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

-------