swefw
PB86- 183076
I EPA/600/2-86/043
| April 1986
FIELD EVALUATION OF HAZARDOUS WASTE PRETREATMENT
AS AN AIR POLLUTION CONTROL TECHNIQUE
I I • by
I I C. C. Allen
Research Triangle Institute
P. 0. Box 12194
Research Triangle Park, North Carolina 27709
and
S. Simpson
G. Brant
Associated Technologies, Inc.
212 S. Tryon Street
Charlotte, North Carolina 28281
EPA Contract No. 68-02-3992, Task 9
Project Officer
Benjamin L. Blaney
Thermal Destruction Branch
Alternative Technologies Division
Hazardous Waste Engineering Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
« HAZARDOUS WASTE ENGINEERING RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
> 1 RPROOUCtO « . •
f f| NATIONAL TECHNICAL
I * INFORMATION SERVICE
I "-• . U.S OfP/mWENT OF COKMEIiCt
•" ^ SPRINGFIEtO. V*. 2Z161
-------
\
f I
S i
t
fc.
TECHNICAL REPORT DATA
(Ftease read Imtrucnom on the reverse before completing/
i. REPORT NO.
EPA/600/2-86/048
4. TITLE AND SUBTITLE
Field Evaluation of Hazardous Waste Pretreatrcent
as an Air Pollution Control Technique
5. REPORT DATE
April 1936
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
C.C. Allen, Research Triangle Institute
S. Simpson a G. Brant, Associated Technologies,Inc.
8. PERFOF.MING ORGANIZATION REPORT NO,
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Research Trinagle Institute, Research Triangle
Park, HC 27709
Associated Technologies, Inc.,Charlotte, NC 28281
12. SPONSORING AGENCY NAME AND ADDRESS
Hazardous Waste Engineering Research Laboratory
U.S. Environmental Protection Agency
Office of Research and Development
Cincinnati, OH 45268
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-02-3992
13. TYPE OF REPORT AND PERIOD COVERED
4/84 - 5/85
14. SPONSORING AGENCV CODE
EPA/600/12
IS. SUPPLEMENTARY NOTES
16. ABSTRACT
Three commonly practiced commercial treatment processes were investigated
for the removal of volatile organic compounds (VOCs) from hazardous waste:
thin-film evaporation, steam stripping and fractional distillation. The
data collected included limitations of the treatment technology, the VOC
removal effectiveness, the characteristics of residuals and cost information.
Three thin-film evaporators were evaluated, ?ach treating different types
of wastes. There was a large variation in the amount of highly volatile
compounds ^emoved by this technique, depending upon the boiling temperature
of the waste stream. Four waste streams were treated in a batch steam stripper
Over 90% VOC removal was obtained in all four cases. Two aqueous/organic
waste streams were treated using fractional distillation and over 90% VOC
removal was obtained in both cases.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
S. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
EPA Form 2220-1 (»-7.M
b-IDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
19. SECURITY CLASS (ThisReport)
9. SECURITY CLASS (THu
UNCLASSIFIED
20. SECURITY CLASS /This page)
UNCLASSIFIED
21. NO. OF PAGES
216
22.
RICE
-------
NOTICE
The information in this document has been funded wholly or in part by the
United States Environmental Protection Agency under Contract No. 68-02-3992,
Task No, 9, to Research Triangle Institute. It has been subject to the
Agency's peer and adminstrative review, and it has been approved for
publication as an EPA document.
ii
-------
I- t
FOREWORD
Today's rapidly developing and changing technologies and industrial
products and practices frequently carry with them increased generation of
solid and hazardous wastes. These materials, if improperly dealt with, can
threaten both public health and the environment. Abandoned waste sites and
accidental releases of toxic and hazardous substances to the environment also
have important environmental and public health implications. The Hazardous
Waste Engineering Research Laboratory assists in providing an authoritative
and defensible engineering basis for assessing and solving these problems.
Its products support the policies, programs, and regulations of the
Environmental Protection Agency, the permitting and other responsibilities of
State and local governments and the needs of both large and small businesses
in handling their wastes responsibly and economically.
This report presents the results of field assessments of three waste
treatment techniques that have the potential for use in control of emissions
of volatile organic compounds from hazardous waste facilities by removing
those compounds from the waste streams. These treatment techniques are
thin-film evaporation, steam stripping, and fractional distillation. The
report is intended for use by government agencies which are considering ways
to reduce emissions from hazardous waste facilities and by facility operator;
and managers who wish to do the sarce. For additional information, please
contact the Alternative Technologies Division of the Hazardous Waste
Engineering Research Laboratory.
WiHitm A. Cawley, Acting Director
' Hazardous Waste engineering Research Laboratory
iii
-------
f
(
ABSTRACT
Three commonly practiced commercial treatment processes were investigated
for the removal of volatile organic compounds (VOCs) from hazardous waste:
thin-film evaporators, steam stripping, and distillation. The data collected
included limitations of the treatment technology, the VOC removal
effectiveness, the characteristics cf residuals, and cost information.
Thin-film evaporators are effective for separating liquids from dissolved
solids and suspended solids. Of the three thin-filn evaporators evaluated,
vent emissions of volatile organic compounds were observed for high-
temperature vacuum operation. For atmospheric operation, there was no
measurable flow. When both volatile and semivolatile wastes were processed in
the thin-film evaporators, the concentrations of volatiles in the residue from
the treatment process were reduced by 90 percent; however, when primarily
volatiles were treated, the concentrations of volatiles in the residue were
not significantly reduced.
Steam stripping and distillation of six different wastes were evaluated;
sludges are more difficult to treat in these processes than in thin-film
evaporation processes, but lower VOC concentrations could be obtained in the
process residues. Concentrations of individual VOCs were measured as the
treatment process continued, and the rate of removal of the VOCs was generally
proportional to the concentration present. The main air emissions source
identified for these processes was the condenser vent, with rates of O.OC29
and 0.0035 g/sec measured for two steam stripping batches. The air emission
rates from the first distillation batch were varieble, with a typical value of
0.06 g/sec. The air emissions from both the condenser vent and the receiver
vent of the other distillation batch were less than 0.002 g/sec.
This report was submitted in partial fulfillment of EPA Contract
68-C?-3992, Task 9, by Research Triangle Institute under the sponsorship of
the U.S. Environmental Protection Agency. This report covers a period from
April 1984 to May 1985, and work was completed as of July 1985.
iv
-------
CONTENTS
Page
Foreword . ill
Abstract iv
Figures ix
Tables xi
1. Introduction 1
Background 1
Purpose of the Program 2
Procedures 2
Scope of the Report 3
2. Conclusions 5
General Conclusions 5
Thin-Film Evaporator Conclusions 6
Steam Stripping Conclusions 6
Distillation Conclusions 7
3. TSDF Air Emissions Sources 8
Type of Sources 8
Size of Sources 12
4. Selection of Pretreatment Processes and Wastes for Evaluation . 15
5. Removal of Volatile Organics From Hazardous Wastes Using
Mechanically Agitated Thin-Film Evaporators 23
Introduction 23
Equipment Design 23
Performance 27
Suppliers 31
Luwa Corporation 34
Design 34
Capital Costs 36
Operating Costs 36
Sources of Emissions to the Environment ............. 38
Applications to Pretreatment ' . . . 38
-------
CONTENTS (Continued)
Page
6. Fractionation Distillation as a Method for Pretreatment
of Hazardous Waste Streams 41
Introduction 41
Process Description . 41
Batch Distillation 42
Continuous Distillation 42
Process Design Considerations 44
Equipment Design 46
Column Shells 46
Column Internals 46
Instrumentation 52
Reboiler 54
Condensers 54
Preheaters/Coclers , 56
Vent Condenser 56
Pumps 56
Equipment and System Suppliers 56
Cost 57
Capital Cost 57
Operating Costs. 59
Sources of Emissions to the Environment 63
Applications to Pretreatment 63
7. Field Test Results: Thin-Film Evaporators 65
Thin-Film Evaporator Field Evaluation at Plant A 65
Process Description 65
Process Effectiveness 66
Process Residuals 68
Process Cost 71
Thin-Film Evaporator Field Evaluation at Plant B 71
Process Description 71
Process Effectiveness 73
Process Residuals 75
Process Cost ..... 75
-------
CONTENTS (Continued)
Page
Thin-Film Evaporator Field Evaluation at Plant C 77
Process Description 77
Process Effectiveness 78
Process Residuals 82
Process Cost 83
8. Field Test Results: Direct Steam Stripping 85
Process Description 85
Process Effectiveness 83
Process Residuals 105
Air Emissions 105
Liquid Residuals Ill
Process Cost 112
General Facility Costs for Plant D 112
Unit Treatment Costs for Each Batch 113
Variation of Unit Costs With Degree of Treatment 115
9. Field Test Results: Distillation 125
Distillation Field Evaluation at Plant B 125
Process Description 125
Process Effectiveness 127
Process Residuals 139
Process Cost 141
Distillation Field Evaluation at Plant E 144
Process Description 144
Process Effectiveness 144
Process Residuals 144
Process Cost 145
10. Summary 146
Applicability of Waste Treatment 146
Liquids 146
Reactive Wastes 146
Sludges 146
Solids 147
vii
-------
ft'-
CONTENTS (Continued)
Page
Effectiveness of Waste Treatment 147
Liquids 1^7
«f j Reactive Waste ..... 147
Sludges 147
Solids 147
Cost of Waste Treatment and Residuals Disposal 147
Thin-Film Evaporation 150
Steam Stripping 150
Distillation 150
Economical Recovery of VOCs 150
Measured Air Emissions 153
Thin-Film Evaporators 153
Steam Stripping 153
Distillation 155
Storage Tanks 155
References 156
Appendixes
A. Summary of Analytical Data 158
B. Summary of Process Data 17C
C. Analytical Procedures 184
D. Quality Assurance 185
E. Cost Effectiveness Estimation Methodology 197
F. Sample Calculations. 200
viii
-------
FIGURES
Page
1 Cumulative distribution of TSDF size 13
2 Vertical thin-film evaporator, cylindrical thermal zone .... 24
3 Vertical thin-film evaporators 25
4 Horizontal thin-film evaporation 25
5 Flow path of thin-film evaporator 26
6 Cross section of thin-film evaporators 28
7 Rotors for thin-film evaporators 29
8 Kontro horizontal design has adjustment for
rotor-shell clearance 30
9 Heat transfer in agitated thin-film equipment for various
services 30
10 Luwa evaporator design data 35
11 Batch distillation 43
12 Continuous distillation 45
13 Sieve tray column. Bubble tray column ..... 48
14 Typical valves used in value tray columns 50
15 Illustration of packed column internals 51
16 Types of packing typically utilized in packed columns 53
17 Shell and tube reboiler 55
18 Plate heat exchanger reboiler 55
19 Time schedule for Chem-Pro preassembled distillation units ... 60
20 Time schedule for Chem-Pro field-erected distillations units . . 60
21 Labor comparison of Chem-Pro preassembled and field erected
distillation units 61
22 Solvent recovery: Mixed chlorinated xylenes 67
23 Thin-film evaporation: Isoprcpanol, xylene recovery 74
24 Solvent recovery: Acetone 79
25 Batch steam stripping process 86
26 Concentrations in the batch of waste as a function of time:
xylene MST 89
27 Concentrations in the batch of waste as a function of tine:
trichloroethane MST 90
28 Concentrations in t.ie batch of waste as a function of time:
1,1,1-trichloroethone 91
29 Concentrations in the batch of waste as a function of time:
Mixed solvent batch 92
30 Cost effectiveness of VOC removal: Batch 1 (Aqueous xylene) . . 116
31 Cost effectiveness of VOC removal: Batch 2
(1,1,1-trichlcroethane/oil) 117
32 Cost effectiveness of VOC removal: Batch 3
(1,1,1-trichloroethane/water) 118
ix
-------
FIGURES (Continued
33 Cost effectivenss of VOC removal: Batch 4 (Mixed solvent/
water)
34 Distillation process (Batch 1 (Aqueous MEK)
35 Distillation process: Batch 2 (Aqueous acetone)
36 Concentrations of VOC in MEK waste during distillation
stripping
37 Concentrations of VOC in acetone waste during distillation
119
126
132
134
135
-------
TABLES
Table Page
1 Limitation of Technologies due to Selected Waste
Characteristics 5
2 Total Quantities Treated, Average Quantities Treated, and
Number of Facilities Treating Hazardous Waste by Each Treatment
Process Type 9
3 Total Quantities Stored, Average Quantities Stored, and Number
of Facilities Storing Hazardous Waste by Storage Process Type . . 10
4 Total Quantities Disposed, Averara Quantities Disposed, and
Number of Facilities Disposing of Hazardous Waste by Each
Disposal Process Type . .• 11
5 Distribution of TSDF Sites 14
6 Ranges of Volatility for Hazardous Waste Components 16
7 RTI VOC Pretreatment Waste Characterization Code. . 18
8 A Collection of Waste Streams Organized by VOC Pretreatment
Waste Characterization Codes 19
9 A Preliminary Selection of Waste Pretreatment Options According
to the VOC Pretreatment Waste Characterization Code 21
10 Typical Operating Characteristics for Thin-Film Units 32
11 Major Agitated Thin-Film Evaporator Producers 33
12 Configurations of Commercial Agitated Thin-Film Units 33
13 Budget Prices--Luwa Evaporators 37
14 Cost Distribution of Agitated Thin-Film Units 37
15 Estimated Operating Costs for Volatile Removal From
Hazardous Waste 39
16 Representative List of Distillation System Suppliers 58
17 Plant A Thin-Film Evaporator Waste Compositions and
Keadspace /Analysis 69
18 Plant A Vacuum Pump Vent Gas Analysis 69
19 Plant A Replacement Capital Costs 72
20 1984 Plant A Operating Cost 72
21 Analysis of Liquid Samples, Thin Film Evaporator, Plant B . . . . 76
22 Analysis of Product Samples, Thin Film Evaporator, Plant B. . . . 76
23 Analysis of Gas Samples from Plant B Vacuum Condenser Vent. ... 76
24 Analysis of Liquid Samples, Thin-Film Evaporator, Plant C . . . . 80
25 Analysis of Gas Samples from Vent of Thin-Film Evaporator,
Plant C 81
26 Analysis of Gas Samples from Product Receiver, Plant C 81
27 1984 Plant C Operating Costs 84
28 Waste Characterization And Process Data 93
29 Waste Characterization of Batch 1 (Aqueous Xylene) 94
30 Waste Characterization cf Batch 2 (1,1,1-Trichloroethane/Oil) . . 95
31 Waste Characterization of Batch 3 (1,1,1-Trichloroethane/Water) . 95
32 Waste Characterization of Batch 4 (Mixed Solvent/Water) 96
xi
-------
TABLES (Continued)
Tab".. Page
33 Waste VOC Concentration During Stripping: Batch 1 (Aqueous
Xylene) 97
34 Waste VOC Concentraions During Stripping: Batch 2
(1,1,1-Trichloroethane/Oil) 97
35 Waste VOC Concentrations During Stripping: Batch 3
(1,1,1-Trichloroethane/Water) 98
I 36 Waste VOC Concentrations During Stripping: Batch 4 (Mixed
I Solvent/Water) 98
37 Headspace Concentrations of VOC as a Function of the Stripping
Time: Batch 1 (Aqueous Xylene) 100
38 Headspace Concentrations of VOC as a Function of the Stripping
Time: Batch 2 (1,1,1-Trichloroethane/Oil) 101
39 Headspace Concentration of VOC as a Function of the Stripping
Time: Batch 3 (1,1,1-Trichloroethane/Water) 102
40 Headspace Concentration of VOC as a Function of the Stripping
Time: Batch 4 (Mixed Solvent/Water) 103
41 Linear Correlation of the Logarithm of the Waste Concentration
with the Stripping Time 104
42 Gas Phase VOC Concentrations: Batch 1 (Aqueous Xylene) 106
43 Gas Phase VOC Concentration: Batch 2 (1,1,1-Trichloroethane/
Water) 107
44 Air Emissions Estimations: Batch 1 (Aqueous Xylene) 109
45 Air Emissions Estimations: Batch 2 (l,l,i-Trichloroethane/Oil) . 110
46 Estimated Unit Cost for the Four Batches Tested 114
47 Cost Analysis for the Direct Steam Stripping of Hazardous
Waste: Batch 1 (Aqueous Xylene) 120
48 Cost Analysis for the Direct Steam Stripping of Hazardous
Waste: Batch 1 (1,1,1-Trichloroethane) 121
49 Cost Analysis for the Direct Steam Stripping of Hazardous
Waste: Batch 3 (Aqueous 1,1,1-Trichlorcethane) 122
50 Cost Analysis for the Direct Steam Stripping of Hazardous
Waste: Batch 4 (Mixed Aqueous) 123
51 Distillation Waste Characterization and Process Data 128
52 Waste Characterization of Batch 1 (Aqueous Methyl Ethyl Ketone) . 129
53 Waste Characterization of Batch 2 (Aqueous Acetone) 130
54 Concentrations of VOC in Acetone Batch 136
55 Concentrations of VOC in Methyl Ethyl Ketone Batch 137
56 Concentrations of VOC in Headspace over Batch Residue as a
Function of Stripping Time: Acetone Batch 138
57 Concentration of VOC in Keadspace over Batch Residue as a
Function of Stripping Time. MEK Batch 138
58 Summary of Distillation VOC Removal Rates 140
59 Air Samples: Methyl Ethyl Ketone Waste Process 142
60 Air Samples: Acetone Waste Process 142
61 Air Emissions: Methyl Ethyl Ketone Process 143
62 Air Emissions: Acetone Process 143
63 Effectiveness of Waste Treatment 148
xn
-------
TABLES (Continued
Table Page
64 1984 Unit Costs Provided by Plant Personnel 149
65 1984 Unit Costs Obtained from the Current Investigation 151
66 The Costs of Steam Stripping as a Function of the VOC Content
of the Waste 152
67 A Comparison of Treatment Versus Disposal Costs 152
68 Condenser Vent Emissions 154
69 Storage Tank Emissions. • 154
xm
-------
SECTION 1
INTRODUCTION
BACKGROUND
The EPA Office of Air Quality Planning Standards (OAQPS) is developing
regulations under the 1976 Resource Conservation and Recovery Act (RCRA) and
its 1984 amendments to control air emissions from hazardous waste treatment,
storage, and disposal facilities (TSDF). The purpose of the a'r emissions
regulations is to protect human health and the environment from emissions of
| ; volatile compounds and particulates.
I The sources of TSDF emissions include storage tanks, treatment processes,
I surface lagoons, landfills, land treatment, and drum storage and handling
! facilities. There are approximately 5,000 TSDF locations in the United States
f where one or more of these activities are in progress. The majority of sites
I are part of industrial facilities, while the rest are commercial facilities
| i that accept wastes from offsite.
! i
f j Research has concentrated upon the characterization of uncontrolled
I emissions from these sources through field measurements and by determining the
| i reliability of emissions models. Recent investigations have identified a
number of options for controlling volatile air emissions from TSDFs. These
include restricting the VOC concentrations of wastes going to sources where
volatile air emission rates would be high, the "pretreatment" of wastes to
remove volatiles, and the use of in-situ (i.e., add-on) control techniques at
the TSDF. In addition, changes in waste management practices (such as holding
tanks instead of ponds) may be a cost-effective control option.
Pretreatment is a viable volatile air emissions control option and is in
current use at several TSDFs. In general, it is attractive because it can be
used by either the waste generator or the TSDF operator to remove volatiles
from the waste before there is much opportunity for air emissions to occur.
Pretreatment may be a cost-effective control technique for TSDF emissions
sources with large surface areas, such as land treatment facilities and
lagoons. In disposal surface impoundments (e.g., evaporation ponds) and
aeration tanks and lagoons, pretreatment appears to be an important option
because these TSDF processes rely on transfer of water or oxygen between the
waste and the atmosphere as part of the disposal or treatment process, making
process covers unattractive.
-------
r
PURPOSE OF THE PROGRAM
The purpose of this investigation was to collect data for the support of
regulations which consider waste pretreatment as an alternative for the
control of volatile air emissions from TSDFs. To the extent possible, these
data were collected from the TSDFs that are treating hazardous wastes or from
the treatment of similar wastes at other industrial/commercial operations in
order to permit a comparison of pretreatment to other emission controls.
Field data on several waste treatment techniques were collected to
determine (I) how efficiently they remove volatiles from hazardous waste
streams, (2) what the removal costs are, (3) how the byproducts from the
pretreatment technologies are collected and disposed, and (4) what limitations
(in terms of waste types, volatile concentrations, etc.) are placed on the use
of such treatment techniques. These data documented the viability for
treating specific waste streams by specific treatment techniques. Data from
treatment equipment manufacturers and engineering judgment were then used to
analyze the applicability of each technique to treat other waste streams and
to determine technology capital, overhead, and management costs.
PROCEDURES
For the processes selected for evaluation, field data were collected at
representative sites for thin-film evaporation (3 sites, 3 batches), direct
injection steam stripping, (1 site, 4 batches), and distillation (1 site, 2
batches). Telephone interviews were held with a number of facilities before
sites were selected for 1-day visits. On that first visit, the operating
practices were discussed with the plant management, and information on process
limitations, process operating conditions, equipment costs, and other factors
was collected.
On the first visit, grab samples were taken while following QA/QC
procedures, and the samples were analyzed in the laboratory. Relevant process
measurements such as velocities, flow rates, and temperatures were recorded or
measured. These data were analyzed to determine process effectiveness, and
the potential for air emissions (in the case of the three thin-film
evaporators) was investigated during the 1-day visit.
Field tests were planned for the processes of steam stripping and
distillation. These tests sampled batches of v/aste as they were being treated
to determine, on a compound-specific basis, the rates of VOC removal as a
function of time. Thin-film evaporators v/ere generally being used to treat
waste streams with an organic content above 30 percent, and removal of lower
concentrations of VOCs from waste (as could be achieved with steam stripping
and distillation) was of primary concern. In addition to monitoring the
concentration in the batch of waste being treated, attempts were made to
monitor the flow rates of all process streams: steam, product, reflux, vent
releases, and others. In some cases, it was not possible to perform the
desired measurements; in these cases, the value was estimated or calculated on
the basis of other measurements.
•s
V,
-------
For air emissions, two techniques were available to measure concentra-
tions: carbon adsorption tubes and evacuated steel containers. The steel
containers were used for the planned field tests since the carbon tubes had a
limited retention of chlorinated Icw-molecular-weight VOCs in the presence of
water vapor. The velocity of air streams was measured with an Alnor
velometer, as were wind velocities. In some cases, a dry gas meter cculd be
used to measure flow rates of vent emissions.
Liquid samples were taken in 40 ml VOC bottles; both the headspace
concentration and the liquid concentration of individual VOCs were measured by
gas chronatography (GC). For some solidified samples, the headspace
concentration was the only measurement of VOCs that was made. The compound
| identification by retention time in the GC analysis was verified by a
j nonquantitative GC/mass spectroscopy procedure.
1 The concentration data was analyzed by calculating the partition
1 coefficient for the compounds in the waste and headspace phases. These values
were compared to the partition coefficients calculated from the process data.
First-order decay equations were used to correlate the removal rates of VOCs
from the waste being treated.
SCOPE OF THE REPORT
This report documents the use and effectiveness of pretreatment as a
method of controlling volatile emissions at TSDFs. The types of pretreatment
options considered include thin-film evaporation, steam stripping, and
distillation. The types of waste streams that were considered are those of
significant volume which potentially emit hazardous emissions from (1) lend
treatment; (2) aeration tanks and lagoons; and (3) treatment, storage, or
disposal surface impoundments. RTI and its subcontractors, Associated
Technologies, Inc. (ATI) and Industrial Environmental Analysts, Inc. (IEA),
i performed the sampling and analyses and the engineering evaluation.
I The major conclusions of the rep3rt are summarized in Section 2. These
include comparisons of the cost-effectiveness of thin-film evaporation, steam
stripping, and distillation as VOC removal techniques. Conclusions about the
amount of residuals produced by each individual process and the process
limitations for use to treat hazardous waste erea also provided. A discussion
of the sources of volatile organic compounds (VOCs) is presented in Section 3.
Section 4 presents a discussion of why the three treatment technologies were
chosen for study. Sections 5 and 6 provide general engineering descriptions
of thin-film evaporation and distillation, respectively. These are intended
to indicate how the processes operate, what types of wastes they can treat,
and, in general terms, what their technical limitations are. For similar
discussions of steam stripping, a recent EPA report entitled "Process Design
Manual for Stripping of Organics" (EPA, 1984) should be referenced.
Section 7 reports findings from 1-day site visits to three facilities
operating thin-film evaporators. Limited sampling and analysis conducted
during these visits on process influent and effluent streams are reported. In
Section 8, the results of a 3-day field assessment of a batch steam stripper
-------
r
at. a TSDF are reported. A similar field assessment of two fractional
distillation units at another TSDF is reported in Section 9. Section 10
summarizes the findings from the previous site visits and field studies and
presents, analyses of the data to make comparisons between the three types of
treatment processes with respect to their suitability for VOC removal from
hazardous wastes.
The appendixes to the report present the analytical results (Appendix A),
process data for the field tests for steam stripping (Appendix B, the
analytical procedures (Appendix C), and quality assurance (Appendix D).
Appendix E presents the methodology used in estimating the cost-effectiveness
of steam stripping, and Appendix F contains selected sample calculations.
-------
SECTION 2
CONCLUSIONS
GENERAL CONCLUSIONS
1. Waste treatment is not being practiced for the purpose of VOC control.
2. Based upon information obtained from field visits including operator
comments and EPA sampling and analysis, Table 1 presents potential uses
of three treatment techniques for VOC removal. All three techniques are
applicable to aqueous, as well as organic and mixed aqueous/organic
streams. However, their appli:ability is limited by other waste
characteristics shown in the table.
TABLE J. LIMITATION OF TECHNOLOGIES DUE
TO SELECTED WASTE CHARACTERISTICS
Thin-film
Waste characteristics evaporation
Over 95% water content
Polymer izable waste
Presence of dissolved solids
Waste containing sludges and tars
Highly viscous waste
A
P
A
A*
N
Batch steam
stripping
A*
P
A*
P
N
Distillation
A*
P
P*
N
N
A Applicable technique
P Potentially applicable
N Not applicable
*Waste/treatment combinations for which field sampling and analysis was per-
formed as documented in this report. /
-------
IT"""
3. The cost of waste treatment is sensitive to the concentration of VOC in
the influent waste. For a fixed percent of VOC removal, costs rise as
the initial VOC concentration decreases. For many wastes, the value of
the recovered VOCs is less than the treatment costs.
4. Treatment of aquecus waste streams for VOC removal can significantly
decrease their disposal cost by making them amenable to discharge to
municipal sewers.
THIN-FILM EVAPORATORS
1. Thin-film evaporators permit the recovery of VOCs from waste materials
containing sludges and tars. Typically greater than 89 percent organic
recovery is achieved. Thin-film evaporators may not adequately process
wastes that are reactive (polymerization) or that contein large pieces of
solid material. VOC removal efficiencies in the residual waste material
were 23 to 99.9 percent.*
2. The overhead product from the thin-film evaporator can be treated by
distillation, carbon adsorption, and other separation or reaction
processes applicable to liquids.
3. The major air emission source from a thin-film evaporator is the vacuum
vent. Under abnormal operating conditions (e.g., inadequate condenser
cooling), the emissions could be significant. Under normal operations,
emissions are expected to be much less than VOCs recovered.
4. Since some liquid must remain in the product bottoms to maintain waste
viscosity, VOC concentations will remain relatively high in some wastes
treated by thin-film evaporation. The VOCs can be selectively removed
from high boiling liquids. However, if the waste is a mixture of low
boiling compounds, significant amounts of these compounds remain in the
bottoms. This suggests that the emissions (g VOC/liter waste) will
not be significantly lower for ths treated waste than for the untreated
waste, although waste volumes will be reduced,
5. Costs of waste treatment using thin-film evaporation ranged from $0.033
to 50.37/L of VOC recovered.
STEAM STRIPPING CONCLUSIONS
1. Steam stripping is effective for reducing the concentration of VOCs to
levels of 0.1 percent or lower. Removal efficiencies of 99 to 99.8
percent were observed.
*A11 processes tested were at recycling firms and the extent of waste stream
treatment was determined by economic considerations, not technical
constraints. Therefore, VOC removal efficiencies reported here are lower
limits for the combinations of processes and waste streams tested.
-------
2. The amount of steam required tu remove the VOCs in waste materials is
greater than predicted from equilibrium partitioning based on vapor
pressure and solubility in dilute aqueous solutions.
3. The rate of volatile removal was logarithmic in nature, with
substantielly longer times required to remove the VOCs at lower
concentrations than at higher concentrations.
4. The air emissions from the steam stripping process tested are much lower
than the amount of VOCs recovered from the waste.
5. Costs of treatment ranged from $0.17 to S0.53/L VOC recovered for streams
that are typically recycled at the facility visited, but were as high as
$4.34/1 for streams of low (<10 percent) VOC concentration.
DISTILLATION CONCLUSIONS
1. The individual VOC components can be removed from the waste material by-
batch distillation. Removal efficiencies of 99 percent and greater were
observed, with resultant VOC concentrations below 0.1 percent.
2. The removal rates of the components are a function of the waste matrix
and the ratio of the rate of steam to the batch size, and they are
generally proportional to the VOC concentration in the v/aste. In the
distillation process with reflux to the column, the more volatile
materials were selectively removed first from the waste.
3. In the two processes tested, air emissions from the process vents
represented only a relatively small fraction (less than 0.2 percent) of
.the VOC present.
4. Costs of treatment were typically in the range of $0.20 to S0.70/L of VOC
recovered, but were estimated to be as high as S1.18/L for streams of low
(<10 percent) VOC concentration.
-------
I
t _
SECTION 3
TSDF AIR EMISSIONS SOURCES
This section discusses the air emissions sources at TSDFs. The number of
TSDFs, the volume of the waste streams, and the major sources of air emissions
are discussed. This overview describes the context in which pretreatment may
be applied to reduce VOC emissions from TSDF processes.
TYPES OF SOURCES
Hazardous waste treatment, storage, and disposal facilities contain a
number of potential sources of air emissions of volatile organic compounds
(VOCs). Treatment sources of VOC emissions include process tanks, cooling
towers, aerated lagoons, and process leaks. Storage emissions include VOC
fugitive losses from lagoons, leaking drums, and tanks. Ultimate disposal VOC
emissions result from operations such as land treatment, surface impoundments,
and landfills. Processes such as surface impoundments are expected to release
a substantial amount of the VOC originally present in the waste, whereas only
a small fraction of the VOC is expected to produce air emissions from storage
leaks. Many potential emission points exist in the lifecycle of waste, from
the point of generation to ultimate disposal. One advantage of pretreatment
is that it can be used at the point of generation to control VOC emissions for
ail subsequent waste management operations.
Table 2 presents the total quantities of waste materials treated, the
average quantities treated* and the number of facilities treating hazardous
waste by each treatment process type employed at TSDFs. Much of the treatment
was in tanks and surface impoundments which can be significant sources of TSDF
emissions.
Table 3 presents the total quantities stored, average quantities stored,
and the number of facilities storing hazardous waste by storage process type.
Again, surface impoundments constituted the major process type. The disposal
quantities, the average quantity disposed, and the number of facilities
disposing of hazardous waste by each disposal process type is presentee in
Table 4. Surface impoundments and injection wells are the major disposal
method.
Relatively large numbers of facilities handled organic wastes, solvent
wastes, and nonhalogenated solid wastes (Dietz, 1984). Many of the facilities
either incinerate or deep-well inject the waste streams containing volatile
organic compounds; however, in many other cases, the waste streams containing
VOCs are treated in treatment tanks, stored in surface impoundments, and
placed in landfills and land treatment facilities, all of which can be
significant sources of emissions.
-------
TABLE
2. 1981 TOTAL ANNUAL QUANTITIES TREATED, AVERAGE ANNUAL
AND NUMBER OF FACILITIES TREATING HAZARDOUS WASTE BY
Treatment Process
Type
Treatment Tanks
Treatment Surface
Impoundments
Incinerators
Other Treatment
PROCESS TYPE (Dietz,
Total Quantity Treated
By Process Type
(Billion (1,000 Metric
Gallons) Tonnes)
8.73 32
16.60 62
0.45 1.7
4.£8a 17a
1984)
Average Quantity Treated Per
Plant by Process Type
(Million (Metric
Gallons) Tonnes)
14.3 53
40.6 151
1.9 7
12. 9a 48a
QUANTITITES TREATED
EACH TREATMENT
Number of Facilities
Treating By Each
Process Type
609
410
240
392b
Based on 355 facilities that were not in combination with the above processes.
Includes 37 facilities that were in combination with the above processes.
-------
1 ' *
I /
f . '
V
TABLE 3. 1981 TOTAL ANNUAL QUANTITIES
NUMBER OF FACILITIES STORING
wyg^^w.ff'^^wy
STORED, AVERAGE ANNUAL QUANTITIES STORED AND
HAZARDOUS WASTE BY STORAGE PROCESS TYPE
(Dietz, 1984)
Total Quantity Treated Average Quantity Treated Per
By Process Type Plant by Process Type Number of Facilities
Treatment Process (Billion (1,000 Metric (Million (Metric Treating By Each
Type Gallons) Tonnes) Gallons) Tonnes) Process Type
Storage Tanks 5.10 19,300
Storage Containers 0.16 600
Storage Surface
Impoundments 14,10 53,400
Waste Piles 0.39 1,500
Other Storage 0.263 980a
3.57 13,500 1,428
0.045 170; 3,577
25.6 96,900 552
2.2 8,300 174
1.9 7,200 139b
aBased on 355 facilities that were not in combination with the above processes.
0 Includes 37 facilities that were in combination with the above processes.
*
,«
I ]
1!
• i
i
i
i
\
•i
"i
3
I
i
i
1
t
\
,
1
1
J
1
1
]
i
-------
TABLE 4. 1981 TOTAL ANNUAL QUANTITIES DISPOSED, AVERAGE ANNUAL QUANTITIES DISPOSED
AND NUMBER OF FACILITIES DISPOSING OF HAZARDOUS WASTE BY EACH DISPOSAL
PROCESS TYPE (Dietz, 1984)
Treatment Process
Type
Disposal Surface
Impoundments
Land Application
(Treatment)
Other Disoosal
Total Quantity Treated
By Process Type
(Billion (1,000 Metric
Gallons) Tonnes)
Average Quantity Treated Per
Plant by Process Type
(Million (Metric
Gallons) Tonnes)
5.10
0.10
0.02
19,300
400
75
44.0
1.4
3.3
166,500
5,300
12,500
Number of Facilities
Treating By Each
Process Type
Injection Wells
Landfills
8.60
0.81
32,500
3,100
99.0
4.1
375,000
15,500
87
199
116
70
7
-------
SIZE OF SOURCES
A few of the TSDFs are substantially greater in size than the other
TSBFs. Also, most of the waste is managed in relatively few of the facili-
ties. Table 5 presents a distribution of TSDF sizes. The distribution of
TSOF sizes is approximately log normal as illustrated in Figure 1, with a
standard deviation of approximately 4. Table 3 indicates that there is a
large variation in facility size. Fifty percent of the TSDFs handle less than
76,000 L (20,000 gal) of hazardous waste per year, while 25 percent handle up
to 10 times this much. Another 5 percent handles over 91 million L (24
million gallons) per year.
The implication of the spread of facility sizes is that the size of the
pretreatment units may vary greatly from site to site. Small TSDFs would be
likely to run batch treatment processes and may have a large variation in
stream composition. The unit costs of treating wastes at large facilities are
likely to be lower than at smaller operations because the large facilities may
be able to both use continuous treatment and benefit from the economies of
large-scale process equipment.
12
-------
2-
1-
1 10
SIZE OF FACILITY (Million Gallons/Year)
FIGURE 1. CUMULATIVE DISTRIBUTION OF TSDF SIZE
(EPA, 1984)
100
-------
r
TABLE 5. DISTRIBUTION OF TSDF SITES
Approximate size of facility
Million gal/yr
Thousand metric
tonnes/yr
Percent of
facilities
greater than size
Normal
probability
function
F(x)
306
130
24
4
0.2
0.02
1,200
500
91
15
0.8
0.08
1
2
5
10
25
50
2.33
2.05
1.64
1.28
0.68
0
14
K* '„ " V
-------
n
SECTION 4
SELECTION OF PRETREATMENT PROCESSES AND WASTES FOR EVALUATION
The major purpose of this report is to document the performance of
treatment techniques which can be used to remove volatile organic compounds
from hazardous wastes. If the volatile organic compounds are removed from the
wastes before the wastes are handled at a TSDF, then the emissions of volatile
organic compounds would be reduced at these facilities.
On the basis of the previous RTI report entitled, "Preliminary Assessment.
of Hazardous Waste Pretreatment as an Air Pollution Control Technique," by
James Spivey, 1984, various pretreatment techniques were identified which were
potentially applicable to removing VOCs from hazardous waste. These
techniques included carbon adsorption, resin adsorption, biological treatment,
distillation, evaporation, chemical oxidation, wet oxidation, ozonalysis,
physical separation, solvent extraction, air stripping, and steam stripping.
The appropriateness of these waste treatment techniques depends upon the
nature of the waste material in some cases. For example, chemical oxidation
and ozonalysis are more appropriate for low concentrations of VOCs, but
distillation is more appropriate for medium or high concentrations of VOCs in
the waste. The processing of sludges is not practical in certain types of
process equipment.
The waste streams of interest for the present study are predominantly
those that are generated in large volumes and for which the treatment,
storage, or disposal of these waste streams can result in significant VOC
emissions to the air. These types of waste are primarily composed of mixed
wastes (organics and water) and aqueous wastesf In the selection of the waste
treatment technologies which would be of most interest, techniques which were
limited in the types of waster, that could be treated or techniques for which
substantial information is already available (e.g., carbon adsorption) were
not emphasized. Commercial techniques which were applicable to a variety of
wastes were investigated because they are potentially available for immediate
;. I application to VOC control problems.
Table 6 presents the different ranges of volatility for hazardous waste
components. Examples of compounds that are in each range are shown.
Pretreatment techniques whichtrely on vapor phase mass transfer such as steam
stripping, air stripping, distillation, and evaporation are much more
appropriate for the volatile and semivolatile components than are the
components of low volatility. Pretreatment techniques such as extraction and
reaction are potentially applicable to each of the components independent of
the vapor pressure.
* For the purposes of this report, aqueous wastes are defined to contain
at least 97% water and little solids.
15
-------
TABLE 6. RANGES OF VOLATILITY FOR HAZARDOUS WASTE COMPONENTS
COMPONENT
VAPOR PRESSURE
(mm Hg)
VOLATILE
Benzene
Acrylonitrile
Carbon Tetrachloride
Methyl Ethyl Ketone
Toluene
Xylene
Tetrachioroethane
SEMIVOLATILE
Phenol
Aniline
Toxaphene
Naphthalene
Tridecane
LOW VOLATILITY
Mercury
Parathion
PCB-1254
Dioxin
16
95
100
90
71
88
10
5
0.34
0.33
0.3
0.1
0.0018
0.00002
0.00008
-------
For the purposes of this study, it was decided to focus primarily on
organic compounds which are considered volatiles based on the ability to
analyze their presence in water through a purge and trap technique (Method
624). The basis for this decision was the fact that it has been found in
earlier studies of emissions from TSDFs (Radian) that over 95 percent of the
emissions from surface impoundments and land treatment areas were volatile
compounds, while semivolatiles contributed little to emissions. While 'waste
streams were selected for study because they contained such volatile
compounds, some streams also contained semivolatiles and potential for
removing them was also determined.
In the selection of waste streams which are produced in large volumes, it
is important to relate the characteristics of the waste with the current
practices of waste disposal. Table 7 presents a VOC pretreatment waste
characterization code which the authors d2veloped to categorize character-
istics related to stream treatability. The three major waste stream
characteristics which were considered for identifying appropriate pretreatment
techniques were (1) the general composition of the waste (aqueous, solid,
organic, or water-organic mixture), (2) the presence of tars or solids, and
(3) the water solubility of the VOCs contained within the waste. Aqueous
streams could be handled entirely differently than solid streams, and the
presence of tars of other solids could limit applicable technology as, for
example, through the contamination of activated carbon used for adsorption.
As another example, the steam stripping of water soluble VOCs would not
necessarily result in an effective pretreatment.
These waste codes were applied to the waste streams in an early version
of the Waste-Environment Technology (WET) Model Data Base (SCS, 1983). Table
8 presents a collection of the waste streams organized by the above VOC
pretreatment waste characterization code. These streams included all the
streams in the SCS reference which contained VOCs (more recent versions of the
data base do not contain additional disposal information and were not
included). Predominantly organic streams were typically landfilled and
incinerated. There is also a strong potential for waste recycling for these
streams. Mixed waste streams undergo a variety of different treatment or
disposal practices. In addition, there was less known about the disposal
procedures for these types of waste. Aqueous wastes either were typically
treated with activated carbon or were biologically oxidized.
Since the process that was used to dispose of mixed and aqueous waste
streams appeared to have high potential for emissions (Table 8), and since
these streams are not frequently recycled, it was decided to concentrate in
this study on these two broad categories of wastes.
In order to identify the waste treatment types that should be studied in
the field, a preliminary assessment of technique applicability by waste stream
class was made. The organic wastes are typically subject to solvent recovery
and incineration, and these were not emphasized in the present study
program; however, there would be a need for treating the mixed waste ana the
aqueous waste, since they are often treated or disposed of in processes which
have a high potential for VOC emissions (e.g., surface impoundments). Table 9
presents a selection of waste pretreatment options according to the
17
-------
TABIE 7. RTI VOC PRETREATHENT WASTE CHARACTERIZATION CODE
1. First Code Letter
S solid, predominantly solids
0 organic, greater than 75% organics
M mixed, aqueous soluble organics
A aqueous, less than 3% of liquid is organics
2. Second Code Letter
T contains tars or high molecular weight material
S contains solids
L liquid stream
5 3. Third Code Letter
V contains very soluble VOCs
S contains soluble VOCs
I contains insoluble VOCs
18
-------
TABLE 8.
A COLLECTION OF WASTE STREAMS ORGANIZED BY VOC
PRETREATHENT WASTE CHARACTERIZATION CODES
MAJOR ORGANIC WET
COMPONENT STREAM NUMBER
(SCS. 1983) (SCSj 1983)
Phthalic Anhydride
Oichloropropanols
Naphtha Una
DiaUylbenzene
Tetrach 1 oro< thane
Ally! alcohol
Trichloroe thane
Trichloroe thane
Cydohexanol
Uichloroarooene
Chloropnenol
Phenol
Sutanol(s)
Trichloroprepane
Tr'chloroethylene
Tetracnloroetnene
Toluene
Toluene
Hex ac h 1 orobenzene
Tricnloroetnylerie
Methyl acetate
Aniline
Trichloroethane (1,1,1)
Acetaldenyae
Acrolein
3enzene
Cyctonexane
Heaach 1 orasutadi ene-
Benzal chloriae
Tetrachloroethane
"richlorooenzene (1.2,4)
3enzene
3initrobenzene-M
texacr.oroetftane
Chloroforn
VicMorodeozene (1.2,*)
Hexachlorooenzene
Jisiethyl aUyianine
"oxapnene
Cyanide
Trichloroe thylene
5enzo(a)pyrene
senzo(a)pyrene
Mercury
Mercury
Oichloroethane (1,2)
Benzene
Acetaldehyde
Acrolein
^henol
Paraldehyde
Methyl metnacrylate
Etnanol
Acetaldehyoe
CarDon Tetrachloride
Methyl etnyl ketone
Methyl ethyl icetone
03.05.02
03.05.07
03.05.09
03.05.09
03.05.03
03.04.30
03.04.04
03.04.05
03.04.18
03.04.23
03.04.29
03.04.13
03.04.24
03.04.02
03.04.03
03.01.04
03.01.90
03.02.90
03.04.12
03.02.01
03.34.22
03. 04. IS
03.04.16
03.04.26
03.04.30
03.04.21
03.04.08
03.04.11
03.04.0',
03.01.05
03.06.03
03.04.19
03.04.09
03.05.01
03.06.01
03.04.01
03.04.31
03.04.32
03.04.32
01.03.02
03.01.06
04.02.01
04. (12. 02
01.01.03
01.01.02
02.02.09
02.02.08
03.04.14
03.04.28
02.02.21
02.02.06
02.02.15
03.04.27
02.02.12
02.02.20
02.02.22
02.02.23
VOC
PRETREATHENT
CODE
STS
SSS
SSI
SS!
SSI
OTV
OTS
OTS
OTS
OTI
on
osv
OSS
OSS
OS I
OSI
OSI
OSI
OSI
OSI
OLV
OLV
OLV
otv
OLV
OLS
OLS
OLI
OLI
OLI .
OLI
OLI
QLI
OLI
OLI
OLI
OLI
OLI
OLI
MSV
MSI
MSI
MSI
MSI
MSI
MLV
MLV
MLV
MLV
MLV
MLV
MLS
MLS
MLS
MLI
MOV
MOV
TOTAL AMOUNT OF
ORGANICS GENERATED TREATMENT/DISPOSAL
(Kg/day) METHOD*
(SCS, -1983) (SCS, 1983)
1929
9920
64925
64925
49572
245
151317
104688
62?7
11925
244
4134
58900
16704
69660
48960
32880
8220
1260
63
274234
4905
4781
3611
245
25440
3180
110523
87513
47360
12326
6945
3914
1920
790
253
189
66
66
1929
17360
1890
876
404
17
136
41798.4
36216
1003.2
360
255
11484
7536
6636
1560
1152
148
landfttl
Landfill, incineration
Landfill, Incineration
Incineration
Landfill, incineration
Landfill, incineration
Incineration
Incineration
Incineration
Incineration
Incineration
Landfill, incineration
Landfill, incineration
Landfill
Inci neration
Incineration
Incineration
Unknown
Incineration, landfill
Incineration, landfill.
deep well injection
Incineration
Incineration
Landfill
Landfill, incineration
Landfill, incineration
Incineration
Incineration
Incineration
Deep well injection.
Incineration
Aerobic digestion
Deep well injection
Incineration
Unknown
Biooxidation
Unknown
Unknown
(Continued)
19
-------
TABLE 8. (Continued)
1AJOR ORGANIC
COMPONENT
(5CS. 1983)
Toluene diisocyanate
3enzene
3enzene
Naphtnoquinone-1,4
»henol
Cyanide
Acetonitrile
Chlordane
"arathion
Naphthalene
Mercury
Toxaphene
Acenaphthcn?
Pentachorophenol
tthylene oxide
Acetic acid
Maleic anhydride
Phenol
Benzene
Phenol
Phenol
Anil ine
Formaldehyde
Carbon Tetracnloride
To1 uene
Benzene
WET
STREAK NUMBER
(SCS, 1983)
03.05.05
02.02.13
02.02.14
03.05.06
02.01.02
01.03.01
02.02.02
02.03.91
02.03.90
03.06.02
01.01.12
02.03.01
02.04.01
02.01.01
02.02.17
02.02.01
02.02.10
02.02.19
02.02.04
02.01.03
02.02.16
02.02.11
02.02.18
02.02.03
02.02.07
02.02.05
VOC
PRETREATHEfT
CODE
ATV
ATS
ATS
ATI
ASV
ASV
ASI
ASI
ASI
ASI
ASI
ASI
ASI
ASI
ALV
ALV
ALV
AlV
ALV
ALS
ALS
ALS
ALS
ALS
ALI
ALI
TOTAL AMOUNT OF
ORGANICS GENERATED
(Kg/day)
(SCS, 1983)
3
232845
178920
5013
8160
1366
81921
81654
81654
16815
1228
138
70
35
43650
32850
8332
4225
942
91392
84000
25600
5250
2
290
3
TREATMENT/DISPOSAL
METHOD*
(SCS, 1983)
Landfill
Biooxidation
Deep well injection.
anaerobic lagoon, landfill
Biooxidation
Biooxidation
Unknown
Aerobic digestion
Biooxiddtion
Unknown
Deep well injection,
activated carbon
*ihe treatment/disposal method is the current treatment/disposal
method identified in the reference.
20
-------
p..* ,.....-. <,*,,*-, -,
. , „. .,.^.,,,^,^,,,.,, :,,,,w,w^^_
f \ -
; TABLE 9. A PRELIMINARY SELECTION OF WASTE PRETREATMENT OPTIONS ACCORDING
\
i Code for
; waste stream
category
ASI
ASS
ASV
- ALI
ALS
2ALV
ATI
/ ATS, ATS
TO THE VOC
Steam stripping
PRETREATMENT WASTE CHARACTERIZATION CODE
Distillation column(s)
System/
stream
Applicable category
w/solid removal or TFE
w/solid removal or TFE
w/solid removal or TFE
Yes
need LCA add on
need LCA add on
w/tar removal or TFE
w/tar removal or TFE
MSI, MSS, MSV w/solid removal or TFE
MLI, MLS, MLV
S
Waste Stream
and DC add on
add on
No
Categorization Code (XYZ):
SI
S2
S2
S3
S4
S4
S5
S6
S7
S8
S9
X
Solid
Organic
Aqueous
Applicable
w/solid removal
w/solid removal
may need LCA add
Yes
Yes
may need LCA add
No
No
w/solid removal
Yes
No
Y
Tar
Solids
Liquids
System/
stream
category
Dl
Dl
on D2
D3
D3
on D4
--
--
D5
D6
--
Z
Very soluble
Soluble
Insoluble
Liquid phase
carbon adsorption
System/
stream >
Applicable category
w/solid removal CA1
MW dependent
w/solid removal CA2
MW dependent
w/solid removal CA2
MW dependent
MW dependent CAS
MW dependent CA4
MW dependent CA4
No -- !
No
No
' ~
No
NO - ;
! j
*
Mixed
Other abbreviations: TFE - thin film evaporation; LCA = liquid phase carbon adsorption; DC « distillation
column; MW = molecular weight.
-------
VOC-pretreatment waste characterization code. Steam stripping, distillation
columns, and liquid phase carbon adsorption were considered for these streams.
Nine waste categories where steam stripping v/as potentially applicable are
identified as S1-S9. Six stream categories which would be appropriate for
distillation are identified as D1-D6. Four different categories believed to
be appropriate for liquid phase carbon adsorption are identified as CA1-CA4.
f Because the carbon adsorption was applicable to only a limited number of
streams, would require preliminary solid removal for most solid waste streams,
and would not necessarily preferentially remove volatile organics, it was
decided to not investigate this technology in the present study. Steam
stripping and distillation were chosen for investigation because of their
potential applicability to a wide range of waste categories and because they
could be used to treat waste of intermediate organic content, 0.5 to 50
percent. As the study progressed, it became evident that thin-film
evaporation (TFE) is most applicable for waste with high organic content and
low-to-intermediate viscosity. The study also sought to find hazardous waste
treatment sites using air stripping, since this technique may be less costly
than steam stripping. Only one hazardous waste site was identified whic1! uses
air stripping for VOC removal from wastes, but a thorough study of the
facility had not been completed at the time of the present report.
22
-------
SECTION 5
REMOVAL OF VOLATILE ORGANICS FROM HAZARDOUS WASTES
USING MECHANICALLY AGITATED THIN-FILM EVAPORATORS
INTRODUCTION
In the selection of potential pretreatment processes for removing
volatile organics from hazardous wastes, agitated thin-film evaporation must
be considered for those wastes that would foul or plug conventional
evaporators. Certain slurries, sludges or viscous wastes which cannot be
handled by flash evaporators, forced circulation evaporators, distillation
columns, falling film evaporators or other conventional equipment can be
successfully processed in agitated thin-film evaporators.
EQUIPMENT DESIGN
Agitated thin-film evaporators are designed to spread a thin layer or
film of liquid on one side of a metallic surface, with heat supplied to the
other side. Heat can be supplied by either steam or heated oil; heated oils
are used to heat the waste to temperatures higher than can be achieved with
saturated steam (>100°C).
T'.ie unique feature of this equipment is not the thin film itself
(falling- and rising-film evaporators use thin liquid layers) but rather the
mechanical agitator device for producing and agitating the film. This
mechenical agitator permits the processing of high-viscosity liquids and
liquids with suspended solids. The agitation at the heat transfer surface not
only promotes heat transfer but also maintains precipitated or crystallized
solids in manageable suspension without fouling the heat transfer surface.
There are two general types of mechanically agitated thin film
evaporators: horizontal and vertical. A typical unit consists of a
motor-driven rotor with longitudinal blades which rotate concentrically within
a heated cylinder.
The vertical design illustrated in Figure 2 is manufactured by Luwa
Corporation (Charlotte, NC) end incorporates a cylindrical thermal zone. Some
manufacturers utilize a tapered thermal zone, as illustrated in Figures 3 and
4.
In the vertical design, Figure 5, product enters the feed nozzle above
the heated zone and is mechanically transported by the rotor and gravity down
a helical path on the inner heat transfer surface.
The evaporator does not operate full of product; the liquid or slurry
forms a thin film or annular ring of product from the feed nozzle to the
23
-------
!• I
A-HMttng Noato
B-HMttngNozzt*
Figure 2. Vertical thin-film evaporator, cylindrical thermal zone.
24
-------
Feed ^.
•*• Vapor
orcora
Cylindrical
Thermal Zone
•»• Vanor
bottoms
Tapered
Thermal lone
Figure 3. Vertical thin-film evaporators.
Cylindrical
Thermal Zone
Tapered
Thermal Zone
Figure 4. Horizontal thin-film evaporation.
25
-------
Heating medium
H
Modular heating bodies
Product outlet
Figure 5. Flow path of thin-film evaporator.
26
-------
I !
product outlet nozzle, as illustrated in Figure 6. Holdup or inventory of
product in a thin-film evaporator is very low—typically about a half-pound of
material per square foot of heat transfer surface.
With typical tip speeds of 900 to 1,200 cm/sec ,(30 to 40 feet per
second), centrifugal forces distribute the liquid feed as a thin film on the
heated cylinder wall, and the wave action produced by the rotating blades
provides rapid mixing and frequent surface regeneration of the thin liquid
layer on the transfer surface.
As illustrated in Figure 7, the rotor may be one of several
"zero-clearance" designs, a rigid "fixed-clearance" type, or, in the case of
tapered rotors, an adjustable clearance construction may be used. The
clearance is the space between the shell and the periphery of the circle
described by the rotor blade tips. One vertical design includes an optional
residence time control ring at the end of the thermal surface to hold back
(and thus build up) the film thickness.
Mechanical construction of units with a fixed clearance rotor requires
machining of both the shell inside diameter and the outside diameter of the
rotor to assure concentricity and dynamic balancing of the rotor, and
consideration of the effects of differential expansion of rotor and shell.
Although clearances may vary in a narrow range determined by viscosity,
surface tension, and thermal conductivity of the material handled, they
usually fall in the range of 0.08 to 0.25 cm (0.03 to 0.10 inches). Such
small clearances emphasize the importance of the machining operations.
Adjustable clearance is available in the tapered shell units by moving
the rotor in or out with respect to the fixed position of the shell (see
Figure 8). This offers some advantage in a pilot plant or commercial unit
where a wide variety of materials will be processed.
"Zero-clearance" or wiping blades are spring-loaded or free-floating and
.are forced against the wall by centrifugal force as the agitator rotates.
This design is primarily for materials that remain in n liquid state at the
final concentration. Kontro, however, has designs suitable for materials that
will be evaporated to the solid state.
PERFORMANCE
Heat-transfer rates between the liquid and the wall (termed its
"U-value"), in most situations, determine the size and effectiveness of
thin-film equipment. Figure 9 shows overall U-values of several dozen
products, differing in latent heat of vaporization, heat conductivity,
viscosity, boiling-point rise, and surface tension. These data were developed
using saturated steam or high-temperature heating mediums. The U-values
include the resistance of a 0.63 cm (1/4 in.) stainless steel wall or a 1.3 cm
(1/2 in.) stainless steel clad wall. A U-value can be used to calculate the
temperature of the heating fluid necessary to achieve a specified flow of
overhead product.
27
-------
I I
i 1
I
Figure 6. Cross section of thin-film evaporators.
28
-------
I
Ztr* C*«'*ocr
Carbon or Ttfton Atp
FaH Clearance
U>« Vhcosity
CI*
-------
titltnol
P'OduCt inlet
pnd double
mechomcof
foce icon
(bothtiKfs!
Keotlng medium inlet
/
^TTP
ttC*ntnfu-
jgol loom
andcn-
Uotnment
sep.fotor
Figure 8. Kontro horizontal design has adjustment for rotor-shell clearance.
T 0<
1 f
\ / ,
/ \
\
—
Steom is heotmg medium:
1 Cur»»
X
A Concentrotion of osueotS
solutions
too
C Distillation at organic*
0 Stripping of 10" t>o
-------
The heat-transfer coefficients are grouped into main applications to
facilitate preliminary evaluations. U-values for concentration or
distillation show the characteristic increase with increased heat flux, due to
the additional circulation of the formed vapor bubbles (nucleate boiling
usually takes place in thin-film evaporators).
A direct comparison of the Figure 9 U-values with coefficients for
conventional equipment would not be too meaningful, since products handled in
falling-film evaporators usually have low viscosities while the values in
Figure 9 are for viscosities ranging up to 10,000 cp under operating shear
(equivalent to 100,000 to 400,000 cp without subjected shear, as measured by
standard instruments in the laboratory).
In many applications of the agitated thin-film evaporator, mass transfer
(not heat transfer) determines the size and capacity of the equipment.
Deodorization, low-boiler stripping, and dehydration are mass transfer
controlled processes. The stripping of volatile organics from hazardous waste
falls within this category, if it is desired to remove the volatiles to low
residual levels.
A volatile component has to be transported within the film to the
interface, then vaporized. Transport of the volatile component within the
film is accomplished by molecular diffusion or by eddy diffusion. Molecular
diffusion, the only possibility in ncnagitated laminar flow, is extremely slow
and decreases linearly with increasing viscosity of the film liquid. Eddy
diffusion can be influenced and increased by adding turbulence to the film.
Values of diffusivities in agitated thin-film evaporators are on the order of
10~6 m2/sec, or 1,000 to 10,000 times greater than the molecular diffusivities
achieved in nonagitated evaporators.
Deodorization is an extreme stripping operation where only traces of a
low-boiling odor impurity in a feedstock must be removed. Again, it is the
diffusion process, not the thermal load, that determines the required surface.
Deodorization can be further facilitated by using a stripping gas (or steam)
to lower the partial pressure of the low-boiling impurities.
Typical operating characteristics for various applications of thin-film
evaporators are listed in Table 10.
SUPPLIERS
There are six major producers of agitated thin-film equipment in the
United States: Buflovak Division of Blaw-Knox, Chemetron Division of Cherry
Burrell, Luwa, Kontrc, Pfaudler, and Artisan. The locations of these
companies are presented in Table 11. The Luwa concern had the first
production units onstream in 1946 in Switzerland. The Pfaudler design is
based on development work by Arthur Smith. Buflovak introduced its design
with a bottom drive in the mid-fifties. The tapered shell designs of Kontro
(Figure 8) evolved from the standard cylindrical designs. Basic design
configurations furnished by the six producers are shown in Table 12. This
table also relates liquid-vapor flow characteristics to design and location of
the vapor separator. Countercurrent flow occurs when the vapor is removed
31
. . . ~~ — -
-------
TABLE 10. TYPICAL OPERATING CHARACTERISTICS FOR THIN-FILM UNITS
SIZE: 1 to 430 square feet of heat transfer surface.
CAPACITY: Steam heated:
- Water evaporation, 60 kW/m2 (50,000 Btu/(hr) (sq ft))
- Organics distillation, 63 kW/m2 (20,000 Btu/(hr) (sq ft))
Hot oil heated: organics distillation, 25 kW/m2
(8,000 Btu/(hr) (sq ft))
OPERATING PRESSURE: Standard (full vacuum to atmospheric pressure)
Special (any positive pressure)
HEATING STEAM IN JACKETS: Up to 1.4 MPa (200 psig)
MAXIMUM HOT-OIL TEMPERATURE: Up to 350°C (650°F)
LIQUID THROUGHPUT: Up to 900-1,100 kg/m2 (200-250 lb/(hr) (sq ft))
PRESSURE DROP (VAPOR FLOW): 0.5 mm Hg
OVERHEAD TO BOTTOMS SPLITS: Up to 100 to 1
RESIDENCE TIME: Uncontrolled, 3 to 10 sec. Controlled. 3 to 100 sec.
PRODUCT VISCOSITIES; Up to 10,000 cp. at operating conditions
BLADE TIP SPEED: Nonscraping blades, 9 to 12 m/sec (30 to 40 ft/sec)
Scraping blades, 1.5 to 3 m/sec (5 to 10 ft/sec)
RECOMMENDED MAINTENANCE: Twice a year, more often when processing extreme
products.
32
-------
TABLE 11. MAJOR AGITATED THIN FILM EVAPORATOR PRODUCERS
Blaw-Knox (Buflovak Division)
Cherry Burrell (Chemetron Division)
Luwa
Kontro
Pflaudler
Artisan Industries, Inc.
Buffalo, NY
Louisville, KY
Charlotte, NC
Orange, MA
Rochester, NY
Wai than), MA
TABLE 12. CONFIGURATIONS OF COMMERCIAL AGITATED THIN FILM UNITS
Configuration
Manufacturer
Vertical, cylindrical shell
integral separator
Vertical, cylindrical shell
external separator
Vertical, cylindrical shell,
internal separator and condenser
Horizontal, tapered shell,
integral separator
Vertical, tapered shell,
integral separator
Horizontal, cylindrical shell,
external separator
Luwa
Chemetron
Buflovak
Pfaudler
Kontro
Kontro
Artisan
33
Liquid-vapor flow
Countercurrent
Countercurrent
Cocurrent
Separated
Co- or
Countercurrent
Co- or
Countercurrent
Co- or
Countercurrent
-------
from the top of the evaporator and the bottoms are removed from the bottom;
both the liquid and vapors flow in the same direction in co-current flow. In
separated flow, the vapors are condensed in an internal condenser.
LUWA CORPORATION
Luwa is a major producer of thin-film evapcr - in the United States.
Plants A and C (Section 8) used Luwa stills to pro.-. waste. The following
section discusses Luwa stills, but the general princi, as may apply to other
producers.
Luwa manufactures vertical thin-film evaporators with cylindrical heating
bodies (see Figure 2). The body of each Luwa thin-film evaporator is mace in
sections, including one or more jacketed thermal sections, a top or vapov*
head, and a bottom discharge cone. The thermal body sections must have t
uniform bore for precise rotor clearance. The sections are rolled, machi.iec,
and honed (or bored), and then are assembled as a unit.
The general design specifications for standard Luwa evaporators are:
Pressure: Full vacuum to about 15 psig
Temperature: To 340 degrees C (650 degrees F)
Heating:
Sizes:
Steam to 1.4 megapascal (200 psig)
Dowtherm to 1.0 megapascal (150 psig)
0.13 to 40 M2 (1.4 to 430 ft2)
of surface
Figure 10 lists the specific design data for all standard Luwa
evaporators.
At the drive end of the rotor, a double mechanical seal is most often
supplied, although single mechanical seals can be used satisfactorily. A
coaxial design like the Crane 151 seal, which is more durable and accommodates
more vibration, has a typical life expectancy of about 4 years. Because of
torque and some axial loading by the multiple belt drive, a self-aligning
roller bearing is recommended; it also has a life of about 4 years. The
bearing and seal can be inspected and serviced without removing the top cover
or rotor.
Reasons for premature failure are:
Double Mechanical Seal
- ioss of cooling fluid
- solids or abrasives in the seal coolant
- chemical attack on elastomeric 0-ring
- abnormal vibration
34
-------
APPROIIUATE DIMENSIONS
IN INCHES (MILLWCTERS)
tVAPORATOft
IKiOEL
LN-0012
LN-0050
LN-0100
IN-0200
LN-0350
IN05CO
LN-0750
LN-1050
IN- 1400
LN-1800
IN 2400
LN-3200
LN-4000
A
44
(1117)
72
I1830i
95
(2413)
128
(3250,
160
(457?,
204
(5181,
243
I6300)
294
(74671
320
181281
34?
(88' 4 1
38'
1966?
435
111050'
472
fi W;
B
3
(82!
7
(ISO.
10
12501
14
(360i
17
(430-
20
(503)
24
(600,
28
(710,
34
(850i
39
COOO)
47
(!20Ci
55
(14001
67
(17001
C
30
(762)
3P
1951)1
54
'•3721
64
(2-33.
!25
IS"*}.
!36a't
ISO
145301
220
I558S'
250
(63501
27i
<66M|
331
("650,
344
(873-1
3'3
19475)
D
34
uJB63i
&'
I2057)
114
(2896,
135
(3429,
132
ii9;2
217
!55i2i
270
16850)
295
(7503i
341
(B6S'I
35£
19300,
43J
(102KH
457
("SOT)
492
(125001
C
10
(2551
20
(500)
23
1585)
27
16851
48
H2'0l
49
(1250)
59
(1500)
67
(1700)
87
(22001
99
(2500)
r.s
(3030,
138
(35001
168
(4250i
HEAT
TRANSFER
SURfACE
SO FT
(•"I
1 4
(0 13)
54
(OSCl
10£
HOi
216
(201
377
(35,
538
(50)
807
(75)
H30
(1051
1507
(140|
1937
-(180)
2532
(240'
3«3
(320)
4304
(400)
APPROXIMATE WEIGHT L3S |«G-
EVAPORATOR
MODEL
LN-0012
LN-OC5CI
LN-0100
LN-0200
LN-0350
LN-0500
LN-07SO
LN-1050
LN-1400
LN-1800
LN2400
LN-3200
LN-4000
COMPLETE ASSEMBLE
EMPTY
143
(651
770
(350,
1015
(460)
1740
(790)
3525
(1600i
4520
:2050i
6950
(3150,
13200
16000)
17200
(7800)
23100
(10500)
33'00
f'StJOO)
48500
122000)
63900
129000,
FLOODED
150
,68,
esc
!385i
1170
I53C'
2270
(1030)
4400
(2000)
6050
I2V50I
9700
14400)
19000
(6600)
2SOOO
(113501
35600
(16150)
5H50
(23200-
79600
(36100)
111400
(50500,
TOP COVER
WT* ROTOR
55
(25)
155
(701
275
(125)
480
(218)
1060
1483)
1150
1520)
1870
(850!
24^5
11100'
4125
(18701
5600
(2630)
8400
(3803i
11000
ISOOO.
15650
JACKCT PRESSURE
AT
SATURATED
STEAM
200
200
150
150
150
ISO
150
150
150
150
150
125
125
AT
«SO
F
150
150
100
100
100
100
100
100
100
100
100
100
100
ROTOR
DRIVE
MOTOR
STAND
ARDHP
1 5
30
to
75
75
1C
15
20
20
25
30
40
SO
A. Overall unit height
B. Inside dUme'er
C. Support elevation
lo bottom nozzle
0, Clearance from
support elevation
fo' removing roto"
E. Maximum width
Figure 10. LUWA evaporator design data.
35
-------
Self-Aligning Roller Bearing
- inadequate lubrication
- mixing of lubricant types
- sporadic lubrication which could permit water or other fluids
into bearing
- abnormal vibration
The internal bearing most commonly used at the discharge end is a nonload
bearing design. The full weight of the rotor is supported at the top, and the
spider-type bearing assembly only centers the rotor. A hardened pin on the
rotor fits into a carbon cup bushing in the center of the spider. This
bearing design is preferred t»-. an external bearing to eliminate a mechanical
seal and bearing at the dirty end of the machine. This design often employs a
small amount of lubricant: oil, process liquid, water, steam, or air.
Typical life of the carbon bushing is 2 years. The bearing can be inspected
or serviced by simply removing the bottom cone.
Premature failure of the bearing can be caused by the following:
loss ot lubricant,
poor lubricating properties of lubricant, or
process upsets with liquid backing up into rotor area.
Original equipment rotors last typically 10 to 30 years with little or no
maintenance.
Capital Costs
Listed in Table 13 are 1984 budget prices for four standard Luwa
evaporators with 316 L stainless steel wetted parts.
Teble 14 compares the cost of the evaporator alone to the system
installed cost for two cases. In the case designated "simple," the heating
medium is assumed to be steam and the system is controlled manually. In the
case designated "sophisticated," the heating medium is a recirculating thermal
fluid and the system is completely automated. The cost of a heat generator
(steam boiler) is not included.
Table 14 gives a very generalized example based, however, on actual
projects involving agitated thin-film evaporators. From Table 14 we can say
that installed cost could range from twice the evaporator-only cost to as high
as four times, depending upon the degree of sophistication.
1 Operating Costs
As an example of operating costs, assume that a hazardous waste
containing 2 percent by weight volatile organics, 88 percent by weight water,
and 10 percent by weight solids is to be processed in a steam-heated thin-film
evaporator to remove the volatile organics. Assume that the limiting factor
36
-------
TABLE 13. BUDGET PRICESa--LUWA EVAPORATORS
Model number
LN-0012
LN-0100
LN-0500
LN-1400
Heating
surface
mz (sq ft)
0.13 (1.4)
1.0 (10.8)
5 (53.8)
14 (150.7)
Budget price
(1984)
$ 16,000
45,000
120,000
210,000
aBudget prices are for used equipment.
TABLE 14. COST DISTRIBUTION OF AGITATED THIN FILM UNITS3
Evaporator
Components as % of evaporator
Main auxiliaries (condenser, pumps,
vacuum system, controls)
Piping, fittings (materials only)
Structural frame
Installation (foundation, erection,
piping, wiring, insulation)
TOTAL INSTALLED COST
Evaporator as % of installed cost
Simple
100
30
10
5
100
245
41
Sophisticated
100
150
20
5
100
375
27
ATI, 1984; does not include steam boiler.
Sophisticated has automatic controls^ simple is manually operated.
37
-------
in the degree of separation is a solids content of 50 percent by weight, above
which the concentrates will not flow from the evaporator.
In this situation the residual level of organics in the concentrates will
depend upon the vapor pressure of the organics, the viscosity of the
concentrates, the speed of the evaporator rotor, the heating temperature, end
the operating pressure, but residual levels of 1,000 ppm and less could be
expected for low-boiling organics. Assume a continuous feed flow rate of 40
liter? (10 gallons) per minute is to be processed. Table 15 lists estimated
operating costs and assumptions.
SOURCES OF EMISSIONS TO THE ENVIRONMENT
A thin-film evaporator system has three streams which exit the
system—the distillate which was evaporated and then condensed back to the
liquid phase, the concentrates stream which was not evaporated, and the
noncondensible gases which are vented from the condenser.
In the case of volatile organics removal from hazardous waste, the
distillate >ill contain the organics which have been removed from the waste.
Depending upon water content and miscibility, it may be possible to separate
the organics and water by decanting. The organics can be incinerated or
reused, depending upon purity. In cases where fractionation of the organics
into individual components is desired, vapors from the thin-fi'im evaporator
can be used as feed for a fractionation column. In some cases, it may be
desirable to pass the distillate of the thin-film evaporator through activated
carbon to remove the organics by adsorption. The bottoms product from the
evaporator may be stripped of organics to levels which will allow it to be
discharged to a municipal wastewater treatment system with the associated
emissions of volatiles to the atmosphere.
The exhaust vent of the condenser contains noncondensible gases
(primarily air) and potentially, a small amount of organic vapors wnich may
pass through the condenser without condensing. This exhaust can be further
processed by passing it through activated carbon to remove the organics by
adsorption.
The exhaust vent of the condenser is connected to a vacuum pump when the
evaporation is under vacuum. The vacuum pump vent is then a source of
potential emissions. At startup, the vent emissions are expected to be
greater than during normal operation.
APPLICATIONS TO PRETREATMENT
Thin-film evaporators can be used to remove volatile organic components
from waste streams. The bottoms of the evaporator will have less volatiles
than the feed, either because of the reduction in concentration or the
reduction in the volume (or both). For mixed (water and organics) waste
streams, a thin-film evaporator can selectively remove the water (the organics
are high boiling) to improve the Btu value of the bottoms so that they c?n be
more easily incinerated, or selectively remove the organics (the organics are
low boiling) so that the organics can be recycled or burned.
38
-------
TABLE 15. ESTIMATED OPERATING COSTS FOR VOLATILE REMOVAL
FROM. HAZARDOUS WASTE
Assumptions:
Feed rate
Feed composition
Feed temperature
Bottoms concentration
Steam pressure
Heat transfer coefficient
Luwa evaporator model
Steam cost
Electricity cost
Labor cost (includes overhead)
Operators requited
Operating period
Operating costs:3
Steam
Electricity
Operating labor
Spare parts
Maintenance
Total
Unit costs:
S/L waste treated
$/L VOC removed .
$/Mg VOC removed0
40 L/min (10 gpm) continuous
10% solids, 2% organics, 88% water
42°C (60°F)
50% solids, less than 1,000 ppm
organics
1 M pascal (150 psig), saturated
4 MJ/hr/nT/°C (200 Btu/hr fr°F)
LN-1400
$4.4/Mg ($2/1,000 Ib)
$0.05 kWh
$20/man-hour
0.25 (automated system)
8,760 hr/yr
$85,000/yr
7,500/yr
45,000/yr
10,000/yr
10,000/yr
$157,500/yr
0.0075
0.37
462
Does not include laboratory analyses and residual disposal costs.
Assumes a density of 0.8 g/cm .
39
-------
In the removal of volatiles from aqueous or mixed hazardous wastes, the
removal efficiency of the volatiles and the residual volatile concentration
will depend upon waste viscosity and concentration, the boiling points of the
volatiles, and evaporator operating pressure and temperature. However,
residual volatile concentrations of less than 1,000 ppm have been achieved
routinely in similar applications, and less than 100 ppm is possible if
conditions ore optimal.
For complete separation of close-boiling components by distillation, a
fractionation column of adequate design can be added to an agitated thin-film
evaporator, which serves as a reboiler. A thin-film evaporator with vapors
flowing countercurrent to the thin liquid film can be expected to have a
fractionation effect of 1.25 to 1.5 theoretical plates, as opposed to the
single plate maximum efficiency of a conventional still-pot reboiler.
40
-------
r.
SECTION 6
FRACTIOMAT10N DISTILLATION AS A METHOD FOR PRETREATMENT
OF HAZARDOUS WASTE STREAMS
INTRODUCTION
Distillation is an operation which has been in wide use throughout the
chemical and petroleum industries for many years. In numerous cases,
distillation is the only feasible method for separation of components in
liquid or gas streams. Distillation systems vary in complexity from simple
batch processes to very sophisticated multiple column units, with the system
configuration depending upon the objectives of the operation.
By far, the most numerous industrial applications of distillation have
been for purification in chemical manufacturing and in processes involving
internal solvent recycle. The increasing difficulty and cost fjr disposal of
chemical wastes, combined with the rising cost of raw materials during recent
years, has made distillation more attractive as a process for recovery of
organic solvents from waste streams which otherwise would be. discarded.
PROCESS DESCRIPTION
Distillation is broadly defined as the separation of more volatile
materials from less volatile materials by a process of vaporization and
condensation. In engineering terminology, the separation of a liquid from a
solid by vaporization is considered evaporation, and the term "distillation"
is reserved for the separation of two or more liquids by vaporization and
condensation.
Basic distillation involves application of heat to a liquid mixture,
vaporization of part of the mixture, and removal of heat from the vaporized
portion. The resultant condensed liquid, the distillate, is richer in the
more volatile components, and the residual unvaporized bottoms are richer in
the less volatile components. Most commercial distillations involve some form
of multiple staging in order to obtain a greater enrichment than is possible
by a single vaporization and condensation operation.
Simple distillations use a single equilibrium stage to obtain separation
and may be either batch or continuous. Simple continuous distillation (also
called flash distillation) has a continuous feed to an equilibrium stage; the
liquid and vapor leaving the stage «re in equilibrium. Flash distillation is
used in applications where a crude separation is adequate. The component
separation in simple distillation is limited by thermodynamic partitioning
constraints, and multiple staging must therefore be used to increase the
component separation.
41
ia^
-------
Multiple staging is achieved by returning part of the condensate to the
top of the column under such conditions that this reflux is brought into
intimate contact with the vapors on their way to the condenser. Either a tray
or a packed column is normally used to provide adequate gas-liquid interfacial
area for mass transfer. The degree of component separation for a given system
configuration is dependent upon the operating conditions, the number of
stages, and the amount of reflux. In continuous distillation, feed is
constantly charged to the column at a point between the top and bottom trays.
•The section of the column above the feed point is called the rectifying or
enrichment section and serves primarily to remove the heavier component from
the upflowing vapor; it enriches the light overhead product. The portion of
the column below the feed point is called the stripping section and serves
primarily to remove or strip the light component from the downflowing liquid.
i < The stripping section thus serves mainly to purify the bottoms product. In
{ i batch distillation, where fractionation is primarily for enrichment of the
t j overhead product, only rectification is involved since there is no stripping
{ - section. Consequently, it is possible to obtain a distillate of high purity,
I '• but the recovery of the less volatile components from the bottoms is poor
I during the removal of the pure component. As the concentration of the more
| volatile component diminishes in the overhead vapor, the reflux ratio must
f increase to maintain a constant distillate composition. A point will be
| reached at which time the overhead vapor contains proportionally more of the
I less volatile components and it becomes necessary to divert the distillate to
an intermediate product vessel so that the desired bottoms concentration can
be achieved by further distillation without contaminating the low-boiVing
producfwhich was initially distilled.
i Batch Distillation
i
} A typical batch distillation operation (see Figure 11) involves charging
\ the feed batch initially to a still. The still is heated usually by steam
coils, utilizing total reflux until the boiling point of the feed is reached
and the column temperatures have stabilized (column head temperatures are the
boiling point of the more volatile component or azeotrope). Takeoff of
overhead product is then started. As the distillation proceeds, the more
volatile component is continuously depleted from the still, building up in the
distillate receiver. The temperature of the feed batch will rise during the
distillation, approaching the boiling point of the less volatile components
near completion. As the distillation proceeds and the concentration of the
more volatile component increases in the overhead vapor, the reflux ratio must
be increased in order to maintain a constant distillate composition. An
alternative may be to maintain a constant reflux if distillate purity
requirements of the more volatile component in the bottoms are required; it
will be necessary to switch to an intermediate product tank and continue
distillation after reflux can no longer maintain an acceptable distillate
composition.
Continuous Distillation
A typical continuous distillation system (see Figure 12) consists of a
reboiler, tray or packed columr, condenser, accumulator, and associated pumps,
piping, and instrumentation. The feed stream enters the column at a point
42
-------
IT"
„._
COLUMN
CONDENSER
HEAT
OUT
REFLUX
OVERHEAD
PRODUCT
BOTTOM PRODUCT
Still with fractionating column: A, still; B, heating coil; C, column; £>, condenser.
Figure 11. Batch distlliation.
-------
between the top and bottom plates. The plate on which the feed enters is
called the feed plate. The section of the column above the feed plate is the
rectifying section while the section below the feed plate is the stripping
section. As the liquid moves down the stripping section, it encounters vapor
moving up the column from the reboiler. The more volatile component is
stripped from the feed, which continues downward to the column bottom and
reboiler. The vapor rising through the rectifying section encounters, liquid
moving down this section as a result of reflux intrcducad at the top of the
column. This enriches the more volatile component in the overhead vapor which
flows from the top of the column through a condenser to an accumulator.
Reflux is pumped from the accumulator back into the top of the column, with
the remainder of the liquid distillate being taken off as overhead product.
The reboiler in a typical continuous distillation process is normally heated
with steam, and the stripped, more volatile component is continuously
discharged from the vessel (see Figure 12).
PROCESS DESIGN CONSIDERATIONS
In order to design or specify an optimum distillation system for a
specified application, the feed stream characteristics and the overhead
product and bottoms product purities must be defined. The feed rate, feed
temperature, feed composition, component relative volatilities, and product
purity determine, to a large extent, the size and cost of the system. The
process design and operating conditions that will provide the optimum
combination of capital and operational costs can be determined.
At offsite TSDFs, a variety of different waste streams will be treated.
Several different columns can be employed at a TSDF, the selection of process
columns depending on waste volume and difficulty of separation.
The number of plates or stages required for a given separation is
dependent on the reflux rate. As the reflux is increased, the required number
of stages falls. An infinite reflux ratio would be required at the minimum
number of stages, and an infinite number of stages would be required at the
minimum reflux required to effect the separation. The optimum process will
obviously lie somewhere in between these extremes. As the reflux rate is
increased, the required number of stages decreases, resulting in a lower cost
for the column, even though the column diameter must increase to maintain an
acceptable pressure drop with the increased vapor flow. However, as the
reflux increases, reboiler steam requirements increase proportionately. Steam
is the single biggest operating cost, accounting for over 50 percent of the
variable cost. In addition, a larger reboiler and condenser are required,
increasing the cost of these components. It has been found that the optimum
reflux ratio generally will be between 1.1 and 1.25 times the minimum
theoretical ratio (King, 1977).
At a TSDF, there is considerable flexibility in the selection of
processing conditions. The reflux ratio, feed rate, and other variables can
be specified to achieve a desired product composition. The feed rate has a
direct effect on the reboiler capacity and the column diameter. The column
diameter must be sized so that the pressure drop across the column will fall
within the optimum operating range. The relative volatility of the feed
44
«i\iBiii^
-------
r,
Vapor
Condenser C
Jk
Rectifying
1
f
a
-"r-- -
1
Accumulator D
" (~ ^=-—
J — ^
>» Q
i 1
r^J ^//t/i
P'ate
Vapor
CuT *-
^^! •
^-1— x/ Liquid [
' \
1
' * r rv acnoms
r-^-U.,.. V roo^-G
1— i > •
1 '
i
Cooler E
H —
~ 'l *
Overhead
product
!>»>fB
u—~-c
mater
earn
orxfensott
Feed
Bottoms
frodxf
CODDDUOUS fractionatiac column with reciifying and stripping sections.
Figure 12. Continuous distillation.
45
-------
r
stream components will directly affect the number of stages required at a set
reflux rate. The feed stream temperature and component physical property
considerations will influence the location of the feed point in the column.
. j EQUIPMENT DESIGN !
After a distillation process has been designed to achieve a specific
j separation for a particular feed stream, there are a number of different
j equipment types available for the system components. Consideration may be
| given to such factors as fouling tendency, corrosivity, throughput,
I versatility, and cost in selecting the best equipment designs for the process.
I The following discussion briefly outlines the many pieces of hardware required
I for a distillation system and the many equipment variations available to meet
; different process conditions. (This section is based on information contained
I in the APV Distillation Handbook.)
i
i
! Column Shells
A distillation column shell can be designed either as a free-standing
module or to be supported by a steel structure. A self-supporting column is
generally more economical. Column fabrication in a single piece (without
I shell flanges) is more economical than with shell flanges, in addition to
I simplifying installation and eliminating potential sources of leakage.
I Columns over 80 feet (24 m) in length have been shipped by road without
| transit problems. Hazardous waste processing may require a flange assembly to
facilitate cleaning.
I While columns of over 3 feet (1m) in diameter normally have been
j transported without trays to prevent dislodgjment and possible damage, recent
j and more economical techniques have been devised for factory installation of
j trays with the tray manways omitted. Manway; are added after the column has
I been erected, and the fitter inspects each tray. The position and number of
] manways are important, especially for systems which require periodic cleaning.
i Packing can be installed prior to shipment in col imns of 20-in. (51 cm)
* in diameter or less which use high-efficiency metal mesh packing. Larger
I columns are packed on site to prevent the packing from compacting during
I transit and leaving voids. Random packing is almost always installed on site.
I
| Additional requirements can include access platforms and interconnecting
I ladders for onsite attachment to free-standing columns.
?
1 Column Internals
j During recent years, the development of sophisticated computer programs
j and new materials has led to many innovations in the design of trays and
1 packings for more efficient operation of distillation columns.
Tray Devices--
There are five basic types of distillation trays: sieve, valve, bubble
cap, dual flow, and baffle trays—each with unique advantages and preferred
46
^ft.-
-------
usages. Sieve and valve type trays currently are most often specified for
tray towers.
The hydraulic design of a tray is a most important factor. The upper
operating limit generally is governed by the flood point although, in some
cases, entrainment also can restrict performance. Entrainment reduces
concentration gradients because some liquid flows up the column, therefore
lowering efficiency. A column also car, flood by downcomer backup. The trays
fill and the pressures increase when the downcomer is unable to handle all the
liquid involved. This can occur with a highly foaming liquid. Flooding is
associated with high tray pressure drops and small tray spacings.
The lower limit of tray operations is characterized by the excessive
liquid weeping from one tray to the next. Unlike the upward transport of
liquid via entrainment, weeping liquid flows in the normal direction, and
considerable amounts can be tolerated before column efficiency is
significantly affected. As the vapor rate decreases, however, a point
eventually is reached when all the liquid •>* weeping, and there is no liquid
seal on the tray. This is known as the dump point, below which there is a
severe drop in efficiency.
Sieve Tray—The sieve tray (see Figure 13) is a low-cost device which
consists of a p'erf orated plate that usually has holes of 0.5 to 2.5 cm (3/16
to 1 in.) in diameter, a downcomer, and an outlet weir. Although inexpensive,
a correctly designed sieve tray can be comparable to other tray styles in
vapor and liquid capacities, pressure drop, and efficiency. Its flexibility,
however, is inferior to valve and bubble cap tr-jys, and it is sometimes
unacceptable for low liquid loadings when weeping has to be minimized.
Depending upon process conditions and alloviable pressure drop, the
turndown ratio of a sieve tray can vary from l.E to 3 and may be occasionally
higher. Ratios of 5 or greater can be achieved only when the tray spacing is
large, the available pressure drop is very high, the liquid loadings are high,
and the system is nonfoaming. For many applications, a turndown ratio of 1.5
is quite acceptable.
It also is possible to increase the flexibility of a sieve tray column
for occasional low throughput operation by maintaining a high reboil and
increasing the reflux ratio. This may be economically advantageous when the
low throughput occurs for a small fraction of the operating time. Flexibility
likewise can be increased by the use of blanking plates to reduce the tray
hole area. This is particularly desirable for initial operation of a column
in situations where it is envisioned that the plant capacity will be expanded
after a few years. There is no evidence to suggest that blanked-off plates
have inferior performance to unblanked plates of similar hole area.
Dual Flow Tray—The dual flow tray is a high-hole-area sieve tray without
a downcomer7The downflowing liquid passes through the same holes as the
rising vapor. Since no downcomer is used, the cost of the tray is lower than
that of a conventional sieve tray.
47
-------
Si
Figure 13. Sieve tray column (left). Bubble tray column (right)
48
-------
IT*
In recent, years, use of the dual flow tray has declined somewhat because
of difficulties experienced with partial liquid/vapor bypassing of the two
• phases, particularly in large diameter columns. The dual flow column also has
| a very restricted operating range and a reduced efficiency because there is no
I cross-flow of liquid.
Valve Tray—While the valve tray dates back to the rivet type first used
in 1922, many design improvements and innumerable valve types have been
introduced in recent years. The wide selection of modern valve types
(see Figure 14) provides the following advantages:
1. Throughputs and efficiencies can be as high as sieve or bubble cap
trays.
2. Very high flexibility can be achieved and turndown ratios of 4 to 1
are easily obtained without having to resort to large pressure drops
at the high end of the operating range.
3. Special valve designs with venturi-shaped orifices are available for
duties involving low pressure drops.
4. Although slightly more expensive than sieve trays, valve trays are
very economical in view of their operating superiority.
5. Since an operating valve is continuously in movement, the valve tray
can be used for light-to-moderate fouling duties. APV has
successfully used valve trays with brewery effluents containing
waste beer, yeast, and other materials with fouling tendencies.
Bubble Cap Tray—Although many bubble cap columns (Figure 13) still are
in operation7 bubble cap trays rarely are specified today because of high cost
factors and the excellent performance of the modern valve-type tray. The
bubble cap, however, does have a good turndown ratio and is suitable for low
ij liquid loadings.
- I
I laffle Tray—The liquid flows down the baffle tray column by splashing
from one"baffTeto the next lower baffle. The gas or vapor rises through this
curtain of liquid spray.
Although the baffle type tray has a relatively low efficiency, it can be
useful in treating waste flows when the liquid contains a high fraction of
solids (see Figure 14).
Packings—
Packing is the most economical method of contacting liquid and gas
streams in distillation columns, particularly small diameter columns. Most
packings can be purchased from stock on a cubic-foot basis. In addition, the
mechanical design and fabrication of a packed column is quite simple (see
Figure 15). Packing is limited in waste treatment functions because of the
tendency to foul and, in APV experience, because of less predictive
performance at low liquid loads or high column diameters because of liquid
distribution problems.
49
-------
^p
(left) Special two-stage valve
with lightweight orifice cover
for complete closing.
(below) Two typical general
purpose valves which may be
used in all types of
services.
Figure 14. Typical valves used in valve tray columns.
50
-------
I iniriri in
^-•*~
Liquid ^^
Distributor
1 iquid
Redistributor
_->,
• * * * • *f *
.",••'.'•.''•'
.* . '* • *. » ' " '
* , . . * * •
. • • . '
-r —
<* * * .
* * •• . • *
"•*.**
« • *
* . .
«. - . * *
*
•
_...„__ Parking
— Packing Support
r
Liquid out
Figure 15. Illustration of packed column internals.
51
-------
The most widely used packings are randomly-dumped packings such as Rashig
rings, Pall rings, and ceramic saddles (Figure 16), These are available in
various plastics, a number of different metals, and, with the exception of
Pall rings, in ceramic materials. While plastic packings have the advantage
of corrosion resistance, the self-wetting ability of some plastic packings
(such as those made of fluorocarbon polymers) sometimes is poor, particularly
in aqueous systems. This considerably increases the height equivalent of a
theoretical plate (HETP) as compared with equivalent ceramic packagings.
High-efficiency metal mesh packings have found increasing favor in
industry during recent years. One type uses a woven wire mesh that becomes
self-wetting because of capillary forces. This helps establish good liquid
distribution as the liquid flows through the packing geometry in a zig-zag
pattern. If properly used, high-efficiency structured packings can provide
HETP values in the range of 15 to 30 cm (6 to 12 inches). This can reduce
column heights, especially when a large number of trays is required. Such
packings, however, are very expensive and each application must be studied in
great detail.
With both random and high-efficiency structured packings, considerable
attention must be given to correct liquid distribution. Certain types of
high-efficiency packing are extremely sensitive to liquid distribution and
should not be used in columns over 2 feet (0.6 m) in diameter. Positioning of
these devices and the design of liquid distribution and redistribution are
important factors that should be determined only by experts,
Instrumentation
One of the most important..requirements of any distillation system is the
ability to maintain the correct overhead and bottoms compositions from the
column by means of proper controls and instrumentation. While manual controls
can be supplied, this approach rarely is used today in the United States.
Manual control involves the extensive use of rotameters and thermometers
which, in turn, involves high labor costs, possible energy inefficiency, end,
at times, poor quality control. Far better control is obtained through the
use of pneumatic or electronic control systems.
Pnuematic Control Systems--
The most common form of distillation column instrumentation is the
pneumatic-type analog control system. Pneumatic instruments have the
advantage of being less expensive than other types, and since there are no
electrical signals required, there is no risk of an electrical spark. One
disadvantage is the need to ensure that the air supply has a very low dew
point (usually -40°C (-40°F)) to prevent condensation in the loops.
Electronic Control Systems—
Essentially, there are three types of electronic control systems:
1. Conventional electronic instruments,
2. Electronic systems with all explosion-proof field devices,
52
-------
i Katcnip rmg>
wddlc
let Pill ring*
I/
it< Berl uddlr (rfiC>cloheli» spira! rinp
^^
^. ^
I/1 Le\Mnf nnp
Figure 16. Types of packing typically utilized in packed columns.
53
-------
3. Intrinsically safe electronic systems.
A clear understanding of the differences between these electronic control
systems is important.
Most distillation duties involve at least one flammable liquid which is
being processed in both the vapor and liquid phases. Since there, always is
the possibility of a leak of liquid or vapor, particularly from pump seals, it
is essential for complete safety that there be no source of ignition in the
vicinity of the equipment. While many instruments such as controllers and
alarms can be located in a control room removed from the process, all local
electronic instruments must be either explosion proof or intrinsically safe.
With explosion-proof equipment, electrical devices and wiring are
protected by boxes or conduit that will contain any explosion that may occur.
In the case of intrinsically safe equipment, barriers limit the transmission
of electrical energy to such a low level that it is not oossible to generate a
spark. Since explosion-proof boxes and conduits are not required, wiring
costs are reduced.
For any intrinsically safe system to be accepted for insurance purposes,
Factory Manual (FM) or Canadian Standards Association (CSA) approval usually
must be obtained. This approval applies to a combination of barriers and
field devices. Therefore, when a loop incorporates such instruments from
different manufacturers, it is essential to ensure that approval has been
obtained for the combination of instruments.
Reboiler
Although there are many types of reboilers, the shell and tube
thermosyphon reboiler is used most frequently. Boiling within the vertical
tubes of the exchanger produces liquid circulation and eliminates the need for
a pump. A typical arrangement is shown in Figure 17.
For certain duties, particularly when the bottoms liquid has a tendency
to foul heat transfer surfaces, it is desirable to pump the liquid around the
heat exchanger. Since boiling can be suppressed by use of an orifice plate at
tha outlet of the unit, fouling is reduced. The liquid being pumped is heated
under pressure and then is flashed into the base of the column where vapor is
generated.
An alternative approach is the use of a plate heat exchanger as a forced
circulation reboiler (Figure 18). With this technique, the very high liquid
turbulent flow induced within the heat exchanger through the use of multiple
corrugated plates holds fouling to a minimum. Meanwhile, the superior rates
of heat transfer that are achieved reduce the surface area required for the
reboiler.
Condensers
Since most distillation column condensers are of shell-and-tube design,
the design engineer has the option of condensing on either the shell or tube
54
-------
r
dase of
column Liquid 3t
I 11 vapor
Bottoms
product
Steero
Shell tube
heat exchanner
Liquid
Typical shell and tube
thermosyphon reboiler arrangement
Figure 17. Shell and tube reboiler.
Figure 18. Plate hpat exchanger reboiler.
55
-------
side. From the process point of view, condensation on the she!i side is
preferred since there is less subcooling of condensate and a lower pressure
drop is required. These are important factors in vacuum duties. Furthermore,
with cooling water on the tube side, any fouling can be removed more easily.
Tube-side condensation, on the other hand, can be more advantageous
whenever process fluid characteristics dictate the use of more expensive,
exotic materials. Capital costs of the unit then may be cut by using a carbon
steel shell.
Preheaters/Coolers
The corrosion characteristics of the waste stream dictate the selection
of plate or shell-and-tube preheaters and product coolers. If the VOCs do not
excessively degrade or c.wel1 gasket materials, a plate heat exchanger is an
extremely efficient preheater. Heat can be transferred from the tops and
bottoms products for this purpose. The plate heat exchanger can be easily
disassembled for cleaning. Heat transfer to aromatic VOCs such as benzene and
toluene normally is accomplished in a series of tubular exchangers.
Vent Condenser
It is normal practice in some distillation systems to use a vent
condenser after the main condenser to serve as an air pollution control
system. Usually of the shell-and-tube type, the vent condenser will typically
have 1/10 the area of the main unit. Chilled water is used to cool the
noncondensible gases to about 7-10°C (45-50°F), although some plants use
process cooling water as a "back-up" system.
Distillation is used to process many VOC fluids that are highly
flammable. It is desirable that explosion-proof (Class 1, Group D, Division
1) pump motors be used. Centrifugal pumps are reliable and can economically
provide the required flow capacities and operating pressures. The bottoms
product of the distillation unit or the reboiler may, however, contain
particulate material necessitating the use of positive displacement pumps.
EQUIPMENT AND SYSTEM SUPPLIERS
A number of companies manufacture distillation equipment and provide
complete package units for specific applications. Most distillation systems
are custom designed because of the large number of variable process factors
that must be taken into consideration. These companies usually have computer
capability for complete system design as well as pilot plant facilities of
varying capability.
Most companies also provide complex distillation systems which include
multicomponent units. In these, solvent extraction, carbon adsorption, and
distillation processes may be integrated in solvent recovery or stream
purification operations.
56
-------
r
In most cases, complete package systems can be provided in preassembled
modular or skid-mounted units. This will usually result in significant
overall installation cost and time reductions.
Some of the major suppliers for distillation systems used in solvent
recovery and VOC removal operations are listed below:
APV Equipment, Inc.—pilot plant facilities for evaporation.
Aqua-Chem, Inc.--pilot plant facilities for evaporation.
Artisan Industries, Inc.—pilot plant facilities for distillation,
solvent extraction.
Chem-Pro Corp.—pilot plant facilities for distillation, stripping,
extraction.
DCI Corp.—provides standard package systems for distillation and
live steam stripping.
Glitsch, Inc.—pilot plant facilities for distillation, scrubbing,
stripping, extraction.
The Pfaudler Company—pilot plant facilities for distillation.
Vara International, Inc.—experts primarily in integrated
adsorption/distillation systems for solvent recovery from gas
streams.
This is only a representative list of equipment suppliers and is not intended
to be a complete listing of all companies providing equipment or systems. The
locations of the companies are presented in Table 16.
COST
Capital Cost
The cost of a distillation system is dependent upon such design variables
as the size and type of reboiler, column height, column diameter, column
internals, degree of automation, and materials of construction. These design
considerations must be determined for the feed stream to be processed and the
component separation which is required. Changes in component relative
volatilities, feed rates, and required product purities can effect large
differences in system costs.
Equipment installation costs vary with the type and size of equipment,
the geographical location, and the site cnaracteristics. Peters and
Timmerhaus (1968) present a general range of installation costs :
-------
n
TABLE 16. REPRESENTATIVE LIST OF DISTILLATION SYSTEM SUPPLIERS
APV Equipment, Inc.
Aqua-Chem, Inc.
Artisan Industries, Inc.
Chem-Pro Corporation
DCI Corporation
Glitsch, Inc.
The Pfauldler Company
Vara International
Tonawanda, NY
Milwaukee, WT
Waltham, MA
Fairfield, NJ
Columbus, OH
Dallas, TX
Rochester, NY
Vero Beach, FL
58
-------
F"
M
f \
K '
indicated. However, APV Distillation Handbook refers to a typical installed
system cost as being 1-1/2 to 2 times the equipment cost.
Custom-built, preassembled process systems or process modules is an
option being offered by most system suppliers and is claimed to provide
significant savings in cost and time. An overall cost savings of 25-40
percent for a preassembled unit is projected in Chemical Processes Brochure,
p. 200, and the reduction in manhcurs and time length for the project are
depicted graphically in Figures 19, 20, and 21.
In one APV Equipment, Inc., distillation system application (described in
APV Distillation Handbook with supplemental information via telephone
conversation with Dr. Cooper, APV), ethanol is recovered from at: aqueous waste
stream containing about 3-percent ethanol in addition to ?;uspende-i sludge and
yeast solids. One major problem is that the liquid tends to foul heat
transfer surfaces. A paraflow plete (corrugated plate) heat exchanger
reboiler, which is less susceptible to fouling than many other types of heat
transfer equipment, is utilized to minimize this problem. The plate
corrugations enhance the liquid turbulence, which assists in shearing the
fouling deposits off the surface. In addition, the paraflow design is easily
cleaned in place and car. be readily opsned if fouling becomes excessive. The
efficiency of the column trays, however, has not been affected by the fouling
tendency of the liquid. APV attributes this primarily to the use of valve
trays as the gas-liquid contacting device. The small valves in the holes en
the trays are continually moving up and down and there is some rotation, thus
helping to prevent buildup of foulant in the area around the holes. The
system performance and cost are illustrated by the following process
information:
Feed rate 265 L/min (70 gal/min)
Feed temperature 100°C (211°F)
Ethanol concentration (feed) 3 percent v/v
Ethanol concentration (distillate) 95 percent v/v
Ethanol concentration (bottoms) <0.02 percent v/v
System equipment cost $1,050,000
(1-1/2 times equipment cost assumed)
Annualized system equipment cost $171,000
(10 years, 10-percent interest)
The bottoms product containing water, solids, and no more than 0.02
percent v/v alcohol, can be readily utilized in related plant byproduct
processes.
Operating Costs
The total operating costs will vary significantly from a small batch
distillation system to a very high throughput continuous process. The total
59
i-- X.
-------
Engi neeri r.g
Procurement
Assembly
SMp
Startup
—i—
10
Weeks
20
30
4O
Figure 19. Time schedule for Chem-Pro preassembled distillation units
(Chem-Pro, 1985}
Engineering
Procurement
Installation Contract
(Bid and Award)
Assembly
Startup
—i—
10
20
Weeks
40
Figure 20. Time schedule for Chem-Pro field-erected distillation units.
(Chem-Pro, 1985)
-------
*.JW «**«W"'W"--'>?>>* W$* r^r^tsrirv^f^' «•
"•"•"tTTS
Man-hours (Thousands)
ooooooooo
f
I;
If ,
cr>
n>
re i—
ro cr
o o
r+ -S
ft)
Q.O
O
Q.3
_i.-O
(/> Qj
n-1
O -h
=)
O
c =r
3 rp
-••3
CO "O
—3
o
ro -i
3 fo
I CU
-D O
3 to
01
• §-
C»o.
ui
. — .en
• 3
o.
ro
o.
a
a
S
Purchasing and Expediting
Construction Supervision
L
-------
cost for processing a large feed volume will be much greater than for the
smaller operation, due primarily to the much greater steam usage. On a unit
cost basis, however, the steam usage becomes the predominant variable cost as
the feed volume becomes large, arid other costs such as labor and maintenance
become less significant. In a small batch operation that may require
considerable operator attention, the unit labor cost can be fairly significant
(due to low volume, high labor) while steam usage would be equivalent on a
unit cost basis to a high-volume process.
In the case of the APV system described in the previous section, the
operating cost may be determined .f some basic assumptions are made. The
following shows the results of the calculation procedure along with the
assumptions employed:
Operating Conditions
Feed rate
Operating hr/year
Ethanol concentration (feed)
Eth-mol concentration (distillate)
Ethanol concentration (bottoms)
Steam rate, per Ib effluent processed
(APV Distillation Handbook, typical for
dilute alcohol streams)
Operating Costs
1. Annual
Steam
Cooling water and electrical
(10 percent of steam)
Operator (0.25 operators at $15/hr)
Maintenance
(10 percent of equipment cost)
Total operating costs
160 L/min (7 gal/min)
8,400 hr
3 percent
95 percent
0.02 percent maximum
6.09 L (0.2 Ib)
$ 58,000
5,800
31,500
70.000
$165,300
2. Unit Volume
Effluent processed
Ethanol recovered
Total Cost
1. Annual
Operating cost
Depreciation and interest
Total cost
$0.025/1 ($0.0065/gal)
$0.177/L ($0.216/gal)
$165,300
171.000
$336,300
62
-------
Iff
2. Unit Volume
Effluent processed S0.035/L ($0.0093/gal)
Ethanol recovered $1.18/L ($0.311/gal)
S1.475/mg ethanol recovered
SOURCES OF EMISSIONS TO THE ENVIRONMENT
The potential for VOC emissions to the air from distillation processes
would include losses via condenser vents, accumulator tank vents, and storage
tank vents.
Emissions from the condenser vent should be primarily noncondensible
gases. However, if the condenser should become overloaded for any reason, VOC
emissions to the air could become significant. A vent condenser may be used
after the main condenser in order to minimize the amount of volatiles emitted.
Emissions from operating and storage tanks would be displacement losses
of equilibrium vapor from the liquid holdup.
APPLICATIONS TO PRETREATMENT
Distillation is applicable to the pretreatment of hazardous waste to
remove VOCs, either as a primary treatment or in conjunction with solids
separation in a thin-film evaporator. Solids should be removed from the waste
stream, either by decanting the liquid from a settler or by processing the
waste in a thin-film evaporator. Solids which remain in the waste in the
distillation column can foul the plates and require process downtime for
cleaning, as well as reducing column efficiency. The distillation column can
be directly connected to the thin-film evaporator so that the vapors from the
evaporator are then fed to the column.
Distillation serves to separate VOC components from each other, VOCs from
oils, or VOCs from water. The separation of components serves to increase the
value of the organic phase, to improve the energy content of a waste, or to
separate components that cannot be easily separated by a thin-film evaporator
or a steam stripper.
Multiple distillation columns are used at TSDFs to provide flexibility
for various types of separations that a TSDF may require. Wastes vary in
volume, the number of trays needed for separation, and in the reflux ratios
needed for separation of the components. The design of the distillation
columns at a TSDF generally should focus more on flexibility of operation
rather than optimum design for a specific application.
Distillation by batches is commonly carried out at TSDFs to creat batches
of the waste which are collected. Continuous distillation is more appropriate
for large volumes of waste generated continuously, which is not variable with
time.
Distillation as a pretreatment process for removal of VOCs from hazardous
waste streams would have potential application in processing (1) liquid
63
-------
•f*-
organic wastes and (2) aqueous organic waste streams. In the case of liquid
organic wastes, a product reclamation value would be necessary in order to
justify the cost of distillation. With no appreciable recovery value, other
treatment methods such as waste heat recovery or incineration would likely
prove more economical. Distillation is currently widely used for solvent
recovery from organic waste streams at commercial and industrial treatment,
storage, and disposal facilities (TSOFs). The processes may or may not use
rectification depending upon the waste stream composition and the desired
degree of purity of the recovered product. Distillation of organic waste
streams usually results in a bottoms residue consisting of sludge or heavy
organics for which there is no use. The most common means of disposal for
these residues is as fuel in cement kilns. In some cases, distillation
bottoms are landfilled.
Distillation of aqueous organic streams may be feasible when an organic
compound is present in concentrations that would result in an appreciable
recovery value or where the combination of organic recovery and wastewater
purification is of significant value. The organic compound would normally be
more volatile than water for distillation to be considered for such
applications. For VOC removal from an aqueous stream, great care is normally
taken to ensure that the aqueous phase has been stripped to a very low VOC
concentration (less than 0.1 percent) to facilitate disposal of the water.
The aqueous phase is often acceptable for disposal in a municipal wastewater
treatment facility if low concentrations of residual organics can be obtained.
Otherwise, the aqueous residue would require disposal in evaporation ponds or
landfarms.
64
-------
BI-
SECTION 7
FIELD TEST RESULTS: THIN-FILM EVAPORATORS
Thin-film evaporators were investigated at three hazardous waste
recycling plant sites: Plant A, Plant B, and Plant C. At Plant A. a high-
boiling organic was separated from a waste (together with a few percent YOC).
At Plant B, there was a mixture of VOCs being recovered from a waste, with a
range of VOC volatility. At Plant C, a highly volatile VOC (acetone) was
recovered from a waste. Plants A and B were operating under vacuum, and Plant
C was at atmospheric pressure. Thin-film evaporators typically only have a
few minutes residence time for the waste material in the evaporator, and grab
samples of the feed, bottoms, and product are believed to characterize the
process during the time that the samples are taken. Thin-film evaporators are
described in Section 5 of this report.
Thin-film evaporators are used in many solvent recovery operations.
Typically, 80 percent of the volatile materials can be recovered, with a
sludge containing residual VOC obtained at the bottom of the evaporator.
THIN-FILM EVAPORATOR FIELD EVALUATION AT PLANT A
A 1-day site visit was conducted at Plant A in order to collect
information on engineering design and cost, along with selected process
samples (Allen, 1984a). The primary activity at Plant A is the recovery of
organic wastes and contaminated chemicals. The company also engages, to a
lesser extent, in waste management for some firms.
The recovery and purification processes involve three VOC recovery
systems:
1. One Luwa thin-film evaporator;
2. One batch fractionatior. distillation column;
3. One continuous feed frcctionation distillation column.
Support facilities include a concrete drum storage and management area, a
cooling water system, an activated sludge wastewater treatment system, an
oil-fired boiler system for steam generation, and a main building providing
housing for offices, laboratories, and locker rooms.
Process Description
The Luwa thin-film evaporator processes organic wastes from the
furniture, chemical, dry cleaning, and paint industries. Hastes processed
include furniture finishing wastes and other organic wastes which could
65
ass&hi; ,.irt
-------
contain sludges. The sludge would include paint films, particulates, and*
insoluble organic materials.
The major requirement for processing in the Luwa still is that the still
bottoms are acceptable for incinerators. This means that the bottoms irust
have an energy content greater than 38,000 kJ/L (100,000 Btu/gal), contain
less than 1-percent chlorides, and be suitable for pumping. The process
economics require that typically greater than 60 percent of the waste be
recovered as volatiles from the still overheads.
There were no apparent mechanical safety problems associated with the
Luwa still, apart from the moving drive belt on the still rotor.
Waste and contaminated solvents and organic byproducts are received at
the site in drums and bulk shipments. Extensive laboratory analyses are
performed in order to ensure consistency with the manifest identification.
The standard operation is to process each batch of chemicals through the Luwa
Evaporator during which 70 to 95 percent of the material is stripped off as
overhead product. The amount stripped off is selected so that the bottoms
product is acceptable in heat value and viscosity for offsite consumption as
fuel. The overhead product may or may not be further refined through
fractionation distillation, depending upon the intended end-use. Distillation
bottoms are shipped offsite and utilized as fuel in cement and expanded
aggregate kilns.
o
? 'The Luwa Thin-Film Evaporator System at Plant A consists of a 4.0 m (43
ft ) heat transfer surface Luwa Evaporator, entrainment separator, condenser,
feed pump, bottoms pump, distillate pump, vacuum system, and instrumentation
(Figure 22). Steam or sometimes hot oil is used as the heating medium. Steam
is typically controlled at about 38°C (100°F) above the boiling point of the
distillate. Evaporator feed rate and system pressure (vacuum) control are
determined based on the material being processed. A typical feed rate is 23
L/min (6 gpm), but may be set over a range of from 4 to 45 L/min (1 to 12
gpm). The cooling water for the overhead product 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 (40"0 gpm). The system is operated under reduced pressure with
either of two vacuum pump systems, a Kinney or Nash, being utilized. The two
Kinney pumps at Plant A are positive displacement oil-sealed units which can
be used to obtain operating pressures down to about 50 urn Hg absolute (28
inches Hg vacuum). The single Nash pump is a water-sealed displacement pump
which can obtain operating pressures of about 350 mm Hg absolute (16 inches Hg
vacuum).
The most persistent maintenance problem involves the bottom rotor shaft
bearing. It is estimated that this bearing is replaced about eight times per
year with a downtime of 6 hours per occurrence. A maintenance overhaul is
performed every 4 or 5 years during which time the rotor edges are redressed.
Process Effectiveness
Based on discussions with Plant A personnel and the engineering judgment
of the authors, the thin-film evaporator at Plant A can be used in the
following applications:
66
... „-.
-------
Feed Stream
3.8 L/min
Hot Oil
205°C
1-
Evaporator
•- Water 20 c
Cooling water
Hot Oil * 1.JOO L/min
fcl -r- 1fi°r
"-" Vacuum "" ' *" Ver:L
Product
3.6 L/min
146UC 0,2 L/min
Figure 22. Solvent recovery: Mixed chlorinated xylenes.
-------
1. Removal of VOCs from organic streams which may contain viscous high
molecular weight organics or solids.
2. Removal of VOCs from sludges such as insoluble organics and
participate 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 than that of water.
5. Removal of water from streams containing relatively high
concentrations of volatile organics of lower volatility than v/ater
(water removed as overhead product).
During the visit to Plant A, a "batch" consisting of mixed chlorinated •
xylenes was being processed through the thin film evaporator (Allen, 1984a).
This waste treated at Plant A represents a class of waste oils containing a
snail amount of solids and approximately 5-percent VOCs. Since this material
contained few solids, approximately 95 percent of the feed was being taken
overhead, with bottoms being acceptable for fuel. The purpose of the
treatment was to recover the chlorinated xylenes.
The data obtained from samples of the feed, the bottoms, and the product
shortly after process startup are presented in Table 17. The headspace
analysis demonstrated that the volatile organic material was removed by the
Luwa still from the bottoms. The following removal effectiveness was
estimated (based upon a material balance using the headspace analysis and
95-percent product recovery, Appendix F):
Methylene chloride 99.91
Chloroform >99.99
1,1,1-Trichloroethane >99.5
Toluene 95.4
\ I Freon TF 80.0
| Based on headspace analysis, the concentrations of the volatiles appear to be
| reduced in the bottoms of the thin-film evaporators. The vapor pressures of
I the more volatile compounds are reduced more than 90 percent by the treatment.
j Process Residuals
There are several potential VOC fugitive emission sources associated with
the operation of the Luwa still. Storage tank emissions are possible to a
limited extent. (The feed tank was covered ouring the test visit.) Leaks
from the transfer lines are possible, together with any potential spills of
materials. The vacuum pumps remove the vapors from the condenser and the
vented vapors are an air emission source. The air emissions from the vacuum
pump system are expected to be relatively low during the normal operation of
.the system, but would be much greater if the condenser were unable to condense
some of the vapors, if air was leaking into the system, or if the still were
68
-------
TABLE 17. PLANT A THIN-FILM EVAPORATOR WASTE COMPOSITIONS
AND HEADSPACE ANALYSIS
Methyl ene chloride
Chloroform
Feed
Liquid
composition
(vol. %)
2.0
1.5
1,1,1-trichloroethane 0.7
Toluene
Mixture of high
1.3
boiling hydrocarbons 94.4
Freon TF
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
Bottoms
Headspace
analysis
(mg/L)
0.03
<0.01
<0.01
0.03
ND
0.24
not detected.
aBottoms solid upon cooling and no solids analysis was performed.
bat 25°C.
TABLE 18. PLANT A VACUUM PUMP VENT GAS ANALYSIS
i i
Chloroform
1,1,1-Trichloroethane
Trichl oroethyl ene
Tetrachi oroethyl ene
Toluene
Ethyl benzene
Xylenes
Process vent
10 minutes
(mg/L)
2.4
3.1
0.87
1.0
27.1
0.31
1.41
Process vent
5 minutes
(mg/L)
17.8
3.05
0.59
1.04
23.5
0.52
2.16
Duplicate
Process vent
5 minutes
(mg/L)
6.1
2.35
0.61
1.01
25.8
0.45
1.9
Field
blanks
1 2
_ _
-
-
— —
-
-
0.1
Sample period.
69
-------
operating with an overhead rate that was greater than the condenser could
handle. When a new waste type is introduced into the still, the bottoms are
collected in an open 50-gallon drum for part of the run. These bottoms are
hot and are a fugitive emission source.
The Nash water ring seal vacuum pump will absorb some of the VOCs in the
condenser gas into the water. A stream of the water containing VOCs would be
transferred to water treatment. The still bottoms are incinerated and the
condensate is either sold as a product or is processed further at the plant. •
The volatile materials, methylene chloride and chloroform, were present
in the feed stream at somewhat higher concentrations than in the product, and
were not present to a significant extent in the bottoms. This suggests a
possible loss of some of the more volatile materials to the atmosphere through
the vacuum pump vent. The exhaust of the vacuum pump did contain some
volatile materials during startup. The following removal effectiveness was
estimated using a material balance from the information presented in Table 17
(based upon the headspace analysis and 95-percent recovery), according to
boiling point:
Estimated Removal
Effectiveness
Methylene chloride (b.p. = 40.7°C) 55%
Chloroform (61.3°C) 97%
1,1,1-Trichloroethane (74°C) -21%
Toluene (110°C) -17%
A negative sign indicates a VOC recovery above 95 percent during
processing. Although there is an expected error of 15 percent or less in the
data, and even greater error in differences and estimates based on gas
partitioning, the data do suggest that the lower boiling VOCs may be subject
to much higher process losses than higher boiling VOCs. In addition, the vent
losses are expected to be higher upon startup.
Carbon adsorption tubes were analyzed for VOCs. Five tubes were
analyzed, one 10-minate sample (85 mL/min), two 5-minute samples, and two
field blanks. The analysis of the gas samples of the vacuum pump discharge
vent is presented in Table 18. Chloroform and 1,1,1-trichloroethane were
present in the vent discharge with statistically significant quantities of the
higher boiling VOCs (toluene, ethyl benzene, xylene). The absence of
methylene chloride captured on the carbon tube and the apparent insensitivity
of the quantity of chloroform captured to the length of sampling time suggest
that the quantities of the more volatile chlorinated compounds may be greater
than reported in Table 18.
These data suggest that Luwa stills operating under a vacuum can have
potentially significant VOC emissions. The emission rate would depend en the
operating conditions of the still. The data reported in Tables 17 and 18 were
taken during startup and the concentrations observed were expected to be lower
after the transitional period of startup.
70
-------
r v
Process Cost
The equipment at Plant A was purchased second-hand and, therefore,
capital costs must be estimated for a replacement system. This estimate is
provided in Table 19. The estimate does riot include cost of a steam boiler.
Table 20 provides cost, estimates of the thin-film evaporator at Plant A.
The typical unit costs for operating this process were 0.0279 $/L waste
treated and 0.033 $/L (41.1 $/Mg) of VOC removed as shown in Table 20.
THIN FILM EVAPORATOR FIELD EVALUATION AT PLANT B
Process Description
The primary activity at Plant B is the reclamation of 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 recycled back to the generator with the remainder
being marketed to suitable end users.
Processing equipment include two Votator agitated thin-film evaporators,
two distillation reboilers, eight fractionation columns, and one caustic
drying tower. Support facilities include 97 storage tanks (one-million-gallon
capacity); two warehouses containing dyked concrete pads for drum storage; an
analytical laboratory; gas-fired steam generation; and an office building. A
fleet of tractors and vacuum tankers is maintained for transporting solvents
and chemicals to and from the plant.
The wastes processed by Plant B are from the chemical, paint, ink,
recording tape, adhesive film, automotive, airlines, shipping, electronic,
iron and steel, fiberglass, and pharmaceutical industries. The types of
chemicals recovered include the following VOCs: alcohols, ketones, esters,
glycols, ethers, chlorinated solvents, aromatic hydrocarbons, petroleum
naphthas, freons, and specialty solvents. Contaminated organic chemicals and
solvents are received in bulk and drum shipments and processed for reclamation
and recycle.
Plant B is able to remove the VOCs from still bottoms, coating residues,
obsolete paints, and inks using thin-film evaporators.
All waste material is either processed first in the thin-film evaporator
or the distillation reboilers. Approximately 90 percent of the incoming
shipments are processed through one of two Votator thin-film evaporators
during which about 80 percent of the material is stripped off as overhead
product. The limiting factor for the amount of liquid which can be recovered
is that the bottoms product must be acceptable in heat value and viscosity for
offsite consumption as fuel. The overhead product may or may not be further
refined through fractionation distillation, depending upon the intended end
use. Thin film evaporator bottoms are shipped offsite and utilized as fuel in
cement kilns.
71
-------
TABLE 19. PLANT A REPLACEMENT CAPITAL COSTS (LUWA EVAPORATOR SYSTEM}'
Luwa Evaporator Model LN-0500
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
a!984 Cost (per Ray Danaher, LUWA Corp., July 23, 1984)
b5.00 square meters (53.8 square feet)--4.00 square meters (43 square feet)
model used at Plant A, no longer available.
TABLE 20. 1984 PLANT A OPERATING COST*
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/hr])
Electrical (45 hp)
Cooling Water 1,500 L/min (400 gpm)
Overhead
Evaporator Bottoms Disposal ($0.05/L [$.22/gal])
Schedule Production
Utilization (88 percent of schedule, 24 hr/day
ANNUAL COSTS
Annual Operating Cost
Capital Recovery Factor (Avg. 10 yr at
10 percent
Total
UNIT COSTS
$/L waste treated
$/L organic recovered (density = 0.8)
$/Mg organic recovered
23 L/min (6 gpm)
85 percent
$10/hr, S66,000/vr
$25,000/yr
S9,000/yr
$24,000/yr
S8/hr, $6C,000/yr
31.70/hr, $ll,000/yr
$4/hr, $26,000/yr
$100,000/yr
$78,000/yr
310 days/yr
273 days/yr
$217,200
35,350
0.0279
0.033
aBased on information provided by Plant A for typical application. See
Appendix F for details on calculations.
72
-------
Each of two thin-film evaporator systems consists of a Votator thin-film
evaporator, entrainment separator, condenser, feed pump, bottoms pump,
distillate receiver and pump, and vacuum system. The two VotatorS are 4.65 m2
(50 ft2) and 5.76 m2 (62 ft2), respectively, in heating surface, steam heated,
and generally operated under reduced pressure. The Votator has an external
bottom rotor bearing which apparently works well and is relatively problem
free. Plant B considers their operational experience with Votator evaporators
to be very good with no major problems being experienced in about 2-1/2 years
of service.
The following cases illustrate alternate treatment technologies at
Plant B.
Case 1:
Case 2:
Case 3:
A mixture of MEK and alcohol in water is fractionated by
distillation. The distillation bottoms are predominantly water
with most of the VOCs removed (99 to 98 percent removal).
Water collects on the top of a separator tank when the waste is
evaporated and condensed. The organic layer on the bottom is
recovered.
Water collects on the bottom of a separator tank, and a light
hydrocarbon layer on the top is recovered.
The waste is steam stripped. If alcohols are present, the
partition of the alcohol is shifted to the xylene and toluene
phase. The organic phase is recovered, and the water is left
in the bottoms.
The Votator unit has the capability to run under vacuum so that lower
processing temperatures c?n be used in the VOC recovery. Degradation
reactions are also reduced by the design of the Votator, since there is an
attempt to minimize the residence time of the feed.
Process Effectiveness
At the time of the visit, a batch consisting of isopropyl alcohol,
xylene, and other VOCs was being processed through the 5.76 m2 (62 ft2)
Votator (Figure 23). The reclaimable product was being stripped off overhead
for further distill;,lion and purification. The bottoms product was maintained
at a concentration suitable for offsite fuel use.
The waste stream being processed with one pass under vacuum in the
Votator contained isopropyl alcohol and xylene. Normally, two passes are used
to process this material, with the more volatile IPA rerroved on the first pass
and the less volatile xylene removed on the second pass. This two-pass
technique would permit more effective capture and separation of the volatiles.
On the batch sample, both IPA and xylene were being recovered on the same
pass. The plant stated that vacuum would not be used normally for the
recovery of VOCs from this waste stream and any vented VOCs would be expected
to be emitted at a lower rate for the system under atmospheric pressure.
73
-------
Feed Stream
Condensate
Condenser
Bottoms Product
Figure 23. Thin-film evaporation: Isopropanol, xylene recovery.
Vent
-------
Samples wire taken at the following points:
Feed stream,
Bottoms product,
Overhead product, and
Vacuum pump discharge (gas).
The analysis of the liquid samples is presented in Tables 21 and 22. The
results from duplicate sampling and analysis is presented in the table. The
vent gas analysis is presented in Table 23. The concentration of VOCs at
equilibrium in the vapor phase is not substantially altered in the bottoms,
although the volume of waste is reduced, hor those vent gas components that
could be correlated with the feed, the concentration of the VOC in the vacuum
pump vent gas is approximately one-half the headspace concentration of the
product.
The process parameters were:
Feed temperature: ambient,
System pressure: 5.5 kilopascal (22-inch vacuum),
Process Temperature: 70°C, and
Steam: 1 megapascal (150 psig).
Process Residuals
The sources of air emissions include storage tanks and the vacuum pump
vent. The Votator system itself is not expected to contribute to air
emissions from leaks, since any leaks would be into the systen since it. is
under vacuum. Any such air leaking into the system would be tvrooved by the
vacuum pump along with some VOC vapors. The rate of venting f,*om the vacuum
pump Is expected to be greater at the beginning of a process run, since there
would be air at atmospheric pressure initially. The vent samples were taken
at the same time that the liquid samples were taken after the process had
stabilized.
The storage tanks were large, fixed roof tanks. The air emissions from
these tanks would be expected to be working losses from the tank vents.
Process Cost
Specific, detailed operating and capital costs were not discussed
primarily because of the relatively large number of, and differences in,
process systems at this facility. In addition, the complexity of the process
operations vary over a wide range, depending upon the particular feed stream
being processed and the intended use of the product.
75
-------
TABLE 21. ANALYSIS OF LIQUID SAMPLES, THIN FILM EVAPORATOR, PLANT B
Feed
Liquid Headspace
composition analysis
(vol. %) (mg/L)
Product Bottoms
Liquid Headspace Headspace
composition analysis analysis
(vol. X) (mg/L) (mg/L)
Isopropyl alcohol
Freon TF
Toluene
Ethyl benzene
Xylenes
38.2
0.6
0.4
11.4
49.2
0.75
38.
0.58
5.5
22.
42.9
0.5
0.3
10.4
45.7
0.69
29.
0.48
10.4
23.
1.6
5.3
0.32
9.C
39.0
Bottoms solid upon cooling and were not analyzed as a liquid.
bat 25°C.
TABLE 22. ANALYSIS OF PRODUCT SAMPLES, THIN FILM EVAPORATOR, PLANT B
Product sample
Liquid Headspace
composition analysis
(vol. X) (mg/L)
Duplicate product sample
Liquid Headspace
composition analysis
(vol. %) (mg/L)
Isopropyl alcohol
Freon TF
Toluene
Ethyl benzene
Xylenes
53.8
0.7
0.4
8.4
34.0
1.1
62.
0.94
5.3
19.
60.3
0.6
0.4
7.0
27.4
1.1
51
0.71
4.8
17.
'at 25°C.
TABLE 23. ANALYSIS OF GAS SAMPLES FROM PLANT B VACUUM CONDENSER VENT
Unidentified (Freon TF?)a
Toluene
Ethyl benzene
Neohexane (Isopropanol?)
Xylenes
Vent sample
(mg/L)
10.24
0.4
2.3
2.23
6.8
Vent sample
(mg/L)
32.4
0.47
2.1
5.7
6.21
There were analytical problems identifying components.
76
-------
Plant B stated that it would not be economical to fractionate waste
streams with less than 6 to 8 percent reclaimable organics. Some order of
magnitude costs provided were:
$0.26 per liter ($1.00 per gallon) operating cost when organic is
stripped as the overhead product.
$0.40 per liter ($1.50 per gallon) operating cost where water is stripped
overhead with the organic being the bottoms product.
It is estimated that 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 the thin-film evaporator bottoms to a cement kiln
are approximately $0.05 to $0.08 per liter (20 to 30 cents per gallon).
THIN-FILM EVAPORATOR FIELD EVALUATION AT PLANT C
A 1-day site visit was conducted at Plant C in order to collect
information on engineering design and cost, along with selected process
samples (Allen, 1984c).
Process Description
Plant C uses thin film evaporation for the reclamation and recycle of
organic solvents. The primary activity at Plant C is the reclamation of
organic solvents and con' ?.;ninated products for recycle or sale. Specialty
solvent blends which are optimized for specific client uses are also produced.
The solvent recovery processes include two VOC recovery systems: a Luwa thin-
filn evaporator and one SRS, Riston Batch Distillation.
Support facilities include a drum storage and management area, a cooling
water system, an oil-fired boiler for steam generation, an air compressor, a
bench-scale Rodney-Hunt thin-film evaporator, storage tanks, and associated
pumps and piping.
The wastes processed by Plant C are from the chemical, plastics, paint,
adhesive film, electronics, and photographic industries. The types of
chemicals recovered included chlorinated solvents, freons, ketones, and
aromatic hydrocarbons. There is currently no vacuum system and consequently
no capability for operating the Luwa evaporator under reduced pressure. This
precludes processing of high-boiling compounds such as naphtha and xylene.
Contaminated organic solvents are received it the plant in drums, and
each drum is coded for identification in the drum storage area. A bench-scale
distillation is normally done on each incoming shipment in order to determine
recovery, efficiency, and materials characteristics. The standard recovery
procedure is to process each batch of chemicals through the Luwa thin-film
evaporator during which 70 to 95 percent of the material is stripped off as
overhead product. However, nonflammable materials may be processed through an
SRS 120 Riston evaporator instead of the Luwa when the capacity is needed.
77
L%.a 'j
-------
This evaporator was designed for recovery of vapor degreasers (chlorinated
solvents) and would not meet standards required for flammable materials.
? 2
The Luwa thin-film evaporator system consists of a 1.0 m (10.8 ft ) heat
transfer surface Luwa evaporator, enfainment separator, condenser, product
tank, feed recirculating tank, and pumps. Steam is used as the heating medium
with the pressure being set typically between 207 kPa (30 psig) and 552 kPa
(80 psig) depending upon the solvent being processed. The Luwa is operated at
atmospheric pressure. The system process pumps are compressed air diaphragm
pumps. A batch of solvent is charged to a 1,700-liter (450-gallon) feed
recirculation tank. The overhead product from the evaporator is pumped into a
1,700-liter (450-gallon) product tank. The evaporator bottoms are pumped back
to the feed tank, end the feed continues to recirculate until a predetermined
VOC removal is reached. Process steam is generated in a 147 kW (15 hp) oil-
fired boiler.
Process Effectiveness
Based on discussions with Plant C personnel and in the engineering
judgment of the authors, the thin-film evaporator at Plant C can be used in
the following treatment applications:
1. Removal of VOC from organic streams which may contain viscous high
molecular weight organics or solids.
2. Removal of VOC from sludges such as insoluble organics and
particulate solids.
3. Concentration of aqueous sludges such as insoluble organics and
particulate sludges.
4. Removal of VOC from aqueous streams where the VOC volatility is
higher than that of water.
5. Removal of water from streams containing relatively high
concentrations of volatile organics of lower volatility than water
(water removed as overhead product).
At the time of the visit, a batch of contaminated acetone was being
processed through the Luwa evaporator (Figure 24). The reclaimed product was
being stripped off overhead and pumped into a product receiver. The bottoms
from the evaporator was being pumped back to the feed tank for recirculation
through the Luwa. The condenser was vented to the outside of the building
through a 1.25-cm (1/2-inch) pipe. Any emissions from this vent would
normally be noncondensibles; but in the event of a condenser overload, it
would be a source of VOC emission. The top of the feed tank and the top of
the product tank consisted of movable lids which were not airtight, and this
is a potential source of VOC emissions.
Grab samples were taken and the results of the analysis are reported in
Tables 24, 25, and £6. The following sampl?» were taken:
78
-------
Feed Stream t
38°C 1,640 L/hr,
Condensate
Condenser
Vapor
57°C
Steam
30 psig
26°C.
Cooling Water
Product
^.Bottoms Product
1,300 L/hr
20°C
Vent
Figure 24. Solvent recovery: Acetone.
-------
TABLE 24. ANALYSIS OF LIQUID SAMPLES, THIN FILM EVAPORATOR, PLANT C
Feed
Liquid Headspace
composition3 analysis
(vol. %) (mg/L)
Acetone
Freon TF
1,1,1-Trichloroethane
Trichloroethylene
Toluene
Ethyl benzene
jg(ylene
Tetrachl oroethyl ene
74.3
0.1
1.5
0.2
0.5
NO
5.9
0.6
378.0
.0
17.9
0.1
0.3
0.1
2.1
2.4
Product
Liquid Headspace
composition analysis
(vol. *) (mg/L)
82.2
i
2.2
0.3
0.9
0.3
2.0
0.5
383.0
2 0
19.1
0.1
0.2
<0.1
0.2
1.6
Bottoms
Liquid Headspace
composition analysis
(vol. ?>) (mg/L)
60.6 308.0
0.1 1.5
0.9 9.2
<0.1 0.1
0.9 4.1
0.3 0.4
<0.1 <0.1
13.6 5.0
- _^ .^ !»•«•
•••••MPBPWBW**^^^^^^^^^^^^^^^^™^™"
ND = not detected.
Approximately 17 percent of the waste was high boiling organics and resins.
-------
TABLE 25. ANALYSIS OF GAS SAMPLES FROM VENT Or THIN FILM EVAPORATOR, PLANT C
Compound
Concentration in Luwa vent
(mg/L)
Acetone
1,1,1-Trichloroethane
Trichloroethylene
Tetrachloroethy1ene
Toluene
Ethyl benzene
Xvlene
0.04
0.003
.008
.003
.002
0.002
0.005
0.
0.
0.
TABLE 26. ANALYSIS OF GAS SAMPLES FROM PRODUCT RECEIVER, PLANT C
Compound
Air concentrations
Test 1 Test 2
(mg/L) (mg/L)
Air emissions
Test 1 Test 2
(g/sec) (g/sec)
Acetone
Freon TF
1,1,1-Trichloroethane
Trichloroethylene
Tetrachloroethylene
Toluene
Ethyl benzene
Xylene
5.3
0.04
0.41
0.04
0.075
0.009
0.016
0.03
7.2
0.04
0.45
0.013
0.059
0.007
0.007
0.016
0.565
0.004
0.043
0.004
0.008
NS
NS
0.003
0.70
0.004
0.043
ND
0.0057
NS
NS
NS
NS = not significant concentrations above field blank.
81
-------
r
Luwa feed stream,
Distillate,
Residue,
Condenser vent (gas), and
Product tank (gas).
In addition to the grab sample, the following process parameters were
: { provided by Plant C during the field test:
1. Feed rate 1635 L/hr (432 gal/hr)
. i 2. Feed temperature 38°C
• I 3. System pressure 760 torr
• ' 4. Vapor temp. 57°C
r 5. Steam pressure 310 kPa (30 psig)
; 6. Jacket (upper) temperature 132°C
7. Jacket (lower) temperature 107°C
8. Condenser water inlet temperature 20°C
9. Condenser water outlet temperature 25°C
• 10. Distillate rate 344 L/hr (91 gal/hr)
11. Bottoms rate (feed - distillate) 1291 L/hr (341 gal/hr)
. | 12. Luwa drive motor amps 1.1
s
: The waste being processed in the Luwa evaporator was acetone containing
• xylene and low levels of chlorinated solvents. The acetone and other low-
* boiling point compounds were somewhat more concentrated in the distillate and
I xylene was enriched in the bottoms. Because of the requirements to maintain
| 1 the resins in solution, the VOCs in the bottoms at the end of the run were not
;r I substantially different in concentration than the VOCs in the feed (although
I | the volume of waste was reduced by about 70 percent).
Process Residuals
The analysis of the air samples from the process vent indicated that no
significant (relative to the field blank) air emissions were observed from the
process vent. The vent pipe was located perpendicular to the wind flow on the
building exterior; and due to the wind gusts, air flowed alternatively in and
out of the vent (as measured by the Alnor veTometer). No odors were detected
at the vent.
The product storage tank has a loose fitting steel top with a gap of
approximately 8-3 cm (3-1 inch) and was 1 meter (4 ft) wide with «n area of
620 cm (0.67 ft ). An MSA portable pump with 0.42 ml/count was used to draw
: air past a carbon adsorption bed. An Alnor Velometer Jr., Type 8100, was used
[ to measure wind velocities. The wind was variable and five readings were
I taken during each air sampling procedure. The windspeed was 172 cm/sec (340
I ft/min) on the first product storage test and 157 cm/sec (310 ft/min) on the
! second product storage test. The volumes of air sampled were 313 and 320 mL
[ 0" tests 1 and 2 of the product storage tank. Acetone was the major component
I ! 82
-------
lost from the storage tank at 0.56 and 0.70 g/sec. Rased on an emission
factor of 2.5 x 10~= g/mol/cm2 sec obtained from th.- air emissions model of
Thibodeaux and Parker (Spivey, 1984), a velocity of 170 cm/sec (340 ft/min),
and a vapor concentration of 383 mg/L (measured at 25°C in the laboratory),
the emissions for a fully exposed surface were estimated to be 2.0 g/sec,
greater than the measured field rates by a factor of 3.
At the time of the test of the product storage tank, the plant was
planning to replace the current receiver tanks with a more enclosed tank
system.
Distillation residues which cannot be utilized as fuel in furnaces or
cement kilns are being solidified with diatomaceous earth in drums for
landfill disposal.
Process Cost
Equipment and operating costs were not readily available at the facility
at the time of the visit. However, some information which may be relative to
cost was provided.
o The Luwa evaporator was purchased rebuilt, and a condenser was
fabricated to Luwa specifications. In retrospect, the owner would
have purchased a package unit from Luwa because of factory
assistance available with installation and startup operational
problems.
2 2
o The 1.0 m (10.8 ft ) Luwa evaporator was reported to be rated at
378 liters/hr (100 gal/hr) distillate by Luwa. A high rate of 606
liters/hr (160 gal/hr) was achieved one time, with rates of 454
liters/hr (120 gal/hr) ootained frequently. Rates for chlorinated
solvents which is the largest volume-type process, averages above
322 liters/hr (85 gal/hr).
The plant is operated on a two-shift, 5-day work week.
two production operators per shift.
There are
Plant C provided limited process costs for the operation of the thin-film
evaporator. Based on the information obtained from the batch being processed
at the time of the site visit, Table 27 provides operating costs and an
estimate of unit costs for waste treatment. The cost per unit of waste
treated is greater than estimated for Plant A which used a larger unit.
83
-------
TABLE 27. 1984 PLANT C OPERATING COSTS
Feed rate a
Overhead product, percent of feed
Labor
VOC recovery
Maintenance Materials
Disposal by landfill
Disposal by incineration
Steam3
Cooling water
Annual system cost
ANNUAL COSTS b
Annual operating costs
Capital recovery factor (10 yr at
10 percent interest)
Total
UNIT COSTS
$/L waste treated
$/L VOC recovered
$/Mg VOC recovered (density = .8 g/mL)
491 L/hr (130 gal/hr) .
70 percent
S0.61/L ($2.31/gal. waste)
344 L/hr (91 gal/hr)
$0.24/L ($50/55 gal. drum)
+ labor and freight
$0.21/L ($44/55 gal. drum)
+ labor and freight
23 L/min
110 L/min (30 gpm)
$125,000
$150,400
$20,343
$171,000
0.048
0.057
71
Provided by Plant C.
^Assumed to equal 1 man/process x $24/hr x 24 hr/day x 300 days.
84
-------
j SECTION 8
i
FIELD TEST RESULTS: DIRECT STEAM STRIPPING
f '
i i
i Plant D is a hazardous waste TSDF that uses steam stripping to recover
I VOCs from wastes. At Plant D, live steam is injected into a recycling waste
v I material. The steam stripping unit used at Plant D was chosen for assessment
f 1 because it can be commercially purchased and is suitable for treating wastes
I on the site of generation as well as offsite. The batch steam stripper can
| handle small batches of waste; wastes which vary in composition can be
! handled. Some reactive wastes can be processed in the unit more successfully
I i than in a thtn-film evaporator due to the lower processing temperatures. The
I , wastes that ere treated must be able to be pumped in the diaphragm recycle
| pumps: this imposes a limitation on viscosity and solids. When the VOC of
f interest is of low volatility and high waste solubility, the concentration in
f the condensed steam may be lower than in the initial waste. The direct steam
I stripper would not be able to recover the VOC in such a system. When the VOCs
I separate as an organic layer in the condensed steam, direct steam stripping
' can be used for recovery. The direct steam stripping process can be used for
i organic and aqueous batches of waste contaminated with solids, sludges, oils,
••• and grease.
PROCESS DESCRIPTION
I Plant D is engaged in the reclamation of organic solvents for recycle and
f sale. A live steam stripping process is used for organic solvent reclamation.
I The steam stripping system is a package unit manufactured by and purchased
I I from DCI, Inc. (DCI, Inc., 5725 W. 79th Street, Indianapolis, IN 46268).
This system is located inside a building which also contains three 3,800-liter
(1,000-gallon) waste solvent storage tanks and three 3,800-liter (1,000-
gallon) product storage tanks. The building is also utilized for drum
storage. There are five 38,000-liter (10,000-gallon) outside storage tanks
which are used primarily for contaminated solvent and residue storage. An
oil/gas-fired boiler system is used for process steam generation. An
analytical laboratory is maintained in the building housing company offices.
The contaminated crganics processed by Plant D are generated mostly by
the chemical, paint, pharmaceutical, plastics, and heavy manufacturing
industries. The types of chemicals recovered include the following VOCs:
ketones, aromatic hydrocarbons, chlorinated solvents, freons, and petroleum
naphthas. The recovered products ma> be recycled back to the generator or
marketed to suitable end users. Generally, 50 to 70 percent solvent recovery
from the waste stream is expected. Residues from the stripping process are
solidified by mixing with sorbents and shipped offsite to be landfilled.
The steam stripping system (a schematic is shown in Figure 25) consists
of a stripping vessel, overhead vapor condenser, distillate receiver,
85
-------
00
Vent
Recovered
VOC
Storage
Tank
Miscible
Solvent
Tank
Vent
Treated
Waste
Tank
Figure 25. Batch steam stripping process.
-------
^
I I
decanter, miscible solvent tank, product storage tanks, a residue tank and
associated pumps and support facilities.
The steam stripper is a 1,900-liter (500-gallon) stainless steel
horizontal tank containing a steam sparger running lengthwise along the
bottom. A recirculating pump discharging into the steam sparger promotes
steam and solvent contact. Steam pressure to the sparger is normally
controlled at 150 kPa (20 psig).
I I The overhead product condenser is a vertical, shell-and-tube, water-
; cooled condenser which discharges into a distillate receiver. The distillate
receiver is vented through a pipe extending through the roof. The distillate
* is pumped frSW the receiver to a decanter from which the aqueous phase flows
by gravity to a miscible solvent tank, with the organic phase being pumped to
a product storage tank.
The process steam is supplied by an oil/gas boiler rated at 3,900 kW (400
I > HP) and 1,000 kPa (150 psig). The boiler is currently being operated on gas,
•; utilizing about 500 kW (60 HP), and producing steam at 690 kPa (100 psig).
[ The distillate condenser/receiver is vented to the atmosphere where
l emissions were expected to be mostly noncondensibles. However, this is a
; . potential source of volatile organic compound (VOC) emissions should the
• condenser become overloaded for any reason. Emissions from storage tank vents
: would represent operational displacement of equilibrium vapor from the tank
; j headspace.
I j The steam stripper is primarily utilized for solvent reclamation,
although some incidental aqueous stream treatment results. On occasion, the
system has also been utilized for treatment of dilute aqueous streams solely
to reduce the organic content to a level that is acceptable to the local
municipal waste water treatment facility.
The typical operation involves charging a 950-liter (250-gallon) batch of
organic solvent to the stripper, steam stripping the solvent overhead, and
decanting the distillate when the solvent is immiscible with water, which is
the case with most of the solvent processed. A typical batch of 950 liters
(250 gallons) can be processed in one hour. The aqueous phase from the
decanter is collected in a 1,100-liter (300-gallon) miscible solvent tank
(MST). The contents of the MST are ultimately processed again through the
stripper for recovery of residual solvent. In many cases, the aqueous
residual in the stripper after the MST contents have been processed is
suitable for discharge to municipal waste water treatment systems. The volume
of organic residues left in the stripper is approximately 260 liters (70
gallons). They ?-re typically burned in an incinerator or solidified and
landfilled. Aoueous residues from the stripper (which contain soluble
organics) are not suitable for use as fuel, since steam condensate reduces the
heat value to an unsatisfactory level. This residue is currently being
solidified and landfilled.
Processing of solvents which are miscible in water may involve multiple
strippings, giving the effect of fractionation, depending upon the degree of
87
-------
dryness required. The concentration of the solvents in the condensate must be
greater than in the batch for this to occur.
PROCESS EFFECTIVENESS
Plant D stripped waste material by direct injection of live steam into a
waste batch. The process of stripping continued until the desired
concentrations were achieved in the waste. Four batches of waste were
evaluated: (1) an aqueous xylene batch, (2) a chlorinated organic-oil
mixture, (3) a chlorinated organic-water mixture (1,1,1-trichToroethane), and
(4) a mixture of solvents and water. A summary of the waste characterization
and process data is presented in Table 28. The individual components in the
waste are presented in Tables 29, 30, 31, and 32.
The results of analyzing waste material during stripping is presented in
Tables 33, 34, 35, and 36. Only the major compounds detected in the analyses
are reported in these tables; compounds below the detection limit are reported
in the field test report (Allen, 1985d). These results are also presented in
log normal plots in Figures 26, 27, 28, and 29, In general, each of the
curves for the VOC removal by the process was linear; an exponential first
order decay was indicated for each major volatile component. Further time in
the stripper would reduce the VOC concentration. The second batch was an
exception; the concentration of trichloroethane in the second batch did not
initially decay at the exponential rate because the other component (methyl
ethyl ketone) boiled off first.
The aqueous xylene batch (Batch 1) contained organic and water phases and
a small amount cf solids (Table 26). The organic phase constituted
approximately 25 percent of the waste volume and was composed predominantly of
xylene. The aqueous xylene batch initially contained 1,260 L and the final
batch size was 1,420 L. Recovery of 333 L of organic distillate was achieved,
with 248 L of aqueous distillate going to the MST. The steam rate was 250-270
kg/hr and the heatup and stripping time was 2.08 hr. The removal rates of all
the compounds appeared to be approximately the same, with the heavier
materials being somewhat more slowly removed (Table 33, Figure 26). A
detailed evaluation is found later in this section. There was removal of all
the VOCs to less than 300 ppm in the final treated waste.
In the 1,1,1-trichloroethane batch (Batch 2), trfo major components were
removed from the organic waste. These were trichloroethane and methyl ethyl
ketone. The 1,1,1-trichloroethane batch initially contained 897 L of VOC in
oil. The final batch volume was 320 L of oil and water. Recovery of 670 L of
VOCs was achieved as organics, and 400 L of aqueous distillate were generated.
The steam flow rate was 250-300 kg/hr and the heatup and stripping time was
1.72 hr. Methyl ethyl ketone was quickly removed from the process with only
23 ppm remaining after 80 minutes (Table 34, Figure 27). The trichloroethane
was somewhat more slowly removed wvth 4,000 ppm or 0.4 percent residual
trichloroethane in the oil material at the end of the batch. Since the methyl
ethyl ketone boiled off first, the concentration of the trichloroethanc did
not drop initially but appeared to decrease much more rapidly in the later
stripping as the batch temperatures increased.
88
-------
Concentrations Relative to Initial C/C
•n
!-••
cr
61
r»
o
e
EU
in
W
(A
c
a
n
n
H-
O
a
a
n
-------
06
CONCENTRATION RELATIVE TO INITIAL, C/Co
OQ
ro
to
H- 3
O O
=r n>
M 3
O rt
n 1-1
O Ql
(D rr
rt H-
3" O
O 3
3 B
ft
rt
m
CB
rt
O
O
i-n
O
to
w
ta
3
O
ft
H-
O
m
^-N
2
T
C
n>
-------
100,000
C£
~ 10,000
in
a
c
o
o
I
1,000
O Methyl ethyl ketone
1,1,1-Trichlorottthane
60
Tine (minutes)
100
Figure 28. Concentrations in the batch of waste as a function of time:
1,1,1 trichloroethane.
91
-------
10,000
1,000 .
oc,
E
C
o
•rl
C
0)
u
o
10
n Toluene
Acetone
§| Xylene
O Trichloroethane
10
20 30
Time (minutes)
Figure 29. Concentrations in the batch of waste as a function
of time: Mixed solvent batch.
f; -'
92
-------
TABLE 28. WASTE CHARACTERIZATION AND PROCESS DATA
Batch 1 Batch 2 Batch 3
WASTE CHARACTERISTICS
Initial volume (L)a
VOC content {mass percent)
Water content (L)a
Oil content (L)c
PROCESS DATA
Stripping time (min)a
Steam rate (L/min)^
VOC stripping ratea
constant {min~ )
Bottoms VOC content (wt. percent)3
VOC recovered (L)a
Stripper residue volume (L)a
Stripper oil content (L)
Stripper water content (L)a
Condensed steam (L)a
Percent VOC removed
aMeasured.
\fe\T timf\ *+V\-*v*f*f\f4 + r\ c4-*»n*\r>/iv» C«rt
l,260b
18.7°
1,253
f
86
4.3
0.073
0.05
333
1,420
f
1,420
248
99.8
T=.hTo D-l
897
74d
f
233
87
4.5
0.060
0.41d
670
320
233
87
400
99.8
564
18
460
f
57
4.5
0.048
1.2
45
545
f
533
180
93
Batch 4
360
3Je
357
f
33
4.0
0.132
0.04
3.2
349
f
34*
140
99
cNot measured, estimated by material balance, density = 0.866 (refer to sample
calculations in Appendix F).
Volume percent, measured.
eSecond phase measured by volume as added to water.
fNegligible.
measured, estimated by enthalpy balance.
93
-------
TABLE 29 WASTE CHARACTERIZATION OF BATCH 1
(AQUEOUS XYLENE)
Number of Phases
Total solids (mg/L)
Water (weight percent)
Oil (weight percent).
VOC (weight percent)
81.3
Negligible
18.7
Aqueous Phase
PH
Density
VOC Analysis;
Acetone
Isopropanol
Methyl ethyl ketone
1,1,1-tri chloroethane
Tetrachloroethene
Ethyl benzene
Toluene
Xylene
Organic Phase
Density (g/cm )c
Composition:
Xylene
Other aromatics
6.1
1.0
39
960
1,040
170
290
360
86
2000
0.866
aThe solids were in the aqueous phase.
bMost of the VOC was apparently in the organic phase.
cBased on density of toluene, not measured.
94
-------
•"*«
TABLE 30. WASTE CHARACTERIZATION OF BATCH 2
(1,1,1-Trichloroethar.e/Oil)
Number of Phases
1
Total Solids (mg/L) 2,800
pH 4-°
Water (volume percent) -
Oil (volume percent) «
VOC (volume percent) 74
Methyl ethyl ketone (volume percent) 7.5
1,1,1-Trichloroethane (volume percent) 66.0
Density (g/mL) I-2
Estimated from pure components.
TABLE 31. WASTE CHARACTERIZATION OF BATCH 3
(1,1,1-Tri chl oroethane/Water)
Number of Phases 2
Total Solids (mg/L) 130
pH J-8
Water (weight percent) 82
Oil (weight percent)
VOC (weight percent) 18 a
Density (g/mL) 1-°
VOC Analysis (Aqueous Phase)
Methyl ethyl ketone (mg/L) 320
1,1,1-Trichloroethane (mg/L) 180,000
Ethyl benzene (mg/L) 44
Acetone (mg/L) . 290
Isopropanol (mg/L) 37
Estimated as the density of water.
95
-------
TABLE 32. WASTE CHARACTERIZATION OF BATCH 4
(Mixed Solvent/Water)
Number of Phases 2
Description of Secondary Phase Xylene, toluene
Total Solids (mg/L)
PH
Water (weight percent)
Oil (weight percent)
VOC (weight percent)
Density (g/cm )
VOC Analysis (Aqueous Phase)
Acetone (mg/L)
Isopropanol (mg/L)
Methyl ethyl ketone (mc/L)
1,1,1-Trichloroethane (mg/L)
Tetrachloroethene (mg/L)
Toluene (mg/L)
Xylene (mg/L)
130
7.0
97
3.1
1.0
6,500
95
112
2,200
55.
86^
4b
aMeasured volumetrically as added to batch initially.
The low values are thought to be due to incomplete initial mixing in the
stripping vessel. Values were not used in stripping rate analysis. The
recovered VOCs were measured, and VOCs were added to the batch.
96
-------
I :
I1
TABLE 33 WASTE VOC CONCENTRATION DURING STRIPPING:
TABU «. wo XYLFN£)
r.nncentration (mg/L)
TCEA TCE EB Toluene Xylene
Process Sample
time (min) number
Acetone I PA
. _
0
15
64
86
^— i— •-
45-3
45-5
45-6
45-7
39
10
<6
<6
960
640
47
<6
in.
1,040
460
70
34
i
170
99
33
20
_ in
290
230
72
<20
in 11 ii •*
360
560
56
100
bb
32
17
42
£,UUU
480
410
270
IPA = isopropanol
EB = ethyl benzene
TCEA = 1,1,1-trichloroethane
TCE = tetrachloroethene MEK = methyl ethyl ketone
TABLE 34 WASTE VOC CONCENTRATIONS DURING STRIPPING:
TABLE 34.1 §1 J-TRICHLOROETHANE/OIL)
f-
__— — — — — — —
Process time
(minutes)
0
128
55
87
{• i
\ t ,
'—
Sample
number
,.^_— — — — — —
45-22
45-24
45-25
45-26
,
—— — — — -
Concentration (mg/L)
1,1,'1-trichloroethane netny etny keun.c
-
660,000 75,000
690,000a 591
22.0003 <7
4,100 <7
__ .
are probably good to within a factor of 5.
97
-------
TABLE 35 WASTE VOC CONCENTRATIONS DURING STRIPPING:
BATCH 3 (1,1,1-TRICHLOROETHANE/WATER)
Concentration (mq/L)
[
I
Process time
(minutes)
0
i
1 22
i 1
i i 43
t
57
Sample
number
45-29
45-31
45-32
45-33
Acetone
290
71
9
<6
MEK
320 .
50
<7
<7
TCEA
180,000
71,000
28,000
12,000
Ethyl benzene
44
30
24
12
I PA
37
<6
<6
<6
MEK = methyl ethyl ketone
TCEA = 1,1,1-trichloroethane
IPA = isopropanol
TABLE 36. WASTE VOC CONCENTRATIONS DURING STRIPPING:
BATCH 4 (MIXED SOLVENT/WATER)
Concentration (mp/L)
Process time
(minutes)
0
10
19
33
Sample number
45-11
45-15
45-16
45-17
Acetone
6,500
270
32
<6
Toluene
86a
170
56
35
Xylene
4a
900
310
120
TCEA
2,200
930
460
230
aThe low values are thought to be due to incomplete initial mixing in the
stripping vessel. Values were not used in stripping rate analysis. The VOCs
were added to the batch, and the recovered VOCs were measured.
TCEA = 1,1,1-trichloroethane
98
-------
*?* ftw
-------
TABLE 37. HEADSPACE CONCENTRATIONS3 OF VOC AS A FUNCTION
OF THE STRIPPING TIME: BATCH 1 (Aqueous Xylene)
Concentrations (mg/lj
Process time
(minutes)
0
15
64
86
Sample
number
45-3
45-5
45-6
45-7
Isopropanol
0.53
0.16
-------
TABI E 38. HEADSPACE CONCENTRATIONS OF VOCa
"AS A FUNCTION OF THE STRIPPING TIME
Batch 2 (1,1,1-Trichloroethane/Oil)
Process time
(minutes)
0
28
55
87
Sample
number
45-22
45-24
45-25
45-26
Vapor concentrations
Methyl ethyl 1,1
ketone
104
0.82
0.032
0.02
(mq/L)
,1-Trichloro-
ethane
440
460
5.8
0.91
aWaste sample at 25°C.
101
-------
TABLE 39. Kc
'* i '
,%. L r* ' y ; A - ,'fl^
" -*"••'***'.
_ »,-^pir.,^T=,,.,|r.p,,«T»^e-, _. ...^
AOSPACE CONCENTRATION3 OF VOC AS A FUNCTION OF THE STRIPPING TIME:
BATCH 3 (1,1,1-Trichloroethane/Water)
Vapor concentrations (mg/L)
Process time
(minutes)
0
22
43
57
Sample
number
45-29
45-31
45-32
45-33
Methyl ethyl
ketone
11
0.2
<0.02
<0.02
1,1,1- trichloro-
Acetone ethane
0.22 440
<0.1 24
<0.1 4.7
<0.1 2.1
Ethyl
benzene
0.24
0.34
0.29
<0.1
Trichloror
ethyl ene
0.3
0.13
<0.1
<0.1
?Waste sample at 25°C.
DNot detected in liqui
iquid above limit of detection.
n
! 1
o
ro
-------
TABLE 40. HEAOSPACE CONCENTRATION3 OF VOC AS A FUNCTION OF THE STRIPPING TIME:
BATCH 4 (Mixed Solvent/Water)
Process time
(minutes)
0
10
19
33
Sample Methylene.
number chloride
. 45-11 0.65
45-15 <0.1
45-16 <0.1
45-17 <0.1
Concentration (mg/L)
Acetone 1,1,1-Trichloroethane Toluene Xylene
8.5 19 7.6 15
0.33 <0.1 0.47 2.8
0.17 - 0.9 0.14 1
<0.1 1.7 <0.1 0.58
?Waste sample at 25°C.
Methylene chloride was present below the detectability limit in the liquid phase (Table 30),
-------
^
TABLE 41. LINEAR CORRELATION OF THE LOGARITHM OF THE WASTE CONCENTRATION
WITH THE STRIPPING TIME
Batch Component
1 Acetone
1 Isopropyl alcohol
1 Methyl ethyl ketone
1 1,1, 1-Trichl oroethane
1 Tetrachl oroethane
1 Ethyl benzene
1 Toluene
1 Xylene
2 Methyl ethyl ketone
2 1,1,1-Trichloroethane
3 Acetone
3 Methyl ethyl ketone
3 1,1,1-Trichloroethane
3 Ethyl benzene
4 Acetone
4 Toluene
4 Xylene
4 1,1,1-Trichloroethane
Correlation
coefficient
-0.817
-0.988
-0.998
-0.9971
-0.97JJ
-0.85°
-O.M?
-0.84b
-0.999
-0.944
-0.993
-1.0C
-0.9971
-0.950
-0.997.
0.908^
0.977d
-0.990
Stripping rate constants
(min"1)
0.0246
0.0581
0.0393
0.0242
0.0296
0.023
0.0088
0.0180
0.1763
0.0648
0.0806
0.0844
0.0468
0.021
0.280
0.127
0.110
0.0684
(dimensionless)3
7.14
16.9
11.4
7.0
6.47
6.58
2.54
5.22
34.8
12.7
10.0
10.5
5.82
2.59
25.4
5.97
7.8
6.21
aThe dimensionless rate is obtained by dividing the rate constant (min ) by the
ratio of the steam rate (L/min) to the amount of waste (L).
The lower correlation coefficients could be due to the presence of an aromatic
layer in the batch.
The values of concentration below the detection limit were not used in these
calculations; only two concentrations were available for the calculation.
The estimated concentration at zero time was used in the correlation: the
vapor pressure was measured and Henry's constant was calculated from the second
sample set.
104
Bw,
-------
the rates of stripping. The steam rate influences the rate of stripping, so
the rate constants are presented in a dimensionless form to account for the
steam rate and batch size. Theoretically, the dimensionless rate is
numerically equal to Henry's constant for equilibrium partitioning.
The mass flow rate of gas leaving the stripper is approximately equal to
the steam rate, particularly for low concentrations of VOCs in waste water.
The ratio of the mole fraction of a compound in the vapor phase to the mole
fraction in the liquid phase is the equilibrium partition constant, K, in
units of atmospheres per mole fraction. Defining [Ca] as the instantaneous
concentration of a compound, the stripping rate constant (min ) for the
removal of VOC from the batch equals the steam rate, S (moles/sec), times K
(moles of compound/mole steam/moles of compound in batch x moles water in
batch), divided by the batch size B (moles water in the batch).
1 d[Ca] SK
TTaT dt = " B~
The dimensionless rate constant is obtained by multiplying the stripping rate
constant by the ratio of the batch size B to the steam rate. For equilibrium
controlled processes, the dimensionless rate constant equals the equilibrium
partition coefficient.
Henry's constants as reported in the literature for the laboratory
measurement conditions are substantially greater for organics in aqueous
streams than was measured in the laboratory for the wastes treated. The
Henry's constants calculated from the headspace analysis and the liquid
analysis (Appendix A) are compatible with the observed low values of the
dimensionless rate constants. The use of Henry's constants given in the
literature to predict stripping rates would seriously underestimate the times
and costs that were observed in the field.
Although the partition coefficients estimated from the stripping data are
much lower than literature values, the coefficients are similar for the same
components in the various wastes. For example, the value of the partition
coefficient of MEK was 11.4 and 10.5 in the aqueous batches. In the organic
waste, the value of the partition coefficient was 34.8. In the aqueous
wastes, the partition coefficient of 1,1,1-trichloroethane was 7.0, 5.82, and
6.21 for Batches 1, 3, and 4, respectively. To estimate the effectiveness of
steam stripping to remove VOCs, it is desirable to measure the partition
coefficient either by headspace methods or by process evaluation.
PROCESS RESIDUALS
Air Emissions
The air emissions were estimated for two batches, the xylene aqueous
batch (Batch 1) and the production waste trichloroethane batch (Batch 2}. The
results of the concentrations obtained from air samples taken frc-n the
condenser vents are presented in Tables 42 and 43. In addition to the
concentrations in the air samples, seme of the equilibrium vapor analyses
obtained on liquid samples are presented in Tables 42 and 43 for comparison.
105
-------
o
cr>
TABLE 42. GAS PHASE VOC CONCENTRATIONS: BATCH 1 (AQUEOUS XYLENE)
Process
time Sample
(minutes) number
NA
59
63
65
74
NA
NA
46-1
46-3
46-2
46-5
45-10
45-9
45-8
Concentration (mg/L)
MEK IPA TCEA TCE Toluene Xylene
Trip blank <0.1 <0.1 <0.1 <0.1 <0.1 0.3
Condenser vent 8.8 2.2 2.1 1.1 6.4 2.3
Condenser vent 7.6 1.8 1.7 .97 7.0 4.3
Condenser vent 6.8 1.7 1.5 .87 6.6 3.7
Distillate 26 5.5 2.2 1.3 11 17
headspace :
Organic composite
distillate 11 25 2.5 2 13 16
headspace
Aqueous composite
distillate 11.2 1.8 0.43 <0.1 0.16 0.36
headspace
EB
0.14
1.2
1.7
1.6
5.2
5.1
0.65
MEK - methyl ethyl ketone
IPA - isopropanol
NA - not applicable
TCE - tetrachloroethylene
TCEA - 1,1,1-trichloroethane
EB - ethyl benzene
.Collected from the receiver
Collected from the organic product storage tank
Collected from the miscible solvents tank
1
•i
-------
TABLE 43. GAS PHASE VOC CONCENTRATION: BATCH 2
(1,1,1-Tri chloroethane/Water)
Process
time Sample
(minutes) number
-2a
46b
47b
43
44
42
125
49-2
49-3
49-4
49-5
45-27
49-1
45-28
Concentration (mq/L)
Methylene
chloride
Condenser vent
Condenser vent
Condenser vent
Product storage vent
Product storage tank
(lab headspace)
HST tanK vent
MST Tank
(lab headspace)
3
<0
0
0
0
<0
0
.0
.1
.12
.2
.28
.1
.88
MEK TCEA Toluene Xylene
104
0.34
2.4
0.22
5
0.64
74
210
22
85
4.5
460
24
560
0.13
0.10
0.18
1.3
0.18
0.62
<0.1
0.17
0.61
0.63
1.8
0.74
1.4
<0.1
aDuring batch heating, 2 minutes before stripping started.
Midcycle sample.
MEK = methyl ethyl ketone
TCEA = 1,1,1-trichloroethane
I
\
\ 107
-------
The results of air emission measurements for Batch 1 are presented in
Table 42. The concentrations of the equilibrium vapor over the distillate
obtained midway through the process corresponded with the gas concentrations
obtained in the vent duriiig the process. The correspondence of the air
emission concentrations in the vent with the headspace concentrations of the
distillate (obtained at approximately the same time) was better than for
either the product storage equilibrium concentrations or the aqueous composite
distillate concentrations.
The average emissions are expected to be a function of both the rate of
flow and the composition drift of the various components. Table 44 estimates
the air emissions on the basis of the average of the results of the tests on
the three air samples taken from the aqueous xylene batch; it was assumed that
the VOC concentration in the air was constant throughout the batch operation.
The volume emitted from the condenser vent was estimated on the basis of the
data obtained from the integrated flow as determined by dry gas meter readings
(Allen, 1985d). In Batch 1 (aqueous xylene) the dry gas meter recorded a 683-
liter (24.2 cubic foot) flow over a time of 108 minutes. Since the process
time (filling to stop stripping) was 125 minutes, the estimated total volume
during the process was 790 liters. The emission factors were estimated by
obtaining the ratio or" the estimated grams of emissions to the total amount cf
wastes charged into the system and the total amount of volatile materials
recovered. Emission rates are calculated by obtaining the ratio of the grams
emitted to the process time (sec). The data that were used in the
calculations are presented in Table 44. The emission rate was 0.0029 mg/sec
for (aqueous xylene) Batch 1. This value should only be used in health effect
calculations with care since the rates v,ould not necessarily apply to
processing other wastes and does not reflect process downtime. The emission
factors in terms of g/g waste are expected to be mere appropriate for
dispersion modeling calculations.
In Table 43, the air emission results from the 1,1,1-trichloroethane
batch (Batch 2) are presented. The concentrations of the methylene chloride,
methyl ethyl ketone, and trichloroethane at the condenser vent were much
greater at the beginning of the process than at the middle of the process.
The observed concentration results from the vapor of the waste being charged
into the stripper being forced up the condenser vent. This is not unexpected
since the concentrations of these components are much greater at the beginning
of the cycle than at midcycle, as .seen in Tables 34 and 35. Even more
importantly, the concentrations for MEK and TCEA correspond very well with
those at t=0 in headspace analysis (Table 38). The concentrations of VOCs in
the product storage vant and the MST vent were substantially lower than the
concentrations obtained at equilibrium within the storage tank.
Table 45 estimates the air emissions on the basis of the average of the
results of the tests on the air samples taken from the second batch. The
volume emitted from the condenser vent was estimated on the basis of the data
obtained from the integrated flow as determined by dry gas meter readings
(Allen, 1985d). In Batch 2 (1,1,1-trichloroethane/oil) a gas flow of 179
liters was monitored over a 47-minute period; the average gas flow was 3.8
L/min. Two air samples were taken from the condenser vent during the
nridcourse of the process. One condenser vent sample was taken during batch
108
<*»&•&**,
-------
:• i
f !
TABLE 44. AIR EMISSIONS ESTIMATIONS: BATCH 1
(Aqueous Xylene)
Concentration
Volume3 Component (mg/L)
Condenser vent 790 L Methyl ethyl ketone 7.7
Condenser vent 790 L Isopropyl alcohol 1.9
Condenser vent 790 L 1,1,1-Trichloroethane 1.8
Condenser vent 790 L Tetrachloroethene 0.98
Condenser vent 790 L Ethyl benzene 1.6
Condenser vent 790 L Toluene 6.7
Condenser vent 790 L Xylene 3.5
Total VOC emissions
Emission factor (g/g waste treated) 1.5 x 10
Emission factor (g/g VOC removed)6 5.7 x 10"
a683 L/108 min x 125 min = 790 L. (See text)
Average of condenser vent concentrations in Table 36.
^Product of concentration and volume.
1.26 Mg waste fed to process.
e333 kg solvent recovered.
Total
Emissions
(g)
6.1
1.5
1.4
0.77
1.3
5.3
2.8
19.2
109
-------
TABLE 45. AIR EMISSIONS ESTIMATIONS: BATCH 2
(1,1,1-Trichlorcethane/Oil)
Time
(min)
46-47
46-47
Volume
391 Lb
391 Lb
Component
Methyl ethyl ketone
1,1, 1-Trichl oroethane
Total VOC Emissions
Emission
Emission
factor (g/g waste treated)
factor (g/g VOC
removed)6
Concentration
(mg/L)
1.36C
54°
21.6 g
2.0 x
2.7 x
Emissions3
(9)
0.53
21.1
10"5
ID'5
?Product of concentration and Volume.
D103 min x 3.8 L/min = 391 L (see text).
^Average of measurements at 46 and 47 minutes (see Table 37).
1,100 kg waste basis.
e800 kg VOC basis.
110
SBSw^-ta-flM*.^^
-------
heating. Initially, the concentrations of the VOCs were substantially greater
than during process mid-cycle (factor of 10 or greater). The volume of gas
emitted during the initial heating is assumed to equal the batch size. These
initial emissions were not used to estimate the emissions because of the lack
of measured gas flow rates during the initial period.
The emission factors were estimated by obtaining the ratio of the
estimated grams of emissions to the total amount of wastes charged into the
system and the total amount of volatile materials recovered. Emission rates
are calculated by obtaining the ratio of the grams emitted to the process tine
(sec). The data that were used in the calculations are presented in Table 45.
The emission rate was 0.0035 g/sec for the 1,1,1-trichloroethane production
Batch 2. This value should be used with care in health effects calculations
using emission rates since the rate does not necessarily apply to other waste
streams and it does not reflect process downtime. The emission factors in
terms of g/g waste are expected to be more appropriate for dispersion modeling
calculations. In both Batches 1 and 2 tested, emissions were only a small
fraction of the VOCs recovered from the process, thus suggesting that the use
of waste treatment does not itself produce significant emissions.
In contrast to the open product storage tank at Plant C, the product
storage vessels at Plant D were enclosed inside a building with a vent to the
roof from each tank. The flow rates from the vents of the product-receiving
vessel and the separator could not be measured with an Alnor velometer because
of the very low flow rates. Since the flow rate was not measurable with the
velometer (less than 25 cm/sec), the volume of the displaced gas was
multiplied by the concentration measured in the gas to estimate the VOCs lost.
The rates estimated from concentrations and working losses in the storage tank
and MST are 0.4 mg/sec and 2.4 mg/sec, respectively. These air emissions from
the storage tanks at Plant D were much lower than from the storage tank
surface exposed to the flow of air at Plant C.
Liquid Residuals
The condensed steam and organics are collected in a receiving vessel and
the liquid is transferred to a decanter. The recovered organic layer is
transferred to a product storage tank. Some solvents can be sold as paint
thinners or cleaners; the 1,1,1-trichloroethane can be sold as a specialty
solvent if stabilized to avoid hydrochloric acid decomposition. Recovered
solvents (nonhalogenated) can be used as fuel or fuel supplements. The
condensed steam containing VOCs is steam stripped to remove the VOCs so that
the treated water can be sent to a public waste water treatment facility.
Batch 3 was an example of how the VOCs could be removed from the condensed
steam used in treating an organic waste. If the treated waste has sufficient
Btu content (e.g., Batch 2), then it can be burned as fuel after the
chlorinated material is removed. Aqueous residues from the stripper are
currently being solidified and landfilled.
-------
r
PROCESS COST
General Facility Costs for Plant D
This section describes cost information provided by Plant D during the
pretest site survey. The cost per unit volume is dependent on the size ot the
unit and the volumetric throughput, as well as the type of waste. The
condensed steam also needs to be treated and this will decrease the
throughput.
Capital Cost (New Unit, 1984)--
DCI Steam Stripping System
Support Facilities and Installation (building,
tanks, boiler, air compressor, cooling water
tower, pumps, lines)
Total Installed
Operating Cost (1984)--
Salaries and Payroll Tax
Maintenance
Material and Supplies
Fuel, Utilities
Monthly Operating Cost
Annualized
Total Annual ized Cost-
Annual Operating Cost
Capital Recovery Factor (10 percent over 1 year)
Total
$80,000
$370,000
$450,000
$ 6,000/mo.
800/mo.
4,000/mo.
5,000/mo
$ 15,800
$189,600
$189,600
73.500
$262,900
Process throughput for the steam stripping system was reported to range
from 130 L/hr (35 gal/hr) to 760 L/hr (200 gal/hr) depending upon the
characteristics of the stream being processed. This throughput does not
include treating the residuals which are produced, but can be used to
determine to a first approximation the range of waste processing costs.
1. At 130 L/hr (35 gal/hr), 24 hr/day, 5 days per week:
Total Waste Processed (L/yr)
(gal/yr)
Total Annualized Cost
Unit Cost ($/L)
($/gal)
824,000
218,000
$262,900
0.319
1.2C6
112
-------
r
2. At 760 L/hr (200 gal/hr), 24 hr/day, 5 days per week:
Total Waste Processed (L/yr) 4,723,680
(gal/yr) 1,248,000
| i Total Annualized Cost $262,900
Unit Cost ($/L) $ 0.056
($/gal) $ 0.211
Unit Treatment Costs for Each Batch
The unit costs of treating each of the four waste batches were estimated
and are presented in Table 46, along with unit solvent recovery costs. The
purpose of this analysis was to determine the cost-effectiveness of steam
stripping these four streams to remove VOCs and to determine if any
conclusions could be drawn about the influence of waste type, batch size,
degree of treatment, or initial waste VOC content on unit costs.
I The annual process operating costs used in this calculation ^as assumed
| to be that derived above ($262,900). The amount of waste which could be
| treated in a year varied from batch to batch. This annualized waste volume
} was dependent upon the volume of waste charged to the system and the time it
| took to process the waste (i.e., the batch cycle time). The annualized waste
' volume that could be treated in the system was estimated as the batch size
I times the ratio of the annual process time (24 hours per day, 5 days per week,
52 weeks per year) to the batch process time. The batch cycle time was taken
to be equal to the sum of (1) the waste stripping time (derived from batch
process data, Appendix B); (2) an assumed 40 minutes for emptying and
recharging the cnit, and for heating each batch; and (3) the time required to
strip the VOC from the MST condensate that was produced by the stripping. The
I derivation of the condensate stripping time is provided in Appendix E.
j The volume of solvent recovered in a batch run was taken to be the
i orgam'cs removed from the waste. It was the difference between the products
i of the waste VOC concentration and waste volume before and after stripping.
I The annualized solvent recovery was calculated from the product of the
I annualized waste volume and the ratio of solvent recovery to batch size.
1
I Unit waste treatment and solvent recovery costs were obtained by dividing
j the annual operating cost by the annualized waste volume and solvent recovereo
l volume, respectively.
I- j
I i A sample calculation showing the derivation of a unit treatment cost is
t j provided in Appendix F.
I i Because each of these batches has different initial VOC content and
I volume and has undergone different degrees of treatment (i.e., percent VOC
I removed), conclusive statements cannot be made from Table 46 about the factors
| influencing treatment cost. However, several qualified conclusions can be
| drawn. Compared to mixed aqueous Batch 1 organic Batch 2 has a charge that
is lower by approximately 30 percent and has a greater initial VOC
concentration (see Table 28). Yet, the unit treatment cost for Batch 2 is 20
113
-------
F
l*W»*-Hf«S!H««|
TABLE 46. ESTIMATED UNIT COST FOR THE FOUR BATCHES TESTED
Waste stream
Waste volume, liters
(gallons)
Organic solvent recovery, liters
(gallons)
Condensate stream, liters
(gallons)
Condensate treatment cost $/litera
Batch cycle time, hours
Process rate (L/hr)
Annual ized waste volume, liters3
(gallons)(l
^Annual ized solvent recovery, liters3
•** (gallons)
Total annual operating cost, $a
Cost per unit volume:3
1) Waste treatment, $/liter
$/gallon
2) Solvent recovery, $/Mg
$/liter VOC
Aqueous
xylene
1,590
(420)
333
(88)
359
(95)
0.01
2.35
677
4,224,000
,116,000)
885,024
(233,793)
262,900
0.06
(0.236)
345
0.324
1,1,1-Trl-
chloroethane
897
(237)
660
(174)
393
(104)
0.05
2.48
362
2,259,000
(596,700)
1,662,000
(439,100)
262,900
0.12
(0.44)
119
0.166
Aqueous 1,1,1-
trichloroethane
564
(149)
95
(25)
256
(68)
0.06
1.88
300
1,872,000
(494,500)
314,500
(83,100)
262,900
0.14
(0.53)
629
0.147
Aqueous
mixed solvent
360
(95)
11
(3)
1,320
(35)
0.06
1.30
276
1,725,000
(455,700)
52,800
(13,900)
262,900
0.15
(0.58)
5,242
0.156
The method of calculation is presented in Appendix E.
3The stripping time plus 40 minutes plus time to treat the condensate.
-------
percent lower than for Batch 1. It is believed that this is due to the fact
that the stripping rate constants for VOC compounds in organic matrices are
higher than for compounds in waste, as is indicated for methyl ethyl ketone
and trichloroethane in Table 41.
While waste treatment unit costs vary by less than a factor of two among
the four batches, solvent recovery unit costs vary by over an order of
magnitude. Much higher costs are incurred for Batches 3 and 4 due to both a
small initial batch size and their lower initial VOC contents.
Variation of Unit Costs With Degree of Treatment
The extent to which solvent recovery and waste treatment costs are
influenced by the degree of treatment (i.e., percent of VOC removed) was
investigated. The analysis was performed on each of the four batches to
determine unit costs for reducing VOC content of wastes by from 68.4 percent
to 99.9 percent. The results of the cost analyses are presented in Tables
47-50. The costs per unit weight ($/Hg) recovered solvent that were derived
in these tables are presented graphically in Figures 30-33.
As with the previous analysis (Table 45), the present analyses used the
batch charge volumes and initial VOC concentrations as observed in the field.
The volume of organic solvent recovered was determined by percent VOC removed
(and solvent density, see Table E-l). The amount of steam condensate that was
produced during the stripping and the resultant time and costs associa'.ed with
condensate treatment were also handled in a similar manner (see Appendix E)
and were accounted for in determining the batch cycle times and, then, the
waste processing rate. As with the analysis in Table 46, a 24-hour-per-day,
5-day-per-week operating schedule was assumed when calculating annualized
waste volumes and annualized solvent recovery volumes.
Several additional elements were added to the cost analysis presented in
Tables 47-50, compared to the Table 46 analysis. First, a stripping rate
constant which was representative of the total VOC component of the waste had
to be derived so that the variation in VOC content with time could be
estimated. Second, the analysis accounted for:
a. the variation in steam load (and cost) with batch cycle time.
b. waste processing credits, i.e., the income that a recycler received
for acquiring waste.
c. variation in treatment residual disposal costs with composition of
the residual.
d. income from sale of the recovered solvent.
The approaches to determining a representative stripping constant and to
estimating the steam load are complex and are presented in Appendix E. The
waste processing credits was assumed to be equal to the landfilling costs
incurred by Plant D, approximately $0.37 per liter. (In fact, the facility
typically charges from $0.50/liter ($100/drum) for waste received by the
115
-------
9IT
k$/Mg VOC TREATMENT
I I I I I I I I I
o>
00
i r* r
ro «o oo
01
i i i; i i i i i
ppppppppp
WN>-jk-*bj^.b'tio-*
i i i i i i i i i i i i i i i i i i i
-------
ill
k$/Mg VOC TREATMENT
-------
811
k$/Mg VOC TREATMENT
I I I I I I
oooooo ooooppppp
b>uii>.w?o^-'O^-'fobiikWb)Njboio
11(11 i I I I r-i I II I
-------
611
k$/Mg VOC TREATMENT
-------
ro
O
TABLE 47. COST ANALYSIS FOR THE DIRECT STEAM STUMPING OP HAZARDOUS WASTE:
BATCH 1 (AQUEOUS XYLENE)
Percent VOC removal
Concentration of residual VOC
Waste volume (L)
Organic solvent recovery (L)
Steam condensate (L)
Condensate treatment tine (ain.)
Condensate treatment ($/L waste)
Batch cycle time (rain.)
Processing rate (L/hr)
Annual! zed waste volume (L)
Annuallzed solvent recovery (L)
Total annual operating cost
Waste collection revenues ($)
at $0.37 /L
Solvent sale credits ($)
at $0.20 /L
Treated waste disposal casts ($)
0.10% or more VOC landfilled
Waste processing cost
Cost per unit volume:
waste treatment ($/L)
solvent recovery ($/L)
solvent recovery ($/Mg)
Incremental cost ($/L solvent)
Waste residual volatility, torr
VOC landfilled with treated waste
VOC sent to POTW with treated waate
68.4
6.636
1590
228
68.4
3.5
($0.01)
56
1610
9974093
1432679
$262.900
3690415
286536
$3,690,415 $2
($23,638)
($.00)
($0.02)
($19)
($0.02)
116.76
31.600%
0.000%
90.0
2.1
159C
301
136.7
6.9
($.00)
72
1216
7502739
1418018
$262,900
2776013
283604
,776,013 $2
($20,704)
($.00)
($0.01)
($17)
($0.01)
36.95
10.000%
0.000%
96.8
0.6636
1590
323
205.1
10.4
$0.01
87
976
6011388
1222500
$262,900
2224214
244500
,224,214 $1
$18,400
$.00
$0.02
$18
$0.41
11.68
3.160%
0.000%
99.0
0.21
1590
331
273.4
13.9
$0.01
103
816
5015653
1042754
$262,900
1855791
208551
,855,791
$54,349
$0.01
$0.05
$61
$1,71
3.69
1.000%
0.000%
,J,,*«. « .*.-».* *.*
99.7
0.06636
1590
333
341.8
17.3
$0.01
119
701
4302146
900596
$262,900
1591794
180119
$215
($1,608,798)
($6.35)
($1.68)
($1,948)
($251.70)
1.17
0.000%
0.316%
99.9
0.021
1590
334
410.0
20.8
$0.02
135
614
3763946
7S0268
$262.900
1393770
158054
$188
($1.288.735)
($0.34)
($1.63)
($1,896)
$18.94
0.37
0.000%
0.100%
( ) indicates credit.
-------
TABLE 48. COST ANALYSIS FOR THE DIRECT STEAM STRIPPING OF HAZARDOUS WASTE:
BATCH ?, (1,1,1-TRICHLOROETHANE)
Concentration of residual VOC
Waste volume (L)
Organic solvent recovery (L)
Stean condensate (L)
Condensate treatment tine (min.)
Condensate treatoent ($/L waste)
Batch cycle time (rain.)
Processing rate (L/hr)
Annual! zed waste voluae (L)
Annual ized solvent recovery (L)
Total annual operating cost
Waste collection revenues ($)
at $0.37 /L
Solvent sale credits ($)
at $0.20 /L
Treated waste disposal costs ($)
0.10* or store VOC landfilled
Waste processing cost
Cost per unit voluae:
waste treatar ;$/L)
solvent recovv. „ ($/L)
solvent recovery ($/Mg)
Incremental cost ($/L solvent)
Waste residual volatility, torr
VCC landfilled with treated waste
VOC sent to POTW with treated waste
68.4
23.364
897
484
87.0
4.9
($0.04)
69
840
5107651
2585288
$262,900
1889831
517058
$1.889,831
($254,158)
($0.05)
($0.10)
($74)
($0.10)
13.33
31.600%
0.000%
P
90.0
7.4
897
597
174.0
9.8
$.00
78
610
3671502
2445220
$262,900
1358456
489044
$1,358.456
($226.144)
($0.08)
($0.09)
($70)
($0.07)
4.22
10.000%
0.000%
'ercent VOC r
96.8
2.3384
897
643
261.0
14.7
$0.03
98
479
2864941
2053063
$262,900
1060028
410613
$1.060.028
($147.713)
($0,05)
($0.07)
($54)
$0.20
1.33
3.160%
0.000%
enoval
99.0
0.74
897
657
347.9
19.6
$0.06
117
395
2349455
1721211
$262,900
869298
344242
$869,298
($81,342)
($0.03)
($0.05)
($36)
$1.06
0.42
1.000%
0.000%
99.7
0.23384
897
662
435.0
24.5
$0.09
136
335
1990803
1468539
$262,900
736597
293708
$736,597
($30,808)
($0.02)
($0.02)
($16)
$3.78
0.13
0.316%
0.000%
T -It'll T~-f —I— =--*_*- J--*-
99.9
0.074
897
663
521.9
28.4
$0.12
155
292
1727434
1277023
$262.900
639151
255405
$86
($631,569)
($0.37)
($0.49)
($372)
($219.05)
0.04
0 . 000%
0.100%
( ) Indicates credit.
-------
-------
TABLE 50. COST ANALYSIS FOR THE DIRECT STEAM STRIPPING OF HAZARDOUS HASTE:
BATCH 4 vMIXED AQUEOUS)
Percent VOC renoval
Concentration of residual VOC
Haste volume (L)
Organic solvent recovery (L)
Steam condensate (L)
Condensate treatment tine (aln.)
Condensate treatment ($/L waste)
Batch cycle time (rain.)
Processing rate (L/hr)
Annualized waste volume (L)
Annualized solvent recovery (L)
Total annual operating cost
Waste collection revenues ($)
at $0.37 /L
Solvent sale credits ($)
at $0.20 /L
Treated waste disposal costs ($)
0.10* or more VOC landfilled
Haste processing cost
Cost per unit volume:
waste treatment ($/L)
solvent recovery ($/L)
solvent recovery ($/Mg)
Incremental cost ($/L solvent)
Waste residual volatility, torr
VOC landfilled with treated waste
VOC sent to POTW with treated waste
68.4
0.9796
380
8
34.9
1.4
$0.01
4i-
431
2665911
56523
$202,900
986387
11308
$986,387
$251.694
$0.09
$4.45
$4.685
$4.45
22.84
31.600*
0.000*
90.0
0.31
360
10
69.8
2.7
$0.03
57
359
2205833
61543
$262,900
816158
12309
$816.158
$250,591
$0.11
$4.07
$4,288
$2.87
7.23
10.000*
0.000*
93.8
0.09796
360
11
104.7
4.1
$0.05
68
307
1880836
56463
$262,900
695909
11293
$94
($444,208)
($0.24)
($7.87)
($8,281)
($184. yS)
2.28
0.000*
3.160*
99.0
0.031
360
11
139.6
5.6
$0.08
75
269
1639571
50318
$262,900
606641
10064
$82
($353,723)
($0.22)
($7.03)
($7.400)
$30.52
0.72
0.000*
1.000%
99.7
0.009796
360
11
174.5
6.8
$0.12
84
239
1452959
44899
$262,900
537595
8980
$73
($283,602)
($0.20)
($6.32)
($6,649)
$96.92
0.23
0.000*
0.316*
99.9
0.0031
360
11
209.3
8.2
$0.16
92
215
1304652
40404
$262,900
482721
8081
$65
($227.837)
($0.17)
($5.64)
($5,936)
$306.98
0.07
0.000*
0.100*
( ) Indicates credit.
-------
truckload to Sl.OO/liter ($200/drum) for smaller loads, making this credit
conservatively low.)
At Plant D, wastewaters which contain less than 0.1 percent organics can
be discharged to the municipal treatment system. The cost for this is
approximately $0.0005 per liter, and this figure was assumed to be the cost
for stripper bottoms disposal in the cost analysis when treatment was
sufficient to reach this level. For bottoms containing over 0.1 percent
organics, it was assumed that the bottoms were sent to a landfill at a cost of
$0.37 per liter.
From Tables 37-40 and Figures 30-33, the following conclusions can be
drawn. The figures of the VOC treatment costs as a function of the percent
VOC removal show a dramatic decrease in cost when the residual waste can be
disposed by sewer to a publicly-owned wastewater facility, instead of
landfilling; it is assumed that wastes containing less than 0.1 percent VOC
can be sent to a sewer. The cost of disposal by sewer is substantially less
than landfilling costs. This analysis suggests that disposal costs and
revenues for waste collr:tion are major components of the cost analysis. The
cost of VGc removal adds to the costs, particularly at high VOC removal rates.
The treatment of organic wastes (Batch 2) is estimated to be more profitable
than for aqueous wastes (Batches 1, 3, and 4).
124
ita.
-------
r.
SECTION 9
FIELD TEST RESULTS: DISTILLATION
This section describes the field evaluation at Plants B and E. The
distillation system at Plant B differs from the direct 1"Ject;'on.s*!jflJ*
! Plant D in that the capacity of Plant B is larger, the steam heats the waste
j indirectly through coils, and the stream of vapors is processed in a
I distillation column.
'T
1 Distillation to treat hazardous waste was investigated since low
'i concentrations of VOCs in water can be removed and recovered as an organic
stream. In batch distillation, the batch can be treated until the
concentrations of VOCs are below specifications. Organic materials can be
recovered with distillation which have significant water solubilities, with
steam stripping, a mixture of water and organic would be obtained from the
process and would require further processing.
DISTILLATION FIELD EVALUATION AT PLANT B
• Process Description
j Contaminated organic chemicals and solvents are received in bulk and drum
shipments and processed for reclamation and recycle. All waste material is
either processed in the thin film evaporator or the kettles. Approximately 90
percent of the incoming shipments are processed through one of two Votator
thin-film evaporators during which about 80 percent of the material is
stripped off as overhead product. The overhead product may or may not be
further refined through fractionation distillation, depending upon the
intended end use. Distillation bottoms are shipped offsite and utilized as
flel in cement kilns. There is an additional discussion of Plant B wastes and
processes in Section 7.
There are eight fractionation distillation systems of varying capability
and capacity at the Plant B facility. The fractionation distillation systems
each consist of a reboiler. a tray column and condenser, an accumulator, and
associated pumps, valves, and piping (Figure 34) Instrumentation includes a
reboiier and column head vapor temperature recorders (multipoint recorder) and
rotameters in the reflux and product lines. The system selected for any
particular separation is dependent upon a number of factors sucn as
throughput, relative vapor pressures, and required purity of the process
streams.
Aqueous phases from distillation operations may be disposed of in a
municipal wastewater treatment facility if the organic concentration is less
than 01 percent by weight; otherwise it is shipped offsite for treatment and
125
-------
r
c
E
— i
q X-N
B
C TH
c
^H tS
^-" *C"
r"- *-_-*
—I O
tt — •
C
1
r
u
01
(0
s
o
•D
C
c
u
1
•~—-'
Reboiler
Steam
Vent A
2'
10
20'
Vent B
Recetiver
15'
Product
Accumulator
to Product
Storage Tank
Figure 34. Distillation process: Batch 1 (Aqueoiu. HEX),
126
-------
disposal. Still bottoms which are unacceptable for fuel are sent to a
landfill.
The reboiler contains a steam coil for heating. The steam stnply header
pressure is controlled at 125 psig, but there is some fluctuation ,n pressure,
due to the nature of the controller. Steam flow is controlled manually by the
operator with a hand valve.
The fractionation column is a tray column. Vapor from the reboiler
enters about midway up the column, and the bottoms from the column recycles
back to the reboiler fay gravity. Reflux is p-o^ided to the top of the column
from the accumulator by a small centrifugal pump. Thft reflux flow rate is
controlled manually with a hand valve by the operator.
The overhead condenser is a vertical, shell-and-tube, water-cooled
condenser. The distillate accumulator (approximately 50 L) is a small tank
from which the column reflux is pumped, with the product overflowing to any of
a number of product storage tanks.
A typical operation involves charging a quantity of material to the
reboiler. The quantity is noted from visual observation of the liquid level
in the reboilar. Steam is then applied to the coil in the reboiler, and the
batch is heated up with the column operating under total reflux. After the
reboiler and column head temoeratures have been lined out, the steam flow and
reflux are set and product takeoff is started. The distillation is continued
until such time as the column head temperature begins to rise indicating that
the volatile component being stripped from the batch is essentially depleted.
The process may either be discontinued at this time or, as is done with some
aqueous streams, stripping may be continued in order to achieve a VOC level of
less than 0.1 percent for disposal as waste water. In the case of the latter,
laboratory analyses are run periodically to determine the VOC content.
The steam flow rate to the reboiler and the reflux rate to the column are
controlled by an operator through manual adjustment of hand valves. The
column head and reboiler temperatures and the distillate rate and appearance
are the primary factors used in control of the process.
The distillate condenser is vented to the atmosphere where emissions
should be mostly noncondensibles. However, this is a potential source of VOC
emissions should the condenser become overloaded for any reason. The reflux
accumulator is also vented to the atmosphere, and emissions from this tank
would represent operational displacement of equilibrium vapor frcm the
distillate.
Process Effectiveness
Two different waste streams were selected for the field evaluation at
Plant B. A summary of the waste characterization and process data is
presented in Table 51. The individual components in these two waste streams
are characterized in Tables 52 and 53. The waste streams were both aqueous
organic and consisted primarily of methyl ethyl ketone and acetone,
respectively. These wastes were obtained directly from the generator. The
127
t- ' "*"
-------
r
TABLE 51. DISTILLATION WASTE CHARACTERIZATION AND PROCESS DATA
Batch 1 Batch 2
, ' 30,000 H.400
Initial volume (L)
VOC (mass percent)3 4-7 23
Water content3 95 77
Process Data
Distillation timeah 13.5 hr 8.0 hr
Steam rate (kg/hr)D 860 "J
Stripper residue volume (L) 28,500 o,ouu
Bottoms VOC content (weight %)c 0.05 0.07
VOC recovered (kg)3 1.400 2.614
% VOC recove-ed* 99 99-8
^^^^^^^^—_^^Mg-aa^mBa»i^^M«e»^a"«**""'^M''o*™i" !• I III ~^""
3Measured
bEstimated.
cEstimated by material balance.
128
-------
TABLE 52. WASTE CHARACTERIZATION OF BATCH 1
(Aqueous Methyl Ethyl Ketone)
Total solids (mg/L) 340
pH 5.5
Water (weight percent) 95
Oil (weight percent) Negligible
VOC (weight percent) 5
Density (g/cra ) 1
Number of phases 1
VOC analysis (Aqueous phase)
Compound Concentration
(mgTTJ
Methyl ethyl ketone 30,000
s i 2,2-Dimethyl oxirane 6,400
I Isopropanol 1,900
f Methylene chloride 3,100
I Methanol 3,500
f . Carbon tetrachloride 1,700
s 1,1,1-Trichloroethane 710
{ Other VOCs 2,130
129
-------
TABLE 53. WASTE CHARACTERIZATION OF BATCH 2
(Aqueous Acetone)
Total solids (rag/L) 7,500
pH - 13
Water (weight percent) 77
Oil (weight percent) Negligible
VOC (weight percent) 23
Density (g/cm ) 1
Number of phases 1
VOC analysis (Aqueous phase)
Compound Concentration
Acetone 212,000
Trichloroethene 9,500
1,1,1-Trichloroethane 2,800
Toluene 2,700
Methyl ethyl ketone 2,300
Isopropanol 440
Aromatics 291
130
-------
objectives in processing these wastes were to reclaim solvent and reduce the
VOC content to a level acceptable for disposal in a municipal wastewater
treatment facility. However, due to some light sludge, oil, or heavy organic
contamination in the acetone stream as received, it was unlikely that the
residue from this batch could be made acceptable for disposal to the sewer and
was sent to Plant F for disposal.
Two distillation systems having the same design but different reboiler
capacities and column diameters were used.
Process data collected during processing included reboiler temperature,
colu-n head temperature, reflux rate, and in one case product rate (see
Appendix B). There were no means to measure the steam flow to the reboiler
during distillation, but approximate rates could be obtained by enthalpy
calculations. The quantity of distillate recovered could not be measured
since there were no liquid level instruments on the product storage tanks.
These tanks were very large, on the order of 37,850 liters (10,000 gallons) or
greater, and in both cases contained product from previous batches.
Methyl Fthyl Ketone Batch—
A 30,280-liter (8,000-gallon) batch of the me'.hyl ethyl ketone (MEK)
waste stream was charged to the raboiler of a 42-'lnch (110-cm) system
(designated by column diameter). This is a 41,635-liter (11,000-gallon)
reboiler, and the column in this system has 30 trays. The 41,635-liter charge
was determined by visual observation of the H-vel in the reboiler by an
operator.
The column was operated under total reflux during the heatup which lasted
until the reboiler and column head temperatures were lined out (became
constant). The system was held under total reflux for about 1 hour before
distillation was started. The reboiler and column head temperatures and the
rate of distillate flow were monitored to determine the progress of the
distillation. The distillation was essentially completed when the reboiler
temperature had reached 100°C and a rapid rise in the column head temperatures
occurred, indicating depiction of the; MEK. However, stripping was continued
until a VOC level of less than 0.1 percent was achieved as evidenced by Plant
B laboratory analyses. The aqueous residue remaining in the reboiler was then
discharged to the sewer for processing through the municipal wastewater
treatment system.
The overhead product was put into a tank in which organic distillate was
being accumulated for further refining prior to being returned to a specific
client.
Acetone Batch—
The 32-inch (81-cm) distillation system (designated by column diameter)
was used for processing this batch. This system includes a 13,248-liter
(3,500-gallon) reboiler and has 30 trays (Figure 35).
An 11,335-liter (3,000-gallon) batch of the acetone waste stream was
charged to the reboiler. The column was operated under total reflux during
the period of heatup and until the reboiler and column head temperatures had
131
-------
c
o
0.81M (32 in.) Distillation Unit
Vent C
30'
i
12'
en
c
o
c
u
Recycle
T-eboiler
Steam
to Product
Storage Tank
Figure 35. Distillation process: Batch 2 (Aqueous acetone).
132
i**.
-------
lined out. The system was held at this condition for about 3-1/2 hours prior
to the start of distillation only for scheduling purposes. The reboiler and
column head temperatures and the distillate rate were monitored during
processing to determine the progress of the distillation.
This waste stream contained some contamination of heavy organics or light
sludge as received. The aqueous residue would normally have been stripped to
a level acceptable for disposal to the sewer, but the unexpected contamination
precluded this option. Therefore, the distillation was completed after the
batch temperature had reached 100°C and the column head temperature began to
rise rapidly indicating depletion of the organic solvent. The final batch
sample was taken at that time.
The distillate was put into a tank in which reclaimed acetone was being
accumulated. No further processing would be required prior to shipment. The
contaminated aqueous residue (reboiler contents) was shipped to another TSDF
for further treatment and disposal at a facility which accepts dilute waste.
The results of analyzing the waste material in the reboiler during
stripping is presented in Tables 54 and 55. These results are also
graphically presented in Figures 36 and 37.
In the acetone batch, the relative rate of removal of all the compounds
was to be approximately the same rate, with the xylene and isopropanol being
somewhat more slowly removed (Figure 37). There was removal of all the VOCs
to less than 700 ppm in the final treated waste, with 99.7 percent removal of
acetone.
As the waste material is stripped, the vapor phase concentration of the
volatile organic compounds at equilibrium in the cool waste decreases as their
concentrations decrease. In Tables 56 and 57, it is apparent that the
volatility characteristics of the wastes are dramatically altered by the steam
distillation process. These volatility characteristics of the cooled wastes
are of concern because of the relation between the vapor pressure of the
volatile components and its release into the atmosphere upon disposal. The
waste material generally showed at least an order of magnitude decrease in the
vapor concentrations at equilibrium with the waste.
In the methyl ethyl ketone (MEK) batch, there were two major components
which were removed from the organic waste: 2,2-dimethyl oxirane (DM0) and
MEK. MEK and DM0 were removed from the process with less than 10 ppm
remaining after 615 minutes. The trichloroethane was somewhat more slowly
removed with 530 ppm or 0.05 percent residual trichloroethane in the water
material after 720 minutes. The methylene chloride was rapidly removed from
the batch with greater than 99.7 percent removal after 215 minutes. The data
from the 1,1,1-trichloroethane were anomalous and unexplained. The other
major components were removed to below the detectability limits of the
process.
The distillation of individual components in a dilute aqueous phase is
expected to occur independently. That is, the concentration decrease of one
component is not expected to be significantly influenced by the composition of
133
-------
10,000
1,000
MEK
81,1 DCE
IPA
D MEOH
2345
Time (rain x ICO)
Figure 36. Concentrations of VOC in MEK waste during
distillation stripping.
134
-------
100,00
10,00
1,00
10
O Acetone
IPA
MEK
Toluene
1,1,1 TCE
100 200 300
Time (minutes)
400
Figure 37. Concentrations of VOC in acetone waste during
distillation.
lob •
jftyfh-
-------
CO
TABLE 54. CONCENTRATIONS OF VOC IN ACETONE BATCH
Sample
number
250039
250041
250042
250043
Process
time
(Minutes)
0
120
240
364
Trichloro-
ethene
(mg/L)
9,500
<10
<10
<10
Toluene
(nig/L)
2,700
820
28
<10
Ethyl
benzene
(mg/L)
91
41
10
<10
Xylenes
(mg/L)
200
110
<10
<10
Acetone
(mg/L)
212,000
140,000
4,100
690
Isopro-
panol
(mg/L)
440
430
110
13
Methyl ethyl
ketone
(mg/L)
2,300
1,300
14
<10
1,1,1-Tri-
chloroethane
(mg/L)
2,800
1,100
32
<10
-------
i;
TABLE 55. CONCENTRATIONS OF VOC IN METHYL ETHYL KETONE BATCH
2,2-01-
Process methyl Isopro- Methyl ene
Sample time MEK oxirane panol chloride
number (minutes) (mg/L) (mg/L) (mg/L) (mg/L)
250030
250033
250036
250037
250038
0 30,000 6,400 1,900 3,100
215 9,400 490 2,500 <10
325 61 <10 52 <10
615 <10 <10 <10 <10
720 <10 <10 <10 <10
1.2-D1- 1,1,1-Trl
chloro- chloro-
ethene Hethanol ethane
(mg/L) (mg/L) (mg/L)
320 3,500 710
310 3,400 550
<10 <10 350
<10 75 1,100
<10 <10 530
- Tri- Carbon
chloro- tetra-
ethene chloride Benzene
(mg/L) (mg/L) (mg/L)
520 1,700 260
990 2,200 500
350 <10 190
110 <10 210
<10 <10 <10
-------
f -_.-
TABLE 56. CONCENTRATIONS OF VOC IN HEADSPACE OVER BATCH RESIDUE AS A FUNCTION
OF STRIPPING TIME. ACETONE BATCH
?
I
f
1
'
Sample
number
250039
250041
" j 250042
• i 250043
Process
time
(Minutes)
0
120
240
364
Acetone
(mg/L)
340
280
7.1
2.1
Methvl
ethyl
ketone
(mg/L)
0.4
0.5
<0.l
-------
r -
the other components. The rate of stripping of dilute compounds is expected
to be proportional to the amount of material present. Therefore, the rate
decreases as the batch is stripped to completion. One of the characteristics
of first-order decay processes is that the logarithm of the concentration is
proportional to time plus a constant. The value of the constant is specified
so that the concentration equals the initial concentration at the beginning of
the stripping process. To verify the first-order decay model, the results
from the batches are plotted as logarithm of the concentration versus time in
Figures 36 and 37.
For distillation systems in which a layer of organic material is being
stripped with an aqueous phase (this was not the case for the two batches
tested), the concentration in the aqueous phase may not fall as rapidly
initially as later in the process. If there is a layer of organic on water,
the organic components present as a liquid would need to be removed by the
stripping process (leaving only an aqueous phase) before very low levels of
these organic components could be obtained in the aqueous phase through the
stripping process. In addition, distillation can provide a relatively clean
"cut" of the more volatile components. The less volatile organics may be
returned by reflux to the top of the column and ultimately to the reboiler.
For organic waste materials, some of the more volatile components are
removed from the reboiler initially. The temperature above the batch beirig
stripped will reflect the equilibrium concentrations of the vapors of these
components.
Table 58 presents a summary of the data analysis of the two batches
evaluated at Plant B. The correlation coefficient was relatively high for
many of the rates of stripping. The steam rate influences the rate of
stripping, so the rate constants are presented in a dimensionless form to
account for the steam rate and batch size.
The data in Figures 36 and 37 were correlated with a least-squares
analysis for the logarithm of the concentration as the independent variable
and with the time of distillation as the dependent variable. The negative of
this slope of the linear correlation is the stripping rate constant. Figures
36 and 37 are obviously not represented by a straight line, within the limits
of the data precision and particularly for the initial part of the removal
curve.
Process Residuals
Air Emissions—
In addition to the removal of VOCs from the waste material, the absence
of significant treatment process emissions is also of interest. The treatment
after disposal would be of little use if substantial quantities of VOCs are
;,-, released to the atmosphere during processing.
[.; The air emission sources from the two different distillation processes
[ were .evaluated at Plant B. The vent locations are indicated in a typical
; j process diagram of the equipment which is presented in Figure 35. In both
Cj columns, the condenser was vented to the atmosphere on the downstream side of
i 139
-------
TABLE 58. SUMMARY OF DISTILLATION VOC REMOVAL RATES
Correlation Striooinq rate constants
Batcf
Acetcne
Acetone
Acetone
Acetone
Acetone
Acetone
MEKb
MEK
MEK
MEK
MEK
MEK
Component coefficient Hinutes-l gQ% Limits' Dimensicnless3
acetone
isoprcpanol
methyl ethyl ketone
1,1,1-Trichlorethane
toluene
xylene
methyl ethyl ketone
isopropanol
methanol
benzene
trichloroethene
2,2-dimethyl oxirane
0.963
0.931
0.927
0.971
0.975
0.931
0.874
0.923
0.763
0.708
0.856
0.967
0.017
0.00986
0.0171
0.0168
0.0166
0.0094
0.0172
0.00947
0.00766
0.00366
0.00526
0.0188
0.0097
0.0796
0.00141
0.0086
0.0079
0.0076
0.0635
0.0099
0.0088
0.00469
0.00432
0.0368
18.9
10.9
19.0
18.6
18.5
10.4
36.0
19.8
16.0
7.6
11.0
39.4
Obtained by multiplying the stripping constant by the ratio of the batch size
to the steam rate.
bThe correlation coefficients are relatively low for the MEK batch because of
the nonlinearity of the data.
140
-------
r
the product flow through a vertical atmospheric vent with a tee near the top
and a vertical pipe with a discharge several meters lower than the tee. The
upper condenser discharge was beneath metal plates in the scaffolding support.
The condenser vent on the 110 cm (42-inch) column was accessible fo^ air
sampling and velocity determinations as planned, but the vent on the condensnr
from the 81-cm (32-inch) column was inaccessible for the measurement of
velocities due to personnel safety consideration. Air sampling was obtained
on the condenser vent and the receiver vent of the 81-cm (32-inch) column at
approximately mid-cycle. With a cross-sectional area of 13 cm2 (0.014 ft2K
the average flow during a 10-minute period was estimated to be less than 1.3
! I L/sec (0.05 ft3/sec).
I I
r "; The product vent of the MEK batch was sampled during filling, and visible
} | fumes were observed with a relatively low velocity. The volumetric flow rate
I I from the product vent was assumed to be equal to the volume displaced by the
I j product. The temperature in the gas vent was measured with a mercury
f ; thermometer where accessible. The temperature of gas emitted from the
- ! condenser vent on the 110-cm (42-inch) column was typically slightly warmer
f than the atmosphere, indicating that some of the flow from the vent could have
* originated with the distillation process.
r The air emissions were estimated on both batches. The results of the
I concentrations obtained from air samples taken from the vents are presented in
t Tables 59 and 60. In addition to the concentrations in the air samples, some
• of the equilibrium vapor analyses obtained on liquid samples are presented in
,' Tables 59 and 60 for comparison. The concentrations of the equilibrium over
| > the distillate obtained midway through the process correlated within a factor
I j of 33 percent with the concentrations obtained from the material in the vapor
f ! phase of the MEK process. The concentrations in the air emission decreased as
t | the batch was processed, and the concentrations in both the laboratory
*. | analysis of headspace over the distillate and the reboiler contents tended to
I | decrease as the batch was processed. The average emissions are expected to be
I ij a combination of both the rate of flow and the composition drift of the
I I various components. Table 61 estimates the air emissions on the basis of the
I | average of the results of the tests on three air samples taken from the
f I aqueous MEK batch. The estimates of air emissions from the acetone batch is
I | presented in Table 62. The volumes emitted from the vents were estimated on
| I the basis of the data obtained from waking losses and vent velocity
measurements. The same procedure was used with the acetone batch and
presented in Table 62. In both cases, the emissions were only a small
fraction of the VOCs recovered from the process.
Process Cost
A listing of waste treatment costs provided by Plant B is presented in
Section 7, Page 77.
141
-------
TABLE 59. AIR SAMPLES: METHYL ETHYL KETONE WASTE PROCESS
Sample
number
202 Field ambient
201 Condenser vent
31 Distillate3
206 Condenser vent
205 Condenser vent
34 Distillate3
204 Condenser vent
203 Accumulator vent
35 Distillate*
Time
10:42A
11:25A
1:14P
1:19P
1:20P
2:49!>
2:05P
2:25P
Methyl
ethyl
ketone
(mg/L)
<0.10
190
130
0.27
0.12
200
6.5
1.2
140
Acetone
(mg/L)
<0.10
55.3
47
0.10
<0.1
7.2
0.16
0.80
2.1
2,2-Di-
methyl
oxirane
(mg/L)
<0.10
488
330
0.12
<0.1
41
0.49
4.9
10
Carbon
tetra-
chloride
(mg/L)
<0.10
15
<0.1
<0.1
<0.1
<0.1
<0.1
0.16
1.4
Methyl
alcohol
(mg/L
<0.10
14.3
11
<0.1
<0.1
6
0.98
<0.1
6.2
Laboratory analysis or vapors in headspacing above distillate. Vent samples
reported above the distillate headspace analysis were taken at approximately
the same time as the distillate. :
TABLE 60. AIR SAMPLES: ACETONE WASTE PROCESS
Sample
number
202
207
203
209
40
210
212
44
Field ambient
Condenser vent
Receiver vent
Product vent
Distillate
Condenser vent
Condenser vent
Composite distillate
Time
10:
10:
11:
11:
2:
2:
10:
3:
31A
42A
39A
OOA
36P
37P
OOA
SOP
Acetone
(mg/L)
<0
1
1
1
4S1
1
0
450
.1
.1
,5
.7
.3
.98
Methyl
ethyl
ketone
(mg/L)
<0
<0
<0
452
<0
<0
<0
0
.1
.1
.1
.1
.1
.1
.21
Chloro- Isopropyl Mettiylene
form alcohol chloride
(mg/L)
-------
TABLE 61. AIR EMISSIONS: METHYL ETHYL KETONE PROCESS
Source Time
Condenser vent 1 10:42A
Condenser vent 2 1:15P
Condenser vent 3 2: SOP
Condenser vent average
Accumulator vent 2:05?
VOC
(wg/L)
762
0.26
8.1
7.1
Flow
(L/s)
0.72
0.41
0.8
0.4a
Emissions
(g/s)
0.55
0.00011
0.0064
0.186
0.00283
aMaximum from working losses: not representative of process
TABLE 62. AIR EMISSIONS: ACETONE PROCESS
Source
Condenser vent
Condenser vent
Condenser vent
Receiver vent
Receiver vent
Time
1 10:31A
2 2:36P
average
10:42A
10:42A
VOC
(mg/L)
1.1
1.3
1.5
1.5
Flow
(L/s)
<1.3
<1.3
-------
r
1
DISTILLATION FIELD EVALUATION AT PLANT E
Process Description
Plant E is primarily engaged in reclamation and recycle of waste organic
solvents and contaminated products. The processes utilized at the Plant E
facility are distillation which includes fractionation and thin-film
evaporation. These are technologies which are of interest for removal of VOC
from waste streams.
Plant E is situated on about 7 acres of land, with the primary activity
being the reclamation and recycle of waste solvents and contaminated products.
Reclaimed solvents which are not recycled back to the generator are marketed
to suitable end uses. Special blends may be produced for these sales.
Approximately 23 million liters (6 million gallons) per year of solvent is
recovered.
The recovery and purification processes involve four distillation
systems:
One Votator thin-film evaporator,
One continuous fractionation distillation,
Two batch fractionation distillations.
Support facilities include a hot oil process heat system, a cooling water
system, and a building providing housing for offices and a laboratory. A
fleet of tank trucks is maintained for transport of all incoming and outgoing
bulk shipments.
Waste organic solvents and contaminated products are received in bulk and
drum shipments and are processed for reclamation and recycle. Each shipment
of chemicals received is analyzed to verify the manifest identification.
The contaminated solvent is processed through one of the four
distillation systems, with the reclaimed product being taken off overhead.
The system selected is dependent upon the characteristics of the particular
stream being processed, and in some cases additional purification may be
required.
Process Effectiveness
The process effectiveness of the distillation process was not evaluated
during the 1-day visit.
Process Residuals
Distillation bottoms are disposed of in one of several ways with the most
attractive option being utilization in fuel in cement kilns. In order to be
acceptable as fuel, a minimum heat value must be maintained, resulting in the
loss of some solvent, and the halogen content must not be greater than 4
144
-------
percent. Still bottoms which are unacceptable as fuel are solidified and
landfilled. In the case of dilute aqueous streams (less than 2 percent
organics), disposal is through offsite land treatment.
Process Cost
The following data were obtained from Plant E personnel:
1. The cost of simple distillation is in the range of $0.21 per liter
($0.80 per gal) of solvent recovered.
2. Reclamation cost could be as high as $0.70 per liter ($3 per gal) if
multiple distillations were required to achieve the desired level of
j purity.
) 3. It is not economical to reduce VOCs in aqueous streams below 2
; percent by fractionation at P^ant E. The cost could be as high as
: $1.50 per liter ($6 per gai)
4. Aqueous waste containing less than 2 percent organics can be land-
treated at a cost of about $0.11 per liter ($0.40 per gal). This
includes $0.022 per liter ($0.08 per gal) transportation cost.
5. It was estimated that the cost to ship waste to Houston
(approximately 1,500 miles) and incinerate would be in the range of
$0.79 per liter ($3 per gal).
i 6. The cost of landfill ing including transport, superfund tax, State
I tax, and landfill fee is now $0.30 per liter ($1,06 per gal) for
| hazardous waste and $0.33 per liter ($1.20 per gal) for waste which
j the State defines as "restricted." The above prices are only
! applicable if the material is already solid. The average price of
j waste disposal is $0.60 per liter ($2.18 per gal), including
I solidification.
145
-------
SECTION 10
SUMMARY
This section summarizes the differences among the various treatment
techniques and identifies similarities of applicability.
APPLICABILITY OF WASTE TREATMENT
i
i Of primary concern in assessing the applicability of a technique for VOC
j removal is the answer to the question, "Can it be used to treat the waste?"
i
J Liquids
Liquids are classified for pretreatment purposes into aqueous wastes,
mixed organic and aqueous wastes, and organic wastes. Other waste
characteristics include dissolved solids, suspended solids, and viscosity. A
thin-film evaporator can treat each of these waste categories, as long as the
viscosity is not greater than 10,000 cp. Some additional processing may ba
needed for the condensed aqueous vapors from the evaporators, which generally
will contain volatiles.
;
I Steam strippers can treat nonviscous liquids, with a limited ability to
i process wastes with dissolved and suspended solids.
i
! Distillation can be used to treat nonviscous liquids with restricted
abilities to process dissolved and suspended solids because of potential
reboiler heat exhange surface ana column plate fouling.
Reactive Wastes
Reactive wastes such as free-radical polymerization monomers must be
stabilized before processing with the thin-film evaporator. Problems ran
occur (Plant A, Plant B) when these materials are processed. Polymerization
in the reboiler of a distillation column is dangerous. A steam stripper can
be successfully used to process some of these reactive wastes because of the
lower temperature of the steam (Plant D).
Sludges
Thin-film evaporators can be used to process sludges (viscosity less than
100 poise). The amount of material recovered depends on the properties of the
bottoms. Sludges generally are net processed in the steam stripping unit at
Plant D and in the distillation units at Plant B.
146
-------
Solids
Unless the solids become liquids when heated, solids are not appropriate
for processing in thin-film evaporators, steam stripping units, or
distillation units. Some problems can occur for the recovery of organics
which solidify upon condensing the vapors. Scraping condensers or upflow
condensers are options for resolving condensing solids problems.
EFFECTIVENESS OF WASTE TREATMENT
The effectiveness of thin-film evaporators, sueam strippers, and
distillation to remove VOCs are compared. Tabl<» 63 presents a summary of the
percent VOC removal obtained for the processes and wastes investigated.
Liquids
Thin film evaporators are mainly used to remove VOCs by volume reduction.
In the processing of high boiling point chlorinated xylenes (Plant A), the
removal of VOCs was very high (>95 percent) but the removal of the chlorinated
xylenes was also high. When the liquid was primarily VOCs (Plant B and Plant
C), the bottoms still contained a high fraction of VOCs, but the volume was
reduced.
In the cases of steam stripping and distillation, the removal of VOCs
from the waste is characteristically high (>99 percent), and the VOCs can be
removed to achieve even lower concentrations in the residual waste by
extending the process time. The oil can be separated from the volatiles by
distillation and steam stripping.
Reactive >-.'aste
Reactive waste treatment was not investigated, but the removal
effectiveness is expected to be comparable to the liquids (if the waste can be
processed).
Sludges
The treatment of sludges was not investigated in thin-film evaporators,
but the removal effectiveness of VOCs from the waste is expected to be limited
(only 1-1.5 theoretical trays) unless the sludge is recycled.
Solids
The effectiveness of treating solids was not investigated.
COST OF WASTE TREATMENT AND RESIDUALS DISPOSAL
Table 64 presents general unit costs for waste treatment that were
obtained from the facilities during this study and which have have been
presented on a plant-by-plant basis in Sections 7-9. These are typical
facility costs, representing approximate 1984 annual averages as estimated by
the operators.
147
-------
TABLE 61. EFFECTIVENESS OF WASTE TREATMENT
Process
Thin film evaporator
Thin film evaporator
Thin film evaporator
Thin film evaporator
Thin film evaporator
Thin film evaporator
Thin film evaporator
Thin film evaporator
Steam stripping
Steam stripping
Steam stripping
Steam stripping
Distillation
Distillation
Plant
A
A
A
B
B
B
C
C
C
C
C
C
B
B
Waste
Oil
Oil
Oil
VOC
VOC
VOC
VOC resin
VOC resin
Aromatic in water
VOC in oil
Chlorinated in H?
VOC in water *•
MEK in water
Acetone in water
Compound
Toluene
Chloroform
Methyl ene chloride
Isopropyl alcohol
Freon TF
Xylenes
Acetone
Xylene
VOC
VOC
0 VOC
VOC
VOC
VOC
Percent
removal
95
>99
>99.91
-113?
86a
-77a
78
23
99.8
99.8
93
99
99
99.8
Percent reduction in vapor pressure of headspace.
148
-------
TABLE 64. 1984 UNIT COSTS PROVIDED BY PLANT PERSONNEL
Liquid disposal, landfarm
Liquid disposal, landfill
Bottoms disposal, landfill
Bottoms disposal, landfill
Bottoms disposal, fuel
Bottoms disposal, fuel
Bottoms disposal, incinerate"
Bottoms disposal, shipping, fuel
Aqueous bottoms disposal,
wastewater treatment
Operating cost organic overhead
Operating cost organic overhead
Operating cost organic as bottoms
Operating cost organic overhead
Operating cost multiple distillation
Operating cost simple distillation
Limit of economical VOC recovery
Limit of economical VOC recovery
so.ir
$0.60?
$0.30a .
S0.36/L + freight0
$0.079-50.13/L + freight
and drumming
$0.05/L:
$0.79/La A
$0.05-50.08/LQ
$0.00005/Le.
$0.26/L VOC°
S0.26/L VOC°
S0.40/L VOCQ .
$0.057-$0.33/L VOCT
50.70/L VOCa
S0.20/L VOCa
6-8 percent
2 percent
aPlant E distillation.
Plant C thin-film evaporator.
cPlant A thin-film evaporator.
Plant B thin-film evaporator.
eRTI estimate, presented for comparison.
Plant D direct steam stripper.
^Includes cost of transporting waste 1,500 miles.
149
-------
Table 65 summarizes waste-stream specific unit costs that were calculated
in the present investigation. Both operating and equipment costs are included
in the annualized costs in this table. Modelled (i.e., hypoethetical
treatment) cost estimates are given for thin-film evaporation and
distillation. In the case of steam stripping and one thin-film evaporator
facility (Plant A), sufficient data were available from the facility to
estimate unit operating costs for actual waste streams and equipment
combinations. Comparison of the unit costs calculated in Table 65 shows that
treatment costs vary as a function of treatment technique, initial VOC
content, and process load (volume/minute).
Thin-Film Evaporation
! Thin-film evaporators typically have a unit operating cost of approxi-
1 mately S0.26/L VOC recovered overhead (both Plants B and C). These costs are
! sensitive to both the concentration of VOCs and the volume of waste treated.
; The costs can increase to S0.30/L VOC recovered (excluding equipment costs)
for a waste with 2-percent VOC or decrease to a low of $0.033/1 VOC recovered
for gr«»atpr than 50-percent VOC recovery.
Steam Stripping
A range of product recovery costs for the steam stripping system at
Plant D was estimated by plant personnel to typically be $0.057-$0.33/L VOC
recovered. This range is approximately the same as for the thin-film
evaporator. However, one of the three waste streams that were being processed
by the plant at the time of the field assessment was estimated to have a
! treatment cost of $0.£3/L, while the fourth (synthetic) stream, which has a
| VOC concentration lower than the plant typically processes, was estimated to
| cost $4.34/1 VOC removed.
i
I Distillation
$
I The costs of distillation are estimated by Plant E to range from $0.20/L
J of recovered VOC for simple distillation to $0.70/1 of recovered VOC fcr
| tiiultiple distillation. The cost of simple distillation is within the same
| range as that estimated for thin-film evaporators and steam stripping.
§ However, the cost for VOC removal from a hypothetical stream having low
s initial VOC content (3 percent) was estimated to be $1.81/L (Table~65),
| indicating that distillation costs may increase significantly as the initial
I VOC content of the waste decreases.
1
I Economical Recovery of VOCs
i A TSDF can treat the waste and recover the VOCs, or the T5DF can dispose
| of the waste. Table 66 presents the results of an analysis of the cost of
j treating various concentrations of VOC in the waste streams with the same
I characteristics as Batch 1 at Plant D. The cost of treating the waste
I increases with decreasui9 VOC content because of the low recovered VOC credits
I possible at low initial concentrations and because the VOC removal rate is
I proportional to the VOC content of the waste.
E
1
! " 150
-------
TABLE 65. 1984 UNIT COSTSa OBTAINED FROM THE CURRENT INVESTIGATION
Thin film evaporator
Modelled6
Annual ized cost
Flow rate (L/M)
VOC content (percent)
Unit Costs
S/L Waste treated
$/L VOC recovered
$/Mg VOC (Density = 0.8)
157.5
40
2
0.0075
0.37
462
Plant Ac
252.6
32
85
0.0279
0.033
41.1
Distillation
Modelled
359.2
260
3
0.035
1.18
1,475
Steam stripping (Plant
Batch 1
262.9
8
26
0.08
0.30
376
Batch 2
262.9
6
74
0.12
0.17
210
Batch 3
262.9
5
18
0.15
0.53
659
D)e
Batch 4
262.9
5
3
0.16
4.34
5,358
Includes waste-stream specific operating costs and annualized equipment costs.
3Source: Table 15
:Source: Table 20
Source: Section 6, COST
^Source: Tables 28 and 46
-------
TABLE 66. THE COSTS OF STEAK STRIPPING AS A FUNCTION
OF THE VOC CONTENT OF THE WASTE
Fraction
VOC content
0.35
0.26
0.10
0.05
0.01
Product credit3
203,260
151,000
58,000
290,000
5,800
$ recovery/1
0.020
0.039
0.071
0.081
0.089
aBased on e waste volume of 2,904,000 L/year, $0.20/1 VOC product recovery
credit, 99.7 percent recovery.
Based on an operating cost of $262,900/'year.
TABLE 67. A COMPARISON OF TREATMENT VERSUS DISPOSAL COSTS
FOR 200 L (55 GALLONS) OF AQUEOUS WASTE CONTAINING 2-5% VOC
Treatment Costs
Treat Stream of 2% VOC by thin-film evaporation8
Treat Stream of 3% VOC by distillation
Treat Stream of 3% VOC by steam stripping.
Treat Stream of 5% VOC oy steam stripping
Disposal Costs
Land Treatment
Landfill
Incinerate
$2
$7
$37
$16
$22
$120
$158
aBasis: Table 65
bBasis: Table 66
CBasis: Table 64, Plant E.
152
-------
Waste treatment costs are compared to selected waste disposal costs in
Table 67. Treatment costs for aqueous waste containing 2-5 percent VOC are
presented, based on Tables 65 and 66. Costs are conpared for treating or
disposing of 200 L (55 gallons) of waste, typical of a disposal decision faced
by many generators. It is assumed that once the waste has been treated to
remove VOCs, the waste is no longer hazardous, i.e., the cost of disposal of
the treated waste is negligibly small. This a good assumption, for instance,
if the treated aqueous effluent is acceptable for discharge to a municipal
sewer. Based upon this analysis, the cost of steam stripping is less than
disposal costs by landfilling and incineration, but competitive with land
treatment.
The various plants suggested that a limit of economic viability exists
for wastes with low VOC contents. These limits range from 2 percent at Plant
F to 6-8 percent at Plant B. The range of costs for economic viability
suggests that either the disposal costs are overestimated or the waste
treatment facilities are utilized at a level which is substantially under
capacity. If the TSDF has an onsite disposal method for dilute VOC treatment,
the disposal costs could be substantially less than commercial rates.
MEASURED AIR EMISSIONS
Air emissions were measured at several of the TSDF processes which were
investigated. The results are presented in Tables 68 and 69.
Thin-Film Evaporators
Of the three thin-film evaporators evaluated, two were operating under
vacuum (Plants A and B) and one was operating under atmospheric pressure
(Plant C). Visible emissions were observed from the vacuum vent (connected to
the process). The data from the vacuum pumps are not supported with measured
volumetric flow rates, but relatively high concentrations of VOCs were
associated v;ith the vacuum pump vent. In contrast, the concentration and flow
rate from the atmospheric thin-film evaporator process were negligible during
the process assessment at Plant C. The emission factor (g/g VOC recovered)
was estimated to be relatively low (less than 2 percent) for the thin-film
evapjrators operating under vacuum, but substantially greater than the vents
from the steam stripping and distillation processes.
Steam Stripping
The condenser on the steam stripping unit at Plant D was vented onto the
roof. The emissions were greater at the beginning of the process and
decreased toward the end of the run as the concentrations of the VOCs in the
batch were reduced. The emissions were 0.0029 g/sec for the first batch and
0.0035 g/sec for the second batch. The emission factors were 5.7 x 10~5 and
2.7 x 10~5 g/g VOC, respectively, for these two batches. This represents only
a small fraction of the VOCs recovered.
153
-------
TABLE 68. CONDENSER VENT EMISSIONS
—
Location
Plant D
Plant D
Plant B
Plant B
Plant A
Plant B
Plant C
Source
Condenser
Condenser
Condenser
Condenser
Condenser
Condenser
Condenser
vent,
vent,
vent,
vent,
vent,
vent.
vent,
Batch 1
Batch 2
MEKa
Acetone
TFEC
TFE
TFEd
Rate
(g/sec)
0.0029
0.0035
0.186
<0.0016
-
-
Emission factor
(g/g voc)
5.7 x 10"5
-5
2.7 x 10 °
-4
6.0 x 10
_7
9.8 x 10
0.026
••
No measurable
emissions
aBatch size = 30,280 L x .044 Kg/L; 1,332 kg, 4,320 sec.
bBatch size = 11,335 L x .227 Kg/L = 3,573 kg, 2,184 sec.
C0.02 x .57 + 0.015 x .99 = 0.026, vacuum process.
d/
Nonvacuum process,
Location
Plant C
Plant D
Plant D
Plant B
Plant B
TABLE 69. STORAGE TANK EMISSIONS
Source
Outside storage tank
Inside vented storage tank
Inside vented MST
Accumulator tank
Receiver vent
Rate
.63 g/sec
0.0004 g/sec
0.0024 g/sec
0.0028 g/sec
0.0015 g/sec
r i
I !
154
-------
Distillation
t Although the process condenser vent for the distillation of the first
batch at Plant B had a much higher emission rate than the condenser vent at
Plant D (0.186 g/sec) by a factor of 60, the emission factor (0.0006 g/g VOC)
was only a factor of 10 greater. The estimated emissions for the second batch
was comparable to the emissions from Plant D (factor of 2 lower), but due to
the greater batch volume at Plant B, the emissions factor was much lower
(factor of 100). The distillation vent emissions were much le^s than the
recovered VOCs.
i Storage Tanks
The air emissions estimates for product and waste storage tanks are
presented in Table 69. In general, the emissions are relatively low compared
to the amount of waste processed. The open storage tank at Plant C had
substantially greater emissions (0.63 g/sec) than comparable enclosed storage
tanks (by a factor of 250), suggesting that enclosing and,venting the tanks
can reduce emissions by greater than 99 percent.
155
-------
REFERENCES
Allen, C. C. and G. Brant. 1984a. Hazardous Waste Treatment for Emissions
Control: Field Evaluations at Plant A, EPA Contract rlo. 68-03-3149, Work
Assignment 25-1 (unpublished).
t
Allen, C. C. and G. Brant. 1984b. Hazardous Waste Treatment for Emissions
Control: Field Evaluations at Plant 3, EPA Contract No. 6J-03-3149, Work
Assignment 25-1 (unpublished)..
Allen, C. C. and G. Brant. 1984c. Hazardous Waste Traatment for Emissions
Control: Field Evaluations at Plant C, EPA Contract No. 63-03-3149, Work
Assignment 25-1 (unpublished).
Allen, C. C. and G. Brant. 1984d. Hazardous Waste Treatment for Emissions
Control: Field Evaluations at Plant D, EPA Contract No. d8-03-3149, Work
Assignment 25-1 (unpublished).
Allen, C. C. and G. Brant. 1984e. Hazardous Waste Treatment for Emissions
Control: Field Evaluations at Plant E, EPA Contract No. 68-03-3149, Work
Assignment 25-1 (unpublished).
Allen, C. C. and G. Brant. 1984f. Hazardous Waste Treatment for Emissions
Control: Field Evaluations at Plant F, EPA Contract No. 68-03-3149, Work
Assignment 25-1 (unpublished).
Allen, C. C. and G. Brant. 1984g. Hazardous Waste Treatment for Emissions
Control: Field Evaluations at Plant G, EPA Contract No. 68-03-3149, Work
|, Assignment 25-1 (unpublished).
I APV Equipment, Inc., 395 Fillmore Avenue, Tonawanda, NY. Distillation
I ; Handbook, DH-682, 2 no. edition.
I i
Berkowitz, J. B., et al. 1978. Unit Operations for Treatment of Hazardous
Industrial Wastes. Arthur D. Little, Inc.
Breton, M., et al_. 1983. Assessment of Air Emissions from Hazardous Waste
Treatment, Storage, and Disposal Facilities (TSDFs) Preliminary National
Emissions Estimates, GCA-TR-83-70-G.
Chem-Pro Equipment Corp., Fairfield, NJ. Preasseirbled Process Plants, Cost
and Time Effective, pp. 200.
Danaher, R. 1984. Luwa, Charlotte, NC, Personal Communication.
156
-------
REFERENCES (Continued)
Dietz, Stephen, et al. 1984. National Survey of Hazardous Waste Generators
and Treatment, Storage, and Disposal Facilities regulated under RCRA in
1981. EPA 530/SU-84-005, April 1984.
Exner, J. ri., ed. Detoxication of Hazardous Waste. Ann Arbor Science, Ann
Arbor, Michigan, 1982, pp. 3-41.
Fischer, R. 1965. Agitated Thin-Film Evaporators, Process Applications,
I | Chemical Engineering, September 13, 1965, 1:186, McGraw-Hill.
M King, C. J. 1977. Separation Processes. McGraw-Hill.
| I ~L jLM^U-iL J-LT ,__— U^U U , _ I . " __. '.. -__
McCabe, W. L., and
-------
APPENDIX A
SUMMARY OF ANALYTICAL DATA
Table A-l presents the name of the plant in the first column, followed by
the location of the sample in the process, and then followed by either the
batch or process from which the sample was taken. The concentration in the
aqueous phase of the sample (or the concentration in the organic phase if the
sample is mainly organic) is presented in the fourth column, followed by the
calculated value of Henry's constant for the sample and the concentration in
the vapor phase. The partition ccefficient (nole fraction, Y, in the vapor
divided by the mole fraction, X, in the liquid) is presented in the seventh
column. The Henry's constant is not defined by the partition coefficient for
complex waste mixtures, as it is for dilute aqueous mixtures. The calculated
values of the mole fraction of the VOCs in the liquid and vapor are presented
in the last two columns.
The following abbreviations are used:
REB - reboiler contents
DISTIL - distillate
MST - miscible solvents tank
STRIP. - stripper contents
CONDEN - condensate
TFE - Thin-film evaporator
3 3
The units of Henry's constants are atm-m /mole, A - m m. The Henry's
constants have been multiplied by 1,000 for presentation in this table. Note
that this is a computer listing and therefore more significant figures are
sometimes shown than are warranted by the precision of the results. Refer to
the text for the correct number of significant figures.
158
-------
TABLE A-1. ANALYSIS OF LIQUID HASTE SAMPLES
PUNT
SAMPLE BATCH LIQUID
NUMBER LOC. OR COMPONENT CCNC.
H CONST.
VAPOR
CONC.
MOLECULAR
K HEIGHT X
If
PROCESS (uig/L) A/H*H3 *E-3 (mg/L) (Y/X)
PLANT 8
PLANT 8
PLANT 8
PLANT 8
PLANT 8
PLANT 8
PLANT'S
PLANT B
PLANT 8
PLANT 8
PLANT 8
PLAST 8
PLANT 8
PLANT 8
PLANT 8
PLANT 8
PLANTS
PLANT B
PLANT 8
PLANT 8
PLANTS
PLANT B
?LANT 8
PUNT B
PLANT 8
PLANT 8
PLANT 8
PLANT 6
PLANT 8
PLANT B
PLANT 8
PLANT B
PLANT 8
PLANT 8
PLANT 8
PLANT 8
PLANT 8
PLANT 8
PLANT 8
PLANT B
PLANTS
PLANT 8
25
.30
25.30
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
:s
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
.30
.30
30
.30
.30
.30
.30
.30
.30
.30
.31
.31
.31
.31
.31
.31
.31
.31
.31
.31
.31
.31
.33
.33
.33
.33
.33
.33
.33
.33
.33
.33
.33
.33
.34
.34
.34
.34
.34
.34
RES 1 METHANOL
RE8 J METHYLENE CHLORIDE
RES 1 ACETONE
RE6 1 ISOPROPANOL
RES 1 OIHETHYL OXIRANE 2,2
RE8 1 METHYL ETHYL KETOfiE
RES 1 TR1CHLOROETHANE 1,1,1
RE8 1 1 CARBON TETRACKLORIDE
RES 1 1 TOLUENE
REB 1 XYLENES
RE8 1 WATER
REB 1 TOTAL ORGAN ICS
DISTIL. METHANOL
DISTIL. METHYLENE CHLORIDE
DISTIL. J ACETONE
DISTIL. 1 ISOPROPANCL
DISTIL. ! 8ROWOOICHLORGMETHANE
DISTIL. . 1 METHYL ETHYL KETONE
DISTIL. 1 TRICHLOROETHANE 1,1,1
DISTIL. 1 TRICHLOROETHENE
DISTIL. 1 CHLOROFORM
DISTIL.
DISTIL.
DISTIL.
RES 2
REB 2
RES 2
UE8 2
RES 2
RES 2
RES 2
REB 2
RES 2
REB 2
RES 2
REB 2
DISTIL.
DISTIL.
DISTIL.
DISTIL.
DISTIL.
DISTIL.
DIMETHYL OXIRANE 2,2
MATER
TOTAL OR6ANICS
METHANOL
METHYLENE CHLORIDE
ACETONE
ISOPROPANOL
DIMETHYL OXIRANE 2,2
METHYL ETHYL KETONE
TRICHLOROETHANE 1,1,1
CARBOf TETRACHLORIDE
TOLUENE
XYLENES
HATER
TOTAL OR6ANICS
METHANOL
METHYLENE CHLORIDE
ACETONE
ISOPROPANOL
CHLOROFORM
METHYL ETHYL KETONE
3500
3100
0
1900
6400
30000
710
1700
18
0
95.27*
47328
40000
2000
71000
1000
10
440000
5000
300
10
210000
23.07%
769320
3400
0
100
2500
490
9400
550
2200
18
0
98. 13*
18658
42000
2000
13000
45000
10
670000
0.00*93
0.00324
S. 00994
0.07696
5.06225
0.01156
0.02052
0.27362
540 90
0.00564
0.01744
0.01358
0.00431
5.13046
0.00606
0.01642
0.05609
2.05218
0.03225
2233.84
0.01086
0.03078
0.01313
0.01550
0.04585
0.03022
0.01026
0.00000
525.10
0.00293
0.00287
0.01137
0.00064
O.S3296
0.00613
0.84
0.49
3.00
0.92
24.00
91.00
0.40
1.70
fl.2<
0.23
25.11
11.00
1.70
47.00
0.21
2.50
130.00
4.00
0.82
1.00
330.00
25.11
1.80
0.21
0.15
1.60
0.37
21.00
0.81
1.10
0.00
0.23
25.11
6.00
0.28
7.20
1.40
0.44
200.00
0.263
0.173
0.532
4.124
3.336
0.619
t.099
14.66
0.028
0.133
0.411
0.320
0.101
121.0
0.143
0.387
1.323
48.41
0.760
0.052
0.595
1.687
0.720
0.849
2.5!3
1.656
0.562
0
0.023
0.053
0.052
0.208
0.011
16.54
0.112
32
85
53
60
99
72.1
133
153.8
92.1
106
18
32
85
58
60
164
72.1
133
153.8
119.4
99
18
32
85
58
60
99
72.1
133
153.3
92.1
106
18
32
85
58
60
119.4
72.1
0.002040
0.000530
0
0.000553
0.001206
0.007752
0.000099
O.OOP20S
0.000003
0
0.987410
0.052930
0.000997
0.051834
0.000706
0.000002
0.258658
Q.C"1593
0.000082
0.000003
0.08990S
0.543133
0.001938
0
0.000031
0.000760
0.000030
0.002378
0.000075
0.000260
O.G00003
0
0.994462
0.071646
0.001284
0.012235
0.040940
O.OOOOC4
0.507264
0.000538
0.000118
0.001061
0.000314
0.004974
0.0259S1
O.OOC051
0.000226
0.000053
0.000344
0.028627
0.007054
0.000410
0.016629
0.000071
0.000312
0.037001
0.000617
0.000109
o.oooni
0.068406
0.028627
0.001154
0.000050
0.000053
0.000547
0.000076
0.005977
0.000124
0.000145
0
0.000044
0.023827
0.003847
0.000067
0.002547
0.000478
O.QQOQ75
0.056926
(CONTINUED)
159
-------
TABLE A-1. (CONTINUED)
SAMPLE BATCH
PLANT NUMBER LOC. OR COMPONENT
PROCESS
TRICKLOROETHANE 1,1,1
1 BROMODICHLQROMETHANE
TOLUENE
DIMETHYL OXIRANE 2,2
WATER
TOTAL OR6ANICS
METHANOL
METHYLENE CHLORIDE
ACETONE
ISOPROPANOL
DIMETHYL OXIRANE 2,2
METHYL ETHYL KETONE
TRICHLOROETHANE 1,1,1
CARBON TETRACHLORIDE
TOLUENE
XYLENES
HATER
TOTAL ORGANIC3
METHANOL
HETHYL ETHYL KETONE
TRICHLOROETHANE 1,1.1
CARSON TETRACHLORIDE
TOLUENE
BENZENE
1 HATER
TOTAL ORGANICS
ACETONE
ISOPROPANOL
ETHYL8ENZENE
METHYL ETHYL KETONE
TRICHLOROETHANE 1,1.1
i TSICHLOROETHENE
2 TOLUENE
2 XYLENES
HATER
TOTAL ORGANICS
CARSON TETRACHLORIOE
BROMOCHLOROMETHAUE
ACETONE
CHLOROFORM
8RCMOOICHLOROMETNANE
2 TRICHLOROETHANE 1,1,1
TRICHLOROETHENE
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PUNT
PLANT
PLANT
PLAN1
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PUNT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PUNT
PLANT
PLANT
PLANT
PUNT
8
8
8
8
8
8
6
B
B
8
8
B
8
B
B
B
8
B
8
B
B
8
8
B
8
B
B
B
B
B
8
3
6
B
8
B
8
8
B
8
8
8
B
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
<5
25
25
25
25
25
25
25
25
25
25
25
25
.34
.34
.34
.34
.34
.34
.36
.36
.36
.36
.36
.36
.36
.36
.36
.36
.36
.36
.37
.37
.37
.37
.37
.37
.37
.37
.39
.39
.39
.39
25.39
25
25
25
25
25
25
25
25
25
25
25
25
.39
.39
.39
.39
.39
.40
.40
.40
.40
.40
.40
.40
DISTIL.
DISTIL.
DISTIL.
DISTIL.
DISTIL.
1
1
DISTIL. 1
RE8
REB
REB
REB
REB
RE8
RES
REB
REB
REB
RES
REB
REB
RES
REB
RES
RES
RES
RES
REB
*E8
REB
RES
REB
REB
Sea
RES
RE8
REB
RES
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
1
1
1
1
1
i
1
1
1
DISTIL.
DISTIL.
DISTIL.
DISTIL.
DISTIL.
DISTIL.
DISTIL.
1
1
2
2
2
2
2
i
2
I
2
2
2
2
2
2
2
2
2
LIQUID
CONC.
H
CONST.
VAPOR
CONC.
(cig/L) A/M*M3 *E-3 (rog/L)
20000
10
15000
95000
9.80*
902020
0
0
0
52
0
61
350
0
22
0
99.951
485
65
10
1100
110
17
210
99.851
1512
212000
440
91
2300
2800
9500
2700
200
76.85*
230031
10
10
850000
10
8000
28000
42000
0
1
0
0
.00083
.27235
.00030
.0068S
5259.27
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
1
0
3
0
0
0
.00000
.00000
.01466
.00000
515. SS
.00000
.24626
.00000
.00000
.00000
.00000
516.02
.03291
.00000
.10374
.00357
.00491
.00000
.00000
.53339
670.57
.46262
.19027
.01183
.89915
.00282
.00110
.00018
0
0
0
41
25
0
0
0
0
0
0
0
0
0
0
25
0
25
340
0
0
0
1
25
1
0
.81
.62
.22
.00
.11
.20
.10
.00
.00
.00
.00
.25
.00
.00
.00
.11
.12
.11
.00
.46
.40
.67
.30
.11
.20
.58
490.00
1
1
1
.90
.10
.50
0.37
0.
MOLECULAR
K WEIGHT X
Y
(Y/X)
015
23.30
0.
0.
0.
0.
0.
005
162
096
0
0
814
0
028
0
13.66
0.
1.
4.
0.
0.
5.
0.
44
21
0.
70
0.
0.
0.
0
0
0
0
028
530
0
823
165
223
0
0
202
031
.71
.61
214
.79
051
019
003
133
164
92.1
99
18
32
85
58
60
99
72.1
133
153. 8
92.1
106
18
32
72.!
133
153.8
92.1
78
18
58
60
99
72.1
133
153.8
92.)
106
13
153.3
129.5
58
119.4
(64
133
153.8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.008208
.000003
.008890
.052382
.297139
0
0
0
.000015
C
.000015
.000047
0
.000004
0
.999917
.000036
.000002
.000149
.000012
0
0
0
0
0
0
0
0
0
0
0.000003
0
0
0
0
0
0
0
0
0
0
0
.000048
.999747
.073603
.fl'00157
.000019
.000686
.000452
.001328
.000630
.000040
.918080
0.000003
0
0
0
0
0
0
.000004
.807110
.000004
.002586
.011594
.015033
0
.000124
.000077
.000049
.008496
.028627
.000128
.000024
0
0
0
0
.000033
0
0
0
.028627
0
.000034
0
0
0
0
.028627
0.120300
0
0
0
0
0
0
0
0
0
0
0
0
0
.000095
.000113
.000103
0
0
.000251
.023627
.000160
.000091
.173374
.000325
.000137
.000231
.000049
(CONTINUED)
160
-------
TAELE A-I. (CONTINUED)
SAMPLE BATCH
PLANT KUMBER LOC. OR COMPONENT
PROCESS
2 TOLUENE
2 TETRACHLOROETHENE
2 HATER
2 TOTAL ORGAN1CS
2 ISOPROPANOL
2 CHLOROFORM
2 ACETONE
2 TETRACHLORCETHENE
2 ETHYL8ENZENE
2 METHYL ETHYL KETONE
2 TRICULOROETHANE 1,1,1
2 TOLUENE
2 XYLENES
2 MATER
2 TOTAL ORGANICS
2 ISOPROPANOL
2 CHLOROFORM
2 ACETONE
2 TETRACHLOROETHENE
2 ETHYLBENZENE
2 METHYL ETHYL KETONE
2 TRJCHLOROETHANE 1,1,1
2 TOLUENE
2 XYLENES
2 HATER
2 TOTAL ORGANICS
2 ISOPROPANOL
2 CHLOROFORM
2 ACETONE
2 TRICHLOROETHANE 1,1,1
2 XYLENES
2 HATER
2 TOTAL ORGAN[CS
2 DICHLOROETHENE
2 DIMbTHYL OXIRANE 2,2
2 ACETONE
2 CHLOROFORM
2 BROJtODlCHLOROMETNANE
2 METHYL ETHYL KETONE
2 TRICHLOROETHANE 1.1.1
2 TRICHLOROETHENE
2 TOLUENE
2 DICHLOROPRQPANE 1.2
PUNT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PUNT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PUiNT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PUNT
PUNT
PLANT
PLANT
PUNT
PUNT
PUNT
PLANT
PUNT
8
B
8
S
8
8
8
B
B
e
8
B
8
B
8
B
B
8
B
B
B
B
8
B
8
B
8
D
8
B
e
B
6
B
B
B
B
8
8
B
B
8
B
25
25
25
25
25
25
25
25
*5
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
.40
.40
.40
.40
.41
.41
.41
.41
.41
.41
.41
.41
.41
.41
.41
.4:
.42
.42
.42
.42
.42
.42
.42
.42
.42
.42
.43
.43
.43
.43
.43
.43
.43
.44
.44
.44
.4«
.44
.44
.44
.44
.44
.44
DISTIL.
DISTIL.
DISTIL.
DISTIL.
RES 2
RE3 2
RE3 2
RE8 2
RES 2
RES 2
RE3 2
RE3 2
RES 2
RES 2
RE8 2
RE3 3
RES 3
RE3 3
RES 3
RE8 3
RES 3
REB 3
RES 3
REB 3
REB 3
RES 3
RES 4
RES 4
RES 4
REB 4
RE3 4
REB 4
REB 4
DISTIL.
DISTIL.
DISTIL.
DISTIL.
DISTIL.
DISTIL.
DISTIL.
DISTIL.
DISTIL.
DISTIL.
[QUID
3NC.
H
CONST.
VAPOR
CONC.
sg/L) A/M»M3 *E-3 (isg/L)
23000
to
0
1
.00000
.51862
(.90110524.98
951040
430
10
140000
10
41
1300
1100
820
110
85.52%
143821
110
10
4100
10
10
14
32
28
10
99.571
4324
13
10
690
10
10
99.93*
733
to
10
840000
10
22000
10
11000
8000
31000
10
0
1
0
0
0
0
0
G
0
0
2
0
0
0
0
0
0
0
0
1
0
0
0
1
0
0
2
0
0
0
0
0
0
.00000
.08766
.04104
.65670
.27529
.OD789
.00392
.04505
.41044
601.86
.00000
.46:-52
.03554
.30035
.41044
.00000
.31424
.00000
.75931
517.54
.00000
.37496
.06246
.45)48
.22574
515.68
.08766.
.57451
.01099
.46262
.00038
.43096
.00095
.03059
.00008
.20522
0
0
25
0
230
0
0
0
0
1
2
25
1
7
0
0
0
0
25
0
2
0
0
25
0
0
450
1
0
0
0
0
0
0
.00
.74
.11
.53
.00
.32
.55
.50
.21
.80
.20
.11
.20
.10
.39
.2C
.49
.37
.11
.67
.10
.22
.11
.11
.53
.28
.00
.20
.41
.21
.51
.23
.12
.10
27
0.
54
2.
32
13
0.
0.
2.
20
0.
K
MOLECULAR
HEIGHT X
Y
(Y/X)
0
.57
191
0
.40
053
.85
.77
394
195
253
.53
030
0
136.4
1.
44
22
17
42
0.
76
3.
25
12
0.
21
11
0.
49
0.
8.
0.
0.
0.
4.
968
.33
.73
0
.40
0
.05
028
0
.34
468
.06
.53
028
.72
.47
219
.19
007
608
019
Oil
001
099
92.1
165
18
60
119.4
58
165
99
72.1
133
92.1
106
18
60
119.4
58
165
99
72.1
133
92.1
106
18
60
119.4
58
133
106
18
97
85
58
119.4
164
72.1
133
153.8
92.1
106
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
a
0
0
0
0
d
0
0
0
0
0
0
0
0
0
0
0
.013753
.000003
.149799
.003143
.000801
.048253
.003001
.000008
.000360
.000165
.000177
.000020
.950867
.000033
.000001
.001276
.000001
.000001
.000003
.000004
.000005
.000001
.998671
.000003
.000001
.000214
.000001
.000031
.999777
.000005
.003005
.725049
.000004
.005715
.000006
.004140
.002604
.016850
.000004
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.000092
.028627
0
.000091
.099070
.000039
.000114
.000142
.000032
.000401
.000425
.028527
0
.300205
.002512
.000048
.000041
0
0.000075
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.000071
.028627
0
.000115
.000713
.000033
.000021
-0;8627
.000112
.OC0067
.139221
.000206
.001)051
.000059
.000078
.000030
0.000026
0.000019
161
(CONTINUED)
-------
TABLE A-l. (CONTINUED)
PLANT
PUNT 8
PLANT 8
PLANT C
PLANT C
PLANT C
PLANT C
PLANT C
PLANT C
PLANT C
PLANT C
PLANT C
PLANT C
PLANT C
PLANT C
PLANT C
PLANT C
PLANT C
PLANT C
PLANT C
PLANT C
PLANT C
PLANT C
PLANT C
PLANT C
PLANT C
PLANT C
PLANT C
PLANT C
PLANT C
PLANT A
PLANT A
PLANT A
PLANT A
PLANT A
PLANT A
PLANT A
PUNT A
PLANT A
PUNT A
PLANT A
PLANT A
PLANT A
PLANT A
SAMPLE
NUMBER LOC.
25.44 DISTIL.
25 44 DISTIL.
ER1 FEED
ER1 FEED
ER1 FEED
ER1 FEED
ERI FEED
ER1 FEED
ERI FEED
ERI FEED
ERI FEED
ER2 DISTIL.
ER2 DISTIL.
ER2 DISTIL.
ER2 DISTIL.
ER2 DISTIL.
ER2 DISTIL.
ER2 DISTIL.
ER2 DISTIL.
ER2 DISTIL.
ER3 FINAL
ER3 FINAL
ER3 FINAL
ER3 FINAL
ER3 FINAL
ER3 FINAL
ER3 FINAL
ER3 FINAL
ER3 FINAL
10 PRODUCT
10 PRODUCT
10 PRODUCT
10 PRODUCT
10 PRODUCT
10 PRODUCT
10 PRODUCT
11 FEED
11 FEED
11 FEED
11 FEED
11 FEED
11 FEED
11 FEED
BATCH
Oil COMPONENT
PROCESS
2 MATE?
2 TOTAL ORGAN 1CS
TFE ACETONE
TFE FREON TF
TFE TftfCHLOROETHANE
TFE TRICHLOROETHYLENE
TFE TOLUENE
TFE ETHYL BENZENE
TFE XYLENES
TFE TETRACHLOROETHYLENE
TFE TOTAL ORGAN ICS
TFE ACETONE
TFE FREON TF
TFE TRICHLOROETHANE
TFE TRICHLOROE1HYLENE
TFE TOLUENE
TFE ETHY. BENZEhc
TFE XYLENES
TFE TETRACHLOROETHYLENE
TFE TOTAL ORGANICS
TFE ACETONE
TFE FREON TF
TFE TRICHLOROETHANE
TFE TRICHLOROETHYLENE
TFE TOLUENE
TFE ETHYL BENZENE
TFE XYLENES
TFE TETRACHLOROETHYLENE
TFE TOTAL ORGAN ICS
TFE CHLOROFORM
TFE KETWYLENE CHLORIO
TFE TOLUENE
TF£ TRICHLOROETHANE
TFE FREON TF
TFE HIGH 80ILESS
TFE TOTAL ORGANICS
TFE CHLOROFORM
TFE HETHYLENE CHLORID
TFE TOLUENE
TFE TRICHLOROETHANE
TFE FREON TF
TFE HIGH BOILERS
TFE TOTAL ORGANICS
QUID
INC.
>g/L)
8.801
912050
743000
1000
15000
2000
5000
0
59000
6000
831000
822DOO
1000
22000
3000
9000
3000
20000
SOOO
885000
606000
1000
9000
1000
3000
1000
136000
9000
766000
0
SOOO
16000
0
18000
939000
982000
15000
20000
13000
7000
0
944030
999000
VAPOR
H CONST. CONC.
A/H*M3 *E-3 (rng/L)
5859.05 25.11 0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.01044
.04104
.02449
.00103
.00123
.00073
.00821
.00956
.04104
.01782
.00068
.00046
.00068
.00021
.00657
.01043
.03078
.02098
.00205
.00274
.00205
.00075
.00935
.00221
.00005
.00171
.00000
.00698
.00174
.0(002
.00032
.00000
378
2
17
0
0
0
2
2
383
2
19
0
0
0
0
1
.00
.05
.98
.10
.30
.10
.10
.40
.00
.00
.10
.10
.20
.10
.20
.60
308.00
1
9
0
0
0
5
4
0
0
0
0
1
0
5
1
0
0
0
0
.50
.20
.10
.40
.10
.00
.10
.14
.97
.04
.14
.50
.00
.10
.70
.01
.11
.06
.00
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
MOLECULAR
K WEIGHT X
(Y/X)
.117 18 0.244612
.135
.532
.317
.013
.015
.009
.106
.137
.590
.256
.009
.006
.009
.002
.094
.110
.325
.222
.021
.028
.021
.007
.098
.011
.000
.008
0
.036
.009
.000
.001
0
58
186.3
133
131
92.)
106
106
165.8
58
186.3
133
131
92.1
!06
106
165.8
58
185.3
133
131
92.1
106
106
165.8
119.3
85
92.1
133
186.3
200
119.3
85
92.1
133
186.3
200
0
0
5
0
0
0
0
0
0
0
0
0
.986933
.000413
.5G86S8
.001176
.004182
0
.042881
.002788
.984.451
.000372
.011490
.001590
.006787
0.001965
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.013106
.002094
.987248
.000507
.006394
.000721
.003077
.000891
.121231
.005129
0
.020879
.034256
0
.019052
.925811
.023836
.044607
.026759
.009977
0
.894818
0.
0.
0.
0.
0.
0.
0.
0.
3.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
Y
028627
13374S
000220
00275?
800015
000066
000019
000406
000297
135514
000220
002947
000015
000044
000019
000038
000193
108977
000165
001419
000015
000039
000019
00096$
000507
000024
000234
000003
000021
000165
0
000877
OC0410
000002
000016
000006
fl
(CONTINUED)
162
-------
TABLE A-1. (CONTINUED)
SAMPLE BATCH
PLANT NUMBER LOC. OR COMPONENT
PROCESS
LIQUID VAPOR MOLECULAR
CONC. H CONST. CONC. K WEIGHT X
(Btg/L) A/M«M3 »E-3 («g/L) (Y/X)
PLANT A
PLANT A
PLANT A
PLANT A
PLANT A
PUNT A
PLANT 8
PLANT 8
PLANT 8
PLANT B
PLANT 8
PLANT 8
PLANT 8
PLANT 8
PLANT B
PUNT 8
PLANT 8
PLANT 8
PUNT 8
PLANT B
PUNT B
PUNT 8
PLANT 8
PLANT 8
PLANT B
PLANT 8
PUXT 8
PLANT B
PLANT 8
PUNT B
PUNT 0
PUNT 0
PUNT 0
PLANT D
PUNT D
PLANT D
PUNT D
PLANT D
PLANT 0
PLANT D
PLANT 0
PLANT D
PLANT 0
\2
12
12
12
12
12
21
21
21
21
21
21
22
22
22
22
22
22
25
25
25
25
25
25
25
26
26
26
26
26
45.03
45.03
(5.03
45.03
4S.03
45.03
45.03
45.03
45.03
45.03
45.03
45.03
45.05
BOTTOMS
BOTTOMS
BOTTOMS
BOTTOMS
BOTTCXS
BOTTOMS
FEED 82
FEED 82
FEED 82
FEED 82
FEED 82
FEED 12
FEED SI
FEED 81
FEED 81
FEED »1
FEED 81
FEED 81
PROD. 81
PROD. 81
PROD. 81
PROD. 81
PROD. 81
PROD. 81
PROD. 82
PROD. 82
PROD. 82
PROD. 82
PROD. 82
PROD. 82
A HASTE
A HASTE
A SiaSTE
A HASTE
A HASTE
A HASTE
A HASTE
A HASTE
A HASTE
A HASTE
A HASTE
A HASTE
MST
TFE
TFE
TFE
TFE
TFE
TFE
TFE
TFE
TFE
TFE
TFE
TFE
TFE
TFE
TFE
TFE
IFE
TFE
TFE
T;E
TFE
TFE
TFE
TFE
TFE
TFE
TFE
TFE
TFE
TFE
1
1
1
CHLOROFORM
METHYLENE CHLORID
TOLUEKE
TRICHLOROETHANE
FRCON TF
TOTAL ORGANICS
ISOPROPANOL
FREON TF
TOLUENE
ETHYL 8ENZENE
XYLENES
TOTAL ORGANICS
ISOPROPANOL
FREON TF
TOLUENE
ETHYL BENZENE
XYLENES
TOTAL ORGANICS
ISOPROPANOL
FREON TF
TOLUENE
ETHYL BENZENE
XYLENES
TOTAL ORGANICS
ISOPROPANOL
FREON TF
TOLUENE
ETHYL BENZENE
XYLENES
TOTAL ORGANICS
METHYLENE CHLORIDE
ACETONE
•SOPROPANOL
METHYL ETHYL KETONE
TRiCHLOROETHANE
TRICHLOROETHYLENE
TETRACHLOROETHYLENE
ETHYL BENZENE
TOLUENE
XYLENES
HATER
TOTAL ORGANICS
ACETONE
0
0
0
0
0
0
382000
6000
4000
116000
492000
1000000
429000
5000
3000
104000
457000
993COO
538000
7000
4000
84000
340000
973000
603000
6000
4000
70000
274000
957000
0
39
960
1040
170
0
290
360
86
2000
99.5H
4945
12
0.00004
0.12997
0.00298
0.00097
0. 00092
0.00033
0.11903
0.00328
0.00114
0.00103
0.00004
0.18176
0.00432
0.00129
0.00115
0.00004
0.17444
0.00364
0.00141
0.00127
0.05738
0.01133
0.08288
0.47079
0.09199
0.26222
1.93287
0.16417
517.86
0.17102
0.01
0.03
0.03
0.01
0.24
0.75
38.00
0.58
5.50
22.00
0.69
29.00
0.48
5.80
23.00
1.10
62.00
0.94
S.30
19.00
1.10
J1.00
0.71
4.80
17.00
0.17
0.11
0.53
4.20
3.90
0.61
1.30
4.60
8.10
16.00
25.11
0.10
0.000
1.582
0.036
0.011
0.011
0.000
1.488
0.041
0.014
0.012
0.000
2.371
0.052
0.016
0.014
0.000
2.332
0.04S
0.018
0.017
3.203
0.626
4.586
26.05
S 090
14. 9i
106.9
9.085
0.028
9.432
119.3
85
92.1
133
186.3
60
186.3
92.1
106
106
60
186.3
92.1
106
106
60
185.3
92.1
IDS
106
60
186.3
92.1
106
106
85
58
60
72.1
133
131
165.8
106
92.1
lOo
18
58
0.522794
0.002644
0.003566
0.089860
0.361134
O.S71914
0.002146
0.002605
0.078478
0,344854
0.687223
0.002879
0:003328
0.060735
0.245832
0.751631
0.002108
0.003248
0.049389
0.193322
0
O.OJ0012
0.000289
0.000260
0.000023
0
C. 000031
0.000051
0.000016
0.000340
0.993964
0.000003
0.000041
0.000007
0.00000$
0.00000)
O.OOOC26
0.00025:
0.004185
0.000129
0.001061
0.004259
0.00023S
0.003194
0.000105
0.001122
O.QOMS2
0.00037S
0.006829
0.000239
0.00102S
0.003673
0.000375
0.0056!T
O.OQ01SS
0.000929
0.003291
0.000011
0.000033
0.00018!
0.001195
0.000631
0.000095
0.000150
O.OOOE9SJ
0.001804
0.00309?
0.028627
0.000035
(CONTINUED)
163
g^**fc'^i-,*^»i
-------
TABLE A-1. (CONTINUED)
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PUNT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PUNT
PUNT
PuANT
PLANT
PLANT
PLANT
PLfNT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
SAMPLE
NUMBER LOC.
0
0
0
t>
0
D
D
D
D
D
0
D
0
D
0
0
D
D
0
0
0
D
0
0
0
D
0
D
D
0
0
D
D
D
D
0
0
D
0
D
0
D
D
45.
45
05
05
45.05
45.05
45
OS
45.05
45
45
45
05
05
05
45.05
45
06
45.06
45.06
45
45
45
45
06
06
06
05
45.06
«5
.06
45.07
45.07
45
.07
45.07
45.07
45.07
45.08
45
.08
45.08
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
.03
.08
.08
.08
.08
.08
.08
.09
.09
.09
.09
.09
.09
.09
.09
MST
MST
MST
MST
MST
MST
MST
MST
MST
MST
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
STRIP
MST
MST
MST
MST
MST
MST
MST
MST
MST
MST
.
.
.
.
.
.
.
> BATCH
OR
PROCESS
1 ISO
1 MET
1 TRI
1 TRI
1 TET
1 ETH
1 TOL
1 XYL
1 WAI
1 TOl
3 1 ISC
3 1 MET
3 TR
3 TE1
3 EN
3 TOl
3 XYl
3 HA1
3 1 TOl
1 ME1
1 ETf
1 TOl
PRODUCT
PRODUCT
PRODUCT
PRODUCT
PRODUCT
PRODUCT
XYl
WAI
TO
ME1
AC
IS(
ME
TR
TE
ET
XY
HA
TO
ME
AC
IS
ME
TR
iR
PRODUCT 1 TE
PRODUCT 1 FT
COMPONENT
ISOPROPANOL
1 METHYL ETHYL KETONE
TRICHLOROETHANE
1 7RICHLOROETHYLENE
1 TETRACHLOROETHYLENE
ETHYL BENZENE
TOLUENE
XYLENES
WATER .
TOTAL OR6ANICS
ISOPROFANOL
1 METHYL ETHYL KETONE
TRICHLOROETHANE
TETRACHLOROETHYLENE
ETHYL BENZENE
TOLUENE
XYLENES
WATER
TOTAL ORCANICS
1 METHYL ETHYL KETONE
ETHYL BENZENE
TOLUENE
XYLENES
WATER
TOTAL ORGANICS
METHYLENE CHLORIDE
ACETONE
ISOPROPANOL
METHYL ETHYL KETONE
TRICHLOROETHANE
TETRACHLOROETHYLENE
ETHYL BENZENE
XYLENES
WATER
TOTAL ORGANICS
METHYLENE CHLORIDE
ACETONE
ISOPROPANOL
METHYL ETHYL KETONE
TRICHLOROETHANE
iRiCHLOROETHYLENE
1 TETRACHLOROETHYLENE
1 FTHYL BENZENE
LIQUID
CONC
H
CONST.
(mg/L) A/M*M3 *E-3
6(0
469
99
0
230
560
32
480
99.75*
2513
47
70
33
72
56
17
410
99.93%
705
34
100
42
270
0
0,
0
.00513
.03301
.04975
0.05889
0.16857
3
0
.59132
.72681
516.60
0.04366
0
0
0
0
0
0
.03694
.06219
.02850
.19*22
.14486
.12013
515.67
0
0
0
0
99.961
446
0
350
11000
6600
1100
160
0
25
0
0
0
0
0
0
98.08%
.03984
.12108
.20522
.13681
515.53
.02404
.04664
.03482
.00802
.26935
.29551
525.41
VAPOR
CONC.
(mg/L)
0,
0,
0
0,
0
4
5
17
25
0
0
0
0
0
0
2
25
0
0
0
1
25
0
0
25
It
0
2
0
0
25
,16
,74
.24
.34
.66
.60
.60
.00
.11
.10
.13
.10
.10
.53
.12
.40
.11
.07
.59
.42
.80
.11
.20
.41
.00
.20
.43
.10
.65
.36
.11
MOLECULAR
K
(Y/X)
0.284
1.830
2.758
3.265
9.346
199.1
40.29
0.028
2.424
2.050
3.452
1.582
10.78
8.043
6.669
O.G28
2.212
6.724
11.39
7.597
0.028
1.316
2.554
1.907
0.439
14.75
16.18
0.028
HEIGHT
60
72.1
133
131
165.8
106
92.1
106
18
60
72.1
133
165.8
106
92.1
106
18
72.1
106
92.1
106
18
85
58
60
72.1
133
165.8
106
106
18
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
X
00019-2
000115
000fl13
0
000025
000095
000006
000081
999467
000014
000017
000004
001)007
000009
000003
000069
999873
000008
000016
0.000008
0.
0.
0.
0.
000045
999920
0
.000110
,003346
0.001671
0.
0.
0.
0.
,000150
,000017
0
.000004
,994698
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
Y
000054
000210
000037
000053
000031
000390
001247
003291
028627
OOC034
000035
000015
000012
000102
000026
000464
028627
000018
000114
000093
000343
028627
000048
000145
0.008550
0.
0.
0.
0.
0.
0.
003187
000056
000259
000125
000069
,028«27
1923S
0
0
0
0
0
1
55
2
2
3
5
.77
.45
.80
.00
.50
.CO
.30
.10
85
58
60
72.1
133
131
165.8
106
0
C
0
0
0
0
0
0
4.
0.
0.
0.
,000185
.000159
.000615
,015654
0.009385
0.
0.
0,
.000313
.000408
.000987
164
(CONTINUED)
-------
TA8LEA-1. (CONTINUED)
PLANT
SAMPLE
NUMBER LOC.
BATCH
OR COMPONENT
PROCESS
PLANT
PLANT
PUNT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PUNT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
?LANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
0
D
0
D
D
0
0
D
D
D
D
0
D
D
D
0
D
D
0
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
0
D
0
D
D
D
0
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
.09
.09
.09
.OS
.10
.10
.10
.10
.10
.10
.10
.10
.10
.10
.10
.10
.11
.11
.11
.11
.11
.11
.11
.11
.11
.11
45.11
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
.11
.13
.13
.13
.13
.13
.13
.13
.13
.13
.13
.13
.13
.14
.14
.14
PRODUCT
PRODUCT
PRODUCT
PRODUCT
MST
MST
MST
MST
MST
MST
MST
MST
MST
MST
MST
MST
A HASTE
A WASTE
A WASTE
A WASTE
A WASTE
A WASTE
A WASTE
A WASTE
A WASTE
A HASTE
A WASTE
A WASTE
CONDEN.
CONDEN.
CONOEN.
CONDEN.
CONOEN.
CONOEN.
CONOEN.
CONDEN.
CONOEN.
CONDEN.
CONOEN.
CONDEN.
CONDEN.
CONDEN.
CONDEN.
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
TOLUENE
XYLENES
HATER
TOTAL ORGANICS
METHYLENE CHLORIDE
ACETONE
ISOPROPANOL
METHYL ETHYL KETONE
TRICHLOROETHANE
TRICHLOROETHYLENE
TETRACHLOROETHYLENE
ETHYL BENZENE
TOLUENE
XYLENES
WATER
TOTAL ORGANICS
METHYLENE CHLORIDE
ACETONE
ISOPROPANOL
METHYL ETHYL KETONE
TRICHLOROETHANE
TRICHLOROETHYLENE
TETRACHLOROETHYLENE
ETHYL BENZENE
TOLUENE
XYLENES
WATER
TOTAL ORGANICS
METHYtENE CHLORIDE
ACETONE
ISOPROPANOL
METHYL ETHYL KETONE
TRICHLOROETHANE
TRICHLOROETHYLENE
TETRACHLOROETHYLENE
ETHYL BENZENE
TOLUENE
XYLENES
WATER
TOTAL ORGANICS
METHYLENE CHLORIDE
ACETONE
METHYL ETHYL KETONE
LIQUID
CONC.
VAPOR
H
CONST.
(mg/L) A/M*M3 *E-3
100.00%
0
0
1000
27000
1E600
2100
0
260
0
0
0
0
0
0
3 75
11 31
95.30*
46974
0
6500
95
112
2200
0
55
86
86
0
99.09%
9134
0
32000
380
420
16000
120
1100
3400
2200
17000
92.741
72620
0
130
11
0
0
0
0
0
1
1
0
0
0
0
0
0
0
0
0
0
300000
515.30
.02463
.00418
.03214
.02150
.00789
.24671
.71555
540.70
.02684
.02160
.05863
.17723
.63431
.02609
.81356
520.05
.01924
.0(1540
.05961
.03976
.23942
.02425
.02113
.05504
.01449
555.65
.02526
0.61953
CONC.
(mg/L)
13
16
25
1
1
5
26
2
1
0
5
11
17
25
0
8
0
0
19
1
1
4
7
15
25
0
30
0
1
31
1
1
.00
.00
.11
.80
.20
.50
.00
.20
.30
.10
.20
.00
.00
.11
.63
.50
.10
.32
.00
.90
.70
.30
.60
.00
.11
.83
.00
.10
.22
.00
.40
.30
3.50
5
12
25
0
0
0
.90
.00
.11
.25
.16
.32
0
1
0
1
1
0
MOLECULAR
K
(Y/X)
.028
.321
.224
.724
.153
.423
4037.
1701.
0
1
1
3
9
.029
.480
.192
.235
.780
35.00
56.62
100.0
0
1
0
3
2
.028
.008
.283
.125
.084
12.55
1
1
2
0
0
1
.271
.107
.885
.759
.029
.366
33.51
WEIGHT
92.1
106
18
85
58
60
72.1
133
131
165.8
106
92.1
106
18
85
58
60
72.1
133
131
165.8
106
92.1
106
18
85
58
60
72.1
133
13!
165. 8
106
92.1
106
18
85
58
72.1
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
X
0
0
1
0
000321
008386
004290
000294
0
000029
0
000000
000001
986676
0
002030
000028
000028
000299
0
000006
000014
000016
0
997574
0
010523
OOC120
000111
002294
000017
000126
000611
000455
003058
982679
0
000041
0.000002
0
0
0
0
0
0
0
0
c
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Y
.002896
.003097
.028627
.000434
.000424
.001831
.007400
.000339
.000203
.000012
.001006
.002451
.003291
.028627
.000152
.003007
.000034
.000091
.002931
.000297
.000210
.00083?
.001693
.002904
.023527
.000200
.010614
.000034
.000347
.004783
.00021?
.000160
.000677
.001314
.002323
.028627
.000060
.000056
.000091
(CONTINUED)
165
-------
TABLE A-l. (CONTINUED)
PLANT
SAMPLE
HUM8ER LOC.
BATCH
OR COMPONENT
PROCESS
PUNT 0
PLANT D
PLANT 0
PLANT 0
PLANT 0
PLANT D
PLANT 0
PLANT 0
PLANT D
PLANT D
PLANT D
PLANT 0
PLANT 0
PLANT D
PLANT 0
PLANT 0
PLANT 0
PLANT D
PLANT 0
PLANT D
PLANT 0
PLANT D
PLANT D
PLANT D
PLANT 0
PLANT 0
PLANT D
PLANT 0
PLANT 0
PLANT 0
PLANT 0
?LANT D
PLANT 0
PLANT 0
PLANT 0
PLANT D
PLANT D
PLANT D
PLANT 0
PLANT D
PLANT D
PLANT D
PLANT 0
45. U
45.14
15.14
45.14
45.14
45.14
45.14
45.14
45.15
45.15
45.15
45.15
45.15
45.15
45.15
45.15
45.15
45.16
45.16
45.16
45.16
45.16
45.16
45.16
45.16
45.16
45.17
45.17
45.17
45.17
45.17
45.17
45.17
45.20
45.20
45.20
45.20
45.20
45.20
45.20
45.20
45.20
45.20
CONOEN
CONDEN
CONDEN
CONOEN
CONOEN
CONOEN
CONDEN
CONDEN
STRIP.
STRIP.
STRIP.
STRIP.
STRl".
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
HST
MST
HST
MST
HST
KST
MST
KST
MST
MST
.
.
t
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4 TRICHLOROETHANE
4 TRICHLOROETHYLENE
4 TETRACHLOROETHYLENE
4 ETHYL BENZENE
4 TOLUENE
4 XYLENES
4 WATER
4 TOTAL ORGANICS
4 ACETONE
4 METHYL ETHYL KETONE
4 TRICHLOROETHANE
4 TRICHLOROETHYLENE
4 ETHYL BENZENE
4 TOLUENE
4 XYLENES.
4 WATER
4 TOTAL ORGAN ICS
4 ACETONE
4 METHYL ETHYL KETONE
4 TRICHLOROETHANE
4 TRICHLOROETHYLENE
4 ETHYL BENZENE
4 TOLUENE
« XYLENES
4 WATEP
4 TOTAL ORGANJCS
4 TRICHLOROETHANE
4 TRICHLOROETHYLENE
4 ETHYL BENZENE
4 TOLUENE
4 XYLENES
4 WATER
4 TOTAL ORGANICS
4 METHYLENE CHLORIDE
4 ACETONE
4 ISOPROPANOL
4 METHYL ETHYL KETONE
TRICHLOROETHANE
TRICHLOROETHYLENE
TETRACHLOROETHYLENE
ETHYL BENZENE
TOLUENE
XYLENES
LIQUID
CONC.
H CONST.
VAPOR
CONC.
MOLECULAR
K WEIGHT X
Y
(mg/L) A/M*M3 *E-3 (mg/L) (Y/X)
2900
0
420
26000
530
1600
96.841
31591
270
7
930
0
300
no
900
99.741
2577
32
7
460
0
190
56
310
99.89%
1055
230
0
35
35
120
99.961
420
0
23000
250
400
10000
0
210
69
23
320
0.21229
0.04886
0.00245
0.17037
0.141C9
532.11
0.02508
0.54322
0.00221
0.04 652
0.05674
0.06385
516.63
0.10902
0.72430
0.04015
0.03240
0.05130
0.06620
515.85
0.15168
0.19349
0.05863
0.09919
515.52
0.02409
0.07388
0.05336
0.07798
0.22476
1.30864
7.76260
0.96196
30.00
1.10
1.00
3.10
4.40
11.00
25.11
0.33
0.18
0.10
0.26
0.68
0.47
2.80
25.11
0.17
0.24
0.90
0.57
0.30
0.14
1.00
25.11
1.70
0.28
0.33
0.10
0.58
25.11
1.00
27.00
0.90
1.04
38.00
2.00
2.30
4.40
8.70
15.00
11.48
2.643
0.132
9.215
7.631
0.028
1.390
30.11
0.122
2.578
3.145
3.539
0.028
6.051
40.20
2.228
1.798
2.847
3.674
0.028
8.423
10.74
3.256
5.508
0.028
1.304
3.999
2.888
4.221
12.16
70.84
420.2
52.07
133
131
165.8
106
92.1
106
18
58
72.1
133
131
106
92.1
106
18
58
72.1
133
131
106
92.1
106
18
133
131
106
92.1
106
18
85
58
60
72.1
133
131
165.8
106
92.1
106
O.OS3403
0
O.OS0046
0.024534
0.838106
0.523279
0.934586
0.030083
O.C3QOO!
0.530125
Q
0.830051
O.ES0033
O.EI0153
O.S3955Q
O.E80009
C.C30001
0.030062
0
0.130032
O.C30010
0.6(10052
O.S39830
0.180031
0
0.380005
o.saooo6
O.C30020
O.S39935
0
0.807324
0.300075
Q.SQ0102
OJ531388
0
O.S80023
O.S0QQ12
0.100004
O.EOOOSS
0.004628
0.000)72
0.000123
0.000500
0.000980
0.002129
0.028627
0.000116
0.000051
0.000015
O.GGOC40
0.000131
O.C001C4
0.003542
0.028627
0.000060
Q. 000068
0.000138
0.000089
0.000058
0.000031
0.000193
0.028627
0.000262
0.000043
0.000063
O.CCQ022
0.000112
0.028527
0.000241
0.009553
0.000307
0.000296
0.005863
0.000313
0.000284
0.000851
0.001938
0.002904
166
(CONTINUED)
-------
TABLE A-1. (CONTINUED)
PUNT
SAMPLE
NUMBER LOC.
BATCH
OR COMPONENT
PROCESS
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PUNT
PLANT
PUNT
PLANT
PLANT
PLANT
PUNT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PUNT
PLANT
PUNT
PLANT
PLANT
PLANT
PLANT
PLANT
PUNT
PLANT
PUNT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
0
D
0
D
0
D
0
D
D
D
0
D
0
0
D
D
0
D
0
0
0
D
0
D
D
D
0
D
D
D
D
D
0
0
0
D
0
0
0
0
0
0
D
45
45
45
45
.20
.20
.21
.21
45.21
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
.21
.21
.21
.21
.21
4 1
.41
.21
.21
.22
.22
.22
.22
.22
.22
.22
.22
.24
.24
.24
.24
.24
.25
.25
.25
.25
.25
.25
.26
.26
.26
.26
.26
.27
.27
.27
.27
.27
.27
MST
MST
PRODUCT
PRODUCT
PRODUCT
PRODUCT
PRODUCT
PRODUCT
PRODUCT
PRODUCT
PRODUCT
PRODUCT
PRODUCT
A HASTE
A WASTE
A HASTE
A HASTE
A WASTE
A HASTE
A WASTE
A WASTE
A HASTE
A WASTE
A HASTE
A WASTE
A HASTE
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
MST
MST
MST
MST
MST
MST
3
3
3
3
3
3
4
4
4
4
4
4
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
HATER
TOTAL ORGANICS
METHYLENE CHLORIDE
ACETONE
METHYL ETHYL KETONE
TRICHLOROETHANE
TRICHLOROETHYLENE
TETRACHLOROETHYLENE
ETHYL BENZENE
TOLUENE
XYLENES
WATER
TOTAL ORGANICS
METHYLENE CHLORIDE
ACETONE
METHYL ETHYL KETONE
TRICHLOROETHANE
TETRACHLOROETHYLENE
TOLUENE
HATER
TOTAL ORGANICS
ACETONE
METHYL ETHYL KETONE
TRICHLOROETHANE
HATER
TOTAL ORGANICS
METHYL ETHYL KETONE
TRICHLOROETHANE
ETHYL BENZENE
TOLUENE
HATER
TOTAL ORGANICS
TRICHLOROETHANE
ETHYL BENZENE
TOLUENE
WATER
TOTAL ORGANICS
METHYLENE CHLORIDE
METHYL ETHYL KETONE
TRICHLOROETHANE
TRICHLOROETHYLENE
TETRACHLOROETHYLENE
ETHYL BENZENE
QUIP VAPOR
INC. H CONST. CONC.
ig/L) A/M*M3 *E-3 (asg/L)
96.57* 533.59 25.11 0
34272
0 1.40
0 20.00
0
40000
4000
6000
6000"
0
270000
62.001
380000
0
0
75000
660003
0
0
26.50*
735000
0
0
0
100.00*
0
7
22000
130
44
97.78*
22181
4100
56
23
99.58*
4179
G
4000
560000
0
0
7000
0
0
0
0
0
0
0
.02822
.01539
.01026
.00239
.00198
831.13
.14228
.01368
1944.54
0
0
0
0
0
0
C
0
0
0
515.30
.09657
.00541
.01579
.04664
526.99
.00455
.03665
.08923
517.47
.12826
.01686
.00053
10
55
3
3
7
13
26
25
0
0
520
440
0
0
25
0
4
460
25
0
5
0
0
25
0
0
A
V
25
0
25
460
0
0
0
.GO
.00
.00
.00
.00
.00
.00
.11
.41
.28
.00
.00
.10
.10
.11
.18
.10
.00
.11
.03
.80
.10
.10
.11
.91
.10
.10
.11
.28
.00
.00
.31
.10
.18
1
0
0
0
0
0
2
0
0
0
5
0
0
2
0
0
2
4
0
3
0
0
MOLECULAR
K HEIGHT X
(Y/X)
.028 18 0.991011
85 0
58 0
.070
.583
.389
.090
.074
.031
.948
.283
.040
.028
.262
.294
.860
.541
.028
.252
.026
.939
.028
.595
.472
.014
72.1
133
131
165.8
106
92.1
106
18
85
58
72.1
133
165.8
92.1
18
58
72.1
133
18
72.1
133
106
92.1
18
133
106
92.1
18
85
72.1
133
131
165.8
106
0
0
0
0
0
0
.007930
.000805
.000954
.014925
0
.067163
0.908222
0
0
U
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.050192
.239442
0
0
.710365
0
0
0
1
.000001
.003035
.000022
.000008
.996931
.000556
.000009
.000004
.999429
0
.001979
.150208
0
0
.002355
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Y
.028627
.000333
.007076
022645
.003486
.000469
.000371
.001355
.002896
.005033
.028627
.000098
.000099
.148007
.067891
.000012
.000022
.028627
.000063
.001165
.070977
.028627
.OOOC09
.000894
.000019
.000022
.028627
.000140
.000019
.000022
.028627
.000067
.007115
.070977
.OC0048
.000012
.000034
167
(CONTINUED)
-------
TABLE A-l. (CONTINUED)
PLAKT
SAMPLE
NUMBER LOC.
BATCH
OR COMPONENT
PROCESS
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
0
0
0
D
0
D
D
D
0
D
0
D
0
D
0
D
D
D
D
D
0
D
D
D
D
0
0
D
D
0
D
D
0
D
D
D
0
0
D
D
0
D
D
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
.27
.27
.27
.27
.28
.28
.28
.28
.28
.28
.28
.28
.28
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.29
.31
.31
.31
.31
.31
.31
.31
.31
.32
.32
.32
.32
.33
.33
.33
.33
45.33
45
45
.34
.34
MST
MST
MST
MST
PRODUCT
PRODUCT
PRODUCT
PRODUCT
PRODUCT
PRODUCT
PRODUCT
PRODUCT
PRODUCT
0 HASTE
0 WASTE
0 HASTE
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
0 WASTE 3
0 WASTE
0 WASTE
C WASTE
0 WASTE
0 WASTE
0 WASTE
0 WASTE
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
STRIP.
MST
MST
2
2
2
2
2
2
2
2
3
3
3
3
4
4
4
4
4
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
TOLUENE
XYLENES
WATER
TOTAL ORGANICS
METHYLENE CHLORIDE
ACETONE
METHYL ETHYL KETONE
TRICHLOROETHANE
TRICHLOROETHYLENE
TETRACHLOROETHYLENE
TOLUENE
WATER
TOTAL ORGAMCS
METHYLENE CHLORIDE
ACETONE
ISOPROPANOL
METHYL ETHYL KETONE
TRICHLOROETHANE
TRICHLOROETHYLENE
ETHYL BENZENE
TOLUENE
XYLENES
WATER
TOTAL ORGANICS
ACETONE
METHYL ETHYL KETONE
TRICHLOROETHANE
TRICHLOROETHYLENE
ETHYL BENZENE
XYLENES
WATER
TOTAL ORGANICS
TRICHLOROETHANE
ETHYL BENZENE
WATER
TOTAL ORGANICS
TRICKLOROETHANE
ETHYL BENZENE
XYLENES
WATER
TOTAL ORGANICS
METHYLENE CHLORIDE
ACETONE
QUID
INC.
H
CONST.
ig/L) A/M*M3 *E-
3000
0
42.60%
574000
0
0
37000
770000
0
0
1300
19. m
808300
0
290
37
1600
180000
0
44
6
27
81.301
182004
71
250
71000
0
30
45
92.861
71395
28000
24
97.204
28024
12000
12
0
98.80%
12012
33
1100
0
.00123
1209.63
0
0
0
.20522
.01492
.c:i58
2688.07
0
0
G
0
0
0
0
0
0
0
0
0
0
0
0
0
.01557
.05546
.70544
.05016
.11194
.85508
.62326
629.96
.02890
.08209
.00694
.23258
.35571
554.92
.00344
.24797
530.16
.00359
.17102
VAPOR
CONC.
-3 (m
0
0
25
0
0
370
560
0
0
0
25
0
0
0
55
440
0
0
0
0
25
0
1
24
0
0
0
25
4
0
25
2
0
MOLECULAR
X WEIGHT X
Y
9/L) (Y/XJ
.18
.74
.11
.88
.45
.00
.30
.15
.10
.10
.11
.23
.22
.10
.00
.00
.30
.24
.25
.82
.11
.10
.00
.00
.13
.34
.78
.11
.70
.29
.11
.10
.10
0.
0.
3.
0.
Q.
0.
0.
2.
034
033
481
253
026
045
729
597
33.03
2.
5.
349
241
40.03
29
0.
1.
4.
0.
12
18
0.
0.
13
0.
0.
9.
.18
029
506
279
361
.12
.54
028
186
.44
028
197
402
0.14
0
0
521.57
.16791
.01213
25
0
0
.11
.27
.65
0.
4.
0.
028
755
343
92.1
106
18
85
56
72.1
133
131
165.8
92.1
18
85
53
60
72.1
133
131
106
92.1
106
18
58
72.1
133
131
106
106
13
133
106
18
133
106
106
18
85
58
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.001162
0
.84)294
0
C
.030245
.341224
0
0
.000331
.527697
e
.000106
.000013
.000473
.028902
0
.000008
.000001
.000005
.970488
.000023
.000066
.010240
0
.000005
.000008
.989655
.003883
.000004
.936112
0.001641
0
0
0
0
.000002
0
.998356
.0000(3
.000669
0
0
0
0
0
0
0
0
0
0
0
fl
0
0
0
0
0
0
0
0
0
0
0
.000040
.000143
.028627
.000212
.000159
.105313
.085407
.000023
.000012
,000322
.028627
.000055
.000077
.OG0034
.015554
.057891
.C00046
.000046
.000055
.OS0158
.028627
.000035
.009284
0.003703
0
0
0
0
0
0
0
0
0
0
.000020
.000065
.030151
.028627
.000725
.000056
.023627
.000324
.COC019
.CC0027
0.028627
0.000065
0
.000229
168
(CONTINUED)
-------
TABLE A-1. (CONTINUED)
PLANT
SAMPLE
NUMBER LOC.
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
PLANT
D
0
U
D
0
D
0
D
0
D
D
D
D
D
0
D
0
D
0
0
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
.34
.34
.34
.34
.34
.34
.34
.34
.34
.35
.35
.35
.35
.35
.35
.35
.35
.35
.35
.35
MST
MST
MST
MST
MST
MST
MST
MST
MST
PRODUCT
PRODUCT
PRODUCT
PRODUCT
PRODUCT
PRODUCT
PRODUCT
PRODUCT
PRODUCT
PRODUCT
PRODUCT
BATCH
OR COMPONENT
PROCESS
3 ISOPROPANOL
3 METHYL ETHYL KETONE
3 TRICHLOROETHANE
3 TRICHLOROETHYLENE
3 ETHYL BENZENE
3 TOLUENE
3 XYLENES
3 WATER
3 TOTAL ORGANICS
3 MtTHYLENE CHLORIDE
3 ACETONE
3 METHYL ETHYL KETONE
3 TRICHLOROETHANE
3 TRICHLOROETHYLENE
3 TETRACHLOROETHYLENE
3 ETHYL BENZENE
3 TOLUENE
3 XYLENES
3 WATER
3 TOTAL ORGANICS
LIQUID VAPOR MOLECULAR
CONC. H CONST. CONC. K WEIGHT X
(ng/L) A/M*M3 *E-3 (mg/L) (Y/X)
160 0.01283
6000 0.11529
560000 0.01246
0
35 0.11727
(80 0.01026
67 0.20216
43.2H 1192.49
567375
0
0
14000 0.17590
730000 0.01771
0
6000 0.00034
38000 0.00010
9000 0.00064
0
20.30% 2538. «
797000
0.10 0.363
34.00 3.294
340.00 0.352
0.21
0.20 3.322
0.24 0.290
0.66 5.726
25.11 0.033
0.89
0.49
120.00 3 070
630.00 0.309
0.40
: ; :.oos
0.19 0.001
0.28 0.011
0.51
25.11 0.044
60 0.000091 0.00003)
72.1 0.002937 0.009677
133 0.148630 0.052461
131 0 0.000032
106 0.000011 0.000038
52.1 0.000183 0.000053
106 0.000022 0.000127
18 0.847437 0.028627
85 0 0.0002U
58 0 0.000173
72.1 0.011125 0.034155
133 0.314484 0.097208
131 0 0.000062
165.8 0.002373 0.000012
106 0.020540 0.00003S
92.1 0.005599 0.000062
106 0 0.000098
18 0.646177 0.028627
169
-------
APPENDIX B
SUMMARY OF PROCESS DATA
This section presents a series of tables listing process information
obtained at Plants B and D during the field evaluation procedures. These
tables provide supplemental information to the information provided in
Sections 8 and 9.
170
rfJa*y&«*£wU^^»A«£3^'»tai^i*i^^^ .iS-bi^lSisfaww
-------
Y
TABLE B-l. PROCESS DATA AT PLANT B
Aqueous Methyl Ethyl Ketone Batch
Initial batch charge
Steam flow rate
Process time
Distillation time
30,000 L 8,000 gal.
820-900 kg/hr 1,800-2,000 Ib/hr
15 hr
12 hr
Time
06:30 A
08:30 A
08:45 A
09:00 A
09:30 A
^10:00 A
- 10:30 A
11:00 A
11:30 A
12:00 A
12:30 P
01:00 P
01:30 P
02:00 P
02:30 P
03:00 P
04:00 P
09:30 P
Batch (reboiler)
Temp. °C
18
80
84
87
90
89
90
90
91
92
95
98
99
100
100
100
104
Column head
Temp. °C
65
75
73
72
68
70
72
73
72
74
74
75
90
96
97
100
-
Reflux
GPM
-
-
14.8
14.8
14.8
11.7
1
-
10.1
10.6
8.6
4.7
5.4
1.6
~
Started heatup.
Batch on total reflux.
Batch on total reflux.
Batch on total reflux.
Started distillation.
Continue stripping to meet VOC
target.
Stripping completed.
Approximate. From enthalpy calculations.
-------
TABLE B-2. PROCESS DATA AT PLANT B
Aqueous Acetone Distillation
Initial batch charge
Steam flow rate
11.400L 3,000 gal.
590-640 kg/hr 1,300-1,400 Ib/hr
Process time (excluding 3.5 hr; batch on hold) 8.0 hr
Time
04:30 A
05:00 A
05:30 A
06:00 A
07:00 A
08:00 A
09:CO A
09:30 A
09:35 A
10:00 A
10:10 A
10:25 A
10:30 A
10:40 A
11:00 A
11:15 A
11:45 A
12:10 P
12:?5 P
01:45 P
02:30 P
03:00 P
03:30 P
03:45 P
Batch (reboiler)
Temp. °C
20
26
62
70
70
71
72
72
-
74
_
-
74
.
74
_
75
76
79
92
98
100
102
102
Column head
Temp. °C
_
-
_
57
56
56
58
57
-
57
-
_
58
-
56
_
56
55
55
54
56
59
65
66
Reflux
GPM
_
-
_
-
-
-
7
9
-
9
-
10.7
10.7
11
11
11
11.2
11.2
11.2
11.3
11.5
6
7
7
Product
GPM
Charged, started heatup.
-
_
Column head temperature lines
out. Holding, total reflux.
-
-
_
0
Taking overhead product.
3
6.5
6
6
5
2.5
5
4.5
4.5
4
2.3
6
0 Steam cut buck, product tank
switch.
2
1
1
Approximate. From enthalpy calculations.
-------
TABLE B-3. STEAM STRIPPER PROCESS DATA AT PLANT D
Batch 1 (Aqueous Xylene)
Initial Batch Charge
Final Batch Volume
Organic (Xylene) Distillate
Aqueous Distillate
Steam Flow Rate
Process Time
1,260 L (334 gal)1
1,420 L (375 gal) .
333 L (88 gal}
248 L (75 gal)
250-270 kg/hr (550-600 lb/hr)'
2.08 hr
Stripper Vapor Distillate Steam Pressure
Time Temp. °C (°F) Rate, L (gpm) kPa (psig)
Comments
04-15 p Steam started to sparger
04^23 p 71 (160) 250 (36)
04:35 p 68 (155) 264 (39)
04:45 p 63 (146) 310 (45)
04:48 p 68 (155
04:50 p 92 (197
04:55 p 93 (200
05:00 p
283 (41) Distillate
9.5 (2.50)
11.4 (3.01)
started over
05:45 p 97 (207) 303 (44)
05:58 p 99 (211) Final vapor temperature
reached;
stripping con-
tinued to reduce trace
VOC
06:15 p 99 (211)
06:20 p
Completed
stripping
Measured volume in MST prior to transfer to stripper. Quantity of organic
distillate indicates that approximately 200 L (60 gallons) of crude xylene may
have been inadvertently charged to stripper.
2
Calculated from heat of vaporization.
Determined from incremental time-volume measurement.
173
-------
r
TABLE B-4. PLANT D CONDENSER VENT FLOW READINGS
Batch 1 (Aqueous Xylene)
f
1
Time
(p.m.)
4:00
4:03
4:10:29
4:12:42
4:16:13
4:20:30
4:24:30
! 4:25:24
[ 4:26:09
| 4:28:40
! 4:33:55
4:48:00
[ i 4:49:10
F
E-
f
1
1
4:51:00
4:52:00
4:54:00
4:56:00
1 5:01:00
5:21:00**
5:25:00
5:38:00
5:42:00
? 5:48:00
*Dry gas meter.
**Samples taken
Reading*
(cu ft)
38.058
38.102
40.90
40.91
40.92
40.93
40.94
40.95
41.01
41.1
41.80
47.4
50.00
56.0
57.3
57.93
58.37
59.00
5S.66
60.30
61.0
61.5
62.25
at 5:10,
Rate
(L/min)
2.6
10.5
0.11
0.08
0.05
.071
.28
3.4
10.1
4.0
10.9
61 .
84.9
36.8
12.7
6.2
3.56
4.53
1.5
3.5
3.5
5:14, 5:16,
Rate
(cu ft/min)
.091
.370
.004
.0028
.0023
.0025
.01
.12
,36
.14
.386
2.16
3.0
1.3
.45
.22
.126
.16
.054
.125
.125
p.m. of vent gas.
Comment
Start filling
Stripper
Finish filling
Start steam
Batch at temperature
174
-------
IT"-
TABLE B-3. STEAM STRIPPER PROCESS DATA AT PLANT D
Batch 2 (1,1,1-Trichloroethane/Qil)
Initial Batch Charge 895 L (236.5 gal)
Final Batch Volume 323 L (85.4 gal)
Organic (1,1,1-Trichloroethane) Distillate 670 L (177 gal)
Aqueous Distillate 397 L (105 gal)
Steam Flow Rate 250-295 kg/hr (550-650 Ib/hr)1
Process Time 1.72 hr
2
Stripper Vapor Distillate Steam Pressure
Time Temp °F Rate, L (gpm) kPa (psig)
1249 p
0103 p
0108 p
0115 p
0124 p
0135 p
0145 p
0151 p
0155 p
0213 p
0232 p
149
150
152
157
205
208
210
211
211
262 (38)
234 (34)
21 (5.61)
269 (39)
(34)
5.1 (1.36)
262 (38)
Comments
Steani on to sparger
Final vapor temperature
reached; stripping con-
tinued to reduce trace
VOC
Stripping completed
1
Calculated from enthalpy of vaporization.
7
Determined from incremental time-volume measurement.
175
-------
TABLE B-6. PLANT D CONDENSER VENT FLOW READINGS
Batch 2 (1,1,1-Trichloroethane)
Time
(p.m.)
1:13
1:17
1:18:47
1:20:03
1:21:04
1:23:06
1:25
1:36
1:46+
1:56
1:57:04
1:57:38
1:58:11
2:05
2:06
2:06:38
2:07:07
2:07:30
2:08:25
Reading
(cti ft)
62.622
62.83
62.93
63.03
63.11
61.70
63.10
64.4
64.8
66.1
66.3
66.4
66.405
66.9
67.1
67.2
67.3
67.4
67.5
Receiver
L/i.iin
1.5
1.6
2.5
2.3
-20
+20
3.4
3.7
5.7
11.3
.28
2.0
5.7
5.7
5.7
5.7
2.8
Vent Rate
cu ft/mi n Comment
10.052
0.057
0.087
0.08 Negative flow start
-.70
+.70
0.12
Samples taken
0.13
0.2
0.4
0.01
0.07
0.20
0.2
0.2
0.2
0.1
*Sampling of vent gas ty-j,
-------
r-
TABLE B-7. STEAM STRIPPER PROCESS DATA AT PLANT D
Batch 3 (Aqueous 1,1,1-Trichloroethane)
Initial Batch Charge
Final Batch Volume
Organic Distillate
Aqueous Distillate
Steam Flow Rate
Process Time
564 L
545 L
45 L
179 L
(149 gal)
(144 gal)
(12 gal)
(47 gal)
272 kg/hr (600 Ib/hr)
0.93 hr
1
Stripper Vapor
Time Temp. °C (°F)
Steam Pressure
kPa (psig)
Comments
0340
0345
0402
0414
0428
0436
68 (155)
97 (206)
98 (208)
99 (210)
99 (211)
262 (38)
262 (38)
276 (40)
Steam on; started stripping
Steam off; stripping completed
177
-------
TABLE B-8. PROCESS DATA AT PLANT D
Batch 4 (Aqueous Mixed Solvent)
Initial Batch Charge
Aqueous and Miscible Solvent Distillate
Immiscible Distillate
Steam Flow Rate
Process Time
360 L (95 gal)
140 L (37.1 gal)
3.2 L (0.85 gal)
227-250 kg/hr (500-550 lb/hr)3
0.83 hr
Time
0525 p
0539 p
0541 p
0615 p
Stripper Vapor
Temp °F
157
205
211
Steam
kPa
241
262
Pressure
(psig)
(35)
(38)
Comments
Started steam to sparger
Stripping completed
Approximate. From enthalpy calculations.
178
-------
TABLE B-9. SAMPLING TIMES FOR BATCH 1 (AQUEOUS XYLENE) AT PLANT D
Sample No. Sample Time Comments
2500-45-4 Waste Feed Stream 0411 p From stripper, prior to start of stripping
2500-45-3 Waste Feed Stream 0412 p From stripper, prior to start of stripping
2500-45-5 Stripper Contents 0505 p Approximately one-third through processing
2500-45-10 Aqueous Phase Distillate 0525 p Mid process aqueous phase distillate
2500-45-6 Stripper Contents 0558 p Approximately two-thirds through processing
2500-45-7 • Stripper Contents 0620 p Completion of stripping
2500-45-8 Aqueous Phase Distillate 0645 p : Batch aqueous distillate drummed off
2500-45-9 Organic Distillate 0650 p Xylene product accumulated for batch,
drummed off.
-------
1ABLF; B-10. SAMPLING TIMES FOR BATCH 2 (1,1,1-TRICHLOROETHANE PRODUCTION) AT PLANT D
Sample No.
2500-45-22
2500-45-23
2500-45-24
2500-45-25
2500-45-26
2500-45-27
2500-45-28
Sample
Waste Feed Stream
Waste Feed Stream
Stripper Contents
Stripper Contents
Stripper Contents
Miscible Solvent Tank
Product Storage Tank
Time
1248 p
1248 p
0128 p
0155 p
0232 p
0310 p
0305 p
Comments
Initial from stripper, prior to start of
stripping
Initial from stripper, prior to start of
stripping
Approximately one-third through process
Approximately two-thirds through process
Completion of stripping
Aqueous phase distillate accumulated for batch
Organic product for batch
00
o
-------
TABLE B-ll. SAMPLING TIMES FOR BATCH 3 (AQUEOUS 1,1,1-TRICHLOROETHANE) AT PLANT D
Sample Ho.
2500-45-29
2500-45-30
2500-45-31
2500-45-32
2500-45-33
2500-45-34
2500-45-35
Sample
Waste Feed Stream
Waste Feed Stream
Stripper Contents
Stripper Contents
Stripper Contents
Aqueous Phase Distillate
Organic Product
Time
0340 p
0340 p
0402 p
0423 p
0437 p
0440 p
0441 p
Comments
Initial from stripper, prior to
stripping
Initial from stripper, prior to
stripping
Approximately one-third through
Approximately two-thirds through
Completion of stripping
Batch accumulation in decanter
Batch accumulation in decanter
start of
start of
process
process
O3
-------
TABLE B-12. SAMPLING TIMES FOR BATCH 4 (AQUEOUS MIXED SOLVENT) AT PLANT D
CO
Sample No.
2500-45-11
2500-45-12
2500-45-13
2500-45-18
2500-45-15
2500-45-16
2500-45-14
2500-45-19
2500-45-17
2500-45-20
2500-45-21
2500-45-46
Sample
Waste Feed Stream
Waste Feed Stream
Distillate
Distillate
Stripper Contents
Stripper Contents
Distillate
Distillate
Stripper Contents
Aqueous Distillate
Organic Distillate
Plant Boiler Feed Water
Time
0523 p
0525 p
0548 p
0548 p
0552 p
0601 p
0612 p
0612 p
0615 p
0630 p
0620 p
Comments
Initial from stripper, prior to start of
stripping
Initial from stripper, prior to start of
stripping
Condenser discharge
Condenser discharge
Approximately one-third through process
Approximately two-thirds through process
Condenser discharge
Condenser discharge
Completion of stripping, final sample
From drum; distillate drummed from receiver
From drum; distillate drummed from receiver
Representative of live steam injected in
stripper
j
-------
TABLE B-13. GAS SAMPLES DURING STEAM STRIPPING AT PLANT D
Container Satrole
number number
A88
A181
A160
A175
A100
A79
A1C4
A84
A154
A173
2500-46-3
2500-46-2
2500-46-5
2500-46-1
2500-49-2
2500-49-3
2500-49-4
2500-49-5
2500-49-7
2500-49-1
Time (p.m
5:10
5:15
5:16
-
12:58
1:46
1:47
1:43
1:15
1:42
. ) Date
12/4/84
12/4/84
12/4/84
-
12/5/84
12/5/84
12/5/84
12/5/84
12/5/84
12/5/84
Batch being
Description processed
Product receiver
Product receiver
Product receiver
Field blank
Product receiver
Product receiver
Product receiver
Product storage
Ambient air
MST tank
1
1
1
2
2
2
2
2
183
-------
I i
!;,
t
I'1
APPENDIX C. ANALYTICAL PROCEDURES
The analyses that were carried out on the process samples are outlined
below.
ONSITE ANALYSES AND MEASUREMENTS
The onsite analyses were limited to temperature measurement, of the liquid
samples. Temperature was measured immediately following sampling at each
sample point. After collection of the 40-mL volatile organic analysis (VGA)
bottle and 1-L samples, a glass jar was partially filled with a sample. The
temperature was then measured using a mercury thermometer. The vent gas
velocity was measured onsite using an Alnor velometer.
OFFSITE ANALYSES
Analysis of Vent Gas Samples
The evacuated stainless steel canister used to sample the vents was
analyzed for volatile organics using the headspace GC method.
Analysis of Liquid Samples
Volatile Organics—
The volatile organic compounds in the liquid samples collected from the
VOC removal processes were identified and quantified by both direct injection
GC and headspace GC. Confirmation of GC peak identification was carried out
by GC/MS on selected satr.ples.
Determination of pH of Liquid Waste Samples—
The pH was measured using a pH electrode as described in EPA Method
150.1.
Determination of Solids Content of Liquid Waste Samples—
To characterize the waste material to determine the potential suitability
of certain individual treatment techniques subject to solids limitations such
as distillation and extraction, the total suspended matter and volatile and
fixed matter in the wast° material were determined. Standard Methods (Methods
209 C, D, E, and G) were used where applicable.
Determination of Density of Liquid Waste Samples —
A class A volumetric flask was filled and the weight of the liquid
determined using an analytical balance.
184
-------
APPENDIX D
QUALITY ASSURANCE
This appendix piesents representative results of quality assurance date .
for the analysi: of samples collected at Plants D and B. The relevant quality
assurance for the grab samples taken on site visits (field blanks, duplicates)
are presented with the data in the field test reports (Allen, 1985d, 1985g).
The general QA/QC project plan is presented in "Quality Assurance Plan:
Hazardous Waste Pretreatment for Emissions Control," P.TI, 1984, and the
site-specific plans are described in "Site Specific Test and QA Project Plan
Addendum Hazardous Waste Pretreatment for Emissions Control: Field Evaluation
Plant D," RTI, 1984, and "Site-Specific Test and QA Project Plan Addendum
Hazardous Waste Pretreatment for Emissions Control: Field Evaluation
Plant B," RTI, 1984.
The analytical work was performed by Industrial Environmental Analysts,
Inc. (IEA).
PRECISION, ACCURACY, AND COMPLETENESS
Clearly defined objectives for precision, accuracy, and completeness of
the measurement data are necessary to ensure the generation of reliable data
of known quality. This project has the analyzed waste samples for a variety
of volatile organic compounds (VOC) from several pretreatment sites. Although
a list of the specific VOCs expected could not be compiled before sampling
because the types of wastes treated in the processes vary from day to day, the
methodology that was selected to carry out the sampling and analyses was
sufficient to characterize the processes to meet the program objectives.
The program's data quality objectives are given in Table D-l.
INTERNAL QUALITY CONTROL CHECKS
RC and GC/MS Analyses
A standard mixture of VOC in nitrogen is prepared in a large glass
container. A sample of the standard gaseous mixture is withdrawn periodically
to serve as the control for the GC headspace analyses. The gas sample from
the standard is analyzed in an identical manner to the gas sample withdrawn
from the headspace (see Page 190).
SL
^: Liquid samples are analyzed by thermocouple detectors with gas
chromatography for samples with high VOC concentrations (>0.1 percent) and by
flame ionization detectors with gas chromatography for aqueous liquid samples
185
-------
CO
en
TABLE D-l. PRECISION, ACCURACY, rt,',J JMPLETENESS OBJECTIVES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
Parameter
Waste material flow rate
Wind velocity
Source dimensions
Gas temperature
Liquid temperature
Waste volume
Gas flow rate
pH of liquid
Liquid density
Solids content of liquids
Water content of liquids
Volatile organic compounds
in vent gas
Volatile organic compounds
in liquid samples
Units
g/sec
cm/sec
cm
°C
°C
m3
m3/sec
PH
g/m3
percent
percent
g/m3
mg/L
Method
mass flowmeter
or calibrated
container
velometer
ruler
thermometer
thermometer
dip stick
calculated
EPA 150.1
gravimetric
209 C, D. E,c
H as applicable
ASTM, 0-1744
evacuated canister/
GC-FIO
GC-FID headspace
GC-TCD direct analysis
GC-MS
Precision
% RSO
10%
30
5
5
5
5
10
±.1 units
1%
10
20
25
25
25
qual.
Accuracy
(% Bias)3
10%
20
3
1
1
2
10
± .1 units
1%
10
10
25
25
25
qual.
Completeness
100%
95%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
ercent bias
observ^p;c^Pected
x 100%
the number of valid data points divided by the number of planned data points expressed as a percentage.
""Standard Methods for the Examination of Water and Wastewater, 15th ed.
-------
with lower concentrations of VOCs. The liquid sample was injected directly
into the chromatograph.
GC/MS was used in a qualitative mode to verify the components identified
by retention time by the gas chromatography. All reported compounds were
verified by GC/MS.
SPECIFIC ROUTINE PROCEDURES USED TO ASSESS DATA PRECISION, ACCURACY, AND
COMPLETENESS
For aach major measurement parameter, the completeness, precision, and
accuracy of the measurement data were evaluated. Completeness is a measure of
the number of acceptable samples or data points actually obtained, divided by
the number which were planned. Ways in which a sample can become "incomplete"
or voided include not collecting the sample, sampling incorrectly, losing or
breaking the sample in shipment, improper sample preservation, consuming the
whole sample in a voided analysis, or outlier data point rejection. The
completeness of the field tests was 100 percent.
Accuracy of volatile organic measurements is assessed on the basis of
percent bias of analyses of performance evaluation samples. Percent bias is
the difference between a measured value and the true value when the latter is
known or assumed. Percent recovery describes either recovery of a known
amount of analyte (spike) added to sample of known value, or recovery of an
analyte of known value from a synthetic or environmental standard.
% Recovery = 100 x observed Va1ue
(standard)
Both percent bias end percent recovery were used to assess the accuracy of GC
determinations.
Precision of volatile organic analyses was assessed on the basis of
relative standard deviation of analyses of triplicate samples collected from
each process. Relative standard deviation is calculated as follows:
% RSD = —~— x 100%
X
where S = standard deviation, and
"X = mean of 3 or more measurements.
If only two samples are available for analysis, precision was calculated as
relative percent difference as shown below:
187
-------
X1 - x?
% RPD = 2x / / x 100%
Al A2
where X, = first result, and
X~ = second result.
Concentration values below the limit of detection of the methods employed
result in the problem of determining how to report such values. This problem,
when it occurred, was handled by reporting that the analysis was below the
minimum detection limit and stating the minimum detection limit, such as <10
mg/L.
A measurement result will be considered an outlier whenever there is an
obvious and documentable failure in the operation of the equipment or an
obvious and documentable error committed by the analyst or operator. The
measurement of the initial xylene concentration in the stripper as being much
lower than the amount added to the waste water is an example of an obvious
mixing error (or solubility problem). In addition, potential outliers in each
data set can be tested using the method presented by F. E. Grubbs,
Technometrics, 11(1), 1 (1969). The 95-percent level of confidence is used as
the basis for acceptance or rejection.
PERFORMANCE AND SYSTEM AUDITS
Systems Audits
The RTI Quality Assurance Officer (RTI/QAO) performed a complete systems
audit of the laboratories prior to field sampling. Additional audits were
performed by S-Cubed, and the U.S. EPA.
Performance Audits
The RTI Quality Assurance Officer provided IEA with liquid samples tc be
analyzed as unknowns (the contents are described in Tables D-2 and D-3). Two
of these samples were analyzed during the course of the project. The Quality
Assurance Officer determined that acceptable levels of accuracy and precision
were achieved based on program data quality objectives. The data quality
objectives are reported in Table D-l, and the results of the audits are
described below.
Results of Audit
Prior to the field sampling at Alternate Energy Resources, an audit
sample was provided to IEA by the RTI/QAO. The sample consisting of several
aromatic purgeable organics (xylene, toluene, benzene) at '-'1-percent
concentrations was analyzed neat by 6C/FID. The results were all within ±10
percent of the expected values. These result^ are presented in Table D-2. An
additional audit sample was submitted for analysis containing a variety of
chlorinated hydrocarbons. With the exception of methylene chloride (solvent
188
-------
f.
I'
I
TABLE D-2. FIRST ACCURACY
DETERMINATION OF
VOC ANALYSIS
Concentration (ms/L^
Compound
Benzene
Toluene
Expected
614
82
Ethyl benzene 230
p-xylene
382
o- and m-xylene 1,064
IEA
589
77.9
212
351
984
%
Bias
-4
-5
-8
-8
-7
TABLE D-3. SECOND ACCURACY
DETERMINATION OF
VOC ANALYSIS
Parameter
number
1
2
3
4
5
6
7
8
9
10
Parameter measured
Methyl ene chloride
1,1-Dichloroethene
Trans-1 ,2-dichloroethane
1,2-Dichloroethane
Carbon tetrachloride
1,2-Dichloropropane
Trichloroethene
Di bromochl oromet.hane
1 » 1 r2 ,2-Tetrachl orcethane
Chlorobenzene
EMSL certified
values
(mg/L)
200
104
276
125
133
200
248
148
253
202
IEA
values
(mg/L)
300
110
290
140
140
180
?40
130
270
200
Percent
bias
401
5.6
4.9
11.3
5.1
-10.5
-3.3
-12.9
6.5
-1.0
The presence of methanol interfered with the analysis.
189
-------
methanol interference) the concentrations which were determined were within
the QA goals (25 percent). The results of the analysis of second audit
samples is presented in Taole D-3. The accuracy of th« GC was checked
regularly during the course of the analysis by injection of a gas reference
sample containing toluene (100 ppm). An analysis of the results of 17 of
these accuracy checks indicated a percent bias of +5.18 percent, with a
standard deviation of 3.09 percent. These results are substantially better
than the project goals of 25 percent.
Duplicate Analyses
Duplicate analyses were carried out regularly on headspace samples,
liquid samples, and the gas samples collected in steel containers. In an
analysis of 28 compounds from samples anal.V7ed in duplicate, the average
percent relative difference was 10.6 percent. The standard deviation on the
distribution of these percent relative differences was 0.19, indicating
that some of these duplicate analyses were less reproducible than others. The
percent relative differences (Table D-l) for three samples were larger than
the QA objectives (two samples of ethyl benzene and cne of tetrachloroethene)
near the detectability limit (3 x level of detection).
In the duplicate analyses of liquid samples (Table D-4), the precision
was poorer than observed for the audit samples; however, this precision was
still generally better than the quality goals of the project (25 percent) for
the higher concentrations of interest (>1,000 ppm). The reproducibility (in
terms of percent relative difference) is poorer for the components present at
the lower concentrations.
In the duplicate analyses of gas samples taken in bteel containers
(Table D-5), the average percent relative difference was 23.8 percent, within
the quality ycals of the project (25 percent). Of the 17 comparisons, the
percent relative difference was greater than 100 percent for one compound, in
the range 15 to 60 percent on 6 compounds, and less than 15 percent on the
remaining 10 samples of the distribution.
Linear Regressions
Linear regressions are used to evaluate the rate of stripping. The
relationship between the logarithm of the concentration and the stripping time
is linear for some of the VOC removal data. It is useful to estimate the
error associated with the calculation of the slope. The least squares method
is used to obtain the slope that minimizes the sum of the squares of the ;
residuals. . .
The slope m and the intercept B are calculated by the following
equation.
M = nIXY - (£X) (£Y)
n(EX2) - (IX)2
190
-------
TABLE D-4. A COMPARISON OF RELATIVE DIFFERENCE
FROM THE ANALYSIS OF LIQUID SAMPLES
Sample
number
45-10
45-31
45-10
45-10
45-10
45-10
45-31
45-17
45-17
45-31
45-31
45-17
45-31
45-10
45-31
45-10
45-31
Compound Analysis 1
chloroform
1,1,1-trichloroethane
isopropanol
1 , 1 , 1-tri chl oroethane
acetone
tetrachl oroethane
chloroform
1 , 1 , 1-tri chl oroethane
xylenes
acetone
xylenes
toluene
ethyl benzene
xylenes
isopropanol
toluene
toluene
83,000
71,000
27,000
2,100
1,000
260
250
230
120
71
45
35
30
11
6
3
3
Analysis" 2
86,000
100,000
28,000
2,200
1,100
250
34
230
4
110
4
3
63
4
14
48
8
Relative
difference
0.035502
0.3391B1
0.036363
, 0.046511
0.095238
0.039215
1.521126
0.000000
1.870967
0.430939
1.673469
1.684210
0.709677
0.933333
0.800000
1.764705
0.909090
Total number of comparisons
Average
Standard
relative difference
deviation
17
0.
0.
758
6838
191
Jiaa
-------
TABLE D-5. A COMPARISON OF RELATIVE DIFFERENCE FROM THE ANALYSIS
OF GAS SAMPLES TAKEN IN STEEL CONTAINERS
Sample
number
38
38
36
36
36
36
36
36
204
203
203
204
203
204
204
204
203
Concentration (mg/L)
Compound
dibromoethane
Analysis 1
550
1,1,1-trichloroethane 530
trichloroethene
benzene
tetrachloroethene
methyl ethyl ketone
isopropanol
toluene
methyl ethyl ketone
1,1-dichloroethane
methyl ethyl ketone
methanol
acetone
isopropanol
1 , 1-dl chl oroethane
acetone
carbon tetrachloride
350
190
170
61
52
22
6.5
4.9
1.2
0.98
0.8
0.58
0.49
0.16
0.16
Analysis 2
300
330
310
210
140
14
50
13
6.5
5.3
1.3
1.2
0.8
0.62
0.52
0.19
0.18
Relative
difference
0.588235
0.465116
0.121212
0.1000CO
0.193548
1.253333
0.039215
0.514285
0 000000
0.078431
0.080000
0.201834
0.000000
0.066666
0.059405
0.171428
0.117647
Number of
Average
Standard
comparisions
deviation
17
0.238
0.3069
192
-------
., --N
where X is the stripping time, and Y is the natural logarithm of
concentration, and n is the number of observations.
Sxx = EX2 - (£X)2/n
Syy = EY2 - (EY)2/n
Sxy = £XY - (zX)(i:Y)/n
2
These standard error estimates can be used to calculate S and r, which
in turn can be used to calculate the confidence interval.
S2 = (Syy - M Sxy)/(n-2)
r = Sxy/(Sxx Syy)0'5
The confidence interval on M may be written as follows.
H ± (ta/2S/(Sxx)0'5
The students' t-distribution has n-2 degrees of freedom; the value depends
upon the degrees cf freedom.
The enalysis of confidence limits for the calculated slopes are presented
in Tables D-6 and D-7. In some cases the confidence limits could apparently
have been substantially reduced by increasing the Dumber of data points
because of the sensitivity of t to the number of data points; in other cases,
the relationship between the logarithm of concentration and time was not
linear. This lack of linearity was particularly true for high concentrations
of VOCs and for VOCs which were refluxed.
Other Procedures
The following procedures were carried out to ensure quality of the field
evaluations:
o Both field and trip blanks were analyzed for the vent gas samples.
o One field blank was generated for the liquid samples collected at
each site.
o One field replicate was obtained and analyzed for the vent gas
sampling from each batch.
o All of the waste liquid samples were collected in duplicate.
193
;,:.... -. v, _._,, ,.«•„.. _u^1_J..,.m ..-.. ~_
•-—••mini ^^^^^^^^M^*^g*^**wlWl'alaaff''ffi'"^^^&fflSS3iS^5fe
-------
TABLE D-6. CALCULATED CONFIDENCE INTERVALS FOR STRIPPING RATES
AT PLANT D
Batch
First Batch:
aqueous xylene
Second Batch:
production batch,
1,1,1-trichloroethane
Third Batch:
1,1,1-trichloroethane,
HST
Fourth Batch:
mixed solvent.
4
Compound
acetone
isopropanol
methyl ethyl ketone
1,1,1-trichloroethane
xylene
tetrachloroethene
1,1,1-trichloroethane
methyl ethyl ketone
acetone
1,1,1-trichloroethane
methyl ethyl ketone
ethyl benzene
acetone
toluene
xylene
1,1,1-trichloroethane
Slope
0.0246
0.058
0.0393
0.0242
0.0180
0.0296
0.0648
0.176
*
0.0721
0.0468,
0.08951
0.0208
0.280
0.127
0.110
0.0684
Confidence
interval
(90%)
±0.096
±0.0191
±0.0056
±0.0038
±0.0244
±0.0140
±0.611
±0.096
±0.022
±0.0073
±0.0193
±0.0140
±0.147
±0.120
±0.050
±0.019
n
3
4
4
'4
4
4
4
3
4
4
3
4
3
4
4
4
The first value less than detection limit included as the detection limit to
reduce estimate of confidence interval.
194
-------
ft J
TABLE D-7. CALCULATED CONFIDENCE INTERVALS FOR STRIPPING RATES
AT PLANT B
Batch
First Batch:
acetone
Second Batch:
MEK
Compound
acetone
isopropanol
methyl ethyl ketone
1 , 1 , 1-tri ch'l oroethane
toluene
ethyl benzene
xylenes
2,2-DMO
isopropanol
methyl ene chloride
methyl ethyl ketone
1 , 1 , 1-tri chl oroethane
methanol
trichloroethylene
benzene
Slope
0.0171
0.0099
0.0172
0.0169
0.0166
0.0092
0.00937
0.0188
0.0095
0.01889
0.01718
0.000232
0.00766
0.00526
0.00366
Confidence
interval
(90%)
±0.0097
±0.00796
±0.0141
±0.0086
±0.0079
±0.0093
±0.0076
±0.0368
±0.0099
±0.042
±0.0635
±0.00192
=0.0088
±0.00432
±0.00496
N
4
4
4
4
4
3
4
3
4
3
3
4
4
4
4
I, I
I
195
1»rV.
-------
S"
I:
t •
f. •
F
The GC/MS was calibrated with terfluorotributylamine. A standard series
of 12 components were injected for a standard library search of compounds.
These standards were verified each day that the GC/MS was used for liquid
samples.
196
-------
APPENDIX E
COST EFFECTIVENESS ESTIMATION METHODOLOGY
This appendix presents details of several components of the
cost-effectiveness estimation methodology which was used (1) to derive unit
treatment costs for the batches as processed in the field (Table 46) and (2)
for analysis of how degree of waste treatment influenced unit treatment costs
(Tables 47-50). The approach described below for handling condensato
treatment was us-?d in both these analyses. The other analytical details which
are presented in this appendix support tha analyses in Tables 47-50.
CONDENSATE TREATMENT
As the batch of waste ii treated in the steam stripper, the condensed
steam is collected in the MST; when the tank is full enough to constitute a
batch, it is treated in the stripper. The stripping time is estimated for the
treatment of this residual and included in the batch process time used In the
cost estimates.
The amount of condensed steam is estimated as the product of the steam
rate and the time that the steam is sparging. It is assumed that this
condensate is treated for three time constants (reciprocal of the rate
I ! constant K) with the same ratio of steam to batch size as for the waste
originally treated. The condensed steam is combined with condensed steam from
other batches to form a 1,200 L batch for stripping. In addition to the
condensate from the waste being treated, the condensate from the treatment of
each condensate must also be treated. The dimensionless rate constant Is
assumed to be the same as for the original waste. The fraction of the
condensate, F, that is generated from repeated condensate treatments that can
be ascribed to an initial batch is theoretically an infinite series.
F = l + f + f2 + f3 + . . . . fn, n ->»,
I I f steam used to treat a batch (L)
| . batch size [IT
sf-
_ 3K (steam rate)
i= | ~ batch size
|" I The series solution of the fraction of steam treated is evaluated for the
I,! I first eight terms. As f approaches 1, the series will become infinitely
f I larg: . The waste treatment system may fail under these conditions; all that
is accomplished by the treatment is that the volume of wastewater increases by
condensed steam dilution.
197
>. .- *
-------
IP"
I
i;
The condensate treatment time is determined by multiplying the batch
cycle time (three time constants plus 40 min) by the ratio of the wastewater
volume (F x condensate) to the batch size (1,200 L).
BATCH CYCLE TIME AND BATCH PROCESS TIME
The batch cycle time is calculated from the first order decay equation
using the average stripping rate constant, K, plus a batch heating and waste
transfer time of 40 minutes. The stripping rate constants were determined
from the logarithm of the ratio of the fielti values of the final VOC
concentrations to the initial VOC concentrations, divided by the stripping
times (as found in Table 27). Taking C(o) as the initial total VOC
concentration for the batch and C(y) as the concentration of VOCs in the batch
after y percent of the volatiles had been removed, then the batch cycle lime
required to obtain this percentage of removal, t(y), is:
t(y) = [K"1 ln(C(0)/C(y})3 + 40.
The batch process time is the sum of the times it takes to treat both the
v/aste, t(y), and the condensate generated during the waste treatment (derived
above).
OTHER PARAMETERS
Other parameters used in the cost analyses of Tables 47-50 are shown in
Table E-l. The waste volume, rate constant for stripping, and the steam rate
were assumed to be the same as for the batch evaluated during the field test.
The maximum batch size for aqueous condensate treatment is assumed to be a
typical batch size, 1,200 L. The product density and molecular weight is
estimated on the basis of the composition of the major VOCs present.
Assumed costs for waste disposal at a landfill or at a publicly operated
municipally treatment works (POTK) are provided, along with the assumed cost
of steam.
The waste residual volatility (torr) is calculated on the basis of the
sum of the concentrations (mg/L) of the VOCs measured in the headspace of the
waste. This concentration is converted to a vapor pressure in the
hypothetical case by dividing the ratio of this concentration of the VCCs in
the vapor in the laboratory at 25°C to the concentration in the liquid in the
laboratory and multiplying by the concentration in the hypothetical case, and
multiplying by a factor of 0.040^ moles/L divided by the assumed molecular
weight M (mg/mol), and multiplying by the vapor pressure of one atmosphere
(760 torr).
198
-------
f " ^1221!!^
V
! TABLE
Waste volume
The rate constant for stripping
The initial concentration
The steam rate
The dimensionless rate constant
I
Fraction VOC removal from
condensate
Maximum condensate treatment
to batch size
VO
Condensate treatment fraction
The product density
Average VOC molecular weight
Waste disposal, landfill
Waste disposal, POTW
Cost of steam
ff^^^^,w^^
E-l. PARAMETERS USED
Units
-1
, K min
percent volatiles
L/min
L-min./L/min.
L
F
g/cc
$/L
$/L
$/L
vmff**
IN THE
Batch 1
1,260
0.073
26
4.33
21.23
95
1,200
0.1646
0.86
106
0.37
0.00005
0.0183
^F^TO^
COST ANALYSIS
Batch 2
897
0.06
74
4.53
11.87
95
1,200
1.3381
1.33
133
O.i7
0.00005
0.0183
Batch 3
564
0.048
18
4.53
5.97
95
1,200
2.0013
1.33
133
0.37
0.00005
0.0183
Batch 4
360
0.132
3.1
4.00
11.88
95
1,200
1.3378
0.95
100
0.37
0.00005
0.0183
-------
I •
K-
I
APPENDIX F. SAMPLE CALCULATIONS
This appendix describes selected calculations performed for calculating
various parameters and factors. Sample calculations for the field evalution
at Plant A is provided in the first section. This is followed by cost-
effectiveness calculations for a thin-film evaporator (Plant A) and
calculations for the cost analysis of steam stripping.
EXAMPLE CALCULATIONS FOR THE FIELD EVALUATION AT PLANT A
Methylene chloride was present at 2.0 percent in the feed. The
concentration in the product was 0.9 percent. There was no liquid analysis of
the bottoms, but there was a headspace analysis of 0.03 mg/L. Assuming the
same partition coefficient for the feed as for the bottoms, the estimated
concentration in the bottoms was (0.03/1.7)(2.0 percent) or 0.04 percent. The
following is a material balance assuming 95-percent product recovery of the
feed, and assuming a 100-L feed..basis. The methylene chloride lost from the
vent is estimated by material balance.
Volume
Bottoms
Vent
100
95
5
Unknown
Percent
methylene
chloride by
volume
2
0.9
0.035
Unknown
Volume
methylene
chloride (L)
2
0.86
0.0018
1.14
The percent rerrcval of methylene chloride from the bottom sludge and
associated liquids is (2 - 0.0018)/2 x 100, or 99.91 percent. The vent loss
is (1.14 L/2.0 L) x 100, or 55 percent.
EXAMPLE CALCULATIONS FOR THIN-FILM EVAPORATOR UNIT OPERATING COSTS (Table 20)
At a feed rate of 23 L/min, the volume of waste treated is 23 x 60 x 24 x
273 or 9,040,000 L over a 273-day year, at 24 hours per day. Since the annual
cost is $252,600, the cost per liter of waste treated is $252,600/9,040,000 L,
or S0.0279/L. The cost of organic recovered (85 percent) equals (S0.0279/L
waste) x (L waste/0.85 L organic) or $0.033. With a density of the organic of
0.8 Mg/m3, the cost of recovering a metric ton of organic is $252,600/9,040
m3/(0.85 m3organic/m3 waste/(0.8 Mg/m3) or $41.10.
200
-------
EXAMPLE CALCULATIONS FOR STEAM STRIPPING
The initial volume charged to the stripper in Batch 1 was 1,260 L. This
contained 0.5 percent organics by weight or 7 L organics (density 0.866
g/cm3). However, there were 333 L of organics recovered as condensate.
Thus, there must have been at least 333 L of organics in the stripper
initially (i.e., the stripper contained organics from a prior run). The
initial water content is 1,260 - 7 or 1,253 L. This would imply an initial
weight fraction of 0.866(333)/(0.866 x 333 + 1,253) or 0.187 (21 volume
percent). The final batch volume is 1,420 L with nearly 500 mg/L or 0.0005
weight percent organics. Therefore, the percent VOC removed is [1 - 1,420 x
O.C005/{333 x 0.866)] x 100 or 99.8 percent.
Batch 1 has a volume of almost 1,590 L (1,253 + 333) and contained 21
volume percent volatiles. The steam rate was 4.33 L/min. The stripping rate
constant is 0.073 min" . The batch cycle time is 40 min heating and transfer
+• 83 min stripping. The condensate is the steam rate x time or 83 rain x 4.33
L/min or 359 L. The condensate treatment time is the prorated treatment time
of a 1,600 L batch of condensate (3 tine constants) x (13.7 min) x (359/1,600)
+ 40 min x (359/1,600) or 18.2 min. The volume of steam generated during this
treatment was (41.1 min)(4.333 L/min)(1,600/1,590) or 179 L. The prorated
amount to the treatment of the 359 L of wastewater was 359 x 179/1,600 or 40.2
L. Repeating the above process» the prorated condensate from treating the
condensate's condensate would be 40.2 x 40.2/359 or 4.5 L. The treatment of
the 4.5 L would be 4.5 x 40.2/359 or 0.504 L. The sum of these amounts of
condensate treated is 359 + 40.2 + 4.5 + 0.504 or 404 L. The fraction f is
404/359 or 1.126.
The cost of the condensate treatment is the average cost of the waste
treated ($262,900) divided by the annual volume (4,157,000 L) tines the total
condensate treated. 40.4 L divided by the volume of the batch (1,590 L), or
S0.016/L waste. This cost of processing the condensate is adjusted by the
solvent recovery costs (S0.20/L recovered VOC) x (0.8216 L VOC/L of waste).
The recovery value is assumed to equal zero in this example.
The above cost of condensate treatment is not used in estimating the
waste processing cost. The batch time is adjusted to accommodate the time for
treating the condensate [123 min. + (18.2 min)(1.126)] or 143 min. The total
operating time is 24 hours per day times 260 days or 374,400 min. At 143 min
per batch, this is the equivalent of 2,617 batches or 4,157,000 L.
p.
*>;;•
5?'
201
------- |