.C, CDA Environmental Protection EPA-600/R-05/035
Vtr r^. Agency AFRL-ML-TY-TR-2002-4506
January 2005
SERDP
Final Report:
Membrane-Mediated Extraction and
Biodegradation of Volatile Organic
Compounds from Air
Stephen W. Peretti and Robert D. Shepherd
North Carolina State University
Raleigh, NC 27695
Russell K. Clayton and David E. Proffitt
ARCADIS Geraghty & Milller
4915 Prospectus Drive
Durham, NC 27713
EPA Project Officer
Norman Kaplan
Office of Research and Development
National Risk Management Research Laboratory
109 T.W. Alexander Drive
Research Triangle Park, NC 27711
Approved for Public Release; Distribution Unlimited
AIR FORCE RESEARCH LABORATORY
MATERIALS & MANUFACTURING DIRECTORATE
AIR EXPEDITIONARY FORCES TECHNOLOGIES DIVISION
139 BARNES DRIVE, STE 2
TYNDALL AFB FL 32403-5323
-------
-------
REPORT DOCUMENTATION PAGE
Form Approved
OMB No. 0704-0188
Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the
data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing
this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302, and to the Office of Management
and Budget, Paperwork Reduction Project (0704-0188), Washington, DC 20503.
1. AGENCY USE ONLY (Leave blank)
2. REPORT DATE
10 Nov 2001
3. REPORT TYPE AND DATES COVERED
Final Technical Report, 951001-010930
4. title and subtitle Final Report: Membrane-Mediated Extraction
amd Biodegradation of VOCs from Air
5. authors Stephen W. Peretti, Robert D. Shepherd, Russell K.
Clayton and David E. Proffitt
5. FUNDING NUMBERS
Contract No: 68-C-99-201
JON 3904A38B
PE: 63716D
6. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES)
ARCADIS Geraghty & Miller
4915 Prospectus Drive
Durham, NC 27713
7. PERFORMING ORGANIZATION REPORT
NUMBER
EPA-600/R-05/035
SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES)
AFRL/MLQL
139 Barnes Drive, Suite 2
Tyndall AFB, FL 32403-5323
SPONSORING/MONITORING AGENCY
REPORT NUMBER
AFRL-ML-TY-TP-2002-4506
11. SUPPLEMENTARY NOTES
Technical monitor: Dr Joe Wander, AFRL/MLQF, 850-283-6240 [DSN 523-6240]
12a. DISTRIBUTION/AVAILABILITY STATEMENT
Public release authorized. Distribution unlimited. Available from
www.serdp.org as a download of the project final report.
12b. DISTRIBUTION CODE
13. ABSTRACT (Maximum 200 words)
This report describes feasibility tests of a two-step strategy for air pollution control applicable to exhaust
air contaminated with volatile organic compounds (VOCs) from painting aircraft. In the first step of the
two-step strategy, the VOC-contaminated exhaust air passes over coated, polypropylene, hollow-fiber
membranes while an involatile liquid (silicone oil, mineral oil, decanol, octanol) is pumped counter-
current through the filters. The organic liquid captures the VOCs, and their concentration in the
circulating liquid increases whenever exhaust air circulates. In the second step, the circulating organic
loop passes through a second set of hollow-fiber membranes that support a culture of microorganisms,
which remove and metabolize the VOCs, on their exterior surfaces. The concentration of VOCs in the
circulating loop oscillates as the painting process starts and stops [Continued on p. ii]
14. SUBJECT TERMS
Air pollution control; aircraft painting; biodegradation; biofilter;
hollow-fiber membranes; VOCs
15. NUMBER OF PAGES
186
16. PRICE CODE
17. SECURITY CLASSIFICATION
OF REPORT
UNCLASSIFIED
18. SECURITY CLASSIFICATION
OF THIS PAGE
UNCLASSIFIED
19. SECURITY CLASSIFICATION
OF ABSTRACT
UNCLASSIFIED
20. LIMITATION OF ABSTRACT
UL
NSN 7540-01-280-5500
Computer Generated
STANDARD FORM 298 (Rev 2-:
Prescribed by ANSI Std 239-18
298-102
l
-------
13. ABSTRACT (CONTINUED)
because VOC capture by the liquid is a fast process whereas removal and metabolization by
microorganisms is a slow process. Despite constaints caused by limited availability of commercial
membrane packages, adequate rates of removal and transport into and out ot circulating octanol
were shown to be adequate to support the proposed technology. Biodegradtion was also
qualitatively validated, although each of the organisms used in these tests selectively metabolized
specific classes of solvents; however; other cultures or sequential treatment stages are expected
to provide satisfactory removal. Scale-up revealed material incompatibility of the membranes and
adhesives with octanol. Silicone oils and vegetable oils were briefly tested as the circulating
organic liquid at the end of the project. Pressure drop also remains as an engineering challenge
unless ventilation exhaust rates are decreased.
11
-------
NOTICCS
uswg <;o'vmxmknt drawings, specifications, mi< k» htk iuta
INCLUDED 11% * KIV mh t V1»M \nu \W rt.WOM OTHER THAN
GOVERNMENT PROCUREMENT DOES M«r IN tM u *> ^WIJGATK THE l-S
CiOVERNMfcNT. Till: ¥\Ct THAT Till CXWtK.vtfl vj MEMULATED OH
SUPPLIED THE DRAWINGS, Ml « H!i"\hh\\ OR otflt-w 11 \ i DOES NOT
LICENSE THE HOLDER OK ANY OTHER PERSON OR CORPORATION; OR
CONVEY ANY R.HJHTX OR PERMISSION TO >1AP€t?f,%<*?!?'l.E, USE, OR SELL ANY
t'.Ul'MCfU^ i \HO\ lint VI I U! Hi (HIM
THIS TECHNICAL REPORT HAS BEEN REVIEWED AND IS APPROVED MM
PUBLICATION,
DONALD II. Ht'CKLE, CM, USJ
Cliifl, Airbasc T«cl»«lc»f!it l>Ms
Tliit report la iimitsifeferitf h» tile of wtentlfk* Ireiiitkill mtmrmmikm ejtcfcM** mm «i»w» tint
cii«*ill»ii Approval or ilijapf »*•! »»f II* Mew #r rtmliiif*,.
We i»! r*t*s-B fo»i« of III* report mile* ewilwcliitl ®r w?fk$ on i ipf'Cifif ctofittfitfit
mpifri- it* return.
0
in
-------
Foreword
The U.S. Environmental Protection Agency (EPA) is charged by Congress with protecting the
Nation's land, air, and water resources. Under a mandate of national environmental laws, the
Agency strives to formulate and implement actions leading to a compatible balance between human
activities and the ability of natural systems to support and nurture life. To meet this mandate, EPA's
research program is providing data and technical support for solving environmental problems today
and building a science knowledge base necessary to manage our ecological resources wisely,
understand how pollutants affect our health, and prevent or reduce environmental risks in the
future.
The National Risk Management Research Laboratory (NRMRL) is the Agency's center for
investigation of technological and management approaches for preventing and reducing risks from
pollution that threaten human health and the environment. The focus of the Laboratory's research
program is on methods and their cost-effectiveness for prevention and control of pollution to air,
land, water, and subsurface resources; protection of water quality in public water systems;
remediation of contaminated sites, sediments and ground water; prevention and control of indoor
air pollution; and restoration of ecosystems. NRMRL collaborates with both public and private
sector partners to foster technologies that reduce the cost of compliance and to anticipate emerging
problems. NRMRL's research provides solutions to environmental problems by: developing and
promoting technologies that protect and improve the environment; advancing scientific and
engineering information to support regulatory and policy decisions; and providing the technical
support and information transfer to ensure implementation of environmental regulations and
strategies at the national, state, and community levels.
This publication has been produced as part of the Laboratory's strategic long-term research
plan. It is published and made available by EPA's Office of Research and Development to assist
the user community and to link researchers with their clients.
Sally Gutierrez, Acting Director
National Risk Management Research Laboratory
iv
-------
EPA Review Notice
This report has been peer and administratively reviewed by the U.S. Environmental Protection
Agency and approved for publication. Mention of trade names or commercial products does not
constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Information Service,
Springfield, Virginia 22161.
v
-------
Abstract
This report describes feasibility tests of a two-step strategy for air pollution control applicable to
exhaust air contaminated with volatile organic compounds (VOCs) from painting aircraft. In the first
step of the two-step strategy, the VOC-contaminated exhaust air passes over coated, polypropylene,
hollow-fiber membranes while an involatile organic liquid (silicone oil, mineral oil, decanol,
octanol) is pumped counter-current through the fibers. The organic liquid captures the VOCs, and
their concentration in the circulating liquid increases whenever exhaust air circulates. In the second
step, the circulating organic liquid loop passes through a second set of hollow-fiber membranes that
support a culture of microorganisms, which remove and metabolize the VOCs, on their exterior
surfaces. The concentration of VOCs in the circulating liquid loop oscillates as the painting process
starts and stops because VOC capture by the liquid is a fast process whereas removal and
metabolization by microorganisms is a slow process. Despite constraints caused by limited
availability of commercial membrane packages, adequate rates of removal and transport into and
out of circulating octanol were shown to be adequate to support the proposed technology.
Biodegradation was also qualitatively validated, although each of the organisms used in these tests
selectively metabolized specific classes of solvents; however, other cultures or sequential treatment
stages are expected to provide satisfactory removal. Scale-up revealed material incompatibility of
the membranes and adhesives with octanol. Silicone oils and vegetable oils were briefly tested as
the circulating organic liquid at the end of the project. Pressure drop also remains as an engineering
challenge unless ventilation exhaust rates are decreased.
vi
-------
Acknowledgments
This research was supported by the U.S. Department of Defense through the Strategic Environ-
mental Research and Development Program (SERDP). This report was prepared under Contract
Number 68-C-99-201 for the U. S. Environmental Protection Agency (EPA), Research Triangle Park,
NC.
This final report describes work performed from 1 March 1998 to 30 January 2001. The Co-
Principal Investigators were first Norman Kaplan then Jack Wasser of EPA's National Risk
Management Research Laboratory (NRMRL) and Dr. Joe Wander of the Air Force Research
Laboratory (AFRL/MLQ).
The authors particularly acknowledge the contributions and guidance of Bradley Smith and Dr.
Robert Hoist of the SERDP Program Office and the review panel members, without whose help this
study would not have been possible.
vii
-------
Table of Contents
Section Page
Notices ii
Foreword iii
Abstract v
Acknowledgments vi
Index of Tables ix
Index of Figures x
Executive Summary xi
1.0 Objective 1
1.1 Background 1
1.1.1 Membrane BioTechnology Development Background 2
1.1.2 Treatment Process Concept 5
1.2 Scope 7
2.0 Technical Objectives 8
3.0 Results at Bench-Scale 10
3.1 Development of Coated Modules 10
3.2 Membranes Coated with Silicone Rubber 13
3.3 Experimental Structure 18
3.3.1 MBT Bench-Scale Separation Contactor 19
3.3.2 Pilot Testing 20
4.0 Results/Data 23
4.1 Separation Module Tests 23
4.1.1 Bench-Scale Tests 23
4.1.2 Pilot-Scale Tests 25
4.2 Biological Treatment System 29
4.2.1 Suspended-Cell Experiments 30
4.2.1.1 Screening Studies 30
4.2.1.2 Xylene Degraders 30
viii
-------
Table of Contents (continued)
Section Page
4.2.1.3 Aliphatic Degraders 31
4.2.1.4 Growth Studies 31
4.2.1.5 MX and XI Growth on /^-Xylene (GS 11, 12) 32
4.2.1.6 Effect of Ethylbenzene on Growth of XI on m-Mylene (GS13) 33
4.2.1.7 Growth of XI on m-Xylene and /^-Xylene (GS14) 34
4.2.1.8 Ml Growth on Butyl Acetate and a 50:50 Mixture of Butyl Acetate
and MEK (GS 15, 19) 35
4.2.2 Flat-Sheet Biofilm Experiments 35
4.2.2.1 Growth of XI on w-Xylene and /^-Xylene (FS 6, 7, 8) 37
4.2.2.2 Growth of XI on m-Xylene and /^-Xylene (FS 9) 39
4.2.3 Hollow-Fiber Membrane Experiments 39
4.2.4 Staged Biotreatment of VOC Mixtures in Lab-Scale Reactor 40
5.0 Conclusions 43
5.1 Separation System 43
5.2 Biotreatment System 45
5.2.1 Biodegradation Range and Extent 45
5.2.2 Problems Arising from Metabolic Regulation 45
5.2.3 Treatment Strategy 46
5.2.4 Implementation 46
6.0 Recommendations 47
Appendices
Appendix A Literature Search and Review A-i
Appendix B Bench-Scale Data B-i
Appendix C Pilot-Scale Data C-i
IX
-------
Index of Tables
Table Page
1 Compounds Successfully Biodegraded 3
2 Growth Rates of MX-2 with Modified Carbon Sources 4
3 Partition Coefficient Values 4
4 Oxygen Uptake Rates for Various VOCs 5
5 PDD-Coated Membrane Results 11
6 Fiber Coating Technique Development 12
7 Air-in-Shell Tests - Cylindrical Parallel-Flow AMT Module 14
8 Octanol-in-Shell Tests - Cylindrical Parallel-Flow AMT Module 15
9 Tertiary Mixtures in the Air Stream 22
10 Typical Flow Rates Used During Testing 22
11 Summary of Bench-Scale VOC Tests 24
12 Summary of Pilot-Scale VOC Tests 26
13 Suspension-Culture Experiments 32
14 Flat-Sheet Biofilm Experiments 36
15 Biomembrane Mass Transfer 40
16 Biotreatment of VOC Mixtures in a Lab-Scale Reactor 41
17 Concurrent Degradation 45
x
-------
Index of Figures
Figure Page
1 MBT System Schematic 2
2 VOC Extraction in the S/C Unit 6
3 Bioextraction of VOCs 6
4 AMT Cross-Flow Module - Overall Dimensions in Inches 16
5 Pressure Drop for AMT Cross-Flow Module 17
6 Separation Module Bench-Scale Test Apparatus 19
7 Process Schematic of Separation Module Bench-Scale Apparatus 20
8 Pilot-Scale System Design 21
9 Flat-Sheet Contactor Schematic 35
xi
-------
Blank Page
Xll
-------
Executive Summary
A. Objective
The objective of this project was to examine the feasibility of capturing and destroying volatile
organic compounds (VOCs) from process or storage exhaust air by extracting the VOCs through a
coated, hollow-fiber membrane into an involatile liquid and then metabolizing them with bacteria
residing on the exterior surface of a second, coated, hollow-fiber membrane.
B. Background
1. VOC Emissions from Large Aircraft-Painting Facilities
Implementation of the Clean Air Act in the form of the National Emission Standards for Hazardous
Air Pollutants for aerospace coating operations (Aerospace Manufacturing and Rework Facilities
NESHAP) imposed a requirement that large aircraft-painting operations either apply an emission
control system to decrease the amount of VOC emitted to the atmosphere by at least a threshold
amount (originally 81 percent) or apply coating materials that contain less than a specified (for each)
threshold amount of VOCs. Strenuous efforts to develop competent low-VOC coatings have not
been uniformly successful, and use of these compliant coatings generally involves compromises in
coat quality and durability and in preparation, application, and curing time and effort.
By most standards, the option to control emissions is by far the superior approach because it
preserves availability of coatings that have been optimized after decades of development and
experience, because it involves no changes to the established painting methods and techniques, and
because the amount of VOCs emitted are less than half that from an equivalent operation conducted
with low-VOC coatings. However, ventilation of an aircraft-painting facility makes inefficient use
of air by continuously ventilating the entire volume of a hangar while painting is conducted
intermittently in a much smaller part of it, so the exhaust volume is large and the level of
contamination is low. Both factors act to drive up the cost to decontaminate these exhausts by
conventional VOC control methods.
The Strategic Research & Development Program (SERDP) issued Statement of Need CP-98, VOC
Control Technology for Aircraft Painting and Depainting Facilities as a call for develop-ment and
evaluation of alternative technologies that, alone or in combination with flow-reduction technologies
(e.g., exhaust recirculation), would decrease the cost to control emissions of VOCs from aircraft
Xlll
-------
painting operations. Heat generation and loss in processes handling flow rates approaching 1M
ft3/min is commonly the main source of cost, fuel consumption, and greenhouse gas and airborne
pollutant generation. As efforts to oxidize organic vapors by catalysis at low temperatures have
found success only with such easily oxidized materials as aldehydes and thioethers, the most-
promising avenue of development is through concentration of the VOCs prior to destruction.
Recovery is possible as well but rarely economical for gross mixtures of solvents.
2. Membrane Extraction and Biotreatment
With appropriate modification of the structural polymeric surface, hollow-fiber membranes (HFMs)
allow selective passage of molecules based on their physical properties. Significant research has
examined separation and biotreatment of VOCs from air streams with varying degrees of success.
A literature search performed during this project led to several pertinent conclusions regarding
membrane configuration, materials of construction, extraction fluid, module operation and control
of biofouling. These conclusions are summarized as follows:
1. Hollow fiber shell-and-tube membrane modules offer the highest possible surface-area-to-
volume ratio, roughly an order of magnitude more than the nearest alternate, the spiral-wound
configuration.
2. Asymmetric, composite membranes composed of a highly porous support membrane coated by
a thin, nonporous, permselective film offer the most effective combination of perme-ability and
selectivity.
3. For gas-liquid systems, it is most efficient to have the coating film contacted by the liquid phase
and the pores filled by the gas. For instances where the coating film is the predominant mass
transfer resistance, the fluid that fills the pores is less important.
4. The composition of the coating film is critical to performance. In decreasing order of
permeability to organic vapors, it was found that poly[(l-trimethylsilyl)-l-propyne] (PTMSP)
is greater than polydimethylsiloxane (PDMS) which is greater than polyalkylsulfone (PAS-16),
other rubbery polymers are greater than fluoropolymers. Although PTMSP exhibits the highest
permeability to organic vapors, its performance decays relatively rapidly over time, so it is not
considered suitable for long-term commercial applications. PDMS is a rubbery polymer that,
along with other silicone rubber derivatives, is the material of choice by researchers involved
in organic vapor separations. PAS-16 is relatively uncharacterized but offers excellent properties
as a rubbery polymer with local crystallinity.
5. Lower temperatures favor organic vapor separations.
6. Removal percentages in excess of 95 percent are possible with membrane extraction systems
operated with gas-membrane contact times on the order of 20 seconds.
7. Low-vapor-pressure oils and alcohols (silicone oil, mineral oil, decanol, octanol) exhibit
excellent solubility and permeability characteristics for organic vapor separation, with silicone
xiv
-------
oil exhibiting optimal performance in composite membranes.
Peretti et al. have filed for a patent for the use of coated HFMs with involatile solvents to capture
VOCs from furniture-finishing industries and to biodegrade the VOCs by a process of direct capture
across a second HFM by bacteria colonizing the external surface.
Figure E-1 depicts the functional elements and flows in the Membrane Extraction and Biotreatment
(MBT) process. Contaminated exhaust gas enters at the bottom left and is stripped of VOCs as it
passes upward through an array of parallel HFMs. The stripped exhaust is released to the
atmosphere, so the amount of VOC in the treated exhaust is the total amount released, and this
amount defines the efficiency of the control device. An involatile solvent circulates inside the FIFMs
opposite to the direction of airflow, so the concentration of captured VOCs increases downward to
the point of entry of the facility exhaust before entering the bottom of the membrane bioreactor
(MBR).
Clean
Ail*
Membrane
separation
Particulate
Prefilier f
—CD—^
Exhaust
iias
f
Sir i p;im L!
Fluid
Su>ra«e
T ank
)
Recycled
Slnppniu
Fluid"
1
Make-up
Nutrient
f 1
Nutrient
Mining
Recycle
Tank
Bitnnem Inane
EH1 Lien I
Figure E-1. MBT System Schematic
The MBR is a similar array of parallel HFMs surrounded by a nutrient medium circulated downward
among the FIFMs to sustain a culture of microorganisms adhering to their surface. The solvent and
its load of VOCs rises inside the HFMs while the microorganisms capture and consume part of the
VOCs. During periods of painting, VOCs will enter the system faster than the microorganisms can
xv
-------
consume them. The stripping fluid storage tank is a reservoir of fluid with an unreacted VOC
concentration that increases gradually during painting and subsides during interludes between
painting episodes. This distribution of peak load serves both to decrease the size of the MBR and
to ensure a fairly constant rate of delivery of VOCs to the MBR.
Operating equations were derived to describe the membrane separation processes for a system with
non-coated hollow fibers and using octanol as the stripping fluid. The final result for the
separation/concentration unit was a design equation that relates concentration, partition coefficient,
membrane surface area, and flow rate to an overall mass transfer coefficient, Ka. The Ka is based on
the overall system driving force and is defined by a sum of resistances model. In the equation shown
below for Ka in the separation/concentration unit, the concentration (C) subscripts A and 0 denote
the air and octanol phases, and subscripts 1 and 2 represent inlet and outlet conditions, respectively.
P is the air/octanol equilibrium partition coefficient, Q is the volumetric flow of the respective
phases (cubic centimeters per second), and Am is the membrane surface area (square centimeters).
In
C
A2
P
-c
O 2
c
A1
c,
Kr
v P
01
A.
1
1
a, QaP.
Experiments were initially conducted using individual pure VOCs that are typical components of
paints to assess mass transfer rates and removal efficiencies. Studies progressed to include VOC
mixtures and real military paint. Degradation of VOCs was measured for individual and mixed
biofilm cultures. All of these experiments were necessary because both the separation and
biotreatment processes are competitive among the species present.
C. Scope
This is the final technical report for SERDP project CP-1105, Membrane-Mediated Extraction and
Biotreatment of VOCs. Results of the first half of the project have been presented at two national
meetings. The associated long abstracts published in the respective meeting proceedings are
included as Appendixes D and E.
xvi
-------
D. Methodology
Initial evaluations of mass transfer coefficients for the membrane module were conducted by
quantifying the removal of each individual VOC [w-xylene, toluene, methyl ethyl ketone (MEK)]
from an air stream. Experiments were performed to determine the effect of airflow rate, stripping
fluid flow rate, air stream VOC concentration, and stripping fluid VOC concentration on overall
mass transfer coefficients. This early work was performed using the Celgard Liqui-Cell module. The
focus of the separation/concentration process development was aimed at producing a module that
provided high-efficiency removal of VOCs from the airstream at high air flow rates and low
pressure drop.
E. Test Description
Numerous biotreatment experiments were conducted to determine efficacy of the proposed biotreatment
module for enhanced VOC removal from the stripping fluid. These studies included the following:
• Screening experiments to identify organisms able to degrade paint VOCs
• Liquid-liquid stripping efficiency of MEK with and without a biofilm present
• Degradation of individual and mixed VOCs with individual and mixed organism biofilms
• Growth of degraders and degradation of single and mixed VOCs to determine strain characterization,
substrate range, metabolic regulation, and organism interactions
• Capacity of dual organisms in staged reactors to degrade mixtures of differently soluble VOCs
F. Results and Discussion
The Celgard Liqui-Cell membrane module, or contactor, was used in the early bench-scale
experiments. Because of the difference in pressures between the air and stripping fluid sides of the
HFMs, some leakage of stripping fluid into the air occurred. Because it is not economically feasible
to pressurize the air side to prevent this leakage, the application of a thin coating of a VOC-
permeable coating to the HFMs, preferably on the inside of the fibers to maximize mass transfer was
considered. Acceptable coated fibers and modules were not commercially available, and work with
a number of vendors to provide a suitable contactor proved to be unsuccessful. An in situ coating
technique for the Celgard contactor was developed but was found to be too time consuming as a
cost-effective approach.
A secondary issue regarding cost-effective contactor design was the relatively high pressure required
to drive the air through the Celgard module. Discussions were held with module vendors to develop
an efficient separation/concentration module with non-porous coated fibers and low pressure drop.
Applied Membrane Technologies (AMT) was selected to provide a cross-flow module design with
fibers externally coated with plasma-polymerized silicone rubber. The coating was found to perform
xvii
-------
adequately, but the fibers themselves elongated when exposed to the octanol stripping fluid. This
led to the substitution of silicone oil as the stripping fluid. Upon developing a feasible module
design, pilot testing was performed to validate its performance.
Two basic sizes were tested. Bench scale tests using 1 and 2 modules were done using the Celgard
microporous hollow-fiber membrane module coated with perfluorodimethyldioxole and tetrafluoro-
ethylene (PDD-TFE). Pilot scale studies with paint vapors were done using the AMT module coated
with silicone rubber in arrangements of 2 and 10 modules. Test flow rates up to 200 crm provided
contact times of less than 0.1 sec with the coated membrane.
In the bench-scale testing, VOC removal rates ranged between 4.4 percent and 73.7 percent. Higher
air side flow, lower oil flow, and lower VOC inlet concentrations were generally associated with
lower removal rates. Pilot-scale test results were variable and sometimes difficult to understand.
Average VOC removal rates were 34-80 percent, while removal of individual compounds ranged
from 17 to 82 percent. As in the bench-scale testing, MEK proved to be the most difficult compound
to extract from the air. A major problem with the cross-flow modules occurred when leaking
modules allowed the fibers to become wetted with the less-viscous silicone oil, causing fiber
elongation and subsequent voids between some fibers and matting of others. This resulted in poor
contact with the air stream and reduced VOC removal rates. Time and funding limitations prevented
actions to address these issues. These problems resulted in a reduction in planned pilot scale stream
from 500 cfm to 200 cfm.
G. Conclusions
The membrane-supported biofilm modules successfully removed VOCs from the recirculating
stripping fluid stream. Degradation of the aromatic compounds investigated (toluene, w-xylene) was
achieved; these compounds were not observed in the aqueous phase above the biofilm. MEK
biodegradation is problematic, appearing to be partially inhibited by toluene and m-xylene. Further
studies are required to ascertain the underlying mechanism.
A fully-integrated pilot system was not successfully demonstrated. Although the test results did not
meet the research goal of 85-95%, the MBT concept showed potential for being developed into a
technically feasible process. However, the MBT concept has been shown to offer several attractive
benefits:
• Continuous biotreatment of VOCs directly from the stripping fluid avoids the mechanical
complexities of sequential medium transfers found in most concentrate-and-treat designs
(e.g., air-to-adsorber-to-lower-volume-air-stream, and thence to final treatment).
• Continuous recirculation of captured VOCs through the biotreatment module provides
xviii
-------
complete destruction of captured VOCs.
• Modularity of the MBT unit allows linear scaling of a large system by connection of n units
in parallel
• Modularity of the MBT unit allows amplification of the net capture-and-removal efficiency
by connection of two or more units in series.
H. Recommendations
Several serious obstacles remain before a practical MBT system can be applied to a painting
operation:
• Available materials of construction must be compatible with each other (fiber, coating,
assembly adhesives), with the stripping fluid, with the microorganisms, and with the VOCs
to be treated.
• Manufacturing techniques for membrane modules must advance enough to allow cost
effective fabrication of low-pressure drop, high efficiency, leak-free modules.
• Microbiological cultures must be identified that can coexist and metabolize all of the VOCs
to be treated at practical rates.
• Some engineering relief may be possible, but the ventilation system must be able to
accommodate a fairly large pressure loss (>25 in. H20) across the HFM array, consistent
with acceptable process economics.
Additional fundamental and applied research is needed to fill out the understanding of these
processes. Design and eventual commercial availability of properly configured and scaled hollow-
fiber membrane modules must occur before MBT or related technologies can be implemented on
a practical scale. Finally, this technology will be compatible only with processes that can tolerate
a moderate pressure (>20 in. H20) loss through its control system, for example, low flow high
concentration sources.
xix
-------
1.0 Objective
The objective of this project was to examine the feasibility of developing a practical VOC control
method using coated, hollow-fiber membranes to extract organic vapors from ventilation exhaust
streams into a circulating pool of an involatile liquid and to deliver the organics at a buffered rate
to a biofilm adhering to the exterior surface of a separate, coated membrane.
1.1 Background
This project was performed in response to the Statement of Need for FY98 SERDP, Com-pliance
New Start Number 2 (CPSON2), entitled, "VOC Control Technology for Aircraft Painting and
Depainting Facilities." Painting and coating operations present a number of environmental problems
and economic challenges. Volatile organic compounds (VOCs) and other hazardous air pollutants
(HAPs) are present in all currently used coatings. The toxic compounds include metals, metal
oxides, and VOCs. Many of these compounds are either direct or indirect health threats; VOCs are
ozone precursors and may be designated as toxic, and many metals and metallic oxides are identified
on toxic compound lists. In response to the Clean Air Act Amendments of 1990, VOCs and HAPs
in coatings are being reduced, thereby reducing emissions of ozone precursors and toxic compounds
from painting operations. However, additional controls are mandated in specific instances, such as
aircraft booths. The National Emissions Standard for Hazardous Air Pollutants (NESHAP) specific
to aircraft painting will force the DoD to either implement volatile hazardous air pollutant (VHAP)
control technology or replace existing coating formulations. Because efforts to develop replacement
coatings have met with only mixed success, implementation of control technology appears to be the
most-promising near-term solution.
Control technology cost primarily depends on contaminated airflow rates. Paint spray booth
exhausts are high-volume streams because an obsolete OSHA standard requiring a minimum
velocity of 100 ft/min) through all booth section areas remains in the public record. Conventional
booth design approaches include no provision for adjusting flowrate, relying instead on using a high
flowrate with clean filters that will remain above the 100-fpm threshold after the filters are dirtied.
If controls are required for VOC destruction, the necessary equipment must be sized for the
maximum exhaust flow rate. As a result, typical booths emit large volumes of air contaminated with
dilute concentrations of VOCs and HAPs. Many current technologies treat the VOCs within the
entire gas volume directly, leading to large-volume incineration, absorption, or biofiltration systems.
These technologies are extremely expensive in terms of both capital and operating expenses. Also,
they often generate hazardous byproduct streams that must be further treated. The system evaluated
in this research was designed to both minimize the treated volume and to concentrate the VOCs
within that treated volume in order to reduce the size and cost of the ultimate control device. These
1
-------
advantages would make this VOC treatment option applicable across a broad range of spray booth
sizes.
Such a VOC control system could eliminate a significant portion of toxic materials emissions from
DoD installations. Past data regarding aircraft service reported in Air Force Times indicate that 5
of the top 10 air discharges that triggered Toxic Release Inventory reporting thresholds from 131
DoD installations were typical paint constituents. Significant reduction of these emissions in a cost-
effective manner is important to DoD's adherence to the 1995 Aerospace NESHAP for Aerospace
Manufacturing and Rework Facilities and to its meeting existing and evolving limits for VOC
emissions in ozone nonattainment areas.
1.1.1 Membrane BioTechnology Development Background
During initial Membrane BioTechnology (MBT) development, tests were performed to assess the
ability of hollow-fiber membrane contactors to separate VOCs from an air stream. A membrane
separation system was constructed to allow contact of VOC- laden air streams with octanol inside
a Hoechst-Celanese Liqui-Cel hollow-fiber module. The entire membrane separation system used
in initial experiments is shown schematically in Figure 1. A stripping fluid, octanol, was passed
through the unit's shell space while air flowed through the fibers. The two phases contacted in
counter-current cross flow. The octanol reservoir was recycled to the module, but air passed through
the system only once. Air flow rates were varied from 10 to 40 L/min, giving a minimum
gas/membrane contact time of 0.004 seconds based on the inside volume of the hollow fibers. The
air-side pressure drop ranged from 0.5 to 2.0 psi, and the total surface area available for mass
transfer was 1.4 m2.
Make-up
Nutrient
Mixing
Tank
T
Ettluent
Figure 1. MBT System Schematic
Exhaust
gas
Stripping
Nutrient
Recycle
Membrane
separation
Storage
Particulate
Pre filter
Recycled
Stripping
Fluid
Biomembrane
2
-------
Biological degradation experiments were conducted with naturally occurring microorganisms
isolated from soil samples removed from a site contaminated with gasoline. Using a mixed
consortium of organisms isolated in liquid culture from soil samples removed from this site,
degradability of model compounds from each of the species found in furniture exhaust gases was
examined (Table 1). Following completion of these initial studies, organism subcultures were
generated for specific compounds. By enrichment of the initial gasoline-adapted consortium in
individual flasks with isobutyl acetate (IBA), methanol (MeOH), methyl ethyl ketone (MEK),
methyl isobutyl ketone (MIBK), w-xylene, and ^-xylene, respectively, consortia capable of
degrading each compound were developed.
Table 1. Compounds Successfully Biodegraded
Acetone
Benzene
Diethylene glycol ethyl ether
Ethylene glycol butyl ether
Formaldehyde
Ethanol
Isobutyl isobutyrate
Isobutyl acetate (IBA) *
Methyl ethyl ketone (MEK) *
Methanol
Methyl n-amyl ketone
Methyl isobutyl ketone (MIBK) *
n-Butyl alcohol *
Styrene
Toluene *
m-Xylene *
o-Xylene *
p-Xylene *
* Indicates compounds common to DoD painting operations
One pure culture, designated MX-2, was established on m-xylene and another, PX-2, was established
on /^-xylene. Other consortia which grew on either IBA, MEOH, MEK, or MIBK consisted of
approximately three different organisms each. Maximum growth rates for the isolated consortia were
determined in shake flask studies containing low-ionic-strength buffer solution supplemented with
the appropriate carbon source. Growth rates obtained with w-xylene in the presence of additional
carbon sources and with pure octanol are given in Table 2.
The partition coefficient is defined as the ratio of concentrations of a given compound in two phases
(octanol/air and octanol/H20) at equilibrium. Before beginning work with Liqui-Cel membrane
modules, partition coefficient experiments were performed to investigate the equilibrium distribution
of /^-xylene, MEK, and MIBK between phases for stripping fluid/air and stripping fluid/aqueous
systems. Octanol was chosen as the stripping fluid. Values of the partition coefficient at different
temperatures are given in Table 3.
3
-------
Table 2. Growth Rates of MX-2 with Modified Carbon Sources
Carbon Source
Specific Growth Rate3
Degradation Rateb
m-Xylene
0.46 hr1
3.60x10"10 mg/(hr-cell)
m-Xylene + 0.2% MeOH
0.45 hr1
3.57x10"10 mg/(hr-cell)
m-Xylene + 0.2% EtOHc
0.40 hr1
2.34x10"10 mg/(hr-cell)
m-Xylene + 500 ppmd octanol
0.46 hr1
3.60x10"10 mg/(hr-cell)
500 ppm octanol (no m-xylene)
0.20 hr1
not applicable
a The specific growth rates are reported as the rate of change of cell dry mass divided by the cell
dry mass.
b The degradation rate is the rate of removal of m-xylene (in mg per mL of medium per hour)
divided by the cell density (in cells per mL).
c Ethyl alcohol.
d Mass/mass
Table 3. Partition Coefficient Values
Temp.
Partition Coefficient [octanol]/[air]
Partition Coefficient [octanol]/[water]
m-Xylene
MEK
MIBK
m-Xylene
MEK
MIBK
6 °C
9865
2181
NAa
NA
NA
NA
O
O
C\J
C\J
7978
1634
22,045
NA
NA
NA
31 °
7703
1344
8721
1021
33
2.1
a NA=not analyzed
Experiments were also conducted to examine the range of compounds degradable by enzymes
present in the meta- and /%/ra-xylene-degrading organisms. As shown in Table 4, these organisms
successfully removed many compounds that are typically difficult to biodegrade. Higher values for
oxygen uptake indicate a compound is being more rapidly degraded.
An additional set of experiments was run to evaluate alternative stripping fluids because
commercially available octanol is fairly expensive. Partition coefficients were determined for m-
xylene in corn oil, sunflower seed oil, and mineral oil at 31 °C. The respective partition coefficients
([oil]/[air]) were 8283, 8244, and 7284, comparable to that of octanol. These oils cost about one-
fourth that of octanol.
4
-------
Table 4. Oxygen Uptake Rates for Various VOCs
Compound
Oxygen Uptake, mmol/min-
mg Total Cell Protein
Strain PX-2
Strain MX-2
p-Xylene
0.34
1.21
m-Xylene
0.31
0.82
o-Xylene
NA
0.31
Toluene
0.17
0.18
Benzene
0.13
0.27
Styrene
0.47
0.55
Benzoic acid
2.34
0.66
Catechol
5.19
1.23
3-Methylcatechol
1.50
1.17
4-Methylcatechol
4.35
0.92
Protocatechuic acid
0.00
0.32
1.1.2 Treatment Process Concept
In the Membrane BioTreatment (MBT) system, organic volatiles are first separated from the air
stream, concentrated, and then completely metabolized by microorganisms. Selective removal and
concentration of VOCs from the exhaust stream enables significant reduction in the volume directed
to the final control device, dramatically reducing equipment size and costs. The system allows for
independent optimization of each process. One process removes organics from the air, and the other
process biodegrades them. The system relies on micro-porous hollow-fiber membrane contactors
to mediate the extraction and concentration of vapors from the air into an organic stripping fluid and
to provide a physical support for degradative microorganisms. A schematic of the MBT system
appears in Figure 1.
Exhaust gases laden with VOCs pass first through a particle filter, which removes solid particles
and any residual atomized droplets of coatings. Next, the gases enter a membrane
separation/concentration (S/C) unit composed of bundles of microporous, hydrophobic fibers. In the
S/C unit, vaporized HAPs and VOCs (represented as dark particles) are transferred from the exhaust
gases into a stripping-fluid medium (potentially octanol, silicone oil, sunflower seed oil, etc.), as
shown in Figure 2. The stripping fluid is chosen to have low volatility, low water solubility, and high
(fluid/air) partition coefficients for the VOCs. The medium serves as a pollutant sink and allows
accumulation of significant HAP/VOC concentrations.
5
-------
Aqueous Nutrient Stream
( > 11^)
• • I •
STRIPPING FLUID
Figure 3. Bioextraction of VOCs
Upon exiting the S/C unit, the stripping fluid is delivered to a biomembrane unit. There, the
stripping fluid circulates past one side of another microporous membrane with VOC-degrading
bacteria in a film on the opposite side of the membrane. Figure 3 illustrates diffusion of VOCs
through the membrane pores (filled with stripping fluid) into the biofilm, in which they are
selectively and completely metabolized by the bacteria. The solvent is then collected in a storage
vessel and ultimately recycled through the S/C. When hydrocarbon pollutants are treated, outputs
from the overall MBT System are clean air, carbon dioxide, and a mixture of water and
nonhazardous cell mass
In the design of a full-scale system for a military paint spray booth, further economic gains can be
realized by reducing the contaminated air volume through the application of partitioned
recirculation. This patented technology was developed through EPA and Air Force funding to reduce
the cost of VOC control by minimizing the treated volume. This technology takes advantage of the
fact that, in horizontal-flow booths, the lower segment of a paint booth exhaust contains more highly
concentrated VOCs, and conversely, the upper segment exhaust contains lower VOC concentrations.
This characteristic allows both the recirculation of a significant portion of the exhaust (30 to 90
percent) without adverse health and safety implications, and a comparable reduction in the size of
the required VOC control device. Adding on this technology enables a two-step reduction in the
volume of VOC-contaminated streams to be treated: first, partitioned recirculation concentrates the
VOCs into a smaller air volume; and second, the S/C unit concentrates the VOCs in a proportionally
smaller volume of stripping fluid for biodegradation.
MBT offers potential unique advantages due to the nature of the control technology and the impact
VOC/VHAP - containing gas
Microporous
Hydrophobic
Membrane
• • • • ~ •
STRIPPING FLUID
Figure 2. VOC Extraction in the S/C Unit
6
-------
of implementation on coating operations. Advantages include the following:
• High VOC destruction: Naturally occurring bacteria consume pollutants as food for growth
and energy.
• Non-Pollutant-Generating Process: MBT is a clean process with no hazardous by-
products.
• Optimized Rates of Removal and Degradation: Having separate processes for removal
and destruction of pollutant compounds allows each to be designed and operated for
maximum efficiency, and equipment size is minimized.
• Adaptability: MBT is fully adaptable to individual sites. S/C units are modular, which
allows the pollutant-removal process to be tailored to site-specific operations, facilities, and
regulatory permit requirements. Selection and optimization of suitable micro-organisms
ensures effective degradation of site-specific HAPs and VOCs.
• Extended Equipment Life: Each module of the S/C units may be changed on an individual
basis, and extra modules may be built into the system and/or kept on site to make
replacement easier.
• Operating Flexibility: The stripping fluid storage vessel allows the MBT system to operate
continuously to control intermittent processes. For example, some coating operations are
single-shift, resulting in eight hours of waste generation followed by 16 hours of down time.
In other instances, painting facilities may be offline for days or even weeks. The storage
tank mitigates the interruptions in waste generation—that is, the tank uncouples waste
generation from biotreatment, which allows the biotreatment process to operate at optimal
levels regardless of spray booth schedules. A VOC feedstock can be manually introduced
to maintain the bacterial colonies during extended interludes between painting episodes.
• Cost-Effective Treatment: Original estimates indicated that this system would be
significantly less costly than other typical VOC control systems for medium and large paint
spray booths.
1.2 Scope
This is the final technical report describing a 2-year project supported by the Strategic
Environmental Research and Development Program [SERDP],
7
-------
2.0 Technical Objectives
The overall objective of this small, pilot project was to validate and extend development of a
potentially cost-effective VOC control system for painting facilities that meets the requirements of
the Aerospace Coatings NESHAP—81 percent reduction in VOCs from noncompliant coatings. This
was believed to be feasible by combining the partitioned recirculation technique for flow reduction
with a novel process that concentrates VOCs for biological treatment. The project was designed as
a two-phase activity consisting of bench- and pilot-scale efforts. The objective of the project in
phase I was to demonstrate that membrane-supported extraction, coupled with membrane-supported
biotreatment, is a technically feasible VOC treatment process for DoD painting emissions. In phase
II, the obj ective was to establish the technical and econo-mical efficacy of this process to treat actual
aircraft painting emissions. The original concept was to use the paint booth facilities at Tyndall AFB
as the pilot test site, but that option was abandoned after technical difficulties caused delays and
added costs. Secondary goals of Phase II included both attention to the effects of particulate fouling
on membrane transfer performance and dissemination of information about the technology by
identifying all DoD sites and organizations that could benefit from this technology and distributing
appropriate technology transfer materials to them.
The original cost estimates for the technology were based on lab-scale performance. In support of
project technical objectives and economic feasibility, it was necessary to answer the following
critical questions:
1. Under conditions of 85-95 percent reduction of VOC emissions, what are the mass transfer
rates of VOCs present in DoD painting and depainting operations (e.g., MEK, MIBK,
xylenes, toluenes)
a) from air to organic solvent via membrane?
b) from organic solvent to aqueous phase via membrane?
2. What are the contact times needed to achieve the above mass transfer rates?
3. Can a membrane-supported biofilm be stably maintained?
4. What are the degradation rates of the above-cited VOCs?
5. Using commercially available membrane units for design purposes, what is the projected
cost of treatment in
a) dollars per cubic feet per minute of air treated?
b) dollars per unit of VOC removal?
The design is to be based on modules capable of controlling streams from 20,000 to 300,000
ft3/min of exhaust treated at typical VOC concentrations found in DoD operations.
6. What is the impact of particle and particle-bound contaminants such as isocyanates on
membrane performance? How does this impact filtration requirements?
8
-------
Answers to these questions were the subject of this project and were necessary to scale-up and
design the process and to determine process economics. Preliminary targets for mass transfer rate
were 10"4 cm/sec for VOCs from air and 3 x 10"10 mg/cell-hr for VOC degradation rate.
9
-------
3.0 Results at Bench-Scale
Details of progress in the bench-scale development phase (prior to pilot-testing) are described in two
publications, "Membrane-Mediated Extraction and Biodegradation of VOCs from Air," reviewed
and accepted 2/25/00, which was presented at the 2000 Spring National Meeting of the American
Institute of Chemical Engineers (AIChE), and "Membrane Biotreatment of VOC-Laden Air,"
reviewed and accepted 5/1/00, which was presented at the 2000 Annual Conference of the Air &
Waste Management Association (AWMA).
3.1 Development of Coated Modules
The project approach outlined in the Work Plan was initiated but was soon altered based on early
findings. One significant area of study, which was not the subj ect of the technical papers, was in situ
coating of fibers in Celgard Liqui-Cell modules. Because significant back pressure was required on
the air side of the contactors to prevent oil seeping through the pores, it was deter-mined that coated
fibers were necessary to make the process cost competitive, but acceptable coated fibers were
commercially unavailable. Celgard did not manufacture any contactors with coated fibers, so work
continued on testing the performance and ease of application of several coatings on smaller, bench-
scale Celgard contactors. Significant effort was aimed at identifying or developing a suitably coated
(nonporous polymer coating on a porous polypropylene sub-strate) separation/concentration
membrane module. Discussions with Compact Membrane Systems (CMS) led to the acquisition and
testing of a module coated on the inside of the lumens (hollow fibers) with an amorphous copolymer
composed of perfluorodimethyldioxole and tetrafluoroethylene (PDD-TFE). Although PDD-TFE
would not have optimum transfer charac-teristics, CMS was the only vendor identified and judged
to be capable of applying in situ coatings to the inside of the lumens, and they would agree to work
with only this material. Methyl ethyl ketone (MEK) and w-xylene were the VOCs used to test air
to octanol VOC mass transfer performance. Three conditions were examined in duplicate for each
compound, for a total of 12 experiments. Air flow rate and VOC concentration were experimental
variables while absorbent (octanol) flow was held constant. Experimental conditions were chosen
to emulate previous work with a Celgard Liqui-Cel module containing hollow fibers coated on their
outside surface with PDD-TFE. The results of the tests (see Table 5) were similar to those of
previous tests using a module coated with PDD-TFE on the outside of the lumens. These results
indi-cated that the major resistance to mass transfer in modules coated with PDD-TFE may be in
the coating. Because PDD-TFE does have a relatively high resistance to mass transfer of VOCs, the
need for a better polymeric coating is needed to improve process economics.
10
-------
Table 5. PDD-Coated Membrane Results
Mass Transfer Coefficients
PDD-Coated Membranes, Shell-Side (outside) Coating
Compound
Transferred
Air Stream
Solvent Stream
Concentration
(ppm)
Flow
(L/min)
Loading3
(ppm/s)
Concentration
(mg/L)
K0
(10"5 cm/s)
44
28
120
6.2
0.85
50
130
6.2
0.91
m-xylene
110
28
290
6.2
6.0
0.96
1.0
64
60
370
6.3
1.3
275
1600
5.9
1.6
270
28
720
550
0.7
MEK
750
28
2000
105
2.0
1050
60
6000
1200
9.4
2200
30
5500
980
4.3
PDD-Coated Membranes, Tube-side (inside) Coating
40
28
110
4080
0.1
m-xylene
230
60
1300
4300
2.0
280
60
1600
850
2.7
470
28
1300
8500
-4.0
MEK
2800
60
16,000
9500
2.0
5000
28
13,000
5500
5.3
Removal Efficiencies
PDD-Coated Membranes, Shell-Side Coating
Compound
Transferred
Air Concentration
(ppm)
Air Flow
(L/min)
Loading
(ppm/s)
Removal
Efficiency
(%)
800
105
8000
74
toluene
900
60
5100
76
1300
30
3700
78
500
60
2800
80
MEK/toluene
250
850
60
30
1400
2400
60
80
650
30
1900
70
Other activities aimed at acquiring a suitable coated module followed. Bend Research, Inc., supplied
two prototype hollow-fiber modules for wet testing of the microporous Rayon fibers typically used
in their scalable, high-flow, low-pressure-drop "box module" configuration. Results of the wet
testing indicated that the pore size was too large for use in the system.
11
-------
Chemica Technologies was contracted to coat a sample of the Celgard polypropylene fiber, and
several small patches of coated material were received and tested as a VOC mass-transfer medium
at North Carolina State University (NCSU). The results indicated satisfactory VOC transfer, but the
coating partially delaminated from the substrate. Discussions continued with Chemica to assess their
ability to improve coating adherence to internal lumen walls, but no satisfactory resolution was
reached.
Celgard recommended a polyalkyl sulfone (PAS-16) and supplied some PAS-16 as well as a method
for in situ fiber coating, the only option available for coating the Celgard module. With advice from
Celgard and Anatrace, the PAS-16 manufacturer, a significant in-house effort with to coat the inside
of the lumens with the PAS-16 polymer was begun. The coating attempts were based on Celgard's
suggested method, which involved pumping a dilute polymer solution through the lumens while
pulling vacuum on the shell side of the module. This was followed by rinsing and drying, followed
by annealing the inside of the lumens with warm nitrogen (-60 °C). Variations in solvent,
concentration of polymer, amounts of nonsolvent additives, operating conditions, annealing
procedures, and drying and cleaning protocols were used. In initial attempts using tetrahydrofuran
(THF) as a solvent (suggested by Celgard), the polymer did not dissolve completely, causing
plugging in many of the lumens. Discussions with Anatrace and Celgard provided several new ideas,
which included substituting toluene as the solvent, heating, blending, filtering, and centrifuging to
improve solubility. After several trials, two of the four general approaches tried seemed to show
promise toward achieving the desired results. These four approaches are summarized in Table 6.
Table 6. Fiber Coating Technique Development
TEST
TYPE
SOLVENT
NON-
SOLVENT
PAS-16
CONC.
<%)
VACUUM
(in. Hg)
CLEANING
DRYING
ADVANTAGES
DISADVANTAGES
1
THF
_
0.5-2
0-29.5
TMF/Acetone mix
to Nrortly purge
1-12 psi Nj
20-60 "C
Initial recommended solvent
Compatibte with module mafl
Poor solvent for PAS
Relatively expensive
2
Toluene
—,
1-5
0-29.5
THFWeetone rrax
to Nronty purge
1-12 psi Nj
20-60 t
Good PAS solvent
Can be mixed witn propanol to
precipitate PAS
Drying more difficult
More clogged lumens
Unsatisfactory permeability
results
3
Toluene
Propane!
40-45%
0-10
Mitogen
1-12 psi N2
Fewer clogged lumens
Faster drying
Predictable permeability results
4
Toluene
Propane!
added to
shell side
-2
0
Mitogen
1-12 psi Nj
Precipitated PAS in pores
Difficult to control fluid
permeation
12
-------
To test for complete coating of the porous substrate, Anatrace suggested pressurizing one side of
the membrane module with a mixture of nitrogen and carbon dioxide. Since carbon dioxide is seven
times more permeable to PAS-16 than nitrogen, an analysis of the permeate could be used to
determine whether coating was complete. Since that procedure is fairly difficult and time consuming
to conduct, a simpler surrogate procedure was selected, at least for the initial tests as the coating
procedure was being refined. This procedure is a permeation test in which nitrogen is admitted into
the lumen side of the modules. The flow rate of nitrogen penetrating the lumen walls and escaping
through the shell side was measured. A new module allows ~21 L/min flow at a selected pressure.
Coating attempts conducted during this research exhibited flows ranging from 21 (uncoated) to less
than 0.35 L/min. A module, flow tested at 0.35 L/min of nitrogen, was then tested and found to
perform, in terms of mass transfer rate, much like the PDD-TFE-coated module. Since the PAS-16
coating is known to be more permeable to the VOCs tested than PDD-TFE, coating thickness was
suspected to be excessive. Samples of lumens were cut out and viewed using a scanning electron
microscope. The samples indicated a coating thickness of ~15 to 20 |im on the inside of the lumens;
~1 |im is the goal for coating thickness. Samples of a module that allowed nitrogen permeation at
-0.7 L/min showed a coating thickness of ~7 to 10 |im. From these preliminary tests, it is estimated
that the proper permeability may be in the 1-to-l ,5-L/min flow range to achieve a 1 -to-2-|im-thicl<
coating. It became evident that, although a reasonably successful technical exercise, this path would
be time-consuming and would not validate a cost-effective approach.
Many discussions were held with commercial companies to determine the best approach to
developing a separation/concentration module with a quality nonporous coating and a low-pressure
drop. Proposals were solicited from Celgard, Bend Research, and Applied Membrane Technologies
(AMT) for bench-scale modules and commitments to support the project with future larger-scale
modules. Bend Research and AMT responded with proposals, and AMT was eventually selected as
the manufacturer of choice.
3.2 Membranes Coated with Silicone Rubber
During searches for coated fiber modules, a company (AMT) was found that had developed modules
for several water-stripping applications based on a coated fiber. They were asked to consider our
application, and they offered a small, cylindrical module to be used for initial testing. The parallel-
flow, stainless steel, cylindrical-membrane module was filled with fibers coated on the exterior with
plasma-polymerized silicone rubber at a nominal thickness of 1 |im. Upon inspection prior to
testing, a concern was raised because of the unknown effectiveness of the air-to-fiber contact area.
AMT suggested that contact efficiency issues could be eliminated in a cross-flow module.
Therefore, though the cylindrical parallel-flow design of the existing AMT module did not lend
13
-------
itself to high efficiency, it was used in preliminary testing to gather data for the design of a cross-
flow module.
Five 48-minute tests were conducted with air flowing through the shell side of the module: three
were conducted using w-xylene as the pollutant, and two were conducted with MEK. Results are
presented in Table 7. The airflow was typically 60 L/min, and VOC removal ranged from 56 to 83
percent with average overall mass transfer coefficients, Ka, of 4.4><10"6 to 5.0xl0"5 cm/sec.
Table 7. Air-in-Shell Tests - Cylindrical Parallel-Flow AMT Module
Characteristic
m-xylene 1
m-xylene 1
m-xylene 1
MEK 1
MEK 2
^ir flow (L/min)
60
60
60
28
60
^vg. inlet VOC air concentration
[molar ppm)
65
190
185
186
1350
^veraqe VOC removal (%)
56
77
70
83
78
Average mass transfer coefficient, K0
(cm/sec)
4.40x10"6
8.30x10"6
1.20x10"5
2.10x10"5
5.00x10"5
In commercial operation, one may expect that contaminated air will flow through the shell side of
a cylindrical design while the stripping fluid is pumped through the tube, or lumen side. The initial
set of tests on the AMT cylindrical module was run in this manner. Because of the distri-bution and
contact shortcomings encountered, AMT suggested that a second series of tests be conducted with
the air flowing through the fibers. Therefore, a second set of tests was run with the air flowing
through the fibers and octanol on the shell side. Twelve runs were conducted using w-xylene as the
pollutant. These shorter (34-min) tests were conducted with airflow rates through the lumens ranging
from 5.6 to 10.3 L/min at pressure drops from 11.5 to 20 inches H20 (292 to 508 mm H20). Results
are shown in Table 8. Each mass transfer rate reported in this table is an average of samples taken
at four time points and has a variance of 0.17.
The Ka values (6.Ox 10"7 to 5.1 x 10"6 cm/sec) for this set of runs were consistently and signif-icantly
lower than for the air-in-shell results. As airflow decreases or as inlet concentrations increase,
average VOC removal (overall) increases, but which is affected by other physical factors, may
be impacted negatively. High removal efficiencies (93 and 97 percent) were achieved with octanol
in the shell, and high mass transfer coefficients (2.1 x 10"5 and 5.Ox 10"5) were achieved with air in the
shell.
14
-------
Table 8. Octanol-in-Shell Tests - Cylindrical Parallel-Flow AMT Module
Parameter
MX7
MX8
MX9
MX10
MX11
MX12
MX13
MX14
MX15
MX16
MX17
MX18
^ir-side pressure drop [in. (mm) H20]
11.5
(292)
11.0
(279)
11.5
(292)
11.0
(279)
16.0
(406)
16.0
(406)
16.5
(419)
16.0
(406)
20.0
(508)
20.0
(508)
20.0
(508)
20.0
(508)
^ir flow [L/min]
5.6
5.6
5.6
5.6
8.6
8.6
8.6
8.6
10.3
10.3
10.3
10.3
^vg inlet VOC air concentration [molar ppm]
68
84
261
684
92
125
457
499
76
105
274
697
Average VOC removal [%]
91
91
80
97
51
44
93
89
60
52
73
85
Average K0 [cm/s]
1.60
x10"6
1.90
x10"6
8.20
x10"7
2.30
x10"6
8.60
x10"7
6.00
x10"7
3.60
x10"6
2.50
x10"6
5.10
x10"6
1.20
x10"6
1.60
x10"6
2.00
x10"6
-------
One problem encountered in all extraction experiments was swelling of the membrane material.
Occasionally, this was accompanied by "sweating" of the octanol through the membrane. It will be
necessary to evaluate alternate stripping fluids as a means to ameliorate this problem.
The information from this testing was used in the decision to develop a cross-flow module designed
and manufactured specifically for this project by AMT (See Figure 4). This module contained
roughly 2.3 m2 of available membrane surface packaged in a module with air contact dimensions
of roughly 3.5 x 10 x 1.0 inches (88.9 x 254 x 25.4 mm). Manufacturing methods required
manufacturing in pairs, so the minimum two modules were procured.
Differences in design between the cross-flow and radial
modules required a different potting material to seal the
fibers into the end caps and plans were to use a urethane.
This urethane material had not been previously used in
applications with octanol, so limited tests were done to
determine chemical compatibility. Initial results were
only partially successful, showing softening of the
urethane in longer periods of exposure. Using heat
accelerated curing and longer setup times improved
performance. These concerns and subsequent extended
testing trials delayed the beginning of VOC tests. The
urethane manufacturer concurrently investigated these
issues, but they were unwilling to develop entirely new
compounds and were unable to create small batches of
existing materials to test. Few options were readily
available.
Figure 4. AMT Cross-Flow Module -
Overall Dimensions in Inches
After assembly of the modules, they were pressure-tested
to prove compatibility with octanol. This was expected
to be an easy test to pass because no compatibility issues
surfaced with the radial module. However, during the test, elongation swelling of the polypropylene
fibers was encountered in the new module. At this point, further testing was done with single fibers
to prevent destroying the remaining module. It was also determined that, after removing the octanol
and cleaning the module, the fibers returned to their original length. A number of tests were
completed with single fibers exposed to octanol, and it was found that the stretching was limited to
about 5 percent of total length. Three different manufacturers' fibers were tested; all produced
similar results.
Disassembly and testing by AMT resulted in destroying one of the two modules that were built.
16
-------
While a solution to the fiber stretching was being sought, the other module was received from AMT
for inspection and airflow testing to determine how the cross-flow module performed in comparison
to the goal of low pressure drop. The module was mounted in the custom steel transition sections,
and a variable-speed blower and electrical controls were connected for the test. The results indicate
that the module was able to pass 20 ft3/min of air at 0.62 inch H20 pressure drop and 72 ft3/min at
less than 4.0 inches H20 pressure drop. Limitations of the test apparatus prevented using higher air
flows. The goal for flow vs. pressure drop was 10 ft3/min at less than 20 in H20, so this design
greatly exceeded this goal. With such a low pressure drop, multiple modules could be arrayed in
series to increase VOC transfer performance, if needed. Inspection of the module and velocity
profiles measured at the face of the fibers indicated that some bypass was created on the long sides
where the fibers lay parallel to the polycarbonate housing. To address this problem, AMT proposed
to create a seal along the two sides to limit the bypass. The fibers also exhibited some tendency to
vibrate during the test, especially at the midpoint of the fiber bundle. AMT provided a modification
to reduce or eliminate this vibration. At the end of the test, the module was returned to AMT for the
modifications.
As indicated in Figure 5, the air pressure drop is negligible for this module, even at high flow rates.
These results represent an order-of-magnitude improvement in pressure drop performance, relative
to the cylindrical module containing identically coated fibers.
0
CM
1
Q.
O
¦O
3
V)
U)
CD
4.5
4
3.5
3
2.5
2
1.5
1
0.5
0
100
80
60
40
20
0
10 20 30 40 50
Flow rate (cfm)
60
70
80
O
CM
X
E
£
Q.
O
w
T5
CD
O
1 cfm = 0.028 m /min
Figure 5 Pressure Drop for AMT Cross-Flow Module
17
-------
The two approaches were available to address the fiber-swelling problem are (1) modify the housing
in such a way that its length could be adjusted after the initial exposure to octanol, thereby re-
tensioning the fibers, and (2) to use a stripping fluid other than octanol. The first method for fixing
the stretching problem would involve redesigning the module to allow for post octanol length
adjustment; this was abandoned due to complexity and an increased leakage risk. Alternate stripping
fluids, including silicone oil and canola oil, were considered. Canola oil was seen as the most-
economic alternative, but it lacked the chemical purity needed during analysis for extracted
compounds. AMT then tested the fiber material with silicone oil in the same manner that had been
used to measure the length changes after exposure to octanol. Single-fiber tests showed that fiber
stretching was not evident after exposure to silicone oil, so a decision was reached to pursue
switching to silicone oil. A change to silicone oil required repeating some of the initial octanol
testing to verify partition coefficients at different temperatures for the com-pounds of interest.
Potential toxicity to the bacteria also required investigation.
The AMT cross-flow module was returned for slight modifications and more tests using silicone oil
in place of the octanol. The tests using silicone oil were successful; no swelling was seen, as had
occurred in the case of octanol. AMT then returned the module after modifications were made, and
a final decision was reached that further work on the separation contactor module would be done
using silicone oil stripping fluid. An oil of viscosity slightly less than that of octanol (5 cs) was
ordered and received. The single module was set up in a new apparatus for bench-scale testing using
a simulated paint stream.
3.3 Experimental Structure
The bench scale S/C and biomembrane units were evaluated separately during the testing of the
cross-flow separation contactor module. During the bench-scale testing, oil doped with the four
target compounds was used in biomembrane effectiveness studies at N.C. State. During the final
three months of the project, the used oil from the pilot-scale testing separation contactor system at
ARCADIS was taken to N.C. State for biomembrane testing.
VOC mass transfer experiments require real-time sampling and analysis of VOC concentrations in
inlet and outlet air and sampling of the stripping-fluid-reservoir VOC concentration for offline
analysis. The VOC stream was created by a custom system configured as shown in Figure 6. The
VOC-laden air stream is created by first injecting the four-component liquid mixture into a fitting
that is heated by a small electric heater. This forces complete volatilization of the mixture before
it is progressively diluted with filtered air until finally being force-mixed by a fan in the mixing box.
During the mass transfer experiments, module inlet-stream VOC concentration was set via
adjustments to the syringe pump delivery and airflow rates. Silicone oil flows were adjusted by a
18
-------
GC ANALYZER
OIL SAMPLING
PORT
FLOW-METER
OIL PRESSURE
GAUGE
SILICON OIL
TANK
OIL PRESSURE
GAUGE
OIL PUMP
MEMBRANE
HEAT
TRACING
FILTER
HOLDER
SAMPLING
PORT
COUPLING
SAMPLING PORT
CLEAN AIR
SWEEP
COUPLING
MIXING
CHAMBER
SYRINGE
PUMP
TRANSITION
TRANSITION
BLOWER
FLOW
PROBE
Figure 6. Separation Module Bench-Scale Test Apparatus
pump-speed setting and were read on a rotameter. The system was allowed 5-10 minutes to stabilize
before testing began. Once the experiment began, samples of the air at the inlet and outlet of the
module (two air samples) were taken at approximately 2%-minute intervals and analyzed online by
a gas chromatograph with a flame ionization detector (GC/FID). Oil samples were also withdrawn
via syringe from an oil-sampling port located between the module outlet and the oil reservoir. These
oil samples were taken at 5-10-minute intervals. The oil samples were then stored, headspace-free,
at 3 °C for later analysis by GC. Analysis of air-side data resulted in a separation contactor removal
efficiency, and oil-side analysis permitted a mass balance analysis of the system and determination
of the oil condition from test to test.
3.3.1 MBT Bench-Scale Separation Contactor
A schematic of the bench-scale system separation contactor is shown in Figure 6 and a process flow
schematic is presented as Figure 7.
19
-------
f CLEAN/AIR
ayrrin
AIR
cfm
T SEPARATION
FLUDOOT
— RLTER t
I N_ET OJTLET
SAIVPLE SAJVPLE
PCRT PORT
i
SfRNGE
PUVP
HEATED \
v 5
TEE
MXING
BCK
IVEIVBRAISE
MODULE
BLOAER
SEPARATION
FLUID IN
VENT
Figure 7. Process Schematic of Separation Module Bench-Scale Apparatus
3.3.2 Pilot Testing
A schematic of the complete MBT system appears as Figure 1 in Section 1.1.1. A process such as
a paint booth produces a dilute stream of VOCs in air. After particles are filtered from the stream,
it enters an S/C unit, which employs bundles of microporous hydrophobic fibers with the VOC-
laden air and stripping fluid flowing across and through the fibers, respectively. The VOCs are
transferred from the exhaust gases into the stripping fluid medium. The stripped air that leaves this
unit is taken to an exhaust stack. After the circulating silicone stripping fluid leaves the S/C unit, it
is delivered to a biomembrane unit is stored in an intermediate storage vessel. In the biomembrane
unit, the stripping fluid is circulated past one side of another microporous membrane module that
has a film of VOC-degrading bacteria on the opposite side of the membrane. VOCs diffuse through
the membrane pores and are selectively and completely metabolized by the bacteria. The stripped
solvent is then collected in a storage vessel and recycled through the S/C unit. Outputs from the
overall MBT system are clean air, carbon dioxide, and a mixture of water and nonhazardous cell
mass.
20
-------
A diagram of the pilot-scale system is shown in Figure 8. The pilot-scale system differed from the
bench-scale description in a number of details, including that the system was much larger and that
the stream was created from spraying actual coatings from a paint spray gun pointed at a target in
a small tabletop paint booth. The coating used for these experiments were acquired from an Air
Force refinishing facility, and its components were determined from Material Safety Data Sheets
(MSDS). However, paints are not composed of pure compounds, and industrial versions of listed
compounds (e.g., MEK) are typically a mix of many compounds. Also, com-ponents below 1
percent are not required to be listed on an MSDS. Therefore, the capture efficiency was measured
by using a total hydrocarbon analyzer on the upstream and downstream air streams rather than by
speciation with a GC/FID. For these same reasons, a hydrocarbon analyzer is the instrument
typically used to measure paint booth control equipment destruction efficiency in commercial
operations. Based on previous work at NCSU, it was assumed that the separation contactor module
could display selective removal behavior. To look for module compound preferences, samples were
also taken for analysis with a GC/mass spectrometer (MS). These samples were taken on charcoal
sorbent tubes at input and output locations adjacent to the sampling ports for the total hydrocarbon
analyzer and were analyzed against a list of typical paint compounds, including those listed in Table
9. The oil analysis was done the same as for the bench scale. Testing followed the matrix outlined
in Table 10.
it'-..
Figure 8. Pilot-Scale System Design
21
-------
Table 9. Tertiary Mixtures in the Air Stream
Mixture Component
% VOCs in Paint
by Volume
% VOCs in Paint
by Weight
Bench Scale
Xylene
8.7
7.5
MEK
18.6
15
Ethvlbenzene
2.9
2.5
Butyl Acetate
11.3
10
Pilot Scale
Paint (AerosDace Coatina')
63.95
50
Table 10. Typical Flow Rates Used During Testing
Test Air Flow
(ft3/min)
Input VOC Concentration
(ppm)
100
200
400
Bench Scale
2
X
X
X
4
X
X
X
8
X
X
X
16
X
X
X
Pilot Scale
20
X
X
X
100
X
X
X
200
X
X
X
22
-------
4.0 Results/Data
The following sections present the results of the experiments performed during this study. Pertinent
data from bench- and pilot-scale tests can be found in Appendices B and C, respec-tively.
4.1 Separation Module Tests
Separation module tests, described below, was performed in a number of bench- and pilot-scale
modules.
4.1.1 Bench-Scale Tests
Eighteen successful tests, summarized in Table 11, were completed using a variety of air flow rates,
VOC loadings, and oil conditions. Average removal efficiencies varied from 4.4 to 73.7 percent, and
peak removal rates exceeded 79 percent twice for ortho-xylene. A question was raised about the
performance of the GC after test nine, so the steps were expanded to include pre and post test
calibrations. Preparation of calibration standards had also been inconsistent prior to this test, so
alternate methods were found to improve consistency. For this reason, there is more confidence in
the results from tests 12 and higher, but tests did just as well with 17 percent of the loading.
Nevertheless, the results show important trends and general information. Higher loadings, such as
in Test 2, improved inlet/outlet removal efficiency. Typical paint booth operations will have
emissions in the range of 200 to 350 ppm, so most of the tests were run within that range. Higher
air flow rates were used in a few tests to determine if increased turbulence improved performance,
but it appeared that any such performance increase was negated by shortened contact time, resulting
in lower removal efficiencies. Two modules arranged in series improved performance over a single
module. Higher oil flow resulted in higher removal rates, but leaks resulted if this flow were set too
high, indicating that materials and the module structure required improvements. Although overall
removal efficiencies for individual compounds were as high as 79 percent, VOC removal rates were
highly variable; MEK removal was often well below acceptable values, especially considering its
prevalence in military coatings. Even though MEK removal was often at or close to zero, the other
three compounds were consistently removed and were relatively comparable from test to test.
The first module received from AMT was the one not damaged during the initial testing. After a few
trial runs with the lab apparatus, the module was tested for VOC removal efficiency. As a starting
point, pressures and flows were chosen based on previous bench-scale investigations atNCSU using
the uncoated fibers of the Celgard module. Oil flow was set at 1.0 L/min and air flow at 2.0 ft3/min.
It became apparent after two tests that the module was leaking oil, and swelling of an internal
silicone rubber bridge was creating packs and voids in the fibers. This substantially reduced the
23
-------
Table 11. Summary of Bench-Scale VOC Tests
Test
Date
Number
of
Modules
Air Flow
(ft3/min)
Oil Flow
(L/min)
Approx. Inlet
Load
(PPm)
Tests
Since
Last Oil
Change
Removal Rate
(%)
MEK
n-Butyl
Acetate
Ethyl-
Benzene
o-Xylene
Avg.
1
4/27/00
1
2.0
1.0
450
0
18.5
47.9
19.6
50.4
34.1
2
4/27/00
1
0.5
1.0
2100
1
66.5
74.9
70.8
79.2
72.9
3
4/29/00
1
2.0
1.0
450
2
1.2
31.5
33.8
42.6
27.3
4
4/29/00
1
2.0
1.0
230
3
-5.1
14.7
12.5
30.8
13.2
5
4/29/00
1
2.0
1.0
850
4
19.2
41.0
44.1
48.9
38.3
6
5/03/00
1
2.0
0.2
300
5
13.0
34.5
42.8
43.4
33.4
7
5/05/00
1
2.0
0.2
170
6
18.6
31.8
38.9
43.2
33.1
8
5/06/00
1
2.0
0.2
450
7
26.3
44.4
50.0
51.9
43.2
9
5/09/00
1
2.0
0.2
350
8
58.0
22.4
27.2
28.6
34.1
10
5/11/00
2
4.0
0.4
350
0
23.5
54.2
58.5
62.3
49.6
11
5/12/00
2
4.0
1.0
350
1
63.5
75.6
76.2
79.4
73.7
12
6/06/00
2
4.0
1.0
200
4
-5.1
29.9
38.9
45.1
27.2
13
6/07/00
2
2.0
1.0
220
5
2.9
34.9
42.4
50.9
32.8
14
6/08/00
2
8.0
1.0
200
6
0.29
9.8
13.4
18.4
10.47
15
6/21/00
2
4.0
1.4
200
0
9.6
40.3
45.1
51.8
36.7
16
6/22/00
2
16.0
1.4
200
1
-5.6
5.5
6.3
11.5
4.4
17
6/22/00
2
8.0
1.4
200
2
-1.1
13.6
16.4
23.5
13.1
18
6/23/00
2
4.0
1.4
200
3
0.96
27.3
29.4
35.5
23.3
-------
contact efficiency and performance was deteriorating, so testing was stopped after Test 5. An
inspection of the module was unable to locate the exact source of the leak, but previous experience
had shown the corners where the module sections were assembled to be suspect. A problem with
the urethane material was assumed to be the culprit. Two modules received later were constructed
similar to the first two, but an alternate potting material was used so that testing could continue
while the urethane problems were investigated further. Tests with these modules were started with
a lower oil pressure and more conservative flow rates (0.2 L/min) to prevent leakage failures.
Removal efficiencies were lower during the next tests, most likely because of the lower driving
forces from the reduction of oil pressure and flow. A number of modifications were made to the
return piping on the apparatus to reduce back pressure in hopes of increasing oil flow without raising
the static pressure. This was successful in bringing flow up to 0.4 L/min with a back pressure of
about 5.0 psi. Results showed an obvious increase in removal efficiency, although still lower than
the target of 80 percent. The two modules were then arranged in parallel on the oil side and in series
on the air side to double the contact time without risking higher oil pressures. The efficiency again
increased, but not as dramatically as had been achieved with higher oil flow rates, so the next step
was to increase the oil flow and pressure. Again, plumbing modifications were made to reduce back
pressure, and flow was increased to 1.0 L/min with an increase in pressure to 12.5 psi. Although
removal efficiency increased, it quickly became apparent that both modules were again leaking oil.
Based on these trials, AMT proposed a number of design modifications to be used in the pilot-scale
version to prevent the leakage problems of the corners and joints.
4.1.2 Pilot-Scale Tests
Pilot-scale testing produced several tough challenges that reduced the number of successful tests.
Initial runs were hampered by inconsistent spray gun paint application, although the apparatus was
designed to maintain a constant trigger position. In addition, only the last five tests were run with
both air-side analytical systems. Upon data review, the comparability of these two analyti-cal
methods and its implications are important enough to focus primarily on these results, shown in
Table 12. The overall removal rate is shown by averaging the GC/MS results for the eight listed
compounds and by the average removal reported by the total hydrocarbon analyzer. For all five tests,
both methods passed pre-and post-analysis calibration checks. However, agreement was achieved
only for Tests 1 and 3, with 7.4 and 16.9 percent differences, respectively. Comparability for Test
5 was lower at 35.6 percent difference, and Tests 2 and 4 show gross differences. Similar trends
were seen for the bench-scale module in that MEK shows much lower results compared to the other
compounds, and removal generally improves with higher oil flow and lower air flow.
To understand the removal results requires a short review of the methods used. During bench-scale
testing the air-side analytical method (GC/FID) agreed closely with the oil-side analysis method in
25
-------
Table 12. Summary of Pilot-Scale VOC Tests
Test/VOC
Load
(PPm)
Air Flow
Rate
(ft3/min)
Oil Flow
Rate
(L/min)
VOC Removal
(%)
Individual Constituents
Average
MIBK
Xylene
2-
Heptanone
MEK
n-Butyl
Acetate
Ethyl-
benzene
GC/MS
Total HC
Analyzer
para
meta
ortho
1/300
200
4.6
49.5
55.2
53.0
33.5
59.9
33.9
49.8
55.7
48.8
45.3
2/260
100
3.85
69.9
73.9
73.9
50.5
77 A
63.2
71.6
73.9
69.3
0.0
3/310
100
1.8
38.6
50.4
47.2
45.2
54.1
35.0
46.3
48.3
45.6
54.0
4/225
100
6
78.6
81.7
81.5
73.8
84.2
75.2
81.3
81.3
79.7
20.0
5/260
140
4.4
26.0
40.8
37.6
36.5
45.2
17.6
33.4
38.1
34.4
24.0
-------
total mass balance. This was largely due to using four pure compounds to create the pollutant
stream, and each method could differentiate and quantify the known four compounds. The change
to pilot-scale brought the use of real paint to simulate a paint-booth exhaust. It was known
beforehand that even compounds listed on the MSDS, such as MEK and xylene, would be impure
compounds that would have significant variations in constituents. For this reason, the air-side
analysis was changed to a total hydrocarbon analyzer because it would be capable of analyzing all
known and unknown VOCs. Because no better alternative existed, the oil-side analysis was
unchanged and analyzed for only the same four constituents identified during the bench-scale
testing. An additional analysis was added for the air side with the knowledge that the module could
turn out to be selective in its removal rate; additional samples were taken parallel with the total
hydrocarbon analyzer to be analyzed by EPA Method 25. These carbon tube samples were extracted
and analyzed on a GC that is normally used for identifying unknown compounds in coatings
formulations. A sample run identified eight compounds on the standard search list.
Pilot-scale test results were variable and, at times, difficult to understand. Initial test results using
the total hydrocarbon analyzer showed inlet and outlet results varying from 20 to 65 percent
removal. Oil analysis, however, almost never agreed with those removal rates, and during later tests,
when charcoal tubes were analyzed by GC/MS, averaged removal rates based on the eight
constituents were always reported higher than those with the other two methods. To understand this,
it is important to remember that each method was different and comparability was not a given. The
total hydrocarbon analyzer measures total carbon atoms, and the relationship between its readout
in parts per million and the other methods requires the use of a multiplier to account for the more-
complex molecular structure of organic compounds. Regardless, with any reasonable factor (e.g.,
4.0 for typical paint compounds), the data from GC/MS barely exceed 1 percent of the reading
produced by the total hydrocarbon analyzer. This implies that the eight compounds that were
identified by GC/MS represented only 1 percent of the total constituent VOC content of the paint,
and while the module was relatively efficient at removing some compounds, especially those
identified by the GC/MS, it was unable to remove other compounds, especially after the oil had been
used. The total hydrocarbon analyzer reported an overall VOC removal rate that was comparable
to the rate reported by GC/MS at times and not comparable at other times.
More problematic was that the oil analysis data rarely reflected the air side removal. In some cases,
the oil showed no removal, while the air side showed removal of 20 percent or more. One
contributing factor is that the oil analytical procedure was developed during the bench-scale studies,
and calibrations were based on removal of four, well characterized, pure compounds rather than the
eight identified in the GC/MS samples, or the many unknown compounds contributing to the total
hydrocarbon analyzer data. However, with even the four known compounds, the GC/MS data did
not always agree with the oil data. A possible explanation is that much of the stripped VOCs did not
27
-------
pass through the fiber boundary but, rather, was adsorbed and temporarily held both on the wetted
outer surface of the fibers and by the static oil that had formed a puddle inside the apparatus during
the tests. Contributing to this theory is a daily operational aspect. Each time a test was completed,
the air fan was left running to prevent migration of the paint vapors to office areas adjoining the
laboratory. During this time, the oil outside the fibers that had adsorbed VOCs was thus desorbed,
and this cleaned oil was again the dominant remover during the next test.
To treat the needed 500 ft3/mim, 20 modules were ordered with the latest in design specifications
of the bench-scale module: 2.3 m2 area and 3.5 x 10 x 1.5 inches (89 x 254 x 38 mm). Although
scaling up by using many small modules was not an ideal approach, project time requirements for
delivery required using existing production methods, and this was the largest size that could be made
with existing production equipment. To reduce bypass around the side, a shelf area was added to
overlap the fibers, and the silicone rubber bridge that became distorted on the first bench-scale
version was dropped. One advantage of the small modules was that, to suit various test scenarios,
they could be arranged in a range of patterns such as multiple banks of module for increased contact
time or parallel arrangements for greater air flow rates. The rest of the apparatus (module mounting
frames, ducting, transitions, and oil piping) was designed with this in mind, and to reduce assembly
time, the transitions and mounting frames were constructed while the modules were in production.
After the 20 modules were delivered, a number of tests were done to determine their basic integrity.
Although AMT tested the modules under pressure with air, no tests had been done with silicone oil.
Therefore, each module was to be unpackaged, visually inspected, and then tested under 5 psi oil
pressure. Based on visual inspection, several of the modules exhibited question-able potting material
with bubbles and extrusions. The bubbles were due to the urethane curing at humidity that was not
ideal. Two with obvious porosity problems were chosen for pressure testing, and each leaked almost
immediately upon start up. This was seen as serious enough to warrant stopping further tests while
AMT addressed the problem. After these were sent back, AMT determined that one was damaged
in shipment and could be repaired. To conserve the short time remaining in the project, a decision
was reached to test some of the remaining modules. The apparatus was easily able to test 10 modules
arranged in two series-arranged banks. Therefore, they were assembled into a reduced-size, pilot-
scale apparatus so testing could begin. This required a reduction in airflow from the planned 500
cfm. The maximum treated stream was 200 cfm.
Meanwhile, AMT was unable to locate the leak in the second returned module. The leak that had
been seen during receiving tests was the result of a slow, but continuous, sweating of oil through the
pores of the fibers. This suggested a difference between AMTs tests and the receiving tests; the
receiving tests had used silicone oil with a viscosity similar to octanol, or about 5 cs, whereas
AMT's tests were done with the more commonly available 50-cs oil. Further study revealed that the
28
-------
thinner silicone oil was acting more solvent-like; it was seeping through the pores of the fibers, both
wetting the outside of the fibers and accumulating, over time, in a puddle at the bottom of the
module. Based on this information, AMT advised that further tests should be limited in pressure and
oil flow rate.
After a number of tests with the 10-module arrangement, the apparatus was disassembled for a
visual inspection, and it was apparent that wetting the fibers had created a problem that had
previously gone unnoticed. The fibers were again distorted much like had occurred during the
bench-scale testing. It was then determined that the distortion was not due just to swelling the
silicone rubber bridge that had been deleted from the design but was also due to individual fibers
swelling much as had occurred in the octanol compatibility tests. The thinner, 5.0 cs oil had acted
on the fibers much like an aggressive solvent, and the stretching was creating voids and packed
areas. Therefore, the modules had an unknown, but certainly not optimal, contact efficiency. At this
time, AMT recommended that we use the remaining 10 modules in another test using the thicker,
50 cs oil. Their reasoning was that the thicker oil would be unlikely to seep through the fibers. After
assembling the remaining 10 modules, a shakedown test was started. To attain the same flow rate
that was used in previous tests required a much higher pressure (about 30 psig) setting, and one or
more modules ruptured, causing a catastrophic loss of oil, and the testing was stopped. In retrospect,
while thicker oil would be less likely to seep through the fiber, its viscosity requires much higher
pressure for any given flow rate. To prevent failure of the urethane potting material would have
required a much lower flow rate, and with lower flow rates, the driving forces would be reduced,
so any advantages may have been lost. Contaminated stripper fluid from the pilot plant was
transferred to the NCSU laboratory for bio-treatment.
4.2 Biological Treatment System
The experiments related to the biological treatment system can be organized into four major
categories of suspended cell experiments performed in shake flasks, biofilm experiments performed
in flat-sheet membrane units, biofilm experiments performed in hollow-fiber membrane units, and
experiments involving hollow-fiber units in series with a stirred tank reactor. These categories
represent qualitatively different objectives.
• Shake flask (suspended-cell) experiments determine organism and mixed-culture
characteristics,
• Flat-sheet biofilm experiments determine the basic behavior of pure- and mixed-culture
biofilms,
• Hollow-fiber-module experiments determine the robustness of the degradative activity and
evaluate biofouling control schemes, and
• Experiments involving the hollow-fiber module operated in series with a suspended-cell
29
-------
reactor demonstrate the capability to simultaneously degrade aliphatic and aromatic VOCs.
Characterization of the performance of the bacterial consortium on mixed VOCs is confounded by
three facts: (1) growth and degradative behaviors of a mixed culture cannot be expected to be the
sum of the behaviors for each organism in the culture grown separately; (2) immobilized-cell
(biofilm) growth kinetics often deviate significantly from suspension-culture (shake-flask) growth
kinetics; and (3) growth of a single organism on a mixture of compounds is not easily related to
growth on the individual compounds.
The complex interactions between species and compounds must be identified and quantified before
one can effectively design and operate a biotreatment module. To adequately describe the behavior
of this system, a series of suspended-cell-growth studies and membrane-supported-biofilm-reactor
experiments were completed. Complete data sets for each experiment can be found in Appendix B.
4.2.1 Suspended-Cell Experiments
A significant number of growth studies were performed during the course of this research, many in
response to the degradative behavior exhibited by the mixed-biofilm reactor system. The first set
of experiments to be described pertain to the need to identify degraders for compounds found in
MilSpec paints; specifically w-xylene, ^-xylene, ethylbenzene, butyl acetate, butanol, MEK and
acetone.
4.2.1.1 Screening Studies
Original screening studies were performed to identify organisms capable of degrading m-xylene, and
two organisms were ultimately established with those capabilities, respectively designated MX-1
and XI. Subsequent screening studies were undertaken to establish degradation of ^-xylene,
ethylbenzene, butanol, butyl acetate, and acetone.
4.2.1.2. Xylene Degraders
The first of these studies involved the screening of MX and XI for growth on /^-xylene and
ethylbenzene. This was done to select one of the two different strains for further studies.
Methods: Cultures were grown under sterile conditions in 250-mL flasks incubated at 30 °C in a
shaker (250 rpm). The flasks contained 50 mL of L-salts medium; individual VOCs were added as
liquid in quantities of 5-7.5 |iL. The flasks were covered with foil and sealed with Parafilm before
incubation. Optical density of the culture at 600 nm was monitored for up to 72 hours to detect
growth. The initial concentrations of/>xylene and ethylbenzene were 150 ppm and 100 ppm, respectively.
30
-------
Results: Both organisms were able to grow on /^-xylene, but only XI grew on ethylbenzene. Based
on these results, X-l was chosen for the staged biotreatment experiment.
4.2.1.3 Aliphatic Degraders
Two organisms that had been isolated previously, designated Ml and AOC-1, were screened for
growth on butanol, butyl acetate, MEK, and acetone. This was done to select the better aliphatic-
compound degrader of the two strains for the staged biotreatment study.
Methods: Cultures were grown under sterile conditions in 250-mL flasks incubated at 30 °C in a
shaker (250 rpm). The flasks contained 50 mL of L-salts medium; individual VOCs were added as
liquid in quantities of 5-7.5 |iL. The flasks were covered with foil and sealed with Parafilm before
incubation. Optical density of the culture at 600 nm was monitored for up to 72 hours to detect
growth. Initial concentrations of all substrates were 150 ppm.
Results:
Organism
MEK
Acetone
Butanol
Butyl Acetate
Ml
+
+
+
+
AOC-1
(-)
+
+
+
Ml grew on all four substances. AOC-1 grew on acetone, butanol, and butyl acetate; but growth on
MEK was very uncertain. Therefore Ml was chosen for the staged biotreatment experiment.
4.2.1.4 Growth Studies
Many growth studies were carried out in conjunction with screening studies to quantify growth rates
and extents of VOC degradation. These growth studies involved the organisms MX, XI, and Ml.
Table 13 summarizes the experiments performed over the course of the project. Only the
experiments that have not been described in previous reports or publications will be commented on
directly; the results of all the experiments can be found in the appendices. Although each experiment
involved specific organisms and substrates, a general methodology was employed for culture growth
and characterization.
Culture growth: Growth flasks were prepared under sterile conditions in screw-top flasks. L-salts
were added to the flask, and the headspace was filled with oxygen. The substrate was added, and
inoculation was effected by adding a sufficient amount of an overnight culture to a total liquid
volume of 150 mL. The flask was sealed with an open-top closure with a PTFE-coated septum, and
a needle pierced through the septum connected to a Teflon valve. A control flask was prepared in
a similar way but without inoculation. Liquid was withdrawn by use of a glass syringe for optical
31
-------
Table 13. Suspension-Culture Experiments
Experiment No.
Organisms
Substrates
Rationale
3S1
X1
m-Xvlene
Strain characterization
3S2
X1
Toluene
Substrate ranqe
3 S3
X1
m-Xvlene, toluene
Metabolic requlation
3S4
M1
Toluene
Substrate ranqe
3S5
M1
MEK, toluene
Metabolic requlation
3S6
M1
MEK, m-xvlene
Metabolic requlation
3S7
M1
Toluene, m-xvlene
Substrate ranqe
3S8
M1
MEK, toluene, m-xvlene
Metabolic requlation
3S9
M1, X1
MEK, toluene
Metabolic requlation
3S10
M1, X1
MEK, toluene, m-xvlene
Orqanism interactions
3S11
MX
p-Xvlene
Substrate ranqe
3S12
X1
p-Xvlene
Substrate ranqe
3S13
X1
m-Xvlene, ethvlbenzene
Metabolic requlation
3S14
X1
m-Xvlene, p-xvlene
Metabolic requlation
3S15
M1
Butvl acetate
Substrate ranqe
3S16
M1
MEK, butyl acetate
Metabolic requlation
density (OD) and VOC measurements. During the sampling, oxygen was added to the flasks to
compensate for the volume withdrawn. The control flask was treated in the same way, but the
samples for OD were discarded.
General analyticalprocedure: The sample withdrawn for VOC analysis (1 mL) was transferred into
a 2-mL vial containing 1 drop of acetic acid to acidify the sample. Extraction of the VOCs was
carried out by addition of pentane (0.1-0.2g) containing heptane as internal standard. The sample
was shaken vigorously for 2 minutes and stored upside down at 5 °C. Analysis was performed with
a Hewlett-Packard 5890 gas chromatograph (GC) equipped with a J&W Scientific capillary column
(DB-624) and a FID detector. A l-|iL portion of the pentane phase was injected into the GC, and
the response was compared to the response from standards prepared in pentane.
4.2.1.5 MX and X1 Growth on p-Xylene (GS 11, 12)
The objective of these experiments was to study the growth and degradation of /^-xylene by the m-
xylene-degraders in order to choose the best organism for xylene degradation in the staged
biotreatment unit. The substrate (/^-xylene) was added so that initial concentrations varied from
100-150 ppm.
Results:
32
-------
Organism
Growth Rate
(h1)
MX
0.29
XI
0.49
The inoculum was grown on m-xylene, so not surprisingly both organisms exhibited a long lag phase
(20-30 hours). The growth rates indicate that XI degrades /^-xylene more rapidly than does MX.
Although suspension culture behavior is not completely indicative of the behavior of organisms
when attached to a surface, XI was selected for further evaluation.
4.2.1.6 Effect of Ethylbenzene on Growth of X1 on m-Xylene (GS13)
The objective of this experiment was to determine whether ethylbenzene has a positive or negative
effect on the rate of degradation of m-xylene by XI. This would indicate whether the pathways
involved in the degradation of these two compounds were synergistic, antagonistic, or non-
interactive.
Methods: Growth flasks were prepared as described above. The initial concentration of m-xylene
was 150 ppm in all flasks. Initial concentrations of ethylbenzene were varied: 0 ppm, 50 ppm, 100
ppm, and 150 ppm. The flasks were closed as described above. Optical densities were followed for
26 hours.
Results: Ethylbenzene appeared to have very little effect on the growth rate. It seems that the initial
lag phase was shorter in the presence of ethylbenzene, indicating that the presence of ethylbenzene
might facilitate w-xylene degradation. Another possibility is that ethylbenzene was degraded as well.
However, no VOC analysis was undertaken.
The presence of ethylbenzene in high concentrations (150 ppm) did affect the apparent growth yield
obtained after 26 hours. For an initial concentration of ethylbenzene of 150 ppm, the final optical
density was 0.17 compared to 0.24-0.26 for cultures exposed to lower concentrations of
ethylbenzene. The growth had reached a stationary growth phase in all the flasks, which was
confirmed by a sample point after 32.5 hours. The lower OD level for the high concentration of
ethylbenzene indicated that the growth of XI on w-xylene was inhibited rather than ethylbenzene
was degraded itself. For further study of the effect from ethylbenzene, VOC analysis should be done
to determine if ethylbenzene is being degraded. The table below shows the growth rates calculated
for each concentration of ethylbenzene, and the optical densities of each culture after 26 hours.
33
-------
Concentration
of Ethlybenzene
(ppm)
Growth Rate
(h1)
Period for
Exponential Growth
(h)
OD after
26 hours
0
0.23
11 - 19
0.24
50
0.20
7-16.5
0.26
100
0.24
7-16.5
0.24
150
0.22
4-16.5
0.17
4.2.1.7 Growth of X1 on /r?-Xylene and p-Xylene (GS14)
The objective of this experiment was to study the growth and degradation of mixtures of m- and p-
xylene by XI. Again, the concern was that there might be negative interactions between the two
degradative pathways or intermediates generated therein.
Methods: Growth flasks were prepared as described above; m-Xylene and /^-xylene were added to
a concentration of 75 ppm each.
Analytical procedure: It was not possible to separate w-xylene from /^-xylene, so a total
concentration of the two xylenes was calculated.
Results: The growth curve indicated that the two substrates were degraded concurrently. The growth
took place within 10 hours, and no lag phase was observed. After 9 hours, m-xylene and /^-xylene
could not be detected in the liquid. The growth rate obtained in this study is compared to growth
rates determined in previous studies of XI growing on w-xylene and /^-xylene as single substrates.
Results show that the growth rate obtained in the presence of both substrates is lower than the one
obtained when only w-xylene is present, but higher compared to the growth rate obtained for p-
xylene.
Substrate
Growth Rate
(h1)
m-Xylene
0.78
/(-Xylene
0.49
w-Xylene + /^-xylene
0.58
34
-------
4.2.1.8 M1 Growth on Butyl Acetate and a 50:50 Mixture of Butyl Acetate and MEK (GS
15, 19)
The objective of these experiments was to study the effect of butyl acetate on the rate of MEK
degradation and of growth of Ml on MEK.
Methods: For the study with butyl acetate, the concentration of substrate in the flask was 150 ppm,
whereas concentrations of butyl acetate and MEK were 75 ppm each for the mixed-substrate study.
Results: The growth curve indicated that butyl acetate and MEK were degraded concurrently. An
effect from the presence of MEK was seen on the growth rate for butyl acetate; the growth rate
obtained for the mixture of butyl acetate and MEK was lower than that obtained for Ml growing on
butyl acetate alone.
Substrate
Growth Rate
(h1)
MEK
0.5
Butyl Acetate
0.70
MEK + Butyl Acetate
0.57
4.2.2 Flat-Sheet Biofilm Experiments
Experiments were performed utilizing a small contactor using a flat, square section of porous
polypropylene substrate. The basic configuration employed in these studies is shown in Figure 9.
250 mL aqueous medium
Film Side
Feed side
250 mL fluid w/VOCs
Figure 9. Flat-Sheet Contactor Schematic
35
-------
The system was operated under countercurrent flow with a flow rate significantly smaller than the
recycle flow rate, resulting in well mixed fluid in the reservoirs and on each side of the membrane
within the contactor. This configuration was adopted because it allowed relatively rapid
establishment of films and determination of degradative activity and because the film could be
sampled directly for characterization (thickness, composition).
Initiation of all of these experiments involved aqueous phases on both sides of the membrane, with
the substrate being provided through the membrane in all cases. Subsequent to establish-ment of a
viable, active biofilm, the feed fluid either remained aqueous or was switched to an oleophilic fluid
(octanol or silicon oil, for example).
Flat sheet biofilm experiments were conducted according to the matrix in Table 14.
Table 14. Flat-Sheet Biofilm Experiments
Experiment No.
Organism(s)
Substrate(s)
Comments
FS1
M1
MEK, toluene
aq/aq
FS2
M1
MEK
aq/octanol
FS3
M1, X1
MEK, toluene
aq/aq
FS4
M1, X1
MEK, toluene, m- xylene
aq/aq
FS5
M1, X1
MEK, toluene
aq/octanol
FS6
X1
p-Xvlene
aq/aq
FS7
X1
m- Xylene
aq/aq
FS8
X1
m- Xylene, p-xylene
aq/aq
FS9
X1
m- Xylene, p-xylene
aq/silicone oil
General experimentalprocedures: During experiments to determine the degradative capacity of the
biofilm developed on the membrane, a 1-L flask containing 800 mL of oxygenated L-salts
(reservoir) replaced the chemostat (CSTR), and a 1-L flask with 800 mL feed solution replaced the
feed bottle. Both flasks were stirred, and flow rates were as given above. For analyses of VOC
concentration, 1.0-mL samples were withdrawn hourly. VOC concentrations in the reservoir and
feed bottle were followed for at least 5 hours.
General analyticalprocedures: The liquid samples were transferred to a 2-mL vial containing 1 drop
of acetic acid for preservation. VOC extraction was carried out by adding pentane (0.1- 0.2g)
containing heptane as an internal standard. The sample was shaken vigorously for 2 minutes and
stored upside down at 5 °C. Analysis was performed with a Hewlett-Packard 5890 GC equipped
with a J&W Scientific capillary column (DB-624) and a FID. A l-|iL portion of the pentane phase
was injected into the GC, and the response was compared to the response from standards prepared
36
-------
in pentane. A total xylene concentration was calculated for experiments involving both /^-xylene and
w-xylene because it was not possible to separate the two xylenes in the GC analysis.
4.2.2.1 Growth of X1 on /r?-Xylene and p-Xylene (FS 6, 7, 8)
The obj ective of these experiments was to study the degradation capacity and biofilm characteristics
of a xylene-degrading biofilm. Degradation of w-xylene and /^-xylene supplied as single substrates
and as part of a mixture were investigated.
Experimental configuration: Two flat-sheet contactors connected in series were run co-currently
with aqueous solution on both side of the membrane. The biofilm side was connected to a CSTR and
the opposite side (feed-side) was connected to a feed bottle saturated with the VOC. A slow airflow
was passed through a pure solution of the VOC (m-xylene or /^-xylene), and the exhaust air was fed
to the bottom of the feed bottle, obtaining a saturated solution. The feed-bottle contained 5 liters of
distilled water. Operational data were as follows:
Parameter
Value
Volume of CSTR
1-L vessel with 500 mL culture
Average flow of L-Salts to CSTR
30 mL/h
Ph in CSTR
7.0
Temperature in CSTR
30 °C
Flow on feed side
7.7 L/h
Flow on biofilm side
34 L/h
The CSTR was inoculated with XI. When exponential growth was obtained in the CSTR, the
contactors were inoculated by passing the suspension over the membrane on the biofilm side.
Results: Experiments were carried out for two levels of /^-xylene and w-xylene. The feed was
prepared from the saturated solution, which made the concentration difficult to control. For the last
two experiments reported below (A, B), the feed solution was made by adding the pure VOCs to
distilled water that had been oxygenated for 10 minutes. An overview of the experiments is shown.
After the experiments with /^-xylene, the biofilm was harvested, and the system was inoculated
again. A very thin biofilm with black and gray spots was observed on the membranes. The second
biofilm was grown for 38 days. After 38 days, the biofilm appeared yellow and was filling out the
space between the support material and the membrane to the thickness allowed by the gasket.
37
-------
Substrate
Level
Age of Biofilm
(days)
/^-Xylene
High
5
7
5
10A
25b
Low
/^-Xylene + m-Xylene
w-Xylene
High
Low (150 ppm)
100 ppm of each
Results: All experiments exhibited very low VOC concentrations in the film-side reservoir. The
concentrations were below 1 ppm except for the first experiment, for which a high concentration (26
ppm) was observed for the very first point and an increase from 0.1 ppm to 4 ppm was observed
during the 10-hour run. Because of the insignificant concentrations in the film-side reservoir, the
removal of VOC from the feed may be considered as the amount degraded in the biofilm.
For the young biofilms (5-7 days), the average removals of /^-xylene and m-xylene from the feed
bottle were in the range 4-11 ppm/h. For the 10-day-old biofilm, the average removal of m-xylene
was 20 ppm/h. However, the removal appeared to take place only during the first 4-7 hours of
exposure to the substrate. After 4 hours, the concentration in the feed bottle established a constant
level and no further degradation was seen. This could be due to oxygen limitation in the system, as
no oxygen was supplied during the experiment. For the 10-day-old biofilm, the feed solution was
oxygenated prior to the experiment, which may explain the higher removal. This could also be due
to growth of the biofilm and, hence, an increase in active biomass. For the same experiment, the
reservoir was replaced by a new oxygenated reservoir after 21 hours. After the replacement,
additional substrate removal was observed, which may confirm that the w-xylene degradation was
oxygen-limited.
When /^-xylene and w-xylene were supplied as a mixture, the biofilm was 25 days old, and an
average removal of 60 ppm/h was observed for the first 2.5 hours. The feed solution was also
oxygenated prior to this experiment, making higher removal possible. However, the removal
obtained in this study (150 ppm) compared to the removal obtained in the study with the 10-day-old
biofilm (80 ppm) should be equal if the amount of oxygen was the critical limitation, since the same
amount of oxygen was supplied in the two experiments. The discrepancy between the two results
may still be oxygen limitation, but a limitation inside the biofilm rather than in the aqueous phases.
38
-------
4.2.2.2 Growth of X1 on /r?-Xylene and p-Xylene (FS 9)
The objective of these experiments was to determine the ability of the biofilm to extract substrate
from an oleophilic feed fluid.
Experimental configuration: Two flat-sheet contactors were connected in series. After establishment
of the biofilm (1-2 weeks), the activity of each contactor unit was determined separately. In each
case, silicon oil was fed to the feed side, and L-salts medium was fed to the biofilm side. The oil and
the L-salts were run counter-current in closed loops. The capacity of the contactor unit was
determined by following the concentration of the substrate in both reservoirs after spiking the oil
with VOC. After determination of the capacity, the biofilm was harvested, and samples for biofilm
thickness (wet weight), cell count, and protein content were collected.
Results: Removal of m-xylene and /^-xylene from the silicone oil was not detectable during the 25-
hour- long experiment.
Contactor
Film Age
(days)
Biofilm
Thickness
(\im)
Protein
Content
(|lg/cm2)
Cell
Count
(cells/cm2)
1
12
14
4.7
24xl06
2
19
95
—
86x10s
4.2.3 Hollow-Fiber Membrane Experiments
A series of experiments using hollow-fiber modules was performed to establish the efficacy of the
proposed biotreatment module for enhanced VOC removal from the octanol. The liquid/liquid
stripping efficiency of MEK (octanol to water) was determined for the biotreatment module both
with and without a biofilm present. Three aqueous (absorbent) flow rates were examined, with a
constant 5000-ppm MEK concentration in octanol and an octanol flow rate of290 mL/min. Samples
were taken in duplicate. Results shown in Table 15 (comparing the 301-mL/min flow, abiotic vs.
biofilm) indicate that the presence of a live biofilm enhanced MEK removal by approximately 43
percent.
39
-------
Table 15. Biomembrane Mass Transfer
Experiment Type
Abiotic
Biofilm
Aqueous Phase
filtered tap water
L-saltsa
Q=n— (mL/min)
116
301
598
301
CU_, (mL/min)
290
MEK in Octanol (ppm)
5000
Mass Transfer Rate (g/m2h)
1.80
2.07
4.45
2.97
a pH-balanced trace-nutrient source for microbiological organisms
4.2.4 Staged Biotreatment of VOC Mixtures in Lab-Scale Reactor
A membrane bioreactor for treatment of a broad range of compounds found in paints for aircraft was
set up. Different levels of water solubility among the compounds suggested a staged bioreactor
comprising a membrane contactor (Liqui-Cel) and a chemostat (CSTR). VOCs were supplied to the
Liqui-Cel in silicone oil. The Liqui-Cel served as the reactor for removal of compounds with low
water solubility, whereas compounds with high water solubility are more likely to be removed in
the CSTR. The objective was to study the performance of the staged bioreactor system and to study
the degradative capacity for mixtures of VOCs with different water solubilities in the bioreactor
system inoculated with two organisms. A limited number of VOCs were chosen, and the removal
of these compounds was monitored. VOC-contaminated oil obtained during air treatment was
treated as well.
Experimental configuration: A Celgard Liqui-Cel unit with polypropylene fibers was used as the
membrane contactor. The volume of the shell side was 0.4 L and of the lumen side was 0.15 L. The
oil passed through the lumen side at a flow rate of 54 mL/min. (3.2 L/h). A CSTR with a liquid
volume of 4 L was connected to the shell side, and the bacterial suspension was passed through the
shell side in a counter-current mode. The flow rate of the aqueous phase was 400 mL/min (24 L/h).
Aerated L-salt medium was supplied to the CSTR at 100 mL/h. The temperature was set at 30 °C,
and pH was adjusted to 7.0. The supply of L-salt medium to the CSTR was turned off during the
experiment, and the outlet and the air vent from the reactor closed. Oil samples were withdrawn
from the oil reservoir, and aqueous samples were taken at the inlet and outlet from the Liqui-Cel.
Ml and XI were cultivated in a 1-L chemostat, and when exponential growth was obtained, the
Liqui-Cel was inoculated by passing the suspension through the shell side. The attached CSTR was
inoculated with Ml, and after reaching exponential growth, the reactor was supplied with L-salt
medium containing MEK until a biofilm was established. The systems were combined, and oil
containing both MEK and m-xylene served as the substrate source.
40
-------
Analytical procedure: All oil samples and aqueous samples containing both MEK and butyl acetate
or paint components were analyzed by ARCADIS. GC analysis was performed on the aqueous
samples for determination of the aromatics, and high performance liquid chroma-tography (HPLC)
analysis was done to determine aqueous MEK concentrations. Samples for aromatics were
transferred to 2-mL vials containing 1 drop of acetic acid for preservation. Extraction of VOCs was
carried out by addition of pentane (0.1- 0.2g) containing heptane as internal standard. The samples
were shaken vigorously for 2 minutes and stored upside down at 5 °C. Analysis was performed with
a Hewlett-Packard 5890 GC equipped with a J&W Scientific capillary column (DB-624) and aFID.
A l-|oL sample of the pentane phase was injected into the GC, and the response was compared to
the response from standards prepared in pentane. Samples for MEK were collected in 2-mL plastic
vials and kept in the freezer until the analysis. HPLC analysis was performed by use of a Spectra
Physics HPLC equipped with a Waters 990 Photodiode Array Detector and a reverse-phase Altima
C-18 column. The mobile phase used was methanol and phosphate buffer (50 mMKH2P4) acidified
with 0.1 percent trifluoroacetic acid. The thawed samples were centrifuged to spin down the cell
mass. Sample volumes of 100 |iL were analyzed.
Experimental procedure: The substrates were chosen to represent constituents in paint were m-
xylene and /^-xylene (aromatics), MEK (ketones), and butyl acetate (B A) (esters). Table 16 describes
the substrate levels and combinations used in the experiments.
Table 16. Biotreatment of VOC Mixtures in a Lab Scale Reactor
Experiment
No.
Substrates
Concentrations
in Oil
(PPm)
Volume Oil
(mL)
Volume Aqueous
Phase
(mL)
SB1
MEK + m-xvlene
500
1000
4000
SB2
MEK + m-xvlene
1000
1000
4000
SB3
MEK + m-xvlene
1500
1000
4000
SB4
MEK + p-xvlene
500
1000
4000
SB5
MEK + p-xvlene
1000
1000
4000
SB6
MEK, BA, m-xylene
500
1000
4000
SB7
Oil with MEK, BA,
ethylbenzene, m-
Various
1800/2000
4000
SB8
Oil with paint
Various
1800/2000
4000
Results: Treatment of MEK and m-xylene in mixture was examined in seven experiments. The
system was closed down after the first three experiments due to operational problems. The system
was set up and inoculated again, and four more experiments were done with MEK and /^-xylene.
41
-------
In the first three experiments, a high removal of m-xylene from the oil was observed. In the aqueous
phase, m-xylene was either not detectable or detectable only at low concen-trations. The waste outlet
and the air vent from the CSTR were not closed during the first three experiments, which may have
contributed to the high removal. Removal of w-xylene at rates up to 18 ppm/h was obtained during
an 8-10 hour period in the next four experiments. Rapid transfer of MEK from the oil to the aqueous
phase was observed in all the experiments. However, the removal of MEK was very low, and an
accumulation of MEK in the aqueous phase was observed. After the experiments with MEK and m-
xylene, /^-xylene replaced w-xylene in the mixture. Removal of /^-xylene was not detectable.
Removal of MEK from the aqueous phase appeared to take place in the range of 5-14 ppm/h.
When mixtures of MEK, BA, w-xylene, and /^-xylene were treated, MEK removal was observed as
well. It was not possible to separate w-xylene from /^-xylene in the GC analysis, so total xylene
concentration was determined. The rate of removal of the xylenes was 36 ppm/h. BA appeared to
be removed as well (standard curve for MEK was used for the GC-analysis).
Treatment of oil containing a mixture of BA, ethylbenzene, w-xylene, and o-xylene showed only
insignificant removal of VOCs from the oil. When oil run in the air treatment system with paint was
treated, only xylene removal was observed. MEK and BA appeared only in the oil run with paint
for two hours and both compounds were removed immediately from the oil.
42
-------
5.0 Conclusions
This study was performed to collect data on a novel hollow-fiber membrane-based technology to
extract and process VOCs from a dilute stream produced by processes such as paint spray booths.
The resultant reactions would form water and C02 as a final exhaust stream. This system was at least
partly developed during the execution of this project, and the components tested were judged by
their capability to produce a system that is economically competitive with such other technologies
as thermal afterburners or adsorption systems (e.g., carbon filters). It is concluded that the process
science is sound. Under tests/evaluations in which process economics were not considered, VOC
removal rates (greater than 70 percent) of certain compounds were measured. Although many of the
problems were technical in nature, many could likely be resolved through further research and
resource allocations.
5.1 Separation System
No suitable separation module currently exists, and the project's development efforts failed to
produce a suitable module to deliver a complete, cost-competitive system for VOC control.
Membrane-contactor developers are currently devoting their time and resources to applications they
deem more promising with brighter payoff potentials. Equipment size is a problem as well.
Microfiber contactor modules are typically developed for small-scale processes, and a module
suitable for extracting VOCs from a large paint booth would require manufacturing capabilities that
are not currently available. A suitable module could be as much as 20 to 50 times the size of the
largest module currently available for any process, and its production equipment would be required
to outfit more than just paint booth processing equipment to meet the necessary economies of scale.
In addition, radial modules are the most-commonly available, but the separation contactor for paint
booth work needs to be a box-type cross-flow module to reduce the air-side pressure drop to
reasonable levels.
A number of microporous fibers were tested for their potential to develop appropriate high-
performance modules. In early testing, a number of modules with uncoated fibers were tested that
produced reasonable results, but pressure balancing between the air side and transfer-fluid side was
determined to be economically unfeasible, so modules with coated fibers were pursued as a solution.
Pore size as well as coating thickness and consistency are the important parameters for producing
the appropriate extraction performance in fibers. The coating must be inert to prevent reactions and
such performance detriments as coating swelling or softening. The coating must adhere to the fiber
walls. A plasma-polymerized silicone-rubber-coated fiber developed by AMT proved to have good
adherence and inertness properties, and this fiber displayed good endurance under a wide range of
pressures. The AMT fiber has a greater potential for module development than any other coated
43
-------
fiber developed or tested in this project. Although testing showed this fiber leaked silicone oil, the
coating thickness could have been adjusted to address this problem if more time had been available.
Other adjustment are also available to address this problem including coating treatments, pore size,
and post-treatment processing steps.
A successful module must be developed simultaneously with testing the intended transfer fluid to
be certain of material and physical compatibility. Before modules are developed, potting and
module-frame materials must be tested with the intended contact fluid using extended contact times
and under the intended operating pressure. Seals such as O-rings or other gaskets must be able to
withstand constant immersion in the intended transfer fluid. Octanol attacked many O-ring materials
that were used on the Celgard modules during early testing, and the only O-ring found to withstand
this fluid was cost prohibitive for production purposes. Subsequently, octanol was found to also
swell the coated fibers that were used in the AMT cross-flow separation contactor. This swelling
caused bunching and voids in the fiber bundle that resulted in air flow channeling and poor contact.
A number of alternate fluids were considered before settling on a replacement for octanol. These
alternatives included canola oil and mineral oils, both of which have too many impurities to allow
accurate GC determination of extracted compounds, and silicone oil, which would be cost
prohibitive in a full-scale facility and would have unknown effects on bugs over a long period.
Silicone oil was chosen as an alternative to octanol but, because of time constraints, was not
thoroughly tested for partitioning performance, and this may have played a big role in the
disappointing removal efficiency test results. In addition, although materials compatibility testing
with a higher-viscosity silicone oil produced acceptable results, performance results with a lower-
viscosity oil, chosen for its similarity to octanol, resulted in unforeseen problems such as weeping,
leaking, and stretched fibers. An optimized module must be closely matched in all characteristics
to its extraction fluid. Based on the experience with this project, a fluid or fiber substitution will
almost always require a complete return to the beginning of the development cycle.
The AMT cross-flow module has the necessary characteristics for acceptable air-side pressure-drop
performance. However, distortion of the fibers caused by small leaks of the thin silicone oil
produced uneven contact that counteracted any performance improvements. Data showed that
performance from a given arrangement of modules is adversely affected by increased airside flow.
Extraction performance drops progressively as the oil absorbs higher quantities of VOCs, but the
results of testing may have showed the limited capacity of silicone oil compared to octanol. This
conclusion seems particularly obvious when seeing that VOC absorbance dropped quickly during
some runs, when the oil was still lightly loaded. On the other hand, increasing oilside flow and
pressure enhances performance. This increased performance is, of course, limited by module
integrity, and running at higher pressures caused failure of several modules.
44
-------
5.2 Biotreatment System
5.2.1 Biodegradation Range and Extent
The screening studies, and subsequent growth studies, indicate that several organisms that were
isolated are capable of degrading the model compounds tested (w-xylene,^-xylene, toluene, methyl
ethyl ketone, butyl acetate). As one would expect, no single organism was capable of simultaneously
degrading all of the compounds used during the project. Therefore, a consortium of organisms was
used to effect degradation of VOC mixtures. The majority of the work performed focused on two
organisms, designated Ml and XI. These organisms were isolated utilizing aliphatic and aromatic
substrates, respectively.
The degree of degradation exhibited by the organisms used in the study was significant. Between
the two organisms, all of the compounds could be reduced to levels below 250 ppb in an aqueous
phase. These levels were exhibited in experiments GS2-GS10, GS14, FS3, and FS5-FS8. In every
case, this VOC level represented greater than 99 percent reduction in the amount of compound
present initially.
5.2.2 Problems Arising from Metabolic Regulation
The experimental results obtained indicate that concurrent degradation of aromatic and aliphatic
VOCs in a single bioreactor is problematic. Cultures containing Ml and XI (individually or in
combination) exhibited roughly concurrent degradation of the aromatic compounds, followed by
degradation of the aliphatics. Table 17 indicates those experiments where concurrent (or nearly so)
degradation or aromatic-aliphatic sequential degradation was observed and identifies the compounds
involved.
Table 17. Concurrent Degradation
Experiment No.
Organism(s)
Compounds
GS3
X1
Toluene, m-xvlene
GS5
M1
Toluene, MEK
GS6
M1
m-Xvlene, MEK
GS7
M1
Toluene, m-xvlene
GS9
M1 AND X1
Toluene, MEK
GS10
M1 AND X1
Toluene, m-xvlene, MEK
GSM
X1
m-Xylene, p-xylene
Ethylbenzene was found to have a predominantly, but not exclusively, inhibitory effect on culture
45
-------
growth and biodegradative activity, as indicated directly by the results of experiment GS13 and as
can be inferred from the results of experiments SB7 and SB8. The results from GS13 indicate that
ethylbenzene stimulates growth to a small degree at aqueous phase levels below 100 ppm, but a
level of 150 ppm prematurely terminates cell growth. The low rate of degradation of aromatic
compounds exhibited in experiments SB7 and SB8 (ethylbenzene present) relative to results from
experiments SB3, SB5, and SB6 (ethylbenzene absent) suggests that aqueous-phase ethylbenzene
levels as low as 40 ppm might have an inhibitory effect on aromatic degradation in biofilm cultures.
This is, however, a supposition; further testing is necessary before this can be reliably concluded.
Fortunately, no such inhibition was noted for degradation of butyl acetate or MEK.
5.2.3 Treatment Strategy
Given the sequential nature that was observed for aromatic/aliphatic degradation, the strategy was
adopted to circulate the aqueous phase of the biofilm reactor through a continuously stirred tank
reactor containing primarily Ml (the aliphatic degrader). As indicated in experiments SB2 through
SB6, this strategy is quite successful in concurrently reducing aliphatics (MEK and/or butyl acetate)
and aromatic compounds (m- or /^-xylene) in the oil phase. This behavior was indicated by staged
biotreatment systems with fresh films (SB4, 6) and films regenerated following EDTA-induced
sloughing (SB3, SB5).
5.2.4 Implementation
The flat-sheet and staged-bioreactor studies indicate that "bioextraction" of VOCs from the carrier
fluid occurs at rates sufficient to maintain active cell growth and activity. These studies also indicate
that the hollow-fiber membrane units available to this project had inadequate surface area to
effectively treat the carrier fluid in anything approaching real time. The hollow-fiber membrane
units do, however, offer effective contacting of the biofilm with the carrier fluid. It would appear
that the next steps in evaluation of this configuration would involve (1) identifying an effective
degrader of ethylbenzene and (2) designing a membrane contactor that would enable sampling of
the biomass so that comparisons can be made between the flat-sheet results (where biomass can be
sampled directly) and the hollow-fiber unit performance.
46
-------
6.0 Recommendations
The technology developed as a result of this project must be technically and fiscally sound before
a decision will be made for future full-scale implementation. Although the technology has been
demonstrated to have technical promise or potential, no results have been produced that warrant
further consideration and study until manufacturing techniques for membrane modules advance
enough to allow the cost-effective fabrication of low-pressure-drop, high-efficiency, leak-free
modules. The potential cost and environmental benefits of such a system do warrant consideration
of further basic development projects aimed at improved fibers, coatings, stripping fluids, and
module designs.
Future efforts should concentrate on fiber and module development. Many options are available for
fiber materials and for coatings (see AMT response), and more are being developed each year,
usually for some specific application that suddenly becomes financially promising. Module
construction is also undergoing groundbreaking development in many other applications.
Paints are also being reformulated, and this will make specific module development more difficult,
especially considering the difficulty experienced in identifying paint constituents. Future work
should pay particular attention to the great difference between common paint constituents, such as
industrial-grade xylenes and pure compounds, when performance testing. The use of a total
hydrocarbon analyzer through each phase will more clearly establish whether performance goals are
being achieved or if many compounds are escaping notice.
6.1 AMT Recommendations
Applied Membrane Technology, Inc.'s objective was to develop low-pressure-drop cross-flow
hollow-fiber membrane modules for use in the MBT system intended for high volumetric flow rates
containing low contaminant concentrations of fugitive target compounds. To provide suitable low-
pressure-drop modules within the project time constraints, both module fabrication and membrane
composition needed to be based as much as possible upon known parameters of module construction
and chemical formulation. In other words, a thorough investigation of alternative materials of
construction that might be more suitable was not feasible during AMT's period of involvement.
AMT's module design was based on a previous project for the US Navy for use in a low-pressure-
drop seawater application. This design was adapted for the ARCADIS VOC air proj ect with several
minor alterations to make it more compatible for air-stripping applications. AMT successfully
designed and built a winding fixture that could wind four cross-flow membrane modules at a time.
(Scale-up to larger volumes would be straightforward based on this design). The modules produced
47
-------
using this design yield a very low air side pressure drop as intended. The module housings were
constructed from polycarbonate due to the need to rapidly tool fixture parts and module components.
Other module housing materials suitable for long-term applica-tions could be utilized in the future.
The hollow-fiber membranes proved to be a more-problematic issue. Preliminary testing and
evaluations of membrane fibers and prototype modules utilizing plasma-polymerized, silicone-
coated, polypropylene fibers indicated adequate flux of VOCs and tolerance for contact with silicone
oil up to 35 to 40 psi. AMT lab evaluations of octanol stripping agents ruled them out due to
excessive swelling and distortion of polypropylene membrane substrates supplied by all vendors.
Membrane compositions for use with silicone-oil stripping agents could be made utilizing stronger
polypropylene substrates produced by Celgard US or AKZO Membrana, although they will exhibit
some degree of swelling depending on pressure, temperature, and the nature of the membrane top
coating. AMT could place copolymer coatings, such as silicone/Teflon or silicone/propylene, onto
these microporous substrates instead of the straight silicone membrane.
The latter copolymer compositions have shown higher resistance to oil permeation and swelling
effects in other AMT lab projects and during evaluation of flat-film samples conducted by outside
research groups. The ability to tailor the coatings' chemical composition offers potential avenues
for future improvements. Other groups encountered similar problems of oil and solvent
swelling/permeation during the successful development of solvent extraction from oil in the de-
waxing process performed by petroleum companies. Mobil Oil and Texaco use several commercial
units based on flat membranes. AMT can make hollow-fiber versions of similar copolymers, which
may merit investigation for applications such as the MBT system. AMT would need to evaluate a
series of potting compounds and determine appropriate potting techniques for assembling bundles
of such coated fibers because adhesion becomes more difficult as the concentration of Teflon or
polypropylene increases in the membrane coatings. It is a doable proposition, but the time
constraints of this project would not have allowed for such trials.
In addition, AMT could evaluate both polymeric and ceramic non-propylene-based substrates. For
example, new ceramic-based hollow fibers of dimensions comparable to conventional microporous
membranes and coated with a silicone/PTFE copolymer membrane via plasma polymerization may
offer the ideal membrane composition for membrane-mediated extraction of VOCs. The ceramic
materials would tolerate broad temperature and pressure regimes and exhibit no swelling behavior,
and the membrane coatings could be tailored to optimize VOC flux while restricting oil permeation.
The obj ective of such an endeavor would be to raise the operating window of the cross flow module
so that higher pressures (i.e., faster liquid sweep rates) could be utilized to enhance module VOC-
stripping performance as well as to fix the hollow fibers in place for air flow distribution without
48
-------
later shifting due to swelling, as observed in the current units.
In summary, the general design and method used to fabricate the cross-flow membrane modules was
found to be suitable for producing the low-pressure-drop units. More adequate allocation of
resources in terms of time and research expenditures directed at optimizing the hollow-fiber
substrate/coating composite formulation may result in superior membrane performance, which, in
turn, would provide a more-economic and commercially viable operating window for the MBT
process.
49
-------
Appendix A
Literature Search
and Review
-------
The success of the VOC treatment technology we are developing is dictated by the performance of the
separation and biotreatment modules. Certain design and operational criteria are recognized as essential
to the efficient and economical operation of each contactor. For the air/VOC separation module, these
criteria include the following:
High surface-area-to-volume ratio
Membrane composition and configuration to maximize VOC transfer
Membrane configuration to minimize leakage of transfer fluid into air stream
Module design to minimize airside pressure drop
Possibility for commercial availability of module
For the bioreactor module these criteria include the following:
High surface-area-to-volume ratio
Membrane configuration to minimize leakage of transfer fluid into aqueous stream
Design and operation to minimize biofouling of the aqueous side
Maximize degradative activity of biofilm
Capability to rapidly degrade a wide variety of VOCs
In consideration of these issues, the following extensive literature search was undertaken to guide the
subsequent development of the process.
AIR/VOC SEPARATION MODULE
Module Type
Although membrane-based gas separation was first commercialized in 1977 for the enrichment of oxygen
from air, the predominant applications currently are for the production of high-purity (99.95%) and
medium-purity (95-99.5%) nitrogen (Prasad, 1994), C02 removal from natural gas (McKee, 1991), and
hydrogen purification for recycle (Shaver, 1991). All of these processes employ asymmetric or composite
polymer membranes (Stern, 1994). These applications have come on the heels of 20 years of intensive
research that saw the development of asymmetric membranes that exhibited high fluxes (Loeb, 1963), a
method for the synthesis of robust hollow fibers (Vos, 1969), and a method for the casting of ultrathin,
high- permeability, nonporous polymer films onto existing membranes (Ward, 1976) [subsequently
improved upon and commercialized by Monsanto (Henis, 1981; Henis, 1980)]. Commercial membrane-
based nitrogen production processes, for example, predominantly utilize hollow-fiber membrane
configurations instead of plate-and-frame or spiral-wound. The rationale for this choice is clear when
considering the relative membrane module areas. Koros and Fleming report that membrane module areas
for the three configurations are: 100-150 ft2/ft3 for plate-and-frame modules, 200-250 ft2/ft3 for spiral-
wound, and 2000- 4000 ft2/ft3 for hollow-fiber modules (Koros, 1993). The advantageous surface-to-
volume ratio attained by hollow-fiber systems was a major factor in our selection of this configuration for
our modules.
Membrane Composition
Membrane transport
To select among the plethora of materials available for membrane construction, it is important to first
understand the molecular basis of flux through porous and nonporous polymeric materials. Because
essentially all commercial gas-separation membrane systems utilize nonporous membranes, the
A-l
-------
description of transport through strictly porous materials will be abbreviated. Flux is the rate of transport
of a chemical species (also referred to as the penetrant) through a given membrane or pore-filling
material. Flux is determined by the chemical potential gradient of the transported species across the
membrane, the membrane thickness, and in the case of nonporous membranes, the solubility of the
species in the membrane material. The mechanism of species flux through nonporous material has been
generally described as solution-diffusion (Koros, 1989; Koros, 1991) and depends upon the solubility (or
condensability) and the diffiisivity of the species in the polymer.
The chemical potential gradient is represented by the partial pressures of the species on either side of the
membrane for gas-gas separations or, in the case of a gas-liquid separation, the "effective" partial
pressure of the species on the liquid side of the membrane. An intrinsic property of polymeric material
relative to species transport is the permeability (/' ,). which is "a parameter equal to the pressure-and-
thickness-normalized flux" (Koros, 1993), as shown in Eq. 1:
I' , = (flux of A per unit arca.)/(Ap , dm) (1)
where Ap , is the partial pressure gradient of species A across the membrane and dm is the membrane
thickness. Permeability is a direct measure of the ease of transport of the species through the membrane,
and it can be written as the product of the solubility coefficient, SA (a thermodynamic parameter) and the
diffusion coefficient /), (a kinetic parameter);
Pa = (SA) (DJ (2)
These coefficients can themselves be complex functions of many variables, including identity and
concentrations of current and past sorbed species as well as temperature (Koros, 1989; Koros, 1991).
What is of importance to those seeking to separate two gas-phase species (A. B) is the separation factor, or
selectivity, (aAB), which is defined in terms of the mole fractions of the two species (y y B) upstream
(subscript 1) and downstream (subscript 2) of the membrane.
*
AB
B 2
y.Jy
B\
(3)
The separation factor can also be written in terms of the permeabilities of the two penetrants and the
relative partial pressure driving forces (Ashworth, 1992; Koros, 1993; Stern, 1994):
Written in this manner, the separation factor can be seen to have contributions derived from penetrant
solubility (SA/SB) and penetrant mobility (DA/DB). Selective solubility is the controlling factor in
processes involving the separation of vapors from gases, such as VOC removal from air (Baker, 1987;
Watson, 1990; Lund, 1996; Deng, 1998). The motion of polymer segments in the membrane controls
mobility selectivity. As these segments move, free volume becomes available to the penetrant and it
"hops" from one intersegmental free volume to another (Kumins, 1968; Frisch, 1983; Zielinski, 1992).
The frequency of these movements is affected by membrane temperature (Kulkarni, 1983), by the size of
the side chains along the polymer backbone, and by whether the polymer is glassy or rubbery (above or
A-2
-------
below the glass transition temperature for that polymer). This theory has been used successfully to
describe the permeability of organic vapors in several rubbery polymers (Fujita, 1968; Suwandi, 1973),
most particularly silicone rubber (Sok, 1992) or poly(dimethylsiloxane) (PDMS).
Membrane experimental configurations
Use of membranes to separate VOCs from a contaminated stream has taken two primary forms:
pervaporation and vapor-phase extraction. In pervaporation, a contaminated aqueous phase contacts the
shell side of a microporous hollow fiber, which has been coated with a nonporous thin film, most often
PDMS. A vacuum is drawn, or a carrier gas flows on the tube side of the membrane, and the VOCs
permeate through and evaporate from the membrane. Vapor-phase extraction, on the other hand,
generally involves a contaminated gas stream with a vacuum or extraction fluid on the other side of the
membrane. The driving force in these cases is the same as described above, the chemical potential
gradient caused by low downstream penetrant concentration.
Porous membranes
When a gas-vapor mixture is made to flow through a porous solid medium, the condensable species
(vapors) have much higher transport rates than the noncondensable species (gases). This phenomenon is
the result of several transport mechanisms, including adsorbed flow (Carman, 1952) and capillary
condensation (Rhim, 1975, Lee, 1986; Qiu, 1991). Forthe condensable gas, capillary condensation can
occur on the insides of the pores at a pressure much less than the ordinary condensation pressure (or at a
concentration much less than normal). Once condensed, the transport of the condensate through small
pores can be regarded as viscous flow, and this flow blocks the transport of the noncondensable gas
through the pores. A mathematical treatment of these phenomena was presented by Qiu and Hwang
(1991) and compared to experimental results using a porous glass membrane.
Nonporous membrane materials
A significant number of investigations have focused on the effect of polymer modifications on transport
properties for both glassy and rubbery polymers. Two generalizations can be made regarding the results
obtained:
1. Permeability and selectivity exhibit an inverse relationship.
2. Rubbery polymers exhibit high permeability and low selectivity; glassy polymers exhibit high
selectivity.
Several exceptions to these "rules" do exist, and it is instructive to investigate more closely the literature
regarding glassy and rubbery polymers separately.
Glassy Polymers
Generally, these polymers exhibit higher selectivity and lower permeability than rubbery polymers.
However, PTMSP (poly [(l-trimethylsilyl)-l-propyne]), a polyacetylene which has a repeating structure
of [-(CH3)C=C(Si[CH3]3)-]x , is a glassy polymer that has been found to have roughly ten-fold higher
permeability than PDMS with only somewhat lower selectivity (Masuda, 1983; Takada, 1985; Masuda,
1988; Ichiraku, 1987). Prior to these results, PDMS was found to have the highest permeability of any
nonporous polymer. The high permeability of PTMSP is attributed to the bulky side groups (-Si[CH3]3)
which generate a large free-volume fraction (estimated to be 0.20-0.27) (Plate, 1991) for penetrant
diffusion. Unfortunately, since glassy polymers are non-equilibrium materials, their excess free volume
A-3
-------
tends to decrease over time and is highly dependent on the processing history of the polymer. As such,
unmodified PTMSP is not suitable for long-term commercial applications.
Significant effort has also been put forth to investigate the substituted polyacetylenes, including analogs
of PTMSP, to determine whether achieving even higher permeabilities with little loss of selectivity and
with enhanced stability is possible. Early efforts, summarized by Odani and Masuda (Odani, 1992),
indicate that polyacetylenes with bulky substituents tend to have the highest permeabilities and that the
substitution of long /7-alkyl groups for the silyl group, or long alkyl groups for a methyl group on the silyl
group, greatly reduces the permeability of the polymer.
More-recent efforts by Nakagawa and co-workers have investigated the modification of PTMSP by
bromination (Nagai, 1994) and by copolymerization (Nagai, 1997). Brominated PTMSP also exhibited a
time-dependent decline in gas permeability for a number of gases (02, N2, C02, C3H8). However, the
decline was less pronounced than exhibited by PTMSP, and once it restabilized, consistent permeability
was observed throughout a 175-day period (Nagai, 1994). Much greater success was achieved by
copolymerization or blending of PTMSP with poly(l-phenylpropyne) (PPP), a glassy polymer with
excellent stable gas permeability but lower permeability than PTMSP (Nagai, 1995). PPP concentrations
up to 10% in either blended or copolymerized form caused less than an order-of-magnitude decrease in
permeability and essentially complete stability over a 30-day period.
Work similar in breadth and focus has been underway for polyimides, poly(ether imides), polypyrrolones,
poly(amide imides), polycarbonates, polysulfones, cellulose acetate and poly(phenylene oxide). While
many of these studies show promise for development of high-performance vapor-separation materials,
few are sufficiently advanced to warrant detailed consideration. Interested readers are directed to Stern
(1994) for a review of early literature in these areas. We will deal directly with polysulfones with specific
mention of poly(alkylsulfone) (PAS-16).
In general, polysulfones are glassy at room temperature and have the high penetrant size selectivity
characteristic of glassy polymers. However, PAS-16 is a rubbery polysulfone copolymer of hexadecene
and sulfur dioxide as shown in Figure 1. It exhibits some side chain crystallinity at room temperature; the
degree of which is dependent upon the thermal processing of the polymer. This makes the permeability
of PAS-16 history-dependent, which is atypical of rubbery polymers (Singh, 1997).
CH3
(CH2)13
-t- c — en2— so2i—
H
Figure A-1. PAS-16 Repeating Unit
Extensive characterization of PAS-16 for permanent gases (02, N2, H2 and C02) indicates that the PAS-16
behaves as a typical rubbery polymer. However, a comparison of permeabilities for PAS-16 and PDMS
indicates that permeability in PAS-16 is roughly one order of magnitude lower than that in PDMS (Gray,
1976; Gray 1984; Singh, 1997), which makes it the next-most permeable-polymer given the high
A-4
-------
permeability of PDMS. The permeability of PAS-16 to organic vapors is likewise high, though some
crystallinity is manifested in these experiments, with permeability decreasing dramatically upon first
exposure to toluene (induces local crystallization) and then remaining constant for a given penetrant
(toluene, «-hexane, acetone and methanol) at different partial pressures (Singh, 1997). The relative
permeability of the top two penetrants (toluene and «-hexane) is temperature dependent, with «-hexane
permeability highest under 40 °C and toluene permeability highest at higher temperatures (Singh, 1997).
This polymer behaves as do the rubbery and ultrahigh-free-volume glassy polymers (PTMSP, for
example) regarding permeability dependence on penetrant solubility (Freeman, 1997), and as such, it
deserves further investigation to establish its suitability for commercial application.
Rubbery Polymers
PDMS (-(CH3)2SiO-)x has incredibly high permeability to a wide variety of gases; more than an order of
magnitude greater than the nearest polymer (with the exception of PTMSP, as previously described). The
high permeability of this polymer is generally accompanied by low overall selectivity for low-molecular-
weight gases. The high permeability of PDMS is attributed to the flexibility of the siloxane (-SiO-)
linkages of this polymer. Research with this polymer has focused on modifications that exhibit higher
selectivity without sacrificing permeability. Results analogous to those obtained for PTMSP were
obtained:
1. Bulkier functional groups on the side chain (replacing a methyl group) decreased permeability by
increasing chain stiffness (Stern, 1987; Lee, 1988). These substituents act by reducing the diffiisivity
of the penetrant in the polymer, not by reducing solubility.
2. An inverse relationship holds between permeability and selectivity.
3. Substituents that induce formation of crystalline domains greatly reduce permeability (Stern, 1987).
4. Specific interactions between the penetrant and the polymer that increase solubility dramatically
increase the permeability of the polymer to that penetrant (Ashworth, 1991).
The overall selectivity of highly permeable silicone polymers is a function of the solubility selectivity
(SA/SB), not the mobility selectivity (/), /),.,) (Stern, 1987). Given this fact, and in light of item (4) above,
a potentially effective method for increasing selectivity in silicone polymers, without impairing
permeability, is to substitute functional groups that interact specifically with selected penetrants.
Membrane Configuration
Composites versus single-polymer (simple) membranes
The transport mechanisms of gases through composite membranes involve two major diffusional
pathways, as shown in Figure A-2 (adapted from Kimmerle, 1991).
Path 1 runs through the nonporous top layer (thin film) polymer, which generally exhibits some
selectivity, then into the pore space of the basement membrane. Path 2 runs through the thin film and
then through the basement membrane polymer (substructure) itself before reaching a pore space. The
basement membrane material is usually selected to provide structural support for the thin film and is of
high porosity to minimize the contribution of basement membrane resistance to the overall resistance to
transport of the penetrant. The key parameters determining the flux through composite membranes under
a given driving force can be developed in the following manner.
A-5
-------
coating
substructure
W 1
Figure A-2. Two Pathways for Transport Through a Composite Membrane
The volume flux of penetrant, J, can be related to membrane structure by substitution of the appropriate
terms into the general relationship that flux is equal to the product of the driving force and area, divided
by the resistance to transport. For path 1, this equation is (Pinnau, 1988):
JL-F =
(A^,
(,r
dL | dl
pL + pp
(5)
where P,x is the permeability of penetrant /' in membrane component x, where x = L denotes the top layer,
x = P denotes the pore and, in equation (6), x = S denotes the membrane substructure material. Ap
denotes the surface area of the pores and cF denotes the thickness of the particular membrane component x
(=L, P). For path 2, transport can be expressed in the following way (Henis, 1981):
(a». xAs )
JLS = 1
i
f ,
cT
PL P:
V / i J
(6)
where As denotes the surface area of the substructure, which is equal to the total membrane surface area,
Am, minus the pore area, Ap. The porosity of the membrane S can be expressed as
s = -
aM
(7)
The total flux of penetrant / through the membrane is given by
tM tL,S , tL,P
(8)
An expression for the total volume flux (per unit area),f. can be expressed by combining equations (5-
8) to give:
jM -
7, A -
L+ P
P P j
v i i s
(l-s)
pl + ps
v /' /'V
\Ap,
(9)
A-6
-------
There are three scenarios to be considered using this equation relative to what controls the performance of
the composite membrane: (1) top-layer controlled, (2) support-structure controlled, and (3) mixed-flux
control. Kimmerle concludes the following regarding membrane configuration:
If the transport properties of a membrane are determined by the selective top layer it can
be optimized in terms of its flux and selectivity by preparing a support structure which
has the highest possible porosity and permeability in the pore structure. The permeability
of the selectivity top layer should furthermore be as high and its thickness as low as
possible. (Kimmerle, 1991)
This conclusion is endorsed by a large number of practitioners in the field (Koros, 1993; Parthasarathy,
1994; Stern, 1994; Prasad, 1994; Nagai, 1997; Zhu, 1983; Blume, 1990; Bessarabov, 1996; Das, 1998;
Poddar, 1996b; Cha, 1996) who have evaluated single polymer and composite membranes for gas
separation and VOC removal from liquid and air streams.
In a recent review of commercial air separation processes, Prasad and co-authors list the
development of ultrathin barrier layers as one of the key technical innovations in the commercial
development of membranes in air separation. "The discovery by Loeb and Sourirajan (Loeb,
1963) of integrally skinned, high-flux asymmetric membranes was essential in transforming
membranes to a technology of significant commercial interest" (Prasad, 1994). Ashworth
provides an excellent illustration of the direct benefit of composite membranes using the
separation of H2 from CO as the goal and polysulfone and silicone rubber as the membrane
materials under consideration (Ashworth, 1992). The following data are taken from his paper.
Table A-1. Effect of Membrane Composition on Permeability and Selectivity
Polymer
Permeability
3 2
(cm (STP)-cm/cm -sec-cmHg)
h2
CO
OtH2/CO
Silicone rubber (SR)
5.2 x 10"8
2.5 x 10 8
2.1
Polysulfone (PS)
1.2 x 10"9
3.0 x 10"11
40
SR/PS (99:1)a
3.65 x 10 8
2.68 x 10 9
13.6
SR/PS (91:9)a
1.08 x 10"8
3.29 x 10"10
32.8
The ratio denotes the relative thicknesses of the two polymers in a bilayer composite.
Obviously, the productivity of the composite membrane is nearly an order of magnitude higher than that
achieved by the pure polysulfone membrane (approximately 80% selectivity, ninefold increase in H2
permeability). Experiments involving the separation of dichloroethane, chlorobenzene and chloroform
from water across polyvinyl acetate, PTFE, or polysulfone membranes (simple or composite) also support
such a conclusion. In all cases, the separation factor was at least an order of magnitude higher for the
A-7
-------
composite membranes than the equivalent simple membrane (Zhu, 1983). In addition, the composite
membranes exhibited greater structural stability (Zhu, 1983).
Blume (1990) compared the permeability and separation factor for trichloroethane, ethyl acetate, acetone
and ethanol in a pervaporation through PDMS and PDMS/polyolefin composite membranes. As
expected, the PDMS membrane exhibited permeabilities that were an order of magnitude greater than
those achieved by the PDMS/polyolefin composite, but the separation factors of the composite were an
order of magnitude greater than that of the PDMS membrane so that the flux of a penetrant through the
membranes was equivalent, and the concentration of penetrant in the downstream gas was much higher
for the composite membrane (Blume, 1990).
Film composition
Baker and co-workers at Membrane Technology & Research Inc. (1987) investigated eight membrane
materials for their permeabilities to acetone, toluene, octane, trichloroethane and nitrogen. Each
membrane was a flat sheet, cast to a thickness of 25-50 |om, and evaluated at 40 °C. Figure A-3 is taken
from Baker (1987) and presents the relationship between toluene permeability and selectivity when
toluene was removed from a nitrogen gas stream by pervaporation. Note that the permeability of PDMS
(silicone rubber) is nearly matched by that of chloroprene (Neoprene™), which exhibits a higher
selectivity than PDMS. Figure A-4, which presents equivalent data for acetone, indicates that silicone
rubber is not uniformly the best rubberized membrane for organic compound permeability (Baker, 1987).
Neoprene
Silicone
Rubber
PVC
(25% DOP)
Silicone
Polycarbonate
Fluoral
Hypalon
Nitnle Rubber
(33% PAM)
PVC
(50% DOP)
Toluene
PVC 40% DOP)
Nitnle Rubber
(21% PAM)
— 103 2
— 10J
— 10'
10°
106
Selectivity («toluene/N2)
Figure A-3. Toluene Permeability vs. Selectivity For Toluene Over Nitrogen
A-8
-------
lO*
£ Silicone
Rubber
Silicone
Polycarbonate
£ io*
£• 103 —
I 102
Nitrite Rubber
(33% PAM)
Fiuoral
(50% DOP)
PVC (25% DOP)
Nrtrile
Rubber
(21%
RAM)
HypaJon *
— 1 o2
8-
Neoprene
Acetone
PVC (40% DOP)
101S
H
TJ
3
3
— 10° 5
10°
101
10®
103
104
105
Selectivity (aacetone/^)
Figure A-4. Acetone Permeability Vs. Selectivity For Acetone Over Nitrogen
Inner versus outer films
The question of placement of the thin film in a composite membrane relative to the liquid and gas phases
is an important one. If the film is on the same side of the membrane as the liquid, then the pores will be
gas-filled, and vice versa. Gas-phase diffusivities are several orders of magnitude larger than liquid-phase
diffusivities (Semmens, 1989). This is significant because the resistances-in-series model has been found
to be applicable to composite membrane transport. If one considers the composite membrane and the
affiliated resistance, shown in Figure A-5, one finds four potential resistances, through fluid boundary
layers, Rfl and Rp, through the film, RF, and through the porous membrane, RpM. Total resistance to
transport, RT, is equal to:
Rt = Rjx + Rp + Rpm + Rj2
(10)
Fluid 1 Film
Porous
Membrane
Fluid 2
Figure A-5. Transport of Penetrant Through Different Resistances of a Composite
Membrane
A-9
-------
Studies of the transport of organics through uncoated pervaporation membranes indicate that the liquid-
side boundary-layer mass transfer resistance is a significant fraction of the total mass transfer resistance
(Raghunath, 1992). In some cases where nonporous films coat the fiber, the film resistance dominates so
that all other resistances can be neglected, as was the case reported for transport of methylene chloride
across a PDMS/polyethylene composite and absorption into either silicone oil 200 fluid or Paratherm NF
mineral oil (Poddar, 1996a). However, using the same experimental system, the film resistance
represented only 64% of the total resistance to transport of acetone (Poddar, 1996a). In that case, the
porous membrane and liquid boundary layer were found to contribute significantly to the overall
resistance. The resistance could have been lowered had the pores been filled with gas (Das, 1998;
Semmens, 1989) instead of oil, a point made clearly by Semmens et al. (1989). In a study of
pervaporation of several organic compounds (CC14, C2C14, C2HC13, CHC13, C2H3C13) in a vapor phase
across microporous polypropylene fibers without a coating film, Semmens found that the gas-filled
membrane pores offered negligible resistance to transfer (Semmens, 1989).
Cha et al. (1997) also studied vapor-phase pervaporation (gas/vapor mixture on one side on the
membrane, vacuum on the other) where the feed side was varied from the film side to the non-film side of
a PDMS-coated polypropylene hollow fiber. Surprisingly, the fastest permeation rates were achieved
with the gas/vapor mixture exposed to the non-skin side of the composite membrane (tube side in this
case). This was attributed to condensation of the vapors within the pores and decreased pressure drop
within the pores (Cha, 1997).
Sirkar and co-workers (Das, 1998) studied trichloroethylene (TCE) removal from water utilizing
hydrophobic, microporous, polypropylene, hollow fibers (Celgard™ X-10) with a plasma-polymerized
silicone coating on the fiber outside (shell side) in pervaporation studies. The aqueous stream was fed
through the fiber (tube side) instead of the traditional shell-side contacting. TCE removal was found to be
significantly lower when the feed was on the shell side compared to tube-side feed. The most plausible
explanation proffered involved bypassing of the fluid flow on the shell side given the tight packing of the
fibers and the relatively low fluid velocity. Such shell-side bypassing has been observed with similar
modules used for liquid-liquid extraction (Tompkins, 1993).
VOC Transport
Solubility
Preferential solubility of different penetrants in the membrane determines the permeability and selectivity
of the membrane. High penetrant-membrane affinity leads to high permeation rates. Several
investigators have evaluated the effect of membrane composition on separation and permeation
performance. It has generally been established that penetrants with solubility parameters close to that of
the membrane material sorb to the polymer and permeate more rapidly than those penetrants with
solubility parameters significantly different from the membrane (Zhu, 1983). Table A-2 contains
solubility parameters (S) for several common polymers and organic compounds. The solubility
parameter, 8, is defined in Eq. (11):
Where Alf is the energy of complete vaporization of the liquid, and V1 is the liquid molar volume. Table
A-3 indicates the separation factors (a) achieved for different penetrant/membrane pervaporation
systems.
(11)
A-10
-------
Table A-2. Solubility Parameters (Zhu, 1983)
Compound
J(cal/cm3)5
Chloroform (CF)
9.3
chlorobenzene (CB)
9.5
1,2 dichloroethane
(DCE)
9.8
water
23.4
Compound
J(cal/cm3)5
PTFE
6.2
polyvinyl acetate
9.3
(PVAc)
polysulfone (PSF)
10.2
cellulose acetate (CA)
11.5
Table A-3. Separation Factors for Membrane/Penetrant Pervaporation Systems (Zhu,
1983)
Polymer
a
CF CB
DCE
PTFE
5-19
7-16
CD
I
CM
PVAc
50-75
46-55
45- 60
PSF
3-10
1-8
5-7
CA
1
N/A
N/A
Watson and Payne (1990) evaluated the separation factor and permeability of silicone rubber membranes
(0.2 mm thickness) for ^-alcohols (Cj-C10) and a variety of other organic compounds in dilute aqueous
solutions using pervaporation. The separation factor (a) was shown to rise monotonically with alcohol
chain length (Watson, 1990), as shown in Table A-4. Included in Table A-4 are diffusion coefficients, D,
measured at 80+ 3 °C in a 0.8 mm thick silicone rubber membrane.
Table A-4. Separation Factor Dependence on Chain Length for n-Alcohols
n-Alcohol
Feed concentration
(%v)
a
D (x10"10mz-sec"1)
Methanol
1.0
9
10.0
Ethanol
1.0
17
7.1
Propanol
1.0
67
6.2
Butanol
1.0
74
5.5
Hexanol
0.5
1050
4.2
Octanol
0.05
3100
3.9
Decanol
0.005
5000
2.5
A-ll
-------
For pervaporation, since the downstream pressure of the penetrant is essentially zero, Eq. (4) reduces to
D S
_ A_ A_
aAB ~ D x . The increase in separation factor exhibited by the ^-alcohols in the face of declining
B B
diffusivities, shown in Table 4, indicates that the increase in separation factor measured through the
s D A
silicone rubber membrane is driven by the selectivity factor, —, not the diffusivity factor, ~J~ . The
B B
results of further investigations are presented in Table A-5. The solubility factor (A ) in Table 5 is based
on the Scatchard-Hildebrand activity-solubility equation (Hildebrand, 1970), which allows the
development of a relationship for the partition coefficient, P, of component between an aqueous
solution (w) and a membrane (m), as defined in Eq. (12):
const x e
^ " 8if ~ (Sm " 4 / ] (13)
As the solubility factor A, defined in Eq. (13), increases, the partitioning of component /' into the
membrane also increases.
Table A-5. Separation and Solubility Factors for a PDMS Membrane Pervaporation System
Compound
Feed concentration
(%v)
a
separation
factor
A (cal/cm3)
solubility factor
Methanol
1.0
9
-2.3
Ethanol
1.0
17
5.0
Phenol
1.16a
97
24
Acetone
1.0
170
35
Octanol
0.05
3100
64
Nitrobenzene
0.1
4200
62
Chloroform
0.01
15,000
50
Benzene
0.1
20,000
69
Toluene
0.01
36,000
81
Ethylbenzene
0.01
43,000
97
Percent mass
The data presented in Tables A-4 and A-5 indicate that, for poly(dimethylsiloxane) (PDMS), solubility
selectivity is the controlling factor in determining permeability, a conclusion which is in agreement with
other findings (Bell, 1988).
A-12
-------
Operating Conditions
Leemann (Leemann, 1996) investigated the temperature and concentration dependence of the
permeability of pure PDMS hollow fibers (OD approximately 1.0 mm, ID approximately 0.85 mm, length
= 250 mm) operated as pervaporation membranes. The permeabilities for toluene, /;-xylene, ethyl acetate,
MEK, acetone and ethanol vapors decrease with increasing temperature, exhibiting Arrhenius dependence
on temperature (Leemann, 1996). This is due to the strongly exothermic nature of absorption of organic
vapors into PDMS (Leemann, 1996). Similar temperature dependence was noted for pervaporation of
carbon dioxide and methane through polypropylene and PDMS membranes coated with fluoropolymer
films (Oh, 1996), acetone and ethanol permeation through microporous glass (Qiu, 1991), and acetone
pervaporation through silicone rubber (Deng, 1998)
Penetrant concentration, as indicated by the partial pressure of the penetrant in the vapor phase, has a
greater effect upon permeability through PDMS at lower temperatures (below 50 °C). At low feed-gas
temperatures, permeability increases with penetrant partial pressure, presumably due to increasing
diffusivity due to polymer swelling (Leemann, 1996). This was noted specifically for PDMS-
polypropylene composite membranes with permeation of toluene, methanol, acetone and methylene
chloride (Cha, 1997). However, studies of gas-liquid pervaporation systems indicate that as organic
concentration increases in the aqueous feed phase, the separation factor falls dramatically for silicone
rubber membranes (Seok, 1987; Ishihara, 1986; Leeper, 1987). The separation factor has also been found
to decrease with increasing aqueous phase concentration of the solute until a limiting value is reached for
a wide variety of organic compounds ( Zhu, 1983; Watson, 1990, Seok, 1987).
The picture is rendered more complex when multiple organic vapors are considered. As different
penetrants exhibit different permeation rates, the more slowly permeating compound will build up on the
upstream membrane surface (Feng, 1992; Haraya, 1987; Psaume, 1988) diluting the faster penetrant at the
membrane surface and resulting in lower permeation rates for the faster penetrant. A parametric study
was undertaken to investigate the significance of this phenomenon, termed concentration polarization
(Feng, 1992). It was determined that (1) concentration polarization is significant for highly permeable
and selective membranes, (2) separation is effective for low vapor content, and (3) for highly selective
membranes, variable permselectivity has little effect on permeant concentration but greatly influences
permeant flux (Feng, 1992).
However, Ji et al. (1994) found no decrease in membrane permeability for multiple VOCs in the liquid
phase (toluene, trichloroethane, methylene chloride) through PDMS; instead, they found that for dilute
liquid mixtures, downstream VOC dilution occurred, increasing permeability of all VOCs by increasing
the driving force (Ji, 1994b). Similar results are reported for polyurethane and polyether-Woc£-
polyamides (Ji, 1994a).
Pressure also plays a role in permeability though the picture is not quite clear. Nagai (Nagai, 1994)
investigated the permeation properties of brominated PTMSP (Br-PTMSP) and untreated PTMSP above
and below the glass-transition temperature. In both regions, for both polymers, permeability decreased
with increasing upstream pressure for propane and carbon dioxide. Conversely, Strathmann reports
increased permeability through PDMS as upstream pressure increases for octane, toluene,
trichloromethane and acetone, up to 10-15 cmHg (Strathmann, 1986).
Membrane-Mediated Absorption
Kamalesh Sirkar's research group has investigated membrane-mediated absorption for VOC removal from
gas streams (Poddar, 1996b; Poddar, 1996a, Poddar, 1997). Hollow fibers constructed of microporous,
A-13
-------
hydrophobic Celgard™ X-10 polypropylene were used throughout. On the fibers of some of these
modules, an ultrathin (approximately 1 |a,m), plasma-polymerized, nonporous, PDMS coating was placed
on the shell side of the fibers (Advanced Membrane Technologies, Inc, Minnetonka, Minn.). A VOC-N2
gas mixture was pumped through the tube side of the fibers, and an extracting liquid was pumped
countercurrently in the shell space. Three absorbents were used; silicone oils 50-cs, 200 (Dow Corning,
Midland, Mich.) and a mineral-oil-based fluid, Paratherm NF (Paratherm Corp., Conshohocken, Pa.).
The VOC-N2 mixtures were supplied as standard cylinders (Matheson, E. Rutherford, N.J.) containing
relatively high VOC concentrations (993 ppmv acetone, 999 ppmv dichloromethane, 514 ppmv methanol,
236 ppmv toluene). At gas residence times between 1 and 1.5 seconds in uncoated fibers,
dichloromethane and toluene exit gas concentrations were lowered to 1-2 ppmv using fresh silicone oil
flowing at approximately 5% of the gas flow rate (Poddar, 1996a). Similar performance was achieved
using Paratherm NF with toluene, but the silicone oil performance was significantly superior for gas
residence times between 0.2 and 0.5 seconds.
Using coated fibers, residence times between 5 and 7 seconds are required to reduce dichloromethane to
similar concentrations in the exit gas with fresh silicone oil. However, the difference between
Paratherm™ NF and silicone oil is negligible to gas residence times down to a single second. A mixed
VOC-N2 stream was evaluated, with the results summarized in Table A-6. What is obvious is that longer
residence times are required for high removal percentage as VOC concentration decreases.
Table A-6 Absorption Data for VOC-N2 Gas Mixture
VOC feed concentration
ppmv
Removal % at given gas contact
times
17 seconds 5 seconds
Acetone
226
93.8
61.5
Methylene
chloride
201
95.5
91.0
Toluene
204
100.0
100.0
Methanol
163
52.7
21.5
Total
794
87.4
70.6
This absorption system was also operated in an absorbent recycle mode with an uncoated hollow-fiber
module used to remove VOC from the gas stream (Poddar, 1996b). After extracting VOC from the air,
the absorbent flows through the coated, hollow-fiber module membrane operated as a pervaporator. The
absorbent then flows into a storage vessel before recirculation through the extracting module. The VOC-
N2 mixtures were supplied in standard cylinders (Matheson, E. Rutherford, N.J.) containing relatively
high VOC concentrations (993 ppmv acetone, 999 ppmv dichloromethane, 514 ppmv methanol, 236
ppmv toluene). At gas residence times of 1-1.5 seconds in silicone oil flowing at approximately 5% of
the gas flow rate through the extraction module, the dichloromethane exit gas concentration was
approximately 50 ppmv and the toluene exit gas concentration was about 25 ppmv (Poddar, 1996b).
Comparison of these numbers to the 1-2 ppmv achieved with fresh absorbant indicates that residence
times greatly in excess of 1 minute will be required for removal percentages near 90%.
For a given VOC and process condition, silicone oil provided higher removal efficiency than
Paratherm™ However, silicone oil exhibits deterioration over an 18 month period causing leakage of the
A-14
-------
oil into the gas stream. For that reason, Paratherm™ is the preferred oil for these applications (Poddar,
1996a).
BIOTREATMENT MODULE
Biofouling, in the form of a biofilm colonizing a solid surface, has been a significant problem for many
important industrial and medical systems, including water supplies, heat-transfer units, ship hulls, and
implanted medical prostheses. To suppress the formation or growth of biofilms in these instances,
application of biocides has been the dominant mode of operation. The following review will instead
focus on efforts involving reduction of biofouling in biotreatment systems, where the emphasis is on
maintenance of an active film, and avoidance of excess biofilm thickness.
Membrane Biofouling
Biological systems for the treatment of contaminated water or gas streams generally involve columns
packed with solids operated either as a static or a fluidized bed. The biomass in these systems is generally
in the form of a biofilm attached to a solid surface. Excess biomass concentration in the bed retards mass
transfer (oxygen, inorganic ions, carbon and energy sources), blocks liquid and gas flow, and generally
leads to a loss in reactor productivity. To maintain the reactor activity, two general approaches to
prevention of biomass clogging have been evaluated; cleaning and metabolic strategies to suppress
biomass growth rate.
Cleaning
Cleaning strategies involve the application of high shear forces or a cleaning treatment to remove a
substantial fraction of the biofilm material. These approaches act by disrupting some aspect of the
biofilm structure. Characklis (Characklis, 1981) was among the first to characterize the shear stresses
(normal and parallel to the surface) that contributed to erosion of biofilms. Rittmann (Rittmann, 1980)
developed a series of equations describing the friction factor and resulting shear stresses generated by
liquid flow through a packed bed. These equations were modified, using data generated by Characklis, to
calculate biofilm loss rates as a function of biofilm thickness, biofilm density, and the shear stress
(Rittmann, 1982).
Stress is applied to solid packing to remove unwanted biomass in several different ways. Wubker
(Wubker, 1997) used a screw stirrer to periodically mix the polyamide-bead packing within a trickle-bed
reactor and found that the amount of biomass removed was directly related to the magnitude of the
applied shear stress (stirrer rotation rate) and the total stress applied (stirring time). Daily stirring was
sufficient to maintain maximal toluene degradation in this reactor. Taylor et al. (Taylor, 1996)
determined that weekly removal, washing, and reinoculation of packing material maximized the
productivity of an ethanol fermentation process. Smith etal. (Smith, 1996) determined that the efficient
operation of a highly VOC-loaded biofilter could be extended indefinitely if a backwash system that
expanded the bed by roughly 40% was used twice weekly. The expansion of the bed led to vigorous
mixing of the beads, shearing film from the bead surface. Backwashing, however, is necessarily limited
to those biofilters containing packing that can be fluidized.
Weber and Hartmans (Weber, 1996) reported chemical washing of a biotrickling filter. Every two weeks
a 0. lMNaOH solution was flushed through the system for 3 hours. Curiously, the toluene removal rate
of the bed after the washing was 50% higher than a similar bed that was unwashed. Loss of activity was
observed immediately following washing, but recovered to pre-wash levels within 24 hours (Weber,
1996). This approach has significant advantages relative to fluidization, because of its applicability to
A-15
-------
beds that are difficult to fluidize and that utilize fluid flow rates much lower than required for fluidization
(Cox, 1998).
Biofllm structure can also be disrupted ionically. The extracellular polysaccharide that forms the bulk of
the biofllm is crosslinked by divalent cations (primarily Ca2+, to a lesser extent Mg2+). Release of biofilm-
bound calcium could lead to biofllm dissolution or detachment from the surface. Gross biofllm
detachment was observed by Turakhia (Turakhia, 1983) following exposure of a biofllm to a pulse of a
calcium chelating agent, (ethylene glycol)bis(2-aminoethyl ether) (EGTA).
Metabolic strategies
All metabolic strategies are predicated on discovering a nutrient limitation that has a minimal negative
effect on biodegradative activity but causes a significant growth rate decrease by increasing the
maintenance energy requirement. For example, Schonduve (Schonduve, 1996) determined that while
using nitrate as a nitrogen source rather than ammonium caused a noticeable decrease in mixed culture
biomass formation rate, the degradation rate was more severely depressed. Smith (Smith, 1996) did find
that for a toluene-degrading mixed culture, the use of nitrate in place of ammonium led to a 50% decrease
in biomass yield without affecting degradation rates at all.
Addition of 0.4MNaCl led to a 32% greater decrease in biomass formation rate than degradation rate
(Schonduve, 1996). Similar results were reported for 30g/L NaCl (Strachan, 1996). Potassium and
phosphate limitation were each shown to increase the specific butanol degradation rate of a mixed culture
while decreasing the biomass yield by as much as a factor of five (Wubker, 1996), and potassium was
shown to increase the specific toluene degradation rate while decreasing the biomass yield (Wubker,
1997).
A-16
-------
References
Ashworth, A. J., B.J. Brisdon, R. England, B.S.R. Reddy and I. Zafar (1991). "The permeability
of polyorganosiloxanes containing ester functionalities." J. Membrane Sci. 56: 217.
Ashworth, A. J. (1992). "Relation between gas permselectivity and permeability in a bilayer
composite membrane." J. Membrane Sci. 71: 169-173.
Baker, R. W., N. Yoshioka, J.M. Mohr and A.J. Kahn (1987). "Separation of organic vapors
from air." J. Membrane Sci. 31: 259.
Bell, C. M., F.J. Gerner, and H. Strathmann (1988). "Selection of polymers for pervaporation
membranes." J. Membrane Sci. 36: 315.
Bessarabov, D. G., E.P. Jacobs, R.D. Sanderson and I.N. Beckman (1996). "Use of nonporous
polymeric flat-sheet gas-separation membranes in a membrane-liquid contactor:
experimental studies." J. Membrane Sci. 113: 275-284.
Blume, I., J.G. Wijmans, and R.W. Baker (1990). "The separation of dissolved organics from
water by pervaporation." J. Membrane Sci. 49: 253-286.
Carman, P. C. (1952). "Diffusion and flow of gases and vapors through micropores." Proc. Roy.
Soc. Ser. 211 A: 526.
Cha, J. S., R. Li and K.K. Sirkar (1996). "Removal of water vapor and VOCs from nitrogen in a
hydrophilic hollow-fiber gel membrane permeator." J. Membrane Sci. 119: 139.
Cha, J. S., V. Malik, D. Bhaumik, R. Li and K.K. Sirkar (1997). "Removal of VOCs from waste
gas streams by permeation in a hollow fiber permeator." J. Membrane Sci. 128: 195.
Characklis, W. G. (1981). "Fouling biofilm development - A process analysis." Biotechnol.
Bioeng. 23: 1923.
Cox, H. H. J., and M.A. Deshusses (1998). "Biological waste air treatment in biotrickling
filters." Current Opinion in Biotechnology 9: 256.
Das, A., I. Abou-Nemeh, S. Chandra and K.K. Sirkar (1998). "Membrane-moderated stripping
process for removing VOCs from water in a composite hollow fiber module." J. Membrane
Sci. 148: 257.
Deng, S., A. Tremblay and T. Matsuura (1998). "Preparation of hollow fibers for the removal of
volatile organic compounds from air." J. Appl. Polym. Sci. 69: 371-379.
Feng, X., and R.Y.M. Huang (1992). "Organic vapor/gas mixture separation by membrane—a
parametric study." Separation Sci. Tech. 27(15): 2109.
A-17
-------
Freeman, B. D., and I. Pinnau (1997). "Separation of gases using solubility-selective polymers."
Trends in Polymer Sci. 5: 167.
Frisch, H. L., and S.A. Stern (1983). "Diffusion of small molecules in polymers." CRC Crit. Rev.
Solid State Material Sci. 11: 123.
Fujita, H. (1968). Organic vapors above the glass transition temperature. Diffusion in Polymers.
New York, Academic Press. 75.
Gray, D. N. (1976). "Olefin/sulfur dioxide copolymers." Polymer News 3: 141.
Gray, D. N. (1984). Polymeric membranes for artificial lungs. Polymeric materials and artificial
organs. American Chemical Society. 151.
Haraya, H., T, Hakuta, H. Yoshitome, and S. Kimura (1987). Sep. Sci. Technol. 22: 1425.
Henis, J. M. S., andM.K. Tripodi (1980). "Multicomponent membranes for gas separation." US
Patent 4,230,463:
Henis, J. M. S., and M.K. Tripodi (1981). "Composite hollow fiber membranes for gas
separation: The resistance model approach." J. Membrane Sci. 8: 233.
Hildebrand, J. Ft., J.M. Prausnitz, and R.L. Scott (1970). Regular and Related Solutions. New
York, Van Nostrand Reinhold Co.
Ichiraku, Y., S.A. Stern and T. Nakagawa (1987). "An investigation of the high gas permeability
of poly(l-trimethylsilyl-l-propyne) " J. Membrane Sci. 34: 5.
Ishihara, K., Y. Nagesse, and K. Matsui (1986). "Pervaporation of alcohol water mixtures
through PTMSP." Makromol. Chem. Rapid Commun. 7: 43.
Ji, W., S.K. Sikdar, and S.-T. Hwang (1994a). "Modeling of multicomponent pervaporation for
removal of volatile organic compounds from water." J. Membrane Sci. 93: 1.
Ji, W., A. Hilaly, S.K. Sikdar, and S.-T. Hwang (1994b). "Optimization of multicomponent
pervaporation for removal of volatile organic compounds from water." J. Membrane Sci. 97:
109.
Kimmerle, K., T. Hofmann and H. Strathmann (1991). "Analysis of gas permeation through
composite membranes " J. Membrane Sci. 61: 1-17.
Koros, W. J., and M.W. Heliums (1989). Transport Properties. Encyclopedia of Polymer Science.
New York, Wiley-Interscience Publishers. 724.
A-18
-------
Koros, W. J., M.R. Coleman and D.R.B. Walker (1991). "Controlled permeability polymer
membranes." Annual Review of Materials Science 22: 47-90.
Koros, W. J., and G.K. Fleming (1993). "Membrane-based gas separation." J. Membrane Sci. 83:
1-80.
Kulkarni, S. S., and S.A. Stern (1983). "The diffusion of CO2, CH4, C2H4, and C3H8 in
polyethylene at elevated temperatures." J. Polym. Sci., Polym. Phys. Ed. 21: 441.
Kumins, C. A. and T. K. Kwei (1968). Free volume and other theories. Diffusion in Polymers.
New York, Academic Press. 107-140.
Lee, C.-L., H.L. Chapman, M.E. Cifuentes, K.M. Lee, L.D. Merril, K.L. Ulman and K.
Venkataraman (1988). "Effects of polymer structure on the gas permeability of silicone
membranes." J. Membrane Sci. 38: 55.
Lee, K.-FL, and S.-T. Hwang (1986). "The transport of condensable vapors through a
microporous Vycor glass membrane." J. Colloid Interface Sci. 110: 544.
Leemann, M., G. Eigenberger, and H. Strathmann (1996). "Vapour permeation for the recovery
of organic solvents from waste air streams: separation capacities and process optimization."
J. Membrane Sci. 113: 313.
Leeper, S. A. (1987). Membrane separations in the production of alcohol fuels by fermentation.
Membrane separations in biotechnology. New York, N.Y., Marcel Dekker. 161.
Loeb, S., and S. Sourirajan (1963). "Sea water demineralization by means of an osmotic
membrane." ACSSymp. Ser. 38: 117.
Lund, L. W., W.J. Federspiel and B.G. Hattler (1996). "Gas permeability of hollow fiber
membranes in a gas-iquid system." J. Membrane Sci. 117: 207-219.
Masuda, T., E. Isobe and T. Higashimura (1983). "Poly l-(trimethylsilyl)-l-propyne: a new high
polymer synthesized with transition-metal catalysts and characterized by extremely high gas
permeability"./. Am. Chem. Soc. 105: 7473.
Masuda, T., Y. Iguchi, B.-Z. Tang and T. Higashimura (1988). "Diffusion and solutions of gases
in substituted polyacetylene membranes." Polymer 29: 2041.
McKee, R. L., M.K. Changela and G.J. Reading (1991). "Carbon dioxide removal: Membrane
plus amine." Hydrocarbon Processing 70(4): 63-71.
Nagai, K., A. Higuchi and T. Nakagawa (1994). "Gas permeation and sorption in brominated
poly(l-trimethylsilyl-l-propyne) membrane." J. Appl. Polym. Sci. 54: 1353.
A-19
-------
Nagai, K., A. Higuchi and T. Nakagawa (1995). "CAS Permeability and stability of poly(l-
tri methyl si lyl -1 -propyne-co-1 -phenyl-1 -propyne) membranes." J. Polym. Sci., Part B:
Polym. Phys. 33: 289.
Nagai, K., M. Mori, T. Watanabe and T. Nakagawa (1997). "Gas permeation proerties of blend
and copolymer membranes composed of 1-trimethylsilyl-l-propyne and 1-phenyl-1-propyne
structures." J. Polym. Sci. PartB:Polym. Phys. 35: 119.
Odani, H., and T. Masuda (1992). Design of polymer membranes for gas separation. Polymers
for Gas Separation. New York, VCH. 107.
Oh, S.-J., and W.P. Zurawsky (1996). "Gas permeation through poly(dimethylsiloxane) -plasma
polymer composite membranes." J. Membrane Sci. 120: 89.
Parthasarathy, A., C.J. Brumlik, C.R. Martin, and G.E. Collins (1994). "Interfacial
polymerization of thin polymer films onto the surface of a microporous hollow-fiber
membrane." J. Membrane Sci. 94: 249-254.
Pinnau, I., H.J. Wijmans, I. Blume, T. Kuroda and K.V. Peinemann (1988). "Gas permeation
through composite membranes." J. Membrane Sci. 37: 81.
Plate, N. A., A.K. Bokarev, N.E. Kaliuzhnyi, E.G. Litvinova, V.S. Khotimski, V.V. Volkov and
Y.P. Yampol'skii (1991). "Gas and vapor permeation and sorption in poly(trimethylsilyl-
propyne)" J. Membrane Sci. 60: 13.
Poddar, T. K., S. Majumdar, and K.K. Sirkar (1996a). "Membrane-based absorption of VOCs
from a gas stream." AIChE Journal 42(11): 3267.
Poddar, T. K., S. Majumdar and K.K. Sirkar (1996b). "Removal of VOCs from air by
membrane-based absorption and stripping." J. Membrane Sci. 120: 221.
Prasad, R., F. Notaro, and D.R. Thompson (1994). "Evolution of membranes in commercial air
separation." J. Membrane Sci. 94: 225-248.
Psaume, R., P. Aptel, Y. Aurelle, J.C. Mora, and J.L. Bersillon (1988). "Pervaporation:
importance of concentration polarization in the extraction of trace organics from water." J.
Membrane Sci. 36: 373.
Qiu, M. M., and S.-T. Hwang (1991). "Continuous vapor-gas separation with a porous
membrane permeation system." J. Membrane Sci. 59: 53-72.
Raghunath, B., and S.-T. Hwang (1992). "Effect of boundary layer mass transfer resistance in
the pervaporation of dilute organics." J. Membrane Sci. 65: 147.
Rhim, H., and S.-T. Hwang (1975). "Transport of capillary condensate." J. Colloid Interface Sci.
52: 174.
A-20
-------
Rittmann, B. E., and P.L. McCarty (1980). "Model of steady-state-biofilm kinetics." Biotechnol.
Bioeng. 22: 2343.
Rittmann, B. E. (1982). "The effect of shear stress on biofilm loss rate." Biotechnol. Bioeng. 24:
501.
Schonduve, P., M. Sara, and A. Friedl (1996). "Influence of physiologically relevant parameters
on biomass formation in a trickle-bed bioreactor used for waste gas cleaning." Appl.
Microbiol. Biotechnol. 45: 286.
Semmens, M. J., R. Qin, and A. Zander (1989). "Using a microporous hollow-fiber membrane to
separate VOCs from water." JournalAWWA : 162.
Seok, D. R., S.G. Kang, and S-T. Hwang (1987). "Use of pervaporation for separating azeotropic
mixtures using two different hollow fiber membranes." J. Membrane Sci. 33: 71.
Shaver, K. G., G.L. Poffenbarger and D.R. Grotewald (1991). "Membranes recover hydrogen."
Hydrocarbon Processing 70(4): 77-82.
Singh, A. (1997). "Gas and vapor sorption and permeation properties of high free volume glassy
polymers." Ph.D. thesis, North Carolina State University.
Smith, F. L., G.A. Sorial, M.T. Suidan, A.W. Breen, and P. Biswas (1996). "Development of two
control strategies for extended, stable operation of highly efficient biofilters with high
toluene loadings." Environ. Sci. Technol. 30: 1744.
Sok, R. M., and H.J.C. Berendsen (1992). "Molecular dynamics simulation of the transport of
small molecules across a polymer membrane." Polym. Prepr. Am. Chem. Soc., Div. Polym.
Chem. 33: 641.
Stern, S. A., V.M. Shah and B.J. Hardy (1987). "Structure/permeability relationships in silicone
polymers." J. Polym. Sci. Part B:Polym. Phys. 25: 1263.
Stern, S. A. (1994). "Polymers for gas separations: the next decade." J. Membrane Sci. 94: 1-65.
Strachan, L. F., L.M. Freitas dos Santos, D.J. Leak, and A.G. Livingston (1996). "Minimisation
of biomass in an extractive membrane bioreactor." Wat. Sci. Tech. 34\ 273.
Strathmann, H., C.M. Bell, and K. Kimmerle (1986). "Development of synthetic membranes for
gas and vapor separation." Pure & Appl. Chem. 58: 1663.
Suwandi, M. S., and S.A. Stern (1973). "Transport of heavy organic vapors through silicone
rubber." J. Polym. Sci., Polym. Phys. Ed. 11: 663.
A-21
-------
Takada, K., H. Matsuya, T. Masuda and T. Higashimura (1985). "Gas permeability of
polyacetylenes carrying substituents."./. Appl. Polym. Sci. 30: 1605.
Taylor, F., M.J. Kurantz, N. Goldberg, and J.C. Craig, Jr. (1996). "Control of packed column
fouling in the continuous fermentation and stripping of ethanol." Biotechnol. Bioeng. 51(1):
33.
Tompkins, C. J., A.S. Michaels, and S.W. Peretti (1993). "Removal of/>nitrophenol from
aqueous solution by membrane-membrane solvent extraction." J. Membrane Sci. 75: 277.
Turakhia, M. H., K.E. Cooksey, and W.G. Characklis (1983). "Influence of a calcium-specific
chelant on biofilm removal." Appl. Environ. Microbiol. 46: 1236.
Vos, K. D., and F.O. Burris (1969). "Drying of cellulose acetate reverse osmosis membranes."
Ind. Eng. Chem., Proc. Res. Dev. 8: 84.
Ward, W. J., W.R. Browall and M. Salemme (1976). "Ultrathin silicone/polycarbonate
membranes for gas separation processes." J. Membrane Sci. 1: 99.
Watson, J. M., and P. A. Payne (1990). "A study of organic compound pervaporation through
silicone rubber." J. Membrane Sci. 49: 171-205.
Weber, F. J., and S. Hartmans (1996). "Prevention of clogging in a biological trickle-bed reactor
removing toluene from contaminated air." Biotechnol. Bioeng. 50(1): 91.
Wubker, S.-M., and C.G. Friedrich (1996). "Reduction of biomass in a bioscrubber for waste gas
treatment by limited supply of phosphate and potassium ions." A pp. Microbiol. Biotechnol.
46: 475.
Wubker, S.-M., A. Laurenzis, U. Werner, and C. Friedrich (1997). "Controlled biomass
formation and kinetics of toluene degradation in a bioscrubber and in a reactor with a
periodically moved trickle-bed." Biotechnol. Bioeng. 55(4): 686.
Zhu, C. L., C.-W. Yuang, J.R. Fried and D.B. Greenberg (1983). "Pervaporation membranes—A
novel separation technique for trace organics." Environ. Prog. 2(2): 132.
Zielinski, J. M., and J.L. Duda (1992). "Predicting polymer/solvent diffusion coefficients using
free-volume theory." AIChE J. 38: 405.
A-22
-------
APPENDIX B
BENCH-SCALE DATA
-------
Table
B-1. AMT Module, Test 1, 0.5ft3/min Air, 1.0 L/min
Oil, 500 ppm
VOCs
;042700 ;
Test 1 i i ; ! : i ?
Test Date : 4/27/00 ¦" I
|Initial Settings: magnehelic (1) ~1.0"H2O;positive flow measurement 50 cc/min;heater set ~280C
• ;magnehelic(2) > 1,0"H20;syringe flow 2.0ml/hr;
n-Butyl
n-Butyl
Ethyl
Ethyl
M EK
M EK
Acetate
Acetate
Benzene
Benzene
o-Xylene
o-Xy lene
Front
Rear
%
Front
Rear
%
Front
Rear
%
Front
Rear
%
Detector
Detector
Difference
Detector
Detector
Difference
Detector
Detector
Difference
Detector
Detector
Difference
150
315
130.062
13.47371
89.126
67.699
24
041245
35
881
26
768'
25
39784
214
818
155
381
27.66854
2
153
145
138.194
9.762643
91.612
70.278
23
287342
36
914
27
742'
"24
84694
221
479
161
184
27.2238
3
143
296
137.945
3.734228
87.884
72.005
18
068135
35
411
28
437
19
69445
"213
098
165
741
22.2231 1
4
148
908
136.583
8.276923
88.65
72.437
18
'288776
35
749
28.79'
'19
46628
214
407
168
397
"21.45919
5
' 145
134
137.084
5.546598
87.979
73.853
16
056104
"35
439
29
334'
17
22678
213
086
'172
465
19.0632
6
142
971
132.215
7.523204
"86.233
73.949
14
245127
34
667
29.51
14
87582
208
276
174
276
16.32449
7
145.94
138.263
'5.260381
88.783
75.424
15
046799
35
689
'30
021'
15
88164
214
788
1 77
675
17.2789
8
143
996
1 35.1 88
6.1 16837
" 87.305
75.6
13
407021
35
105
30
177
14
03789
210
832
1 78
909
15.14144
9
146
268
135.548
7.329012
88.615
75.617
14
667946
35
605
30
226
15
10743
213
969
179
796
1 5.97101
10
134
227
129.722
3.356255
83.79
75.48
9.9176513
33
638
30
224'
' 10
14924
202
525
180
798
10.72806
1 1
136
837
123.254
' 9.926409
82.883
' 73.603
1
196506
33.3
29
676'
10
88288
200
097
178
365
'10.86073
12
150
697
102.9
31.71729
90.417
46.423
48
656779
36
409
19
275
47
05979
218
704
1 15
088
47.37728
13
175
156
1 10.369
36.98817
103.662
44.659
56
918639
41
771
18
712
55
20337
250
329
108
054
56.8352
14
191
428
123.568
35.44936
1 12.785
47.188
58
161103
45
351
19
796
56
34936
271
844
112.28
58.6969
15
166
186
126.673
23.77637
102.004
48.609
52
345986
40
935
20
412
50
13558
246
625
113
708
53.89437
16
136
714
112.95
' 17.38227
87.39
45.901
47
475684
34.99
19
489
44
30123
"21 1
471
109
703
48.12386
17
130
304
108.047
17.08083
81.858
44.409
' 45
748736
32
796
18
732
42
88328
197.83
103
539
47.66264
18
' 142
684
105.631
25.96857
85.938
42.426
50
631851
34
525
18
088
47
60898
' 207.39
"100
128
51.71995
19
162
393
" 121.14
25.40319
96.488
45.4'
52
947517
38
805
19
129
50
70481
232
733
104
992
54.88736
20
148
302
1 19.684
19.2971 1
90.879
45.541
49
888313
36
466
19
156
47
46888
219
457
105
588
51.8867
21
146
432
117.69
19.62822
89.249
44.893
49
699156
35
766
18
913'
47
12017
215.25
103
787
51.78304
22
135
969
' 1 18.304
12.99193
85.13
46.149
45
789968
34
134
19
293
43
47864
' 205
983
106
105
48.48847
23
142.63
116.638
18.22338
86.986
44.625
48
698641
34
868
18
843
45
95905
209.62
102
297
51.19884
24
154
354
"126.517
18.03452
93.55
47.05
49.70604
37
513
19
692
47.5062
225
313
' 106
'584
52.69514
25
142
248
' 122.586
13.82234
88.443
47.209'
"46
622118
35
447
19
741
44.3084
213
'457
107
'435
49.66902
26
139
154
118.97
'14.50479
85.307
46.315'
45
707855
34
188
19
384
'43
30174
"205
426
'105
148
'48.81466
27
138
814
122.3
1 1.89649
86.002
47.694'
44.54315
34
458
19
876
42
31818
207.62
107
965
47.99875
28
' "131
055
117.642
10.23463
81.987
47.276
42.3372
32
867
19
863
39
56552
198
192
107
077
45.9731
29
151
703
125.706
17.13677
90.754
47.488'
47
673932
36
491
19
678
'46
07437
218
836
106.18
51.47965
30
137
'319
"123.105
' 10.35108
85.575
48.112
43
777973
" 34
342
20
001
41
75936
"207
071
108
561
47.57305
31
142
491
122.195
14.24371
86.57
47.598
45
017905
34
767
19
812
43
01493
209
006
106
946
48.831 13
32
141
586
126.176
10.88384
87.1 18
' 49.108
43
630478
34
911
20
349
41
71178
210
171
109
923
47.6983
33
135
562
121.383
' 10.45942
83.925
47.712
43.14924
33
653
19
849
41
01863
202
695
106
913'
47.25425
34
152
744
126.865
16.94273
90.981
48.734
46.43497
36
537
20
208
44
69168
219
093
108
378
50.53334
35
148
702
132.139
1 1.13838
91.053
50.868
44
133636
36
501
21
014
42
42897
219
446
112
885
48.5591
B-l
-------
MEK inlet
MEK outlet
Butyl acetate inlet
Butyl acetate outlet
Ethylbenzene inlet
Ethylbenzene outlet
Xylene inlet
Xylene outlet
x-*
¦X X x x-*-*-*-*
50
X X X >< X X x )< *-x-X-X
* X XX X
10
15
20
Sample #
25
30
35
40
Figure B-1: AMT Module, Test 1, 0.5 ft3/min Air, 1.0 L/min Oil, 500 ppm VOCs
-------
Table B-2. AMT Module, Test 2, 0.5 ft3/min Air, 1.0 L/min Oil, 2100 ppm VOCs
;Test 2
iRun Date: • 4/27/00; ;
^sequence 042700a • • • ;
|Initial settings: ;sy ringe pum p set at 2.0 m l/hr;0.5cfm ;tem p controller
• I290C;magnehilic (1) rding 1 "H20;magnehelic (2) rding >1"H20;
!Gen notes did not experience any software errors during sampling I
ishould have adjusted syringe sam pling rate to a lower setting (1/4 of what it was set) as a result our num bers are
•m uch higher than we expected. ; ;
n-Butyl |n-B uty I : Ethy I iEthyl i
MEK :.MEK 'Acetate :Acetate ; jBenzene :Benzene :o-Xylene ;o-Xylene
Front -Rear \% .Front iRear \% :Front IRear \% .Front 'Rear i%
•Detector
Detector
Difference .
Detector
Detector
Difference
Detector
Detector
Difference
Detector
Detector
Difference
1
681.
.961'.
377
.218
44.
.68628"
410.
.593
101.
.304
75.
.327392;
171.
.399
38.
.443'
77.57105'
1043.
.313;
204.793'
80.3709
2;
^ 702.
.241 ;
330
.247
: 52.
.97241 '
432.
.769
87.
.043
79.88696',
179.
.723
34.
.172:
80.9863'
1091.
.209
184.561
.83.
.08656
3;
625.
.748':
361
.555'
: 42.
.22035'
384.
.748
97.
.123'
74.
.756724'.
160.54
37.
.504'
76.63884'
980.
.284
203.155
79.2759
4
' 675.
.842":
350
.573'
48.
.12796;
389.
.346
96.
.337'
' 75.
.256713':
162.
.387'
37.
.329;
77.01232
: 979.
.173
'203.841
79.
.18233
5
' 709.
.953 :
334
.148
: 52.
.93379-
436.
.724'
84.
.916
80.
.556141''
181.
.329'
33.
.055;
' 81.7707
1101.
.906
"'176.789
83.
.95607
6 ;
511
.676
389
.961 '
23.
.78751 '
339.
.472
107.
.931 '
68.
.206214':
142.
.086
41.
.344-
70.90213'
879.
.358;
' 224.101
"74.
.51539
7
609.
.509";
312
.931
¦ 48.
.65851 ;
343.
"157
94.
.794
72.37591
143.
.154
87.
.598
38.80856
; 859.
.247:
207.1
75.8975
8
' 890
.686:
295
.112
: 66.
.86689
'483.
.275
76.
.617
84
.146294';
198
.553
71.71
63.8837'
1 173.
.823-
1 59.938
86.
.37461
9;
' 693.19
379.
.609'
45.
.23738
475.43
76.
.729'
83.
.861136;
197.
.178
28.
914
85.33609
1221.
.902
144.917
' 88.
.14005
10
657.94
385.
.654'
41
.38462
381
.081
105.
.657;
72.
.274398:
158.
.654
96.
.372
' 39.2565'
959.
.683;
221.517
76.
.91769
1 1 :
; 697.
.476".
363
.508
; 47.
.88236 :
' 447.
.328
8;
3.42'
'80.
.233743;
177.
.678'
33.
.267'
81.2768
1077.
.098 :
176.13
83.
.64773
12,
698.65'
380.
.423
; 45.
.54884
442.
.492'
97.
.394
77.
989659,
175.
.995'
37.
.457
78.71701
1065.
.251
193.751
8
1.81 17
1 3'
•' 632.
.709";
394
.233
¦ 37.
.69126 :
' 407.
.762
104.
.566
74.35612''
"" 162.
.248
40.
501
75.0376
984.
.326'
210.261
'78
.63909
14:
574.
.213:
376
.799
: 34.
.37993
"365.
.677
105.
.395,
71.
.1781 16!
145.
.245
40.
.817
71.89783
: 881.
.118
216.217
; 75.
.46106
15:
672.
.908!
360
.621
: 46.
.40857
416.
.343
93.11'
77.
.636228:
165.
.348
36.
054:
78.19508
997.
.284,
187.441
: 81
.20485
16:
568.
.607!
400.99
29.
.47853,
370.
.812
107.
.192!
71.
.092629:
147.
.615
40.
.731 ;
72.40728
897.
819 ;
211.567
76.
.43545
17
578.
.692:
360
.337
37.
.73251
' 363.
.342
100.
.622'
' 72.
.306532':
144.
.152
38.
.977'
' 72.961 18
870.6
205.926
76.
.34666
18
687
.882:
358
.966
47.
.81576:
424.38
90.92'
78.
.575805:
168.
.437
35.
.188
79.1091
1013.
.724
181.146
"82.
.13064
19:
642.
.209;
' 403
.282'
¦ 37.
.20393
407.
.004
102.
.042'
74.
.928502,
161.
.922
38.
.609
76.1558'
982.
.041 '
197.743
: 79.
.86408'
20'
674.
.566:
386
.024
: 42.
.77447:
421.
.249
98.
.041;
76.
726117
167.
.319
37.
.702 ;
77.46699
1009.
.392'
193.824
80
.79795'
21;
637
.705:
397
.736
37.6301
408.
.842
100.
.912'
75.
.317604:
162.
.608'
38.
.527'
' 76.30682
986.
.365
197.63'
. 79.
.96381
22
617
.51 4:
' 384
.006
37.8142
389.
.518
102.
.316
73.
732664,
154.
.649
39.22
74.63934
935.
.273
203.762
78.
.21363
23'
708;;
375.
.488
46.
.96497
436.
.517
'94.
.715
78.
.302105':
173.
.044
36.
.539
78.88456'
1041.
.451
'186.716
: 82.
.07155
24:
541.
.946;
" 444
.311'
18.
.01563
'368.
.106
112.
.969
69.
"310742:
146.
.544
42.
.327'
71.1 1652
896.
.285;
215.61 1
.75.
.'94392
25:
398
.468;
361
.162'
9.362358
272.
.646
1 11.
.063
59.
.264761'
108.
.163
100.
.795'
6.81 1941
661.
.669'
230.568
65.
.15357'
26'
411.
.61 8;
' 320
.'555
¦ 22
.12318
266.
.'533
101.
.209 ;
62.
.027591'
105.
.'593'
39.
.148'
62.92557
640.
.'938
' 210.336
: 67.1831
27
571.
.609
314
.159
: 45.
.03953
346.
.'246
: 86.
.748;
' 74.
.946137;
136.
.853
34.
.036
75.12952
820.
.518
175.728
78
.58329
28
. 748.
.405;
347
.828
' 53.
.52409'
452.
.524
83.
.199,
81.
.614456;
178.
.974
32.
.'136'
82.04432
1070.
.831
158.686
85.
.18104
B-3
-------
1400
1200
1000
800
" E
Q.
Q.
600
400
200
0
0
5
10
15
20
25
30
Sample #
MEK inlet
MEK outlet
Butyl acetate inlet
Butyl acetate outlet
Ethylbenzene inlet
Ethylbenzene outlet
Xylene inlet
Xylene outlet
Figure B-2. AMT Module, AMT Module, Test 2, 0.5 ft3/min Air. 1.0 L/min Oil, 2100 ppm VOCs
-------
Table B-3. AMT Module, Test 3, 2.0 ft3/min Air, 1.0 L/min Oil, 450 ppm VOCs
Test 3
III I I I I I I I
Initial settings:
positive flow measured 52cc/min;magnehelic (2) rds >1.0"H20;magnehelic (1) ~1,0"H20;oil flow rate set at 110- rotameter
at 110 on rotameter;
Sequence:
042900&o42900a& 042900b
Test Date:
4/28/00
n-Butyl
n-Butyl
Ethyl
Ethyl
MEK
MEK
Acetate
Acetate
Benzene
Benzene
o-Xylene
o-Xylene
Front
Rear
%
Front
Rear
%
Front
Rear
%
Front
Rear
%
run#
Detector
Detector
Difference
Detector
Detector
Difference
Detector
Detector
Difference
Detector
Detector
Difference
1
3.757
32.314
-760.101
5.623
13.999
-148.95963
2.166
5.653
-160.988
14.411
26.11
-81.181
2
3.543
35.619
-905.334
5.246
13.385
-155.14678
2.028
6.132
-202.367
13.751
22.453
-63.2827
3
3.66
45.528
-1143.93
5.003
12.948
-158.80472
1.889
6.048
-220.169
12.645
21.451
-69.6402
4
3.177
44.738
-1308.18
4.618
12.862
-178.51884
1.741
6.088
-249.684
11.704
20.808
-77.7854
5
34.575
47.879
-38.4787
17.828
13.928
21.875701
7.569
6.53
13.72704
44.581
22.939
48.54534
6
4.057
49.392
-1117.45
5.933
14.864
-150.53093
2.194
6.917
-215.269
14.575
25.115
-72.3156
7
2.992
45.53
-1421.72
4.213
13.394
-217.92072
1.555
6.443
-314.341
10.518
21.304
-102.548
8
2.816
45.604
-1519.46
3.782
13.097
-246.29825
1.42
6.416
-351.831
9.803
20.616
-110.303
9
2.7
45.733
-1593.81
3.549
13.069
-268.24458
1.332
6.45
-384.234
9.237
20.344
-120.245
10
97.198
130.086
-33.8361
57.138
39.764
30.407085
23.454
15.821
32.54456
141.676
73.321
48.24741
11
159.303
145.773
8.493249
92.56
50.205
45.759507
37.459
19.681
47.45989
224.81
98.965
55.97838
12
13.58
69.131
-409.065
16.095
26.546
-64.933209
6.243
11.42
-82.9249
40.095
51.837
-29.2854
13
3.745
49.379
-1218.53
6.07
17.445
-187.39703
2.27
8.237
-262.863
15.338
30.816
-100.913
14
4.136
46.849
-1032.71
5.201
15.396
-196.02
1.937
7.487
-286.526
12.81
26.018
-103.107
15
124.961
125.973
-0.80985
71.011
41.594
41.425976
28.929
16.656
42.42456
173.228
79.313
54.21468
16
134.06
136.895
-2.11472
79.728
49.515
37.895093
32.236
19.527
39.42487
194.064
98.636
49.17347
17
133.615
137.501
-2.90836
80.918
51.76
36.03401
32.621
20.401
37.46053
196.848
104.484
46.92148
18
129.124
135.529
-4.96035
79.073
51.745
34.560469
31.829
20.437
35.79126
192.195
104.869
45.43615
19
138.012
140.345
-1.69043
83.759
53.479
36.151339
33.672
21.087
37.37527
203.042
108.343
46.6401
20
139.722
143.644
-2.807
86.295
54.88
36.404195
34.664
21.544
37.84907
209.467
110.974
47.02077
21
132.683
136.536
-2.90391
81.828
53.903
34.12646
32.852
21.27
35.25508
198.336
110.077
44.49974
22
126.854
121.63
4.11812
78.614
54.163
31.102603
31.518
19.206
39.06339
190.18
105.393
44.5825
23
123.283
118.879
3.572269
76.064
55.693
26.781395
30.486
21.182
30.51893
183.871
109.765
40.30326
24
129.445
123.384
4.682298
79.119
56.29
28.854005
31.762
20.995
33.899
191.522
109.63
42.75853
25
131.092
137.907
-5.19864
81.34
56.003
31.149496
32.582
21.206
34.91498
196.867
114.541
41.81808
26
124.473
121.593
2.313755
77.96
55.433
28.895587
31.21
21.175
32.15316
188.668
112.209
40.52569
27
126.922
123.426
2.754448
79.046
54.31
31.293171
31.65
21.477
32.14218
191.087
111.667
41.56222
28
128.208
131.99
-2.94989
79.707
55.551
30.305996
31.894
21.663
32.07813
192.484
113.859
40.84755
29
121.068
122.157
-0.89949
76.174
55.032
27.754877
30.475
21.553
29.27646
184.333
112.633
38.897
30
128.288
123.072
4.065852
79.282
55.14
30.450796
31.783
21.369
32.76594
191.833
112.526
41.34169
31
136.72
138.287
-1.14614
84.999
57.159
32.753327
34.055
22.414
34.18294
205.927
116.906
43.2294
32
138.344
135.802
1.837449
84.908
56.432
33.537476
34.025
22.063
35.1565
205.265
115.167
43.8935
33
143.485
138.543
3.444262
88.735
58.214
34.395673
35.513
22.648
36.22617
214.446
118.697
44.64947
34
132.866
131.286
1.189168
83.492
58.042
30.481962
33.38
22.535
32.48951
201.816
118.431
41.31734
35
140.485
140.769
-0.20216
86.925
58.208
33.036526
34.778
22.853
34.28892
210.01
119.085
43.29556
36
138.949
139.632
-0.49155
86.766
58.337
32.765138
34.715
22.639
34.78612
209.783
118.869
43.33716
37
139.296
130.346
6.425167
86.57
58.524
32.396904
34.62
22.284
35.63258
209.087
118.265
43.43742
38
142.173
130.54
8.182285
88.287
58.595
33.631225
35.291
22.583
36.00918
213.183
119.004
44.17754
39
129.481
129.824
-0.2649
81.965
58.018
29.216129
32.783
22.75
30.60428
198.621
118.828
40.1735
40
135.624
136.905
-0.94452
84.695
58.382
31.06795
33.869
22.86
32.50465
204.659
119.293
41.71133
41
129.345
130.313
-0.74839
80.506
56.508
29.808958
32.206
22
31.68975
194.441
115.524
40.5866
42
135.328
130.21
3.781922
84.266
57.367
31.921534
33.707
22.494
33.26609
203.681
116.686
42.7114
43
134.72
136.918
-1.63153
84.271
58.252
30.87539
33.678
22.784
32.34753
203.618
118.717
41.69622
44
130.53
126.024
3.45208
81.976
57.017
30.446716
32.733
22.25
32.02578
197.842
116.275
41.22835
45
134.244
131.213
2.257829
84.873
58.852
30.658749
33.906
22.755
32.88798
205.262
119.129
41.96247
46
131.524
127.052
3.40014
82.77
58.041
29.876767
33.072
22.435
32.16316
199.922
117.865
41.04451
47
137.039
127.922
6.652851
84.483
58.02
31.323462
33.773
22.48
33.43795
203.613
117.589
42.24878
B-5
-------
250
200
150
E
Q.
Q.
* iK M
100
+— MEK inlet
MEK outlet
+- Butyl acetate inlet
*- Butyl acetate outlet
*- Ethylbenzene inlet
Ethylbenzene outlet
Xylene inlet
— Xylene outlet
10
20 30
Sample #
40
50
Figure B-3. AMT Module, AMT Module, Test 2, 0.5 ft3/min Air. 1.0 L/min Oil, 2100 ppm VOCs
-------
Table B-4. AMT Module, Test 4, 2.0 ft3/min Air, 1.0 L/min Oil, 230 ppm VOCs
Test 4
I I I I I I I
Initial settings:
syringe flow set 1,0m l/hr;0.5cfm air;magnehelic (1) ~0.9"H2O;m agnehelic (2) >1.0"H20
run date: 4/28/00
i
sequence used :
042900c& d
Gen notes:
software error ocurred which automatically aborted sequence-im mediately started new sequence (042900d)
run#
M EK
Front
Detector
M EK
Rear
Detector
%
Difference
n-Butyl
Acetate
Front
Detector
n-Butyl
Acetate
Rear
Detector
%
Difference
Ethyl
Benzene
Front
Detector
Ethyl
Benzene
Rear
Detector
%
Difference
o-Xy lene
Front
Detector
o-Xy lene
Rear
Detector
%
Difference
1
66.225
69.884
-5.5251
41.646
35.677
14.332709
16.685
14.063
15.71471
100.718
69.825
30.67277
2
67.129
68.288
-1.72653
41.59
35.282
15.167107
16.635
14.728
11.46378
100.227
68.875
31.28099
3
69.965
74.479
-6.4518
43.662
36.869
15.558151
17.462
14.659
16.052
105.388
70.835
32.78647
4
68.354
70.326
-2.88498
42.871
35.988
16.055142
17.174
14.212
17.247
103.737
70.683
31.86327
5
65.418
68.88
-5.29212
41.149
36.592
11.074388
16.448
14.414
12.36625
99.33
70.073
29.45434
6
66.901
68.833
-2.88785
41.762
36.329
13.009434
16.704
14.322
14.26006
100.765
69.763
30.76664
7
71.567
73.801
-3.12155
44.6
36.662
17.798206
17.843
15.303
14.23527
107.638
71.795
33.29958
8
70.914
81.521
-14.9576
44.329
37.108
16.289562
17.715
15.4
13.06802
106.949
72.361
32.34065
9
70.793
75.56
-6.73372
44.699
37.48
16.150249
17.858
15.489
13.26576
107.988
73.32
32.10357
10
70.109
71.897
-2.55031
44.093
37.875
14.102012
17.625
14.655
16.85106
106.424
72.307
32.05762
11
68.472
70.538
-3.01729
43.257
36.489
15.646023
17.284
15.197
12.07475
104.508
71.546
31.54017
12
70.272
71.531
-1.79161
43.801
37.9
13.472295
17.516
14.57
16.81891
105.694
72.318
31.57795
13
71.815
71.163
0.907888
44.497
36.734
17.44612
17.807
15.324
13.94395
107.229
72.501
32.38676
14
69.01
70.579
-2.27358
43.289
36.424
15.858532
17.313
14.305
17.37423
104.441
71.315
31.71743
15
68.487
68.816
-0.48038
42.955
36.117
15.918985
17.152
14.861
13.35704
103.49
71.055
31.34119
16
71.998
80.286
-11.5114
44.673
36.783
17.661675
17.865
15.266
14.548
107.7
72.162
32.99721
17
70.486
79.729
-13.1132
44.147
36.861
16.503953
17.624
15.315
13.10145
106.403
72.38
31.9756
18
70.25
75.262
-7.13452
44.188
37.388
15.388793
17.648
15.527
12.01836
106.658
73.016
31.54194
19
71.417
74.997
-5.01281
44.507
37.234
16.34125
17.788
15.546
12.604
107.173
73.028
31.8597
20
69.171
73.45
-6.18612
43.449
36.796
15.312205
17.346
15.351
11.50121
104.73
72.042
31.21169
21
67.824
73.352
-8.15051
42.251
37.244
11.850607
16.864
15.678
7.032732
101.631
71.788
29.36407
22 69.692 70.423 -1.0489 43.512 36.775 15.483085 17.381 14.538 16.35694 104.838 72.291 31.04504
23
68.943
69.619
-0.98052
43.464
37.076
14.697221
17.35
15.291
11.86744
104.828
72.508
30.83146
24
67.639
70.38
-4.0524
42.634
36.358
14.720645
17.022
15.05
11.58501
102.772
71.556
30.37403
25
69.136
70.085
-1.37266
43.088
36.516
15.252506
17.198
14.955
13.04221
103.662
71.583
30.94577
26
67.33
69.073
-2.58874
42.146
36.556
13.263418
16.815
15.252
9.295272
101.488
71.92
29.13448
27
65.916
69.09
-4.81522
41.603
36.284
12.785136
16.577
15.183
8.409242
100.193
71.602
28.53593
28
63.953
69.793
-9.13171
40.672
36.04
1 1.38867
16.214
15.148
6.574565
98.055
71.224
27.36321
29
64.112
74.936
-16.883
40.498
35.877
11.41044
16.154
15.032
6.945648
97.579
70.799
27.44443
30
63.302
68.238
-7.79754
40.176
35.89
10.668061
16.002
15.075
5.793026
96.655
71.13
26.40836
31
69.961
71.137
-1.68094
43.185
36.552
15.3595
17.259
15.33
11.17678
103.894
72.258
30.45027
32
69.135
71.957
-4.08187
43.293
36.947
14.658259
17.267
15.463
10.44767
104.213
73.006
29.9454
B-7
-------
CO
I
00
120
100
20
9- 60
MEK inlet
*- MEK outlet
Butyl acetate inlet
¦*- Butyl acetate outlet
*- Ethylbenzene inlet
Ethylbenzene outlet
-~—Xylene inlet
— Xylene outlet
A A -H v *
*—X—* X v »< v X M » X n x x X
10
15 20
Sample #
25
30
35
Figure B-4. AMT Module, Test 4, 2.0 ft3/min Air, 1.0 L/min Oil, 230 ppm VOCs
-------
Table B-5. AMT Module, Test 5, 2.0 ft3/min Air, 1.0 L/min Oil, 850 ppm VOCs
Test 5
Irun date: : 4/28/00j ? ; ; i ; ;
;sequence used : i0429e&f ; ; ; ; I
;Initial settings: syringe pump set at 3.5ml/hr;positive net sample flow 50cc/min;magnehelic (1) ~0.9"H2O '
j magnehelic (2) >1.0"H2O i ;
; ;n-Butyl ;.n-Butyl Ethyl Ethyl >
;MEK !MEK : ;Acetate Acetate ; ;Benzene ;Benzene ;o-Xylene 'o-Xylene
.Front :Rear .% :Front -Rear ,% Front .Rear ,% jFront ;Rear ;%
Detector
Detector
Difference
Detector ;
Detector
Difference
;Detector
Detector
iDifference ;
Detector
Detector
^Difference
1 •
258.118
: 196.
.099
24.02738
: 158.
541;
77.836
50.904813
63.119'
29.327
: 53.53697",
381.073'
156.69
; 58.88189
2
235.798
| 189.
.425
'19.66641
146.
'143':
85.804
41.287643
' '58.13,
'32.671 '
: 43.79666
351.1
177.338
49.49074'
3
246.903
201.
.'885
18.23307
153.
555;
89.928'
41.435968'
61.018;
34.'1 27''
: 44.0706:
368.879
186.289
.' 49.49862
4)
255.548
: '208.
.471 :
18.42198
I 159.
528;
94.122
40.999699
63.397!
34.963
: 44.8507
383.122!
193.603
i 49.46701
5
" 238.781
193.
-fc.
cn
CO
18.98099
; 148.
642;
89.258
'39.951023
59.05;
33.327
: 43.56139
356.841
186.187
I 47.82354
6
226.632
181.
.656'
19.84539
; 140.
802;
87.869
37.593926'
55.888
32.773
i 41.3595
337.046'
184.331
: 45.30984
7:
228.48
' 189.
.155;
17.21157
: 144.
341 ;
89.08'
38.285033
57.273
33.677
41.19917
346.809;
188.48'
; 45.65308
8
252.031'
202.
.259'
19.74836
155.
649:
91.968
40.913209
61.791
'34.945'
, 43.44646,
372.998'
191.591'
.' 48.63485
9
' 254.606
; 194.
.285
' 23.6919
157.
5261
90.335
42.653911
62.493
34.263
; 45.17306
. 376.575
189.25
; 49.74441
10:
230.948
203.
.771
11.76758
; 148.
972';
95.288'
36.036302
59.075
35.309
, 40.23022;
.' 358.591
197.982
'44.'78891'
11;
12.208
51.
.564:
-322.379
24.
672i
44.703;
-81.189202
9.251 :
18.621
: -1 01.286:
60.193'
100.169
-66.413
12
5.66'
32.
.563
-475.318'
11.
193;
28.586
-155.39176
4.175'
12.375'
: -196.407;
27.278
57.266'
! -109.935
B-9
-------
450
400
350
300
250
200
150
100
50
0
0
MEK inlet
-¦-MEK outlet
-*- Butyl acetate inlet
-*- Butyl acetate outlet
-*- Ethylbenzene inlet
-•- Ethylbenzene outlet
—Xylene inlet
- Xylene outlet
2 4 6 8 10
Sample #
Figure B-5. AMT Module, Test 5, 2.0 ft3/min Air, 1.0 L/min Oil, 850 ppm VOCs
-------
Table B-6. AMT Module, Test 6, 2.0 ft3/min Air, 0.2 L/min Oil, 300 ppm VOCs
;Test 6 ;
^Sequence ;050300b ? ; i ;
;date run: 5/3/00 • • : •
MnitiaI settings: :syringe pumpset 2/0m l/hr;air flow set at 337 ft/m in;heater set at 290C;magnehelic (1) rds 0.9"H2O
magnehelic (2) rds >
,0"H2O;net positive sample flow ~55cc/min;
n-Butyl
n-Butyl
Ethyl
Ethyl
M EK
M EK
A cetate
Acetate
Benzene
Benzene
o-Xylene
o-Xylene
Front
Rear
%
Front
Rear
%
Front
Rear
%
Front
Rear
%
Detector
Detector
Difference
Detector
Detector
D iffe re n c e
Detector
Detector
Difference
Detector
Detector
Difference
17
9
983
7.921
20.6551 1
4.477
3.3
26.289926
1.736
304
24
88479
12
306
6
016
51
1 1328
18
98
127
86.06
12.29733
56.009
28
649
48.849292
20.831
8
704
58
21612
146.81
53
936'
63
26136
19
96
557
88.905
7.924853
56.99
39
845
30.084225
20.984
2
217
41
77945
148
628
77
471'
47.8759
20
96
712
91.91
4.965258
58.05
46
756
19.455642
21.305
4
633
31
31659
151
185
94
923
37
21401
21
94
921
88.473
6.793017
57.282
49
092
14.297685
21.016
5
666
25
45679
149
178
104
412
30
00845
22
88
494
91.3
' -3.17084
53.439
52
224
2.2736204
19.572
6
862
13
84631
138
721
114
357
17
56331
23
98
396
89.574
'8.965812
59.087
52
521
1 1.1 12427
21.609
17.13
' 20
72747
153
301
1 1
3.46'
22
72718
24
"91
496
88.104
3.707266
56.02
53
019
5.3570154
20.502
17.46
14
83758
145
623
122
318
16
00365
25
92
385
86.708
6.144937
56.568
52
421
7.3309999
20.7
7
376
16
05797
147
037
123
229
16
19184
26
96
107
78.1 14
18.72184
58.774
39
726
'32.408888
21.464
2
957
39
63381
'152
668
"" 93
853'
38
52477
27
9
8.44
80.872
17.8464
60.149
37
114
38.296564
21.933
2
006
45
26057
'156
212
86
639
' 44
53755
28
92
791
78.503
"15.39805
57.039
36
182
'36.566209
20.854
639
44
18817
148
273
83
477'
43
70047
29
93
765
78.983
15.76494
57.162
36
952
' 35.355656
20.893
813
43
45953
14
3.29
83
943'
43
39268
30
90
232
77.21
14.43169
55.614
35
496
36.174345
20.289
361
44
00414
144
548
80
838
44
07532
31
89
777
77.286
' 13.91336
55.336
'35
193
36.401258
20.196
238
44
35532
143
835
79
718
44
57677
32
90
082
' 78.004
13.40778
55.626
35
188
36.74181 1
20.301
243
44
61849
144
686
79
281'
45
20479
33
94
605
81.424
13.93267
57.847
35
956
37.84293
21.105
398
45
99384
150
067
80
328
46
47191
34
87
307
76.669
12.18459
53.961
35
305
34.5731 18
19.686
257
42
81723
140
232
79.39
43
38667
35
86
451
77.371
10.50306
53.993
35
339
34.548923
19.721
256
42
92379
140
587
79
196
43
66762
36
8
4.87
74.861'
1 1.79333
52.134
34
241
34.321 172
19.006
0
928
42
50237'
135.44
76
866
43
24719
37
89
215
78.06
12.5035
54.666
35.26
35.499213
"19.956
189
43
93165
141
953
"l 8
574
44
64788
38
86
825
75.605'
12.92255
53.442
35
103
34.315707
19.506
138
42
89962
139
053
78
342
43
66033
39
88
171
76.857
' 12.83188
54.145
35
505
34.426078
19.762
265
42
99666
140
659
78
934
43
88272
40
88
651
77.577
12.49168
54.609
35
486
35.018037
19.918
1 1.24
43
56863
142
028
78
699
44.5891
41
"" 89
224
78.399'
12.13239
54.853
36
225
33.959856
20.007
458
42
73004
142
466
" 80
319
43
62234
42
89
563
76.426
14.66789
54.52
35
674
34.567131
19.917
308
43
22438
141.61
79
273
44.0202
43
89
'622
79.966
10.77414
55.498
36.71
' 33.853472
'20.214
578
42
12281
144.27
31.1
43
78596
44
90
'273
78.773'
12.73914
55.418
36.57
34.01061
20.22
557
42
'84372
144
069
80
863
43
87203
45
' 86
'586
77.493
10.5017
53.212
36
622
31.177178
19.415
578
40.3657'
138
399
8
1.12
41
38686
46
87
'998
78.186
1 1.15025
54.076
36
515
32.474665
19.733
549
41
47367
140
653
"80
883
42
49465
47
86
636
80.28
7.336442
53.638
38
088
28.990641
19.559
2
058
38
'35063'
"139
645
83
487
40
21483
B-ll
-------
CO
I
bO
E
Q.
Q.
*-*
*—*—-X X X * X X x )( X X X X x X X X—*
MEK inlet
MEK outlet
Butyl acetate inlet
Butyl acetate outlet
Ethylbenzene inlet
Ethylbenzene outlet
Xylene inlet
— Xylene outlet
i i i i i i i i i i i i i i i i i i i i ! i i i i i i i i r
<&<£><£ <& ^ $
Sample #
Figure B-6. AMT Module, Test 6, 2.0 ft3/min Air, 0.2 L/min Oil, 300 ppm VOCs
-------
Table B-7. AMT Module, Test 7, 2.0 ft3/min Air, 0.2 L/min Oil, 170 ppm VOCs
;Test 7 ;
;Run Date:: 5/5/00! ? ; i ;
;run start 3:15; • : •
Mnitial settings: id id not verify net pos itive sam pie flow; sy ringe pump set at 1.0 m i/hr; air flow set 341 ;tem p at sy ringe tip
itip 290C;magnehilic (1) rding 1 "H20;magnehelic (2) rding >1"H20; ;
m-Butyl
I'n-B uty I
; Ethy I
:E thy I
:m EK
:MEK
;Acetate
:Acetate
•Benzene
^Benzene
o-Xylene
;o-Xylene
Front
Rear
%
jFront
:Rear
%
; Front
:Rear
%
Front
;Rear
.%
^Detector
IDetector
Difference
iDetector
:Detector
D iffe re n c e
^Detector
jDetector
Difference .
Detector
jDetector
Difference
1
! 52.52
44.
.503
15
.26466
; 32.4
: 21
. 1 52 :
34
.716049
j 12.922
7.42
' 42.57855'
75.912
! 39.721
: 47.
.67494
2
; 52
.909
42.
.708
19
.28027
32.664
20
.832 :
36
.223365
13.003'
7
.447'
; 42.7286;
76.467
40.253
. 47.
.35899
3'
53
.834
42.
.879
20
.34959'
| 32.943
20
.949
36
.408342
13.098
7.5'
42.73935
76.985
40.597'
: 47.
.26635
4
51
.304
41.
.942;
18
.24809
31.672
20
.838
34.20687
12.609
7
.472
40.74074
74.1 12
40.63
.45.
.17757
5
' 50
.703
41.
.517:
18
.11727
31.255
20
.622'
34
.020157
12.421
7
.392
40.48788;
73.018
; 40.323
: 44.
.77663
6
53
.456
43.
.012:
' 19
.53756
32.986
21
.207
35
.709089
I 13.087
7
.579
42.08757
"77.029
41.336'
: 46.
.33709
7
; 51
.818'
42.
.066:
18
.81972
32.131
21
.'101
34
.328219
12.758
7
.548
• 40.8371 2'
74.988
41.317
' 44.
.90185
" 8
; 50
.982
42.
066
17
.48853'
31.461
21
.'101
32
.929659
12.489
7
.548
39.56282
73.376
41.317
43.6914
9'
52
.202
42.
.377'
18
.'821 12
32.312
21
.073;
34
.782743
12.83
7
.533
41.28605
75.46
41.359
45.
.19083
10
51
.247
: 41.
.892;
18
.25473
31.908'
21.22
' 33
.496302
! 12.679
7
.584
40.18456
74.688
41.68'
: 44.
.19452
11'
49
.'504
40.98'
17
.21881'
30.735
20
.982
31
.732552
12.187
7
.507
.' 38.40158
71.726
41.232'
: 42.
.51457
12
50
.'484'
41.
.414;
17
.96609
31.355
21
.246
' 12.392
: 12.392
7.61
38.58941
73.15
41.445'
' 43.
.34245
13
! 53
.219
42.
519
'20.1056
32.835
21
.444,
34;69164
J' ' 12.963
7
.635
• 41.1016:
76.394
41.679
45.
.44205
14
53
.579
43.
.632
18
.5651 1
33.458
; 21
.995
34
.260864'
13.2
7
.817
40.7803
" 77.824
42.75'
45.
.06836
15
j 52
.533
42.
.'575;
1.
8.9557
32.739'
21
.835 :
' 33
.305843
12.906
7
.776
39.74895,
76.015
42.59'
' 43.
.97158
16
49.32
f" 40.
.465;
17
.95418'
" 30.865
21
.274 :
31
.074032
12.165
7
.601
' 37.51747'
71.653
41.73
41.
.76099
17
48
.944'
40.
.199
17
.86736
30.531
; 21
.118
30
.830959
12.034
7
.557
. 37.20293
70.93
41.447
41.
.56633
18
50
.794
"41.
.991;
17
.33079
31.844'
: 21
.'654
31
.999749
12.554
7
. 726 •
! 38.45786
74.035
: " 42.278'
42.
.89458
19
: 53
.114
42.
.093
' 20
.74971
j 32.886
: 21
.612.
34
.282065
; 12.956
7
.712'
' 40.47546
76.326
42.271
; 44.
.61782
20
: 51
.362'
41.
.356'
19
.48133
i 31.936
21
.628 ;
32
.277054
I 12.577
7
.719'
38.62606
74.1 1 1
; 42.356
: 42.
.84789
21
51
.299
' 41.39
' 19
.31617
31.961
21.67;
32
.198617
12.58
7
.741
38.46582
74.097'
} 42.453
42.
.70618
22
52.05
41.
.923
19
.45629'
32.359
21
.907 ;
32
.300133
12.739
7
.819
; 38.62156:
75.021'
42.912
42.
.80002
23
53.31
42.
.394
20
.47646
' 33.134
. 22
.036'
33
.494296
13.031
7
.852
39.74369'
76.848
43.087
43.
.93218
24"
49
.831
41.
.357
17
.00548
31.461
; 22
.026
29
.98951 1
12.386
7
.879
36.38786:
73.16
;' 43.343'
40.
.75588
25
49
.312
; 40.
.455';
17
.961 15
; ' 30.895'
21
.712
29
.723256
12.167
7
.777
; 36.0812
; 71.786
42.635
40.6082
26
; ' 51
.265
41.
.201 :
' 19
.63133
; 31.85
21
.962
31
.045526
12.542
7
.845
37.45017'
73.887
42.97
41.
.84363
27
51.59
41.
.243
20
.05621
"" 32.134
; 22
.039
31
.415323
12.64'
7
.865'
; 37.7769:
74.542'
43.026
; 42.
.27952
28
51
.296
41.
914
'18
.28993
"32.351
22
.142;
31
.556984
12.714
7
.905-
37.82445
' 75.038'
; 43.362
: 42.
.21328
29
50
.909
41.
.575
18
.33468
31.871
22
.272,
' 30
.118289
; 12.508'
7
.958
36.37672'
73.864
43.712
' 40.
.82097
30
f 51
.634
"' 41.
.934'
18
.78607
32.429
22
.536,
30
.506645
"12.755
8
.046
36.91886
75.285
44.319
41.1317
31
50
.322
40.
.311
19
.89388
31.369'
i 22
.096'
' '29
.561032
; 12.347
7
.'917
35.8791 6:
72.779
; 43.592
40.1036
32
47
.509
39.
.588'
1 6
.67263
29.892'
21
.869
26
.839957
11.755
7
.853
33.19439'
69.332'
"43.217
; 37.
.66659
33
46
.676
38.
418-
"17
.69218
: 29.42
21
.'454
27
.076818
11.553
7
.713
33.23812'
68.225
; 42.399
: 37.
.85416
B-13
-------
90
80
70
60
50
40
30
20
10
0
-•- MEK inlet
-¦- MEK outlet
-*- Butyl acetate inlet
-*- Butyl acetate outlet
-*- Ethylbenzene inlet
-•- Ethylbenzene outlet
+ Xylene inlet
— Xylene outlet
r f^ ^—«r 'v
^ ^"'
X x )f )( i/ X—X—X—X—X—X—X—X—X—X—x—x—*—*—*—*—* * * *—* " " " ^ * X -x
x x x -3E—a--*—* *¦ * xxxxx-xxxxxx-*- *--*—g
! ! ! 1 1 1
0 5 10 15 20 25 30 35
Sample #
Figure B-7. AMT Module, Test 7, 2.0 ft3/min Air, 0.2 L/min Oil, 170 ppm VOCs
-------
Table B-8. AMT Module, Test 7, 2.0 ft3/min Air, 0.22 L/min Oil, 170 ppm VOCs
Test 8
Initial settings:
syringe pum p 3.5m l/hr;silicon oil pressure 5.1 psi;oil tern p 71;air flow reading 344; tern p at syringe tip ~290C
magnehelic 1 (rding at sampie port 1)~1.0"H2O;mag. 2 (rding taken near vac. Pump) >1.0"H20
net positive flow verified at 50cc/min
start/stop time
7:00am-9:45am
run date:
5/6/00
Run sequence:
050600, 050600a, 050600b
run#
MEK
Front
Detector
MEK
Rear
Detector
%
Difference
n-Butyl
Acetate
Front
Detector
n-Butyl
Acetate
Rear
Detector
%
Difference
Ethyl
Benzene
Front
Detector
Ethyl
Benzene
Rear
Detector
%
Difference
o-Xylene
F ront
Detector
o-Xylene
Rear
Detector
%
Difference
1
4.552
3.65
19.81547
3.503
3.662
-4.5389666
1.393
1.906
-36.827
8.524
9.151
-7.3557
2
3.289
2.658
19.18516
2.646
3.125
-18.102797
1.046
1.652
-57.935
6.465
8.053
-24.563
3
3.157
6.072
-92.3345
2.545
5.339
-109.78389
0.963
2.293
-138.11
6.079
10.402
-71.1137
4
3.524
6.179
-75.3405
2.519
6.002
-138.26915
1.006
2.44
-142.545
6.483
11.006
-69.7671
5
2.51
5.428
-116.255
1.968
6.017
-205.74187
0.742
2.443
-229.245
4.902
10.983
-124.051
6
2.962
5.456
-84.1999
2.018
5.977
-196.18434
0.774
2.422
-212.92
5.113
10.915
-113.475
7
1.649
5.365
-225.349
1.465
5.93
-304.77816
0.472
2.405
-409.534
3.42
10.808
-216.023
8
1.137
4.698
-313.193
1.499
6.073
-305.13676
0.49
2.449
-399.796
3.592
11.098
-208.964
9
0.834
4.179
-401.079
1.135
5.722
-404.14097
0.375
2.34
-524
2.78
10.47
-276.619
10
137.872
65.605
52.41601
69.982
15.888
77.297019
28.084
5.681
79.7714
163.814
29.695
81.87273
1 1 151.286 102.345 32.34999 89.985 42.128 53.183308 34.858 14.152 59.401 207.138 78.021 62.33381
12
142.512
98.408
30.94757
85.919
44.01
48.777337
33.181
15.136
54.38353
197.311
84.547
57.15039
13
139.46
101.085
27.51685
85.927
45.668
46.852561
33.137
15.72
52.56058
197.591
88.826
55.04552
14
138.875
101.811
26.68875
86.336
46.47
46.175408
33.274
16.052
51.75813
198.545
91.292
54.01949
15
145.758
104.656
28.1988
89.561
47.204
47.294023
34.503
16.323
52.69107
205.751
93.335
54.63692
16
142.474
104.737
26.48694
88.78
47.981
45.95517
34.195
16.804
50.85831
204.471
97.677
52.22941
17
152.286
109.495
28.0991
94.257
49.884
47.07661
36.24
17.443
51.8681
216.196
100.626
53.45612
18
159.376
111.562
30.00075
98.568
50.774
48.488353
37.933
17.637
53.50486
226.656
102.83
54.63169
19
155.164
112.545
27.46707
97.672
51.512
47.260218
37.54
18.121
51.72882
225.039
104.066
53.75646
20
146.13
108.617
25.67098
92.16
50.841
44.833984
35.406
17.617
50.2429
212.006
103.52
51.17119
21
148.17
108
27.11075
93.482
51.22
45.208703
35.859
17.822
50.29979
214.562
104.314
51.38282
22
145.813
110.063
24.5177
92.387
51.622
44.124173
35.485
17.833
49.74496
212.703
104.645
50.80229
23
145.6
110.408
24.17033
92.216
52.007
43.603062
35.394
18.109
48.83596
212.24
105.455
50.31332
24 148.854 1 1 1.876 24.84179 94.21 1 53.133 43.602127 36.144 18.231 49.56009 216.427 105.709 51.1572
25
156.511
115.287
26.33936
97.995
53.933
44.963519
37.605
18.519
50.75389
225.161
106.749
52.58992
26
154.064
114.648
25.58417
96.341
54.222
43.718666
37
18.615
49.68919
221.235
107.161
51.56237
27
155.588
114.328
26.51875
96.976
54.369
43.935613
37.234
18.669
49.86034
222.84
107.549
51.73712
28
154.746
114.354
26.10213
97.155
54.495
43.909217
37.269
18.738
49.72229
223.296
107.871
51.69148
29
151.776
111.986
26.21627
95.588
54.214
43.283676
36.662
18.635
49.1708
219.433
107.392
51.05932
30
148.536
110.742
25.44434
93.973
54.513
41.990785
36.032
18.732
48.01288
215.84
108.106
49.91383
31
149.849
114.368
23.67784
94.943
55.38
41.670265
36.44
19.006
47.84303
218.428
109.504
49.86723
32
150.134
115.314
23.19261
95.226
55.63
41.581081
36.54
19.092
47.75041
218.946
109.932
49.79036
33
145.804
110.928
23.91978
92.654
54.986
40.654478
35.541
18.932
46.73194
213.008
109.302
48.68643
34
146.056
109.193
25.23895
91.772
54.055
41.098592
35.202
18.62
47.10528
210.566
107.482
48.95567
35
145.557
109.421
24.82601
91.851
54.346
40.832435
35.19
18.692
46.88264
210.736
107.935
48.78189
36
149.602
112.035
25.1113
93.961
54.919
41.551282
36.033
18.849
47.68962
215.74
108.557
49.68156
37
148.367
111.632
24.75955
93.637
55.22
41.027585
35.912
18.98
47.14859
215.021
109.227
49.20171
38
143.993
108.654
24.54217
90.839
55.134
39.305805
34.817
19.056
45.26812
208.587
109.495
47.50632
B-15
-------
CO
I
On
250
200
150
E
Q.
Q.
_3lf^-*-*-^)K XXX-a-XKXXXxttXXXXXXtt
100
MEK inlet
MEK outlet
Butyl acetate inlet
Butyl acetate outlet
Ethylbenzene inlet
Ethylbenzene outlet
Xylene inlet
- Xylene outlet
10 15
20
Sample #
25 30 35
40
Figure B-8. AMT Module, Test 7, 2.0 ft3/min Air, 0.22 L/min Oil, 170 ppm VOCs
-------
Table B-9. AMT Module, Test 9, 2.0 ft3/min Air, 0.2 L/min Oil, 350 ppm VOCs
Test 9
36655
Initial conditions:
m agnehelic (1) ~0.9"H2O;magnehelic (2) > 1 "H20;ran at 200ppm&4cfm
= 4ml/hrand 680;oil pressure 5ps
oil temp 78;temp controller setting 290C;sam pie flow through rotam eter 53;
I I I I I I
Observations:
VOC removal was low during this test run. Recommend changing oil and running at a higher VOC conc-
centration.
n-Butyl
n-Butyl
Ethyl
Ethyl
M EK
M EK
Acetate
Acetate
Benzene
Benzene
o-Xylene
o-Xylene
Front
Rear
%
Front
Rear
%
Front
Rear
%
Front
Rear
%
run#
Detector
Detector
Difference
Detector
Detector
Difference
Detector
Detector
Difference
Detector
Detector
Difference
1
0.631
7.321
-1060.22
1.025
10.458
-920.29268
0.316
4.126
-1205.7
2.324
20.191
-768.804
2
0.592
7.327
-1137.67
1.041
10.621
-920.26897
0.311
4.159
-1237.3
2.366
20.18
-752.916
3
0.603
7.17
-1089.05
0.992
10.579
-966.43145
0.294
4.123
-1302.38
2.266
20.134
-788.526
4
0.668
7.031
-952.545
1.185
10.447
-781.60338
0.316
4.083
-1192.09
2.24
19.882
-787.589
5
0.603
6.673
-1006.63
1.042
10.268
-885.41267
0.291
4.025
-1283.16
2.163
19.668
-809.293
6
0.692
6.472
-835.26
0.269
10.166
-3679.1822
0.308
3.995
-1197.08
2.115
19.552
-824.444
7
0.559
6.213
-1011.45
0.99
10.065
-916.66667
0.237
3.935
-1560.34
1.954
19.434
-894.575
8
0.716
5.919
-726.676
1.323
9.928
-650.41572
0.319
3.949
-1137.93
2.117
19.36
-814.502
9
0.72
5.888
-717.778
1.334
9.82
-636.13193
0.294
3.884
-1221.09
2.013
19.127
-850.174
10
0.655
5.662
-764.427
1.084
9.733
-797.87823
0.225
3.859
-1615.11
1.875
19.018
-914.293
11
0.703
5.484
-680.085
1.282
9.617
-650.15601
0.288
3.829
-1229.51
1.923
18.919
-883.827
12
0.382
5.294
-1285.86
1.062
9.441
-788.98305
0.265
3.772
-1323.4
1.913
18.69
-876.999
13
0.636
5.006
-687.107
1.164
9.269
-696.30584
0.292
3.714
-1171.92
1.944
18.44
-848.56
14
0.599
4.828
-706.01
1
9.107
-810.7
0.227
3.674
-1518.5
1.852
18.179
-881.587
15
4.975
9.396
-88.8643
2.71
10.911
-302.61993
1.009
4.316
-327.75
6.049
21.923
-262.424
16
79.607
68.491
13.9636
47.051
31.646
32.741068
18.528
11.399
38.4769
109.293
62.138
43.14549
17
91.54
75.87
17.1182
55.828
39.67
28.942466
21.604
14.238
34.09554
128.364
79.611
37.98027
18
85.396
76.173
10.80027
56.145
41.499
26.086027
21.691
14.988
30.90222
129.922
84.97
34.59922
19
5.554
9.845
-77.2596
6.997
18.707
-167.35744
2.426
7.365
-203.586
15.225
42.504
-179.172
20
3.666
7.675
-109.356
4.297
13.198
-207.14452
1.482
5.336
-260.054
9.175
30.346
-230.747
21
81.17
53.965
33.51608
43.521
22.564
48.153765
17.341
8.406
51.52529
101.028
45.66
54.80461
22
6.463
13.55
-109.655
10.362
21.204
-104.63231
3.631
7.974
-119.609
22.634
45.981
-103.15
23
3.16
6.587
-108.449
3.979
12.834
-222.54335
1.356
5.103
-276.327
8.94
28.567
-219.541
24
2.22
6.21
-179.73
2.988
11.303
-278.27979
0.952
4.57
-380.042
6.383
25.19
-294.642
25
1.888
4.967
-163.083
2.657
10.478
-294.35454
0.805
4.2
-421.739
5.304
22.886
-331.486
26
88.969
72.766
18.21196
52.085
33.297
36.071806
20.459
11.95
41.5905
120.944
66.25
45.22258
27
92.782
78.099
15.82527
56.78
40.623
28.455442
22.058
14.535
34.10554
130.801
82.121
37.21684
28
94.794
73.99
21.94654
58.039
41.684
28.179328
22.596
15.03
33.4838
134.027
86.213
35.6749
29
87.924
72.883
17.10682
55.353
42.146
23.859592
21.43
15.392
28.17545
127.459
89.125
30.07555
30
84.366
73.662
12.68758
53.926
42.669
20.874903
20.805
15.571
25.15741
124.02
90.522
27.01016
31
88.22
76.025
13.8234
56.077
43.14
23.070064
21.649
15.707
27.447
128.937
91.34
29.1592
32
91.059
74.687
17.97955
57.131
43.414
24.009732
22.08
15.832
28.2971
131.048
92.286
29.57848
33
88.697
75.567
14.80321
56.414
43.272
23.295636
21.764
15.808
27.36629
129.557
92.133
28.88613
34
88.202
74.665
15.34772
56.561
43.155
23.701844
21.804
15.772
27.66465
130.092
92.042
29.24853
35
86.866
71.566
17.61334
55.012
42.789
22.218789
21.217
15.661
26.18655
126.134
91.581
27.39388
36
85.83
72.856
15.11593
54.92
43.329
21.105244
21.183
15.855
25.15224
126.024
92.687
26.4529
37
91.721
78.726
14.16797
58.651
45.509
22.40712
22.643
16.523
27.02822
134.872
96.507
28.44549
38
95.437
81.768
14.32254
61.606
46.901
23.869428
23.708
16.987
28.34908
141.706
99.275
29.94298
39
96.661
80.491
16.72857
61.82
46.699
24.459722
23.804
16.925
28.8985
141.991
99.107
30.20191
40
93.705
80.413
14.18494
60.159
46.691
22.38734
23.174
16.958
26.82316
138.243
99.444
28.0658
41
95.434
80.002
16.17034
61.035
46.468
23.866634
23.506
16.888
28.15451
140.175
98.926
29.42679
42
96.435
77.94
19.17872
61.348
45.941
25.114103
23.626
16.737
29.15855
140.753
98.062
30.33044
43
92.36
78.248
15.27934
59.478
46.378
22.02495
22.921
16.909
26.22922
136.655
99.143
27.45015
44
91.043
79.597
12.57208
59.2
47.009
20.592905
22.815
17.129
24.9222
136.329
100.414
26.34436
45
93.232
79.387
14.85005
59.922
46.539
22.334034
23.115
16.991
26.49362
137.928
99.601
27.78769
46
94.528
79.312
16.09682
60.129
46.689
22.351943
23.192
17.039
26.5307
138.103
99.949
27.62721
47
94.751
78.851
16.78083
60.285
46.571
22.74861 1
23.254
16.981
26.976
138.341
99.592
28.00977
48
91.775
77.304
15.76791
59.015
45.955
22.129967
22.793
16.79
26.33703
135.783
98.465
27.48356
49
89.062
74.638
16.19546
57.12
45.269
20.747549
22.028
16.578
24.74124
131.207
97.432
25.74177
50
87.042
75.892
12.80991
56.641
45.856
19.040977
21.846
16.781
23.18502
130.375
98.558
24.40422
51
90.907
78.336
13.82842
58.8
46.44
21.020408
22.663
16.967
25.13348
135.307
99.552
26.42509
52
91.285
77.471
15.13283
58.703
46.252
21.21016
22.595
16.917
25.12945
134.867
99.344
26.33928
53
89.465
76.244
14.77785
57.455
45.528
20.758855
22.179
16.679
24.79823
132.123
97.891
25.90919
54
89.725
75.301
16.07579
57.322
45.067
21.379226
22.133
16.506
25.42358
131.794
96.898
26.47768
55
91.056
73.616
19.15305
57.913
45.056
22.200542
22.36
16.515
26.14043
133.052
97.044
27.0631
56
85.559
72.865
14.83655
55.092
45.063
18.204095
21.262
16.535
22.23215
126.521
97.199
23.1756
57
86.82
74.576
14.10274
56.109
45.279
19.301716
21.651
16.604
23.3107
129.041
97.579
24.3814
58
95.245
78.84
17.224
60.555
45.877
24.239121
23.367
16.77
28.23212
139.344
98.307
29.45014
59
93.165
79.109
15.08721
59.54
46.774
21.441048
22.981
17.105
25.56895
136.827
100.303
26.69356
60
93.709
78.902
15.80104
59.971
46.791
21.977289
23.114
17.112
25.96695
137.831
100.201
27.30155
B-17
-------
CO
I
O©
160
Q. 80
X X X X X X X X
X X X X
10
20
MEK inlet
MEK outlet
Butyl acetate inlet
Butyl acetate outlet
Ethylbenzene inlet
Ethylbenzene outlet
Xylene inlet
- Xylene outlet
30 40
Sample #
50
60
Figure B-9. AMT Module, Test 9, 2.0 ft3/min Air, 0.2 L/min Oil, 350 ppm VOCs
-------
Table B-10. Two Parallel AMT Modules, Test 10, 4.0 ft3/min Air, 0.4 L/min Oil, 350 ppm VOCs
Test 10
Run date 5/11/00
sequence used:
051100a
I
Initial parameters:
m ag. (1) ~0.9"H2O;mag (2) > 1,0"H2O;oil tern p 74C;air tern p74C;tem p controller set 290C;air velocity 639;
oil pressure ~5.1 psi;excess sample flow set ~55 (dual rotameter);syringe set 4ml/hr
run#
M EK
Front
Detector
M EK
Rear
Detector
%
Difference
n-Butyl
Acetate
Front
Detector
n-Butyl
Acetate
Rear
Detector
%
Difference
Ethyl
Benzene
Front
Detector
Ethyl
Benzene
Rear
Detector
%
Difference
o-Xy lene
Front
Detector
o-Xy lene
Rear
Detector
%
Difference
1
1.196
0.841
29.68227
2.549
1.724
32.365634
0.396
0.942
-137.879
1.839
3.353
-82.3274
2
0.518
1.072
-106.95
2.053
1.969
4.0915733
0.224
1.007
-349.554
1.647
3.357
-103.825
3
1.18
2.452
-107.797
2.13
2.233
-4.8356808
0.224
1.063
-374.554
1.623
3.347
-106.223
4
1.079
0.868
19.55514
2.492
1.565
37.199037
0.305
0.845
-177.049
1.681
3.205
-90.6603
5
1.062
0.837
21.18644
2.294
1.48
35.483871
0.277
0.84
-203.249
1.675
3.144
-87.7015
6
0.992
0.763
23.08468
2.114
1.474
30.274361
0.227
0.812
-257.709
1.551
3.074
-98.1947
7
0.726
0.859
-18.3196
1.745
1.731
0.8022923
0.189
0.918
-385.714
1.461
3.086
-111.225
8
1.006
0.594
40.95427
2.26
1.748
22.654867
0.284
0.918
-223.239
1.61
3.053
-89.6273
9
0.46
0.846
-83.913
1.715
1.793
-4.548105
0.178
0.954
-435.955
1.444
3.09
-113.989
10
0.957
0.733
23.40648
2.135
1.414
33.770492
0.252
0.829
-228.968
1.499
2.991
-99.533
1 1 2.289 2.32 -1.3543 2.736 2.212 19.152047 0.429 1.253 -192.075 3.16 4.016 -27.0886
12
1.239
1.178
4.923325
2.234
1.941
13.1 15488
0.273
1.085
-297.436
1.75
3.492
-99.5429
13
3.925
4.432
-12.9172
3.703
2.714
26.708075
0.846
1.401
-65.6028
5.412
4.878
9.866962
14
95.553
56.539
40.8297
56.971
11.345
80.08636
21.585
3.968
81.61686
127.561
18.208
85.72604
15
101.631
67.474
33.60884
64.59
19.195
70.281777
24.366
6.355
73.91858
145.574
31.711
78.21658
16
103.881
70.809
31.83643
67.57
22.666
66.455528
25.427
7.604
70.09478
152.193
39.211
74.236
17
108.447
74.369
31.42364
71.053
24.641
65.320254
26.704
8.383
68.6077
160.135
43.862
72.60936
18
106.268
74.521
29.87447
69.664
25.774
63.002412
26.165
8.831
66.24881
156.784
47.01
70.01607
19
107.389
74.361
30.75548
70.392
26.385
62.517047
26.474
9.016
65.94395
158.739
48.825
69.24196
20
108.873
76.766
29.49032
72.267
27.376
62.118256
27.16
9.337
65.62224
163.193
51.107
68.68309
21
109.67
76.9
29.88055
72.683
27.897
61.61826
27.27
9.515
65.10818
163.499
52.194
68.07687
22
108.389
80.158
26.046
72.683
28.979
60.129604
27.23
9.912
63.59897
163.409
54.163
66.85433
23
111.554
82.145
26.36302
74.933
29.759
60.285855
28.122
10.071
64.18818
168.989
55.401
67.21621
24 110.226 80.289 27.15965 73.647 30.18 59.020734 27.639 10.259 62.88216 165.939 56.627 65.87481
25
109.571
80.121
26.87755
72.776
30.6
57.953171
27.335
10.411
61.9133
164.004
57.554
64.90695
26
109.47
81.358
25.6801
72.566
30.965
57.328501
27.231
10.531
61.32716
163.649
58.135
64.4758
27
109.668
81.099
26.05044
72.899
31.426
56.891041
27.37
10.674
61.0011
164.194
58.859
64.15277
28
101.128
83.763
17.17131
69.849
32.046
54.121033
26.184
10.93
58.25695
158.458
59.934
62.17673
29
100.226
80.463
19.71844
68.236
32.201
52.809367
25.629
10.959
57.23985
154.617
60.554
60.83613
30
102.113
79.441
22.20285
68.725
32.231
53.101491
25.803
10.999
57.37317
155.173
60.783
60.82888
31
103.253
80.731
21.81244
69.402
32.598
53.030172
26.013
11.107
57.30212
156.484
61.279
60.84009
32
103.484
81.486
21.25739
69.592
33.015
52.559202
26.129
11.194
57.15871
156.888
61.819
60.59673
33
103.967
81.257
21.84347
69.918
33.221
52.485769
26.212
11.279
56.97009
157.562
62.223
60.50888
34
100.367
79.535
20.75583
67.711
33.498
50.527979
25.446
11.386
55.25426
152.755
62.634
58.99709
35
101.594
79.006
22.2336
67.97
33.574
50.604679
25.498
11.458
55.06314
152.957
63.063
58.77077
36
104.337
83.383
20.083
70.162
34.422
50.939255
26.334
11.66
55.72264
158.286
64.097
59.50558
37
106.155
83.507
21.33484
71.522
34.988
51.080786
26.835
11.834
55.90088
160.994
64.973
59.6426
38
101.175
80.757
20.18087
68.224
34.93
48.801008
25.589
11.894
53.51909
153.78
65.34
57.51073
39
104.382
82.376
21.08218
69.773
35.147
49.626646
26.198
11.959
54.35148
157.353
65.529
58.35542
40
105.964
81.905
22.70488
70.128
35.778
48.981862
26.322
12.15
53.84089
157.875
66.315
57.99525
41
102.268
82.505
19.32472
69.134
36.021
47.896838
25.932
12.252
52.75335
156.088
66.913
57.13123
42
101.934
82.631
18.93676
68.362
36.519
46.579971
25.618
12.397
51.60824
154.439
67.87
56.05385
43
103.229
82.532
20.0496
68.693
36.679
46.604458
25.738
12.48
51.51138
154.992
68.379
55.88224
44
101.861
83.264
18.25723
68.689
37.183
45.867606
25.766
12.65
50.90429
154.876
69.333
55.23322
45
106.432
84.464
20.64041
70.928
37.233
47.505921
26.635
12.658
52.47607
159.897
69.338
56.63583
46
104.331
84.73
18.78732
70.381
37.744
46.37189
26.414
12.836
51.40456
158.894
70.027
55.92848
47
103.369
82.745
19.95182
69.099
38.102
44.858826
25.927
12.997
49.87079
155.831
70.891
54.50777
48
101.328
85.423
15.69655
68.983
38.605
44.036937
25.856
13.162
49.09499
156.188
71.748
54.06305
B-19
-------
180
160
140
120
E
Q.
Q.
100
x-x-x-*-*-*-*
X X X X. * X ac*"** XXXXXX****:
MEK inlet
MEK outlet
Butyl acetate inlet
Butyl acetate outlet
Ethylbenzene inlet
Ethylbenzene outlet
«- Xylene inlet
- Xylene outlet
10 15 20 25 30
Sample #
35
40
45
50
Figure B-10. Two Parallel AMT Modules, Test 10, 4.0 ft3/min Air, 0.4 L/min Oil, 350 ppm VOCs
-------
Table B-11. Two Parallel AMT Modules, Test 11, 4.0ft3/min Air, 1.0 L/min Oil, 350 ppm VOCs
;Test 11
;Run date 5/12/00
Initial parameters: temp controller set 290C;airvelocity ~ 4cfm;mag (1) ~0.9"H2O;mag (2) >1.0 "H20
i oil flow 1L/min;excess sample flow set~55 (dual rotameter);syringe set 4ml/hr
n-Butyl ;n-Butyl : Ethyl Ethyl <
MEK ;MEK Acetate Acetate Benzene Benzene o-Xylene ;o-Xylene
Front Rear % Front :Rear % .Front Rear .% Front jRear \%
Detector
Detector
Difference
;Detector :
Detector
D iffe re n c e
Detector
Detector
Difference ;
Detector
Detector
Difference
1'
0
.167
0-
100'
0.6;
0
100
0.0804;
0:
100;
0.633
0
100
2
0
.185
0.0944
48.97297
0.
.623;
1
.609
-158.26645'
0.0779
0
.902
-1057.89
0.637'
3
.743'
-487.598
3
0
.161 '
0
100'
0.
541
1
.505
-178.18854
0.0821 :
0
.838
-920.706
0.638
0
.434'
31.97492
4 ;
0
.175
0
.708
-304.571
0.
.581"
1
.807,
-211.01549
0.0783;
1
.022'
; -1205.24
0.594'
3.
.915'
" -559.091
5
0
.165
0
100'
0.
.603';
1
.693
-180.76285
0.0758;
1
.063'
-1302.37,
0.59'
3
.975
-573.729
6
0
.'141
0.
.117!
17.02128
0.
.589,
1
.755
-197.96265
0.0707
0
.943
-1233.8
0.574'
3
.531
; -515.157
7,
0
.164
0"
100
0.
.565';
1
.287
-127.78761
0.0743
0
.798
; -974.024
0.568
3
.459
-508.979
8
0
.166
0 :
100'
0.
.527,
1
.332 :
-152.75142
0.0681
0
.778
-1042.44
0.576'
3
.328'
-477.778
9"
0.17
0
100'
0.
.541
1
.231 '
-127.54159
"0.0657
0
.753
-1046.12;
0.56;
3
.375
I -502.679
10
0
.167
0
100'
0.
551 '
1
.365 :
-147.7314
0.0693
0
.822
; -1086.15
0^566
3
.473'
-513.604
11''
0
.'138
0'
100'
0.
593
1
.522'
-156.66105
0.0672
"O
.908
' -1251.1 9;
0.572!
3
.533'
-517.657
1 2 :
0
.438
0.
.336-
23.28767
0.
.746;
... .
.696..
-127.34584
0.102
0
.984
: -864.706;
0.596.
3
.714
: -523.154
1 3 :
2
.651
1
.749 ^
34.0249
2.
.063:
2
.051 ,
0.5816772'
0.421 :
1
.135
-169.596
' 3.456.
4
.706
-36.169
14
91
.613
28.
.112
69.3144
56.
.414;
8
.005
85.81026
19.254
2
.735
; 85.79516',
129.88
14
.008
: 89.21466
15
98
.666
31.
.338'
68.2383
62.
.'718';
10
.845
" 82.708313
21.261
3
.605
83.04407
144.055
19
.688'
86.333
16
99
.243
32.
.387'
67.36596'
63.42 :
1 1
.947
81.162094
21 .'449
4.03
; 81.21125,
145.064'
22
.469'
84.51097
17;
100
.032
'' 30.
.637'
69.3728
64.
.249':
: ' 11.61
81.92968
21.706'
3
.'974
; 81.6917
146.984
22
.642
84.5956
18-
94.5
29.
.651'
68.62328
" 60.
.395"
12
.265
79.692027
20.353'
4
.131
' 79.70324
137.525'
' 24
.148
82.44101
19,
90
.672
33.68
62.85513
59.
.439,
13
.621 :
77.084069
20.017
4
.561
: 77.21437'
135.864;
26
.633'
80.39731
20
96.57
35.03
'63.72579
63.
.161 :
13
.945'
" 77.921502
21.264
4
.663
;' 78.07092 :
144.393
27
.308
81.08773
21
99
.858
35.
.457'
64.49258
64.
.854:
'1 4
.348
'77.876461
21.827,
4.77
78.14633
' 148.138
27
.837
81.20874
22'
98
.082
32.
.521
66.84305
63.
.905;
13
.233
79.2927
21.485
4
.428
79.39027
145.832
25
.713
: 82.36807
23
9.
8.47
34.
. 1 58 ;
65.31 126
63.
CO
CD
CD
1 3
.686
78.615291
21.586'
4
.515
79.08367
'146.387
26
.379
81.97996
24
101.48
33.
.959;
66.53626
64.88;
14
.239'
78.053329
21.836'
4
.758 :
; 78.21029,
147.465
27
CO
0
-t*
81.07754
25;
85
.385
30
.279;
' 64.53827
56.
.963;
13
.195'
" 76.835841
19.1 52 :
4
.413
" 76.95802
'130.192'
' 25
.873
80.12704
26
91
.338
33.
.577'
' 63.23874
60.
.568;
14
.286,
76.413288
20.385
4
.739
'76.75251'
138.598
27
.571
80.10722
27
93
.357
31.
.704
66.04004
| " 61,75:
14
.047
77.251822
' 20.781'
4
.634'
77.70078.
141.376
27
.457
80.57874
28 ;
100
.211
37.
.655
62.42428
66.06;
15
.774
76.121708
22.253:
5.16
! 76.81212;
151.244;
30
.325
79.94962
29'
102
.656
39.
.426
61.59406
' 67.
.151 :
16.39'
75.592322
22.601
5
.347
! 76.34175;
153.35,
31.62
79.3805
30;
103
.548'
36.
.538
64.71395
67.
.276;
15
.683;
76.688567'
22.657'
5
.104
77.47275:
153.487
30
.468'
80.14946
31 •
89
.665'
32.
.365;
63.90453'
" 58.
.708;
14
.843;
74.717245
19.708'
4
.918
' 75.04567
133.521
28
.991
78.28731
32;
87
.894'
33.
.585
61.7892'
58.
.912;
15
.131
74.315929'
19.798
4
.972
;' 74.88635 1
134.829;
29
.466
78.14565
33;
94
.948
35.
.449
62.66483
62.
.748;
15
.893;
'74.671703
21.113
5
.226'
; 75.24748.
143.379
30
.457
78.7577
34;
' 100
.195
38.
.518!
61.55696'
65.
.998;
16
.743;
74.631049
22.194
5
.461
'' 75.39425;
150.839
32.06
'78.74555
35
"99
.847
37.44'
62.50263'
65.
.791
16
.398
75.075618
" 22.1 13
5
.299
76.03672
' 150.372
31
.394
79.12244
36
98
.724'
35.81
63.72716'
65.
.309;
16
.438
74.830422
21.986'
5
.332'
75.7482
149.493
31
.627
' 78.84383
37'
103
.372'
36.
.414'
64.77383
' 66.
.509,
16
.883
" 74.615466"
22.371
5
.536-
75.25368.
150.98
32
.694'
" 78.34548
38;
89
.169'
35.
871
59.77189'
59.61'
16.98
71.514847
20.037
5
.524
72.431;
136.233
32
.'705'
75.99333
39
94
.252
37.
411
' 60.30747'
62.
.548,
17
.222
72.465946
21.039
5
.595
73.40653
143.013'
32
.766
; 77.0888
4o;
"" 96
.381'
' 35.
.333'
63.34028
63.
.457.
16
.468;
74.048568'
21.337
5.39'
74.73872
144.757
31
.599
78.171
41
98
.367'
37.
.034 :
62.3512
64.
.804'
1 6
.801 ;
74.074131'
21.788
5
.527'
; 74.63283
148.01
32.43'
78.08932
42
101
.793'
38.27'
62.40409'
66.
.393.
17
.483;
73.667405'
22.322
5.7'
74.46465
151.408'
"33
.421
' 77.92653
43;
' 102
.717'
' 34.
.905
66.01828
66.
.874;
"17
.203
'74.275503
22.488 :
5
.643'
74.90662
152.081 '
' 33
.305
78.10049
44;
94
.515
40.
.619
57.02375
62.
.536;
19
.612
68.638864
21.007
6
.327
: 69.88147
1 42.765 :
37
.854
' 73.4851
45
" 94
.499
41.
.676
55.89795
63.
.291 ;
19
.767:
68.768071
21.263
6
.387
69.96191 ;
1 4 4.7 9 6 :
37
.778
73.9095
46 :
98
.897
41.
.753,
57.78133
65.
.393;
19
.723'
69.839279
22.001 :
6
.359
! 71.09677:
1 4 9.5 7 5 :
37
.681 '
74.80796
47'
97
.976
37.6
61.62325
64.
.447:
17
.891
72.239204
21.652
5
.824
; 73.1 01 79:
147.029
' 34
.187
: 76.74812
48
97.06
36.
.067
62.84051
64.
.099;
17
.405
72.84669
21.539
5
.631
73.85673
146.46
33
.412
77.18695
49 ;
99
.215
34.
.905
' 64.81883
65.
123
17
.112
73.723569
21.893
5.58
f 74.51 24;
148.642
32
.978
; 77.81 381
50;
99
.981
: ' 39.
.438
60.55451
64.
.974;
19
.932
69.3231 14'
21.82;
6
.451
70.43538
147.667
38
.622
: 73.84521
B-21
-------
CO
I
bO
bO
180
160
140
120
E
Q.
Q.
100
MEK inlet
MEK outlet
Butyl acetate inlet
Butyl acetate outlet
Ethylbenzene inlet
Ethylbenzene outlet
<—Xylene inlet
Xylene outlet
* * * * ** *
10 15 20 25 30 35 40 45 50
Sample #
Figure B-11. Two Parallel AMT Modules, Test 11, 4.0ft3/min Air, 1.0 L/min Oil, 350 ppm VOCs
-------
Table B-12. Two Parallel AMT Modules, Test 14, 4.0 ft3/min Air, 1.0 L/min Oil, 350 ppm VOCs
;Test 14
;Run date 6/06/00
Initial parameters: syringe set 4.0m l/hr; air 4c fm (1380-1390);sample flow rate 55cc's;injection temp 290C;
:oil pressure set at 25psi(11.2 at the module); : ;
; ; n-B uty I >n-Butyl Ethyl -Ethyl •
MEK ;MEK i ;Acetate ;Acetate ^Benzene ^Benzene ;o-Xylene ;o-Xylene
Front Rear \% |Front ;Rear \% -Front Rear .% jFront jRear \%
Detector
Detector
Difference
Detector
Detector
D iffe re n c e
Detector
Detector
Difference
Detector
Detector
Difference
1:
105.47
96.
.546
8.461174
46.
.818'
22
.832
51
.232432'
12.
.325'
5
.277
: 57.
.18458
39.
.746:
15
.632'
: 60.
.67026
2!
105.
.427
100.
.266
4
.89533
47.38'
24.09
' 49
.155762'
12.
481
5
.569-
55.
.38018
40.
.365'
16
.462'
.59.
.21714
3;
105.
.966'
100.33
5.318687
47.
.578;
24
.867
47
.734247'
12.
.519 ;
5.74
" 54.
.14969
40.
.482 :
16
.91 1
; 58.
.22588
4 -
115
.016
105.
.482
8.289281'
51.
301
25.81
49.68909
13.
.508'
6
.022
; 55.
.41901 ;
43.
.632'
17
.982
¦ 58.
.78713
5
' 300.
.907
184.
.784'
' 38
.59099
129.
.936
25
.698
' 80
.222571
34.11
5
.521
83.
.81413
109.
.754'
15
.308'
86.
.05244
6,
249.
.'146
195.
.881;
21
.37903'
115
.282:
29
.271
74
.609219
"30.
.238
6
.'1 12
• 79.
.78702
9;
3.35'
16
.631'
83.
.08998
7:
106.
.579
108.
.212;
-
1.5322
50.
.156
30
.601'
'38
.988356
13.
.117
6
.932"
47.
.15255
42.
.555;
"" 20
.095
52.
.77876
8
'106.
.103
108.
.816:
-2
.55695'
49'
29
.572
39.64898'
12.
.852
6
.749
.47.
.48677,
41.
.613;
19
.784
: 52.
.45716
'9
11 0.
.194'
111.
.831
-1
.48556'
50.
329
30
.206;
39
.982912
13.
.215;
6
.837'
: 48.
.26334'
42.
.797
" 19
.884'
53.5388
1°;
108.
.751
111.
.051
-2
.11492
49.
.584'
30.4
38.6899'
13.
.017
6
.877
: 47.
.16909
42.
.149
20.03
52.
.4781 1
1 1 !
109.
.137'
1 10.57
-1
.31303'
49.
.243
30
.717
37
^ 621591
12.
.923
6
.944
46.
.26635
41.
.748
20
.117
: 51.
.81326
12'
"107.
.887
110
.557
-2
.47481 '
49.
.189
' 30
.786
37
.412836
12.
.916
6
.963
46.
.09012,
41.
.788
20
.216'
51
.'62248
13
109.
.278
111.
.426;
' -1
.96563
49.
.377'
30
.869 ;
37
.483039'
12.
.961
7
.003
' 45.
.96868;
""41.
.859'
20
.387'
'51.
.29602
14'
108.
.253
110.
.739
-2
.29647'
48.
.737;
31
.175;
36
.034225'
12.
.785
7
.074
44.
.66953
41.28
20.63
: 50.
.02422
15
' 106.
.088
111.
.303
-4
.91573
48.
. 1 38
31
.451
34
.664922'
12.
.645
7
.178
'43.
.23448
40.
.934
20
.991'
48.
.71989
16
106.
.963
113
.362'
-5
.98244'
48.
.702,
32
.115;
' 34.05815
12.
.'802'
7
.316
42.
.85268
41.
.441
21
.317
4;
3.5606
17'
110.
.332
115.
.603
4.7774
50.
.225;
32
.71 2 ;
34
.869089
13.
.185
7
.449
' 43.
.50398
42.
.718
21
.651
49.
.31645
18
111.
.105'
114.
.675
-3
.21318
50.
.136:
33
.131 '
33
.917744
13
. 1 55 :
7
.535
42.7214
42.
. 51 7 :
21
.893
48.
.50766
19
109.
.971
114.
.096 :
-3
.75099
:' 49.
.862';
33
.618
32
.577915
13.
.085
7
.635'
41
.65075;
42.
.346'
' 22
.144
47.
.70699
20
107.
.338
112.
.593
-4
.89575'
48.
.896;
33
.756
30
.963678
12.
.812.
7
.691
39.
.97034
41.
.449'
22
.328'
" 46.
.13139
21
106.
.'586
111.
.945
'-5
.02786
' 48.
.382'
33
.901
29
.930553
12.
.699;
7.73
39.
.12907"
41.
.107
22
.'463
45.
.35481
22'-
105.
.332
112.
.315
-6
.62951
47.
.729'
' 34
.403
27
.920132
12.
.524;
7
.85Z
' 37.
.29639
40.
.555'
22
.821
: 43.
.72827
23
106.
.985
r 115.
.814'
-8
.25256'
49.
.054;
35.02
28
.609288
12.
. 869 ;
8
.005
: 37.
.79625
41.
.731'
23.
.249
44.
.28842
24
109
.767
"116.
.539 ;
' -6
.'16943
49.76;
35
.309
"29
.041399
13.
.051 ;
"8
.074"
38.
.13501'
42.
.206
23
.442
44.
.45813
25;
110.
.551
; 1 15.57
-4
.53999
50.
.304!
35
.538
29
.'353531
13.
.187
8
.145
; 38.
.23463';
42.
.707
23
.715
44.
.47046
26
109.
.347
115.
.847 ;
-5
.94438
49.
.865'
35
.873
28
.059761
13.
.061
8
.216
; 37.
.09517'
42.
.297'
23
.873
43
.55864
27'
110.
.'497
116
.581 ;
-5
.50603
¦ 50
.167'
36
.183'
27
.874898
13.
.159
8
.289-
37.
.00889
42.
.'607'
24.07
43.
.50694
28,
' 117.
.356
119.
.363:
-1
.71018
52.
.779;
36.61
30
.635291
13.
.856
8
.352
' 39.
.72286'
44.
.783
24.
.181
46.
.00406
29
115.04
:' 120.
.343;
4.6097
'52.
.095
'" 37
.604;
27
.816489
13.65
8
.552
' 37.
.34799
44.
.101'
' 24
.683
' 44.
.03075
30
'107.
.769
117.
.988:
-9
.48232
49.
645
38.1 1
"23
.234968
; 12.
.967'
8.71
32.
.82949;
42.
.035'
25
.161
40.
.14274
31
111.
.199
11;
3.72;
' -6
.76355'
50.
.758'
38
.282
24
.579377
13.
.283
8
.749'
;' 34.
. 1 3386
42.
.998
25
.304'
41.
.15075
32'
110.
.964
" 117.
.973;
-6
.31646
"50.
.516
38
.262
24
.257661
13.
.216
8
.762
33.
.70157'
'' ' 42.
.791'
25
.386
40.
.'67444
33
108.
.'632
116.
.986;
" -7
.69018
49.
.734;
38
.276
23
.038565
13.
.024'
8
.782
32.
.57064'
42.
.224;
25
.465
39.6907
34 ;
106.
.497
113.
.877
-6
.92977
48.
.584
38
.201
21
.371233
12.
.734;
" 8
.798
30.
.90938
41
.206
25
.558
; 37.
.97505
35
105.
.282
. 112.
.155
-6
.52818
47.
.979'
38
.387
19.99208
12.
.564
8
.855
29.
.52085;
40.
.633
25
.724
: 36.
.69185
36^
102.
.929'
111.
.133
-7
.97054
47.
.052'
38.27;
' 18
.664456
12.
.304;
'8
.879'
27.
.83648
39.
.825;
25.9
34.
.96547
37
100.
.181
109.
.982
-9
.78329
46.
.052
38
.283;
16.87006
12.
.023;
8
.893
'26.
.03344
38.
.935
25
.963
33.
.31707
38-
'103.
.344'
"113.
.257
' -9
.59224
47.
.655'
38
.827
18
.524814'
12.
.447;
9
.007
¦ 27.
.63718;
40.
.304 :
26
.223
'34.
.93698
B-23
-------
350
300
250
E
Q.
Q.
CO
I
bO
-1^
A A A
+ +++ + + +4 -t- -f 'l I ( :1; T A sb A J; J;
X-X-X^* A A A A X M " " " v *~X * X ***
HP*
X X X )< X X X
200
150
100
ffft I ¦ frl? *x**x*S$$S$$f*E*SE**E** HUUUUUi
10 15 20 25 30
Sample #
35
40
45
MEK inlet
MEK outlet
Butyl acetate inlet
Butyl acetate outlet
Ethylbenzene inlet
Ethylbenzene outlet
Xylene inlet
- Xylene outlet
50
Figure B-12. Two Parallel AMT Modules, Test 14, 4.0 ft3/min Air, 1.0 L/min Oil, 350 ppm VOCs
-------
Table B-13. Two Parallel AMT Modules, Test 15, 2.0ft3/min Air, 1.0 L/min Oil, 220 ppm VOCs
;Test 15 • • ; • • ;
• Run date 6/07/00 • ! r : • " V *" r i ; 7 -
Mnitial parameters: sample flow rate 55cc's;syringe set 2.0 ml/hr;air velocity ~2.0cfm (~500);
| Ethyl
;Benz
iRear
in-Butyl
.n-Butyl
Ethyl
;MEK
;m EK
;Acetate
Acetate
iBenzene
|Front
;Rear
%
; Front
Rear
%
IFront
IDetector
|Detector
Difference
iDetector
Detector
D iffe re n c e
jDetector
1
; 6.
.728
; 9.
.096
-35.1962
2.91 '
1
.099
62.
.233677
0.81
2
13.
.384
34.
.031;
-154.266
5
.574'
10
.914
-95.
.801938
; 1.54
3
11.
.191'
34.
.246:
-206.014
5.
.035'
13
.751 '
-173.10824
1.353
4
7.
.336'
: 31.
.168
'-324.864
3.
.705'
15
.283
-312.49663
0.973
5
6.
.317"
29.
.131;
-361.152
3.
.055'
'16
.239;
-431.55483
0.82
6
0
16.
.805
#D IV/01
0.
.916;
16
.066
-1653.9301
0.195
7
0
4.
.907'
#D IV/01
0'
13
.229
#D IV /01
0.0984
8
' 0
: 1.
.544'
#D IV/01
0";
'10
.756;
#DIV/0l
0.0786
9
0
f 8.
601
#D IV/01
0;
9
.206
#DIV/0l
0.065
10
0
14.
.966 ;
#D IV/01
0
12
.432;
#DIV/0l
0.0883
11
41.27
: ' 35.
.461
1 4.0756
; 15.
.282'
14
.993
1.8
191 1 137
4.183
12
48.
.805
52.
.'175
-6.90503
21.
.061 ;
18
.21 4 :
13.
.517877
5.565
13
38.
.646
; 48.
.678
-25.9587'
17.
.554;
19
.345
-10.
.202803
: ' 4.607
14'
123.
.869"
62.
.'397
49.62662'
46.
.728'
20
.095'
56.
.995806
12.385
15
129.
.843
105.
223
18.96136
; 55.
.723';
24
.799;
"55.
.495935"
14.551
16
117.
.593
! 103.
.599,
1 1.90037
; 51
.623
27
.364
46.99262'
13.426
17
119.69
100.
.921 :
15.68134
: 52.
.435
28
.592
45.
.471536
13.631
18
115.
.268
103.62
10.10515
51.
.258;
29
.534
42.
.381677
13.294'
19
" 121.
.178
109.
.174'
9.906089
54.
391
30.17:
44.
.531264
14.115
20
113.
.387
106.
705
5.893092
51.57;
30.96
"39.
.965096
13.367
21
r 117.
.521
109.
.113;
7.154466"
! 52.
.625
31
.509
40.
.125416
13.648
22
'112.
.831
107.
851
' 4.413681
51.
.287;
31
.716
38.
.159768
13.281
23
115.
.'453
106.
.038;
' 8.154834
: 51.
.497
"31
.819
38.
.21 1935
13.344
24
112.24
f 106.
.683;
4.950998
51.
.037;
32
.071
37.
.161275
i 13.208
25
107.
.861
105.
151
' 2.512493
48.
.986
32
.237:
34.
.191402
12.669
26"
; 111.
.484
"109.
.778
1.530264'
;' 51.
.237
32
.908
35.
.772977
13.248
27
; 105.
.621
107.
.633
-1.90492
48.
.'806'!
33
.159
32.
.059583
12.616
28
: 112.
.309
110.
.408
1.692652
51.
.428'
33
.397
35.
.060667
13.3
29
113
.'297
; "108.
.437;
4.28961
51.
381
33
.549';
34.
.705436
; 13.292
30
r 108.
.576
101.77'
' 6.26842
; 47
.'691
32
.715
31.
.'402151
; '12.379
31
107.
.545
107
.399
0.135757
! ' 48.
.785'
33.54,
31.
.'249359
12.662
32
111.
.265
109.18:
1.873905
; 50.
.511;
33.99'
32.
.707727
13.069
33
111.
.241
; 112.
.387'
-1.0302
i 51.
.243
34
.594 :
32.
.490291
13.246
34
105.
.404'
;' 110.
.399
-4.73891
49.
.118'
34
.774
29.
.203143
12.697
35
113
.319
• 111.
.859'
1.288398
51.
.465
34
.855
32.
.274361
13.318
37
112.
005
| 1 18.
. 259 ,
' -5.58368
; 51.19
34
.465
32
672397
I 13.215
38
110.
.054
;' 116.
.358'
-5.7281"
50.
.602'
35
.111
' 30.
.613414
13.066
39
107.
.841'
108.
451
-0.56565
49.64
35
.183
29.
'123691
12.822
40
; 118.
.185
113
1 42
"4.267039
53.58-
35
.505;
' 33.
.734602'
13.856
41
; 109.
.017
116.
.007 :
' -6.41184
50.
.772'
37
.108
26.
.912471
13.1
42
\ 111.
.357
" 112.
.726"
-1.22938"
51.
.069'
37
.043"
27.
.464803
13.197
43
112
.028
113.
.582
-1.38715
51.
.283'
37
.274-
27.
.317045
13.245
44
109.
.266
111.
.869;
-2.38226
50.
.268
37
.304
25.
.789767
12.977
45
110.
.727
115
.051
-3.9051
51.
.107:
'38
.477;
24.
.712857'
= 13.198
:ene
o-Xylene
;o-Xylene
%
;Front
;Rear
%
ctor
Difference
'Detector
iDetector
Difference
0
.277
65
.80247
; 3
.051
; 0
.605
80.17044
2.23'
-44.8052
5.
03
5
.125
2.972359
3
. 1 69~
-
134.22
4.
.752
7
.707
-62.1843
3.57
: -266.906
3.
.612
" 8
.931
-147.259
3
03
0
f -363.659
3.
.089
; 9
.616
-21 1.298
3.85'
' -1874.36
0.
.932
9.94'
-966.524
3.43
' -3385.77
0.
CD
"3"
9
0
CD
-1964.8
3
0
0
i -3719.34
' 0.
.341
8
.363
-2352.49
2
.672
-4010.77
I
D.28
7
.599'
-2613.93
3
.174
-3494.56
0
.371
8
.572'
-2210.51
3
.'613
i' 13
.62658
; 13.
.417
9
.544
28.86636
' 4
.'236'
23.8814
18.
.133
r 11
.143'
38.5485
4
.519'
1.910137
15.
.096
11
.964'
' 20.74722
4
.675
' 62
.25273
r 39.
.116
12.51
68.0182
5
.548
'61
.87204
46.
.852
14
.712
68.59899
6.13
| 54
.34232
43.32
! 16
CD
O
. 62.12835
6
.434
!52
.79877
43.
CD
O
03
; 17
.402
60.36713
6
.671
49
CO
CD
-J
42.
.938
;' 1.
8.15
' 57.72975
6
.817
: 51
.70386'
45.
.654
; 18
.601
; 59.25658
6
.991
47
.69956'
43.
-t*
03
19
.128
55.77137
'7
.215'
: 47
.'1351 1
' 44.
.043
;' '19
.772
55.10751
7
.181 ;
' 45
.'93028
43.
.031
19
.778
54.03779
7
.236'
45
.'77338'
43.
.059
; 19
CD
03
03
53.57997
7
.299
44
.73804'
42.
.711
20
.212
; 52.6773
7
.346
; 42
.01594
40.
.973
20
.338
50.36243
7
.559-
42
.94233
42.
.946'
20
.893
51.35053
7
.656
' 39
.31516'
40.
.958
21
.184
48.27872
7
.691
42
.17293
43.
.153
21
.276'
50.69636
7
.762
: 41
.'60397
; 43.
.009'
21
.472
50.07557
7
.623
' 38.4199
39.
.878'
21
.117
47.04599
"7
.758
38
.73006
40.
.833
f 21.5
47.34651
7
.853
39
CD
' 42.
.305
! "21
;754
! 48.57818
7
.979
: 39
.76295
42.
.977
; 22
0
03
48.61438
8
.042
36.6622'
41.
.257
22
.291
45.97038
8
.069'
39
JV
03
43.
0
CD
22
-t*
-t*
CD
47.9069
7
.875 ;
40.40863;
42.
.856;
22.11 ;
03
.40862
8
.042';
38.45094
42.
371
22
.664;
46.
.51059
8
.032';
37.35767''
41.
566
' 22
.582
'45.
.67194
8.11"
41.4694
44.
817.
22
.755 '
49.
.22686
8
.527;
34.9084;
' 42.
.541 :
' 23
.552;
44.
.63694
8
.525'
35.40199
42.
744
23.6"
' 44.
.'78757
8.61
34.99434;
42.
887 :
23
.836 ;
' 44.
.42139
8
.611 ;
33.64414;
42.
027 ;
23
.891 ;
' 43.
.15321
9
.082':
31.18654
42.
734
24
.068
43.
.67951
B-25
-------
CO
I
bO
On
140
120
100
E
Q.
Q.
XK-X-XXXXX^XkX X-j|H|frX * * X X * X X
MEK inlet
MEK outlet
Butyl acetate inlet
Butyl acetate outlet
Ethylbenzene inlet
Ethylbenzene outlet
Xylene inlet
Xylene outlet
10 15 20 25 30 35 40 45 50
Sample #
Figure B-13. Two Parallel AMT Modules, Test 15, 2.0 ft3/min Air, 1.0 L/min Oil, 220 ppm VOCs
-------
Table B-14. Two Parallel AMT Modules, Test 16, 8.0ft3/min Air, 1.0 L/min Oil, 200 ppm VOCs
;Test 16
;Run date 6/08/00
Initial parameters: sample flow rate ~50 cc's;mag(1) 1.75"H20;mag(2) > 1,0"H20;air velocity ~8.0cfm
;oil pressure
11.5psi (on module);syringe flow <
3.0m l/hr;0iI sam pling tim e 120 m in.
n-Butyl
• n-Butyl
;Ethyl
^ Ethyl
:m EK
;M EK i
;A cetate
;Acetate
;Benzene
;Benzene
o-Xylene
;o-Xylene
Front
Rear |c
'/o
;Front
;Rear
%
i Front
iRear
%
Front
;Rear
%
;Detector
IDetector ^Difference
:Detector
;Detector
Difference
;Detector
;Detector
Difference
Detector
IDetector
Difference
1
4
.372
; 10.217
-133.692
; 1.
.726
9
.928
-475.20278
; 0
.528
j 2.653
: -402.462'
2.07
! 7
.494
-262.029
2
5
.958
: 12.334;
-107.016'
2.
.579
10
.821;
-319.58123
0
.734
2.869
-290.872
2.744
8
.208'
| -199.125
3
104
.996
98.911
5.795459
44.
.006
30
.255
31.248012
11
.614
7.203
:' 37.98002
37.457
" 20
.648
;' "44.
.87546
4
; 106
.824
99.866;
6.513518'
45.
.573
35
.676
21.716806
11
.917
8.605
27.79223
38.428
| 25
.219
"34.
.37337
5
102
.041
101.48;
0.549779
45.
.093
"' 37
.504
16.829663
11
.758
9.114"
22.48682
38.144
27
.003
29.
.20774
6
104
.018
101.963;
1.97562
45.89
38.17'
16.822837
11
.933
9.321
21.88888
38.668
27
.871'
27.
.92231
7
; 104
.187
97.588
'6.333804
45.
.683'
"38
.365
16.019088
11
.879
9.439
/ 20.54045
38.365
28
.465'
25.
.80477
8
99
.362
J 101.927.
-2.58147'
44.
.865
39
.336 :
12.323638'
11
.636
; 9.63
17.2396,
37.807
29
.057'
23.
.14386
9
"104
. 011"
102.392
1.556566
46.
.309
39
.604"
14.478827
12
.017
9^05
19.23941
38.902
29
.294
24.
.'69796
10
i 105
.478
; "101.654
3.625401'
" 47.
.044
40.09 :
14.781906'
i 12
.198
"" 9.823"
19.4704
39.539
29
.691
24.
.90705
1 1
"100
.835
103.876;
-3.01582
45.
.543
41
.013
9.9466438
11
.'799
10.053
'14.79786
38.31
30.37'
20.
.72566
12
104
.359
104.31 :
0.046953'
46.
.963
41
.309
12.039265
I 12
.174
10.117
: 16.89667
39.52
30
.574
"22.
.63664
13
103
.067
101.524;
1.497084
46.
.281
40
.838
"1 1.760766"
12
10.02
16.5;
38.903
; 30.36'
21.
.95975
14
:' 100
.093
" 102.068;
-1.97316
45.
.314
41
.307
8.8427418
11
.723
10.145'
13.46072
38.045
; 30
.803'
" 19.
.03535
15
101
.703
103.98
-2.23887
46.
.066
41
.339'
' 10.261364"
11
.932
10.137
.' 15.04358
38.765
; 30
.824'
. 20.
.48497
16
103
.974
;" 103.048
0.890607'
46.
.511
"41
.046"
1 1.749909
12
.043
10.112
16.03421
38.974
30
.808
¦ 20.
.95243
17
; 98
.924
j 101.736
-2.84259
; ' 44.83
41
.295
7.8853446
11
.602
10.1 84
12.22203
37.675
; 31
.076
: 17.
.51559
18
! 7
.051
21.364
-202.992
| 7.
.433
25
.178
-238.73268
i 1
.842
6.744
-266.124
6.596
: 21
.383
¦ -224.181
19
: 8
.567
} 17.1 54;
-100.233'
5.
.703
; 18
.361
-221.95336
1.49'
4.986
-234.631;
5.245
16
.016
-205.357
20
96
.129
81.157:
15.5749
" 39.
.473'
; 25
.829
34.565399
10
.'348
6.415
38.00734
33.104'
19
.249
: 41.
.85295
21
98
.473
92.166;
6.404801
42.91
35
.465'
17.350268
11
.144'
; 8.685
22.06569
35.981'
26
.169
27.
.26995
22'
98
.099
100.21 ;
-2.15191'
44.
.061
39
.362
10.66476
11
.414'
9.66
i 15.36709'
37.063
29
.144
21.
.36632
23
5
.387
: " 22.932
-325.691'
; 7.
.155
25
.476
-256.0587
1
.759
; 6.666
-278.965
• ' 6.363
; 21
.'135
-232.155
24
: 12
.629
19.566
-54.9291
6.
.981
" 18
.01 1
-158.00029
1
.824
4.818
-164.145;
6.28
15
.'482
| -146.529
25
0
10.76;
#D IV/01
2.
.155
15
.932
-639.30394
0
.531
4.37
-722.976'
2.345"
13
.937
/ -494.328
26
109
.227
1 15.757
-5.97838
i 50.
.742
38.5;
24.125971
13
.298'
: 9.269
30.29779
42.884
27
.663
: 35.
.49342
27"
2
.216
13.342
-502.076
;' 2.
.244
19
;724,
-778.96613
; 0
.808
5.326
-559.158
3.204
16.67
: -420.287
28
'" 94
.879
;' 83.61
1 1.87723
; 39.
.239'
j 27
.744
29.294834
10
.301
7.092
' 31.15232,
33.137'
: " '2'1
.277'
35.
.79081
29
96
.357
97.082
-0.75241
42.
.824'
: 37
.272,
12.964693
; " 11
.104"
; 9.082
' 18.20965';
36.067
| 27
.213
: 24.
.54876
30
'101
.499
; 99.005 ;
2.457167'
45.
.101
39
.286,
12.893284
: 11
.678
9.658
17.29748'
37.871
; 29
.137
:' 23.0625
31
102
.192
97.82
4.278221
;'" 45.
.622
39
.871
12.60576
"11
.804'
9.853
16.5283
38.298'
29
.973'
21.
.73743
32'
99
.815
;' 100.198
-0.38371'
44.
.847'
40
.987
8.6070417
'11
.595
10.14
12.54851
37.596
; 30
.948
17.
.68273
33
101
.665
101.546
'0.1 17051*
45.
.979
41
.262,
10.259031
11
.881"
' 10.208
14.08131
" 38.619
31
.248
'19.
.08646
34
105
.134
103.295
1.749196
47.
.261
41
.'584'
12.012018
12
.218
10.296
: 15.73089',
39.639
31
.624
20.
.21999
35
103
.982
101.145
2.728357
" 46.
.413
42
.068 :
' 9.3616013
; 11.99
10.431
i 13.0025;
38.852
! 32
.138
17.
.28096
36
: 9'
8.23
; 102.659
-4.50881
44.96
41.87,
6.8727758
11
.618
10.426
I 10.25994
37.782
! 31
.127
:' 17.
.61421
37
103
.712'
102.89;
0.792579
46.
.627
41
.674'-
10.622601'
12.05
10.375
' 13.90041 ;
39.078
32
.001'
18.
.10993
38
" 103
.382'
100.801 ;
2.496566
46.56
41
.532
10.798969
12
.034
10.365'
13.86904'
39.071
32
.014
18.
.06199
39
99
.235
101.77
-2.55454
45.08
42
.219
6.3464951
11
.654
10.517
9.756307
37.847
'" 32
.487
14.
.'16228
40
r 100
.449
104.481
-4.01398
45.
.973
42
.472
7.6153394
11
.864
10.576'
' 10.85637'
"" 38.64'
32
.648
15.
.50725
B-27
-------
CO
I
K>
00
E
a
a
140
120
100
80
60
40
20
* * x * * x
Sample #
~— M EK inlet
MEK outlet
Butyl acetate inlet
Butyl acetate outlet
*— Ethylbenzene inlet
•—Ethylbenzene outlet
Xylene inlet
- Xylene outlet
Figure B-14. Two Parallel AMT Modules, Test 16, 8.0 ft3/min Air, 1.0 L/min Oil, 200 ppm VOCs
-------
Table B-15. Two Parallel AMT Modules, Test 18, 4.0ft3/min Air, 1.0 L/min Oil, 200 ppm VOCs
;Test 18
;Run date 6/21/00
Initial parameters: sam pie flow rate ~55 cc's;air velocity ~ 1.38 (4cfm)
:oil pressure 11.0psi (on module);syringe flow 4.0m l/hr;used "new" silicon oil for this run.
; !n-Butyl -n-Butyl SEthyl Ethyl j
MEK ;MEK i jAcetate iAcetate ;Benzene jBenzene : ;o-Xylene ;o-Xylene
Front Rear \% iFront iRear \% ;Front ;Rear :% ;Front jRear \%
Detector
Detector
Difference
Detector
Detector
D iffe re n c e
Detector
Detector
Difference
Detector
Detector
Difference
1
88
002
74
253
15.62351
38
384
18
968
50
583576
9
834
4
455
54
69799
31
919
12
864
59
69799
2
98
329
78
066
20.60735'
41
959
19
522
53
473629
10
748
4
561
57.5642
34
748
13
265
61
82514
3
100
382
82
108
18.20446
"""44
044
20
361
53
771229
11
273
4
753
57
83731
36
618
13
864'
62
13884
4
104
511
90.61
'13.30099
45.41
27
819
38
738163
11
636
5
045
56
64318
37
795
14
488'
61
66689
5
106
649
"89
638
15.95045
"47
186
25
384
46
204383
12.07
5
244
56
'55344
39
268
15
223'
61
23307
6
93
856
83.16
1 1.3961 8
41
742
22.67
45
690192
10.67
5
246
50
8341 1
34
727
" 15
434
55.5562
7
95
895
81
769
14.7307
42
483
22
552
'46
915237
10
872
5
265
51
57285
35
322
15
583
55
88302
8
109
373
85
859
21.49891
47
383
23
043
51
368634
12
146
5
514
54
60234
39.32
15
846
59.6999
9
95
189
89
645
5.824202
"" 43
777
24
283
44
530233
11
148
5
756
' 48
36742
"36
518
16.6
54
54297
10
95
905
86
355
9.957771'
43
286
24
361
43
720834
11
056
5.78
47
72069
36
053
16
711
53
64879
1 1
97.77
89
541
8.416692
44
145
24
738
43
961944
11
278
5
861
' 48
03157
' 36
775
'16
953
53
90075
12
1 02
199
90.91
1 1.0461
45
585
25
125
'44
883185
11
679
5
938
49
15661
37
989
17
167
54.8106
13
98
691
85
092
"13.77937
43
932
24
762
43
635619
11
237
5
854
' 47
90424
36
552
17
085
"53
25837
14
94
481
"86
745
8.18789'
"42
561
25
279
40
605249
10
874
6
005
' 44
77653
" 35
454
17
391'
50
94771
15
96
441
87
659
9.106086
43
501
25
728
40
856532
11
'128
6
076
' 45
39899
" 36
279
17.62
51
43196
16
95
747
' 88
644
7.418509
43
024
25
962
' 39
656936
11
002
6
'166
43
95564
35
864
17
847
50
23701
17
96
454
89
096
7.628507
43
224
26
093
39
633074
11
039
6
209
43
75396
35
987
17
884
50
30428
18
99
554
90
258
9.337646
44
212
26
628
39
772008
11
301
6
343
43
87222
36
746
18
086
50
78104
19
103
593
90
436'
12.70067
45.5
27
284
40
035165
11
638
6
501
44
13989'
37
815'
18
397
51
34999
20
99
296
92
753
6.589389
44
934
28
513
36.54471
11
467
6
755
41
09183
37
416
18
997
49.2276
21
121
198
108
551
10.43499
54
427
30
621
43
739321
13
906
7
224
4
8.0512'
45
317
19
986
55
89735
22
100
809
106
562
' -5.70683
48
806
31
229
36
014015
" 12
414
7
398
40
40599'
40
971'
" 20
654
49
58873
23
99
665
94
027'
5.656951
45
047
30
085
' 33
214199
11
516'
7
155
'37
86905
37.52
' ' 20
078
46
48721
24
103.82
91
322
12.03814
" 45
887
29
442
35
838037
11
736'
7
073
39
73245
38.08
19
913
47
70746
25
' 95
252
' 90
674'
4.806198
43
298
29
'788
31
202365
11
042
7
122
35
50082
36
048
20
062
44
34643
26
92
'419
89
217'
3.464656
42
174
29
426
30
'227154
10
742'
7
051
34
36045
35
119
20
056
42
89131
27
96
198
91
047'
5.354581
" 43
616
' 29
987
31
247707
11
129'
7
189
35.403
36
265'
20.41
43
71984
28
99
273
' 93
876'
5.436524
44
692
30.2
32
426385
11
413
7
201
36
90528
37
'176'
20
539
44
75199
29
96
'808
90.61
6.402363
43
864
30
157
31.24886
11
192
7
213
35
55218
' 36
515
20
538
'43
75462
30
"93
933
90
179
' 3.996466
42
589
' 30.32
28
807908
10.86
7
296
32
81768
35
441
20
732
41
50278
31
96
878
" 92
791
4.218708
43
709
30
782
29
575145
11
149
7
352
34
05687
36
376
20
975
42
'33835
B-29
-------
CO
I
O
140
120
100
E
Q.
Q.
60
20 *-*-
*-x
xxxxxxxxx x r*-*-*-*-* * x * x x :
«- MEK inlet
¦- MEK outlet
+- Butyl acetate inlet
*- Butyl acetate outlet
Ethylbenzene inlet
»- Ethylbenzene outlet
Xylene inlet
- Xylene outlet
0 5 10 15 20 25 30 35 40 45 50
Sample #
Figure B-15. Two Parallel AMT Modules, Test 18, 4.0 ft3/min Air, 1.0 L/min Oil, 200 ppm VOCs
-------
Table B-16. Two Parallel AMT Modules, Test 19, 8.0ft3/min Air, 1.0 L/min Oil, 200 ppm VOCs
;Test 19
;Run date 6/22/00
Initial pa ram eters: s am pie flow rate ~55 cc's; a ir veloc ity ~4.75 (8cfm );sy ringe flow rate 8.0 m l/hr
i ;oil pressure 11.3psi (on module);Mag(1) ~1.85"H2O;Mag(2)>1.0"H2O
! jn-Butyl >n-Butyl jEthy I j Ethy I
MEK jMEK ;Acetate ;Acetate i ;Benzene IBenzene • -o-Xylene ;o-Xylene
Front ;Rear i% ;Front ;Rear ;% ;Front iRear % -Front IRear ;%
Detector
Detector
Difference
Detector
Detector
D iffe re n c e
Detector
Detector
Difference
Detector
Detector
Difference
1
87
752
5.59
92.4902
31
154
7.001
77.527765
8
579
1
CD
CO
CD
76.81548
27
151
5.785
78.69323
2
"" 98
834
" 89
061
9^888298
40
CD
^i-
24.203
40.186338
10
638
5
768
45.77928
34
-t*
o
CO
' 16.5
52.04604
3
94
184
92
763
1.508749'
40
316
30.443
24.489037
10
516
7
336
' 30.23963
34
138
21^497
37.02912
4
101
746
102
712
-0.94942
44
876
33.737
24.821731
11
678
8
146
30.2449
38
068
24.035
36.86298
5
102
679
98
565
4.006662
44
031
36.08
'18.057732
11
453
8.75
23.6008
37
CD
CD
O
25.96'
29.96277
6
98
251
100
CD
CD
-2.08446
43
699
35.594
18.547335
11
362
8
622
24.1 1547
36
917
25.749
30.25165
7
100
983
97.21
3.736272
45.28
35.73
21.090989
11
758
"8
676
26.21194
38
308
26.28
31.39814
8
95
234
99
253
-4.22013'
42
497
36.26
14.67633
11.01
8
CO
o
20.03633
' 35
869
26.615
25.79944
9
99
-t*
CO
' 1 01
955
' -2.54154
44
533
37.29
16.264343
11.54
9
427
18.31023
37
649
28.1 13
' 25.32869
10
105
346
98
155
6.826078
46
343
37.472
' 19.14205
12
032
9
491
21.11868
39
005
' 28.291
' 27.46827
1 1
91
263
97
CO
CO
-6.6018'
41
543
37.21 1
10.42775
10
741
9
445
12.06592
35
139
28.239
19.6363
12
101
782
103
667
-1.852
'45
-t*
CO
CO
38.602
15.138058
11
786
9
742
17.34261
38
402
29.115
24.18364
13
105
578
9
3.95
6.277823
46
032
38.622
16.097497
11
934
9
748
18.31741
38
619
29.365
23.9623
14
98
515
106
406
-8.00995
45
073
40.197
10.818006
• 11
674
10
124
13.27737
' 38
195
' 30.192
20.953
15
101
CD
o
103
804
-2.36475
45
CD
O
39.628
13.66825
11
885
10
014
15.74253
38
814
30.052
22.57433
16
97
151
101
602'
-4.58153
43
782
39.84
9.0037002
11
234
10
087
10.21008
36
951
30.225
18.20248
17
100
032
103
511
-3.47789
45
353
40.499
10.70271
11
755
10
269
12.64143
38
385
30.568
20.36473
18
103
783
98
832
4.770531
46
622
40.179
13.819656
12
087
10.2
15.61181
39
422'
30.52
22.5813
19
" '96
901
102
216
-5.48498
43
784
40.842
6.7193495
11
326'
10
372
8.423097
36
968
"30.918
16.36551
20
' 102
452
'105
128'
-2.61 195
45
858
41.313
9.91 10297
11
877
10
CD
CD
1 1.8801
38
'689'
31.231
19.2768
21
103
'431
99
416
3.881815
46
118
40.477
12.231667
11
956
10.13
'15.27267
38
819'
30.296
21.95574
22
"95
778
102.6
-7.12272
43
CO
CD
CD
41.546
' 5.360031
11
346'
10
CD
CO
7.474'
'37
128'
" 31.203
15.95831
23
102
836
05
654'
-2.74029
46
293
41.998
' 9.2778606
11
CD
CO
10.61
1 1.46529
39
CD
CD
O
31.583
19.15476
24
104
194
100
933
3.129739
45
502
42.163
7.338139
11
777
10
701
9.136452'
38
o
CO
31.96
16.3526
25
102
005
107
'975
-5.85265
46
444
43.198
6.9890621
12
035'
10.97
8.84919'
39
317'
' 32.688
16.86039
26
104
CO
CD
106
CO
-t*
CO
-1.92599
47
359
42.427
10.414071
12
268
10.8
1 1.96609
39
961'
32.479
'18.72326
27
94
CO
o
CD
105.07
-10.706
44
112
43.054
2.3984403
11
405
10
979
'3.735204
37
366
32.862
12.05374
28
"101
'171
106.11
-4.88183
' 45
978
43.358
5.6983775
11
CO
CD
CO
11
063
7.017986'
38.83'
32.914
15.23564
29
103
834
101
307
2.433692
46
772
42.906
8.265629
12
101'
10
977
9.288489'
"39
426'
32.77
16.88226
B-31
-------
MEK inlet
MEK outlet
Butyl acetate inlet
Butyl acetate outlet
Ethylbenzene inlet
Ethylbenzene outlet
Xylene inlet
- Xylene outlet
0 5 10 15 20 25 30 35 40 45 50
Sample #
Figure B-16. Two Parallel AMT Modules, Test 19, 8.0 ft3/min Air, 1.0 L/min Oil, 200 ppm VOCs
-------
Table B-17. Two Parallel AMT Modules, Test 20, 16.0 ft3/min Air, 1.0 L/min Oil, 200 ppm
VOCs
;Test 20 • • •
!Run date 6/22/00 1
jsequence used 062200c '
Mnitial parameters: jair velocity 16ml/hr (1657 volts);syringe flow rate 16ml/hr; ,
sam pie flow set at 55 cc's; M ag(1) > 4.0"H2O ;M ag(2) > 1,0"H2O; ;
oil pressure at module 11 .Opsi;system oil pressure ~ 25psi ; ;
; ;n-Butyl -n-Butyl Ethyl i E thy I
MEK ;MEK i :Acetate iAcetate ; iBenzene :Benzene -o-Xylene ;o-Xylene
Front ;Rear \% ;Front iRear \% ;Front Rear % -Front iRear .%
Detector
Detector
Difference
Detector
Detector
Difference
Detector
Detector
Difference
Detector
Detector
Difference
1
0
1
251
#D IV/01
0
3
419
#D IV /01
0
101
1
258
-1145.54
0
481
4
407
-816.216
2
94.2
8
1.32
'13.67304
" 39
143
14
503
' '62.948675
10
293
3.84
62.69309
32
703
11.25
65.59949
3
100
263
89
618
10.61708
42
CD
o
30
555
27.946517
11
213
7.81
30.3487
36
'398
23
817
34.56509
4
91
144
91
918
-0.84921
41
958
31
763'
24.298108
10.98
8
116
26.08379
36
301
25
025
' 31.06251
5
96
268
90
056
6.452819
41
758
34.67
16.973993
10
798
8
911
17.47546
35
262
27
03
03
' 21.02547
6
96
069
94
893
1.22412
40
141
35
055
12.'670337
10.33
8
CD
CD
'12.95257
32
369
28
529
1 1.8632
7
94
CO
o
CD
' 93
514
i. 365904
47
131'
35
078
25.573402
12
374
8.96
27.5901 1
42
379
27
875
34.2245
8
81
154
93
845
-15.6382
40
825
37
375
8.4507042
10
572
9
596
9.231933
35
465
' 30
663
13.5401 1
"9
100
559
94
362
6.162551
45
199
36
741
18.712803
11
553
9
404
18.60123
37
544
29
CD
21.56936
10
97
699
99
159
-1.49439
43
812
39
689
9.4106637
11.39
10
139
10.98332
36
518
32
113'
12.06254
1 1
92
-t*
-t*
03
94
316
-2.0206
38
252
38
589
-0.8809997
9
812
9
03
03
CD
' -0.78475
31
751
31
437
0.988945
12
95
587
99
716
-4.31963'
40
184'
39
453
1.819132
10
321
10.02
2.916384
33
734
' 30
CD
O
8.389162
13
94
787
99
CD
o
03
-5.40264
45
001
38
815
"13.746361
11
755
9
912
' 15.67843
38
955
30
993
20.43897
14
94
CN
CO
•CO
'103
119
-8.91088
' 41
141
39
233
4.6377093
10
576
10
016
5.295008
' 34
067
"30
938
9.184842
15
92
419
99
139
' -7.27123'
43
623
39
CD
03
03
8.3327602
11
'274
10
172
' 9.774703
37
746
31
CD
o
CD
16.26662
16
100
865
97
'135
3.698012
44.62
41
154
7.7678171
11
565
10
576
' 8.551665
38.54
33.61
' 12.7919
17
' " 10
036
101
095
-907.324
18
313
40
565
-121.50931
4
627
10
363
-123.968
'17
997
32
654
-81.4414
18
98
o
-t*
CD
121
585
-24.0043'
43
085
25
919
39.842172
11
195
6
657
'40.53595
' 36
368
" 21
015'
42.21568
19
88
711
99
752
-12.446'
41
696'
39
287
5.7775326
10
794
10
059
6.809339
35
962
31
734
' 1 1.75685
20
97
192
94.77
2.491975
43
112
40
141
6.8913528
1 1
261
10
CD
8.587159
37
576
32
267
14.1287
21
97
516
98
534
-1.04393
45
042'
39
707
1 1.844501
11
743
10
'198
13.15677
""37
722
' 32
03
"14.56445
22
" 96
367
101
079
-4.88964
43
825
41
436
5.4512265
11
329
10
595
' 6.478948
37
'147
33
287
10.391 15
23
91.51
96
CD'
03
CD
-5.98732
43.62'
41
484
4.8968363
11
354
10
'651
' 6.191651
37
398
33
537
"10.32408
24
98
372
94
135
4.30712'
43
934
41.01
6.6554377
11
457
10
533
8.064938
37
653
33
054
12.21417
25
97.53
100
278
-2.81759
43
962
41
724
5.0907602
11
375
10
'726
5.705495
37
483
33
679
10.1486
26
94
823
98
003
-3.35362
41
885'
'40
956
2.2179778
10
786
10
543
2.25292
35
583
33.41'
6.106849
27
88
641
9
3.97
-1 1.6526
42
744
41.12
3.7993637
11
1 1
10
572
4.851049
36
678
33
233
9.392551
28
97
423
94
395
3.108096
42
332
41
217
2.6339412
10
943
10
CD
CN
CD
2.896829
35
CD
03
03
33
395
7.20518
29
97
03
CD
99
854'
-2.00217
44
702
41
551
7.0489016
11
617'
"10
703
7.86778
38
239
33.6
' 12.13159
30
96
574
100
745
-4.31897
43.35
42
547
1.8523645
11
155
10
968
1.676378
36
603
34
233
6.474879
31
90
151
" 97
876'
-8.56896
42.7
42
216
1.1334895
11
069'
10
893
1.590026
36
659'
34
253
6.56319
32
99.42
97
CD
o
CD
'1.824583
43
CD,
"' 42
314
2.2590779
11
'199
10.93
2.402
36
307
34
202
5.79778
B-33
-------
CO
I
-1^
140
120
100
E
Q.
Q.
wrtri
MEK inlet
MEK outlet
Butyl acetate inlet
Butyl acetate outlet
Ethylbenzene inlet
Ethylbenzene outlet
Xylene inlet
¦ Xylene outlet
5 10 15 20 25 30 35 40 45 50
Sample #
Figure B-17. Two Parallel AMT Modules, Test 20, 16.0 ft3/min Air, 1.0 L/min Oil, 200 ppm VOCs
-------
Table B-18. Two Parallel AMT Modules, Test 21, 4.0 ft3/min Air, Maximum Oil Flow, 450 ppm
VOCs
;Test 21
;Run date 6/23/00
Initial parameters: s am pie flow rate 54c c's; syringe set 4.0 m l/hr; air velocity 4.0cfm (~680 short ridge);
;oilflowsettomaxoutonrotameter(147-150);Mag(1)0.4"H20;Mag(2)>1.0"H20
; :n-Butyl m-B uty I Ethyl : E thy I
MEK ;MEK • Acetate Acetate : Benzene -Benzene o-Xylene ;o-Xylene
Front Rear % Front 'Rear % Front .Rear .% .Front IRear \%
Detector
Detector
Difference
Detector
Detector
Difference
Detector
Detector
iDifference ;
Detector
Detector
Difference
1
0
13
.367
#D IV/01
0;
21
.175
# DIV/01
0.
.333
6.36 :
; -1809.91'
1.
.528';
19
.716
-1190.31
2:
64.
.951
15.
.655
75.89721
11.
.895'.
22
.'193'
-86.
.574191
3.
.618
6
^654
' -83.9138
9.
. 71 2 ^
19
.985
-105.776
3
192.
.374'
134.
.858
' 29.89801
79.
.345
34
.685'
56.28584'
20.
.934
8
.844'
57.
.75294;
67.
651
26
.281
61.15209
4"
"177.
.908
141.
.566;
20.42741
"* 76.
.279:
41
.891 '
45.
.081871
19.
.989;
10
.427
47.
.83631 ;
64.
.908
30
.654
^ 52.77316
5
159.
.284
142.
. 272 >
10.68029
70.
.565;
44
.332'
37.
.'175654'
18.
.434
11
. 1 41
39.
.56276!
60.
.'145
33
.167
: 44.85493
6 :
147.
.608'
141.
.718
' 3.990299
"66.
181
46.17'
30.
.236775'
17.
.267'
11
.632-
¦ 32
.63451
56.
.'442
34.
.974'
'38.03551
7
' "187.
.124
"158.
.048
15.53836
80.
.764:
48
.066'
40.48586
21.
.135
12.12
42.
.65436,
68.
.647
' 36
.466
:' 46.87896
8
209.
.879
173.
.806
17.18752
90.
,432:
49
.976 :
44.
.736377
23.
.642
12
.519'
47.
.04763;
76.
.'835
37
.552
! 51.12644
9
'157.
.312
156.
.'131
0.750737
73.
476 :
51
.195
30.
.324187
19.
1 49 ;
12
.823'
33.
.03567;
"62.
.953
' "38
.763
"38.42549
10
154.
.252
160.
.913'
-4.31826
70.
.418 =
52
.651'
25.
.230765'
"18.
.325
13
.544
; 26.
.09004;
60.
.038
40
.342
: 32.80589'
11
147.
.451
158.
.278'
-7.34278
68.
.075;
52
.786'
"22.
.459053
17.
.727
13
.511
'23.
.78293;
58.
.125'
40
.832
29.7514'
12'
199.
.244
173.
.059'
13.14218
87.
. 198":
53
.199'
38.
.990573
: 22.
.792
13.6:
; 40.
.32994;
74.
163
40
.954
; 44.77839
1 3 ^
193.64
179.
.061
7.52892
85.
.101 i
55
.542,
34.
.734022
22.
.199
14
.115
36.
.41605;
72.
.226,
42
.251
: 41.50168
14,
204.
.175
177.21 ,
13.20681
88.
.932;
"56
.089
36.
.930464
23.
.224
14
.227
38.7401;
75.
477
42
.707
43.4172
15
212.
.275
180.12
15.1478'
91.
.825:
57
.228!
37.
.677103
23.96
14
.492
; 39.
.51586
77.
.672
43
.388'
44.13946
16'
185.
.903
191.
.249'
-2.87569
84.
.123
59
.379
29.
.414072
21.
.907'
14
.942
. 31
.79349:
71.
.636
44
.495
37.88738
17
208.
.892
'202.
.261
3.174368'
93.
692 :
60
.509 :
35.
.417111
24.
.436
15
.'197
. 37.
.80897
79.82
45
.195
43.37885
18"
202.
.821
202.
.734;
0.042895
91
.208:
61
.411
32.
.669283'
23.79
15
.391
; 35
.30475;
' 77.
.665'
45
.757
41.08414
19
165.
.335'
179.37
-8.48883
75.77:
60
.947
19.
.563152
19
.706;
' 15
.405
21.
.'82584;
64.54;
46
.197'
28.421 13
20 :
161.36
164.
.754
-2.10337
72.
.637'
59
.238
" 18.
.446522
18.
.905
15
.114
20.0529
61.
.758
45.66
26.06626
21
158.
.441
165.
.971
-4.75256
71.
.346:
59
.358 :
16.
.802624
18.
.571 '
15
.107
:' 18
.65274.
60.
.615'
45
.696'
: 24.61272
22'
183.
.288'
190.
.975
-4.19395
81.
.923;
61
.373
25.
.084531
21.
.351
15
.666'
'26.
.'62639;
69.
.698;
46.06'
' 33.91489
23
169.
.199
182.
.179
-7.67144'
77.
.232;
62
.189'
19.
.477678'
20.
.102
15
.788'
21.
.46055
65.
.799
46
.553
29.24968
24
169.
.'133
177.
.753
-5.09658
75.
.214";
"61
.979
17.
.596458
19.
.595
15
.808
'19.
.32636,
63.
.869;
46
.603'
27.03346
25
191.
.006
180.
.675
5.408731
82.
.829;
61
.938
25.
.221843
21.
.629;
15
.818
26.
.86671
70.
.218;
46
.627'
33.5968
26
177.96
180.
.756
-1.571 14
79.
.051 '
62
.759
20.60948'
20.
.584;
15
.921
22.
.65352;
67.
.088;
"" 47
.466
: 29.24815
27
161.
.095
185.
.522'
-15.1631
74.
.957;
63
.982
" 14.
.641728
19.
.489;
16
.138
' 17.
19431
63.
.996 ;
"" 48
.022'
: 24.96094
28
154.
.299
174.
.384;
-13.0169
71.
385:
64
.022
10.
.314492
18.
.556
16
.285
' 12.
.23863;
60.84
48
.448'
' 20.36818
29
'230.
.611'
218.
'196:
5.383525'
101.
.209;
65
.114
35.
.663824
26.
.446
16
.386
33
.03978;
86.
.068
48
.372
: 43.79793
30
216.
.034
199.39
'7.704343'
94.
.797.
66
.031'
30.
.344842'
24.
.742'
16
.589
32
.95207:
80.
.415;
49
.264'
38.7378
31 ,
217.
.209
204.
.263
5.960158
95.
316;
67
.522
29.
.159847
24.
.836
16
.913
' 31.
.90127:
80.
.771 '
49
.928
; 38.18573
32 <
152.
.515
185.
.308
-21.5015
72.
.493:
68
.527,
5.
.470873
1:
3.81
17
.336'
; 7.836257:
61.
.914
50
.919
17.7585
B-35
-------
CO
I
On
250
200
150
E
Q.
Q.
100
MEK inlet
MEK outlet
Butyl acetate inlet
Butyl acetate outlet
Ethylbenzene inlet
Ethylbenzene outlet
Xylene inlet
- Xylene outlet
10 15 20 25 30 35 40 45 50
Sample #
Figure B-18.
Two Parallel AMT Modules, Test 21, 4.0 ft3/min Air, Maximum Oil Flow, 450 ppm VOCs
-------
Flat Sheet Biofilm Experiments
B-37
-------
Table B-19. FS1: Degradation of MEK and Toluene by an M1 Biofilm: Aq/Aq Operation
Time
MEK Concentration
- PPm (v)
Toluene Concentration
- PPm (v)
Days
Feed In
Feed Out
Film Out
Feed In
Feed Out
Film Out
MEK Feed
Toluene
Feed
2
23.19
9.26
3.05
25
0
9
41.56
18.97
3.96
50
0
10
43.16
18.79
3.56
50
0
18
45.03
19.96
3.47
50
0
21
44.41
30.79
2.44
50
0
28
16.15
8.29
1.58
7.71
3.23
0
25
25
29
22.13
18.08
1.6
7.98
6.13
0
25
25
30
18.08
11.35
1.6
6.49
3.94
0
25
25
-MEKfeed in
-MEKfeed out
-MEKfilm out
-toluene feed in
-toluene feed out
-toluene film out
> 30
E
Q.
— 25
15 20
Time (d)
Figure B-19. FS1: Degradation of MEK and Toluene by an M1 Biofilm: Aq/Aq Operation
B-38
-------
Table B-20. FS2: Degradation of MEK by an M1 Biofilm: Aq/Octanol Operation
Time MEK concentration -
PPm (v)
minutes Aq Octanol Concentrations represent
reservoir values
0
0
59.7
15
3
65.4
40
3.25
61.3
65
3.42
59.7
90
3.61
59.1
122
4.03
58.64
152
4.18
54.02
185
4.17
53.4
245
5.36
51.7
305
5.95
52
365
6.36
50
430
7.69
46.9
495
8.69
44
aqueous
octanol
0
100
200
300
400
500
600
lime (min)
Figure B-20. FS2: Degradation of MEK by an M1 Biofilm: Aq/Octanol Operation
B-39
-------
Table B-21. FS3: Degradation of Toluene and MEK by an M1/X1 Biofilm: Aq/Aq Operation
Time
hours
Concentrations
- PPm (v)
MEK feed
MEK film
Toluene feed
Toluene film
0
50
0
50
0
2
45.6
5.5
9.2
0.08
3
29.7
6.6
3.4
0.06
4
25.1
7.9
1.1
0.06
6
25.9
6.9
0.18
0.05
8
27.3
10.7
0.09
0.05
60
MEK feed
MEK film
50
toluene feed
toluene film
40
Q.
Q.
30
20
10
0
0
1
2
3
4
5
6
7
8
9
Time (h)
Figure B-21. FS3: Degradation of Toluene and MEK by an M1/X1 Biofilm: Aq/Aq Operation
B-40
-------
Table B-22. FS4: Degradation of M-Xylene, Toluene, and MEK by an M1/X1 Biofilm: Aq/Aq
Operation
Time Concentrations - ppm (v)
Hours
MEK Feed
MEK Feed
Tol Feed
Tol Feed
Xyl Feed
Xyl Feed
In
Out
In
Out
In
Out
20
27.8
24.1
26.3
20
20.6
14.9
24
27.9
25.4
35.2
15.3
27.4
11.6
47
24.9
26.1
21.2
19.4
16.7
14.9
50
28.4
23.4
17.4
13.5
17.4
13.5
67
26
22.2
19.5
15.1
15.2
11.5
71
33
34.9
20.8
14.9
16.6
11.6
toulene feed in
toluene feed out
Time (h)
Figure B-22. FS4: Degradation of M-Xylene, Toluene, and MEK by an M1/X1 Biofilm: Aq/Aq
Operation
B-41
-------
Table B-23. FS5: Degradation of Toluene and MEK by an M1/X1 Biofilm: Aq/Octanol
Operation
Time Concentrations -
ppm (v)
Hours MEK Feed MEK Film Toluene Feed Toluene Film
1 21.5 3.3 5.1 0
2 24.4 3.4 2 0
3 27.4 2.9 0 0
4 24.9 3 0 0
5 25.5 3.2 0 0
6 24.4 3.2 0 0
-MEK feed
MEK film
-toluene feed
-toluene film
0-
Time (h)
Figure B-23. FS5: Degradation of Toluene and MEK by an M1/X1 Biofilm: Aq/Octanol
Operation
B-42
-------
Table B-24. FS6: Degradation of p-xylene by an X1 Biofilm: Aq/Aq Operation
Time
/^-xylene conc.
- ppm (w)
Time
/^-xylene conc.
- ppm (w)
hours
Feed
Film
hours
Feed
Film
0
80
25.7
0
34.2
0
1
52
0.3
1
53.6
0
2
53
0.1
2
50.4
0.1
3.33
40
0.2
3
51
0
4.5
41.5
0.1
4
41.6
0.2
5.5
40.4
0.8
5.08
45
0
6.5
41.8
1.8
6
33.3
0.3
7.5
41.2
3
7
30.4
0.3
9.5
34.5
4
10
43.5
3.7
feed
film
60
>
E
Q.
Q.
y 40
o
O on
20
0
2
4
6
8
10
12
Time (h)
Figure B-24. FS6: Degradation of p-xylene by an X1 Biofilm: Aq/Aq Operation
B-43
-------
Table B-25. FS7: Degradation of m-xylene by an X1 Biofilm: Aq/Aq Operation
Hours
Feed
Film
Hours
Feed
Film
Hours
Feed
Film
0
60.2
0.2
0
141
0.1
21.3
62.3
0.2
1
64.9
0
1.4
78.9
0
23.1
37.5
0
2
51.8
0
2.1
79.7
0
24.1
47.4
0
3.25
48
0.2
3.1
70.1
0
25.4
41.2
0
4
34.4
0.2
4.1
53.2
1.1
26.1
42.5
0
5
46
0
5.1
61
0
27.1
45.6
0
6
35.6
0.2
6.1
62.7
0.8
28.7
44.4
0
29.4
40
0.2
30.1
40.2
0
70
60
1 40
Q.
Q.
6
c
o
O
20
feed
film
0 ft
0
1
2
3
4
5
6
7
Time (h)
Figure B-25. FS7: Degradation of m-x ylene by an X1 Biofilm: Aq/Aq Operation
B-44
-------
Table B-26. FS8: Degradation of m-xylene and p-xylene by an X1 Biofilm: Aq/Aq
Operation
initial concentration is 100 ppm each
concentration reported is total xylenes
Time
xylene conc.
- ppm (w)
hours
Feed
Film
0
203.2
0
1
78.7
0.2
2.5
42.7
0
3.7
49.7
0.2
4.5
50.2
0.2
20.5
35
0.1
21.5
22.5
0
22.5
23.6
0
23.5
25.5
0
24.6
23.3
0
25.6
24.3
0
27.5
22.3
0
250
200
Feed
Film
Q.
Q.
= 100
0
5
10
15
20
25
30
Time (h)
Figure B-26. FS8: Degradation of m-x ylene and p-xylene by an X1 Biofilm: Aq/Aq Operation
B-45
-------
Growth Study Experiments
B-46
-------
Table B-27. GS1: Growth of X1 on m-xylene
Initial conc. of w-xylene: 250 ppm (v) Initial conc. of w-xylene: 100 ppm (v)
Time OD6OOOD6OO Time
hours 250-1 250-11 hours
OD600 OD600
250-III250-IV
Time OD6OOOD6OO Time
hours 100-1 100-11 hours
OD6OOOD6OO
100-III100-IV
0
0.006
0.011
0
0
0
0
0
0
0
0
0
2
0.005
0.008
4
0
0
3
0
0
4
0
0
3.5
0.016
0.016
5.33
0.002
0
6
0.011
0.041
5.33
0.002
0.003
4.67
0.036
0.039
6
0.002
0
7
0.04
0.101
6
0.007
0.007
5.5
0.062
0.062
6.5
0.002
0.002
7.25
0.052
0.125
7
0.013
0.015
6
0.074
0.08
7
0.007
0.007
7.5
0.06
0.13
8
0.032
0.036
7
0.127
0.156
8
0.016
0.01
7.75
0.073
0.152
9
0.068
0.069
7.75
0.201
0.235
9
0.034
0.025
8
0.083
0.178
10
0.131
0.124
8
0.228
0.262
10
0.062
0.046
8.25
0.104
0.21
10.25
0.161
0.146
8.25
0.266
0.297
11
0.115
0.081
8.5
0.124
0.22
10.5
0.19
0.176
8.5
0.283
0.333
11.5
0.16
0.109
8.83
0.149
0.223
10.75
0.223
0.203
8.75
0.32
0.354
12
0.212
0.15
9
0.174
0.225
11
0.247
0.21
9
0.351
0.378
12.5
0.258
0.19
9.33
0.2
11.25
0.25
0.219
9.25
0.367
0.399
13.08
0.323
0.259
9.5
0.203
11.5
0.25
0.219
9.5
0.388
0.405
13.5
0.361
0.295
9.75
0.397
0.408
10
0.398
0.416
0.45
0.4
0.35
0.3
§ 0.25
u>
O 0.2
0.15
0.05
0
2
4
6
8
10
12
Time (h)
Figure B-27. GS1: Growth of X1 on /n-xylene
B-47
-------
Table B-28. GS2: X1 on Toluene
Time
OD600
tol
hours
ppm
0
0
80.8
10
0.007
11.5
0.012
12.5
0.016
13.5
0.022
60.6
14.5
0.034
48.6
15.5
0.053
16.5
0.063
39.8
17.5
0.111
18.5
0.127
17.4
19.5
0.209
20.5
0.225
0
21.5
0.222
0.25
0.2
0.15
0.05
0
0 5 10 15 20 25
Time (h)
Figure B-28. GS2: X1 on Toluene
B-48
-------
Table B-29. GS3: X1 on Toluene and /n-xylene
Time OD600 tol m-x yl
hours
ppm (w) ppm (\
0
0.001
44.7
44.7
2
0.001
4
0.009
5
0.009
45.6
45.1
6
0.018
6.5
0.032
57.6
57.6
7.5
0.059
36.6
38.5
8.75
0.128
11.7
16.7
9.5
0.234
9.3
10.67
0.28
0.15
0.14
11.75
0.267
0.15
0.12
12.75
0.267
0.02
GS2: X1 degradation of toluene
° 40
10 15
Time (h)
GS3: Growth of X1 on toluene and m-
xylene
GS3: Degradation of toluene and
m-xylene by X1
~.15 -
Time (h)
70
to uene
m-xylene
c 40
O 20 -
Time (h)
Figure B-29. GS3: X1 on Toluene and /n-xylene
B-49
-------
Table B-30. GS4: M1 on Toluene
Time OD600 Time OD600 tol
hours hours ppm (w)
0
0
0
0.001
77.6
12.5
0.002
21.5
0.023
14
0.003
24
0.043
70.2
15.5
0.003
25
0.048
17.5
0.005
26.5
0.063
61.8
19
0.008
27.5
0.069
20.5
0.01
30
0.106
39
24
0.031
31
0.133
25
0.038
32
0.17
16.9
26
0.046
33.42
0.227
27
0.071
34
0.238
0
28
0.078
35
0.232
29
0.088
35.5
0.229
30
0.113
30.5
0.118
31
0.13
31.42
0.139
40
0.3
GS4: Growth of M1 on toluene
10 20 30 40
Time (h)
GS4: Degradation of toluene
by M1
o 40
10 20 30
Time (h)
40
Figure B-30. GS4: M1 on Toluene
B-50
-------
Table B-31. GS5: M1 on MEK and Toluene
Time
hours
OD600 tol
MEK
Time
hours
OD600 tol
MEK
0
0
24.9
46.3
0
0
17.01
179.4
8.25
0.012
8
0.002
13.2
195.2
9.25
0.013
24.7
42.9
10.5
0.002
10.25
0.015
12.5
0.005
9.6
175.7
11.25
0.021
13.5
0.008
12.25
0.025
18.6
34.6
14.75
0.013
7.2
186.6
13.25
0.03
15.75
0.018
14.25
0.033
12.8
27
16.75
0.024
2.3
171
15.25
0.046
17.75
0.034
16.25
0.063
7.27
11.9
18.75
0.046
0.1
159
17.25
0.085
19.75
0.07
18.25
0.141
0.155
6.4
20.75
0.079
0
124.7
19.25
0.167
21.75
0.096
20.25
0.171
0
6.6
22.75
0.099
0.2
101
21.25
0.171
23.75
0.134
22.25
0.165
24.75
0.174
0.1
43.4
25.75
0.189
26.75
0.248
0
0
GS5: Growth of M1 on MEK and
toluene
10 20
Time (h)
30
250
GS5: Degradation of MEK and toluene
by M1
E 150
Time (h)
Figure B-31. GS5: M1 on MEK and Toluene
B-51
-------
Table B-32. GS6: M1 on m-xylene and MEK
Time OD600 m-x yl MEK
hours ppm (w) ppm (w)
0
0.001
33.8 68.1
9
0.056
3.7 68.7
9.75
0.082
10.5
0.089
0 70.1
11.25
0.093
12.25
0.094
13.25
0.102
14
0.108
30
14.75
0.125
15.5
0.133
3.93
16.25
0.145
17
0.15
2.89
17.75
0.154
18.75
0.148
2.58
20.75
1.99
Growth of M1 on MEK
and m-xylene
10 15
Time (h)
20
25
80
70
60
£ 50
E
Q.
5 40
6
.9 30
O
20
10
0
: Degradation of MEK
and m-xylene by M1
¦ m-xylene
A MEK
A A A
10 15 20
Time (h)
25
Figure B-32. GS6: M1 on m-xylene and MEK
B-52
-------
Table B-33. GS7: M1 on Toluene and m-xylene
Time OD600 tol m-x yl
hours ppm (w) ppm (w)
0 0
8.75 0.005 32.6 35.2
9.75 0.008
10.5 0.011 32.4 31.3
11.25 0.015
12 0.024 28.8 23.1
12.75 0.032
13.5 0.044 28.4 14.5
14.25 0.065
15 0.089 27.4 0
15.75 0.109
16.5 0.12 19.2 0
17.5 0.125
18.5 0.132 0.4 0
20 0.154
21 0.151 0 0
Growth of M1 on toluene and
m-xylene
10 20
Time (h)
30
40
35
30
f 25
Q.
£ 20
ci
o 15
O
10
5
0
Degradation of toluene
and m-xylene by M1
A ¦ toluene
¦ 1 A m-xylene
~ ~
10 15 20 25
Time (h)
Figure B-33. GS7: M1 on Toluene and m-x ylene
B-53
-------
Table B-34. GS8: M1 on Toluene, m-xylene and MEK
Time
OD600
m-x yl
tol
MEK
hours
ppm (w)
ppm (w)
ppm (w)
0.25
0
24.5
20.8
19.9
8.33
0.023
27.8
19.2
7.3
9.33
0.038
10
0.051
30.4
14.8
1
11
0.068
11.75
0.071
26.7
12.75
0.076
13.5
0.076
31.7
3.6
0
14.5
0.076
15.5
0.079
16
0.082
17.1
0.1
17
0.097
18
0.125
19.9
0
19
0.129
20
0.128
15
21
0.135
22
0.146
23
0.155
24
0.159
25
0.163
26
0.163
!: Growth of M1 on toluene,
MEK, and m-xylene
O 0.08
10 20
Time (h)
8: Degradation of toluene, MEK, and m-
xylene by M1
35
30
25
I 20 «
Q.
Q.
2 15
o
o
10
5
0
m-xylene
t toluene
• MEK
10 15 20 25
Time (h)
Table B-34. GS8: M1 on Toluene, m-xylene and MEK
B-54
-------
Table B-35. GS9: M1 and X1 on MEK and Toluene
Time
OD600
MEK
tol
Time
OD600
MEK
tol
hours
ppm (v)
ppm (w)
hours
ppm (v)
ppm (w)
0
0.001
49.4
41.7
0
0.005
19.4
3
0.004
53.2
33.7
10
0.068
44.7
0
4
0.007
11
0.073
5
0.01
59.4
32.2
12
0.078
39.7
6
0.014
13
0.089
7.5
0.026
11.9
14
0.097
28.5
8.5
0.039
15
0.104
9.5
0.055
8.8
16
0.113
10.5
0.054
17
0.118
10.1
11.5
0.101
58.3
0
18
0.121
12.5
0.103
19
0.121
7.3
13.5
0.1
58.5
0
22
0.122
7.2
14.5
0.101
24.67
0.125
15
0.1
26.67
0.12
GS9: Growth of M1 and X1
on MEK and toluene
GS9: Degradation of MEK
and toluene by M1 and X1
10 20
Time (h)
30
~ toluene
Time (h)
Figure B-35. GS9: M1 and X1 on MEK and Toluene
B-55
-------
Table B-36. GS10: M1 and X1 on MEK, Toluene and m-xylene
Time OD600 MEK tol m-x yl
hours ppm (w) ppm (w) ppm (w)
0.42
0.001
26.5
18.6
14
3
0.001
4.5
0.004
5.5
0.006
27.7
16.6
10.7
6.5
0.013
7.5
0.023
13.2
7.1
8.58
0.044
9.5
0.056
2.7
0.1
10.5
0.088
11.5
0.089
0
0
12.5
0.095
13.5
0.092
32.1
14.5
0.095
15.5
0.102
19.1
16.5
0.11
17.5
0.119
7.5
18.5
0.118
19.5
0.123
6.6
20.5
0.121
26.17
0.135
5.5
GS10: Growth of M1 & X1 on
MEK, toluene, and m-xylene
<£ 0.08
10 20
Time (h)
30
GS10: Degradation of MEK, toluene, and
m-xylene by M1 & X1
35
30
25
| 20
Q.
Q.
y 15
o
O
10
¦ MEK
A toluene
• m-xylene
A
-•r
10 20
Time (h)
30
Figure B-36. GS10: M1 and X1 on MEK, Toluene and /n-xylene
B-56
-------
Table B-37. GS11: MX on p-xylene
Time
OD600
hours
0
0
7.1
0.002
22.4
0.011
23.6
0.012
25.2
0.02
26.2
0.024
27.2
0.05
28.3
0.052
29.1
0.062
30.2
0.069
31.1
0.094
31.6
0.101
46.2
0.128
GS11: Growth of MX on p-xylene
0.14
0.12
0.1
o 0 08
o
-------
Table B-38. GS12: M1 on p-xylene
Time
OD600
Time
OD600
p-x yl
Time
OD600
p-x yl
hours
hours
ppm(w)
hours
ppm(w)
0
0
0
0
57.8
0
0
122.9
22.3
0.082
10.8
0.064
32
9.4
0.007
113.4
23.5
0.105
12.2
0.091
20.1
33.4
0.008
61.5
25.2
0.129
12.8
0.114
35.4
0.021
26.3
0.136
13.5
0.13
4.8
36.4
0.03
35.3
27.2
0.157
14.2
0.141
0
37.4
0.054
18
28.2
0.151
14.8
0.152
38.4
0.098
9
28.9
0.145
15.5
0.155
39.4
0.114
0
16.6
0.159
40.4
0.148
18.4
0.161
41.4
0.161
21
0.175
42.4
0.177
43.9
0.182
44.6
0.18
45.1
0.182
B-58
-------
Degradation of p-xylene by X1
GS12: Growth of X1 on p-xylene
Time (h)
Time (h)
Figure B-38. GS12: M1 on p-xylene
-------
Table B-39. GS13: X1 on 150 ppm m-xylene with Ethyl Benzene
Ethyl benzene added at concentrations of
0, 50, 100 and 150 ppm
Time OD600 OD600 OD600 OD600
hours 0 ppm 50 ppm 100 ppm 150 ppm
0
0.005
0.004
0.004
0.006
4
0.014
0.021
0.018
0.011
7
0.028
0.02
0.019
0.024
9
0.033
0.028
0.028
0.029
11
0.033
0.036
0.044
0.043
13
0.053
0.057
0.069
0.068
14
0.058
0.064
0.092
0.076
15
0.068
0.087
0.119
0.117
15.5
0.071
0.093
0.134
0.14
16
0.087
0.105
0.154
0.162
16.5
0.102
0.142
0.19
0.181
17
0.124
0.153
0.176
0.172
17.5
0.124
0.161
0.166
0.162
18
0.165
0.172
0.178
0.173
18.5
0.184
0.177
0.16
0.163
19
0.21
0.167
0.167
0.158
19.5
0.218
0.199
0.179
0.158
20
0.197
0.21
0.184
0.16
21
0.317
0.224
0.207
0.174
22
0.198
0.238
0.215
0.164
23
0.224
0.249
0.229
0.171
24
0.258
0.267
0.227
0.171
25
0.243
0.252
0.238
0.166
26
0.243
0.264
0.238
0.171
32.5
0.228
0.282
0.241
0.186
B-60
-------
0.35
0.3
0.25
0.2
0.15
0.1
0.05
0
0 ppm EB
50 ppm EB
100 ppm EB
*—150 ppm EB
0
10
15 20
Time (h)
25
30
35
Figure B-39. GS13: X1 on 150 ppm /n-xylene with Ethyl Benzene
-------
Table B-40. GS14: X1 on a 50:50 Mixture of m- and p-xylene
150 ppm total
concentration
Time
OD600
Xylenes
hours
ppm(w)
0
0.001
61.5
1
0.007
105.9
2
0.013
112.1
3
0.025
104.9
3.5
0.035
94.3
4
0.04
96.1
5
0.083
75.5
6
0.14
51.4
6.5
0.176
25.8
7
0.228
4.6
7.5
0.251
0.1
8
0.257
0.2
8.5
0.268
0.4
9
0.251
0
10
0.228
GS14: Growth of X1 on
m- and p-xylene
GS14: Degradation of m- and
p-xylene by X1
<£ 0.15
Time (h)
120
100
E
Q.
Q.
O
O
Time (h)
Figure B-40. GS14: X1 on a 50:50 Mixture of m- and p-xylene
B-62
-------
Table B-41. GS15/16: M1 on Butyl Acetate (150 ppm) or a Mixture of MEK and Butyl
Acetate (75 ppm each)
Time
OD600
OD600
hours
BA
BA & MEK
0.25
0.006
0.011
1.3
0.018
0.019
2.25
0.028
0.023
3
0.048
0.05
3.7
0.072
0.077
4.3
0.088
0.09
5
0.094
0.102
5.7
0.116
0.151
6.15
0.129
0.19
6.8
0.141
0.212
7.25
0.156
0.236
8.25
0.197
0.276
9.25
0.217
0.206
10.25
0.219
0.239
11.25
0.265
0.209
12.25
0.28
0.199
13
0.266
0.205
0.3
0.25
0.2
o
o
to
Q
O
butyl acetate
butyl acetate and MEK
0.05
0
2
4
6
8
10
12
14
Time (h)
Figure B-41. GS15/16: M1 on Butyl Acetate
B-63
-------
Staged Bioreactor Experiments
B-64
-------
Table B-42. SB1: Degradation of MEK in a Staged Bioreactor - 500 ppm Case
Oil Oil C Aq Oil Oil Aq Aq Oil Oil Aq Aq Aq Aq
Time m-xyl MEK MEK MEK Time m-xyl MEK MEK MEK Time m-xyl MEK MEK MEK m-xyl m-xyl
hours ppm ppm ppm ppm hour ppm ppm ppm ppm hour ppm ppm ppm ppm ppm ppm
in out in out in out in out
0 458.7 * 0 449.8 413.6 350.6 426.1 0 1109.356.0 166.3 177.8 0.00 0.00
7
1 402.0 0 1.17 496.0 365.5 460.5 443.6 1 1055.166.1 215.4 198.9 0.00 0.15
7
2 375.4 0 2.17 463.9 336.1 467.3 446.9 2 1094.158.4 222.3 207.3 0.00 0.00
5
3 332.7 0 3.25 439.4 351.1 473.6 445.0 3 1088.139.2 213.0 197.3 0.00 0.00
5
4 288.2 0 4.33 446.1 351.9 463.1 431.8 4 1047.160.0 216.7 175.7 0.00 0.52
3
5 259.3 0 5.42 409.2 317.4 455.3 417.6 5 1075.158.7 221.8 192.4 0.00 0.00
3
6 215.5 0 9.1 1.6 10.25 378.8 288.6 376.1 353.5 6 1051.164.4 158.2 191.2 0.53 0.00
4
7 172.3 0 2.8 11.42 373.4 270.8 352.0 359.0 7.5 1045.147.4 165.3 190.1 0.66 0.65
3
8 121.6 0 9.1 4.3
-------
012345678
Time (h)
Figure B-42. SB1: Degradation of MEK in a Staged Bioreactor - 500 ppm Case
-------
Table B-43. SB2: Degradation of MEK and /n-xylene in a Staged Bioreactor -1000 ppm
Case
Oil
Oil
Aq
Aq
Oil
Oil
Aq
Aq
Aq
Aq
Time
w-xyl
MEK
MEK
MEK
Time
w-xyl
MEK
MEK
MEK
w-xyl
w-xyl
hours
ppm
ppm
ppm
ppm
hours
ppm
ppm
ppm
ppm
ppm
ppm
in
out
in
out
in
out
0
731.4
260.3
140.6
0
854.8
541.1
157.1
210.5
0
0.36
1.17
634.0
*
129.0
113.1
1
829.6
240.0
249.5
243.5
0
0.35
2.08
577.1
*
133.3
198.8
2
784.6
203.4
270.9
249.5
0
0.48
3.08
545.7
*
173.3
178.0
3
777.9
209.2
259.9
237.3
0
0
4.08
469.9
0
162.5
162.8
4
778.4
198.4
254.2
237.2
0.08
0.46
5.08
405.2
0
106.6
121.2
5
792.0
209.3
257.6
237.6
0.13
0.46
6.08
406.6
0
76.6
84.4
6
710.9
164.1
258.3
256.5
0.24
0.43
7.08
431.0
0
80.6
87.7
7
703.7
186.0
267.6
261.0
0.52
0.56
8
700.3
183.5
257.6
260.7
0.62
0.62
w-xylene not detected in aqueous phase
*Matrix interference/acetic acid
900.0
800.0
700.0
600.0
m-xylene - oil
MEK-oil
| 500.0
Q.
Q.
MEK-in
—^— MEK - out
£ 400.0
o
o
m-xylene - in
m-xylene - out
300.0
200.0
100.0
0.0 ft
0
1
2
3
4
5
6
7
8
9
Time (h)
Figure B-43. SB2: Degradation of MEK and m-xylene in a Staged Bioreactor -1000 ppm
Case
B-67
-------
Table B-44. SB3: Degradation of MEK and /n-xylene in a Staged Bioreactor-1500 ppm
Case
Oil
Oil
Aq
Aq
Aq
Aq
Oil
Oil Aq Aq
Aq
Aq
Time
w-xyl
MEK MEK MEK m-xyl
m-xy\
Time
w-xyl
MEK MEK MEK m-xyl
m-xy\
hours
ppm
ppm
ppm
ppm
ppm
ppm
hours
ppm
ppm ppm ppm
ppm
ppm
in
out
in
out
in out
in
out
0
1646.6
801.8
129.1
327.8
0.08
0.69
0
1166.0
408.9 163.9252.3
0.00
0.00
1
1574.2
261.4
351.8
354.7
0.24
0.59
1
1085.1
175.7312.4 329.8
0.00
0.25
1.83
1538.1
253.3
0.07
0.00
2
1073.2
179.4331.6270.1
0.00
0.00
2.83
1573.9
234.6
354.8
351.7
0.00
0.37
3
1071.0
156.7333.4259.8
0.00
0.54
3.83
1497.5
268.2
0.00
0.19
4
1060.1
167.5 301.3 299.8
0.00
0.51
4.83
1300.9
237.2
312.3
296.0
0.00
0.19
5
1030.5
164.9 296.2311.4
0.35
0.91
5.83
1275.7
98.1
0.00
0.23
6
993.4
152.2 294.0 302.7
0.39
0.49
6.83
1191.5
128.1
286.4
287.7
0.00
0.27
7
1065.4
148.3 288.8 273.0
1.05
1.02
7.83
1110.4
112.3
0.00
0.29
8.83
1208.9
220.0
253.6
258.8
0.00
0.39
1800.0
1600.0
1400.0
1200.0
1000.0
800.0
600.0
400.0
200.0
0.0
~ m-xylene - oil
AMEK- in
A m-xylene - in
S
¦ MEK-oil
X MEK - out
• m-xylene - out
X
10
Time (h)
Figure B-44. SB3: Degradation of MEK and m-xylene in a Staged Bioreactor-1500 ppm
Case
B-68
-------
Table B-45. SB4: Degradation of MEK and p-xylene in a Staged Bioreactor - 500 ppm Case
Oil
Oil
Aq
Aq
Aq
Aq
Time
p-x yl
MEK
MEK
MEK
p-x yl
p-x yl
hours
ppm
ppm
ppm
ppm
ppm
ppm
in
out
in
out
0
554.5
301.5
43.0
75.20
0.00
0.00
1
540.4
100.6
83.6
73.74
0.19
0.00
2
538.3
127.0
95.7
92.21
0.20
0.00
3
533.1
95.5
88.2
88.30
0.32
0.00
4
551.5
91.7
83.4
70.67
0.48
0.10
5
538.5
57.2
80.0
80.03
0.10
6
549.7
51.1
76.6
74.49
0.49
0.45
7.75
534.7
0.0
67.4
63.99
0.44
0.40
600.0
500.0
400.0
E
Q.
5 300.0
6
c
o
° 200.0
p-xylene - oil
MEK - in
MEK-oil
¦X— MEK - out
p-xylene - in
p-xylene - out
100.0
o.o »
o
1
2
3
4
5
6
7
8
9
Time (h)
Figure B-45. SB4: Degradation of MEK and p-xylene in a Staged Bioreactor - 500 ppm Case
B-69
-------
Table B-46. SB5: Degradation of MEK and p-xylene in a Staged Bioreactor-1000 ppm
Case
Oil
Oil
Aq
Aq
Aq
Aq
Time
p-x yl
MEK
MEK
MEK
p-x yl
p-x yl
hours
ppm
ppm
ppm
ppm
ppm
ppm
in
out
in
out
0
1212.6
605.2
29.30
88.54
0
0.13
1
1152.3
172.3
140.20
160.66
0
0.39
2
1174.8
143.3
132.15
147.19
0
0.00
3
1121.6
127.7
136.72
126.19
0.15
0.08
4
1125.2
78.8
116.76
130.93
0.69
0.46
5
1118.5
110.2
94.68
86.99
0.81
6
1141.7
162.9
96.00
96.74
0.95
0.69
1400.0
1200.0
1000.0
800.0
p-xylene - oil
MEK - oil
600.0
¦A—MEK - in
¦X— MEK - out
400.0
200.0
0.0
0
1
2
3
4
5
6
7
Time (h)
Figure B-46. SB5: Degradation of MEK and p-xylene in a Staged Bioreactor -1000 ppm
Case
B-70
-------
Table B-47. SB6: Degradation of m-xylene, p-xylene, Butyl Acetate and MEK in a Staged
Bioreactor
Oil
Oil
Oil
Aq
Aq
Aq
Aq
Aq
Aq
Time
xyl
MEK
BA
xyl
xyl
MEK
MEK
BA
BA
hour
ppm
ppm
ppm
ppm
ppm
ppm
ppm
ppm
ppm
in
out
in
out
in
out
0
1832.8
349.3
301.1
0
0
21.99
50.6
0.45
1.52
1
1761.7
135.9
0.0
0
0
62.17
66.4
0.88
1.11
2
1783.3
0.0
0.0
0.89
0.11
62.85
61.58
0.93
0.94
3
0.98
0.76
58.67
58.38
0.81
0.6
4
1717.1
0.0
0.0
1.08
0.56
51.99
58.04
0.62
0.47
5
1632.2
0.0
0.0
1.43
0.26
52.51
51.54
0.47
0.43
6
1714.3
0.0
0.0
1.51
1.58
48.91
50.76
0.22
0.16
7
1630.1
0.0
0.0
1.5
0.98
50.43
48.06
0.07
0
8
1440.9
0.0
0.0
24
0
0
3.67
2.77
0
0
5
E
Q.
Q.
6
c
o
O
~ xylenes - oil
¦ MEK-oil
A butyl acetate - oil
2000.0
1800.0
1600.0
1400.0
1200.0
1000.0
800.0
600.0
400.0
200.0
0.0
012345678
Time (h)
Figure B-47. SB6: Degradation of /n-xylene, p-xylene, Butyl Acetate and MEK in a Staged
Bioreactor
B-71
-------
Table B-48. SB7: Degradation of VOCs in Oil Generated from Spray Booth Tests
Containing BA, EB, m-xyl, o-xyl
Oil Oil Oil Oil Oil Aq Aq Aq Aq Aq Aq Aq Aq Aq Aq
Time MEK BA EB m-x yl o-xyl MEK MEK BA BA EB EB m-x yl m-x yl o-xyl o-xyl
hour ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm ppm
MEK
BA
EB m-xyl o-xyl
in
out
in
out
in
out
in
out
in
out
0
0.0
256.0 105.0 129.0 355.7
0
0
0
0
0
0
0
0
0
0
1
0.0
114.1
100.7 130.4 340.0
0
0.97
0
0
0
0
0
0
0
0.414
2.08
0.0
0.0
98.1 130.9 333.7
0
4.15
3.50
0
0
0
0
0.352
0
3.08
0.0
0.0
96.4 127.0 332.5
0.81
1.76
2.63
2.54
0
0
0
0
0.348
0.364
4
0.0
0.0
94.1 124.1 328.7
5
0.0
0.0
92.1 126.0 325.4
6.25
0.0
0.0
91.8 129.4 324.6
7
0.0
0.0
95.4 128.0 339.2
23.8
0.0
0.0
91.3 124.4 328.7
400.0
350.0
300.0
| 250.0 J
Q.
3 200.0
6
o 150.0
o >e
100.0 i*
BA
EB
m-xyl
o-xyl
50.0
0.0
0
5
10
15
20
25
Time (h)
Figure B-48. SB7: Degradation of VOCs in Oil Generated from Spray Booth Tests
Containing BA, EB, m-x yl, o-xyl
B-72
-------
Table B-49. SB8: Degradation of VOCs in Oil Run with Paint
SB8
CO
i
O
OJ
Oil
run
for 1
hour
with
paint
31.08.
00
Oil
Oil
Oil
Time
MEK
BA
EB
hour
ppm
ppm
ppm
MEK
BA
EB
0
0.0
0.0
35.9
1
0.0
0.0
35.2
2
0.0
0.0
34.5
3
0.0
0.0
34.4
4
0.0
0.0
34.2
5
0.0
0.0
34.9
6
0.0
0.0
33.7
24
0.0
0.0
33.9
25.75
0.0
0.0
33.6
27.75
0.0
0.0
33.3
Oil
m-x yl
ppm
m-x yl
Oil
o-xyl
ppm
o-xyl
186.5
174.7
171.2
167.6
164.2
165.5
160.4
153.9
143.1
142.5
18.4
38.1
28.7
13.7
38.2
40.9
14.8
35.8
34.4
32.0
Oil
run
for 2
hours
with
paint
05.09.
00
Oil
Oil
Oil
Oil
Oil
Time
MEK
BA
EB
m-x yl
o-xyl
hour
ppm
ppm
ppm
ppm
ppm
MEK
BA
EB
m-x yl
o-xyl
0
132.5
100.8
40.7
183.1
49.6
1
0.0
0.0
40.5
179.9
43.6
2
0.0
0.0
39.8
178.1
41.3
3
0.0
0.0
41.4
185.7
44.5
4
0.0
0.0
39.1
172.8
37.0
23
0.0
0.0
39.9
174.8
34.8
24
0.0
0.0
39.3
171.0
31.6
25
0.0
0.0
38.8
168.4
30.2
25.8
0.0
0.0
38.1
168.4
30.8
11.09.
00
Oil
Time MEK
hour ppm
MEK
Oil
BA
ppm
BA
Oil
run
for 2
hours
with
paint
Oil Oil
EB m-x yl
ppm ppm
EB m-x yl
Oil
o-xyl
ppm
o-xyl
0
131.4
84.1
39.6
158.8
44.9
1
0.0
0.0
39.6
158.4
39.9
2
0.0
0.0
38.7
153.8
34.7
3
0.0
0.0
38.8
156.1
35.2
4
0.0
0.0
39.2
157.2
36.5
5
0.0
0.0
39.2
158.5
34.6
76
0.0
0.0
37.9
312.1
27.0
-------
SB8: Degradation of VOCs in oil run for 1 hour w/paint
200.0
50.0
00.0
X— m-xyl
10
15
Time (h)
20
25
30
CO
i
^1
-1^
200.0
SB8: Degradation of VOCs in oil run for 2 hours w/paint
> 150.0
100.0
O 50.0
•MEK
¦BA
-A—EB
-X— m-xyl
10
15
Time (h)
20
¦ o-xyl
-A A A A
25
30
Figure B-49. SB8: Degradation of VOCs in Oil Run with Paint
-------
APPENDIX C
PILOT SCALE DATA
-------
Table C-1. AMT Modules; Pilot Test 1
Arrangement: Two Banks of Five Modules in Series
Oil: 5 PSIg Delivered, 5.0 Centistokes Viscosity, 3.8 L/min Oil Flow
Air: 44.5 ft3/min
8/29/00
me
Inlet PPM :
Time
Outlet PPM
3:29
PM
-0.88285:
: 3:31
PM
: 2.51195
3:33
PM
-1 2429
: 3:35
PM
2.5634;
3:37
PM:
-1 19145
: 3:39
PM
1.7404
3:41
PM
50 70783
: 3:43
PM
; 61.7152;
3:45
PM
311 5931
: 3:47
PM
: 118.4495:
3:49
pm!
336.0768:
; 3:51
PM
: 135.9379:
3:53
PM!
356.3427:
: 3:55
PM
; 151.3174:
3:57
PM:
324.8637:
; 3:59
PM
: 155.2265:
4:01
PM
284.8976;
; 4:03
PM
153.8892;
4:05
PM
312.0046!
: 4:07
PM
! 163.3021;
4:09
PM
296.4193:
= 4:11
PM
: 164.948:
4:13
PM
293.899:
; 4:15
PM
: 169.0629;
4:17
PM
287.8295
¦ 4:19
PM
; 170.4003;
4:21
PM
281.3485:
f 4:23
PM
: 171.7376!
4:25
PM
274.3017!
: 4:27
PM
: 169.8345:
4:29
PM
258.9737:
: 4:31
PM
; 171.3261;
4:33
PM
261.5455;
4:35
PM
: 173.6922;
4:37
PM
359.8918:
4:39
PM
191.5406
4:41
PM:
563.477!
! 4:43
PM
: 221.8881:
4:45
PM
598.5051:
; 4:47
PM
: 241.4339;
4:49
PM
568.6206;
•' 4:51
PM
; 249.3551!
4:53
PM
581.2225:
: 4:55
PM
259.3852
4:57
PM
570.3694:
: 4:59
PM
: 271.7813:
5:01
PM
548.7662;
! 5:03
PM
' 272.0899;
5:05
PM
554.2184!
: 5:07
PM
: 277.9023;
5:09
PM
574.793:
! 5:11
PM
: 208.8232;
5:13
PM
12.69633;
: 5:15
PM
; 29.51605:
5:17
PM
5.392425;
5:19
PM
: 41.24355
5:21
PM
951.5129;
: 5:23
PM
63.36118
5:25
PM
15.06245'
V 5:27
PM
; 201.5707:
5:29
PM
32.08785:
: 5:31
PM
; 193.2895;
5:33
PM
34.814
5:35
PM
; 179.5559;
5:37
pm!
35.7913
5:39
PM
168.0342
5:41
PM!
33.21948;
: 5:43
PM
: 159.3929;
5:45
PM!
33.52808;
! 5:47
PM
' 150.7515:
5:49
pm;
32.1393
5 5:51
PM
; 141.1844:
5:53
PM
31.47063
i" 5:55
PM
! 135.3207:
5:57
PM
31.93355
: 5:59
PM
: 126.1135:
6:01
PM!
29.51605:
: 6:03
PM
: 118.861:
6:05
PM
28.02438
: 6:07
PM
: 113.5631!
6:09
PM
26.78993;
; 6:11
PM
: 107.8537!
6:13
PM
18.35438:
: 6:15
PM
; 50.91358:
6:17
PM
; 11 87338
: 6:19
PM
; 21.2348
6:21
PM
9.301575
V 6:23
PM
14.34235
6:25
PM
8.272825
: 6:27
PM
: 11.5133;
6:29
PM
7.346975;
: 6:31
PM
10.536
C-1
-------
Concentrations 8/29/00
1000
800
600
400
200
0 -r ~ <
3:3C PM
4:30 PM
5:30 PM
6:30
-200
Time
Figure C-1. AMT Modules, Pilot Test 1, 44.5 ft3/min Air, 3.8 L/min Oil Flow
-------
Table C-2A. AMT Modules; Pilot Test 2
Arrangement: Two Banks of Five Modules in Series
Oil: 7 PSIg Delivered, 5.0 Centistokes Viscosity, 5.5 L/min Oil Flow
Air: 44.5 ft3/min
8/30/00 A
Time
Inlet PPM ;
Time
:Outlet PPM
12:05
pm;
-1 24288
< 12:03
PM
-0.00842
12:09
PM;
-1 29433
12:07
PM
-0.47138
12:13
PIVH
-1 29433
; 12:11
PM
: -0.26563
12:17
PM
-1 19145
: 12:15
PM
-0.21418
12:21
PM:
-1.19145;
; 12:19
PM
-0.21418
12:25
PM:
-1 5515
; 12:23
PM
: -0.47135
12:29
pm;
-1 44865
i 12:27
PM
; -0.31703
12:33
PM!
" -1 2429
12:31
PM
-0.31703
12:37
pm:
-1 2429
! 12:35
PM
-0.31705
12:41
PM:
-1 19145
: 12:39
PM
! -0.16275
12:45
PM:
-1 03713
: 12:43
PM
-0.11128
12:49
pm:
-1 14003
; 12:47
PM
-0.5228
12:53
pm;
-1 34575
:¦ 12:51
PM
-0.00843;
12:57
PM
-1 50005
12:55
PM
; -0.52278
1:01
PM
-1 5515
; 12:59
PM
-0.41993
1:05
PM
-1 24288
; 1:03
PM
-0.5228
1:09
PM
-0.98573:
: 1:07
PM
-0.21415
1:13
PM
-1.50008
: 1:11
PM
; -0.21415
1:17
PM
-1.34578:
i 1:15
PM
-0.16273
1:21
pm'
-1.55153
¦¦ 1:19
PM
-0.3685
1:25
PM
"" -1.5515
1:23
PM
0.043
1:29
PM:
-0.8314
j 1:27
PM
-0.16273
1:33
pm:
-1.39723
: 1:31
PM
-0.62565
1:37
PM:
-1.55148
; 1:35
PM
-0.21418
1:41
pm:
-1.0886
: 1:39
PM
; -0.05988
1:45
pm;
-1.3972;
< 1:43
PM
-0.26558
1:49
PM;
-1.50008
; 1:47
PM
: -0.2656
1:53
pm;
-1.3972
: 1:51
PM
-0.36845
1:57
PM:
-1.14003
: 1:55
PM
-0.0084
2:01
pm;
-1.34575
' 1:59
PM
-0.2656
2:05
pm;
-0.98573;
: 2:03
PM
-0.21418
2:09
pm;
-1.2429
' 2:07
PM
-0.47135
2:13
pm
-1.39723
: 2:11
PM
: -0.41995
2:17
PM;
-1.44863
V 2:15
PM
-0.5742;
2:21
pm:
-1.14
: 2:19
PM
-0.16273
2:25
PM:
-1.0886;
: 2:23
PM
-0.47135
2:29
PM;
-2.16875
: 2:27
PM
-0.6771
2:33
pm:
196.6328
' 2:31
PM
93.24573
2:37
PM:
172.4577
2:35
PM
189.8947
2:41
PM:
-2.06588
f 2:39
PM
-2.37453
2:45
pm;
28.17873
: 2:43
PM
196.1699
2:49
pm;
-3.5575
: 2:47
PM
2.15195
; 2:51
PM
-3.40318
C-3
-------
Concentrations 8/30/00 A
250
200
150
100
PM
2:31 PM
2:38 PM
2:45 PM
2:52 PM
3:00
PM
-50
Time
Figure C-2A. AMT Modules, Pilot Test 2, 44.5 ft3/min Air, 5.5 L/min Oil Flow
—«—Inlet
—¦—Outlet
-------
Table C-2B. AMT Modules; Pilot Test 2B
Arrangement: Two Banks of Five Modules in Series
Oil: 7 PSIg Delivered, 5.0 Centistokes Viscosity, 5.5 L/min Oil Flow
Air: 44.5 ft3/min
8/30/00 B :
Time Unlet PPM
5:29 PM -2.2202
5:34 PM 3.38635
5:39 PM 3.9522
5:44 PM 22.1092
5:49 PM 95.35458
5:54 PM 153.4262
5:59 PM 89.18223
6:04 PM 1001.612
6:09 PM 1001.715
6:14 PM 330.4188
6:19 PM 362.4637
6:24 PM 335.511
6:29 PM 281.8629
6:34 PM 257.8421
6:39 PM 229.9121
6:44 PM 657.9141
6:49 PM 678.2829
6:54 PM 643.2548
6:59 PM 194.061
7:04 PM 92.57703
7:09 PM 9.250125
7:14 PM 60.275
7:19 PM 45.97568
7:24 PM -2.2716
7:29 PM 15.57678
7:34 PM -2.16873
7:39 PM -1.91158
7:44 PM -3.24888
7:49 PM 6.88405
7:54 PM 2.923475
7:59 PM 13.51935
8:04 PM 12.49063
8:09 PM 12.64493
8:14 PM 11.30755
8:19 PM 11.41045
8:24 PM 10.48463
8:29 PM 9.6102
Time jOutlet PPM
5:32 PM 21.02905
5:37 PM 7.44985
5:42 PM 19.7946
5:47 PM 9.661625
5:52 PM 132.5945
5:57 PM 51.89088
6:02 PM 51.0679
6:07 PM 465.2335
6:12 PM 217.0531 i
6:17 PM 210.572
6:22 PM 205.7885
6:27 PM 195.3469
6:32 PM 181.1504
6:37 PM 171.5319
6:42 PM 266.8949
6:47 PM 302.9003
6:52 PM 315.3479
6:57 PM 328.9271
7:02 PM 95.61178
7:07 PM 53.74258
7:12 PM 40.2148
7:17 PM 929.0352
7:22 PM 1.226075
7:27 PM 196.1184
7:32 PM -1.44863
7:37 PM -2.37453
7:42 PM -2.58023
7:47 PM 195.0383
7:52 PM 4.517975
7:57 PM 13.41645
8:02 PM 12.69633
8:07 PM 12.28488
8:12 PM 11.4619
8:17 PM 11.2047
8:22 PM 10.69035
8:27 PM 10.12455
8:32 PM 9.97025
C-5
-------
Concentrations 8/30/00 B
1200
1000
800
600
400
200
5:3C
PM
5:58 PM
6:27 PM
6:56 PM
7:25 PM
7:54 PM
-200
Time
—~— Inlet
¦ Outlet
Figure C-2B, AMT Modules, Pilot Test 2A, 44.5 ft3/min Air, 5.5 L/min Oil Flow
-------
Table C-3. AMT Modules; Pilot Test 3
Arrangement: Two Banks of Five Modules in Series
Oil: 6 PSIg Delivered, 5.0 Centistokes Viscosity, 4.6 L/min Oil Flow
Air: 197 ft3/min
Pilot Test 3 - CEM Data,
Time Inlet PPM
3:59 PM 150.0315
4:04 PM 264.4259
4:09 PM 310.9244
4:14 PM 308.044
4:19 PM 34.45393
4:24 PM 21.85203
4:29 PM 29.36175
Time Outlet PPM
4:02 PM 86.55898
4:07 PM 148.0769
4:12 PM 163.5592
4:17 PM 169.5258
4:22 PM 33.57953
4:27 PM 28.07583
4:32 PM 1001.663
C-7
-------
1200
1000
800
600
2
Q.
Q.
400
200
4:0C
PM
4:07 PM
4:14 PM
4:21 PM
4:28 PM
-200
Time
Figure C-3. AMT Modules, Pilot Test 3, 197 ft3/min Air, 4.6 L/min Oil Flow
-Inlet
-Outlet
-------
Table C-4. AMT Modules; Pilot Test 4
Arrangement: Two Banks of Five Modules in Series
Oil: 6 PSIg Delivered, 5.0 Centistokes Viscosity, 4.4 L/min Oil Flow
Air: 100 ft3/min
Pilot Test 4 - CEM Data, 9/5/2000
ime
Inlet PPM ;
;Tim e
Outlet PPM
3
:34
PM
; 0.19735;
3:
:37
PM
f 0.763125;
3
:39
PM
: 0.557375'
3:
:42
PM
j 0.968875;
3
:44
PM
: -0.57423:
: 3:
:47
PM
0.81455;'
3
:49
PM
; 0.763125 ; '
! 3:
:52
PM'
•; -1.8601
3
:54
PM
; 221.1 165;
; 3:
:57
PM
; 220.7565
3
:59
PM'
: 221.3223;'
4
02
PM
-2.01443
4
:04
PM '
-1.963
; 4:
:07
PM
; 421.4097
4
:09
PM
; 1002.486;
¦ A:
: 12
PM
; 367.7616:
4
: 1 4
PM
221.4766;
; 4:
:17
PM'
-0.3685;
4
: 1 9
PM
j -2.5288:
; 4:
:22
PM
; 1001.149'
4
:24
PM'
I 219.1105;
"4
.21
PM'
220.9622
4
:29
PM
: 221.0136
: 4
:32
PM
•; 168.4971
4
:34
PM
; 0.608825 '
: 4
:37
PM
; 4.312225 :
4
:39
PM '
; -3.30035 :
4:
:42
PM
[ -2.9403;
4
:44
PM
; 196.2213;'
""j 4;
:47
PM
911.804
4
:49
PM
: 910.9296
• ; 4;
:52
PM
: 912.8841
4
:54
PM
; 912.5241 I
; 4:
.51
PM
;' 910.9295;
4
:59
PM'
j 911.9069;
5:
: 14
PM
• 908.8097;'
5
: 1 7
PM
: 385.4817
5
: 19
PM
; 3.919225
5
:22
PM
; 2.898575
: 5:
:24
PM'
r 3.306825;
5
:27
PM
~ 193.8074;'
i 5:
29
PM
; 18.66733
5
:32
PM
j -0.41845
"5:
34
PM
: 179.2634
5
:37
PM
; 3.000675;
i 5:
39
PM
; 4.5316
5
:42
PM
;" 3.51095 ;
5:
44
PM
3.255825
5
:47
PM
; 3.357875
V 5:
49
PM
•; 3.562025
5
:52
PM
; 3.45995
1 5:
:54
PM
; 3.459925
5
:57
PM
| 3.306825!
"1 5:
:59
PM
! 3.51 095:
6
:02
PM
; 3.459925; '
T 6:
04
PM
' 3.00065
6
:07
PM
2.337225
T 6:
09
PM
.. 2.082075.
6
: 1 2
PM
2.643425
6
: 14
PM
3.20475;
6
:17
PM
; 3.613025 :
! 6:
19
PM '
! ' 3.20475
6
:22
PM
3.562
6:
:24
PM
3.562';
6
:27
PM
3.61305
: 6:
29
PM
3.30685
6
:32
PM
; 3.408875;
I 6:
34
PM
; 4.225425
6
:37
PM
: 4.42955:
6:
39
PM
' 7.899675"
6
:42
"PM*
14.38068
' ; 6:
44
PM
: ' 40.1516;
6
:47
PM'
169.5675
' 6:
49
PM'
226.8758
6
:52
PM
;' 760.8183 ""
: 6:
:54
PM '
!' 175.283;
6
:57
PM '
I 465.5501 j
i 6:
:59
PM
: 358.2309
7
:02
PM
; 74.4958
; 7:
04
PM
336.1853
7
:07
PM
' 284.7966
... ^
o
CD
PM
; 215.9041
7
: 1 2
PM
: 200.8498;
} 1.
: 14
PM
; 151.6044;
7
: 1 7
PM
: 257.2906 .
, . ^
:19
PM
•' 1 17.9236,
7
:22
PM
; 157.7792!
1 7:
:24
PM
! 144.0007!
7
:27
PM'
I 19.38175 :
"' 7:
29
PM'
45.5099;
7
:32
PM'
990.1029
; 7:
34
PM
; 997.0432
7
:37
PM
; 68.32098
'7:
39
PM'
488.5653
7
:42
PM '
; 18.66733
T 7:
44
PM
; 47.4491
7
:47
PM
9.9409
: 7.
CD
PM
; 37.39588;
7
:52
PM
8.409975
[" 7:
:54
PM
I 31.27213;
7
:57
PM
; 6.368725 ;
; 7:
:59
PM
: 24.9952!
8
:02
PM'
; 6.6239
; 8:
04
PM
•; 34.12985;
8
:07
PM
; 995.257
I S:
09
PM
1 ' 167.22
8
: 1 2
PM'
50.91923
• 8:
: 14
PM'
^ 59.9518.
8
:17
PM
; 23.9746;
;' 8:
: 19
PM
; 79.95613;
8
:22
PM
i 59.59458
| "8:
:24
PM
: 70.51533
8
:27
PM
; 57.80848';
V ' 8:
29
PM
58.82913;
8
:32
PM
; ' 46.4795
! 8:
34
PM
; 46.37743;
8
:37
PM
^ 45.5609;
8
39
PM
44.02998;
8
:42
PM
39.3861
: 8:
44
PM
• '41.73355'"
8
:47
PM
| 43.92793j
! 8:
49
PM
I 41.58045
8
:52
PM
; 41.88668:
'- 8:
:54
PM
' 42.19285;
8
:57
PM'
j" 39.3861;
: 8:
:59
PM
•' 39.94743
9
:02
PM
41.42738
; "9:
04
PM'
38.87578
C-9
-------
Pilot Test 4 - CEM Data, 9/5/00
1200
1000
800
i 600
Q.
400
200
3:00 PM
4:00 PM
5:00 PM
6:00 PM
7:00 PM
8:00 PM
9:00 PM
Time
Figure C-4. AMT Modules, Pilot Test 4, 100 ft3/min Air, 4.4 L/min Oil Flow
—«—Inlet
—¦—Outlet
-------
Table C-5. AMT Modules; Pilot Test 5
Arrangement: Two Banks of Five Modules in Series
Oil: 5.5 PSIg Delivered, 5.0 Centistokes Viscosity, 3.85 LVmin Oil Flow
Air: 100 ft3/min
ilot Test 5
- CEM Data,
9/6/2000
ime
Inlet PPM i
Time
Outlet PPI
7:02
PM|
1 265575
j 6:59
PM
1 2656
7:07
PM
32 13963
T 7:04
PM
! 1 980025
7:12
PM
141 3981
f 7:09
PM
: 93 68363
7:17
PM
297 1972
J' 7:14
PM
! 237 4904
7:22
PM
273 0083
\ 7:19
PM
: 289 4404
7:27
PM
372 6218
5 7:24
PM
! 269 7423
7:32
PM
364 865
s 7:29
PM
: 399 2602
7:37
PM
52 24605
f 7:34
PM
! 216 6185
7:42
PM
307 7097
= 7:39
PM
: 116 8519
7:47
PM
322 9171
5 7:44
PM
! 315 4155
7:52
PM
65 15703
5 7:49
PM
: 31 1 9453
7:57
PM
994 7977
• 7:54
PM
! 32 70095
8:02
PM
96 03108
I 7:59
PM
: 275 9682
8:07
PM
24 74005
f 8:04
PM
! 57 50228
8:12
PM
16 9833
= 8:09
PM
19 8921
8:17
PM
14 2276
5 8:14
PM
! 14 48273
8:22
PM
12 6456
5 8:19
PM
: 13 00283
8:27
PM
3613025
• 8:24
PM
! 12 69663
8:32
PM
2 2862
I 8:29
PM
: 91 18308
8:37
PM
167 0669
f 8:34
PM
! 0 09185
8:42
PM
0 908375
= 8:39
PM
6 981 1
8:47
PM
179 9269
i 8:44
PM
! 246 625
8:52
PM
288 1647
5 8:49
PM
: 248 9215
8:57
PM
271 0181
• 8:54
PM
! 284 2862
9:02
PM
236 3167
I 8:59
PM
: 253 9736
C-ll
-------
Pilot Test 5 - CEM Data, 9/6/00
1200
1000
800
600
Q.
Q.
400
200
7:0C
PM
7:14 PM
7:28 PM
7:43 PM
7:57 PM
8:12 PM
8:26 PM
8:40 PM
8:55 PM
-200
Time
Figure C-5. AMT Modules, Pilot Test 5, 100 ft3/min Air, 3.85 L/min Oil Flow
-Inlet
- Outlet
-------
Table C-6AMT Modules; Pilot Test 6
Arrangement: Two Banks of Five Modules in Series
Oil: 3.5 PSIg Delivered, 5.0 Centistokes Viscosity, 1.8 L/min Oil Flow
Air: 100 ft3/min
Pilot Test 6 -
Time Inlet
PPM
3:29 PM 366.1821
3:34 PM 325.1082
3:39 PM 334.672
3:44 PM 320.6786
3:49 PM 295.3598
3:54 PM 276.4839
3:59 PM 33.56385
Data, 9/8/2000
Time Outlet
PPM
3:32 PM 361.1989
3:37 PM 254.8396
3:42 PM 247.4906
3:47 PM 247.6919
3:52 PM 229.9234
3:57 PM 89.58748
4:02 PM 50.67798
C-13
-------
1200
1000
800
600
2
Q.
Q.
400
200
3:3C
PM
3:37 PM
3:44 PM
3:51 PM
3:58 PM
-200
Time
~ Inlet
¦ Outlet
Figure C-6. AMT Modules, Pilot Test 6,100 ft3/min Air, 1.8 L/min Oil Flow
-------
Table C-7AMT Modules; Pilot Test 7
Arrangement: Two Banks of Five Modules in Series
Oil: 6.0 PSIg Delivered, 5.0 Centistokes Viscosity, 6.0 L/min Oil Flow
Air: 100 ft3/min
Pilot Test 7 - CEM Data, 9/12/00
Time Inlet
Time
Outlet
PPM
PPM
2:47 PM 225.8462
2:49 PM
183.0105
2:52 PM 234.5039
2:54 PM
187.9937
2:57 PM 229.118
2:59 PM
190.7119
3:02 PM 228.0106
3:04 PM
187.9938
3:07 PM 221.316
3:09 PM
144.4534
3:12 PM 28.42958
3:14 PM
47.70815
3:17 PM 13.32888
3:19 PM
38.4464
C-15
-------
1200
1000
800
600
2
Q.
Q.
400
200
2:4£
PM
2:51 PM
2:54 PM
2:57 PM
3:00 PM
3:03 PM
3:06 PM
3:09 PM
3:12 PM
3:14
PM
-200
Time
Figure C-7. AMT Modules, Pilot Test 7, 100 ft3/min Air, 6.0 L/min Oil Flow
—•— Inlet
¦ Outlet
-------
Table C-8AMT Modules; Pilot Test 8
Arrangement: Two Banks of Five Modules in Series
Oil: 6.0 PSIg Delivered, 5.0 Centistokes Viscosity, 4.5 L/min Oil Flow
Air: 136 ft3/min
Pilot Test 8 - CEM Data, 9/13/00
Time
Inlet
Time
Outlet
PPM
PPM
4:12 PM
-0.16108
4:09 PM
-0.1611
4:17 PM
100.8123
4:14 PM
0.090625
4:22 PM
275.9805
4:19 PM
298.984
4:27 PM
271.5007
4:24 PM
203.7992
4:32 PM
259.6215
4:29 PM
200.0743
4:37 PM
242.2556
4:34 PM
196.8025
4:42 PM
219.0509
4:39 PM
185.1246
4:47 PM
32.15445
4:44 PM
85.35928
4:52 PM
16.90273
4:49 PM
740.3273
4:57 PM
92.10428
4:54 PM
983.5494
5:02 PM
9.352375
4:59 PM
64.77195
C-17
-------
1200
1000
800
600
s
D.
D.
400
200
PM
4:22 PM
4:29 PM
4:36 PM
4:43 PM
-200
Time
Figure C-8. AMT Modules, Pilot Test 8, 136 ft3/min Air, 4.5 L/min Oil Flow
~ Inlet
¦ Outlet
------- |