
precipitate under the supercritical conditions of operation and presumably
will be removed by the mechanical separator shown in Figure 8.2.3. Adverse
8-39
-------
effects such as interference with pump operations, abrasion of internal parts,
and fouling of internal surfaces resulting from existing or formed solids are
possible problem areas but were not considered such by MODAR .
8.2.1.2 Operating Parameters—
The operating conditions are specified by MODAR as follows:
• Form of Feed Materials: Aqueous slurry or solution of organics,
* Temperature Range: 400°-650°C (750°-1200°F)
* Pressure Range: 220-250 atm
• Residence Time Range: Less than one minute
* Energy Type and Requirements: Thermal, to reaction conditions, with
provisions for useful recovery of
latent heat of oxidation.
These conditions are capable of achieving destruction efficiencies in
excess of 99.999 percent. The technology should be applicable to all
solvent/ignitables considered in this TRD. The principal question related to
the applicability of the technology is associated with cost, including the
durability of the system under the harsh supercritical conditions.
8.2.1.3 Post Treatment Requirements—
Because the oxidation reactions go essentially to completion and
provision can be made for neutralization and removal of inorganic products and
feed stock components the post treatment requirements should be minimal. Off
gases from the subcritical treated effluent should be largely C02 and H»0
and liquid effluent residuals will consist mainly of dissolved salts ,at the 10
to 100 ppm levels.
Along with N9, N?0 may also be a possible off gas component from the
SOW oxidation of nitrogen containing organics. A possible N-O component
would not be considered an air contaminant since there is no evidence
involving it in the series of complex chemical reactions producing
photochemical smog.
8-40
-------
Apart from the modest impacts anticipated as a result of ^0 emissions
and the dissolved inorganic salt loading of the liquid effluent the only other
residual stream requiring possible attention is the largely solid inorganic
stream from the separator, EP toxicity could be a characteristic of possible
concern for some wastes.
8.2.1.4 Treatment Combinations-
SCW oxidation systems can be considered for aqueous waste streams
containing one or more weight percent of organic constituents. Below
1 percent, other treatment technologies such as adsorption appear to have a
cost advantage. The highest practical organic content again will depend upon
costs; specifically the cost of SCW oxidation versus incineration for wastes
in the 10 to 20 weight percent and higher range. Largely unproven, the SCW
oxidation system will, if cost effective, function as a finishing technology
discharging effluents that can be expected to meet acceptable levels of
discharge.
8.2.2 Demonstrated Performance
The destruction of organic contaminants is a function of reactor
temperature and residence time. MODAR reports that a reactor temperature in
the range of 600 to 650°C (1120° to 1200°F) and a 5 second residence time are
sufficient to achieve destruction efficiencies of 99.999 percent. Higher
temperatures could be used to reduce the residence time. However, at a
5 second reaction time, the reactor cost is a small fraction of total capital
cost and, thus, there is not much incentive to reduce reactor volume by
2
operating above 650°C.
Theoretically, increasing residence time will also result in increased
destruction efficiency. The oxidation kinetics appear to be first order in
organic concentration. Assuming perfect mixing and first order kinetics at
all concentrations, doubling the residence time could result in a doubling of
the destruction efficiency. Thus, a 99.999 percent efficiency could become
99.99999999 (ten nines).
8-41
-------
MODAR has conducted more than 200 laboratory (bench) and pilot plant
tests in order to study the technical feasibility of SCW oxidation for a
variety of organic contaminants. In most cases MODAR does not attempt to
measure destruction and removal efficiency to the greatest possible
precision. Test objectives are rather to measure the levels of organic carbon
in the liquid effluent, and in most cases, residual levels are below detection
limits of the analytical equipment. Consequently, destruction removal
efficiency, which may be claimed in many of MODAR1s tests, are limited to
between 99.9% and 99.99+% by precision of the analytical equipment (See
Reference 8). When the objective is to demonstrate the maximum degree of
waste destruction, richer feeds and more sensitive analytical equipment are
used. Tests of this sort (e.g., on dioxins) show destruction and removal
efficiencies of more than 99.9999% . Equal or greater destruction
efficiencies could be expected for the solvent and other low molecular weight
organic compounds of concern.
8.2,3 Costof Treatment
The most significant operating cost factor is the cost of oxygen
consumed. Although compressed air can be used as the source of oxygen, the
cost of power as well as the high capital cost of appropriate compressors has
led MODAR to use liquefied oxygen as the primary oxygen source. "Oxygen demand
and heat content of an organic waste are usually directly related, and
therefore the heating value of the waste and waste throughput can be used to
make a preliminary estimate of waste treatment costs.
Table 8.2.2 presents waste treatment costs based on an aqueous waste with
a 10 percent by weight benzene-equivalent and a heat content of 1,800 Btu/lb.
This is the optimal heat content of a cold feed for this process to attain a
reactor exit temperature of 600 to 650°C. Other factors on which the costs in
Table 8.2.2 are based are: the system is installed at the site of the waste
generator; the units are owned and operated by the waste disposer; and the
units are not equipped with power recovery turbines.
8-42
-------
TABLE 8.2.2. MODAR TREATMENT COSTS FOR ORGANIC CONTAMINATED AQUEOUS WASTES
Waste Capacity Processing Cost3
Gal /day
5,000
10,000
20,000
30,000
Ton/ day
20
40
80
120
$0
$0
$0
$0
$/gal
.75 -
.50 -
.36 -
.32 -
$2
$0
$0
$0
.00
.90
.62
.58
$/ton
$180 -
$120 -
$ 86 -
$ 77 -
$480
$216
$149
$139
aBased upon an aqueous waste with 1800 BTU/lb heating value (equivalent to a
10% organic waste). Does not include energy recovery value of approximately
$0.05 per gallon
Source: Reference 6
If the waste has a fuel value of greater than 1,800 Btu/lb, the cost will
be higher per unit of waste processed. In treating a waste with a higher
organic content, it is recommended that the waste is diluted to a 10 percent
benzene-equivalent. Therefore, the increase in cost will be in proportion to
the increase in organic content.
If the waste has a heat content of between 5 and 10 percent benzene-
equivalent, fuel can be added to the waste to bring the heat content up to
10 percent benzene—equivalent without appreciable cost increases. If,
however, the waste is very dilute (2 to 3 percent benzene-equivalent), it is
more economical to use a combination of fuel with regenerative heat exchange.
8.2.4 OverallStatus of Process
8.2.4.1 Availability--
A pilot plant with capacity to oxidize 30 gal/day of benzene equivalent
has been in operation at MODAR's laboratory as well as at a field site since
late 1984. As a result of these activities, the MODAR SCW oxidation process
has been declared commercial and design of the first plant is underway. The
plant will be installed late in 1987 and will treat 10,000 to 30,000 gallons
of aqueous waste per day.
8-43
-------
8.2.4.2 Application—
SOW oxidation would appear to be applicable to aqueous wastes containing
1 to 20 weight percent organics. As noted in previous discussions above,
complete destruction of all organic solvents/ignitables can be anticipated on
the basis of evidence presented by the developer. The high efficiency of
destruction can be related to the unique and stringent conditions associated
with SOW oxidation which unites oxygen and organic contaminants under
relatively high temperatures and pressures.
Restrictive waste characteristics have not been identified in the
literature as a problem. The effect of heteroatoms and their reaction
products can be anticipated and steps taken to essentially eliminate any
deleterious impacts. However, the applicability of solid content wastes to
SOT oxidation systems may be problematical. The effectiveness of removal of
precipitated inorganic salts by the mechanical separators proposed for the
MODAR system may also be a problem. In the absence of particle size and flow
and design data it is difficult to predict mechanical separator performance,
although separation should be enhanced under the low density SOW conditions.
If particles are present, abrasion problems could occur both within the
oxidation system and in any subsequent system designed to recover energy from
the treated stream.
Supercritical fluid technology is also being considered for a number of
applications other than that concerned with the destruction of organic wastes,
e.g., supercritical fluid extractions, including the extraction of adsorbed
components from granular activated carbon. Fluids such as C0_, ethane, and
ethylene can be used at critical temperature and pressure conditions which are
much less severe than those of SOW. However, no data were found which
relates the performance of such systems to the extraction of solvent from
wastes.
8-44
-------
8.2.4.3 Environmental Impacts—
Liquid, solid, and gaseous emissions are generated from the SCW oxidation
process. Gaseous emissions consist primarily of carbon dioxide with smaller
amounts of oxygen and nitrogen gas. Effluent gas cleaning is not required.
N20 is the most abundant nitrogen oxide in the atmosphere. It does not
appear to interact with the nitrogen dioxide photolytic cycle. Any N20
which might be in the gaseous effluent is not classified as an atmospheric
pollutant.
Solid emissions consist of the precipitated inorganic salts. If chlorine
compounds are processed, chloride salts are formed, and similarly sulfur is
2
converted to sulfates, and phosphorous to phosphates.
Liquid effluents consist of a purified water stream. Although no data
are available for solvent contaminants, six nines destruction has been
measured for dioxins. On the basis of these data it is anticipated that
solvents of concern will be found only at the ppb level.
8.2.4.4 Advantages and Limitations—
The developer states that the MODAR process for supercritical water
oxidation of organics is an improvement in:
• enhanced solubility of gases including oxygen and air in water,
which eliminates two-phase flow;
• rapid oxidation of organics, which approaches adiabatic conditions
as well as high outlet temperatures, and very short residence times;
• complete oxidation of organics, which eliminates the need for
auxiliary offgas processing;
• removal of inorganic constituents, which precipitate out of the
reactor effluent at temperatures above 450°C (840°F); and
• recovery of the heat of combustion in the form of supercritical
water, which can be a source of high-temperature process heat.9
The above advantages are generally relative to the wet air oxidation process
which could be considered as an alternative technology to SCW oxidation. The
limitations of the process have yet to be determined through commercial
operation. Potential limitations relate to cost and equipment limitations due
to the stringent temperature and pressure requirements.
8-45
-------
REFERENCES
1. Josephson, J. Supercritical Fluids. Environmental Science and
Technology. Volume 16, No. 10. October 1982.
2. Thomason, T. B. and M. Mode11. Supercritical Water Destruction of
Aqueous Wastes. Hazardous Waste. Volume 1, No. 4. 1984.
3. Quist, A. S. and W. L. Marshall. Estimation of the Dielectric Constant
of Water to 800°, J. Phys. Chem., 69, 3165. 1965.
4. Uematsu, M. and E. U. Franck. J. Phys. Chem. Reference Data, 9(4),
1291-1306. 1980.
5. Franck, E. U. Properties of Water in High Temperature, High Pressure
Electrochemistry in Aqueous Solutions (NACE-4). p. 109. 1976.
6. Sieber, F. MODAR Inc. Review of Draft Section, Supercritical Water
Oxidation. May 16 1986.
7. National Academy of Sciences, Medical and Biological Effects of
Environmental Pollution: Nitrogen Oxides. 1977.
8. Modell, M., 6. Gaudet, M. Simson, G. T. Hong, and K. Biemann.
Supercritical Water Testing Reveals New Process Holds Promise. Solid
Wastes Management. August 1982.
9. Freeman, H. Innovative Thermal Hazardous Waste Treatment Processes.
U.S. EPA, HWERL Cincinnati, Ohio. 1985.
8-46
-------
8.3 OZONATION
Chemical oxidation has the potential for removing from wastewaters
organic materials which are resistant to other treatment methods,
e.g., refractory materials which are toxic to biological systems. Ozone
(0~) is one of the strongest oxidants available, as shown in Table 8.3.1,
which lists the oxidation potential and relative oxidation power of a number
of oxidizing agents. Ozone, as an oxidant, is sufficiently strong to break
many carbon-carbon bonds and even to cleave aromatic ring systems. Oxidation
of organic species to carbon dioxide, water, etc., is not improbable if ozone
dosage and contact times are sufficiently high, although many compounds are
highly resistant to ozone degradation. These compounds, which include oxalic
and acetic acids, ketones, and chlorinated organics, are not affected
significantly by treatment conditions (1 to 10 mg/liter concentration levels
and 5 to 10 minutes contact times) normally used for treating drinking waters
3
or for disinfecting wastewaters.
Ozone has been used for years in Europe to purify, deodorize, and
disinfect drinking water. More recently, it has been used in the waste
treatment area to oxidize phenolic and cyanide wastewaters. Cost
considerations and mass transfer factors limit the use of ozonation to
applications involving 1 percent or lower contaminant concentration levels.
Since oxidation by ozone occurs nonselectively, it is also generally used only
for aqueous wastes which contain a high proportion of hazardous constituitents
versus nonhazardous oxidizable compounds, thus focusing ozone usage on
contaminants of concern. Ozonation may be particularly useful as a final
treatment for waste streams which are dilute in oxidizable contaminants, but
which do not quite meet effluent standards.
8.3.1 Process Description
Ozone is generated on site by the use of corona discharge technology.
Electrons within the corona discharge split the oxygen-oxygen double bonds
upon impact with oxygen molecules. The two oxygen atoms formed from the
molecule react with other oxygen molecules to form the gas ozone, at
equilibrium concentration levels of roughly 2 percent in air and 3 percent in
oxygen (maximum values of 4 and & percent, respectively). Ozone must be
8-47
-------
TABLE 8.3.1. 1ELATIVE OXIDATION POWER OF
OXIDIZING SPECIES
Species
Fluorine
Hydroxyl radical
Atomic oxygen
Ozone
Hydrogen peroxide
Perhydroxyl radicals
Permanganate
Hypochlorous acid
Chlorine
Oxidation
potential,
volts
3.06
2.80
2.42
2.07
1.77
1.70
1.70
1.49
1.36
Relative
oxidation
power3
2.25
2.05
1.78
1.52
1.30
1.25
1.25
1.10
1.00
aBased on chlorine as reference (=* 1.00)
Source: References 1 and 2.
8-48
-------
produced onsite (ozone decomposes in a matter of hours to simple, molecular
oxygen ) and ozonation is restricted to treatment of streams with low
quantities of oxidizable materials. Using a rule of thumb, two parts of ozone
are required per pound of contaminant. A large commercial ozone generator
producing 500 Ib/day of ozone could treat 1 million gallons/day of wastewater
containing 30 ppm of oxidizable matter, or equivalently, 3,000 gallons/day of
2
wastewater containing 1 percent of oxidizable matter. Extensive
information related to the generation of ozone and its application to the
treatment of industrial wastewaters can be found in References 5 through 9.
While direct ozonation of industrial wastewater is possible and is
practiced commercially, other technologies have been combined with ozonation
to enhance the efficiency and rate of the oxidation reactions. These
technologies, which supply additional energy to the reactants, involve the use
of ultraviolet light or ultrasonics. In all cases it is important that mass
transfer across the gas-liquid interface with one or more of the reactants be
facilitated to maximize reaction rates.
8.3.1.1 Pretreatment Requirements—
Due to the nonselective nature of the ozonation reactions it is important
that the concentration levels of nonhazardous, but oxidizable, contaminants in
the feed stream be reduced as much as possible prior to treatment. The strong
electrophilic nature of ozone imparts to it the ability to react with a wide
variety of organic functional groups, including aliphatic and aromatic
carbon-carbon double and triple bonds, alcohols, organometallic functional
groups, and some carbon—chlorine bonds. It is important to recognize that
many functional groups can be present which compete for the ozone reactant and
can add significantly to the cost of the treatment.
The waste to be treated should also be relatively free of suspended
solids, since a high concentration of suspended solids can foul the equipment
normally used to bring about contact between ozone and the aqueous phase
contaminants. When ozonation is combined with UV radiation or ultrasonics, a
high concentration of suspended solids also can impede the passage of UV
radiation or attenuate the energy supplied by ultrasonics to enhance the
oxidation rate.
8-49
-------
8.3.1.2 Operating Parameters—
To effectively bring about the reaction of ozone with reactive
contaminants, it is important that mass transfer of ozone and its reactants
through the gas-liquid interface be maximized. Also, to increase ozone
solubility in water, temperatures should be maintained as low as possible and
pressures as high as possible. However, conditions such as high temperature,
high pH» and high UV light flux favor ozone decomposition. Under these
conditions reactivity rates may increase, although costs may also increase due
to less efficient use of ozone. Decisions will have to be made on a
case—by—case basis to establish the most effective operating conditions*
Several commercial designs are available for the conduct of gas/liquid
reactions which bring reactants into contact as effectively as possible. The
types of reactor designs available range from mechnically agitated reactors to
more complex spray, packed, and tray type towers. Their advantages and
limitations are discussed in detail in many standard texts and publications
(for example, see References 2 through 5).
The process of UV/ozone treatment operates in the following manner. The
influent to the system is mixed with ozone and then enters a reaction chamber
where it flows past numerous ultraviolet lamps as it travels through the
chamber (use Figure 8.3.1). Flow patterns and configurations are designed to
maximize exposure of the total volume of ozone-bearing wastewater to the high
energy UV radiation. Although the nature of the effect appears to be
influenced by the characteristics of the waste, the UV radiation enhances
oxidation by direct dissociation of the contaminant molecule or through
excitation of the various species within the waste stream. In industrial
systems, the system is generally equipped with recycle capacity. Gases from
the reactor are passed through a catalyst unit, destroying any volatiles,
replenished with ozone, and then recycled back into the reactor. The system
has no gas emissions.
Another alternative process involves the coupling of ultrasonic energy
with ozonation. It has been shown that significant increases in the rate of
oxidation can be obtained by the use of ultrasonic energy as opposed to ozone
alone. Experimental details were not available in Reference 3, although
different oxidation pathways were reported operating in the presence or
absence of ultrasonics.
8-50
-------
UV lamps
Spent 03
Gas out
Solid state
controlled
gear pump
Figure 8.3.1.
Schematic of top view of ULTROX pilot
plant by General Electric (ozone sparging
system omitted) (Edwards, B.H., 1983).
Source: Reference 4.
8-51
-------
Regardless of the reaction mechanisms, there appears to be no doubt that
the combination of ozonation with either UV or ultrasonic excitation leads to
increased oxidation rates. Typical design data for a 40,000 gal/day UV/ozone
treatment process are shown in Table 8.3.2. The plant is designed to reduce a
50 ppm PCB feed concentration to a 1 ppm effluent.
As noted, both temperature and pH can have a significant effect on the
rate of ozonation. As an example, work reported in Reference 11 indicates
that the reaction rate of toluene in acidic solutions with pH values of 2
and 3, increases by a factor of two for a temperature increase of 10°C. In
neutral solutions, however, the rate increases about ten times for the same
temperature change of 10°C. The activation energies were calculated to be
11.22 kcal/g mole and 31.26 kcal/g mole respectively, for the acidic and
neutral conditions* Thus, the data show that the ozonation rate of toluene
increases with increasing temperature and decreasing acidity of the solution.
8.3*1.3 Post-Treatment Requirements—
Post-treatment of industrial wastewaters that Have been contacted with
ozone will involve elimination of residual ozone, usually by passing the
effluent through a thermocatalytic unit. Some by-product residuals may be
formed in the feed water and some contaminants, if present, will not undergo
reaction. Compounds considered unreactive include many chlorinated aliphatic
compounds. If these compounds are present in the waste, technologies other
than ozonation should be considered.
8.3.1.4 Treatment Combinations—
Apart from the employment of UV excitation and ultrasonics with the
ozonation process, ozonation can be considered as a finishing step for waste
streams which have been treated by other technologies, principally
biotreatment systems. It has also been tested with some success as a means of
enhancing biotreatability. Although the use of ozonation in combination with
other technologies such as biological treatment is a possible solvent waste
treatment alternative, it is not a demonstrated technology for industrial
wastewaters, despite its extensive use and success in treating and
disinfecting relatively clean drinking waters.
8-52
-------
TABLE 8.3.2. DESIGN DATA FOR A 40,000 GPD
(151,400 L/DAY) ULTROX PLANT
Reactor
Dimensions:
Meters (LxWxH) 2.5 x 4.9 x 1.5
Wet volume, liters 14,951
UV lamps:
Number of 65 watt lamps 378
Total power, KW 25
Ozone Generator
Dimensions:
Meters (LxWxH) r 1.7 x 1.8 x 1.2
gms ozone/minute 5.3
kg ozone/day 7.7
Total power, kW 7.0
Total energy required 768
(KWH/day)
Source: Reference 10.
8-53
-------
8.3,2 Demonstrated Performance
Data for the treatment by ozonation of contaminated solvents and other
low molecular weight organics wastes are sparse. However, relative activities
for several solvents and other low molecular weight organic compounds are
shown in Table 8.3.3. As noted, chlorinated aliphatics must be considered
nonreactive.
12
Another study of the ozonation of petrochemical waste streams
concluded that ozonation was feasible for complete oxidation of only one of
four waste streams studied. This was a stripped ethylene dichloride (EDO)
wastewater. Haste streams deemed unsuitable for treatment by ozonation were
streams from the manufacture of toluene diisocyanate, ethylene glycol, and
styrene. TOG destruction within the EDO stream was 82 percent of a 100 mg/L
feedstresm TOG level after 3 hours reaction time. A weight ratio of ozone to
TOG of 5.6 was required to achieve total Oxidation. Ozone/TOC weight ratios
were higher for the streams deemed not suitable for ozonation. However, an
improvement in biotreatability was noted for all streams.
13
In the case of UV-assisted ozonation, a California study reported
that trichloroethylene concentrations were reduced from 17 ppb to less than
0.1 ppb. Similar results were attributed to ultrasonic energy used in place
of UV radiation. Reference was also made to a UV ozone treatment system being
operated, by Boeing to treat methyiene chloride in water at the 4,000 mg/kg
concentration level.
8.3.3 Cost of Treatment
Table 8.3.4 lists the costs for a 40,000 GPD UV/Ozone plant for which
design data were shown in Table 8.3.2.. Cost estimates were based on
wastewater containing 50 ppm PCB, designed to achieve an effluent PCS
concentration of 1 ppm. Costs were considered to be competitive with
activated carbon. The unit cost for treatment of the waste is greatly
affected by whether or not the cost for a monitoring system is included. The
cost of PCB destroyed is in excess of $IO/pound. PCB data were used for
costing purposes because of its availability. However, the costs will
increase substantially if ozonation is to be used as treatment for a waste
containing 1 percent organic contaminants. This is 200 times the
8-54
-------
TABLE 8.3.3. OZONATION TREATMENT OF SOLVENTS AND IGNITABLES
Compound
Relative reactivity with ozone
Oxidation products
00
m
Priority Solvents
carbon tetrachloride
ehlorobenzene
p-cresol (p-methylphenol)
o-dichlorobenzene
methyl ethyl ketone
nitrobenzene
pyridine
te t rachloroethylene
toluene
trichloroethylene
trlchlorofluoromethane
xylenes (o-» m-, p-)
unreactive
intermediate C02» oxalic aeld»Cl~, o-» m-
& p-chlorophenols and
chlorotartarlc acid
fast; 21,67 mg/L - no aromatlcity
after 17 min. ozonation
100 mg/L - 95% destroyed In 15 min.}
100% destroyed In 30 min. with UV/03
very slow with 031 fast with UV/03 acetone, ethanol and acetate
fast with UV/03/ intermediate with
ozone alone
unreactive at low pH
unreactive
Intermediate
very slow
unreactive
fast
o-, m- & p-nltriphenols
(mostly p-)i all C02 after 50
min. UV/03; only oxalic acid
with ozone alone
C02, OCOOH, oxalic acid
C02, OCOOH, oxalic acid
CcontinuedT'
-------
fABLI 8,3.3 (continued)
Compound
Relative reactivity with ozone
Oxidation products
00
Other Solvents
aeetonitrlle
bis(cbloromethyl)ether
benzene
chloroform
dichlorodifluoromethane
1» 1-dichloroethatte
1t2-diehloroethane
1,1-dlchloroethylene
1,2~dichloroethylene
hexachloroethane
1,1,1,2-tetrachloroethane
1,1,2,2-tetrachloroethane
1,1,2-trichloroethane
probably fast
unreactive
slow
unreactive to ozone but stripped
unreactive
unreactive
unreactive
fast
fast
unreactive
unreactive
unreactive
unreactive
probably HOAc and NO 3
C02, oxalic acid
Cl"", probably COC12
(phosgene) and CX^; from
solution; reactive with ozone/UV
25% of theoretical Cl" after 2
hrs, of W/ozonation
(continued)
-------
TABLE 8.3.3 (continued)
Compound
Relative reactivity with ozone
Oxidation products
Ignltables
acetaldehyde
allyl alcohol
chloroacetaldehyde
formaldehyde
fast
probably fast
intermediate
fast
CH3GOOH
probably HCHO and
HOCH2COOH
C1CH2COOH
formic acid, then C(>2
00
Source: Reference 3
-------
TABLE 8.3.4. EQUIPMENT PLUS OPERATING AND MAINTENANCE
COSTS; 40,000 GPD UV/OZONE PLANT
Reactor $ 94,500
Generator 30,000
1124,500
0 & M Costs/Day
Ozone generator power $4.25
UV lamp power 15.00
Maintenance 27,00
(Lamp Replacement)
Equipment Amortization
(10 years @ 10%) 41.90
Monitoring labor 85.71
TOTAL/DAY $173.86
Cost per 1,000 gals
(3,785 liters) with
monitoring labor $4.35
Cost per 1,000 gals
without monitoring labor $2.20
Source? Reference 10.
8-58
-------
concentration used to develop the costs in Table 8.3.4 Assuming capital
equipment costs follow a simple "sixth-tenths" factor scaling relationship,
the costs of the reactor and generator would be about $3,000,000 (or 24 times
the costs shown in Table 8.3.4) for treatment of this higher concentration.
Scale factors would be variable for the operating and maintenance cost items
listed in Table 8.3.4. However, the net result of scale-up to handle the more
concentrated waste would drastically increase the cost/1,000 gallons treated,
but would also result in far lower costs when calculated on the basis of the
amount of contaminant destroyed. Costs of roughly $10/pound of contaminant
destroyed would be reduced to an estimated $l/pound, assuming comparable
efficiencies. Destruction efficiencies may be adversly affected at higher
concentrations due to mass-transfer and other considerations. Thus, the cost
benefits per pound of contaminant destroyed, as stated above, may not be fully
achievable. Ozone usage and the corresponding costs are dependent on the
concentration of oxidizable species in the waste stream. The amount of UV
radiation used depends on quantum yield which can vary widely depending upon
waste characteristics and process condition. An optimal tradeoff must be made
on the basis of pilot-scale or full scale test results.
8.3.4 Overall Status of Process
8.3.4.1 Availability—
Ozonation equipment is available commercially from several manufacturers
within the United States. The Chemical Engineers' Equipment Guide published
by McGraw Hill lists nine manufacturers of ozone generators and
10 manufacturers of ozonators. The latter classification includes firms that
usually provide the ozone generator, the reactor, and auxiliaries such as the
catalytic unit for destruction of ozone from the treated stream. The status
of UV/ozonation is far less advanced. Processes such as the Ultrox process
have been concerned with highly refractory compounds such as PCBs. Equipment
specifically designed and available for UV/ozonation of industrial
wastewaters, is not available as a standard commercial item.
8-59
-------
8.3.4.2 Application—
Ozonation generally cannot be used as a sole treatment technology for
wastes which are resistant to oxidation such as chlorinated aliphatic
hydrocarbon wastes, and for those wastes containing solvent contaminants which
form stable intermediates that are resistant to total oxidation. Ozonation
appears best suited for treatment of very dilute waste streams, similar to
those streams treated by the ozone based water disinfection processes now used
in Europe. It does not appear to be cost competitive or technically viable
for most industrial waste streams where organic concentration levels are
1 percent or higher. However, it may be viable for certain specific wastes
with high levels of a contaminant of special concern and high reactivity.
8.3.4.3 Environmental Impact—
Assuming adequate destruction of a contaminant by ozonation, the
principal environmental impact would appear to be associated with ozone in the
effluent vapor and liquid streams. However, thermal decomposition of ozone is
effective and is used commercially to destroy ozone prior to discharge.
Unreacted contaminants or partially oxidized residuals in the aqueous effluent
may be a problem necessitating further treatment by other technologies.
Presence of many such residuals will generally result in selection of a more
suitable alternative technology.
8.3.4.4 Advantages and Limitations—
fhere are several factors which suggest that ozonation may be a viable
1 4
technology for treating certain dilute aqueous waste streams: *
* Capital and operating costs are not excessive when compared to
incineration provided oxidizable contaminent concentration levels
are less than 1 percent.
» The system is readily adaptable to the onsite treatment of hazardous
waste because the ozone can and must be generated onsite.
* Ozonation can be used as a final treatment for certain wastes since
effluent discharge standards can be met.
• It can be used as a preliminary treatment for certain wastes.(e.g.,
preceeding biological treatment).
8-60
-------
However, there are limitations which often will preclude use of ozonation
as a treatment technology* These include:
• Ozone is a. nonselective oxidant; the waste stream should contain
primarily the contaminants of interest.
* Certain compounds because of their structure are not amenable to
ozonation,e.g., chlorinated aliphatics.
* Ozone systems are generally restricted to 1 percent or lower levels
of toxic compounds. The system is not amenable to bulky wastes.
* Toxic intermediates may persist in the waste stream effluent.
• Ozone decomposes rapidly with increasing temperature, therefore,
excess heat must be removed rapidly.
8-61
-------
REFERENCES
1. Prengle, H. W., Jr. Evolution of the Ozone/UV Process for Wastewater
Treatment. Paper presented at Seminar on Wastewater Treatment and
Disinfection with Ozone., Cincinnati, Ohio, 15 September 1977.
International Ozone Association, Vienna, VA.
2. Harris, J. C. Ozonation. In: Unit Operations for Treatment of
Hazardous Industrial Wastes. Noyes Data Corporation, Park Ridge, N.J.
1978.
3. Rice, R. G. Ozone for the Treatment of Hazardous Materials. In:
Water-1980; AICHE Symposium Series 209, Vol. 77. 1981.
4. Edwards, B. H., Paullin, J. N., and K. CoghIan-Jordan. Emerging
Technologies for the Destruction of Hazardous Waste - Ultraviolet/Ozone
Destruction. In: Land Disposal: Hazardous Waste. U.S. EPA
600/9-81-025. March 1981.
5. Ebon Research Systems, Washington, D.C. In: Emerging Technologies for
the Control of Hazardous Waste. U.S. EPA 600/2-82-011. 1982.
6. Rice, R. G., and M. E. Browning. Ozone for Industrial Water and
Wastewater Treatment, an Annotated bibliography. EPA-600/2-8U-142,
U.S. EPA RSNERL, Ada, OK. May 1980.
7. Rice, R* G., and M. E. Browning. Ozone for Industrial Water and
Wastewater Treatment, A Literature Survey. EPA-600/2-80-060. U.S. EPA
RSKERL, Ada, OK. April 1986.
8. International Ozone Institute, Inc., Vienna, VA. First International
Symposium on Ozone for Water and Wastewater Treatment. 1975.
9. International Ozone Institute, Inc., Vienna, VA. Second International
Symposium on Ozone Technology. 1976.
10. Arisman, R. K., and R. C. Musick. Experience in Operation of a UV-Ozone
Ultrox Pilot Plant for Destroying PCBs in Industrial Waste Effluent.
Paper presented at the 35th annual Purdue Industrial Waste Conference.
May 1980.
11, Kuo, C. H. Reactions of Ozone with Organics in Aqueous Solutions.
EPA-600/3-85-031. U.S. EPA, ORD, RTF. April 1985.
12. Coco, J. H., et al. Gulf South Research Institute, New Orleans, LA.
Development of Treatment and Control Technology for Refractory
Petrochemical Wastes. U.S. EPA, Ada, OK. EPA-600/2-79-080. April 1979.
13. Radimsky, J., et al. Recycling and/or Treatment Capacity for Hazardous
Wastes Containing Halogenated Organic Compounds. State of California,
Department of Health. September 1984.
8-62
-------
8.4 OTHER CHEMICAL OXIDATION PROCESSES
Chemical oxidation processes are potential options for the treatment of
hazardous organic wastes, including those containing solvent and ignitables.
Previously described oxidation processes such as wet air oxidation and
supercritical water oxidation can be considered to be a type of incineration,
since the ultimate reaction products of most organic constituents treated by
those processes are carbon monoxide and water. Other oxidation processes,
including the ozonation process described in section 8.3, generally do not
result in total destruction of organic waste constituents. They are processes
which are carried out at ambient or moderate conditions of temperature and
pressure, and except for a small number of readily oxidizable compounds, or in
cases where an activator such as ultraviolet light is used, residual
contaminants and oxidation resistant by-products remain after treatment.
These residuals will generally require additional processing. Thus, the
chemical oxidation processes described here are considered pretreatment
processes for compounds in aqueous wastes at the one percent or lower level.
They can also be used as a finishing or polishing step for very dilute waste
streams containing contaminants which are known to be amenable to treatment.
The specific features and applications of other oxidizing agents such as
hydrogen peroxide and potassium permanganates are discussed here. As shown in
Table 8.4.1, these compounds are relatively powerful oxidizing agents as
illustrated by their high oxidation potentials.
8.4.1 Process Description
As shown in Table 8,4.1, hydrogen peroxide, H_02, and potassium
permanganate, KMnO^, are both relatively strong oxidizing agents. Hydrogen
peroxide has been used to treat phenols, cyanides, sulfur compounds, and metal
ions in dilute waste streams. Potassium permanganate is primarily used in the
treatment of phenols. The choice of oxidant is dependent upon such factors as
toxicity, reaction rate, ease of removal of secondary products, simplicity and
cost.
8-63
-------
TABLE 8.4.1. RELATIVE OXIDATION POWER OF OXIDIZING SPECIES
Species
Fluorine
Hydroxyl radical
Atomic oxygen
Ozone
Hydrogen peroxide
Perhydroxyl radicals
Permanganate
Hypochlorous acid
Chlorine
Oxidation
Potential,
volts
3.06
2.80
2.42
2.07
1.77
1.70
1.70
1.49
1.36
Relative
Oxidation
Power*
2.25
2.05
1.78
1.52
1.30
1.25
1.25
1.10
1.00
*Based on chlorine as reference (•' 1.00).
Source: References 1 and 2.
8-64
-------
Oxidation with H^C^ is generally performed in the presence of a metal
catalyst. Typical catalysts include ferrous sulfate, iron wool, nickel salts,
and aluminum salts. The waste is heated and then treated with H~0? while
being agitated. The H909 oxidation tends to proceed quickly under basic
3
conditions. The feasibility of ultraviolet catalyzed H909 oxidation
4
has been studied, but it does not appear to be used on an industrial scale.
Potassium permanganate oxidation is favored under basic conditions.
Raising the pH to the optimum level is accomplished by the addition of lime,
soda ash, or caustic soda. Potassium permanganate has also been used in the
3
treatment of aldehydes, mercaptans, and unsaturated acids. A schematic of
a typical process utilizing pH modification is shown in Figure 8.4.1.
8.4.1.1 Pretreatment Requirements for Different Waste Forms and
Characteristics—
Possible pretreatment required, prior to oxidation with either H-O-
or KMnO,, is filtration to remove oxidizable solids from the waste stream.
This is necessary since these oxidizing agents are not selective in their
reaction with waste constituents.
As noted in Reference 6, chemical oxidation is best suited for aqueous
liquids containing less than one percent of the oxidizable compound. Violent
reactions may occur when oxidizing agents are added to significantly higher
concentrations of easily oxidizable organics. Strong oxidants are relatively
nonselective; therefore, any easily oxidizable material will react. In a very
qualitative way, the reactivity of selected organic compounds with respect to
oxidation is as follows:
1. High reactivity - phenols, aldehydes, aromatic amines, certain
organic sulfur compounds, e.g., thioalcohols, thioethers;
2. Medium reactivity - alcohols, alkyl-substituted aromatics,
nitro-substituted aromatics, unsaturated alkyl groups,
carbohydrates, aliphatic ketones, acids, esters, and amines; and
3. Low reactivity - halogenated hydrocarbons, saturated aliphatic
compounds, benzene.
8-65
-------
500 GALLON
CAUSTIC STORAGE
oo
o>
o>
500 GALLON
WASTE STORAGE
co ca
V
1,500 GALLON
ALKALINE
JACKETED REACTOR
1 1
HEAVY METAL
TREATMENT
I
POLISHING FILTER
EFFLUENT
Figure 8.4.1. Example process flowsheet - oxidation.
Source: Reference 5.
-------
Chemical oxidation is not suitable for wastes containing significant
amounts of compounds or properties that represent an excessive demand for
oxidant, such as BOD, COD, nitrogen, or phosphorus. If the waste matrix to be
oxidized contains a significant amount of material that is more easily
oxidized than the toxic constituents of concern, then oxidant demand exhibited
by the easily oxidized species must be met before oxidation of the
constituents of concern will take place. For this reason, oxidation often has
limited application to solvent containing sludges.
8.4.1.2 Operating Parameters—
Operating parameters common to the reactivity of most oxidizing agents
include temperature and pH. Both H»0? and permanganate'work best under
elevated temperatures. The optimum pH range for oxidation of phenols by
H_O_ in the presence of metal catalysts is 3 to 4, while that for KMnO,
. . 7
increases as pH increases up to a value of 9.5.
In the treatment of phenols, KMnO, cleaves the aromatic ring and forms
a variety of mostly aliphatic acids. The stoichiometric equation indicates
that somewhat more than 9 moles of KMnO, are needed to oxidize each mole of
phenol to COj* On a weight basis then, 16 parts KMnO, per part phenol
would be required for complete oxidation. However, many applications require
only that the phenols be degraded to less toxic acids. In this case, 6 to 7
3
ppm KMnO, per ppm phenol is sufficient to achieve 90% phenol removal.
Very simple equipment is required for chemical oxidation. This includes
storage vessels for the oxidizing agents and perhaps for the waste, metering
equipment for both streams, and vessels with agitators to provide contact
between the oxidant and the waste. Some instrumentation is required to
determine the concentrations of pollutants, pH, and the degree of completion
of the oxidation reaction. The process is usually monitored by an oxidation-
reduction (ORD) potential electrode.
8.4.1.3 Post-Treatment Requirements—
As noted, the chemical oxidation processes discussed here often do not
result in total destruction of the initial contaminants and their
by-products. As a result, further treatment may be necessary to reduce
residuals in the treated waste stream to acceptable levels. In addition, some
8-67
-------
of the residuals generated from chemical oxidation are attributable to the use
of additives such as lime, soda ash, or caustic soda used to raise the pH of
the waste.
As a result, solids removal is usually necessary prior to discharge or
further treatment. Another disadvantage of chemical oxidation for waste
treatment is that it can introduce new metal ions into the effluent.
Potassium permanganate used to treat wastes will be reduced to Mn09 in the
process. This can be reduced by filtration to levels less than 0,05 mg/1 in
the final effluent. On the other hand, oxidation by hydrogen peroxide adds no
harmful species to the final effluent (except perhaps excess peroxide) since
its product is water. However, the reaction of hydrogen peroxide with
chlorinated organics in the presence of UV light may create chloride ions, a
situation which may call for additional processing.
8,4.2 Demonstrated Performance
4
A recent study was conducted at the University of Connecticut to
investigate the destruction of halogenated aliphatics in water by ultraviolet
catalyzed oxidation using H^O. as the oxidant. The effectiveness of this
process was determined for typical halogenated aliphatics, including
trichloroethylene, tetrachloroethane, dichloromethane, chloroform, carbon
tetrachloride and ethylene dibromide. Conventional biological, physical and
chemical wastewater treatment methods are often ineffective in removing these
types of hazardous compounds. The UV catalyzed H2^2 system was
investigated as an alternative treatment method to the use of ozone and UV
light.
The chemistry of the H-0-/UV reaction involves generation of hydroxyl
radicals and other reactive species by the photochemical action of UV light on
H^OO* The hydroxyl radicals attack organic species by extracting a
hydrogen atom or by adding to the double bonds of unsaturated molecules. In
addition, the UV light may also activate certain organic species and make them
more susceptible to attack by hydroxyl radicals. Under suitable operating
conditions the final products are C02, H20, H and Cl , if chlorinated
organics are in the waste stream.
8-68
-------
The destruction of triehloroethylene (TCE) was studied In greatest
detail. The concentrations of TCE, initially 58 ppm, and H-O- were
followed with time and chloride ion was determined at the beginning and end of
each run. The synergistic effect of H-CL plus UV light on the destruction
of TGE is shown in Figure 8.4.2. With H202 and no UV light the reaction
rate was negligible. A moderate rate of reaction was achieved when the waste
was exposed .to UV light with no HLO,, present. In the H_02/UV system
all of the reacted chlorine was converted to chloride ions, showing that no
other chlorinated organics were formed.
Other factors studied were the effect of increasing the initial
concentration of contaminant, increasing the amount of H-O- used,
increasing the temperature in the reaction vessel, and increasing the pH.
Doubling the initial concentration of TCE decreased the rate of oxidation
although low levels of TCE were still achieved. The rate of reaction
increased with increasing initial concentration of H^Oj, probably as a
result of higher concentrations of hydroxyl radicals. The time needed to
reach any specified TCE level was approximately halved by each 10°C
temperature increase. However, the faster reaction rate was achieved at the
expense of larger peroxide consumption. The rate of reaction increased with
increasing pH over the range of 5.5 to 7.9, but the effect of pH was minimal.
The variables studied for the six remaining compounds were temperature
and the presence or absence of H-O-. As with TCE, there was a strong
synergistic effect when the compounds were exposed to both H^O^ and UV
light. However, the presence of ^O^ had a lesser effect on the
destruction of the other compounds. Temperature had a significant effect on
the rate of reaction for all compounds. Figure 8.4.3 compares the rate of
reactions for the compounds studied at 20°C and 30°C, respectively.
8.4.3 Cost of Treatment
Since H-O,, and KMn(L are not used on an industrial scale to treat
organic wastes, cost data were not available. However, in comparing
H«0«/UV treatment to ozone/UV treatment, it is clear that the former may
4
be more economically attractive because:
8-69
-------
0.01
10 20 30 40
TIME , Minutes
50
Figure 8.4.2.
Effect of H2<)2 alone, W alone, and H2<> plus OT on
decomposition of trichloroethylene (fCE) at 20"C, pH 6.8,
Initial ICE - 58 ppm. Initial H202/TGE - 4.5 mols/mol.
8-70
-------
1.0
0.5
=5 0.1
a
"5
g 0.05
O.OI
.Carbon letrachloride
.Tttraehloroethane
Trtctlloroethylene
--'• and
Tetrachlorosfhylene
60 120
TIME , Minutes
180
240
Comparison of rates of reaction of halogenated aliphatics at 20°C.
Carbon tetrachloride
0.01
Tetrochloroethone ond
Ethylene dibromide
40 80 120
TIME , Minutes
ISO
Comparison of rates of reaction of halogenated aliphatics at 30°C.
Figure 8.4.3. Rates of reaction of halogenated aliphatics at 20°C and 3tf°C.
8-71
-------
• The cost of H£02 is much less sensitive to the scale of
operation than ozone.
• ^2^2 can be stored for use on an intermittent basis according to
process demand.
• The H£02 solution can be ready mixed with wastewater, whereas
ozone gas must be transferred into the water by mass transfer from
the ozone gas.
Both oxidants are currently used in the treatment of phenols, so it can be
inferred that they are capable of destructing aromatics. Whether they are
able to treat aromatic solvents or ignitables has not been demonstrated.
8.4.4 Overall Status of Process
8.4.4.1 Availability/Application—
Technology for large-scale applications of chemical oxidation is well
developed and equipment requirements are straightforward and simple.
Application to industrial wastes is well developed for cyanides and for other
hazardous species in dilute waste streams (phenols, organic sulfur compounds,
etc.). Oxidation has limited application to slurries, tars, and sludges.
because other components of the sludge, as well as the material to be
oxidized, may be attacked indiscriminately by oxidizing agents; careful
control of the treatment via multistaging of the reaction, careful control of
pH, etc., are required.
The application of oxidation processes to the solvents and ignitables of
direct concern has not yet been established. Some level of destruction can be
expected but full destruction has generally not been realized except possibly
in the Reference 4 study. This study, performed at the University of
Connecticut, clearly showed that the H 0_/UV process is capable of
treating organic compounds, specifically halogenated aliphatics and
particularly the unsaturated chlorinated compounds. Apparently the hydroxyl
radicals and other reactive species generated by UV radiation attack the
double bonds more readily. Similar studies need to be performed for KMnO .
8-72
-------
8.4.4.2 Environmental Impacts—
The environmental impact of the processes discussed here relate to the
unreacted contaminants and by-products remaining in the waste stream.
Additional treatment usually will be required. Air emissions associated with
the use of hydrogen peroxide and permanganate oxidants will be minimal,
although some care must always be observed when the contaminants are high
vapor pressure solvents and ignitables.
8.4.4.3 Advantages and Limitations—
The advantages of the oxidation processes discussed here result from ease
and simplicity of operation. Disadvantages are the result of incomplete
destruction and the need for subsequent treatment of the oxidized waste stream.
8-73
-------
REFERENCES
1. Prengle, H. W. Jr. Evolution of the Ozone/UV Process for Wastewater
Treatment, Presented at Seminar on Wastewater Treatment and Disinfection
with Ozone. Cincinnati, Ohio. International Ozone Association. Venna,
Virginia. 15 September 1977.
2. Harris, Judith C. Ozonation. In: Unit Operations for Treatment of
Hazardous Industrial Waste, Noyes Data Corporation. Park Ridge, New
Jersey. 1978.
3. Hackman, E. Ellsworth. Toxic Organic Chemicals-Destruction and Waste
Treatment. Park Ridge, NJ, Noyes Data Corp., 1978.
4. Sundstrom, D.W., et. al. Destruction of Halogenated Aliphatics by
Ultraviolet Catalyzed Oxidation with Hydrogen Peroxide. Department of
Chemical Engineering, The University of Connecticut. Hazardous Waste and
Hazardous Materials, 3(1): 1986.
5. Chillingworth, M. A., et al. Industrial Waste Management Alternatives
for the State of Illinois, Volume IV - Industrial Waste Management
Alternatives and their Associated Technologies/Processes, prepared by GCA
Technology Division, Inc. February 1981.
6. U.S. EPA. Treatability Manual, Volume III. EPA-600/2-82-001.
September 1981.
7. U.S. EPA Background Document for Solvents to Support 40 CFR Part 268 Land
Disposal Restrictions, Volume II. January 1986.
8-74
-------
8.5 CHLORINOLYSIS PROCESSES
8.5.1 Process Description
Chlorine is a strong oxidizing agent and is often used in much the same
manner as other oxidants (e.g., ozone, hydrogen peroxide, and potassium
permanganates) to treat low levels of organic contaminants in aqueous waste
streams. However, in the process termed chloronolysis (or chlorolysis in
Germany), chloride is introduced to the waste at high temperatures and
pressures. At temperatures above 500°C, under excess chlorine conditions, the
carbon-carbon bonds of hydrocarbons can be broken and the molecular fragments
can react with chlorine to form low molecular weight chlorinated
hydrocarbons. It is essentially a pyrolytic process carried out in the
presence of chlorine. Typical chlorinolysis reactions are shown in
Table 8.5.1. As shown in the table, the mole ratio of HC1 to carbon
tetrachloride (CC1,) produced varies from zero to two. This ratio can be
compared to a ratio of four for the process based on the direct chlorination
of methane, and will generally impact favorably on subsequent purification and
disposal operations required by the two processes.
Several companies have developed manufacturing processes that are capable
of converting C,, C™, and C3 hydrocarbons and their partially
chlorinated derivatives to chlorinated solvents such as trichloroethylene,
2
perchloroethylene, and carbon tetrachloride. Another company,
Hoechst-Uhde, has patented a process in Germany for converting C to C,
organochlorine compounds to carbon tetrachloride. The operating
characteristics of the four chlorinolysis processed are summarized in
Table 8.S.2.2'3
The Hoechst-Uhde process is the only chlorinolysis process which appears
capable of handling aromatic feedstocks. The higher reactor temperatures
(600°C) and pressures (20 MPa) used in the Hoechst-Uhde process apparently
promote the breakdown of the benzene ring. Two of the other processes operate
at temperatures around 400°C, and at this temperature are not capable of
destroying hexachlorobenzene which is formed as a breakdown product of the
4
benzene ring. The third American process, although it reportedly operates
at 600°C, does not bring about destruction of the hexachlorobenzene, probably
because of less stringent pressure conditions.
8-75
-------
TABLE 8,5.1. TYPICAL CHLORINOLYSIS REACTIONS
00
I
Organic feedstock Process
Hexachloroethane CCl^-CCLvt-Cl--*- 2CC1,
1,2-dichloroethane CH.Cl-CH-Cl+SCl-^- 2CC1/+4HC1
1,2-dichlorobutane CH2CL-CHC1-CH3+8C12 "*3CC14+6HC1
Hexachlorobutadiene CC1 =CC1-CC1=CC1 +5C1 -* 4CC1,
Benzene C,H,+15C1_-* 6CC1.4-6HC1
DO i 4
Hexachlorobenzene C,C1,+9C1_^- 6CC1,
o o i f\
Enthalpy,
AH°298°Kskcal/mole
•*•
15
106
142
96
300
140
Mole ratio,
HC1/CC14
0
2
2
0
1
0
Source: Reference 1.
-------
TABLE 8.5.2. OPERATING CHARACTERISTICS OF CHLORINOLYSIS PROCESSES
Process
Developer
Feedstock
Operating
Conditions
Product Yield
Comments
00
i
Dow
Chemical
Diamond
Al kali-
Pi tts burgh
Plate
Glass
Hoechst
Uhde
Cu Cg, Ca hydro-
carbon products &
their partially chlo-
rinated derivatives.
Ethylene dichloride.
Ethylene dichloride
and other G£
chlorohydrocarbons.
C-| to Cg organo-
chlorine compounds.
600°C
400°C,
using
Fuller's
earth as
catalyst in
fluid bed
400°C, oxy-
chlorination
in a fluid
bed catalytic
reactor
600°C and
20 MPa
94-951 of perchloro-
ethylene and carbon
tetrachloride, 6% of
hexachlorobenzene.
90% yield of perchlo-
roethylene and
trichloroethylene; the
balance estimated to be
hexachloroethane,
hexachlorobutadiene,
hexachlorobenzene 5
tetrachloroethane,
pentachloroethane. •
85% yield of trichloro-
ethylene and perchloro-
ethylene; the balance
probably carbon oxides,
and chlorohydrocarbons
such as hexachloro-
butadiene and
hexachlorobenzene,
>95% yield of carbon
tetrachloride per pass.
Heavy ends consist
chiefly of hexachloro-
benzene.
Hexachlorobenzene is one
of the hard to treat
wastes.
The tetrachloroethane &
pentachloroethane can be
recycled and pyrolized
to trichloroethylene.
The hexachlorobutadiene
& hexachlorobenzene are
expected to be residues
from this process.
Relatively low yield to
useful products. The
hexachlorobutadiene and
hexachlorobenzene formed
are expected to be
residues from this
process.
Very high overall yield
to carbon tetrachloride
if the presence of oxy-
genated chlorohydrocarbons
in the feedstocks is'
limited. The hexachloro-
benzene formed can be
recycled to extinction.
Source: References 2 and 3.
-------
A schematic of the Hoechst-Uhde process is shown in Figure 8.5.1. The
organic feedstock and preheated chlorine are introduced into a high purity
seamless nickel tube reactor which is surrounded by a stainless steel jacket
to withstand the 20 MPa pressures. In the primary section of the reactor, the
feed is heated and the reaction is initiated. Since the reaction is
exothermic, the electric heaters can be turned off after initiation, and the
reaction is allowed to proceed to an adiabatic end temperature of 620°C. The
final temperature is regulated by chlorine addition which must be 20 percent
in excess of theoretical, and is normally maintained at 50 percent in excess
of theoretical. Quenching from 620°C to 500°C takes place in the final
section of the reactor by injection of cold carbon tetrachloride bled from the
3
output stream.
The remainder of the process consists of distillation columns where
reaction products are separated, resulting in CC1, and HC1 streams, and also
a waste stream consisting of phosgene (from oxygen-bearing organics), and
chlorine.
As noted in Reference 1, in principle any liquid chlorinated hydrocarbon
mixture can be used as a feedstock for the Hoechst process. Suitable
feedstocks include residues from the production of vinyl chloride monomer,
chloromethanes, propylene oxide, allyl chloride, perchloroethylene and benzene
3
chlorination. Other wastes which have been considered as feedstocks for
chlorinolysis are pesticide wastes. Of 20 pesticides considered, however,
only 13 contain the desirable elements: carbon, hydrogen and chlorine. These
include aldrin, chlordane, DDT, ethylene dichloride, benzene hexachloride,
heptachlor, landane, o- and p-dichlorobenzene, and perthane. In a previous
study, it was determined that nine of these compounds are manufactured in too
low a volume (1 MM Ib/yr) to be considered as significant feedstocks. Three
others, aldrin, chlordane, and DDT were produced in large volumes, but have
since been banned by the EPA. The conclusions of the study were that the
pesticides industry did not produce significant quantities of waste suitable
2
for chlorinolysis.
8-78
-------
-10 C
*• HCI
HAZARDOUS
WASTE
MATERIALS
00
«-4
VO
WASTE
WATER
REACTOR
HIGH-BOILER CRUDE C(
COLUMN COLUMN
HCI
COLUMN
COCL2
CRUDE CCI4 CAUSTIC SCRUBBER
COLUMN SEPARATOR
DRYER
Figure 8.5.1. Schematic diagram of the Hoechst AG chlorolysis process.-
-------
8.5.1.1 Pretreatment Requirements—
The feed stream for the Hoechst process should be free of solids to both
ensure reaction and to reduce the potential for fouling critical equipment,
including pressure control valves and pumps. The particle size of the
particulates present in waste hydrocarbon streams (e.g., from ethylene
dichloride pyrolysis) are reportedly too fine to remove by conventional
filtration. Consequently, an evaporation or distillation method appears more
2
suitable. Fractional distillation is not required.
The feed streams for the Hoechst process may contain up to 5 percent
nonchlorinated aromatic hydrocarbons. Higher concentrations make it difficult
to limit the design temperature to a maximum of 620°C. Small quantities of
oxygen containing organic compounds can also be tolerated, although the
phosgene by-product produced by chlorinolysis of such compounds can create
serious residual disposal problems.
The presence of elements, other than carbon or hydrogen in a compound,
also could result in handling and corrosion problems. Small amounts of
sulfur, for example, lead to corrosion of any nickel-containing material.
2 3
Therefore, sulfur content should be kept below 25 ppm. '
Nitrogen and phosphorus are also restricted from chlorolysis feedstocks
to guard against the possible formation of nitrogen trichloride (NCI,,) and
phosphorus trichloride. The first of the compounds is explosive, and the
2 3
second is pyrophoric. ' The effect of the presence of inorganic
contaminants such as metals and water is not clear. According to a
representative of the Pittsburg Plate Glass Co., their process is capable of
treating metals but will not handle any water or alcohols.
8.5.1.2 Operating Parameters—
As noted in Table 8.4.2, the Hoechst process operates at temperature and
pressure conditions that are higher than those used in the three United
State's processes. Conditions for operation of the Hoechst process are
further shown in Figure 8.5.1 and described in more detail in References 1
through 5. Graphical representations of the amount of chlorine consumption
for feedstocks of various C1:H:C mole ratios are provided in References 1
8-80
-------
and 7. As noted In Reference 1, exceed chlorine is used in the Hoechst
process in order to operate well outside the explosive range, and to maintain
final temperatures at levels below 620°C to avoid corrosion and other problems,
8.5.1.3 Post-treatment Requirements—
The products generated from the Hoechst-Uhde process are carbon
tetrachloride, hydrochloric acid, and carbonyl chloride (phosgene) if oxygen
is in the feed. The percent yield as a function of feedstock composition is
not known. However, it is assumed that the purity of product is sufficient to
meet that required for fluorocarbon production since about 80 percent of the
carbon tetrachloride produced in the United States is used in the manufacture
of Freon-11 and 12 for refrigeration and propellant usage. Thus, disposal
would not be a problem except in the case of off-specification product.
The hydrochloric acid produced during the chlorinolysis process could be
utilized onsite at a ehloro-organic manufacturing facility. For example, the
options available include oxyhydrochlorination of ethylene to produce saleable
ethylene dichloride or conversion of the acid back to chlorine. Phosgene
produced could be sent offsite for use in the manufacture of isocyanates,
carbonates and polycarbonates provided chlorine, hydrochloric acid, and carbon
2
tetrachloride are present at acceptably low levels.
The Hoechst process, as shown in Figure 8.5.1, utilizes absorption units
to treat waste exit gases. An incineration section (not shown in the figure)
equipped with a scrubber is also provided to dispose of pretreatment residues
which are not fed to the chlorinolysis unit and all wastewater streams
contaminated with traces of chlorinated hydrocarbons. Most trace inorganics,
including metals present in the feedstock, will be separated in the
pretreatment step, and will be found in the incinerator combustion products.
8.5.2 Demonstrated Performanee
The amount of performance data available are minimal, with available
performance data previously summarized in Table 8.4.2. The Hoechst-Uhde
process appears to be the most viable of the four processes because it is able
to process a wide range of wastes, achieves the highest yield of carbon
tetrachloride, and creates a minimal amount of undesired residues.
8-81
-------
8.5.3 Cost of Treatment
The annual operating cost data shown in Table 8.5.3 are based on data
provided in Reference 3 for March 1978. These data have been updated to
reflect early 1986 costs. The adjustments were made using the Chemical
Engineering Plant Cost Index to estimate capital investment costs and a
variety of other sources to determine 1986 material, utility, and labor
costs. The 1986 capital investment cost for the chlorolysis plant processing
25,000 metric tons/year of organic chlorine wastes is estimated to be
$40,000,000, up from about $27,000,000 in 1978. This cost was based on a site
in the Gulf Coast area, with land and startup costs excluded.
Analysis of the cost data as initially provided by Hoechst-Uhde
(Reference 8) indicates that the economic feasibility of using chlorolysis as
a waste disposal alternative is primarily dependent upon the selling price of
carbon tetrachloride and to a lesser extent upon the type of waste available
as feedstock. The annual organic costs provided in Table 8.5.3 were based on
a mixed waste considered by Hoechst to be the base case since a higher solvent
waste (high chloride) content would require preheating of the chlorine
reactant and lead to possible corrosion and material problems.
The annual operating cost of $26,552,000 for the base case will be about
25 percent lower than that required for a feedstock consisting entirely of
vinyl chloride (VC) monomer waste, due to the difference in chloride
requirement. Although this increased costs for the VC waste could be offset
by the increased product credits from the sale of carbon tetrachloride (and
hydrochloric acid), the return cannot be guaranteed.
The volume of carbon tetrachloride produced by the proposed plant
represents about 25 percent of the total market volume, a level which would
cause major perturbations in the price structure. The process, however,
should be competitive with incineration at existing price levels of about
$500/ton of carbon tetrachloride.
8-82
-------
TABLE 8.5.3. ANNUAL OPERATING COST FOR PROCESSING 25,000 METRIC TONS/YEAR
OF A MIXED VINYL CHLORIDE MONOMER AND SOLVENT WASTE AT A
CHLOROLYSIS PLANT
Item ' ' Cost^~J7year
Chemicals
Chlorine: 68,000 metric tons @ $220/metric ton 14,960,000
20% Caustic: 14,500 metric tons <§ $70/metric ton 1,015,000
Methane: 134,500,000 ft3 @ $6/1000 ft3 807,000
Utilities
Electric power: 25,600,000 Kw-hr § $.05/Kw-hr 1,290,000
Steam: 52,000 metric tons @ $13.2/metric ton 690,000
Cooling water: 3.9 x 109 gallons <§ $0.10/1000 gallons 390,000
Operating Labor
Operator, 10 men/shift, tl2.00/manhour 1,050,000
Direct supervision, 4 men, $15.00/manhour 125,000
Maintenance
Maintenance labor, 2% of Total Plant Cost 800,000
Maintenance supply, 2% of Total Plant Cost 800,000
Direct Overhead
(30% of Operating Labor and Supervision) 350,000
General Plant Overhead
(50% of Operating and Maintenance Labor and
3% of Total Plant Cost) 3,175,000
Taxes and Insurance
(1.5% of Total Plant Cost) 600,000
Royalty 500,000
Net Annual Operating Cost 26,552,000
SOURCE: Reference 3
8-83
-------
8.5.4 Overall Status of Process
8.5.4.1 Availability/Application—
Chlorinolysis is primarily used as a manufacturing process. Its
potential as a waste treatment process is seriously limited by its prohibitive
capital cost and its dependency on the carbon tetrachloride market. In their
feasibility studies of the Hoechst-Uhde chlorinolysis process, both reference
2 and 3 investigators concluded that a regional waste treatment facility would
be a possibility. Their survey indicated that such a facility would be best
3
located in the Gulf Coast region.
The Hoechst-Uhde process is capable of treating C. to C,.
organochlorine compounds, including aromatics. Suitable feedstocks include
those residues from the production of VCM, chloromethane, propylene oxide,
3
allyl chloride, perchloroethylene, and residues from benzene chlorination.
8.5.4.2 Environmental Impact—
The air emissions and wastewater discharges from a Hoechst-Unde
chlorinolysis system processing 25,000 metric tons of waste/year were
estimated in Reference 3. The air emissions of volatiles including carbon
tetrachloride, phosgene, and chlorine from the plant were less than 1 ton/year
and were considered environmentally acceptable. Waste water emission data
were based on concentrations and flows resulting from scrubbing operations.
These also were considered to be insignificant from the standpoint of
environmental impact.
8.5.4.3 Advantages and Limitations
Chlorinolysis is not a process that will be used by any waste processing
facility; however, it has excellent resource recovery potential. Limitations
are high capital and operating costs and uncertainties regarding the
availability of feedstock and the economics of cost recovery through product
sales.
8-84
-------
REFERENCES
1. Berkowitz, J. B. Chlorinolysis. In: Unit Operations for Treatment of
Industrial Hazardous Wastes. Noyes Data Corporation, Park Ridge, New
Jersey, 1978.
2. Shiver, James K., Repro Chemical Corporation. Converting
Chlorohydrocarbon Wastes by Chlorinolysis. EPA-600/2-76-270,
U.S.EPA/ORD, Washington, DC. OCtober 1976.
3. Shih, C.C., et al., 1RW, Inc. Comparative Cost Analysis and
Environmental Assessment for Disposal of Organochlorine Wastes.
ledondo Beach, CA. EPA-600/2-78-190, U.S.EPA/ORD, Washington, DC.
August 1978.
4. Arienti, Mark, Thomas Nunno, et al., GCA Technology Division, Inc.
Technical Assessment of Treatment Alternatives for Wastes Containing
Halogenated Organics. Bedford, MA. U.S. EPA/OSW, Washington, DC,
October 1984.
5. Krekler, H. H., Schmitz, and D. Rebhan. The High-Pressure-Chlorolysis of
Hydrocarbons to Carbon Tetrachloride. Paper presented at the National
Conference on the Management and Disposal of Residues for the Treatment
of Industrial Wastewaters. Washington, D.C. February 3-5, 1975.
6. Conversation with Mr. Ken Lee, Pittsburgh Plate Glass Co.,
Pittsburgh, PA. 2 May 1986.
7. Krekler, H. and H. Weber. Chlorolysis of Aliphatic and Aromatic
Compounds to Carbon Tetrachloride. Paper presented at the ACS meeting in
New York. August 1972.
8. Hoechst-Uhde Corporation. Chlorolysis Applied to the Conversion of
Chlorocarbon Residue Possibly Containing Oxygenated Analogs. Draft Final
Report to U.S. EPA. October 1977.
8-85
-------
8.6 CHEMICAL DECHLORINATION
Chemical dechlorination methods have been developed as possible
alternatives to incineration or land disposal for halogenated organic
compounds such as PCBs. Researchers have found that in order to decrease the
degree of toxicity, as well as the chemical and biological stability of
chlorinated compounds, it is not necessary to totally break down the molecular
structure. Instead, the formation of a compound considered harmless and
environmentally safe can be achieved through a reaction system that will
result in the cleavage of C-C1 bonds or the rearrangement of the chlorinated
molecule. Although several different dechlorination methods exist, all of the
processes are based primarily on two technologies; the "Goodyear process"
developed by Goodyear Tire and Rubber, and the NaPEG system developed by the
Franklin Research Institute.
The Goodyear Process was originally developed to reduce PCS laden heat
transfer fluids from slightly above 500 ppm to less than 10 ppm. The reaction
chemistry is based on the use of a sodium-naphthalene reagent to form sodium
chloride and an inert, combustible sludge. The reagent is produced by
2
disolving molten sodium and naphthalene in tetrahydrofuran. However, the
reactivity of metallic sodium with water necessitates the use of an air free
anhydrous reaction vessel to prevent rapid generation of hydrogen or loss of
reagent through the formation of NaOH.
Since Goodyear has decided not to pursue the marketing of this process,
several companies such as SunOhio, Acurex, and PPM Inc, have entered the
field. Generally, they have modified the process by substituting proprietary
reagents for naphthalene, which is a priority pollutant. These processes are
also intended for treatment of PCB contaminated oils (50-500 ppm) and require
pretreatment to remove water and inorganics such as soil. Typically, these
processes cannot handle PCB concentrations greater than 10 percent, and most
are not suitable for sludges, soils, sediments, and dredgings. The exception
is the Acurex process which uses a relatively nontoxic solvent to extract PCBs
from contaminated soils and then destroy them with a proprietary reagent, (see
Figure 8.6.1). These treatment methods may have limited applicability for
other chlorinated organic compounds; however, more research is needed to
determine process feasibility.
8-86
-------
A more promising technology is the NaPEG process, originally developed in
3
1980 by Pytlewski, et al., at the Franklin Research Institute. The intent
was to devise a reaction system that would decompose PCBs and representative
halogenated pesticides in an exothermic and self-sustaining manner. The
dechlorination reagent was formed by reacting alkali metals such as sodium
with a polyethylene glycol (M.W. 400) in the presence of heat and oxygen.
The reaction mechanism involves a nucleophilic substitution/elimination and
the oxidative degradation of chlorine through the generation of numerous free
radicals. The process reactivity can be "tuned" or directed at various
aliphatic or aromatic systems by varying the molecular weight of the
polyethylene glycol. Typical by-products of the reaction are salts such as
sodium chloride, hydrogen, and hydroxylated organic derivatives.
Laboratory studies have shown a 99,99 percent reduction of PCBs in
dielectric transformer oils and 51.9 percent reduction in PCB contaminated,
low moisture soils. In addition to PCBs, the following organohalogens have
also been successfully treated by this method; hexachlorocyclohexane,
hexachlorobenzene, tri- and tetrachlorobenzenes, pentachlorophenol, DDT,
kepone, and chloroethylsulfide. The primary advantages of the NaPEG system
(which is generally referred to now as "APEG" - Alkali Polyethylene Glycoates)
is that the reagent is not based on a dispersed metallic sodium reaction, can
tolerate low levels of water content, and is stable in air. Therefore, the
process may by applicable to soils, dredgings, sediments, and low moisture
sludges. A summary of the chemical dechlorination methods currently available
is presented in Table 8.6.1
Two emerging technologies based on the APEG system are currently under
development at the Galson Research and the Sea Marconi Corporations. The
Galson Research process involves a series of processes for the degradation of
chlorinated benzenes, biphenyls and dioxins from contaminated soils. The
system, which was developed under EPA sponsorship, is based on the more
reactive KPEG (potassium—based) reagent, in conjunction with a sulfoxide
catalyst/cosolvent. A probable reaction scheme is presented in Figure 8.6.2.
Laboratory results of dioxin testing have shown that destruction efficiencies
of 99 percent or greater were obtained using a single soil sample containing
2 ppm of dioxin. Higher destruction efficiencies can be expected for solvents.
8-87
-------
00
00
00
son
washers
Clean solvent
storage
c
o
c e.
•—
r— U
O Ol
Concentrator
reactor
T
Acurex
reagent
Sludge (dioxin free)
Figure 8.6.1. Soil cleaning process schematic.
Source: Acurex Corporation (Cincinnati, Ohio).
-------
TABLE 8.6.1. DECHLORINATION PROCESSES
Process description
Compounds and -forms
of waste treated
Destruction Capabilities
Residuals
Comments
PC 8s
proprietary sodium reagent
used to strip away chlorine
oil is mixed with reagent
and sent to reactor
mixture is then centrifuged
degassed and filtered
« liquid hydrocarbon streams
i.e. PCB contaminated oil
from transformers
* cannot be used on aqueous
or soil wastes
•250 ppm PCB to 1 ppm
* 3000 pptn to below 2 with
several passes
metal chlorides
polvphenyls
treated oil
* mobile, continuous process
if moisture and contaminant
removal required as
pre-treatment
• moderate temperature and
pressure
* pure PCBs destroyed at
150 ail/rain
proprietary sodium reagent
used to strip chlorine
can he used on PCB contaminated * PCB feeds as high as 10%
oils and soils
• contaminated oil is filtered, • also effective on transformer
and mixed with reagent oil contaminated with
OS 2,3,7,8-TCDD
Jg * reaction takes place in
\0 processing tank
effectively treated
2,3,7,8-TCDD reduced
from 200-400 ppt to 40 ppt
treated oil
sodium hydroxide
effluent
polyphenoI sludge
9 mobile, hatch operation
* pretreatment needed to
remove water, aldehydes
and acids from transformer
01 Is
• non-tojcic solvent used to
extract PCBs from soil
APEC
i sodium polyethylene glycol
reagent (NaPEC) used for PCB
PCB oils and soils
TCDD contaminated soils
* Potassium Polyethylene Glycol
(KPSG) used for TCDD , also tested on hexachlorocyclo-
hexane, hexachlorohensene, PCP,
* reagent is added to contamin-
ated material in the presence
of air, and can he sprayed on
DDT, KEPOtJE. Tri- and Tetra-
chlorobenrenes
* PCB destruction 99.99%
9 2,3,7,8-TCDn reduced from
330 ppb to 101 pph
sodium chloride
oxgenated biphenyls
decontaminated
material
hydrogen gas
involves the application of
reagent in the presence of
air or oxygen
water increases reaction
times and decreases the
degree of chlorination
temps, above 100"C required
fast destruction
PPH
* proprietary sodium reagent
used for chlorine stripping
* reagent is added*to contam-
inated oil and left to react
* solid polymer formed is
filtered out
* PCB contaminated oil
t TCDD detoxification w
investigated soon
be
* aqueous waste and soil not
treated
• 200 ppm PCB reduced
to below 1 ppm
solid polymer
decontaminated oil
* mobile* batch process,
700 gal/hr
* polymer is produced at
a rate of 55 gal per
10,500 gal oil treated
* polymer is regulated and
must be landfilled
Source: Reference 6.
-------
ROH
QUYCOt
KOH
POTASSIUM
HYDROXIDE
ALKOXIDE 1,2.4 TR1CHLOROBENZENE
ROK
AlKOXIDE
HOH
WATER
« +KCI
DICHLOR06ENZENE POTASSIUM
GLYCOLETHEB CHLORIDE
CHCHLDROBENCENE
GLYCOl. ETHER
ROH
DICHLOROHYDBOXYBENZEN6 QLYCOL
Figure 8.6.2. Probable reaction mechanism.
Source: Galson Research Corporation
The Sea Marconi's chemical process, called CDP-Process, was first
developed for the decontamination of PCB-laden mineral oils. However, the
system has been more recently applied to materials and surfaces exposed to
contaminants coming from fire or explosion of PCB equipment. The chemistry
involves reaction with high-molecular weight polyethylene glycol in the
presence of a weak base and a peroxide. Hopefully, the continued success of
these studies will result in a viable method for the destruction of toxic
organo—halogens dispersed as solid waste in the environment. No application
to aqueous media can be expected from these processes due to their sensitivity
to water.
8-90
-------
REFERENCES
1. Weitzman, L. et al., "Disposing Safely of PCBs: What's Available, What's
on the Way." Power, February 1981.
2. Gin, W. et al., Technologies for Treatment and Destruction of Organic
Wastes as Alternatives to Hand Disposal, State of California Air
Resources Board, August 1982.
3. U.S. EPA, Treatment of Hazardous Waste, Proceedings of the Sixth Annual
Research Symposium, pg. 72. EPA-600/9-8-011, March 1980.
4. Ibid pg. 197.
5. Telephone Conversation with Charles Rogers, U.S. EPA, Cincinatti, Ohio,
May 15, 1986.
6. Nunno, T. et al., Technical Assessment of Treatment Alternatives For
Wastes Containing Halogenated Organics. 6CA Report to U.S. EPA/OSW, under
Contract 68-01-6871, WA No. 9. 1985.
8-91
-------
-------
9.0 BIOLOGICAL METHODS
9.1 PROCESS DESCRIPTION
Biological treatment processes used for the removal of organic solvents
and other VOCs from industrial waste streams can be divided into two major
categories: 1) aerobic processes, and 2) anaerobic processes. In aerobic
systems, microorganisms use oxygen to biologically oxidize compounds.
Anaerobic systems do not require oxygen and these anaerobics exist and react
in a relatively oxygen free environment. Each of these processes can be
further subdivided into suspended growth or attached growth systems.
Suspended growth systems are characterized by microbes moving freely within
the waste stream or being suspended by mechanical agitation.. Attached growth
systems have layers of microbes attached to a suitable medium that comes into
surface contact with the waste stream. The following section describes the
major biological treatment processes used.
9.1.1 Processes Used for Biotreatment
9.1.1,1 Activated Sludge—
Figure 9.1 presents flow diagrams of representative activated sludge
processes. The basic reactor design categories of activated sludge systems
include conventional, complete mix, and step aeration. By varying the
operating parameters of the systems in Figure 9.1, a system can be defined as
a high rate or an extended aeration system. Other modifications are also
possible. As noted in Section 7.6, the PACT process, the addition of powdered
activated carbon to a biological process, has been used with some success in
activated sludge processes to treat solvent wastes. The following paragraphs
briefly describe the more conventional activated sludge systems and examples
of pure oxygen activated sludge processes.
9-1
-------
Waste sludge
J_ ,SSffia Effluent
499
Supply
Demand
Tank length
Waste
sludge
31
Supply
Demand
Tank width and length
Influent
Waste sludge
t -***
I Settling,
Sludge return
2H
X »
o «
i Demand
Supply
Tank length
Figure 9.1. Flowsheet and plot of oxygen demand and oxygen supply
versus tank length for (a) conventional, (b) complete-mix,
and (c) step-aeration activated-sludge processes.
Source: Reference 1.
9-2
-------
Conventional Activated Sludge—Figure 9.la sbows a flow diagram of a
conventional activated sludge system. Organic waste and recycled sludge are
introduced to a reactor where aerobic bacteria are maintained in suspension.
The waste stream moves in plug flow through the reactor. A rate of diffused
or mechanical aeration supplies the system's oxygen demand and maintains the
bacterial suspension. In conventional systems oxygen supply is constant;
however, a variation, termed tapered aeration, is designed to provide air in
proportion to the reactor's oxygen demand. By aerating more at the reactor
entrance where the oxygen demand is greatest, a lower total air requirement is
obtained. The waste-sludge mixture remains in the reactor for a mean time
known as the hydraulic residence time that is defined as the reactor volume
divided by the volumetric flow rate. The sludge mixture flows from the
reactor to settling tanks where concentrated sludge is separated from the
waste stream. A fraction of the settled sludge is recycled to the reactor and
the remainder is wasted. The recycled fraction is determined by the desired
food to mass ratio [F/M, measured as biological oxygen demand divided by
mixed-liquor suspended solids (kg BOD/kg/MLSS).
Conventional activated sludge processes operate according to a standard
set of factors. These factors are; organic loading - measured as F/M ratio;
sludge retention time - the average time the sludge remains in the system;
hydraulic residence time - the average time the waste sludge remains in the
reactor; and the system's oxygen demand.
As noted, variation in design and operation of activated sludge units can
be made to accommodate waste stream flows and BOD loadings. For example,
extended aeration uses a relatively long hydraulic residence time operated by
keeping the F/M ratio relatively low (generally between 0.05 and
0.25 kg BOD/kgMLSS/day). This system provides a high removal efficiency with
a minimum of excess sludge. Alternatively, the high rate activated sludge
process operates with a shorter aeration period and therefore a higher F/M
ratio. High rate systems operate in a more efficient BOD removal range and
consequently demonstrate a high rate of BOD removal. The shorter aeration
period generates a lower—quality effluent and maximizes wastes sludge.
Complete Mix Aeration—Figure 9.1b represents an attempt to duplicate the
hydraulic regime of a mechanically stirred reactor. The influent settled
sewage and return sludge flow are introduced at several points in the aeration
9-3
-------
tank from a central channel. The mixed liquor is aerated as it passes from
the central channel to effluent channels at both sides of the aeration tank.
The aeration-tank effluent is collected and settled in the activated-sludge
settling tank.
The organic load on the aeration tank and the oxygen demand are uniform
from one end to the other. As the mixed liquor passes across the tank from
inlet ports to the effluent channel, it is completely mixed by diffused or
mechanical aeration.
Step Aeration—Figure 9.1c shows the flow diagram of a step aeration
system. Step aeration is also an improvement on the conventional activated
sludge design. As seen in the flow diagram, the waste feed is introduced at
discrete steps in the reactor. This configuration creates a more uniform
oxygen demand in the reactor and better utilization of constantly supplied
oxygen.
Contact Stabilization—This is a two stage process that provides for
reaeration of the return activated sludge in a separate aeration tank. The
process was developed to utilize the rapid adsorption of soluble and
particulate organic matter by the activated sludge. The adsorptive phase
takes place within one hour of aeration in the contact tank. The settled
sludge is then aerated to oxidize the organics for an additional 3 to 6 hours
in the stabilization tank. Since the sludge is concentrated in the
stabilization tank, the total aeration tank volume of the system can approach
50 percent of the volume needed for a conventional system. The contact
stabilization process is most successful for waste streams containing
nonsoluble organics.
Pure oxygen activated sludge—Pure oxygen activated sludge processes can
be designed using any reactor design and aeration system design. These
systems are a relatively recent development from conventional aeration
systems. Pure oxygen processes are more efficient and can be operated at
higher process loading parameters than similarly designed conventional air
systems. It is often the case that pure oxygen processes can be run less
expensively than air processes in spite of the additional cost of pure oxygen.
9-4
-------
Nitrification — The presence of ammonia in wastewater exerts an oxygen
demand on the treatment system. In nitrification, ammonia is oxidized
biologically to nitrate according to the following processes:
2NH+ + 302
2ND; + 02 Wtrobacter).
In practice, nitrification can be achieved in the same reactor used for
treatment of organic carbon by significantly extending the aeration period, or
in a separate suspended growth reactor that follows an activated sludge
treatment process. Nitrification is accomplished using either air or pure
oxygen.
9.1.1.2 Aerated Ponds and Lagoons —
The aerated pond or lagoon treatment process is analogous to an extended
aeration activated sludge process. An earthen basin is used for the reactor.
Aeration is performed through surface or diffuse aerators and the biomass is
kept in suspension. Historically, aerated lagoons were operated without
recycle, however, modern lagoons recycle the biomass. Ponds and lagoons
typically rely on algal photosynthesis, adequate mixing, good inlet-outlet
design, and a minimum annual air temperature above 5°C to operate
effectively. In general, aerated lagoons are used for low to medium organic
strength wastes.
9.1.1.3 Aerobic Digestion —
Aerobic digestion is an alternative method of treating the organic sludge
produced from various treatment operations. Aerobic digesters may be used to
treat 1) waste activated or trickling filter sludge, 2) mixtures of waste
activated or trickling filter sludge and primary sludge, or 3) waste sludge
from activated sludge treatment plants designed without primary settling. Two
variations of the aerobic digestion process are in use: conventional and pure
oxygen. Thermophilic aerobic digestion is also emerging as a viable
technology.
9-5
-------
In conventional aerobic digestion, the sludge is aerated for an extended
period of time in an open, unheated tank using air diffusers or surface
aeration equipment. The process may be operated in a continuous or batch
mode. Pure oxygen aerobic digestion is a modification of this process in
which pure oxygen is used instead of air. Thermophilic aerobic digestion
represents still another refinement of the aerobic digestion process. Carried
out with thermophilic bacteria at temperatures ranging from 25 to 50°C above
the ambient temperature, this process can achieve high removals of the
biodegradable fraction in very short detention times.
9.1.1.4 Trickling Filters—
A trickling filter consists of a large, highly permeable bed of media
onto which microorganisms are attached. The filter media typically consists
of stones of 1 to 4 inches in diameter in a cylindical design of 3 to 8 feet
tall. The waste liquid is evenly distributed over the top of the filter by a
rotary sprinkler* Filters are constructed with an underdrain system that
collects the treated liquid and any biological solids that have penetrated the
media. The underdrain must be porous to allow air to circulate up through the
media. The treated liquid flows to a settling tank where solids are separated
from the liquid.
As the waste stream is trickled through the filter media the organics are
oxidized by the attached microorganisms. The attached biomass consists of two
regions: an aerobic outer layer whose thickness is determined by the depth of
oxygen penetration into the biomass, and an anaerobic layer extending from the
media surface to the aerobic region. Oxidation of the organics thus occurs
through both aerobic and anaerobic processes.
9.1.1.5 Rotating Biological Contactors—
A rotating biological contactor consists of a series of closely spaced
circular disks of polystyrene or polyvinyl chloride. The disks are partially
submerged in wastewater and rotated slowly through it. Biological growths
become attached to the surfaces and eventually form a thin layer over the
wetted surface area of the disk. The rotation of the disks contacts the
biomass with the organic material in the wastewater and then with the
atmosphere for oxygen adsorption.
9-6
-------
9.1.1.6 Bioaugmentation—
Bioaugmentation refers to the development and use of specialized
organisms known as chemostatic organisms or chemostats to decompose specific
waste streams. Chemostats are cultured to degrade specific wastes and have
the potential to capitalize on specific characteristics of industrial
wastewater such as its high strength, well—defined composition relative to
domestic wastewater, often unusual pH, temperature, and mineral content.
Examples of chemostats include the aerobic fungi Aureobasidium pollutants.sp.
80. 14 (FRI 4197 and PERM BP-l) isolated by Kaneko, et al., that lends itself
to contact oxidation processes and decomposes and removes solvents such as
benzene, xylene, toluene, and autoldehyde, and seven strains of Pseudomons
cepacia var., niagarous (ATC 31939-31945) isolated by Caloruotolo, et al, that
have been found to degrade some chlorinated aromatics to C0_, salt, and
2
HO. Bioaugmentation can be used to supplement any of the treatment
processes described.
9.1.1.7 Anaerobic Digestion—
Anaerobic digestion is one of the oldest processes used for sludge
stabilization. It involves the oxidation of organic and inorganic matter with
anaerobic and facultative organisms in the absence of molecular oxygen and is
carried out in an air—tight container. Anaerobic digestion is actually a
two—step process in which acid forming bacteria convert complex organics to
volatile organic acids followed by methane producing bacteria converting the
volatile organic acids to methane and carbon dioxide. The methane producing
process is the reaction rate limiting step and is highly sensitive to acidic
conditions. Production of methane will gradually decrease as acidity
increases and will cease if the pH drops below 6.5. For this reason the pH of
the digester must be monitored and lime may be added as needed. In a stable
reactor, gas is produced at a rate of about 1 liter per gram of volatile acids
consumed. The final gas produced is generally more than 50 percent methane
with between 30 and 35 percent C0». The remaining gas consists of other
gases such as elemental nitrogen.
The anaerobic bacteria are highly thermophilic and operating temperatures
are generally in the range of 85-90°F. The methane gas is often recovered and
used to keep the reactor operating in this temperature range.
9-7
-------
Anaerobic reactors are highly sensitive to toxic loads, especially to
high concentrations of ammonia, heavy metals, and sulfide. If SO, is
present it will be reduced to H_S which is often a cause of odor problems in
the produced gas.
9.1.2 Operating Parameters
A number of operating parameters cooperatively determine the
effectiveness of biological treatment processes. The most influential
operating factors are:
* Organic loading described by the food to biomass ratio (F/M often
measured in units of kg BOD/kgMLSS * day).
• Sludge retention time or sludge age.
* Hydraulic residence time.
* Fraction of sludge recycled described by the recycle ratio.
• Numerous factors for anaerobic systems such as pH, temperature,
toxic loadings, and biocarbonate alkalinity.
Table 9.1 lists general ranges of operating parameters for many of the
biological processes discussed in the previous section.
9.2 DEMONSTRATED PERFORMANCE
A literature search revealed a tremendous volume of performance data for
biological treatment systems. Unfortunately, the field data are consistently
reported as BOD or COD reduction and only infrequently reported as specific
organic compound reduction. Pertinent organic compound data have been found
for aerobic treatments using activated sludge, aerated lagoons, and in-situ
bioaugmentation. Their removals are reported along with waste character-
istics, treatment process type, and other pertinent data in Table 9.2. The
literature search revealed no solvent/low molecular weight organic removal
data for any anaerobic systems.
9-8
-------
TABLE 9.1. TYPICAL OPERATING PASAMETERS FOR BIOLOGICAL TREATMENT PROCESSES3
Organic Loading (F/M)
Process
Conventional
and CSTR
Step aeration
Contact Contact
stabilization tank
Stabili-
zation
tank
I High-rate
vo
Extended
aeration
Pure oxygen
Nitrification
Anaerobic Conven-
digestion tional
High
Rate
kg BOD5
kg MLVSS-day
0.2-0.6
(0.30)
0.2-0.5
(0.30)
0.2-0.5
(0.35)
0.4-1.5
0.05-0.25
0.4-1.0
10"3-10~2
NH3-N
(3 x ID-3)
Ib BODj Sludge
____ A(?e' 9c
ftj-day- (days)
0.020-0.040 3-14
(0.035) (5)
0.040-0.060 3-14
(0.050) (5)
0.07+ 3-14
(5)
0.075-0.10 0.25-3
< 0.025 >10
0.15-0.25
5-25
(10)
0.04-0.1
(Ib MLVSS/
ft3 'day
0.15-0.40
(Ib MLVSS/
ft3-day
Residence Recycle
Time, Ratio, Produced
8 (hr) R (ft3/lb MLVSS)
4-8 0.15-0.75
(7) (0.30)
4-8 0.2-0.8
(5) (0.30)
0.5-1.5 0.2-1.0
(1.0) (0.40)
3-6
(5)
1-3 1.0-5.0
15-30 0.7-1.5
1-3 0.25-0.5
0.5
720-1440
240-720 0.6-0.7
"*"Based on combined volume of contact and stabilization tanks.
( ) typical operating value.
Source: Reference 3.
-------
The table demonstrates that many priority solvents, some other solvents,
and a few ignitable organic compounds reported to have been successfully
degraded using aerobic biological treatment processes. Activated sludge is
the most commonly reported process and is the most widely available. Aerated
lagoons are reported considerably less than activated sludge, as is the use of
bioaugmentation. The data indicate that high removal efficiencies may be
obtained using properly designed aerobic biological treatment systems.
EPA, in its background document for solvents, has provided performance
data, as shown in Table 9.3. Research articles demonstrating the
biodegradability (60 to 99+ percent removal) of acetone, n-butyl alcohol,
cresols, ethyl acetate, pyridine, and 1,1,1-trichloroethane are also
identified in Reference 7. The biodegradability of other solvents of concern
is expected to be similar.
9.3 COST OF TREATMENT
Widespread use of aerobic biological treatment systems has led to
standardization of their capital and annual operational, materials, and labor
costs based upon system capacity. Expected treatment system outlays can be
determined using Table 9.4 and Figures 9.2 and 9.3. Since the cost data are
in 1971 dollars, updating can be performed using the periodically published
indexed such as the EPA Index, the Engineering News Record Index, and the
Chemical Engineering Index. 1986 costs are roughly 2.5 times those estimated
for 1971. Table 9.4 and Figure 9.2 and 9.3 do not include the additional
costs of seed chemostatic organisms to be used in bioaugmented processes.
More complete and up-to-date cost information can be found in EPA's Estimating
9
Water Treatment Costs. However, the breadth of this document prevents its
inclusion in this section. The data, as presented in Table 9.4 and
Figures 9.2 and 9.3, do show the relative costs and scaling factors used for
various cost elements.
Standardized cost data for anaerobic treatment systems were not found.
An example of a modern anaerobic system is the "Celrobic" process developed by
Celanese. In 1983, a 1.08 million gallon/day influent COD of 3.3g/L,
Celrobic plant in Pampa, TX, incurred outlays of $8.1 million in capital and
9-10
-------
TABLE 9.2. REMOVAL DATA FOR BIOLOGICAL PROCESSES
Chemical
name
Priority Solvents
1,1, 1-trichloroethane
1,1,2-trichloro-
1,2, 2-trif luoroethane
1 , 2-dichlorobenzene
1-butanol
acetone
carbon disulfide
carbon tetrachloride
chlorobenzene
Problem concentrations^
Substrate Nonsubstrate removal
limiting limiting Influent range
(mg/L) (mg/L) characteristics (median)
99- 99
82 ppb 99.9
69- 99
1,000 42 Ib BOD/ day 95-100
per/1,000 ft5
98
70-90
1,000 100-600 ppb 50
70-90
(98)
0.117 ppm 100
6.5 ppb 99.9
38- 99
( 99)
9.05 ppm 99.6
200 ppm 100
Removal
process
1
3
4
1
3
1
2
3
1
3
1
1
4
5
(continued)
-------
TABLE 9.2 (continued)
V£>
t-1
N>
Chemical
name
cresols
ethyl acetate
ethyl benzene
iso-butyl alcohol
(sec-butanol)
methanol
methyl ethyl ketone
methyl isobutyl ketone
methylene chloride
nitrobenzene
Problem concentrations^
Substrate Nonsubstrate removal
limiting limiting Influent range
(mg/L) (mg/L) characteristics (median)
( 99)
95-96
1,000 — 167 mg/L 99.7
1,000 — 95-100
95-100
90-100
154 ppm 99.8
1,000 98.5
42 Ib BOD/ day 75-85
per 1,000 ft5
84
30-50
257 mg/L 99.6
1,000 100-300
69-99
91-97
98
Removal
process
1
1
1
2
3
4
1
1
2
3
4
1
1
(continued)
-------
1ABLE 9.2 (continued)
^O
1
Problem concentrations'*
Substrate Nonsubstrate
Chemical limiting limiting Influent
name (rag/L) (mg/L) characteristics
tetraehloroethylene
toluene
31 ppm
17 g/L
trichloroethene 214 ppb
(1,1,1-triehloroethylene 78 ppb
trichlorof luoromethance
xylene 20-200 ppb
1,250 ppm
1,1 ,2-trichloroethane
1.3 ppm
1 ,2-dichloropropane
aniline
500 ppm
benzene
Percent
removal
range
(median)
55- 99
( 99)
17- 99
95- 99
98.3
99.9
99
100
(96)
93-95
99.8
( 99)
99.7
99
94.5
100
75- 99
90-100
Removal
process
1
3
1
3
4
1
1
1
5
1
6
1
1
1,2
3
(continued)
-------
TABLE 9.2 (continued)
-------
TABLE 9.2 (continued)
10
I
Problem concentrations"
Substrate Nonsubstrate
Chemical limiting limiting Influent
name (mg/L) (mg/L) characteristics
ethanol 42 Ib BOD/ day
(acetaldehyde) per/1,000 ft^
ethyl acrylate 600-1,000 300-600 42 Ib BOD/ day
per 1,000 fr*
ethylene diamine
formaldehyde — 50-100 3,000 ppm
paraldehyde
NOTES:
Removal Process
1 activated sludge
2 completely mixed activated sludge
3 aerated lagoon
4 bioaugmentation
5 biodegradation using mutant pseudoraonas
6 activated sludge with powdered activated carbon
Percent
removal
range
(median)
85-95
95-100
95-100
95-100
95-100
97.5
99
30-50
Removal
process
1
2
3
1
3
Source: References 4, 5, 6, 7, and 8.
-------
TABLE 9.3. AVERAGE PERFORMANCE OF FULL-SCALE BIOLOGICAL TREATMENT
FACILITIES FOR SOLVENTS OF CONCERN (mg/L)
Type of Average influent Average effluent
Compound treatments concentration (range) Concentration (range)
Carbon tetrachloride AS 6.00 (0.192-44.0) 0.010 (NA)
Chlorobenzene AS, AL 9.88 (3.04-49.8) 0.292 (0.017-1.33)
1,2-Mchlorobenzene AS, AL 5.70 (2.08-23.3) 0.302 ( 0.010-1.15)
Ethylbenzene AS, AL 8.45 (2.21-80.0) 0.010 (NA)
Methylene chloride AS 2.30 (1.64-3.91) 0.011 ( 0.010-0.026)
Nitrobenzene AS 0.765 (0.140-2.32) 0.010 (NA)
Tetrachloroethylene AS, AL 0.435 (0.036-2.25) 0.010 ( 0.010-0.019)
Toluene AS, AL, TF 20.9 (2.08-160) 0.066 ( 0.010-1.45)
Trichloroethylene AS 0.231 (0.134-0.484) 0.011 ( 0.010-0.016)
AS « Activated Sludge
AL ** Aerated Lagoon
TF « Trickling Filter
NA = Not Applicable
Source: Reference 7; derived from Office of Solid Waste Analysis of Organic
Chemicals, Plastics, and Synthetic Fibers Industries Data Base.
9-16
-------
TABLE 9.4. ESTIMATED CAPITAL COST FOR WASTEWAfER TREATMENT UNITS3
Treatment Unit
Parameter
Model Cost*
(1971 dollars)
I
l-l
«4
Raw wastewater pumping
Screening, grit removal
and flow measurement
Equalization
Primary sedimentation or
secondary clarification
Aeration-basin
Aeration-diffused air system
Aeration-surface
Trickling filter
Recirculation pumping
Sludge digesters and buildings
Lagoon
Vacuum filtration
Centrifugation
Incineration
Capacity (mgd)
Capacity (mgd)
Volume (mg)
Surface area (in 1000 ft2)
Volume (in 1000 ft3)
Blower capacity (in 1000 cfm)
Capacity (horsepower)
Media volume (in 1000 ft3)
Capacity (mgd)
Sludge volume (in 1000 ft3)
Volume (mg)
Filter area (ft2)
Capacity (gpm)
Dry solids capacity (Ib/hr)
C = 2.6 x 103 (mgd)1'0
C = 27.0 x 104 (mgd)0'62
C = 7.2 x 104 (mg)0'52
C = 2.8 x 104 (A)0'88
C = 4.2 x 103 (V)0-79
C = 9.0 x 104 (cap)0'72
C » 1.0 x 103 (hp)0'89
C - 3.4 x 103 (V)0'84
C - 2.0 x 104 (mgd)0'70
C - 1.4 x 104 (v)0'64
C = 7.2 x 104 (V)0'71
C = 5.9 x 103 (A)0'67
C - 3.3 x 104 (gpm)0'54
C - 1.4 x 104 (cap)0'56
*1986 cost indicies - 2.5 times 1971 values.
-------
20,000
10,000
S BOOO
2000
1000
J I
10 20 SO 100 200
Capacity
500 1000
Figure 9.2. Estimated annual operating and materials costs as a
function of wastewater treatment facility capacity(3)
1986 costs are -2.5 times those shown.
9-18
-------
10,000
5000
2000
o
-C
c
1000
800
200
100
i lit I I 1 I I
lil i I i i i i i i ill i i i i I
10
20
SO 100 200
500 1000
Capacity
Figure 9.3 Estimated annual man-hours needed for wastewater
treatment facility operation.^'
9-19
-------
$400,000 in expected annual operating costs (1982 dollars). This plant was
expected to produce 220 million cubic feet of methane gas annually which
helped reduce its expected operating costs considerably.
Actual treatment costs for organic compound removal from a particular
waste stream will depend upon specific characteristics of the treatment system
design and waste stream. Pertinent data needed for cost estimation are:
waste stream volumetric rate, organic compound constituents and
concentrations, other waste characteristics such as influent BOD, COD, or
level of toxins, treatment design, and overall treatment goal.
9.4 OVERALL STATUS OF BIOLOGICAL TREATMENT
A large number of companies exist that specialize in the design and
construction of biological treatment systems. Aerobic systems are the most
readily available, and their design and operation are complex but manageable.
The total number of biological treatment systems used for organic compound
removal is unknown. However, the total number of facilities using some sort
of aerobic biological treatment for biodegradable wastes is large, in excess
of 2000. The number of companies offering expertise in bioaugmentation and
anaerobic treatment is relatively small, but this segment is expected to grow
rapidly.
Biological treatment of solvents/ignitables is capable of high removal
efficiency and is generally used as the only treatment system or as the final
stage in a series of treatments. A treatment train including wet air
oxidation units, air stripping treatments, or carbon filters may be used to
remove concentrated solvents prior to degradation. No data were found that
describe typical applications of biological treatments for organic solvent
removal, although most organic solvents of concern can be found in all
municipal waste treatment systems.
Aerated treatment systems are generally open surface impoundments and
require large land areas and considerable capital investments. Aerated
processes have the potential to produce significant air (and odiferous)
emissions. Being open to the environment, the treatment plants are subject to
weathering and their design features and biological kinetics are stressed by
precipitation and temperature extremes.
9-20
-------
REFERENCES
1. Metcalf & Eddy, Inc. Wastewater Engineering: TreatmentDisposal Reuse.
McGraw-Hill Book Company, Inc. New York, New York, 1979. 920 pp.
2. Technical Insights, Inc. New Methods for Degrading/Detoxifying Chemical
Wastes. Emerging Technologies No. 18, Technical Insights, Inc.,
Englewood/Fort Lee, New Jersey, 1986. 258 pp.
3. Sundstrom, D.W. and H.E. Klei. Wastewater Treatment. Prentice-Hall,
Inc., Englewood Cliffs, N.J., 1979, 439 pp.
4. U.S. EPA. Treatability Manual, Volume 3, U.S. EPA, Washington, DC,
EPA 600/8-80-042, July 1980.
5. U.S. EPA. Management of Hazardous Waste Leachate, Office of Solid Waste
and Emergency Response, Washington, DC, SW-871, September 1982, revised
edition.
6. Grubbs, R.B.. Enhanced Degradation of Aliphatic and Aromatic
Hydrocarbons Through Bioaugmentation, Proceedings of the 4th Annual
Hazardous Materials Management Conference, Atlantic City, N.J., June 2-4,
1986. pp 131-137.
7. U.S. EPA. Background Document for Solvents to Support 40 CFR Part 26S
Land Disposal Restrictions, Volume II, January 1986.
8. U.S. EPA. Identification and Control of Petrochemical Pollutants
Inhibitory to Anaerobic Processes, U.S. EPA, Washington, DC, 1973.
9, U.S. EPA. Estimating Water Treatment Costs, U.S. Environmental
Protection Agency, Municipal Environmental Research Laboratory,
Cincinnati, Ohio, EPA-600/2-79-162, a & b, August 1979.
9-21
-------
-------
SECTION 10.0
INCINERATION PROCESSES
10.1 OVERVIEW
Incineration is the principal disposal alternative to land disposal for
nonrecoverable, flammable solvent hazardous wastes. Incineration possesses
several advantages as a hazardous waste disposal technology for solvent
hazardous wastes, including the following:
* Thermal destruction by incineration provides the ultimate disposal
of hazardous wastes, minimizing future liability from land disposal;
* Toxic components of hazardous wastes can be converted to harmless or
less harmful compounds;
* The volume of waste material may be reduced significantly by
incineration; and
* Resource recovery (i.e., heat value recovery) is possible through
combustion.
Incineration processes range widely in overall complexity, but
essentially involve a basic oxidation/pyrolysis reaction (for this discussion,
pyrolysis reactors will be considered as a class of incineration processes).
Solvent hazardous wastes are largely, if not completely, comprised of organic
compounds, thus, the basic incineration process involves the following
oxidation equation:
CxHyClz + 02 C02 + H20 + HC1
+ others (including lower hydrocarbons,
C12 gas, CO)
10-1
-------
In this reaction, heat energy first volatlizes the organic constituent,
then begins to disrupt the intramolecular bonds, causing the molecules to
break down (pyrolyze). Oxygen then contacts the organic components and there
reaction ensues. The completeness of the reaction depends upon the combustion
temperature, the reaction time, and the availability of oxygen.
Numerous incineration processes are currently commercially available
which have been demonstrated to be highly effective for the destruction of
solvent hazardous wastes. According to a recent survey more than 5.5
million tons of hazardous wastes were thermally destroyed in 1981, with 1.7
million tons burned in 240 incineration facilities. An increase in
incineration use and capacity can be expected because of the land disposal
bans. The incineration technologies which have found the most widespread
usage are generally those which employ a simple enough process to be
applicable to many different types of wastes. Five technologies in particular
comprise the vast majority of the currently operating incineration capacity
for which data have been developed demonstrating their applicability to the
destruction of solvent and other low molecular weight, organic bearing
wastes. These are:
* Liquid injection incinerators;
» Rotary kilns;
* Fixed hearth incinerators;
* Multiple hearth incinerators; and
* Fluidized-bed incinerators.
Liquid injection, fixed hearth, and rotary kiln incinerations are most
(2)
common. The comparatively new starved—air fixed hearth units are
generally more efficient than standard designs and are often used for solid
hazardous waste disposal. However, this discussion will emphasize liquid
injection and rotary kiln units because of their documented ability to destroy
wastes. Other thermal treatment technologies, including use as a fuel and
several innovative thermal technologies, will be discussed in following
10-2
-------
sections. Technologies such as wet air oxidation and supercritical oxidation
which could be classified as basically thermal in nature, have already been
discussed in Section 8, Chemical Treatment Processes.
The key element in selecting incineration in general, as a technological
option for waste management, and any incineration design in particular, is the'
physical and chemical characteristics of the wastes. As discussed previously,
solvent hazardous wastes have widely varying characteristics, Most commercial
incinerators restrict the types of wastes they will accept based on their
applicability to their particular incineration system. Waste characteristics
dictate pretreatment requirements, materials transport procedures, and
post-treatment techniques. In general, however, at least one incineration
method is applicable to all solvent hazardous wastes, unless one or more of
the following conditions hold such that incineration and its associated
pretreatment/post-treatment are not technologically or economically viable:
11
a. The waste cannot be physically introduced into an incinerator, even
after pretreatment.
b. Constituents are present in the waste which would destroy the
incinerator or result in its rapid deterioration.
c. No site and/or disposal method is practicably available for the
environmentally sound disposal of ashes and other residues.
d. For wastes having a heat content too low to sustain combustion, no
supplementary fuel is available.
e. Agency or Congressionaliy mandated "acceptable risk" levels are too
low to be met by incineration systems.
While incineration as a hazardous waste management technique possesses many
potential advantages, there are also two major potential drawbacks:
environmental impacts, and costs. Incineration has the potential to affect
both air and surface waters via stack emissions and fugitive emissions of
volatile compounds, and the production of solid wastes (ash and scrubber
liquors and scrubber sludges).
10-3
-------
Incineration facilities permitted to operate by EPA under RCRA are
(4)
required to achieve at least a three tiered environmental standard:
I. They must achieve a destruction and removal efficiency (DRt) of
99.99 percent for each principal organic hazardous constituent
(POHC);
i
2. They must achieve a 99 percent HCl scrubbing efficiency or emit less
than 4 Ibs/hr of hydrogen chloride; and
3. They must not emit particulate matter in excess of 0.08 grains/dscf,
(0.18 grams/dscm) corrected to 7 percent oxygen.
Other standards which may affect the decision to incinerate solvent
hazardous wastes include limitations on the generation of Co, SO , NO ,
and toxic air pollutants (e.g., toxic metals) air standards. To become
permitted, an incineration facility must submit to a full scale evaluation of
design and performance. This evaluation includes A trial burn monitored by
EPA, demonstrating the ability to perform to expected levels for various
wastes. Most incinerators are equipped with control systems to limit both
particulate matter generation and acid gas emissions. Since solvent hazardous
wastes commonly contain chlorine, the latter is needed to remove HCl, a major
product of chlorinated organic compound incineration.
Costs of incineration are higher than most hazardous waste management
alternatives. Incineration costs more because of the large energy input
requirements and cost of environmental controls. Costs vary widely depending
upon waste characteristics, incinerator design, and various operational
considerations. Costs of commercial incineration were found to vary from
approximately $lO/lb to $300/lb. The costs associated with constructing an
incineration facility at a generation site are very large. Typically they
range from several hundred thousand to several million dollars. Incineration
may, however, provide a cost benefit in comparison to other disposal
alternatives. Incineration may be used to provide usable heat energy both for
process and space heating purposes*
10-4
-------
10.1.I Chemical Processes Involved in Incineration
The chemical reaction sequence which takes place in the destruction of
hazardous wastes in an incinerator is a complicated process. It involves a
multiple series of decomposition, polymerization, and free radical reactions.
Several intermediate stages may occur before the original matter is completely
oxidized into its final product, depending upon the chemical composition of
the waste and the design and operation of the incinerator. The chemical
process may be considered as consisting of three basic steps, as detailed
below.
1. Application of Heat Energy—
Most incinerators operate in a flame mode (i.e., wastes pass through a
superhigh temperature zone, followed by a combustion zone at high heat). Heat
energy acts upon the constituents of the waste in two ways. First, heat
energy raises the energy level of individual molecules, leading to waste
particle dispersion (e.g., volatilization). Second, heat energy pyrolyzes
molecules, breaking down intramolecular bonds, allowing reactive species to
form.
2. Mixing with Combustion Air/Reactants—
Oxygen is applied to the reactor and is brought into contact with the
reactive species of the waste. The effectiveness of the reactor in bringing
oxygen into contact with the reactants is the key element in the overall
effectiveness of the unit. The dispersion of waste species, e.g., by
atomization, and the turbulence applied by combustion system mechanisms, is
thus a key element increasing the destruction effectiveness of the system.
Gaseous wastes, burn more readily than liquids, dispersed liquid droplets more
readily than liquids in bulk, and liquids more readily than bulk solids.
3. Formation and Separation of Combustion Products—
As oxidation reaction ensues, gaseous and solid byproducts form and begin
to separate. Some solid byproducts and noncombustable solid waste
constituents fall into the bottom of the combustion unit, while gaseous (and
10-5
-------
vaporous) byproducts are exhausted out of the unit toward air pollution
control and heat recovery. However, most incinerators in 1981 were not
equipped with air pollution control devices, probably because these facilties
handled only low ash, nonhalogenated liquid wastes for which control measures
are not usually necessary*
The typical chemical reaction scheme for incineration of hazardous wastes
is shown below:
M
waste mixture
including noncombus table
solid (M)
stoichiometric volume
of oxygen
»• C02 + H20 + HC1
M
most common reaction products
+ [Cl£ + CaH|jClc + others]
and other species including incompletely combusted species,
noncombus tible species
In practical operations, the incinerator is operated to minimize as much
as possible the formation of the second group of products listed above.
The formation of organics as byproducts is considered a consequence both of
inefficient operation and of the contribution of organics from fuels and
reformation. The formation of Cl~ gas, which is highly toxic, is very
undesirable because it is relatively difficult to remove from stack gases by
conventional air pollution control systems. Fortunately, almost all chlorine
is emitted as HC1, but auxiliary fuel is often utilized as much for its
contribution of sufficient hydrogen to suppress Cl« formation as for its
z 7
contribution to the overall heat value of the combustion mixture.
10-6
-------
Although the reaction scheme in incineration is highly complex, the
overall high rate of reaction allows the general reaction scheme to be
described in terms of first order kinetics. The kinetics if the reaction
serve to indicate the importance of several operational factors as shown below:
"dCA = k (C.)
where C^ = concentration of constituent A in the waste
k = reaction rate coefficient
t = time
The reaction rate coefficient is a function of waste and operating
characteristics, as indicated below:
where A = Arrhenius coefficient (characteristic parameter)
E = activation energy (characteristic parameter)
R = universal gas constant
T = absolute temperature^
Thus, the most significant factors impacting the destruction of wastes in
an incinerator include the temperature, time, turbulence, and concentration of
principal constituents. This observation has been supported by practical
experience, although these is no absolute level of these factors that could be
correlated with DUE or PIC formation.
10.1.2 Applicability of Incineration to Solvent Hazardous Wastes
The determination of the applicability of incineration in general, and
specific incineration technologies in particular, to the management of
hazardous wastes is based upon the analysis of waste physical and chemical
characteristics. The overall "incinerability" of specific wastes is a
function of the relative ease with which those materials may be input to the
combustion system, the ignitability and combustibility of the materials during
10-7
-------
the oxidation/pyrolysis process, the relative hazardousness of potential
combustion byproducts (dictating post-combustion handling and control), and
the general impact on the system from their incineration. Several chemical
and physical characteristics of the wastes, must be considered in determining
whether incineration is technically and/or economically feasible, what
incinerator design will handle a waste most effectively, and what form of
pretreatment should be performed to enhance effective performance. By virtue
of their characteristics, solvent hazardous wastes, in general, are considered
highly applicable to incineration.
"Hazardous wastes to be burned in an incinerator, including solvent
hazardous wastes, may be classified into two basic categories relative to
their incinerability as follows:
1. Combustible wastes—which sustain combustion without the use of
auxiliary fuels; and
2. Noncorabustible wastes—which will not sustain combustion without the
use of auxiliary fuels."^
All combustible wastes are obviously applicable to incineration, but this may
not be the best disposal option for such substances. As discussed in the next
section, such wastes may be better handled in fuel burning devices such as
industrial boilers specially designed to burn hazardous wastes, which would
make more effective use of the recoverable heat energy from these substances.
The primary focus of this discussion will be on noncombustible wastes.
Noncombustible wastes exhibit characteristics which limit their
combustibility. Whether or not these limitations will present a technological
or economic barrier to incineration must be determined.
The primary waste characteristics examined to determine the relative
"incinerability" include the following:
• Physical form;
• Heat content/heat of combustion;
• Autoignition temperature/thermal stability;
* Moisture content;
10-8
-------
* Solids content/metals content/inorganic content;
* chlorine content;
* viscosity; and
* corrosivity.
These characteristics impact all phases of the incineration process. A brief
discussion of the impact of each of these characteristics is presented below.
Physical Form—
The gross physical form of a waste is the primary consideration in the
selection of an appropriate technology, including input of feed mechanism.
Solids, liquids, gases, slurries, and sludges all perform very differently in
an incinerator. Physical form may be modified by some pretreatment or may
change under the high heat conditions of the combustion chamber. Physical
form may especially impact the residence time within an incinerator. A liquid
waste may not, for example, be retained long enough in a rotary kiln to allow
for effective combustion. Sludges may not disperse well and may clog a
fluidized—bed. Physical form has some impact on costs of incineration,
particularly in the handling and transport of wastes to the incinerator. In
general, solid wastes are easier to handle and are, therefore, less costly to
(9)
deal with than liquid wastes.
Most solvent hazardous wastes occur as either liquid or slurry. A
relatively small percentage of the volume of solvent hazardous wastes are
present as solids containing spent solvent. The impact of different physical
forms of solvent hazardous wastes on applicability to different incineration
and thermal destruction technologies is not considered significant for some
technologies including rotary kilns, multiple hearth, pyrolysis, and
fluidized-beds, but is significant for others such as liquid injection and
fixed hearth incinerators.
10-9
-------
Heat Content/Heat of Combustion—
The heat of combustion (or heat content) of a substance is defined here
as a measure of the amount of heat energy produced when the substance is
combusted. Wastes with a higher heat content will produce a higher flame
temperature when burned, which will in turn produce higher destruction
efficiencies.
The heat content of wastes has been the most commonly used characteristic
for ranking their ineinerability, Heat of combustion is often used as a
guideline for determining the need for employing auxiliary fuel. Wastes with
a heat content above 8,500 Btu/lb are considered fuels, and can be burned in
facilities regulated under RCRA Subpart D. These wastes can sustain
combustion in most furnaces. Between approximately 2,500 and 8,500 Btu/lb
wastes may require auxiliary fuels to sustain combustion. Below 2,500 Btu/lb,
wastes require auxiliary fuels and, in many cases, other forms of pretreatment
3
before incineration. A good example of such wastes are those with high
moisture contents, which sometimes require dewatering before incineration can
be conducted. High moisture and chloride contents both limit incinerability
(as discussed later in this section). Heat of combustion for solvent wastes
are listed in Table 10.1.
Autoignition Temperature/Temperature Indicators—
Autoignition Temperature (AIT), as well as several other temperature
based experimental parameters, have been used as indicators of relative ease
of incineration. AIT is defined as the temperature at which a waste will
first sustain combustion. In theory therefore, the lower the AIT of a
material, the lower the required combustion temperature, and thus, the more
easily it will be to incinerate. AIT is considered somewhat limited as an
indicator parameter because it does not take into account other waste
characteristics which may limit ease of incineration in specific
ll(e)
systems.
Two other temperature based parameters described in the literature which
have been developed as incinerability indicators are the practical lower limit
for incineration (T__), and the temperature required for 99.99 percent
LtL
destruction (TQO). TTT was estimated from test burn data for a variety of
yy JLJL
10-10
-------
TABLE 10.1. SOLVENT HAZARDOUS WASTE HEATS OF COMBUSTION
Compound
Heat of
Combustion
(Btu/lb)
Compound
Heat of
Combustion
(Btu/lb)
Trichloromonofluoromethane 198.5
Bromoform 234.9
Dichlorodifluoromethane 376.0
Carbon Tetrachloride 436.7
Hexachloroethane 836.5
Chloroform 1,345.2
Tetrachloroethylene 2,142.5
1,1,1,2-Tetraehloroethane 2,502.6
1,1,2,2-Tetrachloroethane 2,502.6
Methylene Chloride 3,063.5
Trichloroethylene 3,132.8
Methyl Bromide 3,486.8
Bis(chlororaethyl)ether 3,546.9
1,1,1-Trichloroethane 3,582.8
1,1,2-Trichloroethane 3,582.8
Methyl Chloride 5,185.4
Chloroacetaldehyde 5,257.3
Ethylidene Dichloride 5,401.0
1,1-Dichloroethylene 5,401.2
1,2-Mchloroethylene 5,401.2
Carbon Bisulfide 5,841.9
Chloromethyl Ethyl Ether 5,947.7
Ethylene Dichloride 6,165.8
1,3-Dichloropropylene 6,193.3
1,2-Dichloropropane 7,184.0
Acrylic Acid 8,170.1
Formaldehyde 8,179.0
Epichlorohydrin 8,224.1
ra-Dichlorobenzene 8,227.2
Methyl Isocyanate 8,509.4
2-Hitropropane 9,667.2
Methanol 9,769.1
Propylene Glycol 10,196.6
Ethylene Diamine 10,440.9
Furfural 10,492.6
Nitrobenzene 10,819.8
Ethyl Acetate 10,984.4
Methyl Methacrylate 11,143.3
Paraldehyde 11,342.3
Ethanal 11,381.0
Ethyl Acrylate 11,785.9
1,4-Bioxane 11,887.0
Oxirane 12,338.9
Acrolein 12,524.8
Acetonitrile 13,278.4
Acetone 13,282.9
Ethylenamine 13,433.5
Allyl Alcohol . 13,711.2
Methyl Ethyl Ketone 14,587.2
Cresols 14,709.9
2-Picoline 14,903.3
Dimethylamine 14,907.2
Pyridine 15,155.3
Isobutyl Alcohol 15,526.8
1-Butanol 15,559.4
Aniline 15,712.9
Ethyl Ether 15,857.1
n-Propylamine 17,217.8
Benzene 17.998.0
Toluene 18,246.0
Ethyl Benzene 18,505.8
Xylenes 18,523.2
Cumene 18,683.8
Cyclohexane 20,080.1
Source: Refer to Appendix A.
10-11
-------
systems. TQQ is dependent on residence time. Analysis of pooled
field test data has indicated that no strong correlation could be found for
DRE as opposed to any of the thermal ranking methods. Research data, however,
suggest the use of an incinerability ranking based on gas-phase thermal
stability data collected in laboratory experiments under oxygen deficient
..... U(e)
conditions.
Comparison of these stabilty data with appropriate field test data
indicates that field thermal stabilty rankings could be predicted from the
laboratory data in 70 percent of the cases evaluated. Additional research is
being conducted to expand the number of compounds for which comparisons can be
made.
Indication parameters, including the Arrehenius coefficient and
activation energy parameters used to establish the reaction rate coefficient
previously discussed, are presented for several of the solvent hazardous
wastes in Table 10.2.
Moisture Content--
Moisture contained in the waste reduces the incinerability of a waste.
In the combustion process, water will absorb heat energy and vaporize, but
will not oxidize or pyrolyze. This will tend to reduce the heat energy
available to assist the combustion. Hater may also absorb combustion
intermediates and waste components and thus limit their availability for
combustion.
The requirement to drive off moisture increases the overall stress on
incineration systems, and increases the operating costs. Certain incinerator
design, including fixed hearth furnaces and rotary kilns, are not equipped to
handle high moisture content wastes. Moisture content may be reduced by
13
dewatering pretreatments, but these tend to be expensive. The most common
way of dealing with high moisture content wastes is to blend them with solid
14
wastes or other high heat content materials.
10-12
-------
TABLE 10.2. CHARACTERISTIC PARAMETERS FOR SEVERAL SOLVENT HAZARDOUS WASTES
Compound
Acrolein
Athyl alcohol
Athyl chloride
Benzene
(mono) chlorobenzenc
,_, 1 , 2-Dichloroethane
o
M Ethanol
w
Ethyl acrylate
Methyl chloride
Methyl ethyl ketone
Toluene
Vinyl acetate
Vinyl chloride
E
A -(eal/g-mole)
3.3 x IQlO
1.75 x 106
3.89 x 10?
7.43 x 1021
1.34 x 1017
4.8 x 1011
5.37 x 1011
2,19 x 1012
7.34 x 108
1.45 x 101*
2.28 x 1013
2.54 x 109
3.57 x 1014
35,900
21,400
29,100
95,900
76,600
45,600
43,100
46,000
40,900
58,400
56,500
35,900
63,300
Autoignition
temperature
(AIT)
("F)
453
713
905
1,044
1,180
775
793
721
1,170
960
997
800
882
Estimated temperature for
(ILL) lower 99.99% destruction (T99)
1 irai t ing ——.—««.«-.——•—«»-.————«.
temperature for At 1-second At 2-second
incineration (°F) residence (°F) residence (°F)
800
1,050
1,150
1,275
1,350
1,050
1,250
1,000
1,500
1,200
1,275
1,150
1,250
1,020
1,176
1,276
1,351
1,408
1,216
1,307
1,132
1,597
1,290
1,340
1,223
1,371
975
1,072
1,200
1,322
1,372
1,173
1,256
1,092
1,514
1,247
1,295
1,164
1,332
Source: Lee et al., June 1982 (Reference 12).
-------
Asb Content (Solids/Metals/Thermally Inert Materials Content)—
The ash content of a hazardous waste is considered a very important
incinerability indicator. Ash content is defined here as that portion of a
waste which is either a noncombustible solid or will form a solid byproduct in
the combustion process. Ash content, therefore, encompasses most of the
suspended solids, metals, and other (primarily inorganic) thermally inert
compounds of a. waste (it is important to note that some solids will combust
readily in the process).
Ash content directly impacts both the overall combustibility of a waste
and the materials handling requirements both before and after the combustion
stage of the process. Wastes with higher ash contents tend to be more
difficult to pump, and will tend to plug liquid atomizers more readily. Ash
components are more difficult to disperse and require more energy input in
handling. Higher ash contents increase the possibility of unburned waste
carryover to the recovered ash stream. Particularly in the case of high heavy
metals content, higher ash contents may lead to higher emissions of
particulate matter or pollutant particles of concern.
Ash content is used to determine the type of incinerator selected, air
pollution control equipment required, ash recovery system required, and is
often directly used in incineration pricing structures. Rotary kiln and
hearth type incinerators are, in general, more applicable to wastes with
higher ash content, while liquid injection and fluidized—bed incinerators are
less applicable. Fluidized—bed incinerators have a particular limitation to
wastes containing sodium salts which tend to clog the bed, leading to process
failure. The costs of incinerating wastes with higher ash content are
higher primarily due to increased air pollution control and ash recovery
costs. Many incineration facilities appear to use ash content as a factor in
determining the price of incinerating wastes. One facility contacted, for
example, indicated that they charged an extra 1 cent per pound per each
percent of ash content. Solvent hazardous wastes generally have low ash
contents. An example of the ash contents of a variety of solvent waste
streams is shown in Table 10.3.
10-14
-------
TABLE 10.3. WASTE CHARACTERISTIC DATA FOR SEVERAL SOLVENT
HAZARDOUS WASTE STREAMS
Waste description
Trichloroethylene spent
solvent waste
1,1,1-Trichloroe thane spent
solvent waste
Methylene chloride spent
solvent waste
Tetrachloroethylene spent
solvent waste
Spent degreasing solvents from
electroplating metal cleaning
Degreaser sludge from
electroplating metal cleaning
Spent solvents, N.O.S.
Trichloroethylene still bottoms
1, 1, 1-Trichloroethane still bottoms
Methylene chloride still bottoms
Still bottoms, N.O.S.
EPA waste Ash content Chlorine content
code (%) (%)
F001, F002
F001, F002
F001
F002
F001
F001
F004, F005
F001, F002
F001, F002
F001
F004, F005
2
2
2
2,
2
20
2
0
20
20
20
65
64
33
68
46
32
0
16
16
8
0
Source: W-E-T Model Hazardous Waste Data Base, 1982 (Reference 18).
10-15
-------
Chloride Content—
The chloride content of a waste is also considered a very important
indicator for incineration. Chloride content relates directly to the
formation of HC1 and possibly Cl gas. The emission of both of these
compounds is regulated by environmental standards and, thus, most wastes
containing more than nominal levels of chloride require air pollution
controls. The chloride concentration is also related to the overall
corrosivity of combustion byproducts. As a result, most incinerators
establish a limiting chloride concentration for their systems. It is common
for this limit to fall under 3 percent by weight. Most incinerators also
appear to have established surcharges for chloride content. One facility
contacted stated that an additional charge of 0.2 cents per pound per each
1 percent of chloride was common practice in the industry.
Some solvent hazardous wastes, as shown in Table 10.3, are halogenated
organics and the chloride content of those wastes is often extremely high.
Assuming the total conversion of all chlorine within the waste to HC1, the
concentration of HC1 in the exit gases under normal combustion can approach
3
1 gram/m of flue gas for a waste containing 2 percent chlorine. This
estimate assumes that the molecular structure and heating value of the waste
or a waste/fuel mixture approaches that of a No. 2 or No. 6 fuel.
Viscosity—
The viscosity of a waste impacts its pumpability and atomization, and
must be considered if incineration in a liquid injection unit is
comtemplated. Liquid injection incinerators will require that wastes be
atomizable to be effectively destroyed and, therefore, must operate within a
13
general viscosity limit. Although the limit varies, depending on
characteristics of the waste burned and the injection system design, the usual
limiting range falls between 50 and 200 SSU. Viscosity also may somewhat
limit the applicability of wastes to fluidized-bed or multiple hearth
incinerators. Fluidized-bed incinerators require that liquid wastes be
"pumpable" in order to be effectively dispersed in the bed. A common
"pumpability" limit is 10,000 SSU at 100°F and assumes that the viscosity will
10-16
-------
be at an acceptable level for atomization when heated. Multiple hearth
incinerators require that a liquid be dispersed by an impeller mechanism for
effective performance. Highly viscous wastes are not well dispersed by such
mechanisms, and they actually may flow through a multiple hearth too rapidly
for effective destruction (commonly falling directly to the ash recovery
conveyor).
Highly viscous wastes are commonly subjected to preheating or blending
with more mobile, compatible solvent wastes.
Corrosivity—
Corrosivity directly impacts upon equipment design and costs, and
materials handling requirements. Certain incinerator designs may not be
appropriate for highly corrosive materials. In particular, fluidized-bed
materials and metal parts can be affected by corrosivity. In general,
pretreatment methods (e.g., chemical neutralization) should be employed before
incinerating corrosives.
10.1.3 Strategy for Assessment of Incinerability
As stated by Martin and Weinberger, incineration is a potential option
3
for the disposal of all hazardous waste. Although no one type of
incinerator exists that can effectively destroy all types of hazardous wastes,
there is likely to be at least one type of incinerator capable of burning any
particular type of waste. Once it has been established that no technological
options exist which may more effectively or economically manage a waste, a
strategy should be adopted to determine the best incineration technology to
handle the material. This strategy should be based upon waste
characteristics. A recommended approach, based upon the strategy designed in
Reference 3, is outlined below:
1. Determine whether the waste can be physically introduced to the
combustion zone as is, or if pretreatment is required. This
determination is based upon physical form and viscosity. For
example, if the waste is a liquid with low viscosity, it can be
atomized and, thus, may best be input through a liquid injection
system.
10-17
-------
2. Determine the overall physical effect of the waste on the
incineration system. This consideration is primarily based upon the
physical form, solids content, and corrosivity of a waste. These
factors may be such that the incineration of the waste will rapidly
lead to process failures due to debilitation of equipment.
Refractory linings, for example, are highly susceptible to chipping
and cracking by large solid particles.
3. Determine if auxiliary fuel should be used. This determination is
commonly made solely on the basis of the heat of combustion. For
example, wastes with a heat of combustion of below 2,500 Btu/lb are
almost always mixed with a fuel or blended with a high Btu waste.
4. Determine the temperature and residence time requirement for
effective combustion. This determination is largely based on
characteristics such as moisture content. Many incinerator designs
operate with a specified time or temperature range, while waste feed
rate may be adjusted.
5. Determine the disposal or handling method required for combustion
byproducts other than gaseous products. This consideration is
largely based upon the solid/metal/thermally inert material
concentration of the waste. Wastes with a. high ash concentration,
for example, may require a continuous ash removal system.
6. Determine if air pollution controls are required. This
consideration is largely based upon the chloride and ash content of
a waste. Most wastes containing more than a very small amount of
chloride will require a scrub'ber to remove acid gases. Need can be
calculated assuming emissions are directly related to input.
7. Determine if relevant environmental standards can be met. This
determination, again, is based upon chloride and/or ash
concentration. Most incinerators operate with a chloride
concentration limit. If the chloride content is too high, the air
pollution control system will not be adequate to limit emissions to
the applicable standard.
8. Determine overall costs of incinerating the waste. After
considering all of the factors detailed above, the relative costs
associated with these technologies should be estimated. It is
important to note that the technology with the lowest base cost may
not be the most cost-effective alternative, should one of the
factors listed above come significantly into play.
A flow chart detailing the general considerations to be made in selecting the
appropriate incineration system. This flow chart is presented in
Figure 10.I.3
10-18
-------
OewateHng
Process
Is
Gross Physical
For* O.K.
Is
Moisture Conten
Too High
Is
Btu Content
Too Low
7
Can
Physical Form
be made O.K.
Modify Physical
Form as Weeded
Solids Removal
Process
Is ^^
Cont«nt\, Y€S fc
Hlgh^^^
[^
Metal
Removal /Reduction
Process
Augment with
High Btu Waste
or Auxiliary Fuel
trtTTer
Alternatives
Figure 10.1. Pretreatment option logical decision flow chart-
10-19
-------
Based on the "standard" measures employed to evaluate the incinerability
of hazardous wastes, solvent hazardous wastes are generally considered good
candidates for incineration. The characteristics which appear to give solvent
hazardous wastes good incinerability include the following:
• Most solvent hazardous wastes are found in the liquid form. Liquid
wastes tend to burn more efficiently because they are easily
dispersed and, thus, mix more readily with combustion air;
• Most solvent hazardous wastes are composed primarily of organic
materials; (i.e., they contain low levels of solids and inert
materials) which oxidize or pyrolyze more readily;
• Solvent wastes often can be blended with many other types of wastes
to render them applicable to a wider range of technologies; and
• Solvent wastes do tend to have high Btu values and relatively low
autoignition temperatures, indicating that they are more readily
combustible.
Solvent hazardous waste characteristics which limit their applicability
to incineration include:
• Air pollution control requirements because chlorine and ash
concentrations can be high; and
• More pretreatment requirements for some solvent hazardous wastes
than for other types of wastes.^
In addition, low flash point wastes may require processing or blending to
reduce the potential hazards related to storage and combustion of these
materials.
10.2 PROCESS DESCRIPTION
There are numerous incineration system designs available to handle the
wide variation of chemical and physical characteristics found in hazardous
wastes. Hazardous waste incineration technologies range from those with
widespread commercial application and many years of proven effective
10-20
-------
performance, to those currently in development. As many as 67 companies may
be involved in the design and development of hazardous waste incinerators,
with more expected as limitations on land disposal of hazardous wastes
increase.
As mentioned previously, there are several incineration technologies
which have become established commercially as the primary options available
for the incineration of hazardous wastes. These technologies have been
demonstrated extensively for a wide range of hazardous wastes. They comprise
about 80 percent (by number) of the U.S. market. a ' They include:
Liquid injection incineration
Rotary kilns
Fluidized-bed incinerators
Fixed hearth incinerators, particularly the starved air or pyrolysis type
units, and
Multiple hearth incinerators.
The first two (and the fixed hearth units) are the most widely used for
the disposal of hazardous wastes. A description of the first three types of
units listed above will be provided here, following a brief description of
basic components common to all incinerators. The hearth type incinerators,
particularly the fixed hearth unit, are also used extensively, but data on
their ability to handle hazardous wastes have not been widely published in the
literature. Discussions of the design and operation of these systems can be
found in the open literature.
All incineration systems are designed in consideration of the four basic
elements of combustion: temperature, time, turbulence, and concentration, as
described previously. Temperature is the most important element of an
21
incineration system. The heat requirements govern the method in which
10-21
-------
heat energy is supplied and sustained within the combustion chamber, and
governs many of the pretreatment operations conducted. Residence time
requirements impact the size of the combustion chamber, as the volume of the
combustion zone must be sufficient to allow for completion of thermal
destruction. Turbulence is strictly a function of incinerator design.
Elements such as baffles, rotation, or changes in direction within the
combustion chamber increase turbulence (and, therefore, enhance mixing of
wastes and oxygen to allow for more effective performance). Concentration
governs the oxygen input requirement, as sufficient air must be supplied to
insure complete combustion of hazardous constituents.
In terras of the chemical process system, there are essentially five
component parts common to any incineration facility, as shown in Figure 10.2
and discussed below.
1. Material Storage and Preparation—Waste materials are received,
analyzed, stored and prepared for input into the incinerator. In
this initial step of the incineration process, the waste
characteristics which may affect the performance of the incinerator
are identified. If necessary, pretreatment operations are conducted
to mitigate these characteristics. In some cases, wastes are
rejected for incineration when pretreatment will not render them
"incinerable" in the technology present.
Common methods of pretreatment include preheating, chemical
neutralization, filtration/sedimentation of suspended solids and
water, and distillation. Many of these technologies are described
in detail in other sections of this document.
2. Waste Feed Mechanism—The waste feed mechanism is not merely the
means by which waste materials are input into the combustion chamber
of an incinerator. Feed mechanisms control the volume of waste
present in the chamber at any moment, and thus control waste
residence time. Feed mechanisms also play a key role in creating
surface area to increase combustion rate and developing turbulence
within the combustion system. Dispersion of wastes is particularly
critical in liquid injection and fluidized-bed incinerators.
3. Combustion System—As described previously, combustion systems
perform three functions: 1) heating of waste materials to vaporize
and pyrolyze them; 2) mixing of wastes with combustion air; and
3) oxidation and subsequent formation and separation of combustion
products.
10-22
-------
TANKFOH TANK FOR TANK FOR BLENDING FUEL OIL
VNATEn. WATER- SPECIAL TANK TANK
SOLUUIE INSOLUBLE BATCHES
LIQUIDS LIQUIDS
LIQUID W9TES
fPQU TANK CAMS,
on TRUCKS
STACK
"O
NJ
UJ
pirs
fj
STEAM
FOR SOLID WASTES FOR PASTY WASL
ASH, SLAQ
ASH/SLAQ TOANSPORT
BTOnAQE
FEED
INCINEHATIOK
HEAT RECOVERY
OFF-GAS CIEAN1MJ.
NEUTRALIW.TIOM
Figure 10.2.
Flow sheet of an incineration plant for hazardous wastes
Source: Babcock Krauss-Maffei Industrieanlagen GMBH
(Revised by P. Adie)22
-------
4. Heat Recovery—Heat recovery systems are often employed with
incineration of hazardous wastes in order to achieve greater cost
effectiveness of the operation as a whole* Generally, heat recovery
is accomplished by either standard type heat exchange equipment or
waste heat boilers which burn the waste byproducts. There are
generally two limitations in heat recovery. First, the cost
benefits of heat recovery must justify the expense of the heat
recovery system, including design, installation, and maintenance.
Second, heat recovery systems should not be used if they lead to a
more difficult waste management problem (i.e., form new pollutants
of concern, or require difficult maintenance, e.g., cleanup of waste
byproducts plugging the heat recovery system).
5. Solid and Liquid Waste Control—Air pollltion control devices and
air pollution control systems are required if the combustion process
produces air pollutants at levels exceeding applicable emissions
standards. Most commonly, the primary pollutants of concern
generated by incineration of hazardous wastes are particluate matter
and hydrochloric acid (HC1) vapor. Air pollution control is often
but not always used at hazardous waste incinerators. Incineration
processes produce solid and liquid waste streams which must be
managed. These streams are usually not hazardous themselves. Ash
produced in combustion is collected either continuously by some
mechanism; e.g., a screw conveyor built into the bottom of the
combustion system, or periodically by manually cleaning the
combustion chamber. Sludges can be produced by air pollution
control or heat recovery systems, and are removed periodically from
the process systems. Liquid wastes produced by air pollution
scrubbers or quench towers are continuously treated. In most cases,
ash may be disposed of in a landfill, as may dried sludges. Liquid
wastes may be subject to waste water treatments before discharge.
10.2.1 Liquid Injection Incinerators
Liquid Injection (LI) incinerators are the most widely used hazardous
waste incineration system in the United States, accounting for 64 percent of
20
the total number of hazardous waste incinerators in use in the U.S. LI
systems may be used to incinerate virtually any liquid hazardous waste,
including most solvent hazardous wastes, due to their very basic design and
high temperature and residence time capabilities. Liquid injection
incinerators represent the most effective system available for most liquid
hazardous waste solvents, from both a technical; (i.e., destruction
efficiency) and economic perspective.
10-24
-------
Liquid injection incinerator systems typically employ a basic, fixed
hearth combustion chamber. Pretreatment systems to blend wastes and fuels, to
remove solids and free water, and to lower viscosity through heating, are
often used in conjunction with liquid injection incinerators. Ash recovery
systems may not be required, at least on a continuous basis, because many
liquid hazardous wastes fired in an LI system contain low volumes of ash or
suspended solids.
The liquid waste feed system is the key element of the LI process.
Liquid injection incinerators operate as "suspension burners", whose
combustion efficiency (and hence destruction efficiency for constituents of
hazardous wastes) is dependent upon the extent to which the feed mechanism can
disperse the liquid waste within the combustion chamber and provide sufficient
area for contacting waste with combustion air. There are two atomizer designs
commonly employed in LI systems, denoted as fluid systems and mechanical
systems. Typical characteristics of several atomizer designs are described in
detail in Reference 23.
Once liquid wastes enter into the liquid injection incinerator and are
ignited at the burner, the turbulence imparted to the waste and good mixing
with combustion air lead to efficient combustion. Combustion temperature
capabilities of the systems can be very high, reaching over 3,000°F in many
cases. Residence times are generally within a 1 to 2 second range, depending
13
upon liquid heat content. Table 10.4 summarizes operating parameters for
typical hazardous waste liquid injection systems.
Applicability of hazardous wastes to liquid injection incinerators is
generally limited by the extent to which they may be atomized, and the
physical effect they may have on the incinerator equipment (mostly notably, on
the atomizer). The primary restrictive waste characteristics of interest are
the liquid viscosity, solids content, and corrosivity. Wastes with low heat
value may also be restricted from burning in a liquid injection incinerator.
A typical limiting value (below which waste must be mixed with a fuel, or a
high heat value material), given by one incinerator operator, is 5,000
Btu/lb.24
10-25
-------
TABLE 10.4. OPERATING PARAMETERS OF HAZARDOUS WASTE
LIQUID INJECTION INCINERATORS
Form of waste feed:
Heat input capacity range:
Heat release:
Operating temperature range:
Residence time range:
Excess air:
Pressure:
* Liquid wastes only.
* Limiting liquid viscosity for atomization
is typically 16,000 centistokes.
• Limiting solids content may be as high as
10% by weight undissolved solids.
* Limiting solid particle size may oe as
high as 1/8-inch diameter.
5 to 150 x 106 Btu/hr.
25,000 Btu/hr.ft3 (typical)
1,000,000 Btu/hr.ft3 (maximum)
1,200 to 3,000°F
0.5 to 2.0 seconds
20% (typical)
For nitrogen-containing wastes, excess air
requirements may be 65 to 95%.
0.5 to 4 in. H£0 (typical)
Source: MITRE, 1982 (Reference 20).
10-26
-------
Liquid viscosity is regarded as the primary limiting waste characteristic
for liquid injection systems since viscosity determines whether or not the
liquid is pumpable and atomizable. Pretreatment may be needed to reduce the
viscosity of wastes to a level where high combustion efficiencies may be
achieved. The two most common viscosity reducing pretreatment operations are
heating and dilution. In some cases, high energy mixing is done to produce a
one- or two-phase emulsion of liquid waste in the carrier media. Energy for
preheating is often supplied by heat recovery systems.
Suspended solids are another restrictive waste characteristics for liquid
injection incinerators. Undissolved solids can impact negatively through
abrasion or plugging. The best available technology to reduce the solids
content is filtration or sedimentation. Filtered-solids may be collected,
washed of any retained liquids by an appropriate solvent, and disposed of
separately. Wash solutions can be incinerated.
Highly corrosive wastes provide a potential limitation to effective
performance of liquid injection systems. However, no pH limits for liquid
injection incineration were described in the available literature other than
those dealing with chloride content, and no detail was provided on chemical
pretreatment of corrosive wastes. It can be assumed that such techniques are
viable for LI systems, however, depending on the characteristics of wastes
handled and process design.
In some cases, the applicability of an LI incinerator may be extended by
the use of multiple injection systems. In this way, an injector may be fitted
to more specific ranges -in waste characteristics allowing a broader range of
overall usage without requiring pretreatment. As discussed earlier, certain
atomization device designs are better suited to more viscous or higher
suspended solids containing wastes than others. In addition, the use of
multiple injection points may allow for concineration of incompatible wastes.
10.2.2 Rotary Kiln Incinerators
Rotary kiln (RK) incinerators have found widespread application in the
U.S. for management of hazardous wastes, both at chemical manufacturing and at
19
hazardous waste facilities. MITRE estimated that rotary kilns comprised
10-27
-------
12.3 percent of the total number of hazardous waste incinerators in
20
operation. Rotary kiln systems are considered the most versatile of the
established incinerator technologies. Liquid, solid, and slurried hazardous
wastes may all be burned in rotary kilns, without extensive adaptation of the
design for specific waste types.
Rotary kiln systems employ a fairly basic design concept. As depicted in
Figure 10.3, the typical rotary kiln system involves two-stage combustion of
waste materials with primary combustion occuring in the rotary kiln followed
by secondary combustion of gaseous byproducts. Heat recovery, ash recovery,
and air pollution control devices are usually included in the overall system.
Combustion byproducts are most often scrubbed for both particulate matter and
acidic byproducts; e.g., HC1. Heat recovery is employed in the majority
20
( 70 percent, according to recent estimates) of cases.
Pretreatment of hazardous wastes is not often required for incineration
in a rotary kiln, because of the great versatility of the system. The most
common preparatory operations conducted at rotary kiln incinerators include
size reduction, mixing of liquid wastes with solid wastes, and chemical
neutralization. Wastes with an average heating value of 4,500 Btu/lb are
reported adequate to sustain combustion at kiln temperatures between 1,600 and
20
1,800°F. In those cases where auxiliary fuel is required, No. 2 fuel oil
20
is used most often. Size reduction of solid waters, via crushing and
grinding operations, is a common preparatory operation. This is often done
both to preserve the life span of the kiln refractory lining and to increase
the combustion efficiency of the system. Mixing of liquid wastes with solid
wastes helps to increase the liquid waste residence time and thus enhance
destruction efficiency. Highly corrosive wastes are often neutralized by
chemical treatment before being fed to the rotary kiln. This helps preserve
the working life of the kiln refractory.
Waste materials, following pretreatraent, are fed to the elevated end of
the rotary kiln. Waste feed mechanisms employed are typically simple hoppers
which feed a regulated amount of material to the kiln. Vendors generally
recommend continuous operation of a rotary kiln, although they may be operated
10-28
-------
o
I
to
ON UNI.
GAS
MONITORS
CONVEYOR
FIBIR PACKS
EPA METHOD 5
SAMPLING TRAIN
KILN EXIT DUCT
AFTERBURNER
HOT DUCT
IIYDRAUD LIME
SLURRY FEED
ASH RESIDUE
SAMPLE
AMPLE
ORIS
FEED WASTE
LIQUID IURNERS
SCRUIBER
LIQUID SAMPLE
DISCHARGE
SCRUBBER WATER
Figure 10.3. Rotary kiln incinerator with liquid injection capability.
-------
25
intermittently. Waste materials flow through the rotary kiln as a
consequence of the rotation-and the angle of inclination. The kiln is often
designed with baffles, which serve to regulate the flow rate through the unit,
generally resulting in increased residence times. The rotation of the kiln
serves to enhance the mixing of waste with combustion air and provides
continuously renewed contact between waste material and the hot walls of the
kiln. Combustion air is fed either concurrently or countercurrently. One
feature of a rotary kiln is that it may be operated under substoichiometric
(oxygen deficient) conditions to pyrolyze certain wastes.
As combustion of the waste progresses, ash flows to the bottom of the
unit and is conveyed the ash recovery system. Gaseous combustion products are
exhausted to the secondary combustion unit.
Secondary combustion generally takes place in a fixed hearth type unit,
where gaseous products of combustion, including incompletely combusted waste
components, combustible waste products, and fly ash are fired. The gaseous
products from the secondary combustion chamber are normally then passed
through heat recovery and air pollution control systems, while ash is
collected and transported to the ash recovery facility.
Most rotary kiln systems are equipped with a multistage scrubber system .
to control particulate matter, acid byproducts of combustion, and oxides of
sulfur and nitrogen. Heat recovery systems are often used not only for the
conservation of energy, but also to reduce temperatures to allowable levels
prior to introduction to the scrubbers. Typical operating parameters for a
rotary kiln system are shown in Table 10.5.
Rotary kilns are generally large systems, and thus require a large
capital expenditure. Due to their energy requirements, the operating costs
associated with rotary kiln systems -may also be higher than other incinerator
system. Their versatility may lead, however, to benefits measurable in
overall reduced costs for hazardous waste management; cost considerations are
further discussed later in this section.
10-30
-------
TABLE 10.5. OPERATING PARAMETERS FOR ROTARY KILNS
Form of Wastes Fed
Thermal Capacity
Typical Overall System Flowrate
gas flow
pressure drop
solid feed rate
liquid feed rate
Combustion Temperature
1st chamber (Rotary Kiln)
secondary chamber
Residence Time
gases
solids
Rotational Speed
Length-to-Diameter
Excess Air
Refractory Life
Liquid, solid, slurry. Virtually any
waste may be fired to a Rotary Kiln.
1 - 150 x 106 Btu/hr (Rotary Kiln)
20,000 Btu/hr (secondary combuster)
47,000 acfm @ 2200°F
10 - 25 in H20
10,000 Ibs/hr
3,000 Ibs/hr
500 - 2300°F
1600 - 2800°F
0.5 - 3.0 sees
Highly variable, depending on
viscosity, angle of inclination,
rotation of kiln
12 revolutions/hr (typical)
2:10 (typical)
60 - 150%
24 - 30 months
Source: MITRE, 1981 (Reference 20).
10-31
-------
10.2.3 Fluidized-Bed Incinerators
Fluidized-bed (FB) incineration systems represent a newer incineration
technology which has not yet made a significant commercial impact in the
established incinerator market. Although fluidized-bed processing units were
developed nearly 50 years ago and have found extensive application both in
chemical processing, and more recently, in sewage sludge incineration, the
development of FB systems capable of destroying wastes containing hazardous
components is still in its early stages. As indicated in MITRE*s 1982
20
survey, only nine fluidized-bed units, representing 2.6 percent of the
total number of hazardous waste systems in operation, had been put into actual
service at hazardous waste processing facilities. The basic fluidized-bed
system, as depicted in Figure 10.4, consists of a refractory-lined vessel; a
perforated plate which supports a bed of granular material and distributes
air; a section above the distributor containing granular solids referred to as
the freeboard; an air blower to move air though the unit; a cyclone to remove
particles and return them to the fluidized bed (not shown in Figure 10.4); an
air preheater; a start-up bed heater; and a system to inject and distribute
the feed in the bed. Fluidized-beds are always oriented vertically. Feed and
air flow are balanced to achieve fluidization in the bed. The fluidized-bed
promotes turbulence and serves as an excellent heat transfer medium, thus
assisting combustion. As will be discussed later, the fluidized-bed material
can be chosen to react directly with combustion production such as HC1, thus
minimizing subsequent air pollution control requirements.
Fluidized-beds are capable of burning all forms of waste, and due to the
high turbulence present, no atomization is required. Combustion air and
auxiliary fuel are introduced from the bottom. The bed is kept at a high
temperature (typically, between 1,250 and 1,750°F), so the waste materials
almost immediately will begin to burn as they mix with the fluidized-bed. The
gaseous products of combustion flow out the top of the incineration units, to
be scrubbed and passed through heat exchangers for heat recovery. The solid
products either remain within the fluidized-bed if they are of approximately
equal mass to the bed particles, fall to the bottom of the bed where they will
eventually be removed by the ash recovery system, or become entrained in the
exit gases, where they will be removed by the air pollution control system.
10-32
-------
SAND FEED
THERMOCOUPLE
WASTE INLET
FLU1DIZING
AIR INLET
.-.'. .-.'. FLU1DIZED
••••••••:•••" SAND BED :•:•:
EXHAUST AND ASH
PRESSURE TAP
SIGHT GLASS
BURNER
TUYERES
STARTUP
PREHEAT
BURNER
FOR HOT
WIND80X
Figure 10,4. Cross-section of a fluidized-bed furnace.
Source: Reference 22,
10-33
-------
Operating parameters for fluidized—bed incineration are shown in
Table 10.6. As shown in the table, operating temperatures are lower than
those found in other types of incinerators. However, the long residence times
and the excellent distribution of thermal energy within the bed are sufficient
to provide excellent destruction efficiency of organic solvents.
The usage of a fluidized-bed incinerator may be limited, by certain
chemical characteristics of a hazardous waste. In general these restrictive
waste characteristics are those properties which may affect the fluidity of
the bed itself. The key to the effectiveness of an FB incinerator is the
ability of the bed to display certain liquid-like physical properties. Those
wastes with characteristics which lead to either an increase or decrease in
bed particle mobility are not suitable for FB incineration. The primary waste
characteristics identified as potentially restrictive include sodium content,
corrosivity, moisture content, and fusable ash content.
Sodium content has been identified as the most significant property of
concern, in determining the applicability of fluidized-bed incinerators to the
treatment of a facility's hazardous wastes. Certain sodium salts, most
notably sulfates and nitrates, may form eutectic complexes with other
inorganic salts present in the bed which serve to bind bed particles together,
25
and thus destroy the fluidity of the bed.
Highly corrosive wastes pose a different threat to the integrity of the
fluidized-bed. The fluidity of the fluidized-bed are dependent upon
maintenance of a certain bed particle density and size distribution. Thus,
reactions which alter these properties are detrimental to the effective
operation of the bed. Corrosion of the bed may therefore lead to a loss of
fluidization and result in significantly lower destruction efficiencies
achievable by the system.
Wastes with very high moisture content may reduce the overall ,
effectiveness of the fluidized-bed system. Wastes containing more than
75 percent moisture, by weight, may require temperatures or residence times
20
which are not practical for an FB system. Pretreatment of wastes to
reduce high moisture content is highly recommended for fluidized-bed
incineration. Numerous standard dewatering techniques may be employed,
including fractionation, filtration, and settling.
10-34
-------
TABLE 10.6. OPERATING PARAMETERS OF FLUIDI2ED BED INCINERATORS
Feed materials:
Capacity:
Operating temperature:
Residence time:
Gaseous
Solids
Pressure drop:
Excess 02'
Air flow rate:
Typical bed thickness:
Air pollution control
Startup and shutdown
Bed particle size:
Granular solids, sludges, slurries are
best but can handle liquids, bulk solids,
as well.
2 to 200 x 106 Btu/hr heat input.
1,600 to l,850°F in combustion zone.
5 to 10 seconds
No limit
90% of height of fluidized bed (in H20)
30 to 50 percent
2.5 to 8.0 ft/sec
6 to 8 ft
Acid scrubber, particulate scrubber,
quench tower.
Rapid startups and shutdowns possible.
Continuous feed not necessary.
20 to 80 mesh
Source: Reference 20,
10-35
-------
The consequences of a high concentration of fusable solid byproducts of
waste combustion are very much the same for fluidized-bed incineration as
those associated with the formation of inorganic salt eutectic mixtures
described earlier. These materials may impair the fluidity of the bed by
binding the granular solids into large, nonfluid solids.
10.3 HAZARDOUS WASTE INCINERATOR PERFORMANCE
The performance of various hazardous waste incineration systems has been
the subject of extensive study. In almost all instances RCRA performance
standards have been achieved despite the tremendous variability of waste
characteristics and operating conditions employed. As noted by EPA in
Reference 26, available data gathered by MITRE Corporation (Reference 27)
demonstrate that all of the solvents of concern assessed in the
14 January 1986 proposed regulations have been, or are currently, destroyed
through incineration technology. Detailed data on the composition of
413 waste streams incinerated at 204 facilities were obtained and analyzed by
MITRE. It was concluded on the basis of this study that incineration is
demonstrated technology for each of the solvent constituents listed under
hazardous waste codes F001 through F005. These data, which show that
199-million gallons of solvent waste were incinerated yearly by the
204 facilities, are shown in Table 10.7. The following subsections will
summarize performance data for each of the major incineration technologies.
However, a summary of recent EPA sponsored facility testing activity is shown
in Table 10.8.(6^
10.3.1 Liquid Injection Incinerator Performance Data
The results of numerous test programs have indicated that the liquid
injection incinerator is a highly effective system for disposing of virtually
all types of liquid hazardous wastes. Liquid injection incinerator data are
available for a wide range of waste types. High performance levels were
achieved based on the three primary measures of performance; i.e., DREs,
particulate emissions, and HC1 emissions.
10-36
-------
TABLE 10.7. SOLVENT CONSTITUENTS PRESENT IN INCINERATED WASTE STREAMS
Constituents
Number of
waste streams
containing
constituent
Amount of
constituent
incinerated
(million gallons)
Acetone
n-butyl alcohol
Carbon disulfide
Carbon tetrachloride
Chlorobenzene
Cresols
Cyc lohexanone
1 , 2-dichlorobenzene
Ethyl acetate
Ethylbenzene
Ethyl ether
Isobutanol
Methanol
Methylene chloride
Methyl ethyl ketone
Methyl isobutyl ketone
N it robenzene
Pyridine
fetrachloroethylene
Toluene
1,1, 1-trichloroethane
1,1,2-trichloro 1,2,2-trifluoroethane
Trichloroethylene
Trichlorofluoromethane
Xylene
80
9
3
9
21
8
4
2
24
6
2
9
95
26
54
10
3
9
19
103
23
2
15
2
78
17.2
32.3
0.128
0.547
6.18
5.92
0.344
0.240
9.62
0.264
0.248
4.50
44.4
5.84
14.0
4.68
0.0057
4.38
6.74
17.3
5.05
0.0008
3.69
0.401
15.2
Total:
199
Source: MITRE Corporation. Composition of HazardousWaste Streams
Currently Incinerated. Prepared for U.S. EPA, Office of
Solid Waste. 1983.
10-37
-------
TABLE 10.8. INCINERATION FACILITIES TESTED
o
I
03
Facility Control device Waste
DREa
(number of
nines)"
Average
particulate
HGI control emissions
(average) (g/dscf)
Commercial rotary kiln-
liquid incinerator
(87 million Itu/hr)
Packed-tower adsorber, Drummed, aqueous, liquid
ionizing wet scrubber organic waste with carbon
tetrachloride, TCEC,
perchloroethylene,
toluene, phenol
Commercial fixed-hearth, Electrified gravel bed Liquid organic and aqueous
two-stage incinerator
(25 million Btu/hr)
Onsite 'two-stage liquid
incinerator
(6 million Btu/hr)
Commercial fixed-hearth
two-stage incinerator
(2 million Btu/hr)
Onsite liquid injection
incinerator
(4.8 million Btu/hr)
filter; packed-tower
adsorber
Packed-tower adsorber
None
Done
aqueous waste with chloro-
form, carbon tetrachloride,
TCE, toluene, perchloro-
ethylene
Liquid organic waste with
carbon tetrachloride,
dichlorobenzene, TCE,
chlorobenzene, chloro-
methane, aniline, phosgene
Liquid organic waste with
TCE, carbon tetrachloride,
toluene, chlorobenzene
Liquid organic waste with
analine, diphenylamine,
mono- and dinitrobenzene
5.3
4.4
4.4
4.7
6.7
99.4%
98.3%
99.7%
<4 lb/hrd
<4 lb/hrd
0.67
0.178
0.027
0.089
0.092
(continued)
-------
TABLE 10.8. (Continued)
Facility
Control device
Haste
Average
DREa particulate
(number of HCI control emissions
nines)** (average) (g/dscf)
Commercial fixed-hearth
two-stage incinerator
(10 million Btu/Hr)
Onsite rotary kiln with
liquid Injection
(35 million Btu/hr)
o
w
VO
Commercial fixed-hearth
two-stage incinerator
(75 million Btu/hr)
None
Venturi scrubber with
cyclone separators and
packed-tower adsorbers
Venturi scrubber
Aqueous and organic liquid 4.8
waste with carbon tetra-
ehloride, TCE, benzene,
phenol, perchloroethylene,
toluene, methylethyl ketone
Liquid organic, paint waste 5.3
and filter cakes with
methylene chloride, chloro-
form, benzyl chloride,
hexachloroethane, toluene,
ICE, carbon tetrachloride
Aqueous and organic liquids 4.6
and solid waste with methy-
lene chloride, chloroform,
carbon te-trachloride,
hexachlorocyclopentadiene,
toluene, benzene, fCE
4 lb/hrd
99.9%
98.3%
0.40
0.01
0,075
Destruction and removal efficiency (mass weighted average for all FOHCs)
bFor example, 99.995% ORE =4.5 nines.
CTCE = trichloroethylene.
"Bo HClj control device; waste is low in total organic chlorine content.
-------
The relationship between baseline levels of performance achievable by a.
liquid injection incinerator and key operating parameters such as combustion
temperature, waste feed rate, and residence time, and key characteristics such
as moisture content, ash content, and heat value, was studied extensively by
Midwest Research Institute (1984). In this study, eight incinerators,
equipped with liquid injection systems were tested. Some were exclusively
liquid injection incinerators, whereas others were combination units including
rotary kilns and fixed hearths.
While full details of process and waste characteristics are provided in
Reference 26, it should be noted that all incinerators tested achieved the
goal of greater than 99.99 percent destruction efficiency for liquid and solid
wastes containing many solvents of concern, including chlorinated solvents.
The results of the MRI study and some key operating parameters are
summarized in Table 10.9. The table also includes some data from another
study (Reference 29) in which DREs were measured for a liquid injection
incinerator at Plant I firing two industrial organic containing wastes.
Again, additional details can be found in the original reference, and in other
references such as 11, 21, 23, 25, and 30. Results of the extensive
Reference 28 study are representative of the total body of incineration
literature as it relates to the destruction of organics. These results are
summarized below.
"1. Extensive analysis of organics emissions data provided the following
insight into the factors affecting DRE:
• DREs for the incinerators tested were generally above
99.99 percent. The average DRE for volatile organic
constituents was found to be 99.992 percent.
• DRE appears to be strongly correlated to concentration of the
POHC in the waste feed. POHCs at higher waste feed
concentrations were observed to be destroyed or removed to a
higher degree. The phenomenon that caused this relationship
was not identified.
* Analyses of data collected on this program showed no clear
correlation between DRE and heat of combustion for POHCs.
10-40
-------
TABLE 10.9. SUMMARY OF RESULTS OF INCINERATOR TEST PROGRAMS
o
Ho, of Average ««»tt Feed Rate
Facility Puna (Ibi/hr)
4,913
Aqueous M*tCe
4,763
Drummed *fa*ee to ChiwbBr
1,797
DrwBaoed W»»te to Kiln
1,567
Plant B 5 Organic «iste
1,854
Aqueous Waste
3,589
Plant C 3 243
Upjohn
PI line D 4 98.25
2ftfMlti
FUnt E 5 HA
Anericcn
Cy«tti«id
iesidence
Ash Chloride Moisture Temperature Tine Heat Input
HiBte Constiturnli % % % *C nee 106 Iba/hr
Carbon Tetrachloride 0.25 0.04 96 60-8?
frichloraeehyletie 1117-1154 $.5
Teceachloroethyleae 2.2? 5*7 23
Toluene
Methylrne Chloride 39.3 5.9 20
HeEhyl Ethyl Ketone
1 , 1, 1-Trichloroethana
1 ., 1 ,2-Trichloro*thane
Chlorofotra 1.1 0.43 95 HA |lA HA
Trichioroethylene 4,8 5.8 40
TefcrachloTQethylene
Toluene
Phenol
NftptHslene
Diethyl phth«l»te
Butyl Benzyl Phth«l*t*
Carbon TecnchlorUe 0.19 21 HA 1116 5.2 6. 2
Trichloroechylcne
Chlorobeozene
- Chloronethane
cr-DichlorobetiKtme
o-Diehlorobennene
1,2,4-frichlorobeosene
Bid(EH)PhthlUte j
Phenyl Itocyanate '
Aniline
Phosgene
Hethyiime Chloride 0.045 1,6 45<5 345-905 0.066 1.6
Girbon teCraehlorUe
Triehlotocthylene
Ta!u*ne
Chlorebencene
Anilln* 0,21 0.017 6.0 62 9999
99*99
(continued)
-------
TABLE 10,9. (Continued)
4>
to
ft)
Ficintr «
Flint T
Plant G
Flint H
Flint I
NA - Not
mi (Ibi/br)
« M9
1 liquid Hnt«
2 2iS
Flint »«»r»
J5I75
Filter C«k«
27J
ISUM Cote
119.4
DCS Coki
126
4 Liquid Orllnle UatEfl
331.65
645.75
'"'s^a'"*
7 Liquid Unt«
270
* Liquid »itt« ml Mo. 2 lu*t
available
Wilt* Conititucftei XX X
L-irbpn Tott.ehlorU. 0.71 0.» 91
Trichloroctliylenc
BenzcnA
T«tncliloro«tliyUne 2.0 1.0 >.4
Toluen*
Kitliyl Ethyl (clone
Plitnol
Klphrbllena
lytyl leneft Phtti«l«C«
kil(EH)flith«l»<
Motliylrae Chloride 0.13 25 2.5
OiloraforM
l,l.l-trichlorotth»ae 17 10 «5.2
Cartinn TetneMoriita
Trichlaroethylenc
TetriehlorortEiyleMe
Totuetie
4-Dichl(iro-2*luten4
Icniyl Clilocidt
Itexichloroetliine
HiplirhlUne
HctbrUoe Chloride J.6 2.5 S.I
Cblorafor* .
1,1,1-Trichlorot'tti.nc O.OS 0.01 9$
fri !ilaroethyl«ne 2.S 1.8
ten eat
Tec Achlaro«Ehylene
Tot cue
Chi (obtnctM
HKI chlcfocyclo^entl^iene
BictElOl'ttthilite
CMor«Jine
Nipth«l«fie
!l»..
-------
* Data compiled from the eight tests were not sufficient to
define parametric relationships between residence time,
temperature, heat input, or 02 concentrations and DUE.
* The data from the eight tests suggest that POHC levels in
scrubber water and ash were generally very low or
nondetectable. These data suggest that the majority of POHCs
are destroyed rather than merely transferred to another media
in the incineration process.
* Some Appendix VIII compounds detected in the stack (primarily
trihalomethanes) appear to be stripped from the scrubber water
by the hot stack gas. Trihalomethanes detected in the scrubber
inlet water were not detected in the effluent water. The
effect can be lower measured/calculated DREs even though the
destruction mechanisms may not be affected.
Evaluation of organic emissions data for compounds classified as
Products of Incomplete Combustion (PICs are defined in this program
as Appendix VIII compounds detected in the stack, which were not
found in the waste feed in concentrations above 100 pg/g) led to the
following observations:
• Stack gas concentrations of PICs were typically as high as or
higher than those for POHC compounds in the stack.
* PIC output rate infrequently exceeded 0.01 percent of POHC
input rate. (The 0.01 percent criterion was proposed in FR
Vol. 45, No. 197, October 8, 1980.)
* The three likely mechanisms that explain the presence of most
PICs are;
a. Poor DREs for Appendix VIII compounds present at low
concentration (<100 pg/g) in the waste feed;
b. Input of Appendix VIII compounds to the system from
sources other than waste feed (e.g., scrubber water); and
c. Actual intermediate products of combustion reactions or
products of complex side reactions including recombination,
» Data from the tests suggest that benzene, toluene, chloroform,
tetrachloroethylene, and naphthalene have a high potential for
appearing as byproducts of the combustion of organic wastes,
• A summary of the PICs detected in this study are given in
Table 10.10.
10-43
-------
TABLE 10.10. PICs FOUND IN STACK EFFLUENTS
PIC
Benzene
Chloroform
Bromodichloromethane
Dibromochlorome thane
Bromoform
Naphthalene
Chlorobenzene
Tetrachloroethylene
1,1, 1-Tr ichloroethane
Hexachl orobenzene
Methylene chloride
o-Nitrophenol
Phenol
Toluene
Bromo ch 1 or ome thane
Carbon disulfide
Methylene bromide
2,4,6-Trichlorophenol
Bromome thane
Chloromethane
Pyrene
Fluoranthene
Dichlorobenzene
Trichlorobenzene
Methyl ethyl ketone
Die thy 1 phthalate
o-Chlorophenol
Pentachlorophenol
2,4-Dimethyl phenol
Number
of sites
6
5
4
4
3
3
3
3
3
2
2
2
2
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Concentrations
(ng/L)
12-670
1-1,330
3-32
1-12
0.2-24
5-100
1-10
0.1-2.5
0.1-1.5
0.5-7
2-27
25-50
4-22
2-75
14
32
IS
110
1
3
1
1
2-4
7
3
7
2-22
6
1-21
Source: Reference 28.
10-44
-------
3. Compliance with the particulate standard of 180 mg/Nm was not
achieved at half of the sites tested. Particulate control devices
were operated at five of the eight facilities, and two of these five
failed to achieve the standard. Two of the three facilities without
control devices also failed the particulate standard. Data from
this study suggest that any facility firing wastes with ash content
greater than 0.5 percent will need a particulate control device to
meet the standard.
4. HC1 emissions were generally easily controlled to meet one of the
two criteria specified in the regulations - less than 1.8 kg HCl/hr
or greater than 99 percent removal efficiency."
In addition to the effluent data discussed above, the study included
analyses of two other residuals, ash and the scrubber liquor from the air
pollution control device at four sites. The results of the analyses are shown
in Table 10.11. The data indicate that both ash and scrubber liquor contain
concentration levels that are below the proposed treatment levels for the
solvents analyzed.
10.3.2 Rotary Kiln IncineratorPerformance Data
Rotary kiln incinerator performance in the destruction of hazardous
wastes has been studied extensively. The results of such studies have
indicated the effectiveness of rotary kiln systems for the destruction of
solvent hazardous wastes.
The available performance data generally focuses on the destruction and
removal efficiency achieved by the RK system relative to certain hazardous
waste constituents ("the Principal Organic Hazardous Constituents", or
POHCs). Also presented are the characteristics of wastes tested, including
the heat content and moisture content of the waste feed, and the key operating
parameters of the system during the test, including combustion temperature,
residence time, and HC1 and particulate matter removal efficiency. The
results of three recent performance tests are summarized in Tables 10.12
through 10.14. In brief, the following results were common to these and other
tests of rotary kiln incinerators.
* Destruction and removal efficiencies achieved by the rotary kiln
generally exceeded 99.99 percent.
10-45
-------
TABLE 10.11. RESIDUALS ANALYSIS AT FOUR FULL-SCALE INCINiEATORS (mg/L)
Carbon
letrachloride
Chlorobenzene
Cresola
Hethylene
Chloride
Methyl Ethyl
Ketone
Tetrachloro-
ethylene
Toluene
1,1,1-tricM.oro-
e thane
Trichloro-
ethylene
Sits A
Scrubber
Ash Liquor
Concentration* Concentration
< 0.010
—
<4.05 <0.001
— < 0.010
<0.010
— <0.010
— <0.010
— <0.010
— < 0.010
Site B Site
Scrubber
Ash Liquor Ash
Concentration* Concentration Concentration*
<0,05-0.05 < 0,001 <0,05
— — <0.05
— — —
— —
-------
TABLE 10.12. PERFORMANCE TEST SYNOPSIS
Description of Program;
Waste Form:
Constituents:
Program conducted at Cincinnati
Metropolitan Sewage District (MSD)
hazardous waste incinerator, to verify
U.S. EPA Trial Burn protocol and conduct
environmental assessment.
Liquid hazardous wastes.
Compound
Chloroform
Carbon Tetrachloride
Tetrachloroethylene
Hexachloroethane
Trichloroethane
Tetrachloroethane
Dichlorobenzene
(several other non-solvent
constituents were also noted)
Number of Test Runs:
Range of DREs Reported
99.99
99.96 - 99.99
99.97 - 99.99
99.99
99.99
99.99
99.99
6 runs were conducted with a "Pesticide"
waste containing chloroform, carbon
tetrachloride, tetrachloroethylene,
hexachloroethane, hexachlorobenzene, and
hexachlorocyclopentadiene *
3 runs were conducted with a "High
Chlorine" waste containing
trichioroethane, tetrachloroethane,
bromodichloromethane, pentachloroethane,
hexachloroethane, dichlorobenzene.
Effect of Varying Waste Feed Rate: •
Feed rate varied between 32.7 and
60.3 kg/min for "Pesticide" waste, and
between 49.0 and 62.1 kg/min for "High
Chlorine" waste.
For pesticide waste, the average DUE
went down as feed rate went up. The
same effect was noted for the "high"
chlorine waste.
The HC1 removal efficiency was also
noted to decrease as feed rate was
increased.
(continued)
10-47
-------
TABLE 10.12 (continued)
Effect of Varying Residence Time: * Range of residence time for pesticide
was 1.7 - 3.7 sees and for high chlorine
was 1.5 - 2.3 sees.
* DREs were found to be directly related
to residence time. As residence time
went up, so did ORE.
• HC1 removal also went up as residence
time went up.
Effect of Varying Temperature: • A direct correlation between combustion
temperature and DRE was noted for the
first 3 pesticide runs, and the 3 high
chlorine runs. As temperatures were
increased, average DRE went up. No
significant correlation was noted for
the second 3 runs of pesticide waste.
* Correlation between combustion
temperature and HC1 removal was
difficult to assess. For pesticide
runs, HC1 removal increased with
temperature. For the high chlorine
runs, HC1 removal decreased as
temperature increased.
Source: Reference 31.
10-48
-------
TABLE 10.13. PERFORMANCE TEST SYNOPSIS
Description of Program:
Waste Form:
Waste Constituents:
Process Characteristics
Effect of Varying Bulk Gas Temperature;
Effect of Varying Kiln Temperature:
Effect of Varying Residence Time:
Test of Rotary Kiln to determine the
effect on performance (measured by
DRE) of varying certain waste and
process characteristics.
Liquid waste. Usual waste mixture
combines liquid wastes (process
effluents) and solid wastes (sludges,
tars, etc.).
Wastes with different chemical
constituents were used inthe program.
Included as constituents were the
following solvent hazardous waste
constituents:
Acetonitrile (U003)
Methyl Ethyl Ketone (U139)
Ethylene DIchloride (U077)
1,4-Dioxane (UlOo)
Toluene (U220)
Methanol (U154)
Chloroform (U044)
Two bulk gas temperatures were used,
1600 and 1850°F. Two test runs were
conducted. No significant correlation
between bulk gas temperature and DRE
was observed. DRE was greater than
99.99 at both temperatures (all test
runs) for U077 and U220, while the
average DRE for U159 and U1U8 was well
below 9-9.99 at both temperatures.
Two kiln temperatures were examined,
1300 and 1600°F. A strong direct
correlation between temperature of the
combustion zone and DRE was noted.
DRE did not average above 99.99 at
either temperature for either U003 or
U154.
DRE was found to increase*when
residence time was increased from 1.2
sees to 3.3 sees. For U003, the
increase was significant, but for
U159, U1U8, U077, and U220, the
increases was less pronounced.
(continued)
10-49
-------
TABLE 10.13 (continued)
Waste Characteristics
Autoignition Temperature:
Heat Content of Waste:
Moisture Content of Waste:
Heat of Combustion (of constituents):
Solubility in Water:
No significant correlation between
autoignition temperature and DRE was
noted. Combustion was conducted at
temperatures above the autoignition
temperature of waste constituents.
DREs were noted to increase when the
heat content of the waste was
increased from 6,000 Btu/lb to 10,000
Btu/lb. This is attributed to a
higher flame temperature and more
rapid evaporation of higher heat
content waste mixtures.
DRE was noted to increase when
moisture content was increased from
18% to 40%.
No significant correlation was found
between DRE and heats of combustion of
waste constituents.
An inverse correlation between
solubility in water and DRE was
noted. This is attributed to a higher
rate of evaporation for those
constituents which are less
intermingled with water in solution.
Source: Reference 32.
10-50
-------
TABLE 10.14. PERFORMANCE TEST SYNOPSIS
Description of Program:
Waste Form:
Constituents:
Compound
1,1,1-Triehloroethane
Trichlorobenzene
Carbon Tetrachloride
2,4-Dichlorophenol
2,4,6-Trichlorophenol
Effect of Varying Waste Feed:
Effect of Varying Residence Time:
Effect of Varying Temperature:
Effect of Varying Excess Oxygen:
Trial Burn program conducted by the State
of Michigan Department of Natural
Resources.
Liquid wastes with solids.
Range of DREs Reported
99.996 - 99.998
99.992 - 99.995
99.996 - 99.999
99.999
99.999
DREs went down as waste feed rate was
increased.
DRE went up as residence time was
increased.
DRE went up as temperature was increased.
DRE went up as excess G£ went down.
Source: Reference 33.
10-51
-------
* A variety of wastes were bandied without need for pretreatment.
• High combustion temperatures could be used.
* Long residence times were achievable.
• DREs were directly related to length of residence time, higher
combustion temperatures, lowering of waste feed rate, and lower heat
content.
* No correlation was found between DRE and certain waste constituent
characteristics such as heat of combustion or autoignition
temperature.
10.3.3 Fluidized-Bed Incinerator Performance Data
Fluidized-bed incinerators have proven to be an effective means of
disposing of many types of hazardous wastes, as shown by a number of
performance evaluation studies conducted by both industry and U.S. EPA.
Several of the fluidized-bed systems in operation have become permitted
hazardous waste units through demonstration of their ability to perform within
guidelines established under RCRA. Trial burn data are available for several
of these units in service which show that fluidized-bed incinerators are
capable of destroying hazardous wastes of widely varying physical and chemical
characteristics to levels exceeding a destruction and removal efficiency of
99.99 percent. The results have further indicated that the FB systems
effectively control emissions of particulate matter and acid (HC1) produced
during combustion of waste materials. The results also show that DREs are
adversely affected, much as the other established incineration technologies
are affected, by low concentration of POHCs in the waste, and by very high
levels of moisture in the waste. The tests have also demonstrated the
specific limitations of FB relative to sodium salt content, corrosivity, and
fusable ash content, as discussed previously. A summary of typical
performance results for a fluidized-bed incinerator is presented in
Table 10.15.
The test program was conducted by plant officials under the supervision
of U.S. EPA representatives who conducted independent sampling and analysis.
The results of the test program were incorporated as part of a RCRA permit
10-52
-------
TABLE 10.15. FLUIDIZED-BED INCINERATOR PERFORMANCE TEST RESULTS
Waste
Run Feed Rate Waste Constituents
No. (Ibs/hr) (Principal Organic Hazardous Constituents)
1 575.65 l,l,2-Trichloro-l,2,2-Trifluoroethane
Trichloromonofluororaethane (TCHFM)
Tetrachloroethylene
Trichloroethylene
1,1, 1-Trichloroethane
2 632.55 l,l,2-Trichl-l,2,2-Trifl
TCMFH
Tetrachloroethylene
Trichloroethylene
g 1,1,1-TCE
V*
W 3 621.97 1,1,2-TC-1,2,2-TFE
TCMFH
Tetrachloroethylene
Trichloroethylene
1,1,1-TCE
Concentration
in Waste Feed
(X by weight)
0.30
1.5
4.3
0.72
5.3
0.29
1.2
3.8
0.68
5.0
0.28
1.9
3.6
0.89
4.9
Particulate
Destruction and Emissions
Removal Efficiency (Ibs/hr)
99.9918 8.95
99.9906
99.9976
99.9965
99.9995
99.9917 3.55
99.9941
99.9991
99.9964
99.9996
99.9912 3.83
99.9923
99.9907
99.9936
99.9994
HC1
Emissions
(Ibs/hr)
0.40
0.73
0.39
Source: Reference 34.
-------
demonstration. During the test program, liquid hazardous waste containing
five halogenated organic constituents (POHCs) and wastewater contaminated with
the same organics were burned. The waste stream was generated by chemical
manufacturing and consisted primarily of waste solvent. Analysis was made of
destruction and removal efficiency (DRE), particulate emissions, and chlorine
removal efficiency. Three test runs were made during the 2-day
demonstration. In general, the results indicate that the fluidized-bed
incinerator tested is capable of achieving performance levels which are within
the guidelines established by the U.S. EPA for permitting. In summary:
* Three test runs were completed.
* The average waste feed rate was 610 Ibs/hr and the average
wastewater feed rate was 800 Ibs/hr,
• The average heating value of the wastes was found to be
13,000 Btu/lb. The average heat input rate, therefore, was
8 x 106 Btu/hr.
* Chloride content of the waste feed averaged 13.8 percent. Ash
content of the waste averaged 4.7 percent.
* Combustion zone temperature averaged 2,201°F.
• DREs for the five POHCs averaged over 99.99 percent.
• Chloride removal efficiency averaged over 99 percent.
* Particulate emissions averaged 5.4 Ibs/hr.
* PICs were not measured in the test.
* Scrubber water was found to contain methylene chloride,
1,1,1-trichloroethane, tetrachloroethylene, toluene, and
ethylbenzene in very low amounts.
* Leachates were analyzed for heavy metals. None were detected in
significant amounts.
10.3.4 Performance of Multiple Hearth and Fixed Hearth Furnaces
Although testing of multiple hearth and fixed hearth incinerators has
been conducted, extensive performance data, other than that reported for two
liquid injection/fixed hearth systems in Table 10.9, could not be obtained for
10-54
-------
this document. A potential source of data of this type may be from
manufacturers of such systems, who operate pilot and test facilities to
demonstrate the effectiveness of the performance of their systems for various
wastes. Contact of manufacturers was made, but no data were available for
this document from the sources contacted.
10.4 COSTS OF HAZARDOUS WASTE INCINERATION
The overall costs associated with the incineration of hazardous wastes
are high relative to other hazardous waste treatment or disposal methods.
Incineration facilities require large capital costs due to the size and
complexity of the systems involved, and the requirements associated with the
handling of hazardous wastes and their combustion products. Operating costs
are high, primarily due to the large energy input required, and also as a
consequence of large raw materials costs and stringent environmental control
requirements. Incineration costs are difficult to specify, in general,
because in each situation the number of factors impacting costs is large.
These factors may be classified fundamentally as follows:
• Waste characteristics;
* Facility design characteristics; and
• Operational characteristics.
The general significance of many of the factors affecting incineration
costs will be discussed in detail below.
Waste Characteristics-
As discussed in a previous section, all aspects of the incineration
process design are related to waste characteristics. The chemical and
physical properties of a waste considered for incineration govern the type of
incinerator selected, the processing capacity, environmental controls
employed, pretreatment employed, required maintenance and equipment lifespan,
and operational parameter levels. Several waste characteristics which
significantly affect the costs of incineration are described below:
10-55
-------
* Physical State—Physical state dictates the type of incinerator and
the type of waste feed mechanism selected. Liquid injection
incinerators, for example, are applicable only to liquid wastes,
Limited data available on prices charged by commercial incinerators
suggest that solid and sludge wastes are more expensive to
incinerate than liquid wastes.
* Heat Value—Heat value is used as a measure of auxiliary fuel
requirement. The higher the heat value of a waste, the less fuel is
required to sustain combustion.
• Rheological Characteristics—The way in which liquid viscosity of a
waste changes with temperature is important in determining the need
for preheating, waste feed mechanism, and incinerator type. Some of
the wastes are easily handled at higher temperatures, while others
maintain viscosities which render then nonpumpable and/or
nonatomizable over practical limits of temperature.
• Water Content—Water content of a waste strongly affects temperature
and destruction efficiency of the combustion system. In some cases,
dewatering of wastes is conducted as a pretreattnent operation.
• Chloride Content—The chloride content of a waste has strong bearing
on the air pollution control methods employed at an incinerator.
High levels of chlorine necessitate acid gas scrubbing and also
require combustion methods which prevent the formation of toxic
chlorine gas.
* Ash Content/Heavy Metals Content—The amount of ash which will be
formed in combustion, and the nature of the ash is related to the
inorganic salt and heavy metal content of a waste to be incinerated,
and greatly affects the particulate matter air pollution control
requirement and the ash collection and disposal system design.
The impact of various waste characteristics on incineration costs may, in
some cases, be measured directly. A survey was conducted of a cross-section
of hazardous waste incineration facilities operating commercially in the
United States, and it revealed that pricing structures are often established
17 18 19 35—38
based on certain waste characteristics. » » » AS shown in
Table 10.16, chloride content and ash content commonly are used to establish
surcharges based on additional air pollution control requirements. The
physical form of the waste may also be seen as leading to price
differentials. In general, solid and sludge wastes cost more to incinerate
than liquid wastes.
10-56
-------
TABLE 10.16. SURVEY OF HAZARDOUS WASTE INCINERATORS - COSTS OF INCINERATION
AND COST IMPACTING FACTORS
o
Ul
•si
Facility
A
B
C
D
E
Incineration
System
Liquid Injection/
Rotary Kiln
Liquid Injection/
Rotary Kiln
Liquid Injection/
Rotary Kiln
Liquid Injection
Liquid Injection/
Rotary Kile
i
i
Costs to Incinerate Hazardous Wastes
Type of Waste
Blendable8 aqueous
Blendable organic
Directly-burned aqueous
Directlj-burned organic
Directly-burned sludges
or solids
Liquids
Solids and sludges
Any
Liquids
Bulk liquids
Drummed liquids
Drummed solids
Cost {$)
0.18/lb
0.2675/lb
0.2050/lb
0.2850/ib
O.S/lb
39/55 gal. drum
125/55 gal. drum
0.25/lb
0.86/gallon
1.93/gallon
230/drum
300/drum
Additional Costs
Basis
Phase separation
Chloride content or
ash content
Chloride content or
ash content
Suspended solids
Metals (e.g. chromium)
Chlorine
Handling fee
"Approval" charge
Cost
0.3475/lb
S/A
0.002/lb per
chloride or
O.Ol/gal per
<*)
each 1% of
ash content
1% content
0.0005/100 ppm/ gallon
0,02/gal per
2 51 drum
ISO/ job
11 chlorine
aBlendable defined as possessing a viscosity of below 10,000 SSI).
Source: References 17, 18, 20, 19, 35-38.
-------
Facility Characteristics—
Facility characteristics; i.e., the design and size of incineration unit
equipment, are measured in terms of capital costs. Capital costs for
incineration facilities are high relative to many other hazardous waste
management technologies, which are generally less complex and sensitive to
thermal and mechanical tolerances. For each incineration technology, there is
a large variation in the designs available commercially, and great differences
in the pricing policies of manufacturers. As a result, it is difficult to
specify a range of costs for any particular type of system.
To determine the cost of a hazardous waste incineration facility, several
key factors must be assessed. The elements contributing to the capital cost
of a "typical" incineration facility are presented in Table 10.17. Several of
the key factors influencing capital cost are listed below:
• Size requirments - flow rates, heat input capacities, exhaust rates,
etc.;
• Equipment lifespan;
• Pretreatment requirements;
• Heat recovery;
• Environmental control requirements;
• Feed mechanisms; and
• Equipment availability.
The size requirements of the system have the most bearing on capital
costs, while the environmental control equipment costs may be the largest
element of the overall capital costs. The capital costs of a particular
hazardous waste incineration system are strongly affected by the overall
availability of that technology. Certain systems, such as liquid injection
incinerators, are manufactured by many companies. Other technologies, most
notably the newer type systems, are manufactured by a few, or in many cases,
only one company.
10-58
-------
TABLED 10.17. ELEMENTS OF CAPITAL COST FOR INCINERATION SYSTEMS
I. Incineration System
A. Waste conveyance
1. Open or compaction vehicles, commercial containers
2. Special design containers
3. Piping, ducting, conveyors
B. Waste storage and handling at incinerator
1. Waste receipt and weighing
2. Pit and crane, floor dump and front-end loader
3. Holding tanks, pumps, piping
C. Incinerator
1. Outer shell
2. Refractory
3. Incinerator internals (grates, catalyst)
4. Burners
5. Fans and ducting (forced and induced draft)
6. Flue gas conditioning (water systems, boiler systems)
7. Air pollution control
8. Stacks
9. Residue handling
10. Automatic control and indicating instrumentation
11. Worker sanitary, locker, and office space
(continued)
10-59
-------
TABLE 10.17 (continued)
II* Auxiliary systems
A. Buildings, roadways, parking areas
B. Special maintenance facilities
C. Steam, electrical, water fuel, and compressed air supply
D. Secondary pollution control
1* Residue disposal 91andfill, etc.)
2. Scrubber wastewater treatment
III. Nonequipment expenses
A. Engineering fees
B. Land costs
C. Permits
D. Interest during construction
E. Spare parts inventory (working capital)
F. Investments in operator training
G. Start-up expenses
H. Technology fees to engineers, vendors
Source; Reference 16b.
10-60
-------
Capital cost data available for hazardous waste incineration systems were
limited. Several manufacturers of incineration systems were contacted, but
most were reluctant to specify costs for their systems because the cost for a
specific application is dependent upon so many different factors. A study
conducted by McCormick, et al., provided cost estimation curves for several
of the established hazardous waste incineration technologies: liquid
injection, rotary kilns, and hearth incinerators. These cost curves,
including estimation curves for heat recovery systems (waste heat boiler) and
acid gas scrubbing systems, are presented in Figures 10.5 through 10.10. This
information was generated in 1982, and has been updated to reflect the changes
in the Chemical Engineering Plant Index between May 1986 and the date for
which costs were estimated in Reference 5, A study conducted by MITRE
Corporation in 1981 in which several visits were made to incinerator
manufacturers generated additional cost data summarized in Table 10.18.
In general, it may be noted that certain hazardous waste technologies are
considered to be more expensive in terms of capital costs than are others.
Rotary kilns are most expensive. Relative capital costs for the five
established incineration technologies are as follows in order of decreasing
cost:
Rotary Kiln
Fluidized-Bed
Multiple Hearth
Liquid Injection
Fixed Hearth
Operating Characteristics—
Numerous factors impact the operating costs of a hazardous waste
incineration facility. The most significant factor governing operating costs
is energy usage. Energy is used in incineration to heat wastes in combustion,
and to operate materials transport and control systems. In many cases, the
energy usage of an incineration system is large enough to justify the costs of
10-61
-------
1,000
500 -
a 200 •
100 —
S so
Severe aervi.ee, chlorinated orgnnics and salts
2 5 10 20
Capacity, I-,.,, (million Btu/hr)
50 100
Figure 10.5. Purchase cost of liquid injection system (May 1982),
10,000
5,000
g
2 2,000
1,000
500
200
100
- 220 (0, Btu/hr)
0.478
l l l
I I
5 10
Capacity,
20
50
8t«/hr)
100
Figure 10.6. Purchase cost of rotary kiln system (May 1982),
Source: Reference 5
10-62
-------
1,000
soo
I 2M
«*•
-------
1,000
SOD
200
3 100
Q
«
•g
a so
20
10
.0.8164
"Qi,
2 S 10 20
Inlet: g«8 flowrate, q ^ (1,000 aefti)
SO
Figure 10.9. Purchase cost of scrubbing systems receiving
500 to 550°F gas (July 1982).
1,000
500
- a 200
I
a,
i
•a
100
so
10
1 1 j 1 I 1 1 II
10 20
50
100 200 300
Inlet gas flowrate, 4 (1,000 acfni)
Figure 10.10.
Purchase costs for typical hazardous waste incinerator
scrubbing systems receiving 1800 to 2200°F gas (July 1982)
Source: Reference 5
10-64
-------
TABLE 10.18. SUMMARY OF COST DATA COMPILED BY MITRE CORPORATION, 1981
Incineration Capacity
Facility Technology (mmBtu/hr)
Capital Cost
Description of Cost Factors
Fluidized Bed
10
o
o\
Ul
2 "Packaged"
Rotary Kiln
3 Rotary Kiln
37.5
700,000
40-50,000/(100 Ibs/hr)
800,000
Without energy recovery.
Scale-up factor for cost
estimation is 0.6 exponent.
Installed cost, including
heat recovery and air
pollution control.
Not installed. Includes 1
item of air pollution control,
Estimated installation cost
was 20 percent.
4 Rotary Kiln
5 Rotary Kiln
80-150
0.5
1.02
1.24
14.1
17.0
90
10-15 x 106
600,000
1.9 - 2.2 x
2.34 - 2.66 x
2.66 - 3.04 x
3.25 - 3.65 x
8.5 x 106
106
106
106
106
Total installed cost.
Total installed.
All not installed.
Total installed.
(continued)
-------
TABLE 10.18. (continued)
Incineration
Facility Technology
Capacity
(mmBtu/hr)
Capital Cost ($)
Description of Cost Factors
10
Liquid Injection 5
Fixed Hearth
Liquid Injection 18
70
Combined Liquid 150
Injection and
Rotary Kiln
Liquid Injection 30
11 Pyrolysis
150,000
300,000
150,000
300,000
500,000
1.5 x 106
2.2 x 10l
400-500,000
3,000 Ibs/hr 1 x 106
6,000 Ibs/hr 4 x 106
Base cost, not installed, no
APC, heat recovery.
Total installed with APC,
Base cost, no APC or heat
recovery, not installed.
Installed with APC.
Not installed.
Total installed cost.
Scale-up factor is exponent -
0.65.
Not installed. No APC, heat
recovery.
Estimated cost of APC given
is 1.2 x 106.
With boiler and scrubber, not
installed.
Including heat recovery, no
APC installed.
Source Reference 20.
-------
installing and operating heat recovery systems. A summary of the more
important operating characteristics, besides energy usage, affecting the costs
associated with incineration is presented below:
* Residence Time—Residence time affects the volume of the combustion
chamber, secondary combustion requirement, and the exhaust rate.
Residence time may be increased by employing devices such as baffles
or recirculation blowers.
* Temperature—Temperature affects the volume and type of the
incinerator refractory lining, volume of insulation for other
systems, and the need for heating and cooling systems.
* Raw Materials Usage—A variety of raw materials are used in
incineration systems, including chemical agents, fluidized-bed
granular material, scrubber and cooling tower water, sorbents, and
oxygen. The comsumption of these materials leads to additional cost
considerations.
* Maintenance—Maintenance requirements for incineration systems is
considered high due to the number of systems involved and the
thermal and mechanical stresses they operate under. The maintenance
of refractory linings is considered a particularly significant cost
consideration.
• Disposal—Disposal of solid and liquid combustion byproducts can be
a very expensive proposition depending on the characteristics of the
materials produced. In some cases, systems are limited in
applicability based on disposal costs of, for example, heavy metals
containing wastes.
Because of the uncertainties in many of the above items, it is difficult to
assign meaningful values to elements of operational costs. These factors,
however, have been considered by operators in assigning differential values
based on waste characteristics (see Table 10.16).
10.5 STATUS OF DEVELOPMENT
19 20
10.5.1 Hazardous Waste IncineratorManufacturing Industry *
Several surveys have been conducted to determine the number of companies
currently involved in the development, manufacture, and installation of
hazardous waste incineration systems. Investigation of the current hazardous
10-67
-------
waste incinerator market in 1982, .indicates that there are approximately
67 companies known to be actually involved. This number may not necessarily
include the number of companies who are devloping newer, more innovative
thermal technologies. The conventional technologies offered by these
commercial companies are summarized in Table 10.19. In general, the following
conclusions, drawn by the MITRE corporation in 1982, are supported by these
data:
• "About 342 incinerators have been put into hazardous waste services
since January 1969. These units were manufactured by 29 companies,
all of which were based in the United States at the time the units
were delivered. Within the past year one of these companies (BSP
Envirotech) was purchased by a West German firm, the Lurgi
Corporation. The count of 342 units is believed to be reasonably
accurate, but cannot be exact for the following reasons:
- A number of small vendor companies have disappeared since
1969. These companies have probably manufactured a few
incinerators which are still in use, but \their existence could
not be determined.
- Incinerators originally sold for hazardous waste disposal, or
for nonhazardous wastes, could be operating, at least part
time, on the other waste.
- Some incinerators have been manufactured strictly in accordance
with a customer's specifications and the manufacturing company
has no knowledge of, or declines to speculate on, the nature of
the purchaser's wastes.
- A few incinerators which have been manufactured since
January 1969 are probably no longer in use. A vendor will not
generally know this.
- A few incinerators manufactured since 1969 cannot fulfill the
design function and are not operating. Vendors will not
voluntarily acknowledge these.
* The most common type of hazardous waste incinerator is liquid
injection, representing 64.0 percent of all hazardous waste
incinerators in service. This type of incinerator is not designed
to operate on liquids containing any significant amount of salts or
other suspended or dissolved solids,
* The next most common types of hazardous waste incinerators are the
fixed hearth (FH) and the rotary kiln (RK), with 17.3 and
12.3 percent, respectively, of the total manufactured. Both of
these types of units will dispose of solid and/or liquid wastes,
plus fumes.
10-68
-------
TABLE 10.19. NUMBER OF HAZARDOUS WASTE INCINERATORS IN SERVICE IN THE U.S.A.
Type
Liquid Injection
Rotary Kiln
Fixed Hearth8
Multiple Hearth
Fluidized Bed
No. of
Companies
Offering
23
17
15
2
9
No. in
H.W. Service
219-231
42
64
7
9
Range of
Capacities
3-300 mmBtu/hr
1-150 mmBtu/hr
200-2500 Ibs/hr
1000-1500 Ibs/hr
N/A
Numerical
Share of
Market (%)
64
12.3
18.5
2.0
2.6
alncludes other hearth-type systems including Pulse Hearth (2), Rotary
Hearth (2), Reciprocating Grate (1).
N/A - Information not available.
Source: Mitre, 1982 (Reference 20).
U.S. EPA, 1983 (Reference 19).
10-69
-------
» Although there are nine companies offering fluidized-bed (FB)
incinerators, only nine such units are in hazardous waste sevice.
Apparently mo'st of these nine companies belive that the market is
potentially good for this technology.
• Two companies are actively marketing fused salt bath technology, but
there are no units in service or under construction yet.
* Of about 219 liquid injection units in service, about 129
(59 percent) were produced by two companies, John Zink and Trane
Thermal. The data furnished by Zink are not well verified. Of
23 companies marketing LI incinerators, eight have sold no units to
date. However, several of the eight indicate that sales are
imminent.
* Of the 17 companies offering rotary kiln incinerators, eight have
sold none to date.
* Of the nine companies offering fluidized-bed incinerators, five have
sold none to date.
• Of a total of 57 companies offering 14 types of incinerators, 28
have sold no units in the United States (Several companies represent
European technology, and all have sold at least one unit, each, in
Europe).
• The fact that 28 (of the total of 57 companies) have not sold any
units to date is indicative of the extent of: (1) new technology
being made available in the United States by both U.S. and foreign
companies; (2) the formation of new corporate ventures in the field
of technology; and (3) efforts by European companies to invade the
U.S. market. It is therefore believed that the market, or
technology, is not static at this point in time.
* Two companies are allegedly developing new technology, which they
would not describe at this time. It is known that other companies
are researching other techniques for hazardous waste incineration,
but these technoques are not described in this report. The new
processes included plasma, microwave plasma, and several unusual
fluidized-bed techniques."^"
10.5.2 Liquid Injection Incineration
Liquid injection incinerators are currently the most common type of
hazardous waste incinerator in service in the United States. In 1982, MITRE
estimated that 219 of these units were in service representing 64.0 percent of
the total. The number of companies reported to produce such units for
hazardous waste service was 23. Many manufacturers of liquid injection and
10-70
-------
other types of incinerators can be found listed in McGraw-Hill's Chemical
Engineering Equipment Buyers' Guide. No major breakthroughs in liquid
injection technology are anticipated.
Although considerable work has been performed to demonstrate the benefits
of cyclonic systems as opposed to conventional systems, due in part to
increased turbulence (and therefore, better mixing of waste and air), it is
questionable as to what impact such designs will have on the LI market.
According to manufacturers, many customers are seeking the simplest system
possible, to minimize maintenance problems. Manufacturers of liquid injection
systems feel that costs will remain low and possibly improve relative to other
13 24
incineration techniques. ' Liquid injection incineration is probably the
lowest cost incineration alternative due primarily to the simplicity of the
design and lack of moving parts requiring large amounts of energy.
The advantages and limitations of liquid injection incineration can be
summarized as follows.
Advantages—
1. Liquid injection incinerators are the most cost effective means of
incinerating atomizable liquid hazardou wastes. Performance tests
indicate IDlEs of over 99.99 percent have been achieved for most
types of liquid wastes.
2. Capable of incinerating liquid wastes containing a wide range of
physical and chemical properties.
3. Continuous removal of ash may not be required.
4. Capable of operating with no particulate control system, except for
moderate to high ash content wastes.
5. High temperature combustion of up to and exceeding 3,000°P is
possible.
6. Capable of high turndown ratio (maximum-to-minimum feed rate) which
provides flexibility in feed requirements.
7. Fast temperature response to changes in fuel flow rate.
8. Simple, adaptable design. Can retrofit a liquid injection system to
an existing incineration system.
9. Virtually no moving parts. Generally low maintenance requirements
and low maintenance costs.
10-71
-------
10. Overall low capital and operating costs.
11. Most widespread application of any incineration technology. Proven
in many applications. Has achieved RCRA permit status.
Limitations—
1. Capable of incinerating only liquid wastes that can be atomized
through a burner nozzle.
2. Nozzles are a source of plugging, erosion, and corrosion.
3. Difficult to incinerate high surface tension liquids.
4. Difficult to control carry-over of unbound liquid droplets.
5. Can experience mixing problems, leading to inefficient combustion.
6. Can experience high emissions of oxides of sulfur and nitrogen, acid
gases, particularly during high temperature operation.
7. May not destroy waste constituents effectively at low concentrations.
10.5.3 Rotary Kiln Incinerators
Rotary kiln incinerators have gained widespread commercial acceptance in
the hazardous waste management industry, despite being one of the more costly
available alternatives. This acceptance is due primarily to the versatility
of rotary kilns. There are many facilities, for example, which currently
employ a rotary kiln as their sole means of disposing of both hazardous and
nonhazardous wastes. In general, it is believed that as more emphasis is
placed on utilizing available alternatives to land disposal of hazardous
wastes, "multipurpose" technologies such as rotary kiln incinerators will gain
more acceptance. In addition, the utilization of rotary kiln technology may
increase significantly if high temperature industrial kiln processes are
utilized as a means of hazardous waste disposal. Among those technologies
currently being studied are cement and lime rotary kiln systems. A great
reduction in cost may be realized by using existing industrial systems for
hazardous waste disposal.
10-72
-------
As indicated in Table 10.19, 17 companies were identified in 1982 as
having developed, and are now actively marketing rotary kiln incinerators
specifically for hazardous waste disposal. Of that number, at least nine
companies have sold units which are currently in service at hazardous waste
19
management facilities. In addition to these companies, there are numerous
other firms who have developed and produced rotary kiln systems for industrial
applications such as aggregate and lime rotary kilns. As the industrial
processes are developed and put into use for hazardous waste disposal, many of
these firms may become more involved in marketing their systems for hazardous
waste disposal.
The advantages and disadvantages of rotary kiln incinerators are as
fo1lows:
Advantages—
1. Will incinerate a wide variety of liquids, slurries, sludges, tars,
or solid wastes, either separately or in combination.
2. Adaptable to a wide variety of feed mechanism designs, including
those for containerized wastes.
3. Characterized by high turbulence, thus provides good mixing of waste
with combustion air, and good dispersion of waste to increase heat
transfer surface area.
4. Can operate at temperatures up to or exceeding 2,500°F.
5. Can control residence time by adjusting rotational speed. Thus,
slow burning materials may be retained for a very long period of
time.
6. Can achieve a turndown ratio (maximum to minimum feed rate) of
approximately 2:1.
7, There are no moving parts within the kiln.
8. Continuous ash removal does not interfere with oxidation of wastes.
9. Requires minimal preparation of wastes.
10. Adaptable for use with wet gas scrubbing system.
10-73
-------
Limitations—
1. High capital costs for installed system, particularly if secondary
combustion and heat recovery are included.
2. Maintenance costs are high due to refractory lining maintenance and
replacement, and repair and maintenance of various rotating parts.
3. High energy costs required due to relatively low thermal efficiency.
4. Limited use for highly corrosive materials which could potentially
damage refractory lining.
5. Air inleakage from end seals is a common operational problem.
6. Some fusable material may collect in kiln.
7. May generate high levels of airborne particulates due to increase
turbulence. Requires additional control for particulate matter.
10.5.4 Fuidized-Bed Incinerators
Fluidized-bed incineration is a processing technology that is finding
increasing commercial application at facilities for the purpose of managing
hazardous wastes. Standard fluidized-bed systems are an established
technology alternative, actively produced and marketed by about ten
companies. It is expected that their relatively small current market share of
approximately 3 percent (as determined by MITRE in 1982), will rise, perhaps
surpassing less efficient systems such as multiple hearth furnaces. However,
due to the fact that their applicability to wastes with widely varying
physical and chemical characteristics may never approach that of rotary kilns,
or that their operating costs for liquid wastes may never be competitive with
liquid injection systems, fluidized-bed incinerators may never become a
predominant technology for the incineration of hazardous wastes.
Of the ten fluidized—bed incinerators known to be in service burning
hazardous wastes, the majority, seven, were sold by one firm. The company has
sold most of these units to petroleum refineries, where they are in operation
disposing primarily of sludges and a limited amount of contaminated sand and
soil and off-specification liquid solvent wastes. Only one FB system was
20
dedicated in 1982 solely to the destruction of hazardous wastes.
10-74
-------
The advantages and disadvantages of fluidized-bed incineration have been
summarized in Reference 9, as noted.
"Advantages associated with fluidized-bed incineration include the
following:
1. General applicability for the disposal of combustible solids,
liquids and gaseous wastes.
2. Simple design concept, requiring no moving parts in the combustion
zone.
3. Compact design due to high heating rate per unit volume (100,000 to
200,000 Btu/hr-ft3 (900,000 to 1,800,000 kg-cal/hr-m3)) which
results in relatively low capital costs.
4. Relatively low gas temperatures and excess air requirement which
tend to minimize nitrogen oxide formation and contribute to smaller,
lower cost emission control systems.
5. Long incinerator life and"low maintenance costs.
6. Large active surface area resulting from fluidizing action enhances
the combustion efficiency.
7. Fluctuation in the feed rate and composition are easily tolerated
due to the large quantities of heat stored in the bed.
8. Provides for rapid drying of high moisture content materials, and
combustion can take place in the bed.
9. Proper bed material selection suppresses acid gas formation; hence,
reduced emission control requirements.
10. Provides considerable flexibility for shock load of waste;
i.e., large quantities of waste being added to the bed at a single
time.
Potential disadvantages include:
1. Difficult to remove residual materials from the bed.
2. Requires fluid bed preparation and maintenance.
3. Feed selection must avoid bed degradation caused by corrosion or
reactions with the bed material.
10-75
-------
4. May require special operating procedures to avoid bed damage.
5. Operating costs are relatively high, particularly power costs.
6. Possible operating difficulties with materials high in moisture
content.
7. Formation of low melting point eutectics is a serious problem.
8. Hazardous waste incineration practices have not been fully developed.
9. Not well suited for irregular, bulky wastes, tarry solids, or wastes
with a fusible ash content.
There are two sources of waste incineration inefficiency associated
with fluidized-bed incineration: (1) incomplete oxidation of the
volatiles and (2) loss of solids which contain unoxidized combustibles.
The incomplete oxidation of solids presents the greater difficulty in
attaining complete incineration because solids generally require a longer
time for complete oxidation than gases at a specific temperature. The
loss of incompletely oxidized solids can occur by elutriation or by
removal with the bed material. The bed material must be removed and
regenerated, continuously or periodically, because of build up of
noncombustibles or attrition of the inert heat carrier. Inadequate
residence time of solid wastes is a major cause of inefficiency of
fluidized-bed incineration."'
10.5.5 Environmental Impacts of Incineration
Incineration processes potentially affect the environment through
generation of air emissions, and liquid, sludge, and solid wastes. As a
result, EPA has established environmental standards of performance for
incinerators in the RCRA permit process. Most incinerators must be equipped
with appropriate air pollution control systems, leading to higher capital and
operating costs. Environmental impacts associated with incineration are,
therefore, a significant factor in the determination of the appropriateness of
incineration as a management option for hazardous wastes.
10.5.5.1 Air Emissions—
Air emissions of pollutants produced in incineration are a primary area
of environmental concern. Emissions may be emitted from the incinerator stacK
and from fugitive emission sources. Emissions from incinerators primarily
10-76
-------
consist of the following "criteria" pollutants: oxides of nitrogen and
sulfur, and particulate matter. Other air pollutants of concern include
undestroyed organics such as benzene, toxic heavy metals (in particles),
hydrochloric acid, and other acid gases.
As part of the ROM permit process, incinerators must demostrate their
ability to achieve various performance levels established by EPA. Among the
performance criteria are emission standards for hydrochloric acid gas and
particulate matter. These standards are:
1. Hydrochloric acid emissions are limited to a rate of 4 Ibs/hr or, if
acid gas scrubbing is employed, a scrubbing efficiency of 99 percent
or greater; and
2. Particulate matter emissions are limited to 0.08 grains per dry
standard cubic feet of flue gas at 7 percent oxygen (180 milligrams
per dry standard cubic meter).
Emissions from incinerators are also regulated under Federal NESHAPs and state
air toxics program standards. These may affect, in particular, the emissions
of heavy metals such as lead or mercury vapors.
Available technologies for the control of emissions from hazardous waste
incinerators includes devices to control emissions of particulate matter, acid
gases, oxides of sulfur, and possibly oxides of nitrogen. Gaseous pollutant
control devices include various wet and dry scrubbers. Both wet and dry
scrubbing systems are effective in removing acid gases, although the dry
scrubbing systems are newer and, as a result, not as well established as the
wet systems. Oxides of nitrogen emissions can sometimes be minimized by the
use of combustion modifications which reduce the peak flame temperature in an
incinerator.
For control of particulate matter, the primary candidates are the wet and
dry electrostatic precipitators (ESPs), ionizing wet scrubbers, and
baghouses. Conventional scrubbers are not very effective in the removal of
fine particulate matter, Particulate matter control devices must be
compatible with the acid removal device. A wet acid scrubber is more
compatible with a wet ESP or the ionizing wet scrubber than with a baghouse.
10-77
-------
Since baghouses are compatible only with a dry gas system, the use of
baghouses on hazardous waste incinerators is not as prevalent as ESP usage.
Properly designed baghouses and ESPs are both effective particulate matter
control devices. Discussion of the various types of emission control devices
used on hazardous waste incinerators and their control efficiency capabilities
can be found in numerous texts and publications dealing with air pollution
control and incineration.
10.5.5.2 Liquid and Solid Wastes Generation—
Wastes formed both in the combustion unit and in pretreatment and air
pollution control systems constitute a potential environmental hazard which
must be properly managed. Presence of hazardous materials in the incinerator
ash, scrubbing liquor, and scrubber sludges is primarily dependent upon two
factors: composition of wastes fed; and destruction effectiveness of the
incinerator. The primary constituents of concern in these residues are
thermally inert materials such as toxic heavy metals. Toxic organic compounds
are generally not a significant contaminant of these streams, owing to good
destruction efficiencies.
Ash from incineration—Incinerator ash formed during the combustion
reaction consists almost entirely of thermally inert materials (metals and
other inorganics) introduced in the waste feed. Ash, not emitted with the
combustion flue gas, generally collects at the bottom of the incinerator
unit. Many incinerator designs include a conveyor system which continuously
removes ash from the bottom of the unit for subsequent disposal.
Contaminated ash is now commonly disposed of in a Class I landfill. As noted
in Reference 28, ash residuals from incineration have been found to be
suitable for landfill disposal. Alternatives to direct landfilling, if
required, could include encapsulation/solidification treatments.
Scrubber Liquor/Scrubber Sludges—Scrubber systems, which directly
contact the gaseous by-products of combustion with liquid (or solid) media,
most commonly water, may produce contaminated liquid or solid waste streams.
10-78
-------
The primary contaminants of such streams are toxic solid particles carried as
fly ash, acids such as hydrochloric acid formed in combustion, and various
organic products of incomplete combustion. The quantity, quality, and types
of liquid wastes formed from the control equipment is dependent on the
constituents of the waste feed, destruction efficiency, and collection
efficiency.
Liquid effluents from scrubber and quench systems usually will undergo
neutralization and removal of solids before they are discharged to local
sewage systems. A very common practice is to discharge these streams to
settling ponds (volatilization of organics from these ponds is not considered
a significant problem). Sludges are commonly treated in a sewage sludge
incinerator, or are dewatered and directly landfilled. Residual analysis of
scrubber liquor and s
of organic materials.
28
scrubber liquor and-sludges have indicated that they are essentially free
10.5.6 Summary—
The advantages and disadvantages of the various incineration technologies
available for the destruction of solvent hazardous wastes are presented in
Table 10.20. In general, most of the common incineration technologies might
be used to burn solvents, depending upon the individual characteristics of the
waste.
10-79
-------
TABLE 10.20. SUMMARY OF INCINERATION TECHNOLOGIES
o
i
03
O
Incineration
Method
liquid Injection
Rotary Kiln
Flutdlzed Bed
Fixed Hearth
Multiple Hearth
Limitations
Feedstock must be Ktcmlznble;
relatively tree of
participates
Requites largr butch
throughput to be practical
or economical
Bequirea large batch
throughput; limited to
liquids or non bulky solidtr;
no sod turn* salt wastes
Requires afterburner;
cdn*t burn liquids if
use continuous ash recovery
Requires afterburner;
can't burn bulky solids,
corrosives
Advantages
Can process all types
of hazardous liquids
Can process virtually
any type o£ waste;
can coinctnerate
various types of wastes
Con process many
wastes types;
good temperature response
in processing
Gun achieve very high
combustion temperatures;
Best for sludge incin-
eration; low capital
cost
Disadvantages
Requires pretreatraent to
remove impurities, heat,
and blend
Requires air pollution
.controls
Requires periodic bed
replacement; requires
air pollution controls
Hot energy-efficient;
requires higher tetnpera-
Possible high maintenance
costs; not energy efficient
Approximate Costs
Capital Operating
Jft-500,000 for 30 nraBtu/hr Jl-250/1000 gal
(Installed, with he«t
recovery snd AIC 1982)
t40,000-50,000/(100 */hr) *2500-1000/
J10-15 x 106 for 80-150 ton/diy
mBtu (total Installed,
1932)
1700,000 for 10 nrnBtu H/A
(total installed, no heat
recovery, 1982)
13-400,000 for 10 mBtu t0.5/lb
(installed, 1982)
H/A H/A
Source; Engineering-Science (Retertmee 10).
-------
REFERENCES
1. "National Survey of Hazardous Waste Generators and Treatment, Storage and
Disposal Facilities Regulated under RCRA in 1981," U.S. Government
Printing Office Order No. 055000-00239-8; EPA: Washington, B.C., 1984.
2. "A Profile of Existing Hazardous Waste Incineration Facilities and
Manufacturers in the United States," PB-84-157072; EPA: Washington, B.C.,
1984.
3. Martin, E. J., and L. W. Weinberger, et al. Practical Limitation of
Waste Characteristics for Effective Incineration, Presented at the
Twelfth Annual Research Symposium on Land Disposal, Remedial Action,
Incineration, and Treatment of Hazardous Waste. Sponsored by U.S.
Environmental Protection Agency, Hazardous Waste Engineering Research
Laboratory, Cincinnati, OH. April 21-23, 1986.
4. Federal Register 1982, 47, 27516-35.
5. McCormick, R. J., et al. Costs for Hazardous Waste Incineration.
Capital, Operation and Maintenance, Retrofit. Acurex Corporation,
Mountain View, CA. Noyes Publications, Park Ridge, NJ. 1985.
6. Oppelt, E.T. Hazardous Waste Destruction, Environmental Science and
Technology, Vol. 20 No. 4, 1986.
7. Hanson, L., et al. Hazardous Waste Incinerator Design Criteria.
EPA-600/2-79-198. TRW, Inc., Redondo Beach, CA. Prepared for U.S.
Environmental Protection Agency, Industrial Environmental Research
Laboratory, Cincinnati, OH. October 1979.
8. Lee, K. C., H. J. Jahnes, D. C. Macauley. Thermal Oxidation Kinetics of
Selected Organic Compounds. In: 71st Annual Meeting of the Air
Pollution Control Association, Houston, TX. June 23-30, 1978.
9. Advanced Environmental Control Technology Research Center. Research
Planning Task Group Study - Thermal Destruction. EPA-600/2-84-025.
Prepared for U.S. Environmental Protection Agency, Industrial Research.
Laboratory, Cincinnati, OH. January 1984.
10. Engineering-Science. Final Report. Technical Assessment of Treatment
Alternatives for Waste Solvents. Prepared for U.S. Environmental
Protection Agency, Technology Branch. November 1983.
11. U.S. EPA. Incineration and Treatment of Hazardous Waste: Proceedings of
the llth Annual Research Symposium. EPA-600/9-85-028. Articles cited
include:
(a) Olexsey, R., G. Hoffman, and G. Evans, "Emission and Control of
By-Products from Hazardous Waste Combustion Processes";
10-81
-------
(b) Gorman, P., and D. Oberacker, "Practical Guide to Trial Burns at
Hazardous Waste Incinerators";
(e) Westbrook, W., and E. Tatsch, "Field Testing of Pilot Scale APCDs at
a Hazardous Haste Incinerator";
(d) Clark, W. D., et al., "Emergency Analysis of Hazardous Waste
Incineration: Failure Mode Analysis for Two Pilot Scale
Incinerators";
(e) Bellinger, B., J. Graham, D. Hall, and W. Rubey, "Examination of
Fundamental Incinerability Indices for Hazardous Waste Destruction";
(f) Kramlich, J., E. Poncelet, W. R. Seeker, and 6. Samuelsen, "A
Laboratory Study on the Effect of Atomization on Destruction and
Removal Efficiency for Liquid Hazardous Wastes";
(g) Chang, D., and N. Sorbo, "Evaluation of a Pilot-Scale Circulating
Bed Combustor with a Surrogate Hazardous Waste Mixture";
(h) Evans, G., "Uncertainties and Incineration Costs: Estimating the
Margin of Error"; and
(i) Graham, J., D. Hail, B. Dellinger, "The Thermal Decomposition
Characteristics of a Simple Organic Mixture".
U.S. Environmental Protection Agency, Hazardous Waste Engineering
Research Laboratory, Cincinnati, OH. September 1985.
12. Lee, K. C., N. Morgan, J. L. Hansen, G. M. Whipple. Revised Model for
the Prediction of the Time-Temperature Requirements for Thermal
Destruction of Dilute Organic Vapors, and It's Usage for Predicting
Compound Destructability. In: 75th Annual Meeting of the Air Pollution
Control Association, New Orleans, LA. June 20-25, 1982.
13. Sultan, Omar. Telephone Conversation with M, Kravett, GCA Technology
Division, Inc. Peabody Engineering, Stamford, CT. February 1986.
14. Cooper, Doug. Telephone Conversation with M. Kravett, GCA Technology
Division, Inc. Rollins Environmental Services, Inc., Deer Park, TX.
February 1986.
15. Hooper, G. V., editor. Offshore Ship and Platform Incineration of
Hazardous Wastes. Pollution Technology Services, Revision No. 79. Noyes
Data Corporation, Park Ridge, NJ. 1981.
16. Edwards, B. H., J. N. Paullin, K. Coughlan-Jordan. Emerging Technologies
for the Control of Hazardous Wastes. Ebin Research Systems, Washington,
D.C., Noyes Data Corporation, Park Ridge, NJ. 1983.
17. Marti, Bruce. Telephone Conversation with M. Kravett, GCA Technology
Division, Inc. Chemical Waste Management, Inc., Chicago, IL.
February 1986.
10-82
-------
18. ICF Incorporated. RCRA Risk/Cost Policy Model - Phase III Report.
Prepared for U.S. Environmental Protection Agency, Office of Solid Waste,
Washington, D.C. 1984.
19. Incineration and Treatment of Hazardous Waste: Proceedings of the 8th
Annual Research Symposium. EPA-600/9-83-003. Articles cited include:
(a) Frankel, J., N. Sanders, and G. Vogel, "Profile of the Hazardous
Waste Incinerator Manufacturing Industry";
(b) Vogel, G. A., Frankel, I., and N. Sanders, "Hazardous Waste
Incineration Costs";
(c) Staley, L. J., G. A. Volten. F. R. O'Donnell, and C. A. Little, "An
Assessment of Emissions from a Hazardous Waste Incineration
Facility"; and
(d) Johnson, S. G., S. J. Yosium, L. G. Keeley, and S. Sudar,
"Elimination of Hazardous Wastes by the Molten Salt Destruction
Process".
U.S. Environmental Protection Agency, Industrial Environmental Research
Laboratory, Cincinnati, OH. April 1983.
20. MITRE Corporation. Survey of Hazardous Waste Incinerator Manufacturers,
1981. MITRE Corporation, METREK Division, McLean, VA. 1982.
21. Cross, F. C. Hazardous Waste Incinerators - Operational Needs and
Concerns. Cross/Tessitore and Associates, P.A., Orlando, FL. In:
Hazardous Waste and Environmental Emergencies - Management, Prevention,
Cleanup, and Control. March 12-14, 1984.
22. State of California Air Resources Board. Air Pollution Impacts of
Hazardous Waste Incineration: A California Perspective. Technical
Support Document. A Report to the California State Legislature.
Prepared by the California Air Resources Board. December 1983.
23. MITRE Corporation, Working Paper. Liquid Injection Incinerator Burner
Performance. WP-83W00393. MITRE Corporation, METREK Division, McLean,
VA. October 1983.
24. Loczi, Elizabeth. Telephone Conversation with M. Kravett, GCA Technology
Division, Inc. Ross Incineration Services, Inc., Grafton, OH.
April 1986.
25. Kiang, Y. H., and A. A. Metry. Hazardous Waste Processing Technology.
Ann Arbor Science Publishers, Inc. Ann Arbor, MI. 1981.
10-83
-------
26. U.S. EPA. Background Document for Solvents to Support 40 CFR Part 268
Land Disposal Restrictions, Volume II. January 1986,
27. MITRE Corporation. Composition of Hazardous Waste Streams Currently
Incinerated. Prepared for U.S. EPA, Office of Solid Waste. 1983.
28. Trenholm, A., P. Gorham, and G. Sungclaus. Performance Evaluation of
Full Scale Hazardous Haste Incinerators. EPA-600/2-84-181. Midwest
Research Institute, Kansas City, MO. Prepared for U.S. Environmental
Protection Agency, Office of Research and Development, Cincinnati, OH.
November 1984.
29. U.S. EPA. Destroying Chemical Hastes in Commercial Scale Incinerators
Facility Report No. 1. SW-122c.l. U.S. Environmental Protection Agency,
Office of Solid Waste, Washington, B.C. 1977.
30. FEDCo, Inc. Evaluation of the Feasibility of Incinerating Hazardous
Waste in High-Temperature Industrial Processes. EPA-600/2-84-049.
Prepared for U.S. Environmental Protection Agency, Industrial
Environmental Research Laboratory, Cincinnati, OH. February 1984.
31. MRI. Trial Burn Protocol Verification at a Hazardous Waste Incinerator.
Prepared for Rockwell International Corporation, Canoga Park, CA.
August 1982.
32. Air Pollution Control Association. Technical Conference on the Burning
Issue of Disposing of Hazardous Wastes by Thermal Incineration.
April 29-30, 1982. Hilton Gateway, Nework, NJ. Articles cited include:
(a) Deneau, K. S., "Pyrolytic Destruction of Hazardous Waste";
(b) Austin, D. S., R. E. Bastian, and R. W. Wood, "Factors Affecting
Performance in a 90 million Btu/hr Chemical Waste Incinerator:
Preliminary Findings";
(c) Bierman, T. J. and J. C. Reed, "Determination of Principal Organic
Hazardous Constituents (POHCs) in Hazardous Waste Incineration"; and
(d) Sesaverns, G. A., D. R. J. Roy, and W. B. Rossnagel, "Air Pollution
Control Technology: For Hazardous Waste Incineration".
33. Darnels, S. L. et al. Experience in Continuous Monitoring of a Rotary
Kiln Incinerator for CO, C02, and 02* Dow Chemical Co. in
Proceedings of 78th Annual Meeting. Air Pollution Control Association.
June 16-21, 1985.
34. Hall, R. R., et al. Union Chemical Trial Burn Sampling and Analysis.
Draft Final Report. GCA Corporation, Technology Division, Bedford, MA.
GCA Report No. GCA-TR-84-22-G. Prepared for Union Chemical Co.,
Uncon, ME. February 1984.
35* Anderson, R. Telephone Conversation with M. Kravett, GCA Technology
Division, Inc. IT Corporation, Martinez, CA. April 1986.
10-84
-------
36. Warren, P* Telephone Conversation with M. Kravett, GCA Technology
Division, Inc. Stablex Corporation, Rock Hill, SC. April 1986.
37. Garcia, G. Telephone Conversation with M. Kravett, GCA Technology
Division, Inc. TWI, Inc., Sauget, IL. April 1986.
38. Bell, R. Telephone Conversation with M. Kravett, GCA Technology
Division, Inc. Systech Corporation, Paulding, OH. April 1986.
10-85
-------
SECTION 11.0
EMERGING THERMAL TREATMENT TECHNOLOGIES
With the passage of the 1984 amendments to RCRA banning the land disposal
of hazardous wastes, thermal treatment of hazardous wastes has become an
increasingly attractive option. Accordingly, there has been a great deal of
interest shown in the development of new technological approaches to thermal
treatment. HWERL has identified 21 "innovative thermal processes for treating
or destroying hazardous organic wastes", many of which are applicable to
hazardous solvent wastes.
Emerging technologies, by definition, are processes which provide an
innovative or specialized approach to problems which have not been solved
effectively by existing technologies. These technologies, therefore, may be
more restricted by waste characteristics, technological complexity, or
economic feasibility than the established systems. On the other hand, they
may prove capable of achieving high destruction and removal efficiency (DRE)
levels, an accomplishment not possible for some established technologies, or
they may provide a radical improvement for a specific application. In
general, -these technologies have not been tested extensively on a full—scale
basis.
Emerging thermal treatment technologies include modifications of
conventional incineration technologies (e.g., the circulating bed incinerator)
as well as more unconventional approaches to thermal destruction, e.g. the
plasma arc pyrolysis system. Other emerging thermal systems such as wet air
oxidation and supercritical water oxidation have been discussed above in the
section dealing with chemical treatment processes. The technologies included
here, and discussed below, are:
11-1
-------
1. Circulating Bed Combustion
2. Molten Glass Incineration
3. Molten Salt Destruction
4. Pyrolysis Processes
5. In Situ Vitrification
11.1 CIRCULATING BED COMBUSTION
Circulating bed combustion (CBC) systems constitute an innovation in
fluidized bed incineration technology. These systems utilize high air
velocities and recirculating granular bed materials to maintain and achieve
combustion of waste under fluidized bed conditions. The circulating bed
material, serves to not only transfer heat energy and increase turbulence, but
can be chosen for its chemical characteristics to bring about reaction and
neutralization of certain products of combustion such as sulfur oxides and
hydrochloric acid. CBC systems are applicable to solids, liquids, slurries,
and sludges, over a wide range of heat values and ash contents. Numerous
performance tests have been conducted which indicate that circulating bed
combustion can achieve very high destruction and removal efficiencies, while
limiting other pollutant emissions to acceptable levels. CBC systems can
offer both technological and economic advantages over established
fluidized-bed incineration systems primarily due to the increased turbulence
of the system. CBC systems operate at higher air velocities, and are not
limited, as are fixed bed units, to the narrow range of design velocities
needed to maintain fluidization while, at the same time, limiting entrainment
and carry over of bed material.
11.1.1 Process Description
The circulating bed combustion process, depicted in Figures II. 1.1,
represents design innovation to standard fluidized bed (PB) incineration
systems. The CBC system is designed to handle all forms of waste, including
solids, liquids, and sludges.
11-2
-------
f f 18 TMiUfiTSNT
Figure 11.1.1. CBC Incineration pilot plant located at GA Technologies,
Source: Reference 2.
11-3
-------
The primary operating unit, the circulating bed combustor, incorporates a
two-chamber design consisting of a combustion chamber and a hot cyclone
chamber, as shown in Figure 11.1.1. The bed material used, limestone
(CaCO_), is fed to the system concurrently with the waste material.
Limestone is used because it readily reacts with sulfur and chlorine compounds
in the waste, to form relatively innocuous salts such as CaCl« and CaSO,.
The general reaction scheme for the CBC process is as shown in Figure 11.1.2.
Waste material is fed to the system either before the combustion loop for
solids and sludges, or just at the start of the loop for liquids. As stated
by the manufacturer, the CBC requires no specialized waste atomization or
other injection mechanism, due to the inherently high level of turbulence in
3
the system which ensures good distribution of waste feed. During operation
of the system, a high velocity stream of heated air (15 to 20 feet/sec.)
entrains the material and carries it up the combustion column. As the waste
flows upward, combustion occurs, and the byproducts are dispersed. The
gaseous products, primarily CO- and water vapor, flow out the top of the
combustor; the acidic byproducts such as HC1 react with the limestone to form
inorganic salts (generally these form as particulates); and, they and other
solid byproducts flow downward through the hot cyclone, in which solids and
gases are further separated. The hot flue gases pass first to a heat exchange
system, then to a particulate control device, before being vented through the
exhaust stack. Ash eventually settles within the combustion column and falls
to a screw conveyor (as shown in Figure 11.1.1) where it is transported to ash
recovery.
The circulating bed combustor is applicable to wastes with varying
physical characteristics. Because the effectiveness of the design is based
upon the development of a high degree of turbulence within the system,
pretreatment systems are usually unnecessary to supplement dispersion of waste
when fed (e.g., atomization of liquid wastes), or render wastes easier to
disperse; e.g., crushing or grinding of solids.
The operating conditions are as shown below.
• Waste Feed: Applicable to any physical form - granular
solids, liquids, sludges, slurries
• Temperature Range: 1400-1600°F (760-870°C)
11-4
-------
REACTANTS INTER
(HYDROCARBONS)
(Cxliy!.*z •» U2 •• S02
(SULHJR COMPOUNDS! ^
/
MEDIATES FINAL PRODUCTS
• ' 4 1
r •*. %n« , r-fSflar ,i
, r
-------
• Residence Time:
Gas Phase: 2-3 seconds
Solids: 10 seconds to 10 hours.
• Capacity (Ibs/hr): See Table 11.1.1
* Energy Type and Requirements:
Thermal: Sensible and latent heat; self-sufficient
for wastes up to 85 percent water content
Electrical: Blower and feeder operation—approximately
30 HP for 2 MMBtu/hr incinerator
Input capacities, shown in Table 11.1.1, are dependent upon the type of waste
fed to the unit. As noted the data were furnished by the developer; commercial
units covering the range of capacities have not yet been constructed.
Waste streams of primary environmental concern in the CBC process are:
(a) the acidic byproducts and organic products of incomplete combustion and
(b) hazardous heavy metals or other solid byproducts remaining in the ash. To
date, performance testing has indicated that acid or FICs in the flue gas
stream are not usually significant. The ash will be handled as a solid
waste. If hazardous materials exist, they will be disposed of in an
3
appropriate manner.
11.1.2 Demonstrated Performance
A pilot—scale CBC was tested by the California Air Resources Board in
4
cooperation with the manufacturer, GA Technologies, in 1983. The testing
involved a surrogate waste mixture which had a heating value of 8000 Btu/lb
and included organic compounds such as xylene, ethylbenzene, toluene,
hexaehlorobenzene, Freon, and carbon tetrachloride. The CBC unit operated at
a capacity of 0.5 MMBtu/hr, and a temperature of below 1600°F (870°C).
A summary of results is shown in Table 11.1.2. Some of the conclusions
drawn by Chang and Sorbo in Reference 4 are presented below:
1. The DRE of volatile and semi-volatile POHCs under less than optimum
combustion conditions met RCRA requirements (99.99% DRE).
11-6
-------
TABLE 11.1.1. CAPACITIES OF CBC SYSTEMS
Water Heat Throughput (Ib/hr) vs. Combuster I.D.
Content Content —
Legend Waste Description % % 16 In. 24 In. 36 in. 48 In. 60 in. 104 In.
PCB Contaminated Soil 10 0 1,260 2,840 6,380 11,340 17,720 53,240
PCB Contaminated Soil 20 0 930 2,080 4,680 8,320 13,010 39,130
Chlorinated Chen. Sludge 80 1,331 440 1,000 2,250 4,000 6,250 18,770
Chlorinated Chem. Sludge 40 4,000 340 770 1,740 3,100 4,840 14,530
Chlorinated Liquid Waste 4 7,606 210 470 1,060 1,880 2,940 8,840
Oil and Solvent Waste 13 11,227 130 280 640 1,140 1,780 5,340
Source: GA Technologies, Technical Bulletin.
-------
l&BLE 11.1.2. SUMMARY OF BAG SAMPLE RESULTS
oc
JJagSample Concentration
Compound Name ~^«_
0934-1035
1100-1124
1345-1418
1443-1623
Time
1630- 1705
1954-2054
2102-2140
2315-0020
0030-0100
POHCs - [ppbl
Trichlorotrifluoroethane
(Freon 113)
Average ORE
(Penetration)
Tetrachloromethane
(Carbon tetrachloride)
Average ORE
(Penetration)
28
0.9994538
5.46E-04
0.31
0.9999952
4.75E-06
0,41
0.9999950
4.98E-06
0.33
0.9999969
3.10E-06
0.46
0,9999969
3.08E-06
0.24
0.9999987
1.26E-06
0.096
0.9999987
1.33E-06
0.47
0.9999948
5.22E-06
0.056
0.9999996
3.82E-07
0.21
0.9999989
1.13E-06
0.042
0.9999994
5.81E-07
0.22
0.9999976
2.39E-06
0.063
0.9999996
3.92E-07
0.27
0.9999987
1.32E-06
0.025
0.9999997
3.17E-07
0.92
0.9999909
9.06E-06
0.045
0.9999997
2.55E-07
1.4
0.9999938
6.23E-06
PICs - Ippb]
Benzene
Total Volatile PIC
without Benzene - Ippb]
P1C-C1 - [ppb as Cl]
(Volatile PIC-Cl/Ci-in)
Fuel flowrate - [LB/HRJ
Air flowrate - {DSCFMj
Total gas flow - I DSCFMJ
Average 02 - l%]
Average CO2 - [%J
Average CO - tppm]
Average CO/CO2
Average THC - [ppml
Average THC/CO2
n.a.
115.8
129.8
1.33E-04
52.7
316
347.6
10.8
7.7
1493
2.10E-02
38
4.40E-04
7,400
2034.1
2039-9
1.30E-03
83.3
311
342.1
6.8
11,1
2630
2.36E-02
384
3.45 E-03
n.a.
4074.0
4079.4
2.87E-03
74.6
307
337.7
4,4
12.7
2792
2.07E-02
321
2.WE-03
n.a.
1528.4
1536.8
1.15E-03
68.4
300
330
5.8
11.8
1406
1.05E-02
224
2.09E-03
rc.a.
611.9
616.3
4.42E-04
73
306
336.6
4.9
12.5
738
5.66 E-03
194
1.51 E-03
37,000
2058.8
2063.8
1.50E-03
56.2
239
262.9
5.8
11.9
1851
1.47E-02
240
1.96E-03
n.a.
511.7
515.7
3.38E-04
61.2
234
257.4
7.7
10.4
523
4.89E-03
n.a.
n.a.
n.a.
255.8
258.9
1.73E-04
70.1
273
300.3
7.3
10.3
260
3.23E-03
49
4.42E-04
n.a.
103.8
108.4
6.46E-05
78.5
273
300.3
8.0
9.3
23
2.56E-04
30
3.21 E4>4
-------
2. Total volatile PIC formation was found to correlate well with CO and
THC, normalized to fuel flowrate (C02)« Penetration of volatile
chlorinated PICs (based on total chlorine content of the fuel)
eKceeded 1 x 10"^. PIC benzene appeared in substantial
concentrations in several samples and was not correlated with any
conventional combustion parameters.
3» The DRE dropped sharply when the bed temperature fell below 700°C.
Temperature appeared to be a major-factor in the destruction of the
fluorinated compounds and a moderate correlation between sulfur
hexafluoride, DRE and temperature was observed.
4. The CBC seemed to behave as a plug-flow reactor, susceptible to
pockets of non-stoichiometric air/fuel mixtures passing through the
bed causing increased PIC formation. This observation suggests the
importance of the fuel feed system on CBC performance and should be
evaluated carefully by permitting authorities.
GA Technologies, the developer, has reported more than 7,500 hours of
performance testing conducted under the auspices of DOE, EPRI (Electric Power
Research Institute), TVA, and a number of commercial sponsors. The system has
been tested with a variety of fuels and wastes to establish the combustion
efficiency and the pollutant removal efficiency of the system relative to
specific waste types. For solvent waste types, the system has generally shown
a DBE above 99.99 percent, and an HC1 capture of 99 percent and above. These
tests were conducted at the company's 2 x 10 Btu/hr test unit in San Diego,
CA. In summary, DREs found for POHCs existing in the organic wastes are as
shown below:
Solvent
Ethylbenzene
Trichloroethane
Vinyl chloride
1,2-trans-dichloroethylene
1,2-dichloroethane
DRE
99.99
99.999
99.9999
99.99
99.99
Temperature (°F)
1600 (871°C)
1600
1600
1600
1600
Additional results, furnished by the developer, GA Technologies, are shown
in Table 11.1.3.
11-9
-------
TABLE 11.1.3. TEST RESULTS ON HAZARDOUS WASTES CIRCULATING BED INCINERATOR PILOT PLANT
CHEMICAL
FORMULA
CCI4
C2CI3F3
C10H1906PS2
C12H7CI3
C6H4CI2
uQlBIll £
V fm *W
C2HCI3
CHEMICAL
NAME
CARBON
TETRACHLORIDE
FREON
MALATHION
PCB
DICHLOROBENZENE
AROMATIC NITRILE
TRICHLOROETHENE
PHYSICAL
FORM
LIQUID
LIQUID
LIQUID
SOIL
SLUDGE
TACKY
SOLID
LIQUID
DESTRUCTION
EFFICIENCY, %
99.9992
99.9995
> 99.9999
(UNDETECTABLE)
> 99.9999
(UNDETECTABLE)
99.999
> 99.9999
(UNDETECTABLE)
99.9999
HCI
CAPTURE, %
99.3
99.7
—
99.1
99
—
99
Ca|CI2 RATIO
2.2
2.4
—
2.2
1.7
.—
1.7
I
h-
C
Source: Reference 2.
-------
11.1.3 Cost of Treatment
The costs of circulating bed incinerators according to GA Technologies
Inc., are equivalent to the costs of conventional fluidized bed systems and
less than those for rotary bed incinerators. Additional cost savings will
also result from control of pollutants, such as those resulting from chlorine
and sulfur in the waste, through addition of dry limestone to the bed. As
shown in Table 11.1.3, chlorine capture efficiencies are reported to exceed
99 percent, a condition that meets EPA incinerator requirements. Presumably,
other EPA requirements for air emissions, such as those existing for
particulates and those being considered for toxics, can also be met and, thus,
the expense of pollution control equipment can be reduced if not eliminated.
11.1.4 Status of Technology
Circulating bed combustion systems are in operation worldwide, for many
process applications. There are no CBC incinerators operating specifically,
however, as hazardous waste incinerators (although, as the manufacturer points
out, many of the wastes disposed of by currently operating CBCs contain
hazardous constituents). A listing of the operating units, submitted by the
company, is shown in Table 11.1.4. GA Technologies is the only manufacturer
3
of CBC technology. In terms of market potential, the company provides the
general comparison between existing technologies and CBC shown in Table 11.1.5.
While the CBC concept and available performance data are promising,
additional data are needed to validate DREs and establish air emission levels
for particulates, PICs, chlorine based pollutants, and other possible toxics.
As noted in Reference 4, plug-flow reactor behavior, if it occurs, could lead
to incomplete combustion and high emission levels of contaminants in the feed.
11.2 MOLTEN GLASS INCINERATION
Molten glass incinerators (MGI) are electric furnace reactors in which a
pool of molten glass is used both as a means of destroying hazardous organic
wastes and/or as a means for encapsulating the solid byproducts of hazardous
waste treatment. The system utilizes furnaces similar to those used
11-11
-------
TABLE 11.1.4. CIRCULATING BED COMBUSTION UNITS
Customer
USA - GA
QA Technologies Inc.
San Diego, CA
USA - Pyropower
Gulf Oil Exploration
Bakersfield, CA
California Portland
Cement Co.
Colton, CA
B. E Goodrich
Henry, IL
Central Soya
Chattanooga, TN
General Motors Corp.
Pontiac, Ml
Colorado Ute Electric
Utility
Startup
1982
operating
1983
operating
1984
under
construction
1985
1985
1986
1987
Fuel
Varied
Coal, coke,
and limestone
Coal and
limestone
Coal and
limestone
Coal and
limestone
Coal, limestone,
and plant wastes
Coal and
limestone
Output
(MMBtu/hr)
2 MW (t)
50
209
123
105
370
1000
Application
Pilot plant
Enhanced oil
recovery
Cogeneration
Process steam
Process steam
Cogeneration
Electrical
generation
Nucla, CO
Foreign - Ahlstrom
Hans Ahlstrom
Laboratory
Karhula, Finland
Pihlava Board Mill
Rnland
Suonenjoki, Finland
Kemira Oy, Finland
Kauttua, Finland
1976
operating
1979
operating
1979
operating
1980
operating
1981
operating
Varied
Peat, wood,
and coal
Peat, wood,
and coal
Zinciferous
sludge
Peat, wood,
and coal
2^
50
22
—
22!
Pilot plant
Cogeneration
District heating
Sludge incineration
Cogeneration
(continued)
11-12
-------
TABLE 11.1.4 (continued)
Customer
Foreign (cont'd)
Hyvinkaa, Finland
Skelleftea, Sweden
Ruzomberok,
Czechoslovakia
Hylte Bruk, Sweden
Koskenkorva Distillery
Finland
Kemira Chemical
Finland
Zellstoff und
Papierfabrik
Frantschaoh AG
Carinthia, Austria
Ahlstrom
Varkaus, Finland
Neste Lampo Oy
Mantsala, Finland
Bord Na Mona
Ballyforan, Ireland
Oriental Chemical Co.
Inchon, Korea
Ostersunds
Fjarrvarme AB
Ostersund, Sweden
Papyrus AB
Kopporfors, Sweden
Metsaliiton Teollisuus
Oy
Aanekoski, Finland
Kereva Power Company
Kereva, Finland
Startup
1981
operating
1981
operating
1982
operating
1982
operating
1982
operating
1982
operating
1983
operating
1983
operating
1983
operating
1984
1984
1985
1985
1985
1985
Fuel
Coal, peat, oil,
and municipal
wastes
Peat, wood,
and coal
Sewage sludge
Peat and coal
Peat and oil
Peat and oil
Bark, brown coal,
and sludge
Wood waste
Coal-water
mixture
and coal
Peat and oil
Petroleum, coke,
and coal
Peat, wood chips,
and coal
Bark, peat, and
coal
Wood waste,
peat,
coal, and oil
Coal and
limestone
Output
(MMBtu/hr)
85
22
—
157
63
173
188
68
10
61
330
85
190
258
102
Application
District heating
District heating
Sludge incineration
Cogeneration
Process steam
Cogeneration
Cogeneration
Cogeneration-
retrofit
Heating-
firetube design
Cogeneration
Cogeneration
Heating
Cogeneration
Retrofit
Utility-heating
Source: Reference 2.
11-13
-------
TABLE 11.1.5. CIRCULATING BED INCINERATOR VS. CONVENTIONAL INCINERATORS
Item
Coat
Capital
Operating
Circulating Bed
$
$
Bubbling Bed
$ + scrubber
+ extra feeders
+ foundations
$ + more feeder
maintenance
+ more limestone
+ scrubber
Rotary Kiln
$$ (double)
+ scrubber
+ afterburner
$$ + more auxiliary
fuel
+ kiln maintenance
+ scrubber
Pollution control
POHCs
C1.S.P
NOX,CO
Upset Response
Effluent
In minimum-temperature
combustor
Dry limestone in combustor
Low due to turbulence,
staged combustion
Slump bed; no pollution
Dry ash
In high-temperature
combustor or afterburner
Downstream scrubber
High: bubbles bypass and
poor fuel distribution
Bypass scrubber pollution
released
Wet ash sludge
In afterburner
Downstream scrubber
High NO,: hot afterburner
Bypass scrubber pollution
released
Wet ash sludge
Feeding
No. of Inlets
Sludge Feeding
Solids Feedsize
Fly-Ash Recycle
1-solid
1-liquid
Direct
Less than 1 in.
Inherent (50 to 100 X
feedrate)
5-solids
5-liquids
Filter/atomizer (5 each)
Less than lk to % in.
Difficult mechanical/
pressure (10 X feedrate
max)
1-solids
2-liquids
Filter/atomizer (2 each)
Larger, but shredded
Not done
Unit size
Land area
Efficiency
Thermal, %
Carbon, %
Feeder, hp
Smaller
>78
>98
Minimum
Larger (over 2X)
<75
<90
High
Larger (over 4X)
<70
High
SI Convenlon: mm - in. X 25.4
11-14
-------
extensively in the glass manufacturing industry. Combustible hazardous wastes
of virtually any physical form or chemical composition may be destroyed
effectively in MGI systems. The system is considered particularly attractive
for the destruction of highly toxic organic wastes, wastes containing heavy
metals, and contaminated soils. Solids introduced with the waste feed and
many solid products of combustion become incorporated in a glass matrix,
rendering them essentially environmentally inert and land disposable. Molten
glass systems, are being studied by two separate firms (Battelle Northwest and
Penberthy Electromelt International) as hazardous waste treatment devices.
The process is considered to have certain technological and economic
advantages over other established incineration technologies.
11.2.1 Process Description
The molten glass furnace is a tunnel-shaped reactor, lined with
refractory brick, in which a pool of glass is maintained in a molten state by
electric current passing through the glass between submerged electrodes. Such
furnaces are used extensively in the glass manufacturing industry. The unit
is designed to withstand temperatures as high as 1260°C (2300°F), and
corrosion by acidic gases. NGI systems, as designed, will be equipped with
heat recovery and air pollution control systems, and can be combined with a
preconditioning heater or primary incineration unit, as depicted in
Figure 11.2.1.
In the absence of a primary incineration unit, wastes can be fed directly
to the furnace chamber, above the pool of molten glass. Solids, slurries, and
highly viscous liquids are usually charged via a screw feeder. Liquids may
also be sprayed into the chamber through nozzles located at the top of the
unit. Combustion air is fed to the system from two locations, one near the
top (as shown in Figure 11.2.1), and the other nearer to the surface of the
pool on the opposite side, in order to maximize the turbulence within the
reaction space. The temperature within the chamber is maintained at 2300°F.
.Residence time of gases within the chamber is about 2 seconds although this
can be increased if desired by reducing load. Residence time of solids within
the glass will be appreciably longer, and is measured in terms of hours.
Several furnace sizes accommodating various waste feed rates are available.
11-15
-------
CONTAMINATED /•
DMT ''
^-— AIR NOZZLES •
\ ft S \ X \ \..\...\'..\ \\ A A \"TT
DUST i ONTO TGUSSI
ROTARY KILN
FOB
VAPORIZATION AND
PRIMARY COMBUSTION
GLASS FURNACE
FOR
COMPLETE DESTRUCTION
OF ORGANIC VAPORS
„., ,~, BRINK FILTER.
ASH .P^HHEAT EXCHANGER OFF-GAS DENSE FIBERGLASS,
•w" «.---» COOLING AND FOISHABLE
SCRUBBING
MAWN
CMECOO
BfSWEH
AlflOVH)
rr
R.'CHOb.WS
LPiMbcrit'
LP
DATC
re-s-a*
SE 7D1C84
PENBERTHY ILECTROMELT
INTERNATIONAL, INC.
OI SOUTH MthSTKCET SEATTLE. WASHINGTON KM
DffWeNO.:
8181 PLUS 8180
Figure 11.2.1. Dirt purifier and hazardous waste Incinerator.
Source: Reference 7.
-------
During operation, volatile waste materials mix with air, ignite, and
react in the space above, and at the surface of, the pool of molten glass.
The solid products of combustion, dirt, and other noncombustible materials
(e.g., heavy metal contaminants or the solid waste being treated) will be
incorporated into the glass bed. Gaseous products flow out of the chamber,
through a series of ceramic fiber filters, which catch most of the particulate
matter. The hot gases, consisting primarily of C02» water vapor, and HC1
(if chlorinated organics are incinerated) then pass through a heat exchanger
for heat recovery (heat is used to warm the combustion air, as shown in
Figure 11.2.1). The exhaust gases flow next to a series of water spray—type
scrubbers. The first spray chamber is designed to use a slightly alkaline
scrubbing liquor, to capture acidic vapors. Water is used in the other spray
chamber (or chambers), to remove remaining particulates and other scrubbable
vapors. The gases are then reheated above the dewpoint, and passed through
charcoal and HEPA filters before being vented out the stack. The entire
system is maintained under negative pressure by means of the exhaust blower.
After a period of usage, the molten glass bed, with the solid waste
materials incorporated, is tapped out of the chamber into metal canisters,
and, after cooling, is sent to a disposal facility. The ceramic filters,
which eventually become loaded with particulate matter, can be disposed of by
dissolving them in the molten glass bed. The glass bed can also be used to
encapsulate the sludge from the spray chambers, and the spent charcoal and
HEPA filters.
Major advantages of the molten glass incineration system are its
applicability to many forms of hazardous waste and the encapsulation of
residuals in a nonleachable glass matrix. Performance testing, and data
generated from commercial usage of MGI units in the chemical processing
industry, while limited, have shown no significant difference in the effective
operation of such systems for wastes of different physical forms and widely
varying chemical composition. However, preheating and chemical treatment of
wastes are often used to aid combustion and reduce system maintenance and down
time. The waste related factors which may be of the greatest particular
concern are moisture content and metals and inorganics content. The
significance of these characteristics are discussed in detail below.
11-17
-------
11.2.1.1 Moisture Content—
A high concentration of water in waste will necessitate additional energy
input to the system and may affect destruction efficiencies. Penberthy has
set a moisture content limit of 20 percent (by weight) for its systems.
Since many solvent wastes contain water at levels higher than this,
pretreatment of the waste will be needed. Pretreatment systems which can be
used include evaporation and sedimentation. Dewatering options may be
somewhat limited for certain solvent wastes, due to characteristics such as
volatility and miscibility with water.
11.2.1.2 Metal Content—
Metals and minerals which are constituents of wastes pose a problem to
the effective operation of molten glass incinerators. Those materials which
are denser than the molten glass will tend to accumulate near the bottom of
the furnace. (Battelle reports that its process, which involves intermixing
of molten glass and waste, achieves 95 percent retention of nonvolatile heavy
metals.) Eventually, due to their electrolytic properties, they may affect
the operation of the metal electrodes. Penberthy has recommended the usage of
sumps to collect and localize settling particles of metal. Such systems have
been found to be effective in reducing the effect of metals on furnace
operation.
11.2.2 Demonstrated Performance
No data, demonstrating DREs or quantifying exhaust gas emissions, are
available for solvent wastes or any other wastes. These data are needed if
this technology, which appears promising in concept, is to be utilized for
hazardous waste treatment.
11.2.3 Cost of Treatment
Costs will depend to an appreciable extent upon the need for pretreatment
and the demands placed on the system used to clean exhaust gases.
11-18
-------
11.2.4 Status of Technology
Molten glass incinerators are available commercially from Penberthy
Electromelt International Inc. for use as chemical processing units. Battelle
Northwest, another company involved in the development of MGI systems, has not
yet produced equipment on a commercial scale. The itenberthy system has not
been sold or permitted specifically as a hazardous waste incinerator to date.
However, despite the lack of information concerning its application to
hazardous wastes the technology would appear to offer certain definite
advantages. Anticipated advantages are as follows:
* Able to achieve significant waste volume reduction
* Able to destroy almost all forms of hazardous waste, largely
independent of physical state or chemical composition
* Operation at high temperature, thus particularly attractive for
highly toxic organic streams and wastes containing long-chain
resinous organics
• Heat recovery and air pollution control built into system
• Solid byproducts transformed via glass encapsulation to
environmentally safe state. The encapsulates are resistant if not
inert to chemical reaction, leaching, and fracture. They probably
can be disposed of in landfills
* System is small in size, can be transportable
* Equipment used is relatively simple, representing basic technology
that has been applied in heavy industry for 30 years
Limitations, also largely conjectural at this stage, include the following:
* Unproven technology. There is no knowledge of long term operation
and maintenance requirements, or how performance would be affected
by long term usage with wastes
• Energy costs and capital costs are relatively high
• Control system as described may be inadequate for exhaust gases of
the type anticipated from hazardous waste destruction
11-19
-------
11.3 MOLTEN SALT DESTRUCTION
Molten salt incinerators involve the combustion of waste materials in a
bed of molten salt. Using the molten salt incineration process, "organic
wastes may be burned while, at the same time scrubbing in situ any
objectionable byproducts of that burning and thus preventing their emission in
Q
the effluent gas stream." Molten salt incinerators were developed by
Rockwell International, specifically to burn hazardous organic wastes and, as
designed, are applicable to both liquid and solid waste streams. However,
wastes with high ash content or a high percentage of water or noncombustible
material are not good candidates for molten salt destruction.
11,3.1 Process Description
The molten salt destruction process has been under development by
9
Rockwell International since 1969. The original intent was to use the
process to gasify coal. A variety of salts can be used, but the most recent
studies have used sodium carbonate (Na^CO,,) and potassium carbonate
(K2C03) in the 1,450°F to 2,200°F (790°C to 1200°C) temperature range.
In addition to the Rockwell process, another molten salt process is under
development. The State of New Jersey in late 1982 issued a contract to the
Questex Corporation of New York to evaluate a mobile, offsite earth
decontaminator (MOSED), a waste treatment unit based on the molten salt
destruction principle. A status report on the development of this device was
presented at the 1985 Hazpro Conference.
As shown in a schematic of the Rockwell process (Figure 11.3.1), the
waste is fed to the bottom of a vessel containing the liquid salt along with
air or oxygen-enriched air. The molten salt is maintained at temperatures of
800-1,000°C (1,500 to 1,850°F).11 The high rate of heat transfer to the
waste causes rapid destruction. Hydrocarbons are oxidized to carbon dioxide
and water. Constituents of the feed such as phosphorous, sulfur, arsenic, and
the halogens react with the salt (i.e., sodium carbonate) to form inorganic
salts, which are retained in the melt. The operating temperatures are low
1 9
enough to prevent NO emissions. *
X
11-20
-------
LIQUID WASTE FEED
COMBUSTION AIR
SOLID WASTE FEED
SALT QUENCHING CHAMBER
EXHAUSTSTACK
AND/OR GAS
CLEANING EQUIPMENT
•''MOLTEN SALT DEMISTER
^SECONDARY REACTION ZONE
MOLTEN-SALT
LEVELCONTROL-
S-MOLTEN SALT
WASTE ENTRANCE
NOT DRAWN TO SCALE
TO SALT RECOVERY
Figure 11.3.1. Molten Salt Combustion System.
Source: Reference 8.
11-21
-------
Eventually, the build-up of inorganic salts must be removed from the
molten bed to maintain its ability to absorb acidic gases. Additionally, ash
introduced by the waste must be removed to maintain the fluidity of the bed.
Ash concentrations in the melt must be below 20 percent to preserve fluidity.
Melt removal can be performed continuously or in a batch mode.
Continuous removal is generally used if the ash feed rates are high. The melt
can be quenched in water and the ash can be separated by filtration while the
salt remains in solution. The salt can then be recovered and recycled. Salt
losses, necessary recycle rates, and recycling process design are strongly
1 9
dependent on the waste feed characteristics. *
11.3.1.1 Waste Characteristics and Pretreatment Requirements—
Molten salt destruction (MSD) systems are limited in their applicability
to various hazardous wastes. Although the system is capable of handling
hazardous wastes in both the liquid and solid state, MSD is in practice
limited to the incineration of hazardous organic wastes which have a
relatively low percentage of solids or inorganics. Slurried wastes and most
"dry" solid wastes (e.g., contaminated soils) are not good candidates for
incineration by MSD* When ash accumulates in the bed, it tends to form a
waste matrix, which eventually affects bed fluidity, the overall transfer of
heat and will eventually limit waste byproduct neutralization within the
molten mass. Thus, 20 percent was determined to be the limit to which the
system could effectively operate.
Wastes with high water content may pose a problem to the effectiveness of
the molten salt destruction process. As moisture content increases, the waste
will require more fuel and combustion air, to the point where the reactor
volume is limited. Thus, many wastes must be dewatered by pretreatment to
ensure that they are effectively destroyed in the MSD.
Discussion with Rockwell indicated that there is no established
pretreatment system designed as part of the MSD system. However, separation
technology for removal of solids and to dewater wastes prior to incineration
in a MSD unit must be considered.
11-22
-------
11.3.1.2 Operating Parameters—
The operating parameters for a molten salt unit are:
• Temperature Range: 800-1000°C (1500-1850°F)
• Residence Time:
Gas Phase approx. 5 seconds
Liquid or Solid Phase hours
• Energy Requirement: Natural gas or oil to heat salt bed;
Auxiliary fuel for noncombustible wastes;
Power for exhaust
• Available Capacity: commercial units available at 20001bs/hr;
Pilot scale in use operating at 2501bs/hr.
• Operating Limitations: Heat generation. MSD requires a cooling
system for the overall unit to prevent
operational failures
11.3.1.3 Post-Treatment Requirements—
Although post-treatment requirements have not yet been defined, it is
likely that treatment will be required to remove products of combustion that
are not scrubbed out of the exhaust gases by the molten salt. These products
of combustion could include particulates, POHCs and PICs. Solid residues
(i.e., used salt) must be reprocessed or disposed.
11.3.2 Demonstrated Performance
Rockwell International has built two bench scale combustors (0.5 to 2
Ib/hr), a pilot plant (55 to 220 Ib/hr), and a portable unit (500 Ib/hr)
(Edwards, 1983). They have also built a 200 Ib/hr coal gasifier based on the
molten salt process. Destruction efficiency tests have been conducted at the
bench and pilot scale levels. While no data were found to demonstrate the DRE
of the molten salt destruction technology for the solvents of concern, data
showing five nines to eleven nines DRE for certain organic compounds have been
obtained.
Many wastes have been tested in the bench scale unit. Chemical warfare
agents GB, Mustard HD, and VX have been destroyed at efficiencies ranging from
99.999988 to 99.9999995 percent. Other chemicals that have been destroyed
11-23
-------
using the molten salt combustion process include; chlordane, malathion, Sevin,
DDT, 2,4-D herbicide, tar, chloroform, perehloroethylene distillation bottoms,
9
tnchloroethane, tributyl phosphate, and PCBs.
The PCB trial combustion data are presented in Table 11.3.1. The
destruction efficiency at the lowest operating temperature 700°C (1,300°F)
exceeded 99.99995 percent. The average residence time of the PCB in the
melted salt was 0.25 to 0.50 seconds, based on gas velocities of 1 to 2 ft/sec
9
through the 0.5 ft of melt.
Hexachlorobenzene (HCB) and chlordane destruction were tested in the
12
pilot plant facility. Feed rates for HCB and chlordane were as high as
269 Ib/hr and 72 Ib/hr, respectively. Bed temperatures ranged from 1,685° to
1,805°F (920°C to 985°C) and residence times were in the 2 to 3 second range.
HCB destruction efficiencies ranged from 99.9999999 to 99.999999999, and
chlordane destruction efficiencies ranged from 99.99999 to 99.999999). The
results of the pilot-scale tests are summarized in Table 11.3.2.
As shown in Table 11.3.2, very high DREs were noted for both compounds.
UC1 emissions were below 100 ppm, and no Cl_ gas or phosgene gas was
detected. Particulate emissions were measured, but were found to be quite
low, and analysis showed that particulate matter was nonhazardous. The
improved performance in the pilot scale reactor was attributed to greater
residence times.
11.3.3 Costs of Treatment
Detailed estimates of costs for molten salt destruction have not been
formulated. Based on the performance of the bench- and pilot-scale MSB units,
it is speculated that general operating costs will be low, but that the
initial capital costs will be high. Molten salt destruction operating costs
should be lower than, established technologies such as rotary kilns. Operating
temperature are low and the system needs not have a complex air pollution
control system and associated appurtenances, (although emission data are
needed to verify this), or ash recovery and transport systems.
11-24
-------
TABLE 11.3.1. PCB COMBUSTION TESTS IN SODIUM-POTASSIUM-CHLORIDE-CARBONATE
MELTS [Edwards, 1983]
Temp
Stochiometric
air
Concentr at ion
of KC1, NaCl
in melt
(wt %)
Extent of PCB
destruction3
Concentrat ion
of PCB in
off-gasa
(yg/m3)
870
830
700
895
775
775
145
115
160
180
125
90
60
74
97
100
100
100
>99. 99995
>99. 99995
>99. 99995
>99. 99993
>99. 99996
>99. 99996
<52
<65
<51
<59
<44
<66
aPCBs were not detected in the off-gas, i.e., values shown are detection
limits.
Reference: Reference 9.
11-25
-------
TABLE 11.3.2. SUMMARY OF PILOT-SCALE TEST RESULTS
PCB
Chlordane
Combustor Feed Rate 20.9 - 122.0
(Ib/hr)
Combustor Off-gas
- mg/m3
— ppmv
Baghouse
- mg/m3
- ppmv
Spent Melt (ppmv)
NOX (ppmv)
HC (ppmv)
2.7 x 10~4 - 7.1 x 10~2
2.3 x 10-5 - 6.1 x 10~3
<6 x 10~6 - 1.6 x 10~4
<5.2 x 10~7 - 1.4 x 10~5
0.001 - 0.104
70 - 125
35 - 110
Particulate (mg/ra3) <6.2 x 10~3 - 0.107
DR1 (%)
11-9's - 9-9's
12.1 - 32.7
5.3 x ID"3 - 6.8 x ID'2
3.? x 10~4 - 4.1 x 10~3
<3.6 x 10~4 - <4.4 x 10~3
<2.1 x 10~5 - <2.6 x 10~4
0.0044 - 1.2
0.5 - 630
0.4 - 60
4.1 x 10~3 - 1.75 x 10-2
8-9's - 7-9's
Note: The pH of the liquid in. a small sampling scrubber in the off-gas line
remained basic throughout the test indicating essentially no HC1
emission.
Source: Reference 1.
11-26
-------
11.3.4 Status of Technology
Molten salt destruction systems are a proprietary design of the Rockwell
International Corporation. Rockwell began development of the MSD system in
1969, obtaining several patents for the technology. By 1980, the system was
made available for commercial-scale application, at a capacity of 2000 Ibs/hr.
for destruction of specific waste types. The company constructed, and
currently maintains three different sized units, including a bench-scale
(2 Ibs/hr) unit, and a pilot-scale (200 Ib/hr) unit, and a full scale
(2000 Ibs/hr) unit, for demonstration of molten salt incineration
capabilities. However, no commercial scale units have been sold by the
company to date. Rockwell has indicated that development of this technology
has been curtailed, due in part to the limited demand encountered. Rockwell
will maintain th€ir demonstration units and considers future development of
MSD a possibility.
As demonstrated in the molten salt destruction process performance tests,
MSD systems have certain distinct advantages as an incineration technology
alternative. The limitations of the system however, may prove to severely
limit its further development.
Advantages—
o Achievement of high destruction efficiencies for many wastes,
including highly toxic and highly halogenated wastes;
« Low NOX and heavy metal emissions
* Retention of halogens and metals in a manageable salt matrix;
* Compact size. The process has few moving parts; and acts as its
own, highly efficient scrubber for acid combustion gases;
* Especially well-suited to wastes whose combustion results in
liberation of acids;
• Improved reliability due to simple design;
• Increased waste throughput possible
11-27
-------
Limit at ions—
• Generally restricted to certain types of organic hazardous wastes;
* Sensitive to high ( 20%) ash content in wastes;
• Molten salt is corrosive to all but specific engineering alloys.
Material and construction costs will therefore be high, and
management of spent salt beds will be difficult;
* No commercial applications to date, thus, no existing record of
long—term performance and operation and maintenance requirements
11.4 PYROLYSIS PROCESSES
Pyrolysis reactors are systems in which destruction of waste contaminants
is accomplished by applying a large thermal input resulting in molecular
decomposition, often down to an elemental or simple molecular form. No
oxidation reactions are involved in these processes. Pyrolysis reactors can
achieve very high destruction efficiencies for wastes, including difficult to
dispose of wastes such as dioxin wastes. A variety of pyrolysis systems have
been developed, including continuous and batch furnace pyrolyzers, the plasma
arc reactor, and the high temperature fluid wall reactor. These are described
below.
11.4.1 Furnace Pyrolysis Systems
11.4.1.1 Process Description—
The pyrolysis system shown in Figure 11.4.1 consists of three major
components: a continuous rotary furnace, a rich fume secondary combustion
chamber, and a heat recovery unit. The furnace is similar to furnaces
employed to treat metals and other materials requiring controlled thermal
treatment. Waste is continuously fed to a rotating belt which passes through
an indirect-fired, oxygen-free pyrolytic chamber. The waste is heated to
between 540°C (1000°F) and 870°C (1600°F). Volatiles in the waste or
resulting from pyrolysis are driven off leaving behind inert materials,
metals, and other inorganics, which are continuously removed from the moving
belt. The volatile gases, containing organic compounds and products of
pyrolysis such as H2 and some HC1 (if chlorine is present in the waste) are
11-28
-------
Waste heat boiler \
Combustion
air biower \
Rotary hearth
furnace %«-
Refractory duct work,.
Rich fume
reactor and
dwell chamber
PYROTHERM™ System
Figure 11.4.1 Continuous Pyrolyzer
Source; Reference 13
11-29
-------
combusted in the rich fume reactor to complete the destruction of any organic
materials present, and then flow through a waste heat boiler or a similar
device used to recover energy. Although, some HC1 formed by pyrolysis is
removed through contact with alkaline components (either in the waste or added
to the feed deliberately for that purpose) it is conceivable that some type of
air pollution control device might be needed to control acid gas emissions.
Reportedly other pollutants such as particulates and nitrogen oxides will not
be a problem because of the low turbulence level within the pyrolysis chamber
and the reducing atmosphere of pyrolysis, respectively.
Although wastes with a wide range of chemical characteristics may be
treated in a pyrolytic incinerator, certain wastes are clearly better candi-
dates than others. As noted by Midland-Ross, the developer, pyrolysis
systems work best for wastes which fall into the following categories:
"1. Too viscous to atomize in liquid incinerators, yet too fluid for
spreader—stoker incinerators.
2. Low melting point materials that foul heat exchangers, spall
refractories, and complicate residue discharge.
3. High residue materials (ash), with easily entrained solids, that
would generally require substantial stack gas cleanup.
4. Material containing priority pollutants with excessive vapor
pressure at incineration temperatures.
5. Any material, drummed or loose bulk, where controlled thermal
treatment is desired to make clean gases for heat recovery or for
discharge to the atmosphere."
Operating conditions for the components of the two types of pyrolyzers (batch
and continuous) produced by Midland-Ross are as follows.
Pyrolyzer
* Temperature Range: 650°-870°C
(1200°~loOO°F)
• Residence Time Range: 15-30 minutes (continuous systems)
4-6 hours (batch systems)
* Auxiliary Fuel Requirements: Natural gas, fuel oils, and/or
electrically-fired
11-30
-------
Rich Fume Incinerator (Reactor)
* Temperature Range: 980°-1200°e+
(1800°-2200°F+)
• Residence Time Range: 1.0-2.0 seconds
Commercial System
* PyroBatch Systems 1,000 Ib/load to 30,000 Ib/load
* lyroTherm Systems 500 Ib/h up
11.4.1.2 Demonstrated Performance—
A Midland-Ross batch pyrolysis system, operated by the McDonnell Douglas
Company in St. Charles, Missouri, was RCRA permitted in 1984 after a series of
trial burns using wastes with five POHCs with 50-70 percent chlorinated
hydrocarbons. Average DREs for the five POHCs during the trial burns were
99.9999 percent (six nines). Removal efficiency for HC1 was 99.9 percent, and
particulate emissions were 0.035 gj
other data appear to be available.
particulate emissions were 0.035 grains per dry standard cubic foot. No
11.4.1.3 Cost of Treatment—
As noted in Reference 1, the developer states the following with regard
to cost.
"Our cost estimates are proprietary information and are supplied
only to customers with whom we have projects. To date, most of our
clients' wastes applications are different from one another, hence
project capital costs are also different. However, inherent
benefits of pyrolytic incineration help our clients realize
significant overall project cost reductions relative to direct
incineration systems."
11.4.1.4 Status of Technology—
As noted, both batch and continuous pyrolysis are supplied commercially
by the Midland—Ross Corporation. The company also maintains a research
facility and offers complete bench and pilot test facilities.
11-31
-------
The pyrolysis systems are particularly suited for sludges and solid
wastes because of the long residence times that can be employed to assist
14
destruction. In addition to the potential to destroy all organics and to
handle difficult waste types, pyrolysis systems, as noted by the developer,
offer the following advantages.
1. Salts and metals (inert materials) with moderate melting points are
not liquified because the pyrolyzer operates at a design temperature
below the melting points of most salts and metals.
2. Since the same salts and metals are normally not vaporized,
refractory spalling, surface fouling, and formation of inert aerosol
condensates are all greatly reduced.
3. Particulate emissions with most types of pyrolyzers are greatly
reduced because the waste is not agitated or contacted with
turbulent gases during pyrolysis, so particulate cleanup devices in
many cases are not needed to meet Federal standards.
4. Waste-borne NOX is reduced in a pyrolysis atmosphere to 82 and
H£0. Hence, NOX emissions from the process are considerably
lower.
5. Chlorinated or halogenated materials (e.g., hydrochloric acid)
typically liberated by thermal treatment of a waste- can be adsorbed
by caustics present in, or added to, the feed prior to pyrolysis.
This often leads to a reduction in emissions of HC1 and SO^ from
50 to 90 percent.
6. Leaching of metals and salts from the carbonaceous residue (char) is
reduced because they are exposed to a reducing atmosphere throughout
the process, and they tend to be physically or chemically tied up in
the char.
7. Overall, gas cleanup equipment is greatly reduced or not required to
pyrolytically treat the same waste materials treated by direct
incineration.
11.4.2 Plasma Arc Pyrolysis
11.4.2.1 Process Description—-
In this process, under development by Pyrolysis Systems Inc. of Welland,
Ontario, waste molecules are destroyed by the action of a thermal plasma
field. The field is generated by passing an electric charge through a low
pressure air stream thereby ionizing the gas molecules and generating
temperatures up to 10,000°C.
11-32
-------
A flow diagram of the plasma pyrolysis system is shown in Figure 11.4.2.
The plasma device is horizontally mounted in a refractory-lined pyrolysis
chamber with a length of approximately 2 meters and a diameter of 1 meter.
The colinear electrodes of the plasma device act as a plug-flow atomization
zone for the liquid waste feed, and the pyrolysis chamber serves as a mixing
zone where the atoms recombine to form hydrogen, carbon monoxide, hydrogen
chloride, and particulate carbon. The approximate residence times in the
atomization zone and the recombination zone are 500 microseconds and 1 second,
respectively. The temperature in the recombination zone is normally
maintained at 900-1,200°C (1650°F - 2190°F).16
After the pyrolysis chamber, the product gases are scrubbed with water
and caustic soda to remove hydrochloric acid and particulate matter. The
remaining gases, a high percentage of which are combustible, are drawn by an
induction fan to the flare stack where they are electrically ignited. In the
event of a power failure, the product gases are vectored through an activated
carbon filter to remove any undestroyed toxic material.
The treatment system that is currently being used for testing purposes is
rated at 4 kg/minute of waste feed or approximately 55 gal/hour. The product
3
gas production rates are 5-6 m /minute prior to flaring. To facilitate
testing, a flare containment chamber and 30 ft. stack have also been added to
the system. The gas flow rate at the stack exit is approximately
,, 3, . . 16
36 m /raxnute.
A major advantage of this system is that it can be moved from waste site
to waste site as desired. The entire treatment system, including a
laboratory, process control and monitoring equipment, and transformer and
switching equipment, are contained on a 45 ft. tractor-trailer bed.
Two residual streams are generated by this process. These are the
exhaust gases that are released up the stack as a flare, and the scrubber
water stream. Since the product gas (after scrubbing) is mainly hydrogen,
carbon monoxide, and nitrogen, it burns with a clean flame after being
ignited. Analysis of the flare exhaust gases, presented in the following
section, indicates virtually complete destruction of toxic constituents.
The scrubber water stream is composed mainly of salt water from
neutralization of HC1 and particulates, primarily carbon. Analyses of the
scrubber water for the waste constituent of concern (e.g., carbon
11-33
-------
off GASES TO RAH;
EMEBGBKT CARBON fUW
Figure 11.4.2. Pyroplasma process flow diagram.
Source: Reference 15.
j.i-34
-------
tetrachloride (CC1,) and PCB in the feed material) have shown that the
constituents were present at low ppb concentrations. The quality of scrubber
water generated would depend on the water feed rate and corresponding product
gaa and scrubber waste flowrates. During a test in which 2.5 kg/min of waste
containing 35 to 40 percent CC1, was fed, to the reactor, a scrubber water
15
effluent flowrate of 30 I/minute was generated.
The reactor as it is currently designed can only be used to treat liquid
waste streams with viscosities up to that of 30 to 40 weight motor oils.
Particulates are removed by a 200 mesh screen prior to being fed into the
reactor. Contaminated soils and viscous sludges cannot be treated.
11.4.2.2 Demonstrated Performance—
The plasma arc system has been tested using several liquid feed
materials, including carbon tetrachloride (CC1,), polychlorinated biphenyls
(PCBs), and methyl ethyl ketone (MEK).
Table 11.4.1 presents the results of three test burns conducted in
Kingston, Ontario using carbon tetrachloride in the feed material. The carbon
tetrachloride was fed to the reactor along with ethanol, methyl ethyl ketone,
and water at a rate of 1 kg of CC1,/minute* The duration of each of these
tests was 60 minutes, and stack gas flowrates and temperatures averaged 32.5
dry standard cubic meter/minute (dscm/min) and 793°C (1460°F), respectively.
As can be seen in the table, the destruction and removal efficiency (ORE) of
CCl, in each of the tests was high, exceeding six nines. In addition, the
concentration of HC1 in exhaust gases was less than the upper limit of 1.8
kg/hr required by RCRA guidelines. The only possible area of concern is that
the concentration of CC1, in the scrubber water is greater than 1 "ppb. As
far as PCBs are concerned, the destruction and removal efficiency in each of
the tests was greater than 6 nines, and in some cases reached 8 nines.
Similar or better results can be anticipated for most solvents of concern.
11.4.2.3 Costs of Treatment—
The approximate capital cost of a unit similar to the one tested would be
in the range of 1 to 1.5 million dollars. More accurate figures will be
available once a commercial unit has been built.
11-35
-------
TABLE 11.4.1. CARBON TETRACHLORIDE TEST RESULTS
Parameter
Chlorine Mass Loading (%)
Scrubber Effluent
CCl4(ppb)
mg/hr
Flare Exhaust
CC14 (ppb)
mg/hr
NOX
ppm(v/v)
Ibs/hr
CO
ppm(v/v)
Ibs/hr
HC1
mg/dscm
kg/hr
Destruction Removal Efficiency
Test 1
35
1.27
2.29
0.83
12.1
106
1.02
48
0.28
(1)
(1)
99.99998
Test 2
40
5.47
9.85
0.43
4.9
92
0.69
57
0.26
137.7
0.25
99.99998
Test 3
35
3.26
5.87
0.63
7.2
81
0.02
81
0.37
247.7
0.44
99.99998
Source: Reference 15.
11-36
-------
11.4.2.4 Status of Technology—
The construction and testing of the plasma arc system is jointly
sponsored by the New York State Department of Environmental Conservation
(NYDEC) and the U.S. EPA Hazardous Waste Engineering Research Laboratory
(HWERL). The project is comprised of four phases, which are:
Phase 1: Design and construction of the mobile plasma arc system by the
contractor, Pyrolysis Systems, Inc. (PSI).
Phase 2: Performance testing of the plasma arc system at the Kingston,
Ontario, Canada test site.
Phase 3: Installation of the plasma arc system and additional
performance testing at Love Canal, Niagra Falls, N.Y.
Phase 4: Demonstration testing, as designated by NYDEC.
Phase 1 took place in 1982 and Phase 2, the results of which have been
presented above, was completed in early 1986. Phase 3 will be initiated later
in 1986.
The plasma technology is being jointly marketed by Westinghouse Electric
Corporation Waste Technology Services Division and PSI. Once the system has
been properly tested, they plan to lease these units to companies or
organizations that require the system for waste clean up. The current system
is only designed to handle liquid wastes. Future plans by PSI and Westinghouse
include the design of units which could handle contaminated soil and other
solid wastes.
11.4.3 High Temperature Fluid Wall (HTFW) Destruction -
Advanced Electric Reactor
11.4.3.1 Process Description—
The HTFW factor was originally developed by Thagard Research of Costa
Mesa, California. However, the J.M. Huber Corp. of Borger, Texas has
developed proprietary modifications to this original design. The reactor,
called the Advanced Electric Reactor (AER), is shown in Figure 11.4.3. The
reactor is a thermal destruction device which employs radiant energy provided
by electrically heated carbon electrodes to heat a porous reactor core. The
11-37
-------
1 EXPANSION BELLOWS
2P/~ji/i/cB FFcriTt-iBm tr*i-i
"L/VrCri rccLJ I rirl\JLt\3r1
COOLING MANIFOLD — .
4. POWER ^--^"e
FEEDTHROUGH<^I ',
ASSEMBLY ^^
_^_, •— i
6, END PLATE, !
i
i
e/rf fr^Tonn/r ——
, LLL Lf 1 1 1 ULJL ™" ~"~ ""•" 1
10. RADIATION
f-trAT QW/FL D - - ... •!
11. HEAT SHIELD
INSULATOR
12. COOLING JACKET
"•••^
f"
*
"*»,
^*-"
1
-V.
-sf
1
?
4
ym
**-
^
'•«•
«^
i^
**f
^
• '#
!%••
1
<^
%»•
&'
•>
t»
•*r
>••
^
*•*
af
J
txiT
i
i
"A
*
j
-q
\
be
c
1
«j
"SJT
1
•IIHBH
3
jfi-
•—•i" •.
<9
"^c
«
1
't
^
n
•>
. .
1 -
y
i •
i^
4
, , i *
II
'
n •*«
13f^""
"S.
3. POWER CLAMP
5. RADIATION
DEFLECTOR
7. ELECTRODE
CONNECTOR
9. POROUS CORE
73. RADIOMETER PORT
BLANKET GAS INLET
(TYPICAL)
Figure 11.4.3. Mvanced Electric Reactor [Huber].
11-38
-------
heated core then radiates heat to the waste materials. The reactor core is
isolated from the waste by a blanket of gas formed by nitrogen flowing
radially through the porous core walls.
The only feed streams to the reactor are the waste material and the inert
nitrogen gas blanket. Therefore, the destruction is by pyrolysis rather than
oxidation. Because of the low gas flow rate and the absence of oxygen, long
gas phase residence times can be employed, and intensive downstream cleanup of
off gases can be achieved economically.
* Destruction via pyrolysis instead of oxidation significantly reduces the
concentrations of typical incineration products such as carbon dioxide and
oxides of nitrogen. The principal products formed during treatment of
halogenated solvents would be hydrogen, chlorine (if calcium oxide is added to
the reactor, calcium chloride is formed instead), hydrochloric acid, elemental
carbon, and free-flowing granular material.
A process flow diagram for the AER is shown in Figure 11.4.4. The waste,
if it is a solid, is released from an air tight feed bin through a metered
screw feeder into the top of the reactor. If it is a liquid, it is fed by an
atomizing nozzle into the top of the reactor. The waste then passes through
the reactor where pyrolysis occurs at temperatures of approximately 4500°F
(2480°C) in the presence of nitrogen gas. Downstream of the reactor, the
product gas and waste solids pass through two post-reactor treatment zones,
the first of which is an insulated vessel which provides additional high
temperature (2000°F or 1090°C) and residence time (5 seconds). The second
post-reactor treatment zone is water-cooled, and its primary purpose is to
cool the gas prior to downstream particulate cleanup.
Off gas cleaning equipment includes a cyclone to collect particles which
do not fall into the solids bin, a bag filter to remove fines, an aqueous
caustic scrubber for acid gas and free chlorine removal, and two banks of five
parallel activated carbon beds in series for removal of trace residual
organics and chlorine.
The stationary pilot scale reactor which has been used for testing
various wastes at Huber1s Borger, Texas facility consists of a porous graphite
tube, 1 foot in diameter and 12 feet high, enclosed in a hollow cylinder with
a. double wall cooling jacket. This pilot unit is capable of processing 5000
tons/yr of waste. Huber also has a 3 inch diameter mobile unit which has been
transported to hazardous waste sites for testing purposes.
11-39
-------
POST
REACTOR
ZONES
SOLID
WASTE
BIN
AIR TIGHT FEED BIN
MOUNTED ON A HOPPER
METERED
SCREW FEEDER
ELECTRIC
REACTOR
SAMPLE POINT I
/ FAN BAG FILTER
SAMPLE POINT 2
SLIDE VALVE
CYCLONE
Jl
SLIDE
VALVE
ACTIVATED
CARBON BEDS
MAKEUP WATER
AND NaOH
STACK
CAUSTIC
SCRUBBER
Figure 11.4.4.
High temperature fluid wall process configuration for the
destruction of carbon tetrachloride [Huber].
11-40
-------
The AER cannot currently handle two-phase materials (i.e., sludge); it
can only burn single-phase materials consisting of solids, or liquids, or
19 21
gases alone. * Generally, a solid feed must be free flowing,
nonagglomerating, and smaller than 100 mesh (less than 149 micrometers or
19
0.0059 inches). However, depending on the required destruction, solids
smaller than 10 mesh may be suitable. Soils should be dried and sized before
being fed into the reactor.
The Huber process is not cost competitive with standard thermal
destruction techniques (such as the rotary kiln) for materials with a high Btu
T 8 21
content. * It is cost-effective for wastes with a low Btu content (e.g.,
chlorinated solvents) because unlike standard thermal destruction techniques,
the Huber process does not require supplementary fuels to obtain the necessary
Btu content for incineration.
The operating parameters as described by References 19 and 21 are as
follows:
Residence Time
(100 mesh solids)
Gas Flow Rate
Gas Phase
Residence Time
(at 2500°F or 1370°C)
O.I seconds
500 scfra for 150 ton/day
5 seconds
11.4.3.2 Demonstrated Performance—
In 1983, Thagard conducted a series of tests on PCB-contaminated soils
22
using a 3-inch diameter research reactor. The results of these tests
showed an average DRE of 99.9997 percent. The destruction efficiency was
found to be independent of the feed rate in the 50 to 100 g/m range at
2343°C. Pyrolysis products other than carbon and hydrogen chloride were not
detected using a GC with electron capture detection. It was concluded that
the method for dispersing the feed into the reactor needed improvement.
Problems with slagging in the reactor occurred that were believed to be
related to the small diameter of the reactor and also to the design of the
fluid wall flow. After modifications, additional tests on a 6-inch prototype
reactor were conducted by Thagard using hexachlorobenzene dispersed on carbon
22
^articles; 99.99991 percent destruction efficiency was achieved,
11-41
-------
J. M. Huber Corporation purchased the patent rights and made further
21
improvements to the process. The J.M. Huber Corporation then began tests
in its stationary reactor system which has a diameter of 12 inches. Included
in this system are: an insulated post-reactor vessel, a water-jacketed
cooling vessel, a cyclone, a baghouse, a wet scrubber, and an activated carbon
bed. Several research burns have been conducted with this system. Results
and operating parameters for pertinent burns are summarized in Table 11.4.2.
A series of four trial PCB-burns were conducted during September 1983
using a synthesized mixture of Aroclor 1260 and locally available sand to
1 19
obtain a total concentration of 3000 ppm PCBs. * After treatment, the
sand had a PCS content ranging from 0.0001 to 0.0005 ppm (0.1 to 0.5 ppb).
The destruction and removal efficiency was measured to be 99.99960 to
99.99995 percent. Additional studies were conducted with the 12 inch diameter
reactor using soils contaminated with octachlorodibenzo-p-dioxin (OCDD) and
carbon tetrachloride. Seven nines DRE (99.99999 percent) were reportedly
achieved at feed rates up to 2500 Ibs/hr.
11.4.3.3 Cost of Treatment—
Operating costs will vary depending on the quantity of material to be
processed and the characteristics of the waste feed. Pretreatment may be
necessary for bulky wastes having a high moisture content. Typical energy
requirements for contaminated soils range from 800 to 1000 kwh/ton.
Cost estimates for processing contaminated soil at a site containing more
than 100,000 tons of waste material were approximately $365 to 4565/ton in
1985. The cost breakdown for this estimate was 12 percent for maintenance,
7 percent labor, 29 percent energy, 18 percent depreciation and 34 percent for
other costs (permitting, setup, post-treatment, etc.). * These costs
have recently been updated. The new costs are expected to be released in
1986.21
11,4.3.4 Status of Technology—
Huber maintains two fully equipped reactors at their pilot facility in
19
Borger, Texas (Schofield, et al., 1985). The smaller reactor, which is
equipped for mobile operation, has a 3-inch core diameter and a capacity of
0.5 lb/min.. The larger reactor is commercial scale with a 12-inch core
11-42
-------
TABLE 11.4.2. SUMMARY OF OPERATING PARAMETERS AND RESULTS
FOR HUBER AER RESEARCH/TRIAL BURNS
Condition
PCBs
(Sept. 1983)
CC14
(May 1984)
Dioxins
(Oct/Nov 1984)
Reactor Core
Temperature (°F)
Waste Feed
Rate (Ib/min)
Nitrogen Feed
Rate (scfm)
X-DRE
4100
15.5-15.8
147.2
99.99999
3746-4418
1.1-40.8
104.3-190.0
99.9999
3500-4000
0.4-0.6
6-10
99.999
Source: References 19 and 23.
11-43
-------
diameter and a capacity of 50 Ib/min. Both of these reactors are used
primarily for research purposes. In May 1984, the Huber reactor was certified
by the EPA under TSCA to burn PCBs wastes. Recently, the U.S. EPA and the
Texas Water Commission jointly issued J.M. Huber Corporation a RCRA permit
which authorizes the incineration of any non-nuclear RCRA. hazardous waste
(including dioxin-containing wastes) in the Huber Advanced Electric
96
Reactor. This was the first commercial permit issued under RCRA for
4
treating dioxin-containing wastes. The J.M. Huber Corporation intends to use
the permit for research and development of a full-scale transportable AER.
Huber does not intend to operate a hazardous waste disposal operation, but
rather to construct and market stationary and/or mobile units for use by
21
companies or organizations involved in hazardous waste destruction.
11.5 IN SITU VITRIFICATION
In situ vitrification (ISV) was originally developed by Battelle Pacific
Northwest Laboratories as a means of stabilizing in-place high level nuclear
waste. More recently, however, ISV has been studied as a means of destroying
soils contaminated with chlorinated organic wastes, including PCBs and dioxin
wastes, and heavy metals. The system was patented in 1983.
In situ vitrification converts contaminated soils, or sludges, into a
solid glassy matrix through melting by joule heating. As depicted in
Figure 11.5.1, the process begins when graphite electrodes are placed into the
ground in a square array. A conductive path is established by placing
graphite over the soil between the electrodes. Electrical current is passed
between the electrodes, creating high temperatures (1,700°C or 3,100°F) which
melt the soil, and pyrolyze the organic waste constituents. Gaseous effluents
which are produced are collected by a hood over the area and are exhausted to
off-gas treatment systems. When pyrolysis is complete, current is shut off
and the mass cools to form a glass like material. A picture of the system is
presented in Figure 11.5.2, showing the enclosed hood.
Battelle engineers have developed 30 kW, 500 kW, and 3750 kW size units.
The small unit produces up to a ton of vitrified mass per setting, the 500 kW
unit produces approximately 10 tons per setting, and the large unit produces
400 to 800 tons per setting.)
11-44
-------
POROUS
GLASS
$Bmf^smw>%£
GRAPHITE
AND FRIT
STARTER
MELTING
ZONE
VITRIFIED SOIL/WASTE
Figure 11.5.1. Operating sequence of in situ vitrification.
Source; Reference 27.
-------
~~^*Z^:Ji$#&$fc™'~~*'^ . 'illMir
Figure 11,5.2. Off-gas containment and electrode support hood.
Source! Reference 27.
11-46
-------
The cost estimates reported by PNL, and discussed below for TRU wastes
treated by the ISV process, account for charges associated with site
preparation, consumable supplies such as electrical power, and operational
28
costs such as labor and annual equipment charges. Specifically, for
variations in manpower levels, power source costs, and degree of heat loss, it
was determined that the costs for TRU waste vitrification ranges from lt»0 to
3
360 $/m to vitrify to a depth of 5 meters. These costs are a function of
many variables, but are most sensitive to variations in the amount of moisture
in the soil and the cost of electrical power in the vicinity of the process.
Figure 11.5.3, developed by PNL, illustrates the variation in total costs as a
function of both electrical power costs and the moisture content of TRU soil
experimentally treated. The vertical line represents the value beyond which
it is more cost effective to lease a portable generator.
Recently, PNL has assessed the cost implications for ISV treatment of
three additional waste categories; i.e., industrial sludges and hazardous
waste (PCB) contaminated soils at both high and low moisture contents.
Representatives at PNL indicated that for industrial sludges with moisture
contents of 55 to 75 percent (classified as a slurry), the total costs would
3
range from 70 to 130 4»/m . Additionally, treatment of high (greater than 25
percent) moisture content hazardous waste-PCB contaminated soil would cost
3 3
approximately 150 to 250 |/m versus costs of 128 to 230 $/m for low
(approximately 5 percent) moisture content PCB contaminated soil.
As these recent data and past TRU waste cost data suggest, the moisture
content of the contaminated material treated is particularly important in
influencing treatment costs; high moisture content increases both the energy
and length of time required to treat the contaminated material. Furthermore,
PNL representatives suggest that treatment costs are also influenced by the
degree of off-gas treatment required for a given contaminated material, i.e.,
ISV application to hazardous chemical wastes will likely not require as
sophisticated an off-gas treatment system as would TRU waste treatment.
PNL has recently assessed the treatment of and costs associated with
hazardous waste contaminated soils. Specifically, during the summer of 1985,
tests were conducted for the Electric Power Research Institute (EPRI) on PCB
11-47
-------
4OO
300
- 100
t>
1
1OO
Portable
Gen«rator
468
Bactrical Rates JC/kWh)
10
12
Figure 11.5.3. Cost of in situ vitrification for TRD wastes as functions
of electrical rates and soil moisture [Fitzpatrick, 1984],
11-48
-------
contaminated soil. While the draft report on these tests has been completed,
it has not been published and/or made available to date. However, an EPRI
project summary publication, dated March 1986, entitled "Proceedings: 1985
EPRI PCB Seminar" (EPRI CS/EA/EL 4480) has recently been made available to
EPRI members. Preliminary results suggest that a destruction/removal
efficiency (DRE) of six to nine nines was achieved from the off-gas treatment
system overall, and that a vitrification depth of 2 feet was achieved.
Additional information will soon be available to the public. PNL expects to
continue with research in the area of hazardous waste soils.
11-49
-------
REFERENCES
1. Freeman, H. M. Innovative Hazardous Waste Treatment. EPA-t>OQ/2-83-049.
U.S. Environmental Protection Agency, Hazardous Waste Engineering
Research Laboratory. Cincinnati, OH. APril 1985.
2. GA Technologies, Inc. Material received in correspondence with M.
Kravett, GCA Technology Division, Inc., San Diego, CA. April 1986.
3. Rickman, W. Telephone conversation with M. Kravett, GCA Technology
Division, Inc., GA Technologies, Inc., San Diego, CA. April 1986.
4. Chang, D. and N. Sorbo. Evaluation of a Pilot Scale Circulating Bed
Combustor with a Surrogate Hazardous Waste Mixture. In: Incineration
and Treatment of Hazardous Waste: proceedings of the llth annual
research symposium. EPA-600/9-85-028. U.S. EPA, HWERL, Cincinnati, OH.
September 1985.
5. Rickman, W., N. Holder and D. Young. Circulating Bed Incineration of
Hazardous Wastes. Chemical Engineering Progress. March 1985.
6. Hotaling, D. Telephone conversation with M. Kravett, GCA Technology
Division Inc., Penberthy Electromelt Corporation, Seattle, WA. February
1986.
7. Penberthy Electromelt Corporation. Material received in correspondence
with M. Kravett, GCA Technology Division, Inc., Penberthy Electromelt
Corporation, Seattle,.WA. February 1986.
8. U.S. EPA. Assessment of Incineration as a Treatment Method for Liquid
Organic Hazardous Wastes (Background Report Series). U.S. Environmental
Protection Agency, Office of Policy, Planning, and Evaluation,
Washington, D.C. March 1985.
9. Edwards, B. H., J. N. Paullin, K. C. Jordan. Emerging Technologies for
the Control of Hazardous Wastes. Noyes Data Corporation, Park Ridge, New
Jersey. 1983.
10. Leslie, R. H. Development of Mobile On-Site Earth Decontaminator, In:
Proceedings of the Hazpro '85 Conference, Baltimore, Maryland. May 1985.
11-50
-------
11. Kohl, A. Telephone conversation with M. Kravett, GCA Technology
Division, Inc., Rockwell International Corporation, Canoga Park, CA.
February 1986.
12. Johanson, J. G., S. J. Yosim, L. G. Kellog, and S. Sudar. Elimination of
Hazardous Waste by the Molten Salt Destruction Process. In:
Incineration and Treatment of Hazardous Waste, Proceedings of the Eighth
annual Research Symposium. EPA-600/9-83-003, April 1983.
13. Midland-Ross Corporation. Material received in correspondence with
M. Kravett, GCA Technology Division, Inc., Midland-Ross Corporation,
Surface Combustion Division, Idaho, OH. March 1986.
14. Daiga, V. Telephone conversation with M. Kravett, GCA Technology
Division, Inc., Midland-Ross Corporation, Surface Combustion Division,
Toledo, OH. March 1986.
15. Kolak, Nichlos P., Thomas G. Barton, C.C. Lee and Edward P. Peduto.
Trial Burns - Plasma Arc Technology. EPA Twelfth Annual Research
Symposium on Land Disposal, Remedial Action. Incinerator and Treatment
of Hazardous Waste. Cinncinnati, Ohio. April 21-23, 1986
16. Barton, Thomas G. Mobile Plasma Pyrolysis. Hazardous Waste. 1 (2)
pp. 237-247. 1984.
17. Haztech News. Plasma Arc Technology Used to Atomize Liquid Organics. 1
(5) pp. 33-34. 1986.
18. Boyd, J., H.D. Williams, and T.L. Stoddard. Destruction of Dioxin
Contamination By Advanced Electric Reactor. Preprinted Extended Abstract
of Paper Presented Before the Division of Environmental Chemistry,
American Chemical Society, 191st National Meeting, New York, New York:
Vol 26, No 1. April 13-18, 1986.
19. Schofield, William R., Oscar T. Scott, and John P. DeKany. Advanced
Waste Treatment Options: The Huber Advanced Electric Reactor and The
Rotary Kiln Incinerator. Presented HAZMAT Europa 1985 and HAZhAT
Philadelphia 1985.
20. GCA Technology Division, Inc. Draft Report: Identification of Remedial
Technologies. Prepared for U.S. EPA, Office of Waste Programs
Enforcement, under EPA Contract No. 68-01-6769, Work Assignment No.
84-120. GCA-TR-84-109-G(0). March 1985.
21. Boyd, James. J.M. Huber Corporation. Telephone Conversation with Lisa
Farrell, GCA. Technology Division, Inc. January 28, 1986; April 3, 1986;
May 1, 1986.
22. Horning, A.W., and H, Masters. Rockwell International, Newbury Park,
California. Destruction of PCB-Contaminated Soils With a
High-Temperature Fluid-Wall (HTFW) Reactor. Prepared for U.S. EPA,
Office of Research and Development, Municipal Environmental Research
Laboratory, Cincinnati, Ohio. EPA-600/D-84-072. 1984.
11-51
-------
23. Roy F. Weston, Inc. and York Research Consultants. Times Beach,
Missouri; Field Demonstration of the Destruction of Dioxin in
Contaminated Soil Using the J.M. Huber Corporation Advanced Electric
Reactor. February 11,1985.
24. Lee, Kenneth W., William R. Schofield, and D. Scott Lewis. Mobile
Reactor Destroys Toxic Wastes in "Space". Chemical Engineering.
April 2, 1984.
25. Freeman, Harry M. Hazardous Waste Destruction Processes. Environmental
Progress. Volume 2, Number 4. November 1983.
26. Hazardous Materials Intelligence Report (HMRI), First Commercial Dioxin
Incineration Permit Granted to J.M. Huber. January 24, 1986.
27. Buelt, J. L. and S. T. Freim. Demonstration of In-Situ Vitrification for
Volume Reduction of Zirconia/Lime Sludges. Battelle Northwest
Laboratories. April 1986.
28. Oma, K. H. et al. 1983. In-Situ Vitrification of Transuranic Wastes:
Systems Evaluation and Applications Assessment. PNL-4800, Pacific
Northwest Laboratory, Richland, Washington.
29. Fitzpatrick, V. F., et al. 1984. In Situ Vitrification - A Potential
Remedial Action Technique for Hazardous Wastes. Presented at the 5th
National Conference on Management of Uncontrolled Hazardous Waste Sites,
Washington, DC
30. Buelt, J. L. Battelle Memorial Institute, Pacific Northwest
Laboratories. Telephone conversation with Michael Jasinski. GCA
Technology Division, Inc. 1986.
31. Buelt, J. L. et al. An Innovative Electrical Technique for In-Place
Stabilization of Contaminated Soils. Presented at the American Institue
of Chemical Engineers 1984 Summer Meeting in Philadelphia, Pennsylvania.
Pacific Northwest Laboratory, Richland, Washington. 1984.
11-52
-------
SECTION 12.0
USE AS A FUEL
Numerous studies have been conducted to identify existing industrial
combustion processes which have the capability of destroying hazardous wastes
through use as a fuel. If hazardous wastes can be used, either in conjunction
with, or instead of, the primary fossil fuels currently used, this would
provide a dual cost benefit of both eliminating the cost of disposal and
lowering consumption of expensive, potentially unavailable sources of energy.
The general guidelines established for identifying suitable high
temperature industrial processes (HTIPs) are that they should be capable of
achieving levels of performance which are consistent with the requirement
established for hazardous waste incinerators; these requirements (as specified
in the Federal Register 1982, 47, 27516-35) are:
* at least 99.99% destruction and removal efficiency (DKE) for each
principal organic hazardous constituent (POHC) in the waste feed;
* at least 99% removal of hydrogen chloride from the exhaust gas if
hydrogen chloride stack emissions are greater than 4 Ib/h; and
• particulate emissions not exceeding 0.08 grains/dry standard cubic
foot (dscf), corrected to 7% oxygen in the stack gas.
PEDCo has identified over 100 HTIPs which are capable of operating at
temperatures exceeding 1200°F in the metallurgical, chemical, mineral, and
sewage sludge processing industries. Of that number , 15 separate
technologies are considered to be prime candidates for thermal destruction of
hazardous wastes, based on preliminary test programs. These technologies are
listed in Table 12.1.
Performance testing has demonstrated the ability of several HTIPs to
achieve effective levels of hazardous waste destruction. The studies have
also indicated, however, that the applicability of most HTIPs to hazardous
12-1
-------
TABLE 12.1. THERMAL TECHNOLOGIES CONSIDERED APPROPRIATE FOR
BURNING HAZARDOUS WASTE AS FUEL
Process or reactor
Tunnel Kiln
Oil Furnace
Reverberatory Furnace
Blast Furnace
Multiple Hearth Roaster
Suspension Roaster
Fluidized-Bed Roaster
Blast Furnace
Open Hearth Furnace
Long Rotary Kiln
Short Rotary Kiln/Preheater
Rotary Kiln
Melting Furnace
Fluidized-Bed Furnace
Multiple-Hearth Furnace
Industry
Brick
Carbon Black
Primary Copper
Primary Lead
Primary Zinc
Iron & Steel
Lime
Aggregate
Glass
Sewage Sludge
Exit and maximum
temperature
°C
260 -
870 -
1,300 -
700 -
200 -
930 -
950 -
1,100 -
1,200 -
680 -
1,160 -
370 -
620 -
760 -
480 -
1,180
1,400
1,400
1,200
980
1,010
1,740
1,870
1,800
1,900
1,840
1,150
1,480
870
980
Average
residence
time
(sees)
4.3
1.1
2.2
5.9
>10
>10
>10
1.1
2.0
8.3
7.6
3.6
4.1
1.4
0.5
Source: Reference 1.
12-2
-------
waste destruction is limited by many factors. Primarily, these limitations
include the negative impact of hazardous waste combustion byproducts on HTIP
equipment and their potential environmental impact. Not all HTIPs will be
adequately equipped with air pollution controls, other effluent (liquid and
solid waste) controls, and material transport systems needed for hazardous
waste combustion. The applicability of wastes exhibiting certain
characteristics to usage as a fuel is, therefore, limited in some cases.
These primary restrictive characteristics include high water, chlorine, and
•metal/inorganic content, and high liquid viscosity.
Of the various high temperature industrial processes studied, three
technologies have demonstrated particular promise in achieving adequate levels
of performance for a variety of wastes and have, therefore, received the most
attention to date. These include: industrial boilers; industrial rotary
kilns including aggregate (cement, asphalt) processing kilns and lime
processing kilns; and blast furnaces. These units are generally applicable
for the combustion of many of the organic solvents and other low molecular
weight organic compounds considered in this document. According to
Reference 2 approximately 380 million gallons/year are burned as supplemental
boiler fuels. The characteristics of many of the solvent hazardous wastes are
such that usage as a fuel is considered a technically feasible and
economically attractive management alternative. Those characteristics which
contribute to the use of solvent hazardous wastes as a fuel include their
significant energy content, as indicated in Table 12.2, the fact that many of
the solvents may be pumped and atomized in liquid injection burners, and their
ability to blend with a wide variety of fuels and other wastes. Most solvent
wastes, in fact, will sustain combustion without the use of auxiliary fuel.
Solvent hazardous wastes do exhibit, however, certain characteristics
which limit their application to specific high temperature industrial process
technologies. The high temperature industrial processes in which hazardous
wastes may be burned as a fuel are, in general, more limited in the types of
waste streams they can handle effectively than are hazardous waste
incinerators. Generally, they are not equipped with extensive air pollution
controls or ash recovery and handling systems. Other technical limitations to
burning hazardous wastes in HTIPs include:
12-3
-------
TABLE 12.2. REPRESENTATIVE HEATING VALUES OF VIRGIN AND SIENT SOLVENTS
classification
Aliphatics
Aronatics
Esters
M Ketones
i
Ethers
Halogenated
Alcohols
Range i
20,000
17,000
9,000
12,000
12,000
500
8,000
0% solids
- 22,000
- 19,000
- 15,000
- 15,000
- 15,000
- 8,000
- 16,000
0% solids
21,000
18,000
12,000
13,000
13,000
5,000
12,000
Typical 1
5% solids
19,950
17,100
11,400
12,350
12,350
4,750
11,400
leatlng value*
10% solids
18,900
16,200
10,800
11,700
11,700
4,500
10,800
» (Btu/lb)
15% solids
17,850
15,300
10,200
11,050
11,050
4,250
10,200
20% solids
16,800
14,400
9,600
10,400
10,400
4,000
9,600
*Assumes zero heating value of solid material in spent solvent; contaminants such as greases could
add to values shown.
Source: The Pace Company, Reference 2.
-------
Possibly more frequent shutdown for boiler cleaning due to fouling
of the boiler tubes;
High flue gas-exit temperatures needed to prevent condensation of
acidic components; and
Safety problems associated with low boiling point/ignitable solvents.
In general, the assessment of the potential for a specific waste to be
destroyed by using it as a fuel in a specific system is based on
identification of certain key characteristics. These characteristics may oe
classified as follows:
* Those which restrict the ability of a system to effectively
distribute the waste within the combustion zone;
* Those which limit the ignitability and continuous combustibility of
a waste;
* Those which promote the generation of gaseous emissions and/or
liquid and solid effluent streams which are difficult to manage; and
* Those which affect the overall quality of the product.
The primary characteristics to be identified in this regard are summarized
below:
Physical Form—The physical form of a waste dictates the manner in
which it may be input to the system, and the relative ease with
which it will burn.
Btu (Heat) Content—Wastes must exhibit high heats of combustion to
be considered as a fuel. A. common standard used to determine
whether a waste may sustain combustion adequately for this purpose
is 8,500 Btu/lb, thus a waste with a Btu value below 8,500 probably
cannot be used as a fuel without blending.
Chlorine Content—Chlorine presents a limitation to the process both
due to the general low combustibility of highly chlorinated
substances and due to the composition of byproducts of combustion.
Most HTIPs are not equipped with air pollution controls which can
adequately handle acid gases produced when chlorinated wastes burn,
nor can they withstand the corrosive attack of hydrochloric acid on
linings and internal surfaces. A chlorine content of 3 percent is
considered a maximum.
12-5
-------
• Metals/Ash/Organic Salt Content—Wastes which contain high levels of
solid, thermally inert materials are not generally good candidates
for usage as a fuel, due to their negative environmental impact
(particulate emissions), and possible impact on the quality of the
products. Solids in the fuel feed also tend to have a deleterious
effect upon HTIP equipment, for example, fouling of boiler tubes,
* Water Content—Water is a hindrance to effective combustion, and may
also affect product quality. High moisture content wastes,
therefore, are generally not good candidates for combustion as a
fuel.
* Flash Point—Safety considerations require that highly ignitable
components be removed prior to their storage and introduction into
combustion systems.
The specific characteristics of wastes which present restrictions to the
applicability of the technologies focused upon in this document will be
discussed in the next section.
12.1 PROCESS DESCRIPTIONS
12,1.1 Industrial Boilers
Boilers are used primarily to produce steam, for use in processing or
space heating. The population of industrial boilers in service may number
3
over 300,000 in the United States, and consists of many different types of
systems, available in sizes ranging from several thousand to several hundred
million Btu/hour, utilizing a wide variety of fuels, mechanisms for fuel feed,
and heat transfer systems.
Two types of boilers are considered appropriate for the burning of
hazardous wastes: water tube and fire tube boilers. These systems are
considered appropriate because combustion of wastes and production of
combustion byproducts are physically separate from the heat transport media,
thus preventing cross—contamination of the media and allowing for greater
control of the combustion process. Both water tube (WT) and fire tube (FT)
boilers employ a "shell-and-tube" arrangement. In a water tube boiler, wastes
are burned within a combustion zone through which tubes containing flowing
water (or steam) are passed. Hot combustion gases contact the outer (metal)
walls of the tubes, imparting thermal energy which is transferred to the water
12-6
-------
flowing inside. High pressure, high temperature steam is thus produced within
the tubes. WT boilers are most commonly found in the 10 mm Btu/hour or larger
4
range. They are capable of burning any physical form of fuel, and are
consequently applicable to a wide variety of waste types.
Fire tube boilers employ essentially the opposite configuration to water
tube boilers. Fuel combusting within the tubes produces thermal energy which
heats water through which the tubes are passed. FT boilers employ liquid
fuels only, and are generally available in smaller sizes. FT boilers may
be more subject to structural failures when large variations in operation or
fuel feed occurs.
Both water tube and fire tube boilers are usually capable of achieving
combustion temperatures and residence times which are sufficiently long to
achieve high levels of waste destruction. When used as a fuel in a boiler,
wastes are typically blended with conventional fuels to achieve a final
waste/fuel mixture which has a heat content above some specified.level.
Waste/fuel mixture ratios vary widely, from as low as 5 to 10 percent by
weight to 100 percent waste. The waste/fuel ratio selected depends upon
the suitability of the waste for use as a fuel, as determined by several
factors. These factors include the compatibility with fuel handling and
burner/boiler systems, physical form of -the waste, heat content, moisture
content, concentration of sulfur, halogens, metals, and other noncombustible
materials, and liquid viscosity. Many boilers burning wastes with fuels have
established certain baseline limitations relative to these factors. For
example, the maximum acceptable chlorine content for boilers is typically set
at 3 percent by weight. Typical pretreatment methods in addition to
blending to achieve heat content requirements are operations such as
sedimentation, screening and other separation techniques to reduce water and
sediment and to eliminate, if necessary light ends which represent a storage
and combustion hazard.
Blending of hazardous solvent wastes with conventional fuel oils can also
serve the major purpose of meeting regulatory emission requirements, thus
avoiding the need to install costly air pollution control devices. Boilers
firing conventional fuels rarely employ air pollution controls because
combustion efficiencies are high and the ash content (<0.1 percent) and
chlorine content (<;0.01 percent) of the fuels are low. Thus, uncontrolled
12-7
-------
particulate and hydrochloric acid emissions are low and generally below any
applicable regulatory limits. Although the regulatory requirement for
incinerators are stringent, blending with conventional fuels appears to be a
viable means of meeting these regulatory requirements for some wastes.
Asssuming that ash is the sole source of particulate emissions, the regulatory
requirement that particulate emissions can not exceed 0.08 grains/dry standard
cubic foot, corrected to 7 percent oxygen in the flue gas, can be met by a
fuel containing about 0.28 percent ash. This estimate assumes that ash is the
sole source of particulate emission, that all ash is emitted, the Btu content
of the fuel is 19,000 Btu/lb, and that the chemical composition of the fuel is
similar to that of a number 6 fuel. This level of particulate emissions is
equivalent to an emission factor of 67 nanograms per Joule, a value somewhat
higher than EPA emission factors of 37 and 6 nanograms per Joule for
uncontrolled boilers burning residual and distillate oil, respectively. While
the above estimate is only approximate, it shows that blending may be a viable
option for some wastes. For example, a 10 percent blend of a one percent ash
content waste with distillate oil would probably meet particulate emission
requirements, and even higher blending levels may be possible.
Similary, most nonhalogenated wastes could meet the incinerator chlorine
emission level of 4 Ib/h. For example, a 10,000,000 Btu/h boiler could burn a
0.4 percent chlorine content, 10,000 Btu/lb fuel and meet the regulatory
requirement. However, a one percent blend of 1,1,1 Trichloroethane (100
percent) with a 19,000 Btu/lb fuel oil would exceed the 0.41b/h emission
level. Halogenated solvents, because of both low Btu content and high
chlorine levels, are not suitable for use as boiler fuels. However, they can
be destroyed in controlled incinerators and can, as noted below, be burned in
certain industrial processes where the HC1 emissions are adsorbed and/or
provide a. process benefit.
12.1.2 Industrial Kilns
There are several industrial processes which employ rotary kiln units
that are similar to units whose hazardous waste destruction capabilities have
been studied in depth. These industrial kilns include those used in the
aggregate processing (cement and asphalt) and lime industries. The interest
12-8
-------
in these systems is based upon two major factors. First, they operate at
temperature ranges and residence times required for effective destruction of
hazardous waste constituents. Second, they provide a large and widely
distributed source of capacity; thus reducing transportation costs and
materials handling considerations.
Another possible advantage to be gained from the use of wastes cofired as
fuels in industrial kilns relates to the benefits to be realized from the
firing of halogenated compounds. As noted in Section 4, cement kiln can use
2
up to 1.25 percent of their design feed as chlorine waste. Blast furnaces
are also reportedly burning chlorine containing wastes to prevent build up of
alkalis on furnace walls.
A brief overview of prime candidate industrial kiln systems is presented
below.
Lime Kilns—
Lime production processes use high temperature kilns to produce CaO
("quicklime") from limestone (CaCO,,). A number of different thermal process
designs are used in lime production, with rotary kilns accounting for
approximately 90 percent of production capacity. According to the
U.S. Bureau of Mines, 153 lime plants were in operation in 1980.
The process involved in lime production is relatively simple. Limestone
is fed to the elevated end of the kiln and is discharged as quicklime at the
lower end. Combustion air flows upward, countercurrent to the lime, and is
exhausted to air pollution controls (dust collectors such as ESPs or
baghouses) at the feed end of the kiln. Most kilns, according to PEDCo, are
fired with pulverized coal, but also can fire natural gas and oil.
Adaptability to both solid and liquid hazardous wastes, therefore, is
excellent.
The potential effectiveness of the lime kiln in destroying hazardous
wastes used as fuel is considered high. This is primarily due to the high
temperatures (~2300°F) of operation. However, waste composition is important
since impurities introduced into the feed such as alumina, silica, and iron in
the waste may lead to formation of undesirable inorganic salts and affect the
utility of the product. For example, chlorine in the waste may react to form
calcium chloride, the formation of which may lead to not only fouling of the
12-9
-------
product, but also to clogging of the kiln when temperatures go below 1400°F.
HC1 formation may also lead to corrosion in ducts and in the air pollution
control equipment (baghouses, for example, may not be able to handle corrosive
gases).
Aggregate Kilns
Aggregate kilns are used to produce mixtures such as cement and asphalt.
Typical raw materials used to produce cement or asphalt include:
* Natural lightweight aggregates - prepared by crushing and sizing
pumice, volcanic cinders, tuff, scoria, and breccia;
» Byproduct lightweight aggregates - prepared by crushing and sizing
foamed and granulated slag (e.g., from blast furnace operations in
the iron and steel industry), coke breeze, and cinders; and
* Manufactured structural lightweight aggregates - prepared Dy
pyro-processing clay, shale, or slate.i
The process involved in aggregate processing is very similar to that used in
lime manufacturing. Dry raw materials are reacted within a kiln whose
o
operating temperature falls between 2050 and 2300°F . Typical residence
times are 2 to 4 seconds. Excess air usage is high, usually exceeding
19
100 percent. * The temperatures and residence times of the processes are
high enough to ensure adequate levels of destruction. Unlike lime, aggregate
quality is not as significantly impacted by hazardous waste combustion
byproducts. Thus, sorption of certain byproducts within the aggregate can »e
tolerated. Burning of hazardous wastes in aggregate kilns appears to be
limited primarily by the effect combustion byproducts will have on the kiln
and on air pollution control requirements.
Hazardous waste containing fuels have been used by Systech Corporation of
Q
Xenia, Ohio in their cement kiln operation. Their specifications restrict
the chlorine content of waste to less than 3 percent by weight metal content
(lead, barium, zinc, and chromium) to less than 4,000 ppm, and ash content to
less than 7 percent. A Btu content of no less than 10,000 Btu/lb is also
specified. Specifications are shown in Table 12.3. Other information
compiled by Radian Corporation setting water property restrictions for
aggregate kilns is very similar (see Table 12.4).
12-10
-------
TABLE 12.3. OPERATING SPECIFICATIONS; SYSTECH CEMENT KILN PROCESS
Parameter
Waste category 1
Waste category 2
Waste category 3
Applicability
Heat content
Viscosity
Chloride content
Metals (lead,
barium, zinc,
chromium)
Ash content
Water (separated)
Will be accepted 100
percent of the time
10,000 Btu/lb, minimum
<100 cp
<3 percent, by weight
<4,000 ppm each
<7 percent
<7 percent
Accepted with some
blending at the facility
8,500 - 10,000 Btu/lb
100 - 200 cp
3-5 percent
4,000 - 6,000 ppm each
7-10 percent
1-2 percent
*Material under 8,500 Btu/lb must have less than 5 percent chlorine.
**Limit to pumpability.
Source: Systech Corporation, 1986 (Reference 9).
Blending is required before
shipment
6,000* - 8,500 Btu/lb
200 - 330** cp
5-10 percent
6,000 - 10,000 ppm each
10 - 15 percent
2-3 percent
-------
TABLE 12.4. OPERATING SPECIFICATIONS OF COMMERCIAL AGGREGATE KILNS
BURNING WASTES
Parameter
Facility-1
Facility-1
Heat content
Halogen content
Viscosity
Suspended solids
Metals content
Ash content
Sulfur content
Water content
Physical form
Must be >80,000 Btu/gal (high
enough to sustain combustion).
<5 percent, by weight, chlorine
may damage resin refractory and
scrubber due to corrosivity.
Blending is done to wastes
>3 percent chlorine.
200 cp
<10 percent, by weight, must pass
through a 1/8 inch screen.
Cannot exceed state emission
regulations.
<8 percent
<2.5 percent
No separated streams
Liquids and sludges only.
No solids.
>90,000 Btu/gal
Source: Reference 10.
12-12
-------
12.2 DEMONSTRATED PERFORMANCE
The performance of industrial boilers and other high temperature
industrial processes, firing solvents and other low molecular weight organics,
as measured by emissions, can be considered comparable in many cases to that
shown by hazardous waste incineration technologies. This is true even though
environmental regulations do not currently require equivalent performance.
The high level of performance of the conventional fuel burning systems is
attributable to the high temperatures and long residence times achieved in the
combustors without regard to their potential use as hazardous waste combustors.
Air emissions, contaminated scrubber liquors and sludges, and combustion
ash generated during the combustion of hazardous wastes in HTIPs can pose
significant problems. Clearly, the significance of certain pollutants is
attributable to both process design and waste characteristics. The
determination of the actual environmental impacts associated with the
combustion of any specific waste in a specific system will have to be assessed
on a case-by-case basis.
Numerous studies of the effectiveness of hazardous waste fuel burning
technologies have been conducted. In this section, a number of the
performance tests conducted, primarily those carried out under sponsorship of
the U.S. EPA, will be discussed, and key factors influencing performance will
be examined.
12.2.1 Industrial Boiler Tests
The performance of industrial boilers burning hazardous wastes has been
studied extensively. The primary body of information on this technological
alternative was provided in a U.S. EPA sponsored study conducted by Acurex
Corporation in 1984. This study and others have indicated that industrial
boilers are capable of achieving high levels of performance as measured by
hazardous waste destruction (DREs of more than 99.99 percent) with suitable
waste feeds. These results are attributed to the fact that the boilers have
high combustion temperature capability and long residence times. However,
even for these units, emissions of particulate matter and hydrogen chloride
can be excessive for high ash and chlorine fuel blends burned in boilers which
are not equipped with appropriate emissions control devices.
12-13
-------
Investigation of site-specific test data has revealed certain possible
trends in DREs, emissions and PIC formation. The results of several boiler
studies are described in more detail below.
Acurex Corporation, 1984 —
In the most comprehensive study conducted to date on hazardous waste
destruction in industrial boilers, Acurex Corporation evaluated the
performance of 11 different industrial boilers cofired with hazardous wastes.
Boilers were tested for destruction and removal efficiency (DRE), emission of
particulate matter, acid gas and organics, and formation of products of
incomplete combustion (PICs). A variety of wastes streams, all of which
contained hazardous solvents or ignitables, were cofired in these boilers, as
described in detail in Reference 5 and shown in Table 12.5. In summary,
wastes streams fired contained a variety of low molecular weight organic
compounds (POHCs) in wastes, representing wide ranges in chlorine content,
moisture content, and ash content. Operation of the boilers with "typical"
industrial conditions of heat input, waste/fuel ratio, and excess air was
attempted.
As summarized in Table 12.6, the DREs achieved during this test program
were at a uniformly high level, including those for the chlorinated solvents
that are generally considers difficult to destruct. Average DREs for carbon
tetrachloride, trichloroethylene, 1,1,1-trichloroethane, chlorobenzene,
benzene, toluene, tetrachloroethylene, and methyl methacrylate exceeded
99.99 percent DRE. DREs for carbon tetrachloride were generally lower than
those measured for other POHCs, but were generally greater than the standard
of 99.99 percent.
The test results were interpreted to indicate that DREs achievable by
industrial boilers are limited primarily by "nonsteady" boiler operation
including fluctuating waste feed rates and varying excess air levels present
in the combustion zone. Such conditions are often characterized by unstable
combustion (fluctuating combustion zone temperatures), high particulate
emissions, and soot formation. Carbon monoxide levels were identified as a
means of monitoring of boiler operation. Lower DREs were noted when CO levels
exceeded 80 to 100 ppm in stack effluent.
12-14
-------
TABLE 12,5 BOILER SUMMARY
Site
Boiler type
Prinary
fuel(a)
Haste description
control
device
Operational conditions
Kaeler CP 308-hp (10,000 Ib/hr
of steam) watertufee boiler
Cleaver-Brooks 250-hp
(8,400 Ib/hr of steam)
flretube boiler
Mood wastei 4 coflre testa Multiclone
chips, bark - Creosote sludge containing for
and sawdust chlorinated arooatics Including partleulate
pentachlorophenol, phenol, collection
naphthalene, and fluorenc
Typical wood boiler operation with high
excess sit and high combustible emissions.
Baseline fuel contaminated with creosote.
Boiler poorly Instrumented.
1 baseline 3 coflre tests
test - Alkyd waste water with paint
natural gas resin containing toluene,
xylenes, and several aclda
None
Low load tests. Several waste feed problems
caused by inefficient mixing of waste and
plugging of screens. Fluctuations In waste
feed flow.
IS)
I
Ul
Babcock & Hilcox 29-kg/s
(230,000 Ib/hr of steam)
multibumer watertube
1 baseline 3 cofire tests
teat - Phenolic waate containing
natural gas phenol, alkyl-benzenes, and
long~chala aromatic and
aliphatic hydrocarbons
None
Low boiler load and high excess air.
Ho operational transients.
D Babcock & Vllcox 11.4-kg/a
(90,000 Ib/hr of steam)
multlburner watertubea
1 baseline
test
no. 6 oil
3 coflre tests None'1
- Haste stream no. 1
- Mixture of methanol xylenea
and tetrachloroethylene
Burner problems experienced with waste
stream no. 1. Waste feed interruption due
to filter plugging. No transients with
waste stream no. 2.
3 coflre teats
— Waste stream no. 2
- Mixture of toluene and
bia(2-ehloroethyl) ether
Combustion Engineering 1 baseline
13.9-kg/s (110,000 Ib/hr) of test
steam single burner packaged no. 6 oil
watertube and natural
gas
1 cofire test
- Haste atream no. 1
- Mixture of methyl
methaerylate, aod
fluxing oils
6 cofire tests
- Waste stream no. 2:
waste stream no. 1
spiked with
- Carbon tetrachlorlde
- Chlorobenzene
- Triehloroethylene
1 coflre teat
- Haste stream no. 3
mixture of toluene and
methyl aethacrylate
None
Smoke emissions and transients experienced
with spiked waate stream no. 1. ' Generally
higher excess air required during coflring.
Smoke generation sensitive to orientation
of waste fuel guns and surges in waste
flowrates.
(continued)
-------
IABLE 12.5 (continued)
ro
Site Boiler type
f Babcock & Hilcoz 7.6-kg/»
(60,000 Ib/hr of steaa)
aultlburner uatertube
6 Johnston modified flretube
boiler
5.0 kg/s (40,000 Ib/hr of
steaa or 1,200-hp). Thermal
heat recovery oxidlzer (THROX)C
H Combustion Engineering
tangential NSPS coal-fired
boiler 3.2 kg/s (250,000 Ib/hr)
of superheated steaa
I- Foster Wheeler AG252 forced
draft, bent tube boiler
7.8 kg/s (62,000 Ib/hr of
steaa)
PrlBiry
fual(s)
1 baseline
test
no. 6 oil
Hone
Natural gas
used only
for startup
1 baseline
test
Pulverized
bituminous
coal
1 baseline
test staged
1 baseline
test unstaged
natural gas
Waate description
3 coflre testa
- Purge thinner containing mined
methyl eaters, butyl celloaolve
acetate, aroaatic hydrocarbons,
aliphatic hydrocarbons.
Spiked with chlorobencene
trlchloroethylene and carbon
tetrachlorlde
3 primary firings
- Mixture of chlorinated hydro-
carbons containing up to
55 percent by weight chlorine.
Major components:
- Bis(2-chloroisoproByl) ether
- Eplchlorohydrin
Spiked with carbon tetrachlorlde
3 coflre tests
- Crude methyl acetate. Spiked
with trlchloroe thane, carbon
tetrachlorlde and chlorobenzene
1 coflre staged test,
1 coflre unataged teat
- Liquid waste containing nitro-
benzene, aniline benzene.
Spiked with carbon tetrachloridc
trlchloroethylene, chlorobenzene
and toluene
Emission
control
device Operational conditions
None Improper setting of burner* caused several
flame-outs independent of waste feed.
Two chloride Steady-state operation. No primary fuel
recovery/ burned.
removal Hater
scrubber
columns in
series
Cold side High boiler load with steady state operation.
electrostatic Low waste/coal heat Input.
precipitator
None Nominal load. Ho significant boiler
transients. Damage to waste feed pumps
caused several pump replacements.
North American 3200X kg/a
(200-hp) packaged flretube
boiler
None
6 coflre tests with 2 different
trlchloroethylene concentrations
- Carbon tetrachlorlde, aono-
chlorobenzene, trlchloro-
ethylene, and toluene
None
Half and full loads high and normal EA.
No significant boiler transients or Impacts.
(continued)
-------
TABLE 12.5 (continued)
Site
K
Prlnary
Boiler type fuel(s)
Coabustion Engineering VU-lO 1 baseline
balanced draft, uatertube test
boiler 7.6 kg/s (60,000 Ib/hr) no. 6 oil
of steara
Haste description
1 cof Ire teat
- Light and heavy oil nlxtureB
Spiked with carbon tetra-
chlorlde, tclchloroethyleae,
and chlotobenzene
Emission
control
device Operational conditions
Noae Nominal test load with no significant boiler
operational transients.
"Boiler originally coker-coal fired converted to oil burning.
''Some paniculate collected by existing hopper cavities.
cPatented process for heat generation and chemical recovery of highly halogenated hydrocarbons.
-------
TABLE 12,6. SUHHARY OF PREs rOE VOLATILE PKQC
POHC Site B Site D
Carbon tetrachloridc
Triehloroethylene
1,1,1 -Tr ichloroethane
Chlorobenzene
ho
I Benzene
00
Toluene 99.9992-
99.9999-
99.991 (99.9996)
Tetrachloroethylene 99.994-
99.9992
(99.998)
Hethy line trhacry late
Mass-weighted average 99.991 99.994-
99.99990
(99.998)
Site E
99.9990-
99,9998
(99.9996)
99.994-
99.9995
(99.998)
99.995-
99.99990
(99.998)
99.97
99.95-
99.997
(99.991)
99.95-
99.9990
(99.995)
Site r
99.980
99.9990
(99.995)
99,98
99.998
(99.996)
99,96-
99.992 .
(99.98)
99.90-
99.97
(99.95)
99.90-
99.9990
(99. 9«)
Site G Site H
99.9950 99.970
99.9990 99.9994
(99.998) (99.98)
99.97-
99.9996
(99.994)
99.990-
99.997
(99.992)
99.995- 99.97-
99.9990 99.9996
(99.998) (99.991)
Site I
99.9990-
99.9993
(99.9993)
99,99990-
99.99992
(99.99991)
99.997-
99,9990
(99.998)
99.97-
99.98
(99.97)
99.998
99.97-
99.99992
(99.998)
Site J
99.997-
99.9998
(99.9990)
99.9980
99.99993
(99.9996)
99.8-
99.97
(99.95)
99.9990
99.9997
(99.9990)
99.8-
99.99993
(99.9990)
Site K Range
99.97-
99.9998
99.9998
99.98-
99.99993
99.99990
99.97-
99.9996
99.8
99.99992
99.99992
99.97-
99.996
99.996
99.90-
99.99996
99.99996
99.994-
99,9992
99.95-
99.997
99.996- 99.8-
99.99996 99.99996
(99.9997)
Weighted
average
99.9992
99.999*
99.994
99.992
99.990
99.998
99.998
99.991
99.998
Koee: Numbers in parentheses represent the site-average DEE for the POHC.
Source; Reference 5-
-------
The relationship between DRE and several other key factors was also
examined. These factors included waste fuel FOHC concentration. NO
' x
formation, surface heat release, and PIC formation. Of those factors,
correlations between concentration, surface heat release, and PIC formation
were noted.
The relationship between concentration of POHCs in the waste feed (fuel)
and ultimate DREs was found to be consistent with DRE versus concentration
correlations noted for established hazardous waste incineration technologies.
DREs decreased when the concentration of POHCs in the waste feed decreased.
DREs correlated negatively with PIC formation, thus linking PIC formation more
directly with fuel combustion as opposed to POHC destruction.
Surface heat release rates were generally correlated with DREs. DREs of
less than 99.99 percent were found to correspond to surface heat release rates
9
of less than 60,000 Btu/hour ft . This result indicates that lower boiler
heat input loads may be likely to result in lower DREs, and that the
temperature dependence of POHC DREs may be more significant than furnace
residence time.
The test results associated with Site G may be of particular importance,
because it is the only boiler unit tested which is specifically designed to
fire hazardous wastes without auxiliary fuel, and is the only boiler which is
equipped to control acid gas emissions. The boiler system consists of a
fire-tube boiler retrofitted with two scrubber columns in series. The first
column is designed to recover halogen, while the second is designed for acid
emission control. Liquid wastes are injected into the unit, which is started
up by heating with natural gas. Liquid hazardous wastes are fired without
auxiliary fuel if heat content is above 6,000 Btu/lb and are fired with
natural gas if below 6,000 Btu/lb.
The wastes fed to the boiler during the test program consisted primarily
of chlorinated organics, as shown in Table 12.7. Carbon tetrachloride was
added to this waste mixture for this test program. The concentration of
3
CC1, was reported to be 40-50 x 10 ppm by weight. Chloride content of
the waste was found to range from 36 to 48 percent.
Destruction and removal efficiencies were calculated for carbon
tetrachloride, and several other POHCs found in the waste stream. The
calculated DREs for CC1,, as shown in Table 12.8, ranged from 99.992 to
12-19
-------
TABLE 12.7. COMPOSITION OF TYPICAL WASTE FEED - SITE G
Compound
Weight percent
Bis(2-chloroisopropyl)ether
Propylene chloride
Epichlorohydrin
Propylene chlorohydrin
Trlchloropropylene
Dichloropropylene
Propionaldehyde
Heat content
40.7
30.7
17.2
5.4
3.2
1.4
1.4
9,250 Btu/lb
12-20
-------
TABLE 12.8. SUMMARY OF POHC DREs, PERCENT -- SITE G
Weighted
Average for
All Three
POHC Test 1 Test 2 Test 3 Tests
Bis(2-chloroisopropyl)ether >99.9999 >99.9999 >99.9999 >99.9999
l-Chloro-2-propanol >99.9999 >99.9999 >99.9999 >99.9999
t-l,3-DicMoropropylene > 99.9999 > 99.9999 > 99.9999 > 99.9999
Epiehlorohydrin >99.9999 >99.9999 >99.9999 >99,9999
Carbon tetrachloride 99.9990 99.9951 99.9989 99.988
Propionaldehyde 99.963 >99.998 99.750 99.687
12-21
-------
99.9995, averaging 99.998 percent. CC1, DREs never went below the standard
of 99.99 percent. DREs of other POHCs were generally higher than those for
CC1, . Analysis of semivolatile organic DREs was also conducted, and the
average DRE was found to exceed 99.9999 percent.
9
Engineering—Science Study, 1984 —
Engineering-Science conducted a field test program at a packaged
fire-tube boiler burning a mixture of toluene and chlorinated solvents. Two
blends were fired, No. 1 containing 98 percent by weight of toluene, and No. 2
containing 97 percent by weight of toluene. The chlorinated solvents were
carbon tetrachloride, trichloroethylene and chlorobenzene.
The boiler was a. small capacity fire-tube boiler, with a high heat
o
release capacity (92,000 Btu/hr ft ) and high energy liberation rate
3
(164,400 Btu/hr ft ). The boiler was selected as a "worst-case situation
for destroying hazardous wastes" in a boiler. The boiler operated at high
combustion temperatures, short residence times, and an excess air range of 17
to 50 percent. Liquid waste injection was performed through a forced air
venturi nozzle designed to fire gas or distillate oil. Six different
operating load and excess air conditions were employed during the tests.
DREs, as shown in Table 12.9, for toluene, carbon tetrachloride, and
trichloroethylene ranged from 99.9976 to 99.9999 Chlorobenzene DREs were
lower, ranging from 99.948 through 99.978. Conditions were not adequate to
achieve effective chlorobenzene destruction, but the boiler was able to
effectively destroy the other three sample compounds. However, correlations
between DRE and excess air and operating load were not apparent. The high
combustion temperature achieved at the nozzle may be the key to the high DREs
observed, despite the short residence times experienced.
GCA Study (Reference 7, 1984)—
A test program was conducted for the U.S. EPA/OSW to characterize the
performance of six small commercial sized boilers (0.5 to 12.5 x 10 Btu/hr
input) burning waste oil containing hazardous constituents. The waste oil
burned throughout the test program was prepared from a supply of used oil
purchased in bulk from a commercial vendor. The "basestock" oil (Higher
Heating Value - 16, 350 Btu/lb) was spiked with several organic chemical
compounds, including: 12-22
-------
TABLE 12.9. RESULTS OF ENGINEERING-SCIENCE INDUSTRIAL BOILER TEST PROGRAM
Test no.
1
3
5
2
4
6
Average
Boiler load/
excess air %
half/38
half/31
half/50
full/38
full/41
full/17
DREs
Toluene
99.9997
99.9992
99.9991
99.9991
99.9994
' .9991
99.9993
CC14
99.9979
99.9990
99.9993
99.9989
99.9998
99.9992
99.9990
DRE (%)
TCE
99.9999
99.9995
99.9976
99.9998
99.9999
99.9996
99.9994
•
PCB
99.952
99.978
99.972
99.946
99.948
99.973
99.961
12-23
-------
• chloroform • trichlorobenzene
* 1,1,1-trichloroethane * 1-chloronaphthalene
* trichloroethylene * 2,4,5,-trichlorophenol
* tetrachloroethylene
(perchloroethylene)
These compounds were blended with the base oil at the time of the test*.
The concentration of POHCs spiked into the waste was varied widely both from
one boiler site to another, and for separate tests conducted at the individual
sites. Concentrations of each FOHC ranged between 1,500 and 10,000 ppm.
Stack gases were collected and sampled for organics and certain specific metal
and HC1 emissions. The destruction and removal efficiency was calculated for
each FOHC for each test run, as were metals and HC1 emissions.
DRE results are presented in Table 12.10. As shown, DRE consistently
exceeded 99 percent^ but seldom were as high as 99,99 percent. DRE correlated
fairly closely to concentration, as was noted in the Reference 5 study.
Emissions of particulate matter, HC1, and a variety of toxic heavy metals
were measured in the gas stream. In keeping with data found in several
studies of waste oil combustion, roughly 60 percent of the lead and 75 percent
of the ash were emitted in the gas stream. Chlorine emissions, primarily as
hydrogen chloride, were close to stochioraetric, accounting for 86 percent of
the chlorine in the feed.
12.2.2 Industrial Kilns and Other High Temperature Industrial Processes
The performance of various high temperature industrial unit processes
12
firing hazardous wastes as a fuel has been subject to limited study. Most
of the analyses have focused upon the industrial rotary kilns, which appear to
represent the most commercially viable hazardous waste management
alternative. In general, the studies * * have shown that such units
are capable of effectively destroying POHCs.
DREs appear to be strongly correlated to the concentration of a POHC in a
waste: the highest DREs for a constituent are associated with the largest
concentration. In general, highly chlorinated waste constituents were cited
as presenting the greatest difficulty in achieving high DREs.
12-24
-------
TABLE 12.10. CALCULATED DESTRUCTION AND REMOVAL EFFICIENCIES (PERCENT)
H
ro
to
Ul
A
Boiler Data
Rated Capacity
(106 Btu/hr) 0.5
Fuel Feed Rate
(gal/hr) 3.19
Volatile Compounds
Chloroform 99.65
Trlchloroethane 99.78
Trlchloroethylene 99 . 45
Pe rchlo roethylene 9 9 . 74
Semivolatile Compounds
Trichlo robenzene 99 . 84
1-cM.oronaphthalene 99 . 95
2,4,5-trichlorophenol >99.97
Average
by
C D E F G compound
2.4 2.4 3.4 4.2 12.5
24.05 13.14 14.6 23.6 23.1
99.91 99.96 99.90 99.94 99.95 99.88
99.95 99.97 99.37 99.80 99.93 99.80
99.92 99.89 99.85 99.92 99.87 99.82
99.91 99.86 99.73 99.85 99.96 99.84
99.98 99.96 99.90 >99. 96 99.89 >99.92
99.95 99.95 >99.94 99.98 99.92 >99.95
>99.99 >99.92 >99.98 >99.97
Source: Reference 7.
-------
Emissions of partieulate matter and hydrogen chloride appeared to be
reasonably well controlled by the systems, although extensive emissions data
were seldom provided. Most of the systems tested did not employ an acid gas
scrubbing system for HC1 control. Emissions control is considered a primary
limitation to using hazardous wastes as fuels, due to the expense involved in
establishing and maintaining such systems. A probable contributor to the
relatively low volume of emissions noted in the references is that the wastes
burned in the tests did not have large chloride or ash contents. This
represents the expected typical use—as—fuel application, as high chloride and
ash contents tend to negatively impact the quality of the product of these
processes.
Emissions of other pollutants of concern, such as heavy metals, were not
discussed in great detail. As with highly chlorinated wastes, wastes
containing higher concentrations of heavy metals probably would not be good
candidates for use—as—fuel because of high partieulate emission levels and
potentially negative impacts upon product quality.
A brief summary of the major studies of using hazardous wastes as fuels
in industrial processes, for which documentation is available, is presented
below.
Engineering-Science Study (Reference 13)
Three separate asphalt plants were tested by Engineering-Science, to
determine their effectiveness in destroying waste oils spiked with
tetrachloroethylene (a.k.a., perchloroethylene) and chlorobenzene.
With the exception of metal analysis, composition data for the recycled
oil fuels were not provided. Metals concentration in the waste fuel, with the
exception of chromium and lead, were generally well below levels that would
impact on levels in the aggregate produced by the plant.
DBEs were measured at two of the plants. The measurement techniques
employed during this test program were not adequate to provide a clear
representation of DRE. DREs at Plant A were higher than those obtained at
Plant B. At Plant A, an average DEE of 99.99 percent was achieved for
perchloroethylene and chlorobenzene. At Plant B, the RCRA incinerator permit
standard of 99.99 percent DRE was never achieved for perchloroethylene or
chlorobenzene.
12-26
-------
Florida Solite Company (Reference 8, 1984)
Florida Solite Company operates an aggregate kiln fired with fuel
containing crushed coal and waste organic liquids. The liquid wastes consist
primarily of solvents, alcohols, ethers, still bottoms, and chlorinated
hydrocarbons. Average DREs reported for the five test runs were higher than
the standard of 99.99 percent, for each POHC. Toluene appears to be the
easiest to destroy and remove, while tetrachloroethylene ("perc") is the most
difficult.
Liquid waste fuel was reported to comprise 54 percent by weight of the
fuel mixture burned during this test. The waste fuels contained a variety of
organic constituents, including four POHCs; methyl ethyl ketone, methyl
isobutyl ketone, toluene and tetrachloroethyiene. The waste fuels were
relatively high in heat content (above the typical "limit" of 8,500 Btu/lb
many companies employ to determine the need for auxiliary fuel) but also high
in ash content (6 to 15 percent), and relatively low in chloride content.
San Juan Cement Company (Reference 14, 1984)
The San Juan Cement Company manufactures Portland cement in a kiln fired
with No. 6 oil and waste liquids. Wastes tested in this program (see
Reference 14) consisted of typical waste shipments to the plant* Six waste
fuel batches were used during the test program. All but one had heat contents
well in excess of 10,000 Btu/lb. Three POHCs were in the waste oils;
methyiene chloride, chloroform, and carbon tetrachloride (CC1,). The
overall chlorine contents of the wastes burned during an extensive test
program were high, ranging from 6.5 to 35 percent.
The destruction and removal efficiencies measured during this program
were generally below the RCRA incinerator permit standard of 99.99 percent.
Methyiene chloride appears to have been destroyed most effectively, while
CCl, was not effectively destroyed in the San Juan system. ORE was found to
be linked to POHC concentration.
The San Juan cement kiln inability to achieve high destruction or removal
efficiency was partially attributed to the lack of an adequate injection
mechanism for the waste fuel, which was introduced unatomized, and also to the
high level of chlorine in the waste.
12-27
-------
Rockwell Lime Company
Rockwell Lime Company operates a kiln, fired with petroleum, coke and
natural gas. To determine the feasibility of replacing natural gas fuel with
a hazardous waste fuel, tests were conducted using a waste fuel spiked with
tetrachloroethylene and triehloroethylene. The test results indicate that the
lime kiln process is capable of achieving high DREs (>99.99 percent), while
v 8
emissions remain lower than specified in RCRA permit standards.
Blast Furnace Test (Reference 11, 1985)
Liquid organic wastes are used at one blast furnace as a source of heat
and carbon content to supplement or replace coke. The blast furnace has been
retrofitted with a liquid injection system to feed waste materials to the
combustion zone. Combustion zone temperatures exceed 3000°F.
Liquid wastes blended with No. 6 oil were fed at approximately
60 galIons/minute, with recirculation. The waste employed for this test
contained 10 POHCs, including high levels of toluene, o-xylene,
tetrachloroethylene, triehloroethylene and 1,l-dichloroethylene. The total
chlorine content of the waste was found to be 835 ppm by weight.
High levels of destruction were achieved for most of the POHCs. Lower
DREs were noted, however, for benzene and naphthalene, DUE correlated somewhat
with concentration; with the exception of triehloroethane, high DREs were
achieved for those compounds which were found in high concentration in the
waste feed.
12.3 COST OF TREATMENT
The use of hazardous wastes as a fuel could provide significant economic
benefits. The amount of virgin fuel consumed would be greatly reduced,
resulting in potentially large savings in overall fuel costs. Waste use in
boilers and industrial process equipment could also greatly reduce or
eliminate capital costs for conservation of waste disposal alternatives,
particularly if suitable process equipment is already in place at industrial
facilities. If this equipment can be used, thereby avoiding the need for
additional waste treatment facilities, large cost savings can be realized.
f
12-28
-------
Estimation of the costs associated with hazardous waste fuel burning
involves assessment of equipment capital costs, maintenance and labor costs,
raw materials costs, disposal costs, liability costs, transportation costs,
$
and energy costs* The cost's may vary quite a bit from case to case, depending
upon waste characteristics and system design. Some of the more important cost
considerations are discussed below.
12.3.1 Capital Costs
The capital costs of the high temperature industrial process facilities
suitable for combustion of hazardous wastes may be very high, depending upon
the heat input capacity and the characteristics of wastes they may handle. In
general, however, the costs of designing, purchasing, and implementing those
systems is not in the multimillion dollar range found for hazardous waste
incinerators. Even greater savings in capital expenditure can be realized if
existing equipment can be retrofitted for hazardous waste combustion.
As stated by McCormick and V
components to the retrofit cost:
As stated by McCormick and Weitzman, there are three primary
1. Addition of waste storage and feeding equipment;
2. Modification of combustion systems to handle additional physical
stresses of burning hazardous wastes (e.g., corrosion resistant
parts)| and
3. Installation of air pollution control devices capable of handling
additional particulate matter, HC1, and other pollutants as required
by regulations.
The costs involved in storage and feeding, and combustion system modification
can be high—storage tanks equipped specifically to handle hazardous waste
liquids may cost in the $35,000 to $70,000 range for capacities ranging from
5,000 to 20,000 gallons, but the most significant retrofit cost element is the
air pollution control system. HTIPs are generally not equipped with APC
systems which have the capabilities required for control of hazardous waste
combustion products. It is likely that some additional APG will be required,
depending upon waste characteristics. In those cases where both the
12-29
-------
particulate loading and HC1 loading require significant control, it may be
desirable from a cost perspective to utilize AFC systems which are capable of
controlling both; e.g., a high energy scrubber.
Actual capital cost data are readily available frt»m APC equipment
vendors. Estimation of the engineering design, construction, and other costs
may be done using standard cost estimation texts or through use of guidelines
provided in Reference 15.
12.3.2 Fuel Cost Savings
The money saved as a result of substituting wastes for fuels is the most
significant cost factor associated with this management alternative. Fuel
savings may be estimated through an energy balance. As shown below, the
amount of fuel saved is equivalent to the amount of fuel required to produce
an equivalent amount of heat energy when burned at the level of efficiency the
system is capable of attaining for that waste:
H E x
_ w w w
Xf= HfEf
where: Xf - quantity of fuel, Ibs/hr
Hf = heat content of fuel, Btu/lb
Ef = HTIP combustion efficiency when fired with fuel
xw = quantity of waste fuel, Ibs/hr
Ey = heat content of waste fuel, BTU/lb
Ew =* HTIP combustion efficiency when fired with waste fuel
The fuel cost savings is then calculated as follows:
Cf = xf Pf - xwPw
where P = unit price of fuel or waste, $/lb
Cf *» cost savings, $/hr
12-30
-------
The correspondence between amount of fuel replaced and dollars saved may
not be exactly linear, as the long term effects of burning hazardous wastes
will tend to result in lower combustion efficiency and higher maintenance
costs. Maintenance costs comprise a potentially significant portion of the
overall operating costs of a waste fuel burning system. The two factors that
contribute to higher maintenance requirements are the corrosivity of the waste
and/or acid gas byproducts, and the degree to which plugging or fouling of the
system; (e.g., by metallic or resinous waste constituents) will occur. Air
pollution control systems often require the highest level of maintenance, and
thus contribute most significantly to maintenance costs.
12.4 STATUS OF DEVELOPMENT
12.4.1 Availability/Application
Although the use of waste oil as a fuel in industrial, commercial, and
even residential boilers has been a subject or interest for decades, the
practice does not appear to have grown appreciably. Interest fostered by the
high oil prices of past years has dropped as oil has become cheaper and more
plentiful. Technological requirements necessary to achieve 99.99 DRE appear
to be within the capability of most commercially available combustion
equipment. However, each waste must be considered in view of its specific
characteristics to determine if viscosity and chemical content, for example,
are consistent with good combustion and acceptable emissions. The technology
for predicting technical performance and emissions is well developed.
However, the costs of new equipment or the retrofit of existing equipment must
be contrasted with other available alternatives. These include offsite
disposal in commercial incinerators or industrial process units.
While in most cases hazardous wastes will be burned as a fuel at the
generator site, the disposal of hazardous wastes by combustion in a permitted
commercial kiln or boiler site exists as an alternative to destruction in a
hazardous waste incinerator. Only a handful of commercial HTIP facilities are
currently in operation in the U.S., burning hazardous waste. Because these
facilities derive an economic value from burning the wastes, the prices they
12-31
-------
charge for hazardous waste disposal are well below those of hazardous waste
incinerators. The types of wastes which may be burned, however, is somewhat
limited.
A company involved in this management alternative is the Systech
Corporation, which operates cement kilns in several states that are capable of
hijth level hazardous waste destruction. As noted earlier, Systech sets limits
on the characteristics of wastes they can burn, primarily because of the
effect waste burning will have on the quality of their cement. Charges for
hazardous waste destruction will vary depending upon these characteristics.
Presently they range from roughly 15 to 25 cents per gallon, based on 5,i>00
gallon tank lots* The actual prices depend on several factors, including the
location of the facility.
12.4.2 Environmental Impact
The combustion of hazardous wastes as fuels or as constituents of fuels
in high temperature industrial processes produces both air emissions and
liquid and solid effluent streams. Generation of pollution by these processes
must often be controlled, either through implementation of process equipment,
such as air pollution control (APC) systems, or through the restriction of
wastes that may be burned in a system. Environmental impacts associated with
combustion or incineration often, therefore, constitute a primary determinant
of the applicability of a specific system to a specific waste.
As indicated previously, many of the HTIPs selected as primary candidates
for this application effectively control the emission of particulate matterj
however, the control of other pollutants, most notably HC1 vapor emissions,
often presents significant problems. Environmental control is often
prohibitively expensive for smaller facilities, requiring large capital
expenditures for equipment, disposal and handling costs, labor and training.
EPAs regulations on burning hazardous wastes in HTIPs were changed as a
result of the 1984 RCRA Amendments. Previously, any hazardous waste burned to
recover energy value was exempted from regulation as a hazardous waste (and
thus were not required to be handled or achieve emissions levels for
incinerators established under EPA guidelines). In consideration of the
potential risks to human health and the environment posed by the activities
12-32
-------
involved, the regulations were changed such that wastes must meet certain
criteria to become exempted. Recent and proposed Federal regulatory
developments include the following:
• Section 204 of the Hazardous and Solid Waste Amendments of 1984
(HWSA) amended RCEA Sections 3003, 3004, and 3010 to require
notification and regulation of persons burning hazardous wastes or
hazardous waste derived fuels;
• By February 8, 1986, notification must have been furnished to EPA or
to Directors of authorized State programs by owners and operators of
any facility burning any hazardous waste or used oil for purposes of
energy recovery;
• By November 8, 1986, the Administrator must publish such standards
as may be necessary to protect human health and the environment by
regulating owners and operators of facilities burning any fuel
containing any hazardous waste;
* "Hazardous waste" for this purpose is defined in HWSA as including
any commercial chemical product listed in 40 CFR 261.33 which is
burned as a fuel in lieu of its originally intended use; and
• EPA may exempt from hazardous waste management standards facilities
which burn "de minimis" quantities of hazardous waste for energy
recovery, if the combustion takes place at the generation site, and
environmentally sound practices are otherwise employed.
The regulation of hazardous waste combustion in HTIPs at the State level
is focussed upon the applicable combustion units and the generation of air
emissions. Generally, the facilities in which hazardous waste fuels may be
burned are specified in the regulations, for example, in Massachusetts the
burning of hazardous waste fuel is prohibited except in:
a. An industrial and utility boiler or an industrial furnace permitted
or licensed by the state for that burning;
b. A hazardous waste incinerator licensed pursuant to 310 CMR 7.00 and
30.000; or
c. A cement kiln located within the boundaries of a municipality with a
population less than 50,000 (based on the most recent census
statistics) if such cement kiln is in full compliance with all
requirements of 310 CMR 30.000 and 7.08 applicable to hazardous
waste incinerators.1°
12-33
-------
No special conditions apply to the liquid or solid effluent streams from
those processes. Thus, if scrubber sludges or bottom ash from a system
burning wastes as fuel itself contains hazardous constituents, they must be
disposed of in the same manner as similar materials generated differently;
e.g., stabilized and in a secure landfill.
Air emissions standards applicable to combustion of hazardous wastes in
HTIPs vary from State to State. Ho State currently has implemented a
comprehensive air toxics emission program, although several States appear to
be developing such standards. Regulation of air toxics would have significant
impact upon hazardous waste fuel burning particularly since HTIPs may not
achieve destruction or control levels comparable to those of incinerators.
12-34
-------
REFERENCES
1. PEDCo Environmental, Inc. Evaluation of the Feasibility of Incinerating
Hazardous Waste in High-Temperature Industrial Processes.
EPA-600/2-84-049. Prepared for U.S. Environmental Protection Agency,
Industrial Environmental Research Laboratory. Cincinnati, OH. February
1984.
2. The Pace Company. Solvent Recovery in the United States. 1980-1990.
Prepared for Harding Lawson Associates. Pace Company Consultants and
Engineers, Inc. Houston, TX January 1983.
3. U.S. EPA, Fossil Fuel-Fired Industrial Boilers - Background Information
for Proposed Standards. Draft EIS. U.S. Environmental Protection
Agency, Office of Air Quality Planning and Standards. Research Triangle
Park, NC. June 1980.
4. Keeier/Dorr-Oliver Boiler Co., Inc. Correspondence sent to M. Kravett,
GCA Technology Division, Inc. February 1986.
5. Acurex Corporation. Engineering Assessment Report — Hazardous Waste
Co-firing in Industrial Boilers. Volume I and II, EPA-600/2-84-177.
Prepared for U.S. Environmental Protection Agency, Office of Research and
Development. Cincinnati, OH. November 1984.
6. Fred C. Hart Associates, Inc. Impact of Burning Hazardous Waste in
Boilers. Prepared for SCA Chemical SErvices, Inc. Fred C. Hart
Associates, Inc. New York, NY. August 1982.
7. Fennelly, et al. Environmental Characterization of Disposal of Waste
Oils in Small Combustors. Final Report. Prepared for U.S. Environmental
Protection Agency, Office of Solid Waste and Office of Research and
Development. May 1984.
8. U.S. EPA. Incineration and Treatment of Hazardous Waste: Proceedings of
the Tenth Annual Research Symposium, EPA-600/9-84-022. Articles cited
include: (a) Fennelly, McCabe, and Hall, "Environmental Characterization
of Combustion of Waste Oil in Small Commercial Boilers"; (b) Peters, Day,
and Mourninghan, "Disposal of Hazardous Waste in Aggregate Kilns"; (c)
Peters and Mourninghan, "Effects of Disposal of Hazardous Wastes in
Cement Kilns on conventional Pollutant Emissions"; (d) Day and
Mourninghan, "Evaluation of Hazardous Waste Incineration in a Lime Kiln";
12-35
-------
(3) Castaldini, et al, "Field Tests of Industrial Boilers Co-firing
Hazardous Wastes"; (f) Adams, Hartman, and Wagoner, "Field Tests of
Industrial Boilers and Industrial Processes Disposing of Hazardous
Wastes"; (g) Chehaske and Higgins, "Summary of Field Tests for an
Industrial Boiler Disposing of Hazardous Wastes". U.S. Environmental
Protection Agency, Office of Research and Development. Cincinnati, OH.
September 1984.
9. Systech Corporation. Correspondence sent to M. Kravett, GCA Technology
Division, Inc. Systech Corporation, Xenia, OH. February 1980.
10. Radian Corporation. Report on telephone conversations between Ron
Dickson, Radian Corporation, and representatives of two aggregate kiln
operating firms. Radian Corporation, McLean, VA. July 1985.
11. U.S. EPA. Incineration and Treatment of Hazardous Waste: Proceedings of
the Eleventh Annual REsearch Symposium. EPA-600/9-85/028. Articles
cited include: (a) Branscome, Westbrook, and Mourninghan, et al,
"Summary of Testing at Cement Kilns Co-firing Hazardous Waste"; (b)
Adams, et al, "Evaluation of Hazardous Waste Destruction in a Blast
Furnace". U.S. Environmental Protection Agency, Office of Research and
Development. Cincinnati, OH. September 1985.
12. Oppelt, E. T. Hazardous Waste Destruction. Environmental Science and
Technology, Vol. 20, No. 4. 1986.
13. Cottone, L.E., and D.A. Falgout. Summary Test Report - Sampling and
Analysis of Hazardous Waste and Waste Oil Burned in Three Asphalt
Plants. EPA Contract No. 68-03-314g. Prepared for U.S. Environmental
Protection Agency, Office of Research and Development, Cincinnati. OH.
14. U.S. EPA. Incineration and Treatment of Hazardous Waste: Proceedings of
the Ninth Annual Research Symposium. EPA-600/9-84-015. Article cited:
Peters and Hughes, "Evaluation of Hazardous Waste Incineration in a
Cement Kiln at San Juan Cement Company". U.S. Environmental Protection
Agency, Office of Research and Development. Cincinnati, OH. July 1984.
15. McCormick, R.J., and L. Weitzman. Preliminary Assessment of Costs and
Credits for Hazardous Waste Co-firing in Industrial Boilers. EPA
Contract No. 68-02-3176. U.S. Environmental Protection Agency.
Cincinnati, OH. October 1983.
16. Commonwealth of Massachusetts. Hazardous Waste Regulations for
Massachusetts. Public Hearing Draft Amendments and Additions to
Regulations. Commonwealth of Massachusetts, Department of Environmental
Quality Engineering. Boston, MA. Spring 1986.
12-36
-------
SECTION 13.0
LAND DISPOSAL OF RESIDUALS
Land disposal of solvent wastes is not likely to be achievable without
risk to human health and the environment unless treatment is undertaken prior
to disposal. Treatment such as pyrolysis and incineration lead to destruction
of the waste constituents of concern and the residuals (e.g., ash and scrubber
wastes) have proven to be nonhazardous (see Section 10). Other treatments
such as carbon or resin adsorption effectively reduce concentration levels in
aqueous media to below standards felt to be nonhazardous. However, the
constituents in this case are not destroyed but are merely collected and
concentrated on the sorbent surface. The sorbent then must be treated,
possibly incinerated, to eliminate the risks associated with the presence of
these hazardous constituents. In the "absence of the possibility of land
disposal of waste in a location that will be demonstrably supportive of human
health and the environment (e.g., deepwell injection), another option is to
treat the waste to immobilize the waste constituents for as long as they
remain hazardous. This method of treatment, based on fixation or
encapsulation processes, is a possibility for some solvent wastes; however, it
is more likely a treatment that will be undertaken to ensure that residuals
from other treatment processes can be safely disposed. Certain of these
residuals could be found hazardous for reasons other than solvent content;
e.g., their heavy metal content may lead to positive tests for EP toxicity.
In such cases, encapsulation may be needed to eliminate this characteristic.
As will be noted, fixation and encapsulation processes have not been
demonstrated for solvent wastes, and it is not likely that high levels of
organics (>20 percent) can be effectively treated by these techniques.
The following discussions will summarize available information concerning
immobilization techniques, namely chemical fixation or encapsulation.
Chemical fixation involves the chemical interaction of the waste with a
13-1
-------
binder; encapsulation is a process in which the waste is physically entrapped
within a stable, solid matrix. Despite the interest shown in immobilization
techniques and some generalizations made concerning their applicability to
solvent wastes, there are little, if any, data provided in the literature.
Most techniques described must be considered physical processes and few
techniques can be considered to represent chemical fixation. Even if fixation
could be demonstrated, little is known concerning the long term stability of
the matrix and the possible breakdown products over time. In situ
vitrification, a process described in Section 11, would appear to provide both
destruction of hazardous constituents with encapsulation in a matrix of long
term durability. It and the other processes described below, will require
further study to demonstrate their effectiveness for solvent wastes. However,
much will depend upon the regulatory criteria now being established by EPA for
fixation/encapsulation processes.
13.1 SOLIDIFICATION/CHEMICAL FIXATION
Solidification can be used to chemically fix or structurally isolate
solvent and ignitable wastes to a solid, crystalline, or polymeric matrix.
The resultant monolithic solid mass can then be safely handled, transported,
and disposed of using established methods of landfilling or burial.
Solidification technologies are usually categorized on the basis of the
principal binding media, and include such additives as: cement-based
compounds, lime—based pozzolanic materials, thermoplasts, and organic polymers
(thermosets). The resulting stable matrix produces a material that contains
the waste in a nonleachable form, is nondegradable, cost effective, and does
not render the land it is disposed in unusable for other purposes. A brief
summary of compatibility and cost data for selected waste solidification/
stabilization systems is presented in Tables 13.1 and 13.2.
Cement Based Systems
These systems utilize type I Portland cement, water, proprietary
additives, possibly fly ash, and waste sludges to form a monolithic, rock-like
2
mass. In an EPA publication, several vendors of cement based systems
13-2
-------
TABLE 13.1. COMPATIBILITY OF SELECTED WASTE CATEGORIES WITH DIFFERENT WASTE
SOLIDIFICATION/STABILIZATION TECHNIQUES
W
I
Treatment Type
Waste
component
Organics ;
1 . Organic
solvents and
oils
2, Solid organ-
lea (e.g.,
plastics,
resins, tara)
Inorganics :
1, Acid wastes
Cement
based
Hay impede
setting, may
escape as
vapor
Good — often
increases
durability
Cement vill
neutralize
acids
Line
based
Many impede set-
ting, may escape
as vapor
Good — often
increases
durability
Compatible
Thermoplastic
solidification
Organics may
vaporize on
heating
Possible use as
binding agent
Can be neutral-
ized before
incorporation
Organic
polymer
(UF)*
Kay retard set
of polymers
Kay retard set
of polymers
Compatible
Surface
encapsulation
Must first be
absorbed on
solid matrix
Compat ible™ssany
encapsulation
materials are
plastic
Can be neutral-
ized before
incorporation
Self-
cementing
techniques
Fire danger
on heating
Fire danger
on heating
May be neu-
tralized to
form sul-
Classification and
synthetic mineral
formation
Wastes decompose at
high temperatures
Wastes decompose at
high temperatures
Can be neutralized
and incorporated
2, Oxidlzers
3. Sulfates
4. HalIdes
Compatible
Compatible
Hay retard set- Compatible
ting and
cause spallIng
unless special
cement is used
Easily leached
from cement,
nay retard
setting
5. Heavy metals Compatible
6. Radioactive
materials
Compatible
Kay retard set,
most are
easily leached
Compatible
Compatible
Kay cause May cause Hay cause
matrix break matrix break deterioration
down, fire down of encapsulat-
ing materials
Kay dehydrate Compatible
and rehydrate
causing
splitting
Hay dehydrate Compatible
Compatible
Compatible
Acid pll solu-
blllzes metal
hydroxides
Compatible
Compatible
Compatible
Compatible
Compatible
fate salts
Compatible if nigh temperatures
sulfates may cause unde-
are present able reactions
Compatible
Compatible in nan?
cases
Compatible if Compatible In many
aulfates cases
are also •
present
Compatible if Compatible in nany
sulfates cases
are present
Compatible If Compatible
sulfates
are present
* Urea-Formaldehyde resin.
Source: Reference X.
-------
TABLE 13.2. PRESENT AND PROJECTED ECONOMIC CONSIDERATIONS FOR WASTE SOLIDIFICATION/
STABILIZATION SYSTEMS
Type of treatment
system
Cement-based
Pozzolanic
Thermoplastic
(bitumen-based)
Organic polymer
(polyester system)
i-» Surface encapsulation
*f (polyethylene)
Self-cementing
Glassif ica t ion/mineral
synthesis
Major
Materials required
Portland Cement
Lime Flyash
Bitumen
Drums
Polyester
Catalyst
Drums
Polyethylene
Gypsum (from waste)
Feldspar
Amount of ma- Cost of Ma-
Unit terial required terlal required
cost of to treat 100 Ibs to treat 100 Ibs Equipment Energy
Material of raw waste of raw waste Trends in price costs use
$0.03/lb
$0.03/lb
$0.05/lb
$27/drum
$0.«/lb
jl.ll/lb
?17/drum
Varies
**
$0.03/lb
100 Ib
100 Ib
100 Ib
0.8 drum
43 Ib of
polyester-
catalyst mix
Varies
10 Ib
Varies
$ 3.00 Stable Low Low
$ 3.00 Stable Low Low
$18.60 Keyed to oil Very high High
prices
$27.70 Keyed to oil Very high High
prices
$ 4.50* Keyed to oil Very high High
prices
** Stable Moderate Moderate
— Stable High Very high
* Based on the full cost of $91/ton.
** Negligible but energy cost for calcining are appreciable.
Source: Reference 1.
-------
reported problems with organic wastes containing oils, solvents, and greases
not miscible with an aqueous phase. For although the unreactive organic
wastes become encased in the solids matrix, their presence can retard setting,
3
cause swelling, and reduce final strength. These systems are most commonly
used to treat inorganic wastes such as incinerator generated wastes and heavy
metal sludges»
Lime Based (Pozzolanic) Techniques
Pozzolanic concrete is the reaction product of fine-grained aluminous
siliceous (pozzolanic) material, calcium (lime), and water. The pozzolanic
materials are wastes themselves and typically consist of fly ash, ground blast
furnace slag, and cement kiln dust. The cementicious product is a bulky and
heavy solid waste used primarily in inorganic waste treatment such as the
solidification of flue gas desulfurization sludge. However, biological and
paint sludges have been treated, although high concentrations (greater
than 20 percent) of organics tend to prevent the formation of a high strength
4
product.
Thermoplastic Material
In a thermoplastic stabilization process, the waste is dried, heated
(260-450°F), and dispersed through a heated plastic matrix. Principal binding
media include asphalt, bitumen, polypropylene, polyethylene, or sulfur. The
resultant matrix is resistant to leaching and biodegradation, and the rates of
loss to aqueous contacting fluids are significantly lower than those of cement
or lime based systems. However this process is not suited to wastes that act
as solvents for the thermoplastic material. Also there is a risk of fire or
secondary air pollution with wastes that thermally decompose at high
temperature.
Organic Polymers (Thermosets)
Thermosets are polymeric materials that crosslink to form an insoluble
mass as a result of chemical reaction between reagents, with catalysts
sometimes used to initiate reaction. Waste constituents could conceivably
13-5
-------
enter into the reaction, but most likely will be merely physically entrapped
withrn the crosslinked matrix. The crosslinked polymer or thermoset will not
soften when heated after undergoing the initial set. Principal binding agents
or reaetants for stabilization include ureas, phenolics, epoxides, and
polyesters. Although the thermosetting polymer process has been used most
frequently in the radioactive waste management industry, there are
formulations that may be applicable to certain organic contaminants. It is
important to note that the concept of thermoset stabilization, like
thermoplastic stabilization, does not require that chemical reaction take
place during the solidification process. The waste materials are physically
trapped in an organic resin matrix that, like thermoplastics, may biodegrade
and release much of the waste as a leachate. It is also an organic
material that will thermally decompose if exposed to a fire.
New Technology
An EPA sponsored study recently indicated that most solidification
processes in current use (silicates, lime, and cement), including those
described above, stabilize contaminants through microencapsulation rather than
chemical fixation. Microencapsulation is a process that entraps micro and
macroscopic particles individually as the fixative solidifies. An inorganic
polymer that is a candidate for true chemical fixation is the HWT product
series marketed by International Waste Technologies. The HWT series is a set
of inorganic, irreversible colloidal polymers which improve on a successful
Japanese approach which has been used in Japan for over 10 years.
In the HWT fixation process, there is a two-step reaction in which the
toxic elements and compounds are complexed first in a rapid reaction and then
permanently complexed in the building of macromolecules which continue to
generate over a long period of time. Step one of the detoxification reaction
is the blending of contaminants and HWT chemicals to achieve a homogeneous
state ao that all the toxic compounds are exposed. This blending generates
irreversible colloidal structures and ion exchanges with toxic metals and
organics. Step two is the generation of an irreversible, three-dimensional,
raacromolecule which provides the crosslinking framework. The vendor claims
that both inorganic and organic wastes are treatable in either concentrated or
13-6
-------
dilute form, although pretreatment may be necessary. Table 13.3 shows the
effect of the inorganic polymer on samples of PCB and PGP. The levels of
toxic compounds before and after treatment were determined by EPA approved
laboratory testing. A company spokesman indicated that data on the
effectiveness of HWT on concentrated trichloroethylene still bottoms will be
available in the near future.
TABLE 13.3. SUMMARY OF TEST RESULTS ON TOXIC ORGANICS
Toxic
•
organic
(yg/L)
PCB
PCP
HWT - 20
• « .
weignc
percent
15
15
15
15
Concentration
Untreated
1,140
1,800
9,200
11,000
(pg/L)
Treated
0.006
o.oey
0.337
450
Source: International Waste Technology.
International Waste Technology has estimated average treatment levels by
HWT compounds run between 8—15 percent by weight of waste with HWT compounds
costing between 12-25«{/lb. The company estimates that heavy metal electric
arc furnace dust could be treated for $19/ton while chemical still bottoms
(halogenated hydrocarbons, benzene compounds, phenols in pure state) would
cost $90-100/ton in materials costs for low volumes of waste. The bases for
these cost estimates are not entirely clear. As a fixant for low molecular
weight organics, it would appear that HWT amounts far greater than 8 to
15 percent by weight of waste would be required. At an assumeable level of
50:50 HWT/waste, costs would range from $120-250/ton for HWT material with
additional costs required for transportation, processing, and disposal.
13.2 MACROENCAPSULATION
Encapsulation is often used to describe any stabilization process in
which the waste particles are enclosed in a coating or jacket of inert
material. A number of systems are currently available utilizing
13-7
-------
polybutadiene, inorganic polymers (potassium silicates), portland concrete,
polyethylene, and other resins as macroencapsulation agents for wastes that
have or have not been subjected to prior stabilization processes. Several
different encapsulation schemes have been described in Reference 6. The
resulting products are generally strong encapsulated solids, quite resistant
to chemical and mechanical stress, and to reaction with water. Wastes
(nonsolvent) successfully treated by these methods and their costs are
summarized in Tables 13.4 and 13.5. The technologies could be considered for
stabilizing organic wastes but are dependent on the compatibility of the
organic waste and the encapsulating material. Additional research is needed
concerning the interaction of organic wastes and stabilization materials and
the durability of the matrix, if the safe disposal of wastes and treatment
residuals is to be realized through these processes. EPA is now in the
process of developing criteria which stabilized/solidified wastes must meet in
(8)
order to make them acceptable for land disposal.
13-8
-------
TABLE 13.4. ENCAPSULATED WASTE EVALUATED AT THE U.S. ARMY WATERWAYS
EXPERIMENT STATION
Code No.
100
200
300
400
500
600
700
800
900
1000
Source of Waste
SO scrubber sludge, lime process, eastern
xcoal
Electroplating sludge
Nickel - cadmium battery production sludge
SO scrubber sludge, limestone process
eastern coal
SO scrubber sludge, double alkali process
eastern coal
SO scrubber sludge, limestone pro'cess,
western coal
Pigment production sludge
Chlorine production brine sludge
Calcium fluoride sludge
SO scrubber sludge, double alkali process,
western coal
Major Contaminants
^Q ^1O
V/CL* iJW
Cu, Cr
Nl, Cd
Cu, SO
Na, Ca
Ca, SO
Cr, Pe
Na, Cl'
Ca, P~
Cu, Na
— s
h /OV-o
*» 3
, Zn
= S
, so,,-/so3-
= —
, CN
", Hg
, SOj, /S03
TABLE 13.5. ESTIMATED COSTS OF ENCAPSULATION
Process Option Estimated Cost
Resin Fusion:
Unconfined waste $110/dry ton
55-Gallon drums $0.45/gal
Resin spray-on Not determined
Plastic Welding $253/ton =
(80,000 55-gal drums/year)
Source: Reference 6.
13-9
-------
REFERENCES
1. Guide to the Disposal of Chemically Stabilized and Solidified Waste,
EPA SW-872, Sept. 1980.
2. Environmental Laboratory U.S. Army Engineer Waterways Experiment Station,
Survey of Solidification/Stabilization Technology for Hazardous
Industrial Wastes, EPA-600/2-79-056.
3* McNeese, J.A., Dawson, 6.W., and Christensen, D.C., Laboratory studies of
fixation of Kepone contaminated sediments, in "Toxic and Hazardous Waste
Disposal", Vol. 2 Pojasek, R.B. Ed., Ann Arbor Science, Ann Arbor,
Michigan. 1979.
4. Stabilizing Organic Wastes: How Predictable are the Results? Hazardous
Waste Consultant. May 1985 pg. 18.
5. Thompson, D.W. and Malone P.G., Jones, L.W., Survey of available
stabilization technology in Toxic and Hazardous Waste Disposal, Vol. 1,
Pojasek, R.B., Ed. Ann Arbor Science, Ann Arbor Michigan. 1979.
6. Lubowitz, H.R. "Management of Hazardous Waste by Unique Encapsulation
Processes." Proceedings of the Seventh Annual Research Symposium.
EPA-600/9-81-002b,
7. Newton, Jeff, International Waste Technology, Personal communication,
1986.
8. C. Wiles, Hazardous Waste Engineering Research Laboratory, U.S. EPA,
private communication; and Critical Characteristics and Properties of
Hazardous Waste Solidification/Stabilization, HWERL, U.S. EPA, Contract
No. 68-03-3186 (in publication).
13-10
-------
SECTION 14.0
CONSIDERATIONS FOR SYSTEM SELECTION
Waste management options consist of three basic alternatives: source
reduction, recycling/reuse, use of a treatment/disposal processing system or
some combination of these waste handling practices (see Figure 14.1).
Recovery, treatment, and disposal may be performed onsite in new or existing
processes or through contract with a licensed offsite firm which is
responsible for the final disposition of the waste. Selection of the optimal
waste management alternative will ultimately be a function of regulatory
compliance and economics, with additional consideration given to factors such
as safety, public and employee acceptance, liability, and uncertainties in
meeting cost and treatment objectives.
Many of the technologies discussed in previous sections can be utilized
to achieve high levels of solvent removal or destruction; however,
practicality will limit application to waste streams possessing specific
characteristics. Since many processes yield large economies of scale, waste
volume will be a primary determinant in system selection. The physical and
chemical nature of the waste stream and pertinent properties of its
constituents, including many of those properties identified in Appendix A,
will also determine the applicability of waste treatment processes. Treatment
will often involve the use of more than one technology in a system designed to
progressively recover or destroy hazardous constituents in the most economical
manner. Incremental costs of solvent removal will increase rapidly as low
concentrations are attained.
14-1
-------
SOLVENT
FEED
N)
REUSE OF
RECOVERED
PRODUCT
SOLVENT HASTE
GENERATING
PROCESS
REUSE OR
RECYCLING
SYSTEM
SOURCE
REDUCTION
I
TREATMENT/
DISPOSAL
SYSTEM
ON-
HAZARDOUS
WASTE
DISCHARG
Figure 14.1. Solvent Waste Management Options
-------
14.1 GENERAL APPROACH
All generators of hazardous solvent wastes will be required to undertake
certain steps to characterize regulated waste streams and to identify
potential treatment options. Treatment process selection should involve the
following fundamental steps:
1. Characterize the source, flow, and physical/chemical properties of
the waste.
2. Evaluate the potential for source reduction.
3. Evaluate the potential for reuse or sale of recycled solvent and
other valuable waste stream constituents.
4. Identify potential treatment and disposal options based on technical
'feasibiltiy of meeting the required extent of solvent removal or
destruction. Give consideration to waste stream residuals and
fugitive emissions to air.
5. Determine the availability of potential options. This includes the
use of offsite services, access to markets for recovered products,
and availability of commercial equipment and existing onsite systems.
6. Estimate total system cost for various options, including costs of
residual treatment and/or disposal and value o£ recovered solvent
product. Cost will be a function of items 1 through 5.
7. Screen candidate management options based on preliminary cost
estimates.
8. Use mathmatical process modeling techniques and lab/pilot scale
testing as needed to generate detailed treatment system design
characteristics and processing capabilities. The latter will define
product and residual properties and identify need for additional
treatment.
9. Perform process trials of recovered product in its anticipated end
use applications or determine marketability based on projected
stream characteristics.
10. Calculate detailed cost analysis based on modeling and performance
results.
11. Final system selection based on relative cost and other
considerations; e.g., safety, acceptance, liability, and risKs
associated with data uncertanties.
Key system selection steps are discussed in more detail below.
14-3
-------
14.2 ASSESSMENT OF ALTERNATIVES
Waste Characterization
The first step in identifying appropriate waste management alternatives
to land disposal involves characterizing the origin, flow, and quality of
generated wastes. An understanding of the processing or operational practices
which result in generation of the waste forms the basis for evaluating waste
minimization options. Waste flow characteristics include quantity and rate.
Waste quantity has a direct impact on unit treatment costs due to economies of
scale in treatment costs and marketability of recovered products. Flow rate
can be continuous, periodic, or incidental (e.g., spills) and can be
relatively constant or variable. This will have a direct impact on storage
requirements and treatment process design; e.g., continuous or batch flow.
Waste physical and chemical characteristics are generally the primary
determinant of waste management process selection for significant volume
wastes. Of particular concern is whether the waste is pumpable, inorganic or
organic, and whether it contains recoverable materials, interfering compounds
or constituents which may foul heat or mass transfer surfaces. Waste
properties such as corrosivity, reactivity, ignitability, heating value,
viscosity, concentrations of specific chemical constituents, biological and
chemical oxygen demand, and solids, oil, grease, metals and ash content need
to be determined to evaluate applicability of certain treatment processes.
Individual constituent properties such as solubility, vapor pressure,
partition coefficients, thermal stability, reactivity with various biological
and chemical (e.g., oxidants and reductants) reagents, and adsorption
coefficients are similarly required to assess treatability. Finally,
variability in waste stream characteristics will necessitate overly
conservative treatment process design and additional process controls. This
will adversely affect processing economics and marketability of recovered
products.
14-4
-------
Source Reduction Potential
As discussed in Section 5.0, source reduction potential is highly site
specific, reflecting the variablity of industrial waste generating processes
and product requirements* Source reduction alternatives which should be
investigated include raw material substitution, product reformulation, process
redesign and waste segregation. The latter may result in additional handling
and storage requirements, while differential processing cost and impact on
product quality may be more important considerations for the other
alternatives. Source reduction should be considered a highly desireable waste
management alternative. In the wake of increasing waste disposal and
liability costs, it has repeatedly proven to be cost effective while at the
same time providing for minimal adverse health and environmental impact.
Re eyeIing Potent iaI
As part of the waste characterization step, the presence of potentially
valuable waste constituents should be determined. Economic benefits from
recovery and isolation of these materials may result if they can be reused in
onsite applications or marketed as saleable products. In the former case,
economic benefits result from decreased consumption of virgin raw materials.
This must be balanced against possible adverse effects on process equipment or
product quality resulting from buildup or presence of undesirable
contaminants. Market potential is limited by the lower value of available
quantity or demand. Market potential will be enhanced with improved product
purity, availability, quantity, and consistency.
Identifying PotentialTreatment and Disposal Options
Following an assessment of the potential for source reduction and
recycling, the generator should evaluate treatment systems which are
technically capable of meeting the necessary degree of solvent removal or
destruction. Guideline considerations for the investigation of treatment
technologies are summarized in Table 14.I. The treatment objectives for a
waste stream at a given stage of treatment will define the universe of
14-5
-------
TABLE 14.1. GUIDELINE CONSIDERATIONS FOR THE INVESTIGATION OF
WASTE TREATMENT TECHNOLOGIES
A. Objectives of Treatment;
Primary function (pretreatment, treatment, residuals treatment)
Primary mechanisms (destruction, removal, conversion, separation)
Recover waste for reuse (fuel, process solvent)
Recovery of specific chemicals, group of chemicals
Polishing for effluent discharge
Immobilization or encapsulation to reduce migration
Overall volume reduction of waste
Selective concentration of hazardous constituents
Detoxification of hazardous constituents
B. Waste Applicability and Restrictive Waste Characteristics:
* Acceptable concentration range of primary & restrictive waste
constituents
* Acceptable range in flow parameters
* Chemical and physical interferences
C. Process Operation and Design;
Batch versus continuous process design
Fixed versus mobile process design
Equipment design and process control complexity
Variability in system designs and applicability
Spatial requirements or restrictions
Estimated operation time (equipment down-time)
Feed mechanisms (wastes and reagents; solids, liquids, sludges,
slurries)
Specific operating temperature and pressure
Sensitivity to fluctuations in feed characteristics
Residuals removal mechanisms
Reagent requirements
Ancillary equipment requirements (tanks, pumps, piping, heat
transfer equipment)
* Utility requirements (electricity, fuel and cooling, process and
make-up water)
Reactions and Theoretleal Considerat ions;
• Waste/reagent reaction (destruction, conversion, oxidation,
reduction)
Competition or suppressive reactions
Enhancing conditions (specify chemicals)
Fluid mechanics limitations (mass, heat transfer)
Reaction kinetics (temperature and pressure effects)
Reactions thermodynamics (endothermic/exothermic/catalytic)
(continued)
14-6
-------
TABLE 14.1 (continued)
1. Process Efficiency;
* Anticipated overall process efficiency
o Sensitivity of process efficiency to:
- feed concentration fluctuations
- reagent concentration fluctuations
- process temperature fluctuations
- process pressure fluctuations
- toxic constituents (biosystems)
- physical form of the waste
- other waste characteristics
• Acceptable range of fluctuations
F. Emissions and Residuals Management;
* Extent of fugitive and process emissions and potential sources
(processing equipment, storage, handling)
Ability (and frequency) of equipment to be "enclosed"
Availability of emissions data/risk calculations
Products of incomplete reaction
Relationship of process efficiency to emission data
Air pollution control device requirements
Process residuals (cooling and scrubber water, bottom ash, fly ash,
fugitive/residual reagents, recovered products, filter cakes,
sludges)
• Residual constituent concentrations and leachability
• Delisting potential
G. Safety Considerations:
• Safety of storing and handling wastes, reagents, products and
residuals
Special materials of construction for storage and process equipment
Frequency and need for use of personnel protection equipment
Requirements for extensive operator training
Hazardous emissions of wastes or reagents
Minimization of operator contact with wastes or reagents
Frequency of maintenance of equipment containing hazardous materials
High operating temperatures or pressures
Difficult to control temperatures or pressures
Resistance to flows or residuals buildup
Dangerously reactive wastes/reagents
Dangerously volatile wastes/reagents
14-7
-------
candidate technologies. Restrictive waste characteristics (e.g.,
concentration range, flow, interfering compounds) and technological
limitations of candidate treatment processes will reduce this to a list of
potential applications for a specific waste. Consideration must be given to
pretreatment options for eliminating restrictive waste characteristics,
process emissions, residuals and their required treatment, and opportunities
for by-product recovery. System design will be based on the most difficult
compound to remove or destroy.
A number of approaches to selecting potential treatment technologies for
solvent waste streams have been proposed , Many of these references also
provide cost information to assist the user in making a final determination of
the cost effectiveness of a process. One scheme that specifically addresses
the management of solvent bearing wastes is that proposed by Blaney in
Reference 3. Management alternatives including recycle/reuse, destructive
treatments such as those resulting from thermal oxidations, and treatments for
the removal of solvents prior to land disposal are reviewed. The reference
discusses the applicability of these waste management alternatives to solvent
waste streams having various physical chaaracteristics. Several waste
treatment techniques are described including incineration, agitated thin film
evaporation, fractional distillation, steam stripping, wet oxidation, caroon
adsorption, and activated sludge biological treatment*
Blaney discusses approaches to treating three broad categories of solvent
bearing waste: I) aqueous and mixed aqueous/organic liquids, 2) organic
liquids, and 3) sludges. As defined here, aqueous streams have water contents
of 95 percent of higher, while organic streams are described as containing
50 percent or more organic liquids. Mixed aqueous/organic streams fall in
between. Sludges are streams with solids content greater than 2 percent.
Decision charts for aqueous and mixed aqueous/organic liquids and for organic
liquids waste stream treatment are provided in Figures 14.2 and 14.3.
Discussion of these charts in Reference 3 identifies some possible treatment
options and stresses the importance of the possible need for treatment of
residuals.
The treatment processes potentially applicable to the three broad
categories of waste are shown in Table 14.2. The identification of
potentially applicable treatment processes should be considered as tentative
14-8
-------
SLUDGE
SLUDGE
TREATMENT
AND/08
DISPOSAL
DESTRUCTION
OR RECOVERY
SLUDGE
OFFGASES
AND OVERHEAD
SLUDGE
AQUEOUS OR
MIXED WASTE
YES
INCINERATE
ASH
PHYSICAL
SEPARATION
ORGANIC
FRACTION
ORGANIC
RECYCLE 08
TREATMENT
(Fig. 14.3)
AQUEOUS
FRACTION
PRELIMINARY
TREATMENT
AQUEOUS
STREAM
ORGANICS
IRANSFORMATIO
OR REMOVAL
TRANSFORMATION
ORGANIC
COMPONENT
TRANSFORMATION
REMOVAL
ORGANIC
COMPONENT
SEPARATION
AQUEOUS
STREAM
OFFGASES
& SLUDGES
AQUEOUS
STREAK
POLISHING
TREATED AQUEOUS
STREAM
Figure 14.2. Simplified decision chart for aqueous and mixed aqueous/organic
solvent waste stream treatment.
Source: Reference No. 3.
14-9
-------
INCINERATE
ASH
•K
INCINERATE
«*-
INCINERATE
ORGANIC
RESIDUE
ORGANIC
LIQUID HASTE
REUSE OR
INCINERATE
AS IS?
GROSS
SOLIDS
REMOVAL
NEEDED?
REUSE
OR
INCINERATE?
ORGANICS
SEPARATION
REQUIRED?
COMPONENT
SEPARATION
AQUEOUS OR
SLUDGE STREAM
TREATMENT
AND/OR
DISPOSAL
REUSE
PHYSICAL
SEPARATION
ORGANICS
SLUDGE
SLUDGE
TREATMENT
AND/OR DISPOSAL
ORGANIC PRODUCT
REUSE
Figure 14.3. Simplified decision chart for organic liquid solvent
waste stream treatment*
Source: Reference No. 3.
14-10
-------
TABLE 14.2. TREATMENT PROCESSES POTENTIALLY APPLICABLE TO SOLVENT WASTES
1 Aqueous and
mixed aqueous/
Process organic wastes
Preliminary Treatment
pH adjustment
Dissolved solid precipitation
Phase Separation
Solids removal
Drying
Organic fraction
Organic Component Separation
Air or stream stripping
Carbon adsorption
Fractional distillation
Resin adsorption
• Solvent extraction
Organic Compound Destruction
Incineration
Biological degradation
Chemical oxidation
Wet oxidation
Supercritical water
Stabilization/Solidification
Y
Y
Y
NA
Y .
Y
Y
Y
Y
Y
Y
Y
Y.
Y
Y
NA'
Organic
wastes
NA
NA
Y
Y
Y
Y
NA
Y
Y
Y
Y
NA
NA
NA
NA
NA
Sludges
NA
NA
NA
Y
Y
Y
NA
Y
NA
Y
Y
NA
NA
Y
NA
Y
Y = Yes
NA = Generally not applicable.
Source: Adapted from Reference 3
14-11
-------
since the treatments used will depend upon specific waste stream
characteristics not fully defined by the three general waste categories and
the purpose of the treament. In addition, other innovative and emerging
technologies described in previous sections of this document could also be
considered as applicable processes for some of these waste categories.
Concentration of organic solvent within the waste categories is a
principal determinant in assessing the applicability of a treatment process.
Concentration ranges for which treatment processes are generally applicable
are shown in Figure 14.4. Generally, techniques used for wastes with organic
concentrations over 10 percent are applicable to lower concentrations as well,
but other processes are generally more economical. Other waste
characteristics which affect process selection are waste .viscosity and solids
content and pontaminant type, volatility, and solubility. Viscosity is
important in that it indicates whether the waste stream is sufficiently fluid
to undergo treatment. If not, high temperature to improve flow properties or
treatment such as incineration in a kiln may be required. The presence of
excess solids can cause plugging of certain equipment such as packed towers
and necessitate solids removal prior to treatment. Dissolved solids may also
require removal if they precipitate or otherwise interfere with process
performance. Solubility and volatility are indicators of the ease of removal
of a volatile compound by processes such as distillation or stripping.
Finally, the type of contaminant will play a role in process selection.
Certain types of compounds may be susceptible to reaction and degradation, and
may, as in the case of halogens, produce corrosive byproducts and be
inherently low in Btu value.
A list of hazardous solvents and other low molecular weight organic
compounds and their amenability to biological and chemical treatment and
incineration is provided in Table 14.3. The information provided in this
table is necessarily general since the characteristics of the solvent matrix
will greatly influence performance. Reference 10 presents a similar table
assessing the applicability of waste treatment processes to various model
waste streams and their constituents. A letter grade is provided for every
combination of treatment technique and waste stream constituent. Order of
magnitude (+50 percent) costs are presented for example waste streams and
technologie s.
14-12
-------
Drying
H
Thin Film Evaporation
Fractional Distillation
I
I
Chemical Oxidation
• ^_ ^_ i , { tm
.4 J
Steam Stripping
_ _ . _ __ _._ _ _ _ _ • *
—w v~ , , ^ ^ _nn^« |
Incineration
Solvent Extraction
, mmmm m mng* ^_ ^^ B[ nj||_ _J^«™™™^III^^M««I™II^^I^M««™™«^«IIIIIM«™»
V
Air Stripping
Resin Adsorption
J«« H _,m .: •
Carbon Absorption .
•_ ^« t
Qzona/UVt Radiation
I —— — — — —
.j-EGENP
• COMMERCIALLY APPLIED
POTENTUt EXTENSION
I I I 1 I I I
Wet Air Oxidation
1 1
I
Supercritical Water
i i i i il
0.01
0.05
0.1
0.5 1.0
INITIAL % ORGANICS
10
Figure 14.4. Approximate ranges of applicability of treatment techniques as a
function of organic concentration in liquid waste streams.
Source: References 1 and 11.
50
100
-------
TABLE 14.3. LIST OF HAZARDOUS CONSTITUENTS AMD TREATMENT OPTIONS
Constituent
Solvents of Concern
1, 1,1-Trichloroethane
1,1, 2-Trichloro-l ,2,2-
Tr if luroroe thane
1 ,2-Dichlorobenzene
1-Butanol
Acetone
Carbon Disulf ide
NCarbon Tetrachloride
Ch lorobenzene
Cresols
Cyc lohexanone
Ethyl acetate
Ethyl ether
Ethyl benzene
Isobutyl alcohol
Methanol
Methyl ethyl ketone
Methyl isobutyl ketone
Methylene chloride
Amenable to
conventional
biological
treatment
NA
NA
Y
Y
Y
N
N
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Amenable to
aqueous
chemical
treatment
NA
NA
N
N
Y
N
N
N
Y
N
N
N
N
N
N
N
N
N
Amenable to
incineration
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
(continued)
14-14
-------
TABLE 14.3 (continued)
Amenable to Amenable to
conventional aqueous
biological chemical
Constituent treatment treatment
Nitrobenzene
Pyridine
Tetrachloroethylene
Toluene
Tr ich loroethy lene
Trichloromono-f luorometbane
Xylene
Other Solvents
1 , 1, 2-Trichloroethane
1 , 2-D ich loropr opane
1 , 3-Dichloropropene
(cis and trans isomers)
1,4— Dioxane
Acetonitrile
Aniline
Benzene
Chloroform
Cyclohexane
Dichlorodif luoromethane
Ethylene dichloride
Y
Y
Y
NA NA
NA NA
Y
NA NA
Y
Y
NA NA
Y Y
Y
Y
Y
Y
NA
Amenable to
incineration
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
(continued)
14-15
-------
TABLE 14.3 (continued)
Amenable to Amenable to
conventional aqueous
biological chemical
Constituent treatment treatment
Furfural
Propylene glycol
Tetrahydrofuran
o-dichlorobenzene
Other Low Molecular Weight Organic
Acrolein
Acrylic acid
Allyl alcohol
Allyl chloride
Cumene
Dimethylamine
Epichlorohydrin
Ethyl acrylate
Ethylene diamine
Ethylenimine
Formaldehyde
Methyl Methacrylate
Y
Y
Y
Compounds
Y Y
Y Y
Y Y
Y Y
NA
Y
NA NA
Y
Y Y
Y
Y
Y
Amenable to
incineration
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y = Affirmative
N « Negative
NA =* No information available.
14-16
-------
The advantages and limitations of the treatment processes discussed in
this document are summarized in Table 14.4. Incineration and other thermal
destruction processes are discussed first in the table because of their
general applicability to the treatment of solvent wastes. As noted by Blaney
and others, incineration may well prove to be the ultimate disposal method, at
least for solvent sludges for which solvent recovery is impractical.
Incineration will also be the major method used to dispose of still bottoms
following recovery operations. However, the extent to which incineration will
be used for these difficult to treat wastes will depend to some extent on the
technical and regulatory requirements that will be imposed on
solidification/stabilization.
Some of the technologies discussed in Table 14.4 are not generally
intended to be used as final treatment processes. Agitated thin film
evaporation and distillation, for example, are concerned primarily with
recovery/reuse. Others like wet air oxidation and chemical oxidations are
pretreatment processes than can be used to make a waste amenable to a
finishing step such as biological treatment.
Ultimately, the selection of a specific treatment system from the list of
potentially applicable processes will depend on cost, availability, and site
specific factors. These considerations are discussed below.
Availability of Potential Management Options
The availability of each component of a waste management system will
affect its overall applicability. Existing available onsite treatment process
capacity (e.g., wastewater treatment system, boiler), ancillary equipment,
labor, physical space, and utilities will have a signigicant impact on the
economic viability of a treatment system. Purchased equipment must be
available in sizes and processing capabilities which meet the specific needs
of the facility. Offsite disposal, recovery, and treatment facilities and
companies purchasing saleable products must be located within a reasonable
distance of the generator to minimize transport costs. In addition, they must
have available capacity for the waste type and volume generated. Finally,
time constraints may eliminate certain treatment processes from consideration
as a result of anticipated delays in procurement, permitting, installation, or
start-up.
14-17
-------
TABLE 14.4. SUMMARY OF SOLVENT TREATMENT PROCESSES
I
;s
Process
Incineration
Liquid injection
lac inera t.ion
Rot«ry kiln
incineration
Fluidiied bed
incineration
Fixed/ multiple
hearth*
Applicable waste streaauj
Alt pumpable liquids
provided waatei can be
blended to Btu level of
8500 Btu/lb. Sooe solids
removal «ay be necessary
to avoid plugging noitles.
All Hastes provided Btu
level 1* nalntained.
Liquidi or nonbulky
•olid,.
Can handle a wide
variety of wattes.
State oE developnent
Eatliiated that over 219
uniti are in ute. Koit
widely uaed incineration
technology.
Ovec 40 units in service;
•oat veraatile (or waste
destruction.
Mine unita reportedly
in operation-circulating
bed unita under
development «
Approximately 70 unita
in use. Old technology
ferfornsnce
Excellent destruction
efficiency (>99.99Z).
Blending can avoid
problem associated
uith reiidualt, e.g., HC1.
Excellent destruction
efficiency (> 99.991).
Excellent destruction
efficiency <>99.99%).
Perforaumce *ay be
marginal for hazardoua
teaiduala generated
TSf, pciaaibly lone PICi,
and HC1 if halogenated
organica are fired. Only
•inor aah.it aolida removed
in pretreatnent proeeaaea.
Require! APCDt. Reaiduala
ahould be acceptable if
charged properly.
Aa above.
Aa above.
for rainicipal vaate
eoBbuation.
uaatea, particularly
halogenated wastes.
Use A» A Fuel
Industrial kilns
Generally all w«ste«, but
Btu level, chlorine content,
and other ivpurity cdntent
nay require blending to
control charge characteristic*
and product quality.
Only a few units now
burning hacardoua waste*
Usually excellent
destruction efficiency
(>99.99I) because of
long residence tinea and
high temperatures.
Requires APCDs.
Residuals should be
acceptable.
Mign temperature
industrial boilers
All pumpatile fiuida. out Several units tn use,
should blend halogenated
organica. Solids removal
particularly important to
ensure stable burner operation.
deraonat rated
high ORE
was tea tmiit be olenaea
to meet e»i«iion
•tandard* (or TSP and HCl
with APCDa.
(continued)
-------
TABLE 14.4 (Continued)
Process
Other Thermal Technologies
Circulating bed
coabuttor
Molten glass
incineration
Holten salt
des t ruction
Furnace pyrolysis
unitf
Plasm* arc
pyrolyais
fluid wall
advanced
electric
reactor
In situ
vitrification
Liquids or nonbttlky
solids.
Almost all vsstes, provided
moisture and metal impurity
levels are within
liasitaticmi*
Hot auitable for high
(>20Z) ash content
wastes*
Host designs suitable
for all wastes.
Present design suitable
only for liquids.
t
Suitable for all vaates
if solids pret rested to
enaure free flow.
Techn ique for treat ing
possibly be extended to
aiurries* Alao uae aa
solidification proceaa.
Stage of development
Only one U.S. sssmifac-
hacardous waste.
Technology developed
for glass manufacturing
Not available yet as a
Technology under develop-
ment since 1969* but
further development on
hold.
One pyrolysis unit ftCRA
permitted. Certain
designs available
comercUlly.
Commercial design appears
in&inent, with future
modi ficationa planned
for treatment of sludges
and solids.
Ready for cotmerclal
development. Teat unit
permitted under RCfU.
work planned.
Performance
high efficiencies
(>99.99X>.
Ho performance data
available, but DREs
should be high
(>99«99%>.
Very high destruction
efficiencies for
organlcs ( six nines
for PCBa).
Very high destruction
efflcienciea possible
(>99.99X). Possibility
of PIC formation.
six nines in teats with
solvents.
Efficiencies have
exceeded six nines.
No data available, but
B8E« of over six nines
reported.
Residuals generated
Bed material additives
can reduce HCl emissions.
Residuals should be
acceptable.
Hill need APC device for HCl
and possibly PICa; solids
retained (encapsulated) in
molten glass*
Needs some APC devices
to collect Material not
retained in salt. Aah
disposal way be a
probleia.
TSP emissions lower than
those from conventional
will need APC devices for
produce an unacceptable
tarry residual*
Requires APC devices for
HCP and TSP, need, flare
for H2 and CO
destruct ion.
Requires APC devices for
TSP and HCl; Chlorine
removal may be required.
Off gas systen needed
to control paisaiofis
to air. Aoh contained
in vitrified soil.
(continued)
-------
TABLE 14.4 (Continued)
Process
Applicable vine stream
Stage o[ development
Performance
Residuala generated
Phyaical Treatment Methods
Distillation
M
O
Evaporation
SCe*a Stripping
Air Stripping
Liquid-Liquid
Extraction
Carbon Adaorption
Resin Adforption
Tlii* ia a proceaa u*ed to
recover and separate *olventa
Fractional distillation will
require lotidi removal to
avoid plugging colunna.
Aftitated thin filii unita
can tolerate higher level*
of solid* and higher
viacoaltiea than other
typea of itilli.
A aimple distillation
proceas to remove volatile
organica from aqueoua *olu-
tiona. Preferred for low
concentration! and aolventa
with low solubilities.
Generally uaed to treat
low concentration aqueoua
atreama.
Generally suitable only for
liquids of low aolid content*
Suitable for low aolid,
low concentration
aqueous waate atreama.
Suitable for low aolid
watte atreana. Consider
for recovery of valuable
aolvent.
Technology well developed
and equijment available
fron many auppliera;
widely practiced technology.
Separation dependa upon
reflux (99«- percent
achievable). Thii {•
a recovery proceat.
Technology la well developed Thia ia a lolvent
and equipment ia recovery proceaa.
available from aeveral Typical recovery of
aupplieri; widely 60 to 70 percent.
practicei technology.
Technology well developed
and available.
Technology well developed
and available*
Technology well developed
for induatrial proceaaing.
Technology well developed;
used aa polithing treatment.
Technology well developed
in induatry for apeciitl
reain/aolvent cosbinatiotia.
Applicability to vaate
etreava not devmatrated.
Not generally considered
a final treatment, but
can achieve low reaidual
solvent level*.
Hot generally conaidered
a final treatment, but
nay be effective for
highly volatile uaatea.
Can achieve high efficiency
separation* for certain
solvent/waste combinations.
Csn achieve low levels of
residual solvent in
effluent.
Can achieve low levela of
reaidual aolvent ia
effluent.
Hottoma will uiually contain
levela of aolvent In encea*
of 1,000 pi>»; comlensate
•ay requice further treatment.
Bottoms will contain
appreciable aolvent.
Generally auttable
for incineration.
Aqueoua treated stream
will probably require
polishing. Further
concentration of over-
head steai* generally
required.
Air emiaaiona ftay
require treatment.
Solvent solubility in
aqueoua phase ahould
be monitored.
Adsorbate must be
processed during
regeneration. Spent
carbon and wastewater
may also need treatmen
Adsorbate »u»t be
procened during
regeneration.
(continued)
-------
TABLE 14.4 (Continued)
I
to
Process
Applicable waste stream!
Stage of development
Performance
Residual* generated
Chemical Treatment Proceaaes
Wet air
oxidation
Supercritical
water oxidation
Suitable for aqueoua
liquids, also possible
for slurries. Solvent
concent rat iona up to 1SX.
For liquids and slurries
containing optimal
High temperature/
pressure technology,
widely used as pretreatment
for municipal sludges, only
one manufacturer.
Supercritical conditions
may impose demands on
Pretreatment for
biological treatment.
Some compounds
resist oxidation.
Supercritical conditions
achieve high destruction
Some residues likely which
need further treatment.
Residuals not likely to
he s problem. Halogens
Oxonation
Other chemical
oxidation
processes
Chlorinolyals
Dechlorination
concentrations of about
10Z solvent.
Oxidation with ocone
(possibly asaiated by (UV)
suitsble for low solid,
dilute aqueous solutions.
Oxidising sgents nay be
highly reactive for
specific conatituents
in aqueous solution.
Suitable for any liquid
chlorinated motes.
Dry soils and'solids.
Biological Treatment Methods
Aerobic technology suitable
for dilute waates although
some constituents will be
resistant.
system reliability.
Cosnercially available
in 1987.
Mow used ss a polishing
atep for wastewatera.
Oxidation technology well
developed for cyanidea
and other species (phenols),
not yet establiahed for
general utility.
Proceas produces a
product (e.g., carbon
tetrachlerlde). Hot
likely to be available
Not fully developed.
Conventional treatmenta
have been used for years.
efficiencies (>99.99X)
for all constituents.
Mot likely to achieve
residual aolvent level* in
the low ppm range for
moat waste*.
Not likely to achieve
residual aolvent levels in
the low ppm range for
moat wastes.
Not available.
Destruction efficiency
of over 99X reported
for dioxin.
Hay be used as final
treatment for specific
wastes, may be pretreat-
ment for resistant species.
can be neutral iced in
process.
Residual contamination
likely; will require
additional processing of
off gases.
Residual contamination
likely; will require
additional proceaaing.
Air snd wastewater
emisaions were
estiaatcd as not
significant.
Reaidual contamination
seems likely.
Residusl contamination
likely; will usually
require additional
proceaaing.
-------
Management System Cost Estimation
The relative economic viability of candidate waste management systems
will be the primary determinant of ultimate system selection. This must be
evaluated on the basis of total system costs which includes the availability
of onsite equipment, labor and utilities, net value of recovered products and
treatment/disposal processing costs. Costs for a given management system will
also be highly dependent on waste physical, chemical, and flow
characteristics. Thus, real costs are very site specific and limit the
usefulness of generalizations. The reader is referred to the sections on
specific technologies (Sections 7.0 through 13.0) for data on costs and their
variability with respect to flow and waste characteristics. Major cost
centers which should be considered are summarized in Table 14.5.
Modeling System Performance and Pilot Scale Testing
Following this preliminary cost evaluation which will enable the
generator to narrow his choice of waste management options, steps must be
taken to further finalize the selection process. These could involve the use
of mathematical models to predict design and operating requirements. However,
models often sacrifice accuracy for convenience and are not always adequate
for complex waste streams. Laboratory data, or pilot plant and full-scale
data, may ultimately be needed to confirm predicted performance. In fact,
some data may be needed as model inputs for predicting system behavior.
Processes which rely on Henry's Law Constant are a good example of the
need for experimentally documented data. Removal efficiency approximations
using Henry's Law Constant based on a ratio of pure compound vapor pressure to
its solubility often overestimate stripping by as much as two order of
2
magnitude. However, if Henry's Law Constant is obtained experimentally
using headspace analysis and batch stripping methods, it can be effectively
used to estimate equilibrium partitioning behavior.
Many models are useful for predicting constituent behavior in separation
processes. These models are based on thermodynamic equilibrium partitioning
and may also include kinetic factors to establish separation performances.
Perry's Chemical Engineers1 Handbook and other Chemical Engineering textbooks
14-22
-------
TABLE 14.5 MAJOR COST CENTERS FOR WASTE MANAGEMENT ALTERNATIVES
A. Credits
- Material/energy recovery resulting in decreased consumption of
purchased raw materials
- Sales of waste products
B. Capital Costs*
- Processing equipment
- Ancillary equipment (storage tanks, pumps, piping)
- Pollution control equipment
- Vehicles
- Buildings, land
Site preparation, installation, start-up
C. Operating and Maintenance Costs
- Overhead, operating, and maintenance labor
- Maintenance materials
- Utilities (electricity, fuel, water)
- Reagent materials
- Disposal, offsite recovery and waste brokering fees
- Transportation
- Taxes, insurance, regulatory compliance, and administration
D. Indirect Costs and Benefits
- Impacts on other facility operations; e.g., changes in product
quality as a result of source reduction or use of recycled materials
- Use of processing equipment for mangement of other wastes
*Annual costs derived by using a capital factor:
CRF .
Where: i = interest rate and n = life of the investment. A CRF of 0.177 was
used to prepare cost estimates in this document. This corresponds to
an annual interest rate of 12 percent and an equipment life of
10 years.
14-23
-------
1 O_-1 £.
are sources of information about such models. Standard analytical
packages are also avialable to predict the fate of waste stream contaminants
as they are exposed to unit operations such as stripping and distillation.
1^
For example, the Process Program developed by Simulation Sciences, Inc. was
used recently to assess the fate of contaminants in waste oil as they flowed
TJ
through a waste oil re-refining process. The Process Program allows
simulation of most chemical separation processes for which the degree of
completion is determined by thermodynamic equilibrium. The particular program
used did not allow simulation of operations involving mass transfer and
kinetics in addition to equilibrium. However, such programs are available and
must be used when kinetic factors prevent thermodynamic equilibrium from being
established; e.g., processing wastes with high viscosity -and low solvent
concentration.
In many cases models are useful in predicting behavior and can be used in
place of costly laboratory testing. Models are also useful in assessing
relative performance and costs of various approaches to treatments and the
incremental costs of achieving increasingly stringent treatment concentration
levels. Many suppliers of separation equipment use models to optimize design
and operations parameters and to scale treatment processes. The use of models
and other methods for assessing process performance are described in Perry's
and techical articles, publications, and textbooks.
The need for experimental data will depend upon the complexity of waste
stream/process interactions. Equipment manufacturers are often able to
provide experimental equipment and models to establish process parameters and
cost, including the costs required for disposal of residuals.
14-24
-------
REFERENCES
1. Allen, C. C., and B. L. Blaney. Techniques for Treating Hazardous Waste
to Remove Volatile Organic Constituents. Research Triangle Institute for
EPA HWERL. EPA-600/2-85-127 PB85-218782/REB. March 1985.
2. Allen, C. C., and B. L, Blaney, Techniques for Treating Hazardous Waste
to Remove Volatile Organic Constituents. JAPCA, Vol. 35, No. 8.
August 1985.
3. Blaney, B. L. Alternative Techniques for Managing Solvent Wastes.
Journal of the Air Pollution Control Association, 36(3): 275-285.
March 1986.
4. Ehrenfeld, J., and J. Bass, Arthur D. Little, Inc. Evaluation of
Remedial Action Unit Operations at Hazardous Waste Disposal Sites.
Cambridge, MA, Noyes Publication.
5. Bee, R.W., et al. The Aerospace Corporation. Evaluation of Disposal
Concepts for Used Solvents at DOD Bases. Report No. TDR-0083(3786)-01.
February 1983.
6. U. S. EPA Technologies and Management Strategies for Hazardous Waste
Control. U. S. EPA Office of Techology Assessment. 1983.
7. U. S. EPA Superfund Strategy. OTA-ITE-252, U. S. EPA Office of
Technology Assessment. April 1985.
8. White, R. E., Busman, T., and J. J. Cudahy, et al.,IT Enviroscience, Inc.
New Jersey Industrial Waste Study (Waste Projection and Treatment).
Knoxville, TN. EPA/600/6-85/003. May 1985.
9. Michigan Department of Commerce. Hazardous Waste Management in the Great
Lakes: Opportunities for Economic Development and Resource Recovery.
September 1982.
10. Spivey, J. J. et al., Research Triangle Institute. Preliminary
Assessment of Hazardous Waste Pretreatment as an Air Pollution Control
Technique. U. S. EPA/IERL. 15 March 1984.
11. Engineering-Science. Supplemental Report on the Technical Assessment of
Treatment Alternatives for Waste Solvents. Washington, D. C.: U. S.
Environmental Protection Agency. 1985.
12. Perry, J. H. et.al. Chemical Engineers* Handbook. Sixth Edition, McGraw
Hill. 1984.
13. Henley, E., and J. D. Seader. Equilibrium-Stage Separation Operations in
Chemical Engineering, John Wiley and Sons, Inc., New York. 1981.
14-25
-------
14. McCabe, W. L., and J. C. Smith. Unit Operations of Chemical Engineering
(Third Edition), McGraw-Hill Book Company, New York, 1979.
15. Treybal, R. E. Liquid Extraction, Second Edition, McGraw-Hill Book
Company, New York, 1963, pp. 359, 376.
16. Holland, C. D. Fundamentals and Modeling of Separation Processes,
Prentice Hall, New York. 1975.
14-26
-------
APPENDIX A
PERTINENT PROPERTIES OF THE ORGANIC SOLVENTS
AND OTHER LOW MOLECULAR WEIGHT ORGANICS
A-l
-------
APPENDIX A
PROPERTIES OF THE ORGANIC SOLVENTS IN
WASTE NUMBERS F001 , F002, F004, AND F005
EXPLANATION OF SOME PROPERTIES DATA
The data presented in. the Appendix A data sheets for the organic solvents
in Classes F001, F002, F004, and F005 have been compiled from a variety of
sources. A brief description of some of the properties data (as presented in
the Treatability Manual (EPA-6QO/ 2-82-001 a) ) will clarify some of the
information provided in the data sheets, including the units used, and explain
the utility of individual parameters.
Henry's Law Constant
Henry's law constant is the relative equilibrium concentration of a
compound in air and water at a constant temperature and is defined by the
following equation:
where: K = Henry's law constant, m? atm mol~l
P ™ compound's vapor pressure in atmospheres
S = compound's solubility in water in moles per cubic meter
The constant is an expression of the equilibrium distribution of a compound
between air and water. The constant indicates qualitatively the volatility of
a compound and is frequently used in equations that attempt to predict
"stripping" of a compound from aqueous solution. Increasing values of the
constant favor volatilization as a fate mechanism and indicate amenability to
steam or air stripping.
log Octanol/Water Partition Coefficient
The log octanol/water partition coefficient or log P is the equilibrium
distribution of a compound between two immiscible solvents, n-xjctanol and •
water. It is defined by the following equation:
A-2
-------
c
T
Log P = Log
C. H.O
A, 2
where: C. ft = concentration of compound in n-tjctanol phase
A, u
C. H.,0 = concentration of compound in water phase
A, z
Log P varies with temperature. The temperature of determination is
assumed to be 25°C, although in many cases the temperature and method of
determination are not known,
Log P measures the affinity of a compound for octanol and water phases.
It is a useful parameter for predicting the bioconcentration potential of
compounds and sorption of compounds by organic soils where experimental values
are not available. It is also used to determine the applicability of solvent
extraction as a treatment alternative. Increasing values favor strong
bioac cumulation, adsorption, and solvent extraction potentials.
Carbon Adsorption Data
Batch equilibrium carbon adsorption isotherm data can be used to estimate
the relative effectiveness of carbon in adsorbing organic compounds. The
adsorption isotherm is the relationship, at a given temperature and other
conditions, between the amounts of a substance adsorbed and its equilibrium
concentrations remaining in solution.
Carbon adsorption data can be plotted according to the Freundlich
equation. This is an empirical equation that is widely used and has been
found to describe adequately the adsorption process in dilute solution. The
Freundlich equation has the form:
Data can be fitted to the logarithmic form of the above equation, which
has the form:
X
log ^ = log K * 1/ra log Cf
where: X - G - C- = initial concentration of solute minus final concentration
of solute in solution at equilibrium, mg/L
M = weight in grams of adsorbent (carbon) per liter
C_ - final concentration of solute in mg/L
K = intercept at Cf = 1 (log Cf = 0)
1/m - slope of the line
A-3
-------
For dilute solutions in this study, this equation yields a straight line
with a slope of 1/ra and an intercept equal to the value of K when Cf = 1
(log Cg » 0). The intercept is roughly an indicator of adsorption capacity
and the slope, l/» of adsorption intensity. The concentration of compound on
the carbon in equilibrium with a concentration Cf is given by the X/M value,
expressed as rag compound/gram of carbon.
The adsorbability is defined as the carbon dose required to reduce a
pollutant concentration from concentration a_ to concentration b_. The data
here are reported for the reduction from I mg/L to 0.1 mg/L, to serve as a basis
for comparing individual compounds.
Possible Treatment Methods
Possible treatment methods have not been provided in the data sheets.
Incineration is a possible alternative for all, although a better definition
of waste characteristics is needed to assess possible alternatives.
A-4
-------
CHEMICAL NAME: Carbon tetrachloride
CHEMICAL FORMULA: CC14
CAS NO.: 56-23-5
RCRA ID: FUOl
CHEMICAL/PHYSICALPROPERTIES
MOLECULAR WEIGHT:
MELTING POINT, °C:
BOILING POINT, °C:
VAPOR PRESSURE, torr @ 20°C:
LIQUID DENSITY, g/ml § 20°C:
VAPOR DENSITY (air = 1.0):
WATER SOLUBILITY, mg/1 @ 20°C:
LOG OCTANOL/WATER COEFFICIENT. K /
3 °'
HENRY'S LAW CONSTANT, atm-m /aol:
DIPOLE MOMENT, D:
DIELECTRIC CONSTANT @ 20°C:
FRACTIONAL POLARITY:
SOLUBILITY PARAMETER, cal°"5/cm1"5
HEAT OF COMBUSTION, Kcal/mol:
HEAT OF VAPORIZATION, Kcal/mol:
FLASH POINT, °C:
153.82
-22.9
76.75
89.6
1.59
5.32
785
2.64
3.02 x 10
0
2.24
0.0
8.6
37.3
7.63
None
«•* *J
DEGRADATION
BIODEGRADATION: Probably occurs at an extremely slow rate.
PHYSIOCHEMICAL DEGRADATION: Oxidation is not a significant fate;
hydrolysis and photolysis are too slow to be significant; volatilization
is the primary transport process from the aquatic environment.
ADDITIONAL INFORMATION
CARBON ADSORPTION: K = 11; l/n = 0.83 Carbon dose = 550 mg/i
POSSIBLE TREATMENT METHODS:
A-5
-------
CHEMICAL NAME: Chlorobenzene CAS NO. : 108-90-7
CHEMICAL FORMULA: CCl RCRA ID: FOU2
CHEMICAL/PHYSICAL PROPERTIES
MOLECULAR WEIGHT: 112.56
MELTING POINT, °C: -45.6
BOILING POINT, °C: 132
VAPOR PRESSURE, torr i 20° C: 8.8
LIQUID DENSITY, g/ml § 20°C: 1.11
VAPOR DENSITY (air = 1.0): 3.88
WATER SOLUBILITY, rag/1 @ 20°C: 488
LOG OCTANOL/WATER COEFFICIENT, K , : 2.84
3 °'w -3
HENRY'S LAW CONSTANT, atm-m /mol: 3.93 x 10
DIPOLE MOMENT, D: 1.69
DIELECTRIC CONSTANT S 20° C: 5.71
FRACTIONAL POLARITY: 0.058
SOLUBILITY PARAMETER, cal°"5/cm1'5 : 9.5
HEAT OF COMBUSTION, Kcal/mol: 763.9
HEAT OF VAPORIZATION, Kcal/mol: 9.07
FLASH POINT, °C: 29
DEGRADATION
BIODE GRADATION: Probably will eventually biodegrade, but not at a
substantial rate unless microbes present are already growing on another
hydrocarbon source,
PHYSIOCHEMICAL DEGRADATION: No information available on oxidation in
ambient waters; probably will not hydrolyze in ambient waters due to the
extreme difficulty with which aryl halides undergo nucleophilic
substitution; no information available on photolysis; probably
volatilizes from water to the atmosphere at a relatively rapid rate.
ADDITIONAL INFORMATION
CARBON ADSORPTION: K = 91; 1/n = 0.99 carbon dose = 92 mg/1
POSSIBLE TREATMENT METHODS:
A-6
-------
CHEMICAL NAME: m-cresol CAS NO.: 108-39-4
CHEMICAL FORMULA: CH3C6H4OH RCRA ID: F004
CHEMICAL/PHYSICAL PROPERTIES
MOLECULAR WEIGHT: 108.15
MELTING POINT, °C: 11.5
BOILING POINT, °C: 202.2
VAPOR PRESSURE, torr @ 20°C: 0.04
LIQUID DENSITY, j?/tnl <§ 20°C: 1.03
VAPOR DENSITY (air = 1.0): 3.72
WATER SOLUBILITY, mg/1 @ 20°C: 23,500
LOG OCTANOL/WATER COEFFICIENT, K. , : 1.96
3 °'w 6
HENRY'S LAW CONSTANT, atm-m /moI: 1.4 x 10
DIPOLE MOMENT, D: 1.54
DIELECTRIC CONSTANT @ 20°C: 11.8
FRACTIONAL POLARITY:
SOLUBILITY PARAMETER, cal°*5/cm1'5: 10.2
HEAT OF COMBUSTION, Kcal/mol: 880.5
HEAT OF VAPORIZATION, Kcal/mol: 13.48
FLASH POINT, °C: 86
DEGRADATION
BIODEGRADATION: Readily iodegradable.
PHYSIOCHEMICAL DEGRADATION: No information available on oxidation; not
known to hydrolize; no information on photolysis.
ADDITIONAL INFORMATION
CARBON ADSORPTION:
POSSIBLE TREATMENT METHODS:
A-7
-------
CHEMICAL NAME: o-cresol CAS NO.: 95-48-7
CHEMICAL FORMULA: CH3C6H4OH RCRA ID: F004
CHEMICAL/PHYSICAL PROPERTIES
MOLECULAR WEIGHT: 108.15
MELTING POINT, °C: 30.9
BOILING POINT, °C: 191
VAPOR PRESSURE, torr @ 20°C: 0.24
LIQUID DENSITY, g/ml @ 20°C: 1.05
VAPOR DENSITY (air » 1.0): 3.72
WATER SOLUBILITY, mg/1 @ 20°C: 25,000
LOG OCTANOL/WATER COEFFICIENT, KQ , : 1.95
HENRY'S LAW CONSTANT, atm-m /mol: 1.4 x 10~
DIPOLE MOMENT, D: 1.41
DIELECTRIC CONSTANT @ 20°C: 11.5
FRACTIONAL POLARITY:
SOLUBILITY PARAMETER, cal^/cm1'5: 10.2
HEAT OF COMBUSTION, Kcal/mol: 882.6
HEAT OF VAPORIZATION, Kcal/mol: 12.49
FLASH POINT, °C: 81
DEGRADATION
BIODEGRADATION: Readily biodegradable.
PHYSIOCHEMICAL DEGRADATION: No information available on oxidation; not
known to hydrolize; some degradation by direct photolysis in aqueous
media.
ADDITIONAL INFORMATION
CARBON ADSORPTION:
POSSIBLE TREATMENT METHODS:
A-8
-------
CHEMICAL NAME: p-cresol CAS NO. : 106-44-5
CHEMICAL FORMULA: OCHH RCRA ID: F004
CHEMICAL/PHYSICAL PROPERTIES
MOLECULAR WEIGHT: 108.15
MELTING POINT, °C: 34.8
BOILING POINT, °C: 201.9
VAPOR PRESSURE, torr @ 20°C; 0.11
LIQUID DENSITY, g/ml @ 20°C: 1.03
VAPOR DENSITY (air = 1.0): 3.72
WATER SOLUBILITY, m«/l <§ 20°C: 24,000
LOG OCTANOL/ WATER COEFFICIENT, K , : 1.94
,. Of w «
HENRY'S LAW CONSTANT, atm-m /mol: 1.4 x 10
DIPOLE MOMENT, D: 1.54
DIELECTRIC CONSTANT @ 20°C: 9.9
FRACTIONAL POLARITY:
SOLUBILITY PARAMETER, cal0*5/^1*5 : 10.2
HEAT OF COMBUSTION, Kcal/mol: 882.5
HEAT OF VAPORIZATION, Kcal/mol: 13.61
FLASH POINT, °C: 86
DEGRADATION
BIODEGRADATION: Readily biodegradable.
PHYSIOCHEMICAL DEGRADATION: No information available on oxidation; not
known to hydrolize; no information on photolysis.
ADDITIONAL INFORMATION
CARBON ADSORPTION:
POSSIBLE TREATMENT METHODS:
A-9
-------
CHEMICAL NAME: Cresylic acid CAS NO.: 1319-77-3
CHEMICAL FORMULA: CH3C6H4OH RCRA ID: FU04
CHEMICAL/PHYSICAL PROPERTIES
MOLECULAR WEIGHT: 108.15
MELTING POINT, °C: 10.9-35.5
BOILING POINT, °C: 195-205
VAPOR PRESSURE, torr @ 20°C: 0.24
LIQUID DENSITY, g/ml @ 20°C: .1.03-1.04
VAPOR DENSITY (air = 1.0): 3.72
WATER SOLUBILITY, ms/1 @ 20°C: 25,000
LOG OCTANOL/WATER COEFFICIENT, K , : 2.04
3 0/w 6
HENRY'S LAW CONSTANT, atm-m /mol: 1.4 x 10
DIPOLE MOMENT, D: 1.5
DIELECTRIC CONSTANT <§ 20°C: 10-12
FRACTIONAL POLARITY:
SOLUBILITY PARAMETER, cal°*5/cm1'5: 10.2
HEAT OF COMBUSTION, Kcal/mol: _880
HEAT OF VAPORIZATION, Kcal/mol: _13
FLASH POINT, °C: 81
DEGRADATION
BIODEGRADATION: Similar to cresols.
PHYSIOCHEMICAL DEGRADATION:
ADDITIONAL INFORMATION
CARBON ADSORPTION:
POSSIBLE TREATMENT METHODS;
A-10
-------
CHEMICAL NAME: o-Dichlorobenzene CAS NO.: 95-50-1
CHEMICAL FORMULA: C6H4CL2 RCRA ID: F002
CHEMICAL/PHYSICAL PROPERTIES
MOLECULAR WEIGHT: 147.00
MELTING POINT, °C: -17.6
BOILING POINT, °C: 180.5
VAPOR PRESSURE, torr @ 20°C: 1
LIQUID DENSITY, g/ml i 20°C: 1.30
VAPOR DENSITY (air = 1.0): 5.05
WATER SOLUBILITY, mg/1 @ 20°C: 145
LOG OCTANOL/WATER COEFFICIENT, K , : 3.38
~ o/w 3
HENRY'S LAW CONSTANT, atm-m /mol: 1.94 x 10
DIPOLE MOMENT, D: 2.50
DIELECTRIC CONSTANT @ 20°C: 7.50
FRACTIONAL POLARITY: 0
SOLUBILITY PARAMETER, cal°"5/cml*5: 10.0
HEAT OF COMBUSTION, Kcal/mol: 671.8
HEAT OF VAPORIZATION, Kcal/mol: 10.94
FLASH POINT, °C: 74
DEGRADATION
BIODEGRADATION: Sufficiently resistant to biodegradation to make
volatilization more important.
PHYSIOCHEMICAL DEGRADATION: Resistant to autooxidation by peroxy radical
in water; oxidation by hydroxyl radicals occurs in the atmosphere;
hydrolysis is not important; photolysis probably occurs slowly.
ADDITIONAL INFORMATION
CARBON ADSORPTION: K = 129; 1/n = 0.43 carbon dose = 19 mg/1
POSSIBLE TREATMENT METHODS:
A-ll
-------
CHEMICAL NAME: Isobutanol
CHEMICAL FORMULA:
CAS NO. : 78-83-1
ECRA IB: FOOS
CHEMICAL/PHYSICAL PROPERTIES
MOLECULAR WE IGHT:
MELTING POINT, °C:
BOILING POINT, °C:
VAPOR PRESSURE, torr @ 20° C:
LIQUID DENSITY, g/ml @ 20°C:
VAPOR DENSITY (air = 1.0):
WATER SOLUBILITY, mg/1 @ 20°C:
LOG OCTANOL/WATER COEFFICIENT, K ,
1 °/
HENRY'S LAW CONSTANT, atm-m /mol:
. DIPOLE MOMENT, D:
DIELECTRIC CONSTANT <§ 20°C:
FRACTIONAL POLARITY:
SOLUBILITY PARAMETER, cal°*5/cm1*5
HEAT OF COMBUSTION, Kcal/mol;
HEAT OF VAPORIZATION, Kcal/mol:
FLASH POINT, °C: •
74.12
-114.7
99.5
9
0.81
2.55
95,000
0.83
2.2 x 10
1.66
18.7
0.11
10.7
638.2
10.94
27
DEGRADATION
BIODEGRADATION: No information available.
PHYSIOCHEMICAL DEGRADATION: No information available for oxidation,
hydrolysis, or photolysis.
ADDITIONAL INFORMATION
CARBON ADSORPTION:
POSSIBLE TREATMENT METHODS:
A-12
-------
CHEMICAL NAME: Methylene chloride
CHEMICAL FORMULA: CH2C12
CAS NO.: 75-09-2
RCRA ID: F001, FUU2
CHEMICAL/PHYSICAL PROPERTIES
MOLECULAR WEIGHT:
MELTING POINT, °C;
BOILING POINT, °C:
VAPOR PRESSURE, torr @ 20°C:
LIQUID DENSITY, g/ml @ 20°C:
VAPOR DENSITY (air = 1.0):
WATER SOLUBILITY, mg/l @ 20°C:
LOG OCTANOL/WATER COEFFICIENT, K
HENRY'S LAW CONSTANT, atm-mVmol:
DIPOLE MOMENT, D:
DIELECTRIC CONSTANT @ 20°C:
FRACTIONAL POLARITY:
SOLUBILITY PARAMETER, cal°"5/cml*5:
HEAT OF COMBUSTION, Kcal/mol:
HEAT OF VAPORIZATION, Kcal/mol:
FLASH POINT, °C:
o/w'
84.94
-95
39.75
3b2.4
1.33
2.93
20,000
1.25
3.19 x 10~3
1.60
9.08
0.12
9.7
106.8
7.57
None
DEGRADATION
BIODEGRADATION: Probably occurs, but at an extremely slow rate.
PHYSIOCHEMICAL DEGRADATION: Oxidation in aqueous phase probably not
important; hydrolysis probably not significant, neither are photochemical
reactions in aqueous media. Volatilization is major pathway for loss
from water.
ADDITIONAL INFORMATION
CARBON ADSORPTION: Ks 1.3; 1/n « 1.2 Carbon dose = 10,000 mg/1
POSSIBLE TREATMENT METHODS:
A-13
-------
CHEMICAL NAME: Methyl ethyl ketone CAS NO.: 78-93-3
CHEMICAL FORMULA: CH3COC2H5 RCRA ID: F005
CHEMICAL/PHYSICAL PROPERTIES
MOLECULAR WEIGHT: 72,11
MELTING POINT, °C: -86.35
BOILING POINT, °C: 79.64
VAPOR PRESSURE, torr @ 20°C: 71.2
LIQUID DENSITY, g/ml @ 20°C: 0.81
VAPOR DENSITY (air = 1.0): 2.5
WATER SOLUBILITY, tag/1 @ 20°C: 100,000
LOG OCTANOL/WATER COEFFICIENT, K , : 0
3 °'w -5
HENRY'S LAW CONSTANT, atm-m /raol: 4.35 x 10
DIPOLE MOMENT, D: 2.7
DIELECTRIC CONSTANT @ 20°C: 18.5
FRACTIONAL POLARITY: 0.510
SOLUBILITY PARAMETER, cal°'5/on1'5: 9.3
HEAT OF COMBUSTION, Kcal/mol: 582.3
HEAT OF VAPORIZATION, Kcal/mol: 8.15
FLASH POINT, °C: -1
DEGRADATION
BIODEGRADATION: Biodegradable.
PHYSIOCHEMICAL DEGRADATION: No information available on oxidation,
hydrolysis, or photolysis.
ADDITIONAL INFORMATION
CARBON ADSORPTION:
POSSIBLE TREATMENT METHODS:
A-14
-------
CHEMICAL NAME: Nitrobenzene CAS NO.: 98-95-3
CHEMICAL FORMULA: C6H5N02 RCRA ID: F004
CHEMICAL/PHYSICAL PROPERTIES
MOLECULAR WEIGHT: 123.11
MELTING POINT, °C: -5.6
BOILING POINT, °C: 211
VAPOR PRESSURE, torr @ 20°C: 0.15
LIQUID DENSITY, g/ml i 20°C: 1.2
VAPOR DENSITY (air = 1.0): 4.2
WATER SOLUBILITY, rag/1 @ 20°C: 1,900
LOG OCTANOL/WATER COEFFICIENT, K , : 1.85
o/w
HENRY'S LAW CONSTANT, atm-m3/mol: 2.4 x 10~5
DIPOLE MOMENT, D: 4.22
DIELECTRIC CONSTANT @ 20°C: 35.74
FRACTIONAL POLARITY: 0.63
SOLUBILITY PARAMETER, cal0*5/^1*5: 10.0
HEAT OF COMBUSTION, Kcal/mol: 739.2
HEAT OF VAPORIZATION, Kcal/mol: 12.17
FLASH POINT, °C: 88
DEGRADATION
BIODEGRADATION: Slow, but could be significant in the absence of
appreciable photolysis.
PHYSIOCHEMICAL DEGRADATION: Oxidation is highly improbable; it is not
known to hydrolizej photolysis may be significant if the compound is
adsorbed on humus near the air/water surface. Volatilization is unlikely
to be significant in transport.
ADDITIONAL INFORMATION
CARBON ADSORPTION: K = 68; 1/n = 0.43 carbon dose = 36 mg/1
POSSIBLE TREATMENT METHODS:
A-15
-------
CHEMICAL NAME: Pyridine
CHEMICAL FORMULA: CH < (CHCH)2 > N
CAS NO.: 110-86-1
RCRA ID: F005
CHEMICAL/PHYSICAL PROPERTIES
MOLECULAR WEIGHT:
MELTING POINT, °C:
BOILING POINT, °C:
VAPOR PRESSURE, torr § 20°C:
LIQUID DENSITY, g/ml @ 20°C:
VAPOR DENSITY (air = 1.0):
WATER SOLUBILITY, mg/l @ 20°C:
LOG OCTANOL/WATER COEFFICIENT, K
o
" HENRY'S LAW CONSTANT, atm-m /mol:
•DIPOLE MOMENT, D:
DIELECTRIC CONSTANT § 20°C:
FRACTIONAL POLARITY:
SOLUBILITY PARAMETER, cal^/cm1*
HEAT OF COMBUSTION, Kcal/mol:
HEAT OF VAPORIZATION, Kcal/mol:
FLASH POINT, °C:
79.10
-42
115.3
20
0.98
2.73
Miscible
0.64
2.4 x 10
2.19
12.5
0.174
10.7
658.5
9.65
23
-5
DEGRADATION
BIODEGRADATION: Likely to be persistent in the abiotic environment of
most ground waters.
PHYSIOCHEMICAL DEGRADATION: No information available on oxidation,
hydrolysis, or photolysis.
ADDITIONAL INFORMATION
CARBON ADSORPTION:
POSSIBLE TREATMENT METHODS:
A-16
-------
CHEMICAL NAME: Tetrachloroethylene
CHEMICAL FORMULA: Cl2C:CCl2
CAS NO.: 127-18-4
RCRA ID: FUOl, FUU2
CHEMICAL/PHYSICAL PROPERTIES
MOLECULAR WEIGHT: 165.85
MELTING POINT, °C: -22.7
BOILING POINT, °C: 121
VAPOR PRESSURE, torr @ 20°C: 14
LIQUID DENSITY, g/ml @ 20°C: 1.63
VAPOR DENSITY (air =1.0): 5.8
WATER SOLUBILITY, tng/l @ 20°C: 150
LOG OCTANOL/WATER COEFFICIENT, K , : 2.88 '
HENRY'S LAW CONSTANT, atm-m /mol: 2.82 x 10~
DIPOLE MOMENT, D: 1.32
DIELECTRIC CONSTANT @ 20°C: 2.46
FRACTIONAL POLARITY: 0
SOLUBILITY PARAMETER, cal°*5/cm1*5: 9.35
HEAT OF COMBUSTION, Kcal/mol: 162.5
HEAT OF VAPORIZATION, Kcal/raol: 9.24
PLASH POINT, °C: None
DEGRADATION
BIODEGRADATION: Potentially biodegradable; probably slow; conflicting
results.
PHYSIOCHEMICAL DEGRADATION: Oxidation occurs slowly in aquatic
environment; hydrolysis is probably too slow to be significant;
photolysis probably does not occur; volatilization is the primary
transport process.
ADDITIONAL INFORMATION
CARBON ADSORPTION: K = 51; l/n = 0.56 carbon dose = 64 mg/1
POSSIBLE TREATMENT METHODS:
A-17
-------
CHEMICAL NAME: Toluene CAS NO.: 108-88-3
CHEMICAL FORMULA: C6H5CH3 RCRA ID: F003
CHEMICAL/PHYSICAL PROPERTIES
MOLECULAR WEIGHT: 92.13
MELTING POINT, °C: -95
BOILING POINT, °C: 110.6
VAPOR PRESSURE, torr @ 20°C: 28.4
LIQUID DENSITY, g/ml @ 20°C: 0,87
VAPOR DENSITY (air = 1.0): 3.14
WATER SOLUBILITY, mg/1 @ 20°C: 470
LOG OCTANOL/WATER COEFFICIENT. K / : 2.69
3 0/w -3
HENRY'S LAW CONSTANT, atm-m /taol: 5.93 x 10
DIPOLE MOMENT, D: 0.36
DIELECTRIC CONSTANT @ 20°C: 2.39
FRACTIONAL POLARITY: 0.001
SOLUBILITY PARAMETER, cal°*5/cm1*5: H.9
HEAT OF COMBUSTION, Kcal/mol: 934.2
HEAT OF VAPORIZATION, Kcal/mol: 8.58
FLASH POINT, °C: 7
DEGRADATION
BIODEGRADATION: Relative importance of biodegradation cannot be
determined.
PHYSIOCHEMICAL DEGRADATION: Oxidation is probably not important as
aquatic fate; atmospheric photo-oxidation subordinates all other fate
processes. Hydrolysis not aquatically significant; volatilization is
significant transport process.
ADDITIONAL INFORMATION
CARBON ADSORPTION: K = 26; 1/n » 0.44 carbon dose = 96 mg/1
POSSIBLE TREATMENT METHODS:
A-18
-------
CHEMICAL NAME: I,I,l-Trichloroethane
CHEMICAL FORMULA: C13CCH3
CAS NO.: 71-55-6
RCRA ID: F001, FUU2
CHEMICAL/PHYSICAL PROPERTIES
MOLECULAR WEIGHT:
MELTING POINT, °C:
BOILING POINT, °C:
VAPOR PRESSURE, torr @ 20°C:
LIQUID DENSITY, g/ml <§ 20°C:
VAPOR DENSITY (air » 1.0):
WATER SOLUBILITY, mg/1 @ 20°C:
LOG OCTANOL/WATER COEFFICIENT, K ,
3 °'
HENRY'S LAW CONSTANT, atm-m /mol:
DIPOLE MOMENT, D:
DIELECTRIC CONSTANT S 20°C:
FRACTIONAL POLARITY:
SOLUBILITY PARAMETER, eal°*5/era1"5:
HEAT OF COMBUSTION, Kcal/mol:
HEAT OF VAPORIZATION, Kcal/mol:
FLASH POINT, °C:
133.41
-30.41
74.1
9b.O
1.34
4.6
950
2.17
4.92 x 10
1.78
7.5
0.069
8.5
213.3
8.01
None
-3
DEGRADATION
BIODEGRADATION: Probably occurs, but at an extremely slow rate.
PHYSIOCHEMICAL DEGRADATION: Oxidation not significant; hydrolysis
probably too slow to be significant; photolysis not significant;
volatilization is primary transport process in aquatic environment.
ADDITIONAL INFORMATION
CARBON ADSORPTION: K = 2.48; 1/n = 0.34 carbon dose = 800 mg/l
POSSIBLE TREATMENT METHODS:
A-19
-------
CHEMICAL NAME: Trichloroethylene
CHEMICAL FORMULA: C12C:CHC1
CAS NO.: 79-01-6
RCRA IB: FUUL, FUU2
CHEMICAL/PHYSICAL PROPERTIES
MOLECULAR WEIGHT:
MELTING POINT, °C:
BOILING POINT, °C:
VAPOR PRESSURE, torr @ 20°C:
LIQUID DENSITY, g/ral @ 20°C:
VAPOR DENSITY (air = 1.0):
WATER SOLUBILITY, ms/1 @ 20°C:
LOG OCTANOL/WATER COEFFICIENT, K
3
HENRY'S LAW CONSTANT, atrn-rn /mo1:
DIPOLE MOMENT, D:
DIELECTRIC CONSTANT § 20°C:
FRACTIONAL POLARITY:
SOLUBILITY PARAMETER, cal /cm :
HEAT OF COMBUSTION, Kcal/raol:
HEAT OF VAPORIZATION, Kcal/mol:
FLASH POINT, °C:
A
o/w
131.39
-73
87.19
57.9
1.94
4.53
1,000
2.29
1.17 x 10"
0.8
3.42
0.005
9.2
230.0
8.32
None
DEGRADATION
BIODEGRADATION: Potentially biodegradable; probably slow; conflicting
results.
PHYSIOCHEMICAL DEGRADATION: Oxidation occurs slowly in aquatic
environment; hydrolysis is probably too slow to be significant;
photolysis probably does not occur; volatilization is the primary
transport process.
ADDITIONAL INFORMATION
CARBON ADSORPTION: K = 28; 1/n = 0.62 carbon dose = 130 mg/1
POSSIBLE TREATMENT METHODS:
A-20
-------
CHEMICAL NAME; Trichlorofluororaethane
CHEMICAL FORMULA: CC13F
CAS NO.: 75-69-4
RCRA ID: FU01, F002
CHEMICAL/PHYSICAL PROPERTIES
MOLECULAR WEIGHT:
MELTING POINT, °C:
BOILING POINT, °C:
VAPOR PRESSURE, torr @ 20°C:
LIQUID DENSITY, g/ml § 2Q°C:
VAPOR DENSITY (air - 1.0):
WATER SOLUBILITY, mg/l @ 20°C:
LOG OCTANOL/WATER COEFFICIENT, K / :
3 ' o/w
HENRY'S LAW CONSTANT, atm-m /moI:
DIPOLE MOMENT, D:
DIELECTRIC CONSTANT @ 20°C:
FRACTIONAL POLARITY:
SOLUBILITY PARAMETER, cal°*5/cm1"5:
HEAT OF COMBUSTION, Kcal/mol:
HEAT OF VAPORIZATION, Kcal/mol:
FLASH POINT, °C:
137.4
-111
23.8
667.4
1.5
4.7
1,125
2.53
5.83 x 10"
0.45
2.41
8.0
6.42
None
DEGRADATION
BIODEGRADATION: Volatility probably precludes biodegradation.
PHYSIOCHEMICAL DEGRADATION: Oxidation not a significant fate; hydrolysis
too slow to be significant; photolysis not significant; volatilization is
primary transport process.
ADDITIONAL INFORMATION
CARBON ADSORPTION: K = 5.6; l/n = 0.24 carbon dose = 280 mg/1
POSSIBLE TREATMENT METHODS:
A-21
-------
CHEMICAL NAME: 1,I,2-Trichioro-i,2,2-trifiuoroethane CAS NO.: 76-13-1
CHEMICAL FORMULA: FC12CCC1F2 RCRA ID: FUUl, FUU2
CHEMICAL/PHYSICAL PROPERTIES
MOLECULAR WEIGHT: 187.38
MELTING POINT, °C: -36.4
BOILING POINT, °C: 48
VAPOR PRESSURE, torr @ 20°C: 270
LIQUID DENSITY, g/ml @ 20°C: 1.56
VAPOR DENSITY (air = 1.0): 6.5
WATER SOLUBILITY,'mg/1 @ 20°C: 10
LOG OCTANOL/WATER COEFFICIENT, K / : 2.0U
•3 O/W
HENRY'S LAW CONSTANT, atm-m /moI: 4.3 x 10
DIPOLE MOMENT, D:
DIELECTRIC CONSTANT @ 20°C: 2.41
FRACTIONAL POLARITY:
SOLUBILITY PARAMETER, cal°'5/aa1'5: 7.4
HEAT OF COMBUSTION, Kcal/mol:
HEAT OF VAPORIZATION, Kcal/mol: 7.12
FLASH POINT, °C: None
-2
DEGRADATION
BIODEGRADATION: No information available.
PHYSIOCHEMICAL DEGRADATION: No information available on oxidation,
hydrolysis, or photolysis. Volatilization is most important transport
mechanism.
ADDITIONAL INFORMATION
CARBON ADSORPTION:
POSSIBLE TREATMENT METHODS:
A-22
-------
TABLE A-l. PHYSICAL AND CHEMICAL PROPERTIES OF SOLVENTS AND OTHER
LOW MOLECULAR WEIGHT ORGANICS.
Constituent
8glv«*t« ol_ C0|c«rii
1.1, l-trichloroeth>n«
1 t 1 ,2-triehloetri ,2 .2-
t r it loor o«£h»ne
l-,l,-4icMoro««,n.
l-but*ool
Acetpse
C«rboo DUulfide
Cart>Q& T*tr»chloride
ChlQTobeoiflit*
Cr**oU
y^
i
N5 Cyclohexanona
w
Ithyl acetate
ttnyl ether
ethyl b«MM«
lie-butyl alehobel
EPA Chemical
code formula
U226 C2U]Cl]
Z049 CfCljCTjCl
B071 Ct«4Clt
V031 CHjtCHj^OH
IM02 CHjCOCilj
'022 C32
Mil CC15
0037 C6H5C1
U052 C6H4CHjO«
U057 C(H|QO
V112 CHjCOOCjH;
U117 C2B;OCj>s
ZOiS C6H5CjH5
DIM CCB3)jClCll2ai
Holecular
veight Boiling
(g/M) point CO
133.4 74.1
187.38'- 41
147.01 173
74 117.25
58 56.2
76 46.25
153.82 76.54
112.56 1)2
108 o: 192
•: 202,8
p: 201.9
98 156
»» 77
74 34.51
106.16 136.25
74 9».5
Vapor
pressure
(t*a Bg « 25'C)*
125
270*0
400M
2.21
6.5
200
266
357
ll»
11. 7M
o: 0.43
.: O.lt
f. 0.16
4.57
82.2
85
540
10
10
12
Solubility
in water Activity
(•g/L g 25'c)« coefficient
720 382
HA N*
123 HA
91,000 42.9
•iacibla 24.6
2,540 1
1,160 HA
488 10,500
o: 31,000 209
•: 23,500
l>: 24,000
23,000 236
79,000 105
60,500 NA
140*5 27,000
95,000 43
Specific
gravity
(g 20/4*C>*
1.33
1.56
1.28
0.81
0.79
1.263
1.594
1.106
o: 1.041
at; 1.038
p: 1.034
0.95
0.90
0.713
0.166
0.798
Henry'a Lav
constant
(at* =3 not*1)
0.03
0.482
1.94x10"'
7,10-6
8.9.1 OT*
».*«io-*
2.5xlO-5
0.015
0.023
3.93«10~3
ot 2x10-*
•; 1x10"*
ps 9.5x10-'
2.56xlO"5
4.1x10"'
iUt«ur*
«.69*10-*
8,7.10-1
1.03x10-5
2.2x10-*
Oetanol-watar
part it ion
coefficient
320, 309
t
436.5
l.telO1
7.58
0.5S
69.2
144.5
912
436
765
«: 91.2
102,3
5: 83.2
87.2
6.46
4.57
5.37
5.89
6.76
1,412
4.47
6.76
Heat of flash Heat of
combustion point vaporixaticra
(U/koU 9 2S*C>* CO** (KJ/aole 9 2S'C)«
1,110 Hona
MA None
673 6)
2,675 36-38
1,790 -IS (CC)
1,031 -30 (OCX
156 Mom
3.108 23
o: 36* o: 81-83
«: 368 .: 86 (CC)
p: 369 «: 86 (cc)
HA 63 (CC)
2,246 7.2 (CC)
2,727 41 (CC)
4,565 18 (CC)
2,*70 27.8
33.5
29.77
43.71
45.9
31 .»7
28.10
34.6
31.?
40.3
o: 52
.: 56
pi J6
42.9
34.13
29.06.
38.92
45. H
.25
.42
,95
*
(continued)
-------
fABLE A-l (continued)
£F& Chtmieal
K4thinol 0154 CH30H
tothyl «thyl k*t<5c« U159 CHjCOO^CH)
Ktth
| Thrieh lor OTwoofluor cx*et htna 0121 ^Ij^
N3
j&%
t»
Xyttae 1239 CgBjCCBj)^
mT«K
(|/K) p>Ut CO {_ H, f 2S*O*
32 45 113 .»
72.1 71.6 I0«
100 117-119 7.5
84 93 40 345^
500»
123 210.8 0.209
79 115.5 20
165.83 121 14^0
92, IS 110.6 21.7
131 .39 *7
90
137 23.7 768
796
106 o! 144.4 os 2.77
*: 139 «: 3.20
f. 138.4 p: 3.15
SoluHUty
Miicill. 1,53
2.70xW5 5i
I7,000ao 794
95.06
1,900 2,958
Xiieible 1,622
20020 382
534.8 11,100
114
WOO20 3«2
1,100 1
175 15, 1U
(1 20/4«C)»
0.796'5
O.S05
0.801
1.326
1.20»
0.952
1.623
0.867
1.464
1.56
o: 0.880
•: 0.871
p: 0.861
dt. 7 Mr1)
l.l«10'f
2.7I10"6
4.35xir5
5.41x10-*
3.19xlB"3
2.4xlO-s
l!3xlO-5
7.10-*
2.36x10-5
0.0153
6.66x10-3
9.U10"1
11.7XIO"1
5.83xlO-J
5 27x10-'
a: 2.5SxlQ"3
CM!! let tot
0,15
O.f*
1.82
Itt
It. I
17.8
70.8
75.8
4,36
10.96
759, 400
620, 537
263, W4
3)9
o: 588.8
•: 1584.9
f. 1412,5
IMC »! fle>fc B«t ol
ee«&«titioR polite ¥c»«rU«ciea
(KI/Ml> ( Ji*O« (*«*« (U/K>1* 1 3$*C)»
726 12 (CC)
2,444 1.7
XA 20
604.6 Hem
3,092 88 (CC)
Z.782 75 (CC)
825 Iteo-
3,508 4
956 Hose
HA Hone
0! 4567 os 11 (CO
»: 4553 «: 25 (CC)
P: 4556 p: 25 (CC)
39.23
37. JI
34.10
KA
31,7
50.91
40.37
38.66
39.2
35.9
34,79
26.88
o: 41.83
.: 41.44
p: 41.04
(continued)
-------
TABLE A-l (continued)
Conatituent
Other Solvent*
1,1,1 ,2-f etraehloroethane
1,1,2,2-tetrachloroethanB
1 , 1 ,2-Trichloreethane
1,1-Dichloroethylene
1,2-DiehloroetnyUne,
trnna iooaer
l.^-Dichloropropane
1.3-DicU«ro*ropu«
(ei« and1 Erans Isoaera)
1 ,4"Diehl0«o-2-tmtflne
^ 1 ,4-Diox*ne
1^2 2™Sthosty ethaaol
m
2-»itropr0(,,ne
2-Picolint
Aefltoaitrile
Aniline
Benzene
E? A Cheaieal
code fonmlla
0208 C2H2d4
d209 C2H2d4
U227 Cj»jCl3
U078 CH2CC12
t!079 C2H2C12
U083 CjHgC^
U084 C3H4C12
0074 C4H4C12
11108 C4H802
U359 BO(CH2>20C2H5
U171 C)H7HOj
U191 051(411013
W03 CHjCS
11012 Cgi!^nl2
U013 C6H6
(g/H)
167,85
167.85
133.41
97
96.94
112.99
110.97
125
88
90.12
»9
93
41
93
78.12
Boiling
point (*C)
130
138
146.2
113,8
31.9
47.5
96.4
104.3 (cU)
112 (tr.no)
154, 15»
101
135
120
128.8
81.6
184
80.1
Vapor
(•> Kg 8 25"O*
13.9
5.9
9
24.8
630
600
318
49.6
2S20
4.0
37
38»
,30
17.5
10
100
0,85
0.75
74
Solubility
taj/t 8 2S*C>*
20020
2.90020
4.50020
l.OOO25
3,200
440
6002°
2,7002»
2,700-2,800
9.701
6»107
Soluble
17
51,000
2,2x10*
8.4jclQ^
34,000
1780
coefficient
381
382
382
382
382
HA
HA
NA
96.6
HA
HA
2,181
10.89
94
2,080
Specific
(S 20/4-O*
1.J4
1.59
1.44
1.218
1.256
1.156
1.217 (cU)
1.224 (trana)
1.1S3«
1.033
0.931
.992
0.9515
0.79
1.02
0.876
Henry's Law
(atm «r iaol"*)
0.01 1
j.fciir*
7.42»10-*
l.SxlO"*
U9l*10->
5.32.10-S
1.31,10-J
1.33.10"3
6.78X10-5
7.14»10-'
M
0.121
2.4«10-5
tlrtJS'
3.07x10"*
2,6x10-*
5.5*10-3
Octanol-vater
partition
coef Eicient
1100
245, 363
117, 148
HA
3.4,102
105, 191
100, 95.5
HA
0.38
0.288
NA
11.48
0.46
7.94
9.55
135
Hut of
combustion
(KJ/.ole 8 J5'«*
976
976
1110
HA
1217
1886
1597
HA
J431
HA
1999
HA
1,265
3,396
3,267
Fl»ih Belt of
point vaporization
(*C)** (KJ/mole ( 2S*C>*
Hone
Hone
Hone
-17.8 (OC)
2
4, 16
27,35 (l>«l!
iaomers)
HA
5 to 18
44 (CO
49 (OC)
24
38.9
12.8
76 (CO
-11
38.9
36,5
41.49
38.3
30,17
30.30
35.24
KA
HA
35.65
HA
39,65
41.56
34.20
47.11
34,085
42.503
(continued)
-------
TABLE A-l (continued)
HeltcttUr
If* eh«*!e«l vtlsbt loiliat
lit (chlKraMtlijrl) 'Oat 1016 CjUsCljO 114.96 104
Bro«>fom V22) CHIrj 253.3! 149.S
Chloroform *044 CHCij 119 61.7
Cjeleh«»«ii» UOJ* CjHjj 14 11
Diehlorodifluprooethane 11075 CCljfj 120.92 -29*S '
Ethyl e«rb«Mt« 1)238 CsfyHOj 89.11 184
EthyUat iiehloride W)77 CjH^Clj 98.96 83.5
tthylideat dichloridt U076 C2B4C12 98.96 57.3
Fur«o B124 04840 68 31.36
1
NS Fur6««l U125 C4HjOCKO 96 161.7
CTi
Bextehloro«chai)t U131 C2Clg 236.74 186 0
777 _ IK
riopylcae gljreol PlOO CHjOOBCHjOU ?6.1 188.2
' **°
o-dichlorobftneeni) 0070 C^H^lj 14? . 01 180. 5
V.for telukllltj
ft*M«r« fit v«t*r Aetitit?
(— 8t ( J5'c)« (if/I, ( :S'C)« cMdlcltet
30»
5.6
16020
200
120"
4, no
0.36
80.4
234
634
596
S3
0.4"
0.22°
149
71
1.5
11,000
3,190
7,100
55
210
9.6x10*
2«10°
8,7M
1,000
10,000
13,000
5022
Hi.ciblc
ilsi.io5
145
7.»
HA
,S
MA
HA
KA
382
312
SA
HA
382
HA
HA
52,500
gravlcy C99 Hi'1) CMlficlrat
1.J15
2.89
1.483
0.779
1.329
0.986
1.235
1.175
0.937
1.16
•2.82
1.038
0.111
1.305
2.WO-*
5.32I10T*
3.39*10"'
0.178
2.75
0.40
2tlO"8
9.14«10"*
4.26I10'3
5.7«10-'
3.6»W6
8.10.10-5
2.4hlO-3
HA
i.os.ur4
*.9ilO"s
1.93,10-3
2.4
199.5
93
M
KA
HA
30
63
NA
M
4.2*10*
2.4»10S
0.045
KA
3.6xl03
2.3»103
B««t at rlnh lU.t at
cMbultln faint v.pcrlt.ileo
(U/Ml* ( 25"C)» {*«*« (rj/«l. ( a$*C}»
9*7
138
J73
3,919
KA
HA
1,411
1,242
2,0*2
2,340
460
1,803
2S02
2,810
31
•Xetw
Kim*
-18 (CC)
KA
HA
13
-6
-M (CC)
60 (CO
6t (CC)
Hone
99 (OC)
-17.2
66.1
M
40.47
31.38
92,76
26.JO
HA
33.3
30.2
27.17
KA
49.0
56.80
HA
45.79
(continued)
-------
TABLE A-l (continued)
Constituent
Ijraitsblea
l~Kethylbut«diene '
2,2'~Eioxlr*ne
2~«ettiyl«iridii»
Acrolain
Acrylic acid
Allyl alcohol
Allyl chloride
Chloroaeetaloehyae
Chldroaetnyl ststhyl ether
Chloro&re&o
Cuaeae
Dimthylajniae
DlprapyUaine
Bpichlorohydria
Echanal
Ethyl aerylate
IthyUae diamine
EPA Oieaicil
code formula
1)186 C5hl8
1)085 CHjtOHjOz
P067 CjHjN
P003 CHjCHCHO
1)008 CHjCIICOOH
POOS CH2CHCH20H
Z010 CH2CRC«2CI
P023 CjMjClO
U046 CHjOOtiCl
X009 CH2CCl«iC«2
1)055 Cg»12
0092 (0)3)2X1,
DUO (C3H7)2l™
0041 CjHjClO
U001 CtjCHO
0113 CH]CRCOOC2Bj
P053 H»2C«2CB2'M2
(ijp>
Holecul«r
weight
(8/H)
68
86.09
57.11
56
72.06
58.09
76.53
78. SO
81
88.S4
120.19
45.08
101.2
92.53
44.1
100.11
78.1 hydr<
60.1
•sltydrout
Boiling
point (*C) d
42
144
20
53
141
96-97
44-45
86
59.5
59.4
152.7
7.4
109-111
116.5
20.2
100
ite 118 hydrate
Vapor
m H§ | 25*c)*
414
7.5
92.0
258
3.2"
28.1
J4020
44030
lOOJO
214
115.4
4.6
1,500
30
1220
223"
916
40
lift10 nhjrdmw
i20 kyonte
Solubility
Cog/L g 25'O*
870
B.lxlO7
3.1x10*
280,000
3.5X105
(2,000
100
Soluble
2.5x10*
Slifhely
soluble
50»»
Very coluble
12,000
66,000
60.00020
S.oxlO5
20,000
15,000
coefficient
8A
HA
HA
HA
10.7
HA
HA
4.2
DA
MA
45,082
1
KA
HA
15.6
414
HA
Specific
jrwfity
(S 20/4*c»»
NA
KA
KA
0.842
1.0616
0.854
0.938
1.19"
1,061
0.958
0.862
0.68°
0,738
1.180
0.78J18
0.924
0.963 hydretia
0.»99*
anhjrdroul
Benry'a tatf
constant
(atm B3 sol"1)
0.0424
1.02xlO"8
2.22xlO"5
6.7»xlO-5
1x10"'
3.47x10-*
DA
10-3
9.12x10"*
M
0.0146
S.83xlO"4
3.32xlO"4
3.13x10-5
9.5x10-5
3.5xUT4
HA
Octeaol-vater
paftitioa
coefficient
HA
KA.
KA
HA
2.04-2.69
HA
HA
0.42
M
HA
4,571
0.42, 0.95
53.7
0.42
2.69
HA
HA
Heat of
coaboation
C«/«ole S 25*«*
NA
HA
KA
1,630
1,368
1,851
HA
959
KA
HA
5,2U
1,743
HA
1,770
1,1(6
2,742
1,893
FUah Heit of
point vapori*at£ot>
<•«•* (lU/nole * 25*CJ"
-42,1
M
-10
-18 (OC)
68 (OC)
21 (CC)
-11
-------
TABLE A-l (continued)
Coaitfturat
letyltaivU*
ElbylMtlucxylct*
Fomildehydt
GlyeUyUlMiyte
HttluciylMtarite
Methyl turcmide
Methyl chloride
Methyl cHlarQcur&emtMi
Hathyl BMSthacryUtc
Oxircne
Fstftl4eh?c3e
lhioneth«nGl
a-fcopyl*Bine
tfA
»54
gin
U122
U126
nisi
UOZ9
U045
ms6
i)162
U115
UIB2
UIS3
UI94
fomulu
W
XA
«20
CHOCKoctij
CH2C(CH3)CH
CBj»r
0301
W»2
CMjCWKjJCOOCH;
CjB<,0
56«I203
CH3SH
CHjCH2a,jraj
HolicuUr
"(t/M)
41,07
U*
30.03
72
67,09
»S
50.49
94.50
} 100.11
44.05
132,16
48.11
59. U
loillat
point CO
56
117
-21
-79.SJ20.. HS
97
90.3
3.56
-24.2
Jl
101
U
128
6.2
49
V.por
tmst t 25'O*
200
1*
760-»'-5
42.6
65
5.3M
3.80028
5.09230
113
40
1,294
25.3»»
1520
245"
im ««t«r
(mill, t S5'C)»
HUclkU
19,000
Hliclblc
Very coluble
U
900
6,27«103
Slightly
•Oluble
7,«riO*
Very aoluble
120,000"
23,200
Hiiciblt
to*Klelt>t
XA
XA
1,46
KA
144
HA
95,06
NA
161
HA
XA
NA
NA
Sfteltle itoarjf't l»
0.832 1.1*10-3
XA l,4SxlO-*
0.815 S.lilO"4
HA 5.8«10-'
0,800 0.392
1.730° 5.26K10"3
0,916 0.04
HA NA
0,936 «,6ilO"5
0.887 3.63xW5
0.994 3.7isl05
0.868 4*10-3
0.718 2,0»10"5
ptrtieltm
ee«fflc{«at
XA
•A
0.13
KA
XA
HA
8.*
m
NA
HA
2
HA
l.«l, 2.J4
Hut o(
-------
APPENDIX B
MANUFACTURER PROFILES:
SOLVENT DISTILLATION EQUIPMENT
Source: Naval Energy and Environmental Support Activity, NEESA 20.3-013.
Assessment of Solvent Distillation Equipment, Appendix A.
December 1985.
B-l
-------
MANUFACTURER PROFILES
(l) Alternative Resource Management Inc.
P. 0. Box 1265
Miami, Oklahoma 74355
(918) 540-2511
Still Type: Continuous feed
Solvents Designed For: Explosion-proofing and vacuum attachments are
optional. Therefore, depending upon the still obtained, any solvent (flammable
or nonflammable) with a boiling point up to 500°F can be distilled. Standard
models operate at atmospheric pressure and are not explosion-proof and,
therefore, can distill nonflammable (halogenated) solvents with boiling points
up to approximately 350°F.
Water Separator: Optional
Thruput/Capacity; ASM can build stills with thruputs ranging from 5 to 200 GPH.
Materials of Construction: Stainless steel construction is available as an
option.
Safety Features: Vapor temperature and boiler temperature thermostats are
standard. Liquid level controls and an automatic feed pump is also standard.
A condenser flow switch turns the unit off if there is a lack of cooling water.
Heating and Cooling Options: Standard models are heated by a steam jacket.
Electrical heating is optional.
Utility Requirements: 440-480V 3-phase power is required as well as cooling
water, drainage, and compressed air to run pumps.
Available Models: Specifications and base prices for ARM's standard solvent
stills are listed in Table A-l. Explosion proofing, vacuum attachments, water
separators, and stainless steel construction are all optional and will raise
the prices accordingly. Stills larger than those listed are available, but
must be custom designed and built.
B-2
-------
TABLE A-l
A.R.M. STILLS
MODEL THRPFUT PRICE
5 5 GPH t 4,800
15 15 GPH $10,200
25 25 GPH $24,600
50 50 GPH $43,850
B-3
-------
(2) Baron-Blakeslee, Inc.
2001 North Janice Avenue
Melrose Park, Illinois 60160.
(312) 450-3900
Still Type: Continuous feed
Solvents Designed For: The stills are not explosion proof and they operate at
atmospheric pressure (no vacuum available). Therefore, they can distill only
nonflammable (halogenated) solvents with a boiling point below approximately
350°P. Units are designed specifically for either chlorinated solvents or
"Freons",
Water Separator: Standard
Thruput/Capacity: The stills are generally large (60-300 GPH for
chlorinateds, 10-1200 GPH for "Freons").
Materials of Construction: Full stainless steel construction is standard.
Safety Features: A high vapor temperature thermostat is standard. A high
boiler temperature thermostat is optional and should be obtained. High and
low liquid level controls are optional, as is the automatic feed pumping
system. These items must be purchased in order for the still to operate
safely as a continuous feed unit.
Heating and Cooling Options: Electric or steam heated units are available.
Cooling is normally done by water, but a refrigeration unit is available; on
some of the models designed for "Freons".
Utility Requirements: 230V 3 phase or 230V single phase for electrically
heated models, pressurized steam for steam heated models, cooling water, and
drainage.
Remarks:" Filtration systems and dessicant dryers are available if high purity
solvents are required.
Available Models: Some specifications for their smaller still? are given in
Table A-2.
B-4
-------
TABLE A-2
BARON-BLAKESLEE SOLVENT STILLS
MODEL
NRS-60
MRW-20
MRW-60
MRR-10
MRR-20
MRR-60
SOLVENTS
Chlorinated
Freons
Freons
Freons
Freons
Freons
THRUPUT
60 GPH
20 GPH
60 GPH
10 GPH
20 GPH
60 GPH
COOLING
Water
Water
Water
Re frig.
Refrig.
Re frig.
L" x W" x H"
64 x 44 x 81
47 x 36 x 71
56 x 43 x 81
35 x 30 x 61
40 x 43 x 73
75 x 52 x 97
PRICE
S 6,915
$ 5,200
fe 5,710
£ 5,250
£ 8,380
til, 940
B-5
-------
(3) Branson Cleaning Equipment Company
Parrott Drive
Shel*on, Connecticut 06484-9987
(203) 929-7301
Still Type: Continuous feed units. Designed primarily for integral use with
Branson vapor degreasers, but can be used independently as well.
Solvents Designed For: The stills are not explosion proof and they operate at
atmospheric pressure (no vacuum available). Therefore they can distill only
nonflammable (halogenated) solvents with a boiling point lower than
approximately 350°F.
Water Separator: Standard
Thruput/Capacity" Branson stills will distill waste solvents at thruputs from
8 to 96 GPH depending on the model and the solvent being distilled.
Materials of Construction: Full stainless steel construction is standard.
Safety Features: High vapor temperature and high boiler temperature
thermostats are standard on all models. A low liquid level control is also
standard. A feed pump and a high liquid level control are optional and must
be purchased in order to operate the still safely as a continuous feed unit.
Beating and Cooling Options: Electric or steam heated units are available.
Cooling is normally done by water, but refrigeration units are available if
desired.
Utility Requirements: Depending on the heating and cooling options chosen,
the following many be needed: 230V 3 phase (electrically heated models), 230V
single phase (steam heated models), pressurized steam, water, and drainage.
Remarks: 4 dessleant dryer is available if high purity "Freons" are required.
Available Models: Specifications and prices for some of Branson's smaller
stills given in Table A—3,
B-6
-------
TABLE A-3
BRANSON SOLVENT STILLS
THRUPUT (GPH)
MODEL TCI FREON TF
SHOW
S111W 9.3 14.9
S120W
S121W 21.3 31.3
S110R
S111R 9.3 14.9
S120R
S121R 21.3 31.3
FREON TMC COOLING
7.8 Water
Water
16.1 Water
Water
7.8 Refrig.
Refrig.
16.1 Refrig.
Refrig.
L" x W" x H"
26 x 22 x 51
26 x 22 x 51
26 x 22 x 51
26 x 22 x 51
26 x 42 x 51
26 x 42 x 51
26 x 42 x 51
26 x 42 x 51
PRICE
$3,025
$3,125
$3,520
$3,595
$5,635
$5,938
$7,435
$7,710
B-7
-------
(4) Brighton Corporation
11861 Hosteller Road
Cincinnati, Ohio 45241
(513) 771-2300
Still Type: Continuous feed units. They make two basic stills. The first
type has a flat-bottomed boiler with manual cleanout. The second type has a
cone—shaped boiler bottom with scraper bars that continuously clean the boiler
and discharge the sludge automatically.
Solvents Designed For: The stills are explosion-proof and are equipped with
vacuum attachments. Therefore, virtually any commonly-used organic solvent
with a. boiling point up to 500°F can be distilled. The scraped cone bottom
boilers are recommended for high solids applications.
Water Separator: Optional ($440)
Thruput/Capacity: Brighton stills range in thruput from 7.5 to 200 GPH.
Materials of Construction: Standard material is primarily carbon steel
throughout, except for copper condenser tubes. Stainless steel construction
is available as an option (see Section 4.5).
Safety Features: Boiler temperature is controlled by regulating the amount of
steam entering the jacket that surrounds the boiler. Boiler temperature must
be monitored with the thermometer mounted on the boiler. The stills contain
level controls for continuous operation.
Heating and Cooling Options: Heating is normally performed by steam using a
jacket that surrounds the boiler. However, for solvents with an atmospheric
boiling point over 350°F, a heat transfer fluid (oil) must be used. This
requires the purchase of an optional electric fluid heating system. Condenser
cooling is performed by water.
Utility Requirements: Depending on the heating and cooling options used, the
following may be required: electricity, steam (100 psig), water, and drainage.
Available Models: Base prices for their smaller standard models are given in
Table A-4.
B-8
-------
TABLE A-4
BRIGHTON SOLVENT STILLS
THRUPUT FLAT SCRAPED CONE
(GPH) BOTTOM BOTTOM
7.5 $18,230 $22,830
25 $22,870 $29,690
50 $28,950 $37,455
B-9
-------
(5) Cardinal Corporation
P. 0. Box 4234
Wilmington, DE 19807
(302) 656-9446
Still Type: Batch. Cardinal stills are unique in that the collection drum is
the afcill pot. Waste solvent is collected in a 55-gallon steel drum. When
full, the drum is placed inside the unit and an electric heater is clamped
around the bottom.
Solvents Designed For: The stills operate at atmospheric pressure. They are
available with or without explosion proofing. Therefore, depending upon which
still is obtained, any solvent (flammable or nonflammable) with a boiling
point lower than approximately 350"F can be processed.
Water Separator: Not available.
Thruput/Capacity: There are two basic models. One model (-10) can process one
drum of spent solvent in an 8-hour shift, the other model (-20) can process
two drums simultaneously in 8 hours, or can be run as a one drum unit.
Materials of Construction: Since the collection drum doubles as the boiler,
stainless steel construction is not possible. However, since this drum is
disposed of, this is not a concern.
Safety Features: A thermostat automatically turns the unit off when the vapor
temperature reaches a preset limit.
Heating and Cooling Options: The stills are heated electrically and cooled by
refrigeration units.
Utility Requirements: 240V, 3-phase electrical service.
Remarks: Explosion-proofing is Class I, Group D, Division 2 (See Section 4.1.1).
Available Models: Specifications and prices for Cardinal's solvent stills are
listed in Table A-5.
B-10
-------
TABLE A-5
CARDINAL SOLVENT STILLS
DRUMS EXPLOSION
MODEL
MC-10
AC- 10
MC-20
AC-20
PER SHIFT
1
1
2
2
PROOF
NO
YES
NO
YES
PRICE
$16,300
*16,400
117,700
£17,900
B-ll
-------
(6) BCI Corporation
5752 West 79th Street
Indi-anapolis, Indiana 46268
(317) 872-6743
Still "Type: DCI stills are either direct steam injection units or hybrid units
that can be operated as either a direct steam injection or a steam jacket
system. The stills operate on a batch basis, but are filled, turned on, and
turned off automatically. This results in "semi-continuous" operation. Still
bottoms are discharged automatically as well.
Solvents Designed For: The stills are explosion-proof. Direct steam injection
can be used to distill virtually any immiscible solvent (see section 3.3 for a
full discussion of direct steam injection). In addition, the hybrid units can
be switched to convection (steam jacket) heating to distill miscible solvents.
Therefore, DCI units can be used to distill virtually all the commonly used
organic solvents.
Water Separator: Standard on all models.
Thruput/Capacity: Depending on the model chosen, DCI stills can operate at
thruputs from 10 to 1,000 GPH.
Materials of Construction: Full stainless steel construction is standard.
Safety Features: A boiler temperature thermostat turns the still off
automatically and also triggers the automatic discharge of the still bottoms.
The stills also have overflow protection (level controls) in the boiler and in
the condenser.
Heating and Cooling Options: As mentioned before, heating is performed by
either direct steam injection or by steam jacket (hybrid units). Cooling is
performed by water.
Utility Requirements: Depending on the model purchased, the following may be
required: 110V, 60 cycle power ("DG" units), 220V 3 phase power ("Dl" units),
steam, compressed air (80 psi for all models), water, drainage.
Remarks: The smaller stills are available with carbon steel construction at
lower cost (see Section 4.5).
Available Models: Some specifications and base prices for DCI's smaller units
are given in Table A-6.
B-12
-------
TABLE A-6
DCI SOLVENT STILLS
THRUPUT EXPLOSION NON-EXPLOSION
MODEL (GPH) L" K W" PROOF PROOF
DIRECT STEAM INJECTION ONLY;
DG-10-SST 10 58 x 43 $18,840 $16,275
DG-25-SST 25 58 x 43 $29,395 $25,545
D1-50-SST 50 54 x 56 $33,745 $29,240
HYBRID UNITS:
DG-10-SST-HY 10 58 x 43 $22,230 $20,665
DG-25-SST-HY 25 58 x 43 $34,745 $30,895
D1-50-SST-HY 50 54 x 56 $43,600 $39,090
B-13
-------
(7) Diafci Incorporated
131 Prince Street
New York, New York 10012
(212) 505-0677
Still Type: Batch, filled through top cover.
Solvents Designed For: Disti stills are explosion proof and are available with
vacuum attachments. As a result, virtually all commonly-used organic solvents
can be distilled.
Water Separator: Optional
Thruput/Capacity: Boiler capacities (batch sizes) vary from 7 gallons to
1,000 gallons. Thruputa vary from 3 GPH to 300 GPH.
Materials of Construction: Standard construction is with mild steel, but
stainless steel is available as an option (see Section 4.5).
Safety Features: A double safety thermostat shuts the unit down when the
boiler temperature reaches a pre-set limit. This feature is used as an
automatic turn-off system.
A flow switch ensures that the stills will not operate unless cooling water is
flowing through the condenser.
Heating and Cooling Options: Heating is performed by an electrically heated
heat transfer fluid (hot oil) jacket. Cooling is performed by water.
Utility Requirements: The units discussed here require three phase AC power
of any voltage* Cooling water and drainage are also required.
Remarks: Disti stills are manufactured in West Germany by a company named
D. W. Renzmann.
Available Models: Specifications and prices for some of Disti1s stills are
listed in Table A-7.
B-14
-------
TABLE A-7
DISTI SOLVENT STILLS
MODEL
DW-25
DW-50
DW-100
M20N
CAPACITY
(GAL)
7
.5
30
60
THRUPUT
(GPH)
3-5
5-8
8-16
10-20
L" x W" x H"
37 x 26 x 35
51 x 27 x 47
55 x 28 x 50
55 x 53 x 65
BASE
PRICE
$ 5,700
$ 9,850
$13,800
$17,850
Options For "DW" Series Skills:
Vacuum unit - $2,500
Water separator - $1,650
Combination vacuum/water separator - $3,300
Stainless steel construction - $3,900
Options For Model MN20:
Vacuum unit - $2,890
Water separator - $1,650
Combination vacuum/water separator - $3,690
Stainless steel construction - $5,600
B-15
-------
(8) Finish Engineering Company
Erie, Pennsylvania
Government Distributor:
Extratec Corporation
18220 Allwood Terrace
P.O. Box 533
Olney, Maryland 20832
(301) 924-5150
Still Type: Batch. Filled through top cover.
Solvents Designed For: Finish stills are explosion proof and are available
with vacuum units. Therefore, they can be used for recycling any solvent
(flammable or nonflammable) with a boiling point up to 500°F (virtually all
organic solvents).
Water Separator: Not available.
Thruput/Capacity: The stills are basically designed to process one batch per
shift. Batch sizes range from 15 to 500 gallons. The two most popular still
sizes are 15 and 55 gallons.
Materials of Construction: The boilers are Teflon-coated, stainless steel,
which means that still bottoms will not stick to the sides or the bottom. Ml
other parts are stainless steel.
Safety Features: A thermostat shuts the unit off when the boiler reaches a
prc-set temperature. Redundant temperature controllers disallow any single
spot form reaching 365°F.
Heating and Cooling Options: Most Finish stills are electrically heated. The
55-gallon capacity still is available with steam heat (a portable boiler
package is available). Cooling is performed with water.
Utility Requirements: Depending upon the still obtained, the following may be
required: 110V power, 220V power, steam, cooling water, and drainage.
Remarks: A small steam boiler system is available if steam heating is desired
but in-house steam is not available.
Available Models: Finish Engineering currently has a GSA contract for it's
solvent stills. The stills are covered under FSC Class 6640 and contract
number GS-OOF-79500, which expires 1 August 1986. Special item numbers,
specifications, and GSA prices for Finish Engineering's "LSH series stills are
given in Table A—8.
B-16
-------
TABLE A-8
FINISH ENGINEERING STILLS
SPECIAL
ITEM NO.
0517
7011
7257
7259
7260
9831
9831
MODEL
LS-jr
LS-15
LS-15V
w/Vacuira
LS-55
LS-55V
w/Vacuum
LS-55-ST
LS-55-ST
w/Vacuura
CAPACITY
(GAL)
3-5
15
15
55
55
55
55
HEATING
Elec
Elec.
Elec.
Elec.
Elec.
Steam
Steam
L"
30
30
34
34
34
34
x W"
x 44
x 66
x 56
x 78
x 56
x 78
x H"
x 36
x 39
x 69
x 60
x 65
x 65
PRICE
4 2,845
1 5,344
1 9,305
$13,252
419,047
118,895
* 24, 690
B-17
-------
(9) Hoyt Corp.
Forge Bead
Westport, Massachusetts 02790
(800) 343-9411
in Ma. (617) 636-8811
Still Type: Continuous feed stills. Boilers have scraper-agitator bars for
removing solids from the boiler walls. Still bottoms are discharged
automatically.
Solvents Designed For: Explosion-proofing is available. The stills operate
at atmospheric pressure (no vacuum available). Therefore, any organic solvent
(flanmable or nonflammable) with a boiling point lower than approximately 350°F
can be distilled.
Water Separator: Optional.
Thruput/Capacity: Hoyt Solvo-Salvagers will distill solvents at a rate of 4
to 8 GPH.
Materials of Construction: Full stainless steel construction is standard.
Safety Features: A high boiler temperature thermostat is standard. Liquid
level controls are also standard and the unit will automatically shut down for
lack of feed solvent. The still will not operate without the condenser cooling
water flowing or with a low thermal oil (heating medium) level.
Heating and Cooling Options: Heating is performed by a heat transfer fluid
(hot oil) jacket. The oil is heated electrically. Cooling is performed by
water.
Utility Requirements: 220V electricity, 2 gallons per minute of cooling water,
and drainage are needed.
Available Models: The Solvo-Salvager is available in one basic model.
B-18
-------
TABLE A-9
HOYT CORPORATION STILLS
MODEL THRUPUT L" x W" x H" PRICE
IPS 4-86PH 30 x 30 x 92 $14,500
B-19
-------
(10) Lenepe Equipment Co.
P. 0. Box 285
Manasquan, New Jersey 08736 -
(201) 681-2442
Still Type: Continuous Feed.
Solvents Designed For: The stills are not explosion proof and they operate at
atmospheric pressure. Therefore, they can be used to distill nonflammable
(halogenated) solvents with a boiling point below approximately 350°F.
Water Separators: Standard.
Thruput/Capacity: Depending upon the unit obtained, Lenepe stills can recycle
solvent at rates from 1.5 GPH to 150 GPH.
Materials of Construction: Full stainless steel construction is standard*
Safety Features: High vapor temperature and high boiler temperature
thermostats are standard. A low liquid level control switch is also standard.
An automatic solvent level control and a feed pumping system are optional and
should be obtained in order to safely operate the stills as continuous feed
units.
Heating and Cooling Options: Heating on all models is electric. Cooling is
performed by either water or by refrigeration units.
Utility Requirements: Electricity is required, as well as, water and drainage
(water cooled models).
Remarks: A dessicant dryer assembly is available as an option for situations
when extraction of solvent components by water may occur.
Available Models: Specifications and base prices for some of Lenepe*a smaller
models are listed in Table A-10.
B-20
-------
TABLE A-10
LENEPE SOLVENT STILLS
THRUPUT
MODEL (GPH) COOLING L" x W" x H" PRICE
AW2 3-10 Water 20 x 26 x 38 $3,495
AR2 3-10 Refrig. 20 x 41 x 38 *4,945
BW6 14-40 Water 35 x 26 x 49 16,935
BR6 14-40 Refrig. 35 x 58 x 49 *9,68S
B-21
-------
(11) Phillips Manufacturing Co.
7334 North Clark Street
Chicago, Illinois 60626
(312) 338-6200
Still Type: Continuous feed.
Solvents Designed For: The stills are not explosion-proof and they operate at
atmospheric pressure (no vacuum attachment is available). Therefore, Phillips
stills are designed to recycle nonflammable (halogenated) solvents with a
boiling point below approximately 350°F.
Water Separation: Standard.
Thruput/Capacity: Phillips stills vary in thruput from 1 to 1,000 GPH.
Materials of Construction: Standard stills are manufactured with galvanized
mild steel. Stainless steel construction is optional.
Safety Features: All models are equipped with automatic level controls.
Electric models are equipped with a boiler temperature thermostat.
Heating and Cooling Options: Phillips' stills are heated by steam,
electricity,
or gas. Cooling is normally performed by water, but refrigeration units are
available as an option.
Utility Requirements: Depending on the model obtained, the following may be
required: electricity, steam, gas, water', and drainage.
Available Models: Specifications and prices for some of Phillips' smaller
stills are listed in Table A-ll. The prices listed include the extra cost for
stainless steel construction.
B-22
-------
TABLE A-ll
PHILLIPS SOLVENT STILLS
MODEL (GPH) HEATING L" x W" x H" PRICE
RS-3 3 Elec. 26 x 21 x 49 $2,121
RS-10 10 Steam 30 x 28 x 65 $4,753
RS-10 10 Elec. 30 x 28 x 65 $4,973
RS-10 10 Gas 30 x 28 x 65 $5,341
RS-50 50 Steam 48 x 40 x 84 $6,378
RS-50 50 Elec. 48 x 40 x 84 $7,822
RS-50 50 Gas 48 x 40 x 84 $8,202
B-23
-------
(12) Progressive Recovery Inc.
1976 Congressional Drive
St. Louis, Missouri 63146
(314) 567-7963
Still Type: Batch. "SC" models have flat bottomed boilers. "LSR" models have
a. conical shaped bottom that is continually scraped for high solids
applications.
Solvents Designed For: PRI stills are explosion-proof, and vacuum assist is
available on all models except the SCjr. Therefore, the stills are capable of
recycling any solvent with a boiling point up to 500°F (virtually all organic
solvents).
Water Separator: Optional.
Thruput/Capacity: FRI stills are available with capacities from 5 to 710
gallons. Thruputs range from 1 to 260 GPH.
Materials of Construction: Full stainless steel construction is standard.
Safety Features: The electrical heating elements are controlled by a double
safety thermostat. A flow switch ensures that the unit will not operate for
lack of cooling water. A vapor temperature thermostat in the condenser
provides backup protection to this switch.
Heating and Cooling Options: Heating is performed by an electrically heated
heat transfer fluid (hot oil) jacket or a steam jacket. The stills are cooled
by water except for the SCjr model, which.is air cooled.
Utility Requirements: Either 220V power, 440V power, or steam is required,
along with water and drainage.
Available Models: Specifications and base prices for some of PRI's smaller
stills are listed in Table A-12.
B-24
-------
TABLE A-12
PRI SOLVENT STILLS
MODEL
SCjr
SC50
SC100
SO 200
LSR-8.SE
LSR-8.5
CAPACITY
(GAL)
5
35
35
60
55
55
THRUPUT
(GPH)
1-2
4-6
7-9
12-17
10-12
12-15
HEATING
Electric
Electric
Electric
Electric
Electric
Stean
PRICE
$ 4,745
* 9,675
*11,895
*17,195
*44,965
$44,965
The cost of a vacuum assist unit for any of PRI's stills is
approximately $3,000.
B-25
-------
(13) Randall Manufacturing Company Inc. (RAMCO)
32 Montgomery Street
Hilla-ide, New Jersey 07205
(201) 687-6700
Still Type: Continuous feed.
Solvents Designed For: The stills are not explosion-proof and they operate at
atmospheric pressure (no vacuum is available). Therefore, they are designed to
distill nonflammable (specifically halogenated) solvents with a boiling point
lower than approximately 350°F.
Water Separator: Standard.
Thruput/Capacity: Ramco stills are available with thruputs ranging from 25 to
200 GPH.
Materials of Construction: Full stainless steel construction is standard.
'Safety Features: Standard safety features include a high vapor temperature
thermostat and a high boiler temperature thermostat. A flow sensing switch
shuts the unit down if water is not flowing through the condenser. The
boiling chamber has a level control that shuts the unit down if the solvent
level is too low.
Heating and Cooling Options: Ramco stills are available in either electric or
steam heated models. Cooling is performed by water*.
Utility Requirements: Depending on the heating option chosen, the stills may
require electricity or steam, along with cooling water and drainage.
Available Models: Specifications and base prices for some of Ramco1a smaller
stills are given in Table A-13.
B-26
-------
TABLE A-13
RAMCO SOLVENT STILLS
THRUPUT
MODEL (GPH) HEATING L" x W" x H" PRICE
R25 25 Steam 66 x 46 x 78 $6,975
R25 25 Electric 66 x 46 x 78 $7,496
R50 50 Steam 82 x 52 x 78 $7,719
R50 50 Electric 82 x 52 x 78 $8,547
B-27
-------
(14) Recyclene Products Inc.
1910 Trade Zone Blvd.
San Jose, California 95131
(408) 945-8600
Still Types: Batch. Disposable bags are used to line the boiler during
operation. When the batch is complete, the bag (which now contains the still
bottoms) is disposed. An automatic feed system is available to pump the
solvent into the boiler.
Solvents Designed For: Recyclene stills are explosion proof. They operate at
atmospheric pressure (no vacuum attachment is available). Therefore, the
stills can be used to recycle any solvent (flammable or nonflammable) with a
boiling point lower than approximately 350"?.
Water Separator: Not available,
Thruput/Capacity: Recyclene stills are available with boiler capacities of
either 20 gallons or 35 gallons. Thruput range from 0.5 to 35 GPH.
Materials of Construction: Stainless steel construction is standard except
for the condenser, which is made of a copper-nickel alloy.
Safety Features: A double safety thermostat shuts the unit off automatically
when the boiler temperature reaches a preset limit. A condenser overheat
thermostat shuts the still down if the cooling water is not running. A safety
interlock prevents the opening of the boiler lid while the system is running
or is hot. A high liquid level control prevents over-filling of the boiler if
the auto-feed option is chosen.
Heating and Cooling Options: Heating is performed by an electrically heated
heat transfer fluid (hot oil) jacket. The stills are cooled by water.
Utility Requirements: Recyclene stills require 240V power, cooling water,
drainage*, and 60 to 110 pai compressed air (the still lid is raised
pneumatically).
Remarks: Recyclene stills were formerly sold under the name of Zerpa
Industries.
Available Models: Recyclene currently has a GSA contract for its solvent
stills. The stills are covered under FSA Class 4940 and contract number '
GS-007-79421 which expires 23 May 1987. Special item numbers, specifications
and GSA prices for Recyclene stills are listed in Table A-14.
B-28
-------
TABLE A-14
EECYCLENE SOLVENT STILLS
SPECIAL
ITEM NO.
8051
8137
8138
8139
8149
MODEL
RS-20
RS-35
RS-35
W/Auto-fill
RX-35
RX-35
CAPACITY
(GAL)
20
35
35
35
THRUPUT
(GPH)
0.5-3
4-12
4-12
10-35
L"
30
51
51
51
x W"
x 43
x 51
x 51
x 51
x H"
x 43
x 50
x 50
x 50
PRICE
$11,077
$16,514
$21,147
$21,399
8141
W/Auto-fill 35
Options and Accessories
10-35
51 x 51 x 50
$26,031
B-29
-------
(15) Unique Industries, Inc.
11544 Sheldon Street
Sun Valley, California 91352
(213) 875-3810
Still Type: Continuous feed.
Solvents Designed For: The stills are not explosion proof and they operate at
atmospheric pressure (no vacuum is available). Therefore, they can distill
only nonflammable (halogenated) solvents with*a boiling point below
approximately 350°F.
Water Separator: Standard
Thruput/Capacity: Vapo-Kleen stills can recycle solvents at rates ranging
from 12 GPH to 110 GFH.
Materials of Construction: Full stainless steel construction is standard.
Safety Features: High vapor level and high boiler temperature thermostats are
standard. Automatic liquid level controls are also standard.
Heating and Cooling Options: The stills are normally heated electrically.
Steam or gas heated models can be custom built at a higher cost. Cooling is
performed by water or refrigeration units.
Utility Requirements: 240V, 3-phase power is required along with possibly
steam, water, and drainage.
Available Models: Specifications and base prices for the smaller Vapo-Kleen
stills (electrically heated) are given in Table A—15.
B-30
-------
TABLE A-15
VAPO-KLEEN SOLVENT STILLS
THRUPOT
MODEL (GPH)
1100-10W
HOO-10SA
UOO-20W
1100-20RA
1100-30W
1100-301A
12
12
21
2l
41
41
COOLING
Water
Eefrig.
Water
Refrig.
Water
Refrig.
L" x W" x H"
43 x 38 x 59
56 x 38 x 59
40 x 37 x 62
70 x 37 x 62
50 x 41 x 62
87 x 41 x 62
PRICE
£ 5,720
$ 8,600
$ 7,500
£10,750
$ 8,500
$12,750
B-31
-------
(16) Venus Products Inc.
1862 Ives Ave.
Kent, WA 98032
(206) 854-2660
Still Types: Batch. Venus stills have hoses with fittings that connect
directly to 55-gallon drums of waste solvent. The solvent is automatically
pumped into the unit, distilled, and the clean solvent is deposited into a
clean drum. There are two basic units (the SR-5 and the SR-20).
Solvents Designed For: The stills are explosion proof and can, therefore, be
used for both flammable and nonflammable solvents. They operate at atmospheric
pressure (no vacuum attachment is available). They are designed to distill
solvents with a boiling point up to 210°!'.
Water Separator: Not available.
Thruput/Capacity: Model SRS-5 has one hose for a waste solvent drum and one
hose for a clean drum. It is designed to process this one drum in an 8-hour
shift (a thruput of approximately 7 GPH). Model SRS-20 has four hoses for waste
solvent drums and four hoses for clean drums. It is designed to automatically
process these four drums in an 8-hour shift (a thruput of approximately 27 GPH).
Materials of Construction: Model SR-5 comes standard with mild steel
construction. Model SR-20 comes standard with aluminum construction. Stainless
steel construction is available on both units as an option.
Safety Features: Standard safety features include a vapor temperature
thermostat and a boiler temperature thermostat. The unit is also shut down if
the heat transfer fluid level is too low.
Heating and Cooling Options: Heating is performed by an electrically heated
heat transfer fluid (hot oil) jacket or electric immersion heaters. Model SRS-5
is cooled by water. Model SRS-20 is cooled by a refrigeration unit mounted on
top of the still.
Utility Requirements: The stills require 240V power, compressed air (to drive
the pumps), cooling water (Model SR-5), and drainage.
Remarks: Venus stills are designed to be located outdoors. Roofing is provided
with the units.
Available Models: Specifications and base prices for Venus' two basic stills
are listed in Table A-16.
B-32
-------
TABLE A-16
VENUS SOLVENT STILLS
BOILER THRUPUT
MODEL CAPACITY(GAL ) (DRUMS/SHIFT) L" x W" x H" PRICE
SRS-5 15 1 44 x 44x 168 til,000
SRS-20 600 4 44 x 56 x 168 $21,000
B-33
-------
(17) Westinghouse Electric Corp.
Industrial Equipment Division
Route 32
P.O. Box 300
Sykesville, Maryland 21784
(301) 795-2800
Still Type; Continuous feed.
Solvents Designed For: The stills are not explosion proof and they operate at
atmospheric pressure* Therefore, they are designed to distill nonflammable
(halogcnated) solvents with a boiling point lower than approximately 350°F.
Water Separator: Standard.
Thruput/Capcity: Capacities range from 18 to 85 gallons with thruputs from 15
to 55 GPH.
Materials of Construction: Stainless steel construction is standard.
Safety Features: A boiler thermostat protects against excessive temperature
in the boiler liquid. A vapor temperature thermostat in the condenser shuts
the unit down if cooling is inadequate'. Level controls automatically maintain
the solvent level in the boiler.
Heating and Cooling Options: The stills are heated electrically and cooled by
either water or a refrigeration unit (optional).
Utilitiy Requirements: Standard requirements are 220/A40V 3 phase power,
cooling water, and drainage.
Available Models: Specifications and prices for Westinghouse stills are listed
in Table A-17.
B-34
-------
TABLE A-17
WESTINGHOUSE SOLVENT STILLS
THRUPUT
MODEL (GPH) L" x W" x H" PRICE
SRS15 15 31 x 39 x 56 $ 6,680
SRS30 30 47 x 47 x 72 $ 9,880
SRS60 55 51 x 50 x 92 $14,165
B-35
U . S . GOVERNMENT PRINTING OFFICEi 1 98 7-7 48 " I 2 ' /« O7 ! 7
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
BULK RATE
POSTAGE & FEES PAID
EPA
PERMIT No. G-35
Official Business
Penalty for Private Use, $300
Please make all necessary changes on the above label,
detach or copy, and return to the address in the upper
left-hand corner.
If you do not wish to receive these reports CHECK HERE D.
detach, or copy this cover, and return to the address in the
upper left-hand corner.
EPA/600/2-86/095
------- |