United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EPA-456/R-9&003
May 1995
Air
Survey of Control
Technologies for Low
Concentration Organic
Vapor Gas Streams
control
technology center
-------
EPA-456/R-95-003
Survey of Control Technologies for Low
Concentration Organic Vapor Gas Streams
CONTROL TECHNOLOGY CENTER
V)
-, Sponsored by
Information Transfer and Program Integration Division
Office of Air Quality Planning and Standards
U.S. Environmental Protection Agency
Research Triangle Park, NC 2771 1
and
Air and Energy Engineering Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 2771 1
r'. 3. Environmental Prntertion Agency
ittpun 5, Library (PL-12.T
7 , Wtst Jackson Bc^ievoiJ, 12ih Floor
Chicago, IL 60604-3590
May 1995
-------
EPA-456/R-95-003
May 1995
Survey of Control Technologies for Low
Concentration Organic Vapor Gas Streams
Prepared by:
Robert A. Zerbonia
James J. Spivey
Sanjay K. Agarwal
Ashok S. Damle
Charles W. Sanford
Research Triangle Institute
Research Triangle Park, NC 27709
EPA Contract No. 68-D1-0118
Work Assignment No. 114
RTI Project No. 5846-114
Project Manager
Robert J. Blaszczak
Information Transfer and Program Integration Division
Office of Air Quality Planning and Standards
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Prepared for
Control Technology Center
U.S. Envionrmental Protection Agency
Research Triangle Park, NC 27711
-------
IV
-------
EPA REVIEW NOTICE
This report has been reviewed by the Control Technology Center (CTC)
established by the Office of Research and Development (ORD) and the Office of
Air Quality Planning and Standards (OAQPS) of the U.S. Environmental
Protection Agency (EPA), and has been approved for publication. Approval does
not signify that the comments necessarily reflect the view and policies of EPA,
nor does mention of trade names, organization names, or commercial products
constitute endorsement or recommendation for use.
This document is available to the public through the National Technical
Information Service, Springfield, Virginia, 22161, (800) 553-6847.
-------
VI
-------
PREFACE
The Control Technology Center was established by the U.S.
Environmental Protection Agency's (EPA's) Office of Research and Development
(ORD) to provide technical assistance to State and local air pollution agencies.
Several levels of assistance can be provided when appropriate. These include
the following:
• CTC HOTLINE provides quick access to EPA expertise, information,
and assistance on matters relating to control technology
(919/541-0800).
• Engineering Assistance Projects provide more in-depth assistance to
State and local agencies when needed to address a specific
pollution problem or source.
• Technical Guidance Projects address problems or source categories
of regional or national interest by developing technical guidance
documents, computer software, or presentation of workshops on
control technology issues.
• Federal Small Business Assistance Program (SBAP1 coordinates
efforts among EPA centers participating in the Federal Small
Business Assistance Program to assist State SBAPs.
• RACT/BACT/LAER Clearinghouse (RBLC1 bulletin board system
(BBS) provides access to more than 3,100 pollution prevention (P2)
and control technology determinations addressing over 200
pollutants. Select the RBLC from the technical BBS menu on the
OAQPS Technology Transfer Center (TTN) BBS (919/541-5742).
• CTC BBS on the OAQPS TTN provides around-the-clock access to
all CTC services, including downloadable copies of many CTC
products. Select CTC from the TTN BBS Technical BBS menu
(919/541-5742).
VII
-------
• CTC NEWS is a quarterly newsletter published by the CTC. It
contains updates on all CTC activities including the RBLC and
Federal SBAP. Call or write the CTC to get on the CTC NEWS
mailing list.
This document was funded by EPA's Control Technology Center and
prepared by Research Triangle Institute (RTI). This document is the result of a
request for technical assistance from the State and Territorial Air Pollution
Program Administrators and the Association of Local Air Pollution Control
Officials (STAPPA/ALAPCO) to identify control technologies that are effective
on treating gas streams with low concentrations of volatile organic compounds
(VOC) and/or organic hazardous air pollutants (HAP). This document presents
the results of a series of studies conducted to identify commercially available
control technologies applicable to low organic concentration gas streams.
Technical and economic background information relevant to the control
technologies is presented by technology type. Performance of the air pollution
control devices is documented in the form of source test reports or permits
issued by State or local air pollution control agencies. The document with the
information and data presented provides the basis for evaluating the availability
and efficacy of air pollution control devices in reducing organic emission in low
concentration, high flow rate gas streams.
VIII
-------
TABLE OF CONTENTS
Section Page
PREFACE vii
LIST OF TABLES xiii
LIST OF FIGURES xv
1 INTRODUCTION 1
1.1 BACKGROUND 1
1.2 SCOPE 3
1.3 CONCLUSIONS 5
1.4 RECOMMENDATIONS 8
2 INCINERATION 11
2.1 CATALYTIC INCINERATION 13
2.1.1 ARI's Fluid-Bed Catalytic Incinerator 16
2.1.1.1 Pilot Plant Tests 16
2.1.1.2 Wurtsmith Air Force Base 19
2.1.1.3 McClellan Air Force Base 21
2.1.2 Anguil 22
2.1.3 Monsanto Enviro-Chem 22
2.1.4 CSM 23
2.1.5 Amcec 23
2.1.6 Alzeta 23
2.1.7 Thermo Electron 26
2.1.8 Catalytica 26
2.1.9 Costs 26
2.2 REGENERATIVE THERMAL INCINERATION 28
2.2.1 Smith Engineering Systems 29
2.2.1.1 Source Test Data 29
2.2.1.2 Permit Conditions 34
2.2.2 Reeco 35
2.2.2.2 Source Test Data 37
2.2.2.3 Permit Conditions 37
2.2.3 Other Manufacturers 39
2.2.4 Costs 3.9
2.3 RECUPERATIVE HEAT RECOVERY 39
2.4 FLARES 39
2.5 BOILERS AND PROCESS HEATERS 44
IX
-------
TABLE OF CONTENTS (continued)
Section Page
3 ADSORPTION 45
3.1 NONREGENERABLE ADSORPTION SYSTEMS , . 46
3.1.1 Principle of Operation , 46
3.2 REGENERABLE FIXED BED ADSORPTION SYSTEMS 47
3.2.1 Principle of Operation 47
3.2.2 Applications 49
3.3 MODIFIED ADSORPTION SYSTEMS 51
3.3.1 Principle of Operation 51
3.3.2 Applications 51
3.3.3 Met-Pro KPR System 53
3.3.4 CADRE (Calgon, Inc.) 58
3.3.5 Catalytica 61
3.3.6 Munters Zeol . . , 62
3.3.7 Amcec . . 68
3.3.8 Dedert/Lurgi Cyclosorbon 68
3.3.9 HONEYDACS" System (Daikin Industries) 71
3.3.10 Diirr Industries System 73
3.3.10.1 Permit Conditions 77
3.3.10.2 Source Testing Data 81
3.3.11 Eisenmann System 82
3.3.12 Purus System 82
3.3.13 Kelco System 82
3.3.14 EC&C System 88
3.4 COSTS FOR ADSORPTION SYSTEMS 88
4 ABSORPTION , , 93
4.1 QVF GLASTECHNIK 94
4.1.1 Principle of Operation 94
4.1.2 Applications 96
4.1.3 Costs . , , ,,........ 96
4.2 QUAD . 97
4.2.1 Principle of Operation 97
4.2.2 Applications 100
4.2.3 Permit Conditions ,.,,,..,... 100
4.2.4 Costs 100
4.3 DAVIS PROCESS SYSTEM , 103
-------
TABLE OF CONTENTS (continued)
Section Page
5 OTHER COMMERCIAL TECHNOLOGIES 105
5.1 ULTROX D-TOX SYSTEM 105
5.1.1 Principle of Operation 105
5.1.2 Applications 108
5.2 ENHANCED CARBON ADSORPTION 111
5.2.1 Principle of Operation 111
5.2.2 Source Test Results 113
5.2.2.1 Northrop Corporation, Pico Rivera, California . 113
5.2.2.2 General Dynamics, Pomona, California 113
5.2.2.3 General Dynamics, Rancho Cucamonga,
California 114
5.2.3 Permit Conditions 116
5.2.3.1 Northrop Corporation, Pico Rivera, California . 116
5.2.3.2 General Dynamics, Pomona, California 116
5.2.3.3 General Dynamics, Rancho Cucamonga,
California 116
5.2.4 Costs 117
5.3 CONDENSATION 117
5.3.1 Liquid Nitrogen Systems 117
5.3.1.1 Airco Gases Systems 117
5.3.1.2 Edwards Engineering System 119
5.3.2 NUCON System 119
5.4 FLAMELESS THERMAL OXIDATION 119
5.4.1 Thermatrix System 122
5.4.2 Alzeta System 122
5.5 BIOFILTRATION 125
6 EMERGING TECHNOLOGIES 129
6.1 CORONA DISCHARGE PROCESSES 129
6.2 HETEROGENEOUS PHOTOCATALYSIS 132
7 REFERENCES 137
Appendix A - Organizations Contacted for Control Technologies for Gases
Containing less than 100 ppm Inlet 0V Concentration A-1
Appendix B - Cost Tables B-1
Appendix C - Summary Table of Control Devices Installed on High Flow,
Low Concentration Organic Vapor Streams in the U.S C-1
XI
-------
XII
-------
LIST OF TABLES
Table Page
2-1 Summary of Field Studies of Catalytic Oxidation for Control of Gases
with less than 100 ppm Inlet 0V Concentration 15
2-2 Feed Stream Composition (in ppm) Tested Using ARI System 18
2-3 Destruction Efficiencies for the Different Mixtures Using
ARI System 18
2-4 Effect of Inlet Concentration and Temperature on Destruction
Efficiency for ARI System 20
2-5 Catalytic Destruction Efficiency for Trichloroethylene Conducted at
Wurtsmith AFB using ARI's Fluidized-Bed Catalytic Incinerator ... 20
2-6 Summary of Wurtsmith AFB's Catalytic Oxidation Test Results for
ARI System 21
2-7 Fluidized-Bed Catalytic 0V Incineration Results of a Study Conducted
at McClellan AFB using ARI's Fluidized-Bed Catalytic Incinerator . 22
2-8 Catalytic Oxidation Costs 28
2-9 Summary of Field Studies of Regenerative Thermal Oxidation for
Control of Gases with less than 100 ppm Inlet 0V Concentration . 30
2-10 Source Test Results for the Smith RTO at Louisiana Pacific's
Hanceville, Alabama, OSB Plant 33
2-11 Source Test Results for the Smith RTO at Louisiana Pacific's Urania,
Louisiana, OSB Plant 33
2-12 Smith RTO System Test Results, Digital Equipment Corporation,
Cupertino 34
2-13 Smith RTO System Test Results, Mobil Chemical Company,
Bakersfield 35
2-14 Reeco Regenerative Thermal Incinerator Test Results at Sites in NJ
and CA 38
2-15 Cost Effectiveness for Reeco Regenerative Thermal Incineration .... 42
3-1 Summary of Field Studies of Nonregenerable Carbon Adsorption for
Gases Containing less than 100 ppm Inlet OV Concentration .... 48
3-2 Concentration of Inlet Gas at Verona Well Field Site 49
3-3 Modified Adsorption Systems 54
3-4 Field Data for MET-PRO KPR System 57
3-5 Summary of Field Studies of CADRE Adsorption/Incineration System
for Gases Containing less than 100 ppm Inlet OV Concentration .... 60
3-6 OV Abatement Systems Using Munters Hydrophobic Zeolites 65
3-7 Composition of Organic Solvents Versus Efficiency for the
HONEYDACS™ System 74
3-8 Results of Test of Durr Industries System 76
3-9 Comparative Operating Costs for Durr Systems 79
xiii
-------
LIST OF TABLES (continued)
Table Page
3-10 Diirr Industries Comparative Costs 80
3-11 Eisenmann Adsorption System Field Installations 85
3-12 Field Data for EcoBAC™ System 90
3-13 Application of the EC&C System by Industry Type and Materials
Treated 91
3-14 Cost Effectiveness of Adsorption Processes 92
4-1 Cost Effectiveness for QVF Absorption Process 97
4-2 Summary of Field Studies of QUAD System for Gases Containing
less than 100 ppm Inlet 0V Concentrations 101
4-3 Results of 0V Removal from Wastewater Plant Using the QUAD
System - Site 1 102
4-4 Results of 0V Removal from Wastewater Plant Using the QUAD
System - Site 2 102
4-5 Results of 0V Removal from a Compost Facility Using the QUAD
System 102
4-6 Results of 0V Removal from a Dewatering Facility Using the QUAD
System 103
4-7 ODOR/OV Emission Control Systems Costs 103
5-1 Summary of Field Studies of Other Commercial Technologies for
Gases Containing less than 100 ppm Inlet 0V Concentrations . . 106
5-2 Summary of Test Results on Ultrox D-TOX System Without Ozone . 109
5-3 Summary of Test Results on Ultrox D-TOX System with Ozone ... 110
5-4 Continuous Monitoring Results—Terr-Aqua Unit at Northrop
Corporation 114
5-5 Results of Terr-Aqua Enviro Systems' Air Pollution Control
Equipment at General Dynamics, Pomona Division 115
XIV
-------
LIST OF FIGURES
Figure Page
2-1 Schematic Diagram of a Catalytic Incineration System 14
2-2 Ari Fluid-Bed Catalytic Incinerator 17
2-3 Monsanto Enviro-Chem Dynacycle Regenerative Unsteady State
Catalytic Oxidizer 24
2-4 Alzeta Adiabatic Radiant Burner 25
2-5 Schematic of Smith Engineering Systems' Regenerative Thermal
Vapor Incinerator 31
2-6 Schematic of Reeco's Regenerative Thermal Incinerator 36
2-7 Reeco's Regenerative Thermal Incinerator—Horizontal Flow Design . . 40
2-8 Diirr Regenerative Thermal Incinerator—Horizontal Flow Design .... 41
2-9 Schematic of the Basic Components of a Steam-Assisted Elevated
Flare System 43
3-1 General Process Flow Diagram of an Adsorption Process for 0V
Recovery 50
3-2 Rotary Carousel System 52
3-3 KPR System Flow Chart 55
3-4 CADRE Adsorption-Regeneration Process 59
3-5 Munters Zeol System 63
3-6 Munters1 Hydrophobic Zeolite Showing Inlet Solvent Concentration
Versus Adsorption Capacity 64
3-7 Munters1 Hydrophobic Zeolite Showing Relative Humidity Versus
Adsorption Capacity 64
3-8 Amcec HYBRID Adsorption/Oxidizer Process 69
3-9 Supersorbon Solvent Recovery Plant 70
3-10 Dedert/Lurgi Cyclosorbon Modified Adsorption System 72
3-11 Schematic Diagram of a Diirr Industires Rotary Concentrator 74
3-12 Outlet Concentration Profiles for Carbon Honeycomb Blocks for
Solvent Mixture 75
3-13 Durr Industries Concentration Versus Flow Rate Application Chart .. 78
3-14 The Eisenmann Rotary Adsorber 83
3-15 The Eisenmann Rotary Adsorber Coupled With Condensation System
for Solvent Recovery 84
3-16 PADRE™ Schematic: Soil or Water Remediation Vapor Treatment
System 86
3-17 Kelco VAPOREX™ System 87
3-18 ECC EcoBAC™ System 89
4-1 QVF Process 95
4-2 Schematic Diagram of an Atomized Mist System - QUAD System . . 99
4-3 Typical TriplexT"Scrubber System Operational Flow Diagram 104
xv
-------
LIST OF FIGURES (continued)
Figure Page
5-1 Schematic Diagram of Pilot-Scale Ultrox D-TOX UV-Catalytic Test
System 107
5-2 Schematic of Terr-Aqua Enviro Systems' Air Pollution Control
System , 112
5-3 Vapor Recovery System: Refrigeration Liquid Nitrogen Sections ... 118
5-4 Refrigeration Vapor Recovery System: Components 120
5-5 The Braysorb® Process Regeneration Flow Diagram 121
5-6 Thermatrix System's Porous Surface Radiant Burner 123
5-7 Alzeta Adiabatic Radiant Burner 124
5-8 Alzeta VOC Flameless Thermal Oxidizer 126
5-9 Schematic of an Open Single-Bed Biofilter System 127
6-1 Schematic of a Packed-Bed Corona Reactor 130
6-2 Schematic of Pulsed Corona Reactor 131
6-3 Toluene Destruction Variac 44, 40 Pf, 1 L/min (48 ppm Toluene) . . 1 33
6-4 Effect of Increasing Voltage on Power Usage and Destruction
Efficiency for the Packed-Bed Corona System 134
6-5 Experimental Heterogeneous Photocatalysis System Used for
Experiments at NCSU 135
XVI
-------
SECTION 1
INTRODUCTION
1.1 BACKGROUND
A commonly applied approach to control of organic vapor emissions from
stationary or point sources is the application of add-on control devices. Several
different air pollution control technologies can be applied to sources of organic
air emissions (once they are covered, enclosed, or vented) to recover or destroy
the pollutants. In general, application of a particular technology depends more
on the emission (gas) stream under consideration than on the particular source
type. Selection of applicable control techniques for point-source organic
emission abatement is made for the most part on the basis of stream-specific
characteristics and the desired control efficiency. A key stream characteristic
that affects the applicability of a particular control technology is the
concentration of organics in the gas stream.
This document presents the results of a series of studies conducted to
identify commercially available control technologies suitable for application to
low organic concentration gas streams. Initially, OAQPS's Emmision Standards
Division conducted a study to survey and document the performance of control
technologeis applicable to gas streams containing low concentrations (i.e., less
than 100 ppm) of organic vapors* (see EPA-RTI contractor report, "Survey of
'The term "organic vapors" (0V) is used in this report to characterize the
broad range of organic compounds that might be found as constituents in a gas
stream that is vented, exhausted, or otherwise emitted to the atmosphere. The
term "0V" as used in this report would generally include those organic
constituent categories that are specifically defined and carry a precise statutory
or regulatory denotation (e.g., "volatile organic compounds" [VOC] as defined in
part 51 of Title 40 of the Code of Federal Regulations', "hazardous air
pollutants" [HAP] as identified in Title III, Section 112(b), of the Clean Air Act
Amendments of 1990; or "volatile organics" [VO] as measured by Method 25D,
59 FR 19402, April 22, 1994).
1
-------
Control Technologies for Low Concentration Gas Streams," Research Triangle
Institute (RTI) September 1993). The study also evaluated technical and
economic aspects of control systems specifically designed for low organic
concentrations. To the extent possible, results of source tests reflecting
operation of the control technologies at actual facilites were used as the primary
source of documentation of performance. However, at that time, there was
very little available information on field tests of technologies in use on low
organic concentrations gas streams. In many cases, manufacturers claims on
the effectiveness of the technologies were not supported by test data. As a
result, the actual test data included in the original report were limited. In this
first phase of the work, information and data were obtained from various offices
within the EPA including the Office of Air Quality Planning and Standards; the
Office of Research and Development, and the Office of Solid Waste; the EPA
Regional Offices; several State and local air pollution control agencies; and
numerous equipment manufacturers and vendors. Appendix A presents a list of
the organizations contacted regarding control technologies for gases containing
low organic concentrations.
As a continuation of the initial work in this area, the EPA's Control
Technology Center (CTC) supported a study to identify control technologies that
have been documented effective on low concentration/high volume flow
streams. The work was presented as an appendix to the original report, i.e.,
Identification of Permitted Control Technologies for Low Concentration Gas
Streams, as a continuation of EPA's work in this area and utilized the material
and knowledge gained in compiling the original report, (i.e., "Survey of Control
Technologies for Low Concentration Gas Streams," RTI, September 1993). The
object of the second phase of the study was is to identify permitted control
devices that have been installed and demonstrated to be effective for low
concentration organic vapor (0V) gas streams, particularly those with high air
flow rates. Low concentration is assumed to mean 100 ppm or below, although
-------
some control devices currently controlling higher OV concentrations are included
if they are feasible for lower concentration OV gas streams or are of particular
interest. High flow rates are assumed to be those above 100,000 cfm,
although some devices currently controlling lower flow OV gas streams are
included. The demonstration of control device performance is either in the form
of source test reports or permit conditions issued by State or local air pollution
control agencies. In this second phase of the study, information and data were
obtained primarily from two sources. First, equipment manufacturers and
vendors were contacted in order to identify locations where low
concentration/high flow rate devices have been installed and tested. Next, State
and local air pollution control agencies were contacted to request both permit
and source test information on these particular devices.
Permitted control devices are presumably associated with Federally
enforceable pollutant reductions, and include devices that are installed on
full-scale facilities rather than bench-scale applications. Devices installed
pursuant to a consent order prior to permitting are also included.
The performance of some of these air pollution control devices has been
documented through a compilation of source tests and those results are
summarized under the appropriate technology. Source tests in most cases were
conducted using reference or equivalent methods, and observed by a
representative of an air pollution control agency. Performance results obtained
by other test methods are also included in the final report, and such results are
noted; however, a rigorous evaluation of each testing protocol was not made as
a part of this study.
1.2 SCOPE
Although there are a number of control technologies in use for gases with
high organic concentrations, not all are applicable at low concentrations. There
are also other technologies which, in principle at least, can remove or destroy
-------
organic vapors from gas streams but are less cost effective at low
concentrations. For the purposes of this document, technologies such as
membranes and recuperative thermal oxidation systems fall into this category.
For example, recuperative thermal oxidation is very useful for control of
hydrocarbon gases at inlet concentrations of around 1,500 to 3,000 ppm
because the heat of combustion of these gases is sufficient to sustain the high
temperatures required without addition of expensive auxiliary fuel. At 100 ppm,
however, large amounts of auxiliary fuel are needed and recuperative thermal
oxidation, though in principle an effective control technique, generally is not
economically feasible. Biofiltration, though perhaps applicable to low organic
concentration gases, is also not within the scope of this study.1 A brief
description of the biofiltration process is included in Section 5.0 for background
information. The technologies that were evaluated for this document include
the following:
• incineration
- catalytic
- regenerative thermal
• adsorption
- nonregenerable
- modified regenerable (including adsorption/incineration)
• absorption
• other commercial technologies
- UV/ozone catalytic oxidation
- enhanced adsorption
• emerging technologies
- corona destruction
- heterogeneous photocatalysis.
The results of the studies are summarized and presented in this document
by technology beginning in Section 2.0. In general, the overall performance of
-------
the control technologies was found to be poorly documented relative to EPA
standards. For example, the lack of rigorous analyses (using test data) of the
inlet and outlet gas compositions in all but a few cases makes a comprehensive
evaluation of these technologies difficult.
The cost-effectiveness ($/ton 0V removed) of these technologies was
calculated using the model gas streams and general methods described by the
OAQPS Control Cost Manual.2 Though cost information for model streams was
requested from vendors of all technologies, responses were few. Therefore, to
compare the technologies on a common basis, capital and operating costs were
calculated using published values and vendor-supplied cost factors to the extent
possible on four model gas streams. These were 100 and 10 ppm benzene in
air and 100 and 10 ppm tetrachloroethylene in air, all at a flow rate of
10,000 scfm. Estimates of total annualized cost for the various technologies
based on the model gas streams are presented in Appendix B.
Because of the uncertainty associated with both the emission reduction
and the costs, the accuracy of the cost-effectiveness values in some cases is
probably no better than order of magnitude and thus conclusions should not be
drawn about relative cost effectiveness when differences are small, e.g.,
between regenerative thermal incineration and regenerable fixed-bed adsorption
for 100 ppm benzene. Overall, cost-effectiveness values range from $2,000 to
about $67,000 per ton of 0V removed (in 1991 dollars). These relatively high
values reflect the very dilute concentrations of interest. As expected,
cost-effectiveness values are much higher for lower concentrations.
1.3 CONCLUSIONS
1. The control of low concentration organic gas streams is currently one
of the most dynamic segments of the air pollution control technology
industry. The technologies as well as their applications are undergoing
rapid change and development. Since originally compiling the
-------
information and data for this document, performance data have become
more available and most recent indications are that the
cost-effectiveness of some of the technologies has improved. For this
reason, some of the data and information in the document may be
outdated.
2. Commercially available technologies exist for control of gas streams
containing less than 100 ppm 0V. Destruction and removal efficiencies
> 99 percent have been demonstrated at a number of sites for each of
the technologies discussed here. As expected, the lower the
concentration the higher the cost-effectiveness of the controls.
3. Based on the number of commercial installations, adsorption-based
processes are most widely applicable to low concentration gases. A
recent development by several vendors is the pairing of adsorption and
desorption steps that concentrate the 0V with a separate step to treat
the concentrated 0V. These systems are specifically designed for low
concentrations.
4. Concentrating adsorption systems (including but not limited to rotary
carbon or zeolite absorbers from Diirr and Munters Zeol) are
increasingly proposed by vendors in conjunction with incinerators (or
other devices) to control low concentration, high flow 0V streams.
These adsorption systems are more widely demonstrated in Europe,
probably because of more stringent regulations.
5. !n addition to adsorption-based processes, other technologies are being
used specifically for low concentration gases. These include
absorption/stripping process and UV/ozone catalytic oxidation.
Preliminary evaluation suggests that absorption may be competitive
with the more widely used adsorption-based processes at
concentrations close to 100 ppm. Insufficient cost information was
available to evaluate the cost effectiveness of UV/ozone technologies.
-------
6. Thermal or catalytic incinerator systems with regenerative heat
recovery are being proposed more widely by vendors. Regenerative
heat recovery is often more cost effective than recuperative heat
recovery for systems with flows above 50,000 scfm.3 A number of
combinations of these regenerative systems are available, although not
all are demonstrated at the concentration (and flow rate) examined in
this study. Pure thermal oxidizers without heat recovery were not
proposed for low concentration, high flow 0V streams by any vendor
contacted.
7. There is a trend for vendors to collaborate on proposal to provide
"best-of-breed" combinations of devices to make up a (case-specific)
control system. An example of this is a system proposed using a Durr
rotary concentrator, an Anguil recuperative incinerator, and a Johnson
Matthey catalyst. Numerous such systems are proposed and are
available with a performance warranty.
8. The development of these modified or hybred systems and devices is
proceeding at a rapid pace. These devices are generally installed on
new sources or existing sources affected by newly implemented
regulation, and so this rapid technological development appears to be
largely driven by the implementation of new and existing regulations.
9. Twenty-five (25) control devices for low concentration, high flow
OV gas streams are currently known to exist in the U.S. All are either
permitted, being permitted, or installed under a consent order.
Documentation in the form of permits and source test results was
requested for ail these devices. A table containing the relevant details
on these devices (e.g., inlet concentration, flow-rate, industrial
application, and location) is provided in Appendix C.
10. A need exists to more rigorously document the performance of field
systems. Reliable and complete inlet/outlet gas composition
-------
measurements taken at conditions of practical interest are scarce. Data
reported on many field tests are incomplete or inconsistent. In
addition, the acquisition of performance data on these devices was
hindered by the reluctance of some vendors to disclose the identity of
their clients, and the limited access to state and local air pollution
control agencies's files.
1.4 RECOMMENDATIONS
1. Further research on documenting the performance of control devices for
low concentraiton 0V streams with high flows should be conducted;
this would include continued gathering of information on field tests of
technology in use on low organic concentration gas streams, especially
those with high flow rates. Much of the information requested was not
received. Therefore, collection of additional permit information and
field test data will likely require commitment of resources, e.g., it may
be necessary to travel to various local, State, or regional air offices to
collect the information directly. Visits to a few State and local air
pollution control agencies may be the quickest and least expensive
method of gathering such data if detailed and/or extensive
documentation is desired.
2. Several of the technologies applicable to low concentration, high flow
streams now have a better defined cost history. The capital and annual
operating costs reported in the original study were, in large part, based
on either EPA estimates or vendor estimates because of the limited
number of these technologies in actual field applications in 1991. The
number of these devices in full scale operation has dramatically
increased over the past few years and many of these technologies now
have several years of operating history. Updated capital and operating
cost information could be obtained that would better reflect current
8
-------
actual costs. Additional cost effectiveness studies also could be
performed to determine which technologies are most cost-effective for
low 0V concentrations and high gas flow rates. This may involve
formally requesting vendors to develop quotes. Detailed costs were not
available for the modified regenerable adsorption systems. The fact
that they are being commercially used does, however, suggest that
they are of comparable cost to the conventional regenerable systems.
3. Consideration should be given to conducting field tests of some
demonstrated devices to better document performance at realistic
conditions and as a means of broadening the concept of availability (for
use during standards setting). The modified (or hybred) adsorption
systems and the alternative design, i.e., horizontal flow, regenerative
thermal oxidizers appear to be good candidates for performance testing.
4. There is an increasing number of technologies being applied to control
of indoor air pollution in large buildings (e.g., the ozone/catalyst system
developed by Union Carbide4'5). The very low concentration of indoor
air contaminants (typically around 1 ppm) and large flow rates in
buildings make these technologies of interest for study. However,
these technologies were not evaluated as a part of this study, though
they may be of particular interest for concentrations near 1 ppm OV.
-------
This page is intentionally left blank.
10
-------
SECTION 2
INCINERATION
Incineration is an oxidation process which ideally converts organic
compounds, whether hydrocarbon or oxygenated, to C02 and H20. If the
organic is halogenated, the corresponding halogen acids will be formed as
products of combustion. Incineration is widely used for the destruction of a
wide variety of 0V. It is best suited to applications where the gas stream has a
consistent flow rate and concentration. There are two main types of
incinerators: thermal and catalytic. In thermal incineration, the OV-containing
stream is heated to very high temperatures to oxidize the organic compounds in
the gas phase. In catalytic incineration, a catalyst promotes the oxidation
reaction on its surface (i.e., solid-gas interface) at lower temperatures by
providing alternative reaction pathways that have faster rates than the
corresponding gas-phase reactions. A thermal incinerator burns the 0V at very
high temperatures, usually in the 750 to 1,000 °C range; catalytic incinerators
operate between 350 and 500 °C.
To save fuel, a heat exchanger often is used to recover the valuable heat
generated during incineration by preheating the inlet gas. Thermal incinerators
without heat recovery are not known to be used to control high flow, low
concentration 0V gas streams. This is due to the high rate of fuel consumption
in pure thermal systems when compared to systems using heat recovery or
catalysis. Numerous companies such as the John Zink Co.6 (now including
McGill) market small thermal incinerators for the control of low concentration
OV gas streams, but they are typically for low flow applications such as small
air stripper outlets.
Depending upon the type of heat recovery unit, incinerators are further
classified as (1) regenerative or (2) recuperative. Thermal and catalytic oxidizers
are available with or without recuperative or regenerative heat recovery.
Regenerative thermal incinerators consist of a flame-based combustion chamber
11
-------
that connects two (or three) fixed beds containing inert (e.g., ceramic) packing.
Incoming gas enters one of the beds where it is preheated. The heated gas
flows into the combustion chamber, burns, and the hot flue gases flow through
the packed beds which capture, store, and permit recovery of the heat
generated during oxidation. The packed beds store the heat energy during one
cycle and then release it as the beds preheat the incoming OV-laden gas during
the second cycle. Up to 95 percent of the energy in the flue gas can be
recovered in this manner. The packed beds, in effect, are direct contact heat
exchangers.
A recuperative thermal incinerator uses a shell and tube heat exchanger to
transfer the heat generated by incineration to preheat the feed stream.
Recuperative thermal incineration has a much lower thermal efficiency and as a
result it is far less economical for low 0V concentrations. The lack of
recuperative thermal incinerators in high flow, low concentration 0V streams is
probably driven by the high operating costs for these systems. Recuperative
thermal incineration is not considered further in this document. The regenerative
thermal incinerator is better suited for low concentration 0V streams because its
higher thermal efficiency makes it more economical at low 0V concentrations7;
these systems are discussed in Section 2.2.
Catalytic incinerators modify the flame-based incinerator concept by adding
a catalyst to promote the oxidation reaction, allowing faster reaction and/or
reduced reaction temperature. This may allow more cost-effective operation at
low 0V concentrations than even regenerative thermal incineration. A faster
reaction requires a smaller vessel, thus reducing capital costs; and low operating
temperatures generally reduce auxiliary fuel requirements, thus reducing
operating costs. Catalytic incineration, however, is not as broadly applicable as
thermal incineration because of its greater sensitivity to pollutant characteristics
and process conditions. Design and operating considerations are therefore
critical because the catalyst may be adversely affected by high temperatures,
12
-------
high concentrations of organics, fouling from paniculate matter or polymers, and
deactivation by halogens or certain metals.
2.1 CATALYTIC INCINERATION
Figure 2-1 shows the schematic of a catalytic incinerator system.8 The
OV-containing gas is first indirectly preheated by the exhaust gas. For the low
concentrations of interest here, supplemental fuel is used to further preheat the
gas, usually in an open flame burner, to the reaction temperature. The gas then
passes over the catalyst, where the OV is oxidized. The operating temperature
to achieve a particular destruction efficiency depends on the concentration and
composition of the OV in the emission stream and the type of catalyst used.
Commercial catalysts usually consist of noble metals or metal oxides. The type
of catalyst used depends on the type of OV. For example, some noble metal
catalysts may be poisoned by chlorinated OV, even at the very low
concentrations of interest here. In such cases metal oxides that are more
resistant to halogenated compounds must be used.
Several companies market catalytic incinerators for OV destruction. These
companies include ARI Technologies, Wheelabrator, Hunington Energy Systems,
Anguil, Monsanto Enviro-Chem, CSM, Amcec, Alzeta, and Thermo Electron.
However, the only information obtained on systems used for concentrations less
than 100 ppm is for units made by ARI Technologies, Inc. (Palatine, IL). This
group of companies markets a full range of incinerators, including the fluidized
bed Econ-Abator* catalytic oxidizer. These systems are available with
recuperative heat exchangers for smaller flows, and regenerative heat
exchangers for larger flows. Other vendors of catalytic incinerators were
contacted, but none reported having field units operating on gases with OV
concentrations less than 100 ppm. Table 2-1 shows the available information
on catalytic incinerators used for OV concentrations less than 100 ppm. No ARI
systems are known to
13
-------
Emission Source
Supplementary
Fuel Catalytic Incinerator
Stack
Heat Exchanger
(Optional)
Figure 2-1. Schematic Diagram of a Catalytic Incinerator System
14
-------
Table 2-1. Summary of Field Studies of Catalytic Oxidation for Control of Gases
with less than 100 ppm Inlet OV Concentration
en
Vendor
ARI
ARI
ARI
ARI
Site
Mobile Unit8
Wurtsmith
AFB, Mlc
Wurtsmith
AFB, Mlc
McClellan
AFB, CA
Gas flow
(scfm)
500
1,200
1,200
348-691
Inlet
concentration
(ppm)
3-200
1-2
2.1
6-44"
Outlet
concentration
(ppm)
NRb
0.01
.072
.05-0.3
Destruction/
removal
efficiency (%)
72-98
98-99
96. 6"
>99
Comments
Six gas streams
tested
Major contaminant is
trichloroethylene
Different feed gases
from groundwater air
stripper
Paint spray booth
emissions
References
Palazzolo et
al., 19869
Hylton, 199010
Hylton, 199011
Ritts, et al.,
199012
"Pilot scale tests conducted in RTP, NC, using a mobile unit.
bNot reported.
cUnit was tested on several gases from an operating air stripper.
"These are results for a second test at this site using a different
plus one CFC.
"Calculated assuming an average molecular weight of 80. Feed
100 ppm CO was present in the exhaust.
feed which contained benzene, toluene, and three C, - C2 chlorocarbons
consisted of hydrocarbon and oxygenated hydrocarbon compounds. Up to
-------
be installed in the U.S. on high flow, low concentration organic vapor streams.13
(Note: The permit for the 3M Company's facility in St. Paul (No. 23GS-93-OT-1)14
lists an ARI, Econ-Abator fluidized bed catalytic oxidizer as in-use control
equipment. The unit is listed as having a 95% design construction efficiency and
an exhaust (stack) gas flow rate of 19,511 scfm (43,774 acfm at 680°F). This
unit serves several emission units at the plant that are reported in the permit as
Emission Point No. 2.
2.1.1 ARI's Fluid-Bed Catalytic Incinerator
ARI markets a fluid-bed catalytic incinerator for 0V oxidation. This system
uses a chromia-alumina catalyst suitable for oxidation of both hydrocarbons and
halogenated compounds. The catalyst is in the form of small beads through which
the gas passes in an upward direction. Figure 2-2 shows a schematic of the unit
which, in principle, is similar to other catalytic processes as shown in Figure 2-1,
the only difference being that the gas flows upward through a fluid bed of catalyst.
The gradual attrition of the catalyst is claimed to avoid catalyst deactivation by
continually exposing fresh catalyst surface. The purchased equipment costs of the
ARI units are somewhat higher than fixed-bed units of the same size,15 but the ARI
catalyst is one of the few commercially available that is designed to oxidize
chlorinated 0V.
2.1.1.1 Pilot Plant Tests
ARI's fluidized bed has been tested for the destruction of low concentration OV
streams.16 The particular system was designed to handle 500 scfm. Various feed
streams with inlet OV concentration ranging from 3 to 200 ppm were investigated
to determine overall destruction efficiency. Two types of feedstreams were
studied: one containing only chlorocarbons and the other containing a mixture of
hydrocarbons and trichloroethylene. Table 2-2 summarizes the composition of the
five streams used in this work.
Table 2-3 presents the destruction efficiencies for the different feedstreams
that are less than 100 ppm as a function of space velocity and inlet temperature.
16
-------
Process Gas
Inlet jfe
Catalyst Bed
Temperature Indicating
Controller
Blower
Preheat Burner
Figure 2-2. ARI Fluid-Bed Catalytic Incinerator
17
-------
Table 2-2. Feed Stream Composition (in ppm) Tested Usina ARI Svstem
Compounds
Trichloroethylene
1,2 Dichloroethylene
Vinyl chloride
1,2 Dichloroethane
1 , 1 ,2-Trichloroethane
Tetrachloroethylene
Benzene
Ethylbenzene
Pentane
Cyclohexane
Total concentration
Mixture
1 2 3
6.3 2.7 1.8
8.5
7.5
-
.
-
1.5
5.6
11.5
14.1
14.8 35.4 9.3
4
10
-
-
10
10
10
-
-
-
.
40
5
50
-
-
50
50
50
-
-
-
.
200
Table 2-3. Destruction Efficiencies for the Different Mixtures Using ARI System
Feed concentration Space velocity Temperature Destruction
Mixture' (ppm) (h1) (°F) efficiency (%)
1 13.1
16.7
19.2
2 39.1
40.8
33.3
3 12.4
9.92
10.7
4 50.2
57.1
42.2
10,300
10,100
6,840
10,800
10,500
7,790
11,300
10,500
7,350
10,600
10,200
7,430
648
792
794
654
807
653
648
947
656
653
953
649
86
93
95
93
98
96
89
98
92
72
96
75
alt is assumed that slight differences in the total feed concentration between these
mixtures and the compositions given in Table 2-2 simply reflect the slight differences in
various gases blended to make the mixture. Results for mixture 5 are not present because
the concentration is greater than 100 ppm.
18
-------
In general, the trend shows increasing destruction efficiency with increasing
temperature and decreasing space velqcity, as would be expected. The highest
destruction efficiency was observed for mixture 2, containing mostly hydrocarbons,
and the lowest efficiencies were observed for mixture 4, containing mostly
chlorocarbons. This reflects the generally more rapid oxidation of hydrocarbons
compared to chlorocarbons. The effect of feed concentration and temperature on
the destruction efficiency was also investigated in this work. Table 2-4 shows that
the destruction efficiency for this technology is independent of the inlet
concentration, i.e., similar destruction efficiencies are observed even when the feed
concentration is varied by a factor of 4 at both temperatures studied.
In addition to the total oxidation products, some incomplete oxidation products
were observed in the effluent stream. For example, oxidation of mixture 3
produced 1,1,1-trichloroethane (0.03 ppm) and tetrachloroethylene (trace to
0.01 ppm). Similarly oxidation of mixture 4 produced 0.09 to 0.73 ppm of
1,2-dichloroethylene. Formation of such partial oxidation products is possible in all
incineration processes, though such products are particularly hard to detect at the
fow inlet concentrations of interest here.
2.1.1.2 Wurtsmith Air Force Base
The Wurtsmith Air Force Base has been operating an ARI fluidized-bed catalytic
incinerator to treat contaminated air produced from groundwater air strippers.17
The incinerator, designed to treat 1,200 scfm, has been fully operational only since
October 1989. The major contaminant present in the feed gas stream is
trichloroethylene (TCE) at inlet concentrations of 1 to 2 ppm.
Table 2-5 summarizes the trichloroethylene destruction efficiency as a function
of temperature. Significant destruction of TCE was obtained in the preheater,
which is an open flame natural gas burner. The overall destruction efficiency (in
the preheater and in the incinerator) was greater than 98 percent at all
temperatures. Small quantities of some additional compounds including benzene,
19
-------
Table 2-4. Effect of Inlet Concentration and Temperature on
Destruction Efficiency for ARI System
Catalyst temperature
(°F)
653
660
953
953
Space velocity
-------
Table 2-6. Summary of Wurtsmith AFB's Catalytic Oxidation Test Results for
ARI System
Component
TCE
Benzene
Toluene
1,2-Dichloroethene
Trichlorofluoromethane
1,1-Trichloroethane
Chloroform
Feed
concentration
(ppm)
2.0100
0.0007
0.0647
0.0511
0.0008
0.0004
0.0006
Effluent
concentration
(ppm)
0.0511
0.0170
0.0024
0.0003
0.0014
0.0000
0.0001
Efficiency
(%)
97.5
-2172.7
96.3
99.5
-77.1
100.0
77.8
97 percent, it was negative for benzene and trichlorofluoromethane suggesting that
some of these compounds may be produced during oxidation. The total destruction
efficiency was 96.6 percent. The catalyst lost some activity with time on stream.
For example, the concentration of TCE in the stack increased from 0.012 ppm to
0.051 ppm in 5 months, corresponding to a decrease in TCE destruction efficiency
from 99.4 to 97.5 percent.
2.1.1.3 McClellan Air Force Base
A pilot plant test of fluidized-bed catalytic incineration was conducted at the
"Big Bertha" paint spray booth in Building 655 at McClellan Air Force Base,
California.18 Tests were conducted with varying inlet 0V concentration,
temperature, and total flow rate. Table 2-7 summarizes these results. The
concentration of OV has been reported in terms of Ib/h. Since the feed
concentration of individual OV was not reported, the concentration cannot be
presented in terms of ppm. However, assuming that the average molecular weight
of the OV was 80 g/gmol, the concentrations of the individual OV shown in
Table 2-7 vary from 6 to 44 ppm. The OV destruction efficiencies for all the tests
were greater than 99 percent. Of all the OV present in the feed stream, only
toluene was detected in the exhaust. The other compounds were completely
21
-------
Table 2-7. Fluidized-Bed Catalytic OV Incineration Results of a Study Conducted
at McClellan AFB using ARI's Fluidized-Bed Catalytic Incinerator
OV concentration (Ib/h)
In
0.13
0.11
0.26
0.1
0.055
0.28
Out
0.0013
0.00024
0.00024
0.00025
0.00042
0.0022
Destruction
efficiency (%)
99
>99.8
>99.9
>99.8
>99.2
99.2
Temperature
<°F)
698
950
950
1000
807
775
Flow
(scfm)
466
348
446
402
691
611
CO concentration in
the exhaust (ppm)
43
58
56
45
40
99
Fuel usage rate
(Btu/h)
370,000
434,000
490,000
525,000
548,000
498,000
oxidized. In addition to C02, produced by complete combustion, some amount of
CO was also present in the exhaust.
2.1.2 Anguil
Anguil manufactures various systems including oxidizing with regenerative
heat exchangers, recuperative heat exchangers, catalysts, and concentrators.
Projects are usually in the 100 to 35,000 scfm airstream flow rate range; however,
the company has recently expanded its product line to include equipment in the
100,000 scfm range due to market demands.19 Anguil has no system known to be
installed currently in the U.S. on high flow, low concentration organic vapor
streams.20
2.1.3 Monsanto Enviro-Chem
Monsanto Enviro-Chem manufactures a wide range of control devices
including the DynaCycle regenerative unsteady state catalytic oxidizer (RUSCO)
which has been demonstrated to provide 99 percent reduction of OV from oriented
strand board manufacture. RUSCO has been used for the removal of sulfur dioxide
with oxygen over vanadium, titanium, and tungsten oxides; oxidation of carbon
monoxide with air over copper, chromium, and iron oxides; destruction of C1f C4,
C6, and C9 alcohols; destruction of phenols, formaldehyde, hydrogen cyanide,
22
-------
acrylonitrile, ethyl acetate, cyclohexanone, and other compounds. It has not,
however, been demonstrated in the U.S. at high flow rates.21
RUSCO is a technology first demonstrated in Russia. It consists of a
three-layered fixed bed (see Figure 2-3). The center layer is the catalyst, and on
both ends are layers of inert ceramic material functioning as heat
absorbers/desorbers. When the temperature of the downstream layer reaches a
certain temperature, the flow is reversed, and the inert layers switch their heat
absorber/desorber function. The reaction is said to occur in a narrow zone in the
catalyst layer, which suggests that the rate of reaction is very high. Oxidation is
essentially complete for reversible and irreversible reactions, and efficiencies
measured below 100 percent appear to be due only to the short time required for
the switching valves to cycle. Unlike steady-state devices, temperatures in a
RUSCO device never approach the theoretical equilibrium, and so the RUSCO
device is self-optimizing.22
2.1.4 CSM
CSM manufactures mainly catalytic oxidizers, but has none installed in the
U.S. for control of low concentration, high flow 0V streams.23
2.1.5 Amcec
Amcec has no catalyst based control systems operating on low
concentration, high flow 0V streams in the U.S.24
2.1.6 Alzeta
Alzeta manufactures a broad line of air pollution control devices, including
the Alzeta Adiabatic Radiant Burner, which is an inward firing incinerator that is
f
reported to produce much less oxides of nitrogen compared to conventional burners
(see Figure 2-4). Alzeta markets this incinerator with a zeolite concentrator wheel
from Munters. However, no systems installed on low concentration, high flow 0V
streams in the U.S. are documented.25 These systems are described in more detail
in Section 5.4.2 of this report.
23
-------
i
Exit Valve
Plenum
Inert
bed
Catalyst
bed
Inert
bed
Plenum
Inlet Valve
Figure 2-3. Monsanto Enviro-Chem Dynacycle Regenerative Unsteady
State Catalytic Oxidizer
24
-------
Exhaust
VOC-laden
stream
Preheated
Bypass
Flow
Recuperator
Figure 2-4. Aizeta Adiabatic Radiant Burner
25
-------
2,1.7 Thermo Electron
Thermo Electron has no control systems operating on low concentration, high
flow OV streams in the U.S.26
2.1.8 Catalytica
Catalytica has no control systems operating on low concentration, high flow
OV streams in the U.S.27
2.1.9 Costs
Costs for catalytic oxidation units were developed using the methodology
given in the OAQPS Control Cost Manual.28 All costs presented here are calculated
from factored estimates and are given in detail in Appendix B. Though several
vendors were contacted about supplying costs, none responded.
Costs were developed for four cases:
• 100 ppm benzene,
• 10 ppm benzene,
• 100 ppm tetrachloroethylene, and
• 10 ppm tetrachloroethylene.
All these cases are for continuous streams; other stream conditions are:
• OV in clean air,
• 10,000 scfm,
• 70 percent relative humidity,
• 70 °F inlet temperature, and
• 95 percent destruction efficiency.
Assumption for the cost calculations are as follows:
• Catalyst replacement for streams containing benzene is required every
3 years and for streams containing tetrachloroethylene every 2 years,
• Operating temperature for the benzene-containing gas is 399 °C
(750 °F) and for the tetrachloroethylene-containing gas is 427 °C
(800 °F),
26
-------
• 95 percent overall destruction or removal efficiency,"
• 8,000 h/yr operation,
• 104 h"1 space velocity,
• 10,000 scfm (283,000 L/min), and
• 70 percent heat recovery for the unit.
All other costs (operating and supervisory labor, maintenance costs, and
indirects) are calculated as shown in the OAQPS Control Cost.29 Total capital
investment (TCI) is calculated based on purchased equipment costs (PEC) and
79 percent installation costs.30 PECs are a function of volumetric throughput
(scfm) and percent heat recovery and are given in cost curves presented in the
OAQPS Control Cost Manual.31
The total annualized costs (TAG, $/yr) and cost-effectiveness are shown in
Table 2-8. As expected, the TAG does not vary significantly with OV
concentration. This is because the OV concentration is too low to affect the usage
of auxiliary fuel, which is the single largest operating cost. Other costs contributing
to TAG depend almost entirely on the size of the unit, measured as volumetric
throughput (scfm), which is fixed for the sample case here. TAG for the
chlorinated OV is slightly higher than for the hydrocarbon benzene due to the lower
"The destruction/removal efficiency of 95 percent was selected to
represent the lower end of the range of control efficiencies required for the
control of organic vapors by EPA regulations. In many cases, EPA requires
higher control efficiencies especially in those situations where incineration is the
technology serving as the basis of the standard. The incineration-based
technolgies discussed in this document have demonstrated control efficiencies
of 98 percent or higher and therefore are applicable when a higher performance
standard (e.g., 98%) is required by regulation. Conducting the analysis at
95 percent as opposed to 98 percent also can impact the cost-effectiveness
calculation because in many cases the additional organics control can be
achieved at little or no cost. Cost-effectiveness values would therefore be lower
at this higher control efficiency.
27
-------
Table 2-8. Catalytic Oxidation Costs
Tetrachloroethylene
Benzene concentration concentration
100 ppm 10 ppm
Total annualized 271,900 274,400
costs,3 $/yr
Cost 5,500 55,400
effectiveness,3
$/ton OV removed
100 ppm 10 ppm
290,700 291,500
5,900 58,800
"All costs are in 1991 dollars rounded to nearest $100; calculations are based on
10,000 scfm flow rate. Other assumptions discussed in text.
heat of combustion and more frequent catalyst replacement. The
cost-effectiveness varies inversely with concentration, reflecting the assumption of
constant removal efficiency regardless of inlet concentration (see Table 2-4).
2.2 REGENERATIVE THERMAL INCINERATION
In regenerative thermal incineration, the contaminated air enters the system
through a heated (ceramic) packed bed which preheats the gas to near its final
oxidation temperature. The preheated air then enters a combustion chamber where
it is further heated to oxidize the OV. The hot clean (flue) gas exiting this chamber
passes through a second (ceramic) packed bed cooled in an earlier cycle. This bed
absorbs most of the heat; thus, cooling the gas before it is discharged to the
atmosphere. A third (ceramic) packed bed may simultaneously be purged of any
exhaust still contaminated with inlet OV emissions. This heat exchange cycle is
repeated, alternating between the three (ceramic) beds for heating, cooling, and
purging operation. Thermal energy recoveries as high as 95 percent can be
achieved with a regenerative thermal incinerator. The alternation between
chambers/beds typically results in somewhat lower destruction efficiencies than are
achieved in a conventional recuperative thermal vapor incinerator, generally below
99%.32 The lower destruction efficiency for regenerative thermal incinerators has
been attributed in part to valve leaks within the system.
28
-------
Smith Engineering Systems (Ontario, CA) and Reeco (Morris Plains, NJ)
market regenerative thermal systems and are the only vendors identified here
whose systems have been used on low concentration gases. Smith Engineering
provided test results from two sites in California used for the destruction of low
concentration 0V; other source test results and permit information were obtained
from State agencies. Similarly, test results from two sites for 0V oxidation using
Reeco system are presented although Reeco did not disclose the exact location of
these sites. Industry provided field results are summarized in Table 2-9.
2.2.1 Smith Engineering Systems
Smith Engineering Company (Smith, SEC) manufactures recuperative,
regenerative, and catalytic oxidizers. Figure 2-5 shows the schematic of a
regenerative thermal incinerator made by Smith Engineering Systems (Ontario, CA)
to destroy 0V from contaminated air.33 It consists of three ceramic packed beds
that are alternately heated and cooled during the heat exchange cycle. Smith
regenerative incinerators have been used at numerous sites for the destruction of
low concentration 0V emissions. They have seven regenerative oxidizers installed
jn the U.S. on high flow, low concentration organic vapor streams.34 All are at fiber
board manufacturing facilities owned by Louisiana Pacific. These devices were
installed pursuant to orders of consent, and will be permitted at some time in the
future.35
2.2.1.1 Source Test Data
Louisiana Pacific Corporation (LPC ) in Hanceville, Alabama has 3 Smith
regenerative thermal oxidation (RTO) systems at the facility. Two control
^emissions from 5 oriented strand board (OSB) Dryers and 1 unit controls emissions
from a process vent. These units were tested for air emissions in June 1994 by
Environmental Monitoring Laboratories, Inc. Testing was performed to determine
emissions of particulate matter, volatile organic compounds (VOC), formaldehyde,
nitrogen oxides, and carbon monoxide. Testing was simultaneously performed at
the RTO inlets and outlets in order to determine removal efficiency. The test report
29
-------
Table 2-9. Summary of Field Studies of Regenerative Thermal Oxidation for Control of Gases with less than 100 ppm
Inlet OV Concentration
Vendor
Smith
Engineering
Systems
Smith
Engineering
Systems
Reeco
CO
O
Reeco
Site
Digital Equipment
Cupertino, CA
Mobil Chemical
Company Bakersfield,
CA
Morris Plains, NJ
c
Gas flow
(scfm)
24,332
38,000
4,529
19,475
Inlet
concentration
(ppm)
63-80"
100
69
96
Outlet
concentration
(ppm)
0.2-1. 2'
1-2
0.9
0.9
Destruction/
removal efficiency
(%) Comments
99.8
98-99 Organic
contaminant
was isopentane
98.7 "
98.9
References
Smith
Engineering,
1 99036
Smith
Engineering,
1 99037
NETAC, 199138
Pennington,
199139
NETAC, 199140
Pennington,
199141
•As CH4
"OV include acetone, butyl acetate, ethyl acetate, toluene, and xylene.
'Given by the vendor only as "in California."
dOV include acetic acid, isophthalic, trimellitic anhydride, and tri-methylbenzene.
-------
Figure 2-5. Schematic of Smith Engineering Systems' Regenerative
Thermal Vapor Incinerator
31
-------
summary42 shows that overall removal/ destruction efficiency for the RTO system
serving the OSB dryers was 99.3 percent. The dryers' RTO inlet VOC loading was
reported as 263.6 pounds per hour (Ib/hr) and the outlet was measured at
1.25 Ib/hr at the West RTO unit and 0.62 at the East unit, for a total outlet loading
of 1.87 Ib/hr. The Press Vent RTO had an inlet loading of 147.4 Ib/hr with and
outlet loading of 0.34 Ib/hr. RTO outlet VOC concentrations (in ppm) were
reported in the testing summary; however, inlet VOC concentrations were not
contained in the summary information received from the State. Outlet
concentrations are provided in Table 2-10.
The Louisana Pacific Corporation conducted air emisison tests at the
LP Waferboard Plant in Two Harbors, Minnesota in February of 1989.43 This
facility is reported to utilize a regenerative thermal oxidizer (RTO) as a VOC control
device. A copy of the test results summary received from the Minnesota Pollution
Control Agency's Division of Air Quality44 indicated only the emission rate results in
terms of concentration and mass; the summary did not provide details on the
characteristics of the gas streams controlled by the RTO (e.g., VOC concentration
or gas stream flow rate).
Source testing of the Louisiana-Pacific Corporation's Urania, Louisiana OSB
Plant MDF Dryers' South RTO and North RTO units was performed by Armstrong
Environmental, Inc., in January 1994.45'46 Inlet sampling was done simultaneously
with outlet sampling in order to determine removal efficiency. A summary of the
available test results is presented in Table 2-11. The VOC control efficiency for the
individual RTO units was not reported in the limited information obtained from the
State Air Quality Division, but the overall control efficiency was calculated using
available information to be about 98.25%.
Table 2-12 shows the results for the tests of the Smith RTO system
conducted at Digital Equipment Corporation, Cupertino, CA.47 In all four tests, the
overall destruction efficiency was above 98 percent. The feed stream contained a
mixture of several 0V.
32
-------
Table 2-10. Source Test Results for the Smith RTO at Louisiana Pacific's
Hanceville, Alabama, OSB Plant48
DRYERS RTO
WEST RTO
EAST RTO
RTO INLET
REMOVAL EFFICIENCY
VOC as C
Ib/hr ppm
1.25 9.10
0.62 5.10
263.60
99.3
Ib/hr
0.12
0.13
5.52
HCOH
ppm
0.34
0.43
—
95.4
PRESS VENT RTO
RTO OUTLET
RTO INLET
REMOVAL EFFICIENCY
VOC as C
Ib/hr ppm
0.34 1.60
147.40
99.8
Ib/hr
0.03
1.81
HCOH
ppm
0.06
—
98.3
Table 2-11. Source Test Results for the Smith RTO at Louisiana Pacific's Urania,
Louisiana, OSB Plant.49 50
VOC Emissions (Ib/hr as C)
VOC Emissions (ppm as C)
HCOH Emissions (Ib/hr)
HCOH Emissions (ppm)
Volumetric Flowrate (acfm)
Volumetric Flowrate (dscfm)
Stack Temperature (°F)
Outlet
North RTO
0.22
5.4
0.41
1.19
101,615
72,697
224
Outlet
South RTO
0.08
0.5
3.28
1.23
111,751
81,914
217
Inlet*
Loading
17.10
176.7
6.11
7.84
192,381
166,065
118
8 Data were only available for the total inlet gas stream that is controlled by the
2 RTO systems from Smith Engineering.
33
-------
Table 2-12. Smith RTO System Test Results, Digital Equipment Corporation,
Cupertino
Test
No.
1
2
3
4
Inlet OV
(Ib/h)
4.83
4.22
3.80
4.28
OV concentration* (ppm)
Inlet
80
70
63
71
Outlet
1.3
0.2
0.9
0.8
Temperature
(°C)
816
816
816
816
Destruction
efficiency
(%)
98.1
99.8
98.4
98.8
aOrganic carbon as CH4
Flow = 24,332 scfm
The results of a Smith system at Mobil Chemical Company, Bakersfield, CA
are shown in Table 2-13.51 The feed stream contained 100 ppm of isopentane.
Again, 98 to 99 percent destruction efficiency was achieved at 816 °C. The
thermal efficiency of this system was 94.7 percent.
2.2.1.2 Permit Conditions
A review was conducted of the Air Permit issued to Louisiana Pacific
Corporation (Number 702-0027-X008, issued February 8, 1994), by the Alabama
Department of Environmental Management, Air Division, for the board press
system with regenerative thermal oxidation at LPC's Hanceville Plant.52 Item 21
in the permit states that the VOC emission rate shall exceed neither 4.74 Ibs/hr
and/or 0.087/lbs per thousand square feet of board, measured in accordance with
40 CFR Part 60, Appendix A, Method 18, 25, 25A or 25B. Item 22 states that
the VOC collection (destruction) efficiency across the RTO shall be at least
95 percent, item 23 states that the regenerative thermal oxidizer's combustion
chamber operating temperature shall not fall below 1400 degrees Fahrenheit.
Emission limits also are established in the permit for RTO formaldehyde,
34
-------
Table 2-13. Smith RTO System Test Results, Mobil Chemical Company,
§akersfield
OV Isopentane
Inlet concentration, ppm 100
Outlet concentration, ppm 1-2
Destruction efficiency, % 98 - 99
Temperature, °C 816
Flow rate, scfm 38,000
diphenylmethane diisocyanate (MDI), and phenol emissions. No limits are placed
on the RTO unit with regard to inlet or outlet VOC concentration.
The LPC's Hanceville Plant's Permit Number 702-0027-X014, issued
February 8, 1994, for the No. 1-5 Rotary Drum Wood Wafer Dryers with Two (2)
Regenerative Thermal Oxidation Systems has a number of permit conditions
specific to the RTO's and the organic gas streams controlled by these units.53
Item 24 states that the VOC emission rate shall exceed neither 24.89 pounds per
hour and/or 0.553 pounds per ton of dry wafers when up to three dryers are
operating (oxidizer exhausts to be sampled simultaneously). Item 26 states that
the VOC collection (destruction) efficiency across the multidone and RTO shall be
at least 95%. Item 27 has a requirement that neither regenerative thermal
oxidyers' combustion chamber operating temperature shall fall below 1400 °F.
The information and test results obtained from the Louisiana Air Control
Commission (in their submittal, dated 08/09/94) did not contain any information
or data relevant to the RTO units reported to be in operation at the
Louisiana-Pacific Plywood facility located in Urania, Louisiana.
2.2.2 Reeco
Reeco manufactures a full range of control devices, and may be most often
associated with regenerative oxidizers. Figure 2-6 shows a schematic of the
regenerative thermal incinerator made by Reeco (Morris Plains, NJ). It is similar in
35
-------
Figure 2-6. Schematic of Reeco's Regenerative Thermal Incinerator
36
-------
principle to that made by Smith. These regenerative thermal systems have been
used at various sites for destruction of low concentration OV gas streams.54'55
They have four regenerative oxidizers installed on high flow, low concentration
OV streams in the U.S.56
2.2.2.2 Source Test Data
The 3M St. Paul Tape Plant in St. Paul, Minnesota, contracted Precision
Environmental to test the Reeco regenerative thermal oxidizer destruction
efficiency in June 1993.57 The summary of the test results indicates that an
average control device efficiency of greater than 95% was achieved for a variety
of inlet VOC loading rates, ranging from a high of about 1,400 Ibs/hr to a low of
about 1 20 Ibs/hr. No inlet gas stream characteristics were included in the
available summary.
Table 2-14 shows the results of two additional Reeco systems operating on
gas streams containing less than 100 ppm OV. These units are installed in New
Jersey and California. Actual sites and customer names are considered
confidential by Reeco. The gas streams to both units contained a mixture of OV.
The destruction efficiency in both cases was above 95 percent. These units were
said to represent typical performance of Reeco systems with chamber flushing
and valve sealing features.
2.2.2.3 Permit Conditions
Air Emission Permit No. 23GS-93-OT-1, dated March 1993, for a pressure
sensitive tape and label manufacturing plant operated by 3M company in St. Paul
and issued by the Minnesota Pollution Control Agency's Air Quality Division,
contains a description of a Reeco regenerative thermal oxidizer in Section 1.3 of
the permit.58 The Reeco unit is listed as serving Emission Point Number 1 that
consists of a large number of ovens and dryers. The maximum inlet capacity of
the RTO unit is listed in the permit as 5,600 Ibs/hr of solvent; with a design
destruction efficency of 95%. The inlet gas stream characteristics are not
provided. The exhaust (stack) gas flow rate is reported as 270,000 scfm
37
-------
Table 2-14. Reeco Regenerative Thermal Incinerator Test Results at Sites in NJ
and CA
Unit *1 (in NJ) Unit #2 (in CA)
0V Acetone, butyl acetate, Acetic acid, Isophthalic,
ethyl acetate, toluene, and Trimellitic anhydride, and
xylene trimethylbenzene
Flow 4,529 scfm 19,475scfm
Inlet concentration
Outlet concentration
Destruction efficiency
69 ppmv
4.2 Ib/h
0.9 ppmv
0.05 Ib/h
98.7 %
96 ppmv
20.2 Ib/h
0.9 ppmv
0.22 Ib/h
98.9 %
(428,000 acfm at 380°F). The permit does not specify a unique VOC emission
limit for the sources served by the Reeco unit; the emission limit specified in the
permit is an aggregate value that covers numerous emission points at the facility.
This emission limit is formatted in terms of tons per year (i.e., 4,596 tons/yr). No
limits on the gas stream characteristics are contained in this permit for the Reeco
unit.
The engineering evaluation submitted as part of NUMMI's application for an
air permit (Application Number 3611, Plant Number 1438) from the Bay Area Air
Quality Management District contained, as part of the BACT Evaluation, a
discussion of recent New Source Review (NSR) Projects at other similar
facilities.59 Mentioned in the discussion was the Reeco regenerative thermal
oxidizer at the General Motors plant in Arlington, TX. The report states that
BACT for the first topcoat spray booth at this plant will be a Reeco RTO. The
required destruction efficiency of the RTO unit is 93%; no recirculation or solvent
concentrating equipment will be used. The RTO unit is reported to have a
capacity of 429,000 acfm and is an existing unit installed on the previous topcoat
spray booth to meet RACT requirements.
38
-------
2.2.3 Other Manufacturers
Regenerative thermal oxidizers, such as those made by Durr, Huntington,
and Eisenmann, are not discussed in any detail in this document even though this
technology has recently undergone considerable development. In principle, these
systems, which are similar in design to the regenerative thermal oxidizer systems
previous described, will oxidize low concentration gases; Somary (1993)60 claims
that the Eisenmann regenerative thermal oxidizer can be used for concentrations
as low as 100 ppm, although there are apparently no such field installations.
It is also of interest to note that an alternative design for regenerative
thermal oxidizers has recently become commercially available. This design
involves what is termed as horizontal flow and results in a much more compact
unit (see Figures 2-7 and 2-8) that is capable of handling small to moderate gas
flows. The main advantage offered by the alternative design is that the unit
requires less space and, as a result, is amenable to retrofit situations.
2.2.4 Costs
Costs were provided by Reeco for the model gas streams and are
summarized in Table 2-1 5. Details are given in Appendix B. No cost data were
obtained for the Smith Engineering System, Durr, Huntington, or Eisenmann
regenerative thermal oxidizing systems.
2.3 RECUPERATIVE HEAT RECOVERY
Recuperative heat recovery is offered by nearly all incinerator vendors, but
is generally not cost-effective compared to regenerative systems above
50,000 scfm.61 No currently documented control system for low concentration,
high flow 0V streams uses recuperative heat recovery.
2.4 FLARES
Flaring is an open combustion process in which the oxygen is supplied by
the air surrounding the flame. Flares are either operated at ground level (usually
39
-------
Figure 2-7 . Reeco's Regenerative Thermal Incinerator - Horizontal Flow Design
40
-------
Burner
Fuel
Combustion Air
Refractory
Lining
Regenerative
Heat Exchanger
Purge Air
Solvent-laden
Exhaust Air
Purified
Exhaust Air
Figure 2-8. Diirr Regenerative Thermal Incinerator - Horizontal Flow Design
41
-------
Table 2-15. Cost Effectiveness for Reeco Regenerative Thermal Incineration
Tetrachloroethylene
Benzene concentration concentration
Total annualized costs,8
$/yr
Cost effectiveness,3 $/ton
OV removed
100 ppm
181,000
3,700
10 ppm
181,000
38,100
1 00 ppm
245,900
5,000
10 ppm
252,300
50,900
"All costs are in 1991 dollars rounded to nearest $100; calculations are based on 10,000 scfm flow rate.
Based on Pennington (1991) for Reeco system. Details are given in Appendix B.
with enclosed multiple burner heads) or they are elevated. Elevated flares often
use steam injection to improve combustion by increasing mixing or turbulence and
pulling in additional combustion air. Properly operated flares can achieve
destruction efficiencies of at least 98 percent. Figure 2-9 is a schematic of the
basic components of a steam-assisted elevated flare system. The U.S. EPA has
developed regulations for the design and operation of flares to ensure that high
destruction efficiencies are achieved (40 CFR 60.18)62; design requirements
include specification of tip exit velocities for different types of flares and gas
stream heating values (i.e., greater than 7.45 MJ/scm [200 Btu/scf]). The flare is
a useful emission control device and can be used for most nonhalogenated
organic streams. However, low volumetric flows and low organic concentrations
are conditions that do not favor the use of flares. In the case of low organic
concentration gas streams, supplemental fuel costs generally eliminate flares as a
viable control alternative; flares have no heat recovery capability. In addition,
because flaring is an open combustion process, it is very difficult and
economically impracticable to directly measure emissions from a flare. No
currently documented control system for low concentration, high flow OV
streams in the U.S. uses a flare.
42
-------
Steam
nozzles
Pilot
burners
Helps prevent flashback
Gas collection header
and F
transfer line A
_£~J~lj~Ji
Knock-ou
drum
Steam
line
Ignition
device
Air line
Gas line
Drain
Figure 2-9. Schematic of the Basic Components of a Steam-Assisted
Elevated Flare System
43
-------
2.5 BOILERS AND PROCESS HEATERS
Boilers, process heaters, and other existing combustion devices can be
used as control devices to limit organic emissions by incorporating the vent
stream into the inlet fuel or injection through a separate burner, or by feeding the
stream into the boiler or process heater, etc., as combustion air. Where
applicable, use of existing combustion devices can achieve high destruction
efficiencies for organic emissions at a reasonable cost.63
The parameters that affect the thermal efficiency of a boiler or process
heater are the same parameters that affect the efficiency of these units when
they function as air pollution control devices. These are combustion temperature,
residence time, inlet organic concentration, compound type, and flow regime (i.e.,
mixing). A series of U.S. EPA-sponsored studies of organic vapor destruction
efficiencies for industrial boilers and process heaters have been conducted.64 The
results of these tests showed 98 to 99 percent overall destruction efficiencies for
0V; however, none of the tests involved low-concentration (i.e., less than
100 ppm) organic gas streams. No currently documented control system for low
concentration, high flow 0V streams in the U.S. using boilers or process heaters
was identified.
44
-------
SECTION 3
ADSORPTION
Adsorption is one of the most widely applied control technologies for
organic vapors (0V). In the adsorption process, organics are selectively
collected on the surface of a porous solid. Activated carbon is by far the
adsorbent most often used for low organic gas concentrations because of its
low cost and relative insensitivity to water vapor at relative humidities below
about 50 percent.65 Other common adsorption media include silica and
alumina-based adsorbates. In addition, recently developed hydrophobic
zeolites have been incorporated into systems which, in principle, are similar to
those based on carbon. The basic principles of adsorption for separation of
gas mixtures are described in a number of texts and are not discussed
here.66'67'68 Because adsorption processes simply separate the contaminant
(0V) from the gas stream, adsorption processes must be used in conjunction
with other unit operations to recover or destroy 0V.
The carbon adsorption capacity for organics is affected by the
concentration of organics in the gas stream. Carbon manufacturers generally
have equilibrium data for specific compounds and their specific carbon types.
For virtually any adsorbate, the adsorption capacity is enhanced by lower
operating temperatures and higher organic concentrations. As the
concentration of an organic constituent in the gas stream decreases, it
becomes more difficult to adsorb the constituent on activated carbon. In
theory, activated carbon can be tailor-made to remove pollutants at very low
organic concentrations. However, a carbon adsorption system designed to
achieve a 95 percent control efficiency for a given organic present at
1,000 ppm may not achieve a 95 percent control efficiency for the same
constituent present at a lower concentration, e.g., 10 ppm.
Carbon adsorbers are essentially constant outlet concentration devices;
prior to breakthrough, outlet concentration generally remains constant through
45
-------
an adsorption cycle even though the inlet concentration may vary more than
an order of magnitude.69 Inlet concentrations are typically limited by the
adsorption capacity of the carbon bed or by safety problems. The maximum
practical inlet concentration is usually about 10,000 ppm. Outlet
concentrations around 50 ppm can be routinely achieved with state-of-the-art
systems; concentrations as low as 5 to 10 ppm can be achieved with some
compounds.70 For organic concentrations above 100 ppm, carbon absorbers
can achieve control efficiencies of at least 95 percent, and control levels of
97 to 99 percent have been demonstrated in many applications. Theoretically,
fresh activated carbon should remove nearly all organics from an air stream
containing organics at concentrations of 100 ppm and less; but performance
data, which are quite limited, indicate that high removal efficiencies are not
attained in a significant number of cases handling these low organic
concentration.71'72'73 The reasons for the low removal efficiencies have not
been clearly established.
There are two types of adsorption systems that can be used for removal
of 0V from gas streams. These are nonregenerable (e.g., carbon canisters)
and regenerable (e.g., fixed bed systems).
3.1 NONREGENERABLE ADSORPTION SYSTEMS
3.1.1 Principle of Operation
These systems typically consist of one or two fixed beds of adsorbent
(e.g., granular carbon). The OV-containing gas flows upward through one bed.
The 0V is adsorbed over a period of time until breakthrough occurs. In
practice, the outlet gas stream is seldom monitored to determine this
breakthrough point, though regulatory compliance requirements are changing
this.74 In most cases, the bed is simply replaced on a time schedule
determined by calculating the bed life from the inlet concentration and the
working capacity of the carbon bed. Once breakthrough occurs, the carbon is
46
-------
returned for reactivation to the vendor or a central reactivation facility, at
which point other emissions may be generated during the reactivation process.
In principle, nonregenerable systems are simple and find many
applications in cases where the flow rate and/or OV concentration is low.
Table 3-1 summarizes the performance of nonregenerable adsorption system at
sites treating inlet gases containing less than 100 ppm OV. Nonregenerable
adsorption systems are especially attractive if, in addition, the OV is difficult to
desorb. These applications include odor control and control of indoor air. As a
guideline for their use, Stenzel and Bourdeau state that nonregenerable
systems are economically feasible when the carbon life is expected to exceed
3 months.75
An example of such a system is the Verona Well Field site, at which
two 314 ft3 carbon beds are used to control 5,350 cfm offgas from an air
stripper.76'77 The total concentration of the inlet gas to the carbon bed is
shown in Table 3-2. Assuming 5 wt% working capacity of the carbon before
breakthrough,78 it can be shown that this bed will last 257 days of continuous
use.
It is important to note that at concentrations approaching 100 ppm and
above, the capacity of nonregenerable systems may not be sufficient to remain
on line for a reasonable time. In the above example, for instance, a 100 ppm
inlet concentration would require replacement every 2.6 days. No currently
documented control system for low concentration, high flow OV streams in
the U.S. uses nonregenerable adsorption systems.
3.2 REGENERABLE FIXED BED ADSORPTION SYSTEMS
3.2.1 Principle of Operation
A fixed-bed regenerable system consists of two or more vessels, each
containing adsorbent. Figure 3-1 shows a general flow scheme.79 One vessel
is on line while a second is being regenerated, usually with low pressure
47
-------
Table 3-1. Summary of Field Studies of Nonregenerable Carbon Adsorption for Gases Containing less than
100 ppm Inlet OV Concentration
CO
Vendor
NR
NR
Cargonh
Various'
Calgon
NR
Site
Verona Well
Field; Battle
Creek, Ml
Verona Well
Field; Battle
Creek, Ml
Tyson's Dump
NR
Tyndall AFB,
FL
Verona Well
Field; Battle
Creek, Ml
Newark AFB,
OH
Gas flow
(scfm)
5,500
NR
170
175
NR
5,350
2,800
Inlet
concentration
(ppm)
0.48
0.18
20.2
80-95C
7-93
1.0'
4-292fl
Outlet
concentration
(ppm)
NR
NR
.017
15"
NR
ND
ND
Destruction/
removal
efficiency (%)
12.8
99.97
81-84
NR
>99
>99h
Comments
low removal due to
carbon being saturated
paint bake oven
emissions
Chlorinated C2
compounds and
aromatics were tested
C2 chlorinated
compounds from air
stripper
emissions were Freon
1 1 3 and traces of
trichloromethane
References
Byers, 198880
PEI, 1989 p. 61 B1
Vancil et al., 198782
Schuliger, 198383;
Urbanic and Lovett,
1 97484
Lubozynski et al..
1 98885
Stenzel and
SenGupta, 198586
Ayer and Wolbach,
1 99087
NR = not reported.
ND = not detected.
"Outlet concentration reported to be below detection limits; detection limits not given.
bCarbon regenerated by heated gas rather than steam.
cAs C6.
"Outlet consisted mostly of C, and C2 compounds.
"A number of different carbons were evaluated.
'Calculated from measured value of contaminants in groundwater to air stripper.
"Concentration varied due to cleaning schedule for operations whose emissions were vented to the adsorber. The time weighted average
concentration for the 27-hour test period was 126 ppm.
"Data reported for Carbon Adsorber 3 (Table 9, p. 27-28) for the period 0900 September 14 through 1200 September 15 during which (p. 41)
breakthrough did not occur. Detection limits for spectrophotometer used to measure outlet concentration is not given. A >99% ORE is based
on an assumed detection limit of 1 ppm and a time averaged inlet concentration of 126 ppm.
-------
Table 3-2. Concentration of Inlet Gas at Verona Well Field Site
Concentration in gas inlet to
Contaminant carbon bed, ppm
cis-1,2-dichloroethylene .55
1,1,1-trichloroethylene .17
tetrachloroethylene .10
trichloroethylene .08
1,1 -dichloroethane
total 1.0
steam, though hot inert gas can be used. In practice, three beds are
sometimes used; the third is dried while the second is being regenerated.
At the low concentrations of interest here, steam usage rates
(Ib steam/lb 0V) are somewhat higher than for higher concentrations of 0V
because the adsorbed organic is more difficult to desorb. SenGupta and
Schuliger give a steam usage rate of 5 to 20 pounds steam per pound 0V
recovered for concentrations below 100 ppm compared to 2 to
5 pounds/pound for 0V concentration around 500 ppm.88 The working
capacity of the carbon is somewhat lower at low concentrations also, roughly
2 to 10 wt% at 0V concentrations below a few hundred ppm compared to
5 to 1 5 wt% at higher concentrations.89
3.2.2 Applications
This type of regenerable system is limited to situations in which the 0V
either can be easily recovered by, for example, condensation of the steam/OV
mixture produced during the regeneration cycle (or cooling of the inert gas/OV
mixture) or can be disposed of at a minimal cost. In most cases this is not
true, and there exists a need to destroy the 0V after the regeneration cycle.90
This need, coupled with the development of novel carbon absorbers, has led to
the development of modified regenerable adsorption systems.
49
-------
Organic-Containing
Gas Stream
Gas
Conditioning
On-Line
Sorbent
Bed
-*-Clean Gas
[Organic]
^Sensor,
Regenerating
Gas
Off-Line
Sorbent
Bed
Organic + Regenerating
Gas
-^Organic (liquid)
-H3yproducts
Figure 3-1. General Process Flow Diagram of an Adsorption
Process for OV Recovery
50
-------
3.3 MODIFIED ADSORPTION SYSTEMS
A recent development in adsorption processes is systems designed
specifically for control of low concentration (i.e., less than 100 ppm) OV gas
streams that are based on an adsorber followed by treatment of the
concentrated OV in the regenerated gas. There are basically three treatment
options:
• Discharge. This method simply transfers the OV from the
workplace to outdoors; it is apparently used in some industrial
applications, primarily for odor control or worker exposure
reasons.
• Incineration. Either thermal or catalytic incinerators can be used
to oxidize the desorbed OV, which typically has as high as 10 to
1 5 times the OV concentration of the inlet gas (and also a
correspondingly lower flow rate).
• Recovery. The OV in the desorbed gas can be recovered by
condensation or other techniques.
3.3.1 Principle of Operation
The most promenant example of these modified adsorption based
systems is a rotary carousel system (Figure 3-2). In these systems, one sector
of the carousel is being used for adsorption while another sector is being
regenerated (or desorbed) with hot gas. As the carousel turns, any one
position alternately adsorbs OV from the gas and is then regenerated. There
are several variations of these rotarary carousel systems on the market,
differing primarily in the way the desorbed OV-containing gas is treated.
3.3.2 Applications
For OV concentrations of interest here, these modified adsorption
systems offer the advantage of essentially concentrating the OV from less
than 100 ppm in the vent gas to the range of 500 to 2,000 ppm in the
regeneration gas. Of course, the flow rate of the regeneration gas is
correspondingly lowered. This higher concentration/lower flow rate
regeneration gas can then be treated in a number of ways. One vendor states
51
-------
Clean Air
Hot Desorption Air
Solvent Laden Air
Desorbed Gas
Figure 3-2. Rotary Carousel System
52
-------
that their system is specifically designed for inlet concentrations in the 50 to
100 ppm range, at which neither nonregenerable nor regenerable fixed-bed
adsorption systems are entirely suitable.91
Vendors offer several treatment options for this concentrated
regeneration gas. These include thermal incineration, catalytic incineration,
and a fixed-bed regenerable carbon bed system. The choice among these
depends on the concentration and chemical nature of the OV. In general,
thermal incineration is best for higher concentration gases consisting of many
different OV constituents, above for example 1,500 ppm, and catalytic
incineration is best at the lower end of the concentrated regeneration gas
range, from 500 ppm up to about 1,500 ppm. A regenerable carbon bed is
best used when the OV is of value and can be recovered and reused on site,
which generally means single component gases.
Survey of the industry and the literature revealed 12 modified adsorption
systems commercially available in 1991/1992, 5 of which use a rotary
carousel adsorber with some type of downstream oxidation or recovery.
These are described individually in Table 3-3.
3.3.3 Met-Pro KPR System
The KPR system (Figure 3-3) consists of a rotary carousel, made of
microporous activated carbon fiber, designed to adsorb low concentration
organics. These organics are then desorbed at a concentration roughly 5 to
15 times greater than that of the inlet gas. This desorbed gas, containing the
concentrated contaminant, can be catalytically oxidized, or, in some
applications, fed to a second regenerable carbon adsorption system where it is
recovered. Catalytic oxidation is the method of choice for complex,
multicomponent OV mixtures, whereas a second carbon adsorption system is
best when the OV consists of only one or two compounds and has some value
when recovered. When a catalytic incinerator is used, the heat from the
53
-------
Table 3-3. Modified Adsorption Systems
Ul
Trade Name
KPR
CADRE
Zeol Rotor
Concentrators
£VCC
Hybrid
Cyclosorbon
Honeydacs
Durr
Eisenmann System
Padre
Vaporrex
EcoBac
Vendor
Met-Pro
Harleysville, PA
Calgon
Pittsburgh, PA
Munters Zeol
Amesburg, MA
Cataiytica
Mountain View, CA
Amcec
Oak Brook, IL
Dedert Corp.
Olympia Fields, IL
Daikin Industries
Durr Industiries, Inc.
Plymouth, Ml
Eisenmann, Inc.
Crystal Lake, IL
Purus, Inc.
San Jose, CA
Kelco Group, Inc.
Rayham, MA
EC&C Environmental
Tempe, AZ
Adsorbent
Rotary carbon carousel
Fixed carbon beds
Rotary hydrophobic zeolite
Noncarbon
Fixed carbon beds
Fixed beds, activated
carbon or zeolite
Rotary carbon carousel
Rotary carbon or Zeolite
carousels
Rotary carbon carousel
Fixed bed polymer sorbent
Faxed carbon beds
Fluidized carbon bed
Regeneration options
Catalytic oxidation
carbon adsorption
Thermal oxidation
Incineration or carbon
adsorption
Catalytic incineration
Thermal incineration
Thermal incineration
Catalytic incineration/
conventional recovery
Thermal or Catalytic
incinceration
Catalytic or thermal
incineration, or conventional
recovery
Condensation
Condensation
Condensation
State of
development
Commercially
available
Commercially
available
Commercially
available
Developmental
Commercially
available
Commercially
available
Commercially
available
Commercially
available
Commercially
available
Commercially
available
Commercially
available
Commercially
available
-------
Purified
Exhaust Air
Fan
Hot Air
(Desorption Air)
Solvent-laden
Exhaust Air
Control
Panel
Pre-filter
Rotor Drive
Rotor Unit
(Cylindrical)
Solvent-laden
Desorption Air
Figure 3-3. KPR System Flow Chart
55
-------
combustion is used to indirectly heat the air used for desorption of the rotary
carousel. This, in turn, keeps energy costs down.
Table 3-4 shows data reported by Met-Pro for the KPR system for gases
with inlet concentrations < 100 ppm. The KPR system has also been used to
control gases with higher inlet concentrations, but these data are not reported
here. Tests 5 and 6 are the only ones for which emissions from both the rotary
carousel adsorber and the catalytic incinerator are reported. If these results are
general, it suggests that the mass emission rate (Ib/h emitted) from the adsorber
is less than that of the catalytic incinerator, though it must be kept in mind that
the inlet gas concentration to the catalytic incinerator is 5 to 15 times higher than
to the adsorber. Thus, even if the adsorber and incinerator have comparable
removal efficiencies, higher mass emission rates from the incinerator would be
expected.
Met-Pro KPR has installed several high flow, low concentration control
systems world-wide. Most of the reported applications of the KPR system are for
control of paint spray booth emissions. They are typically used in aerospace
painting. Thirty systems are reported in use as of 1988, with sites in the United
States, Europe, and Japan. One new U.S. installation is at the LTV aircraft
manufacturing facility in Dallas, TX. This facility has been field tested.92 Sizes for
the KPR system range from roughly 10,000 to 100,000 scfm. In such
applications, care must be taken to filter the air upstream of the rotary carousel to
remove ordinary particulates and high molecular weight compounds. Both of
these will quickly deactivate the carbon fiber used in the rotary carousel.
Particulates are removed in two stages: one removing large particles (>5 //m),
and another using cloth filters to remove small particles (<5 ;/m). High molecular
weight organics (boiling point >400 °F), which will not readily desorb from
carbon, are captured in a thin carbon bed upstream of the carousel that is either
disposed or regenerated off-site.
56
-------
Table 3-4. Field Data for MET-PRO KPR System
CJI
Test
no.
1.
2.
3.
4.
5.
6.
7.
8.
Flow rate
(scfm)
1 8,400
1 8,400
11,300
9,700
56,100
86,260
105,000°
70,000°
Adsorber Inlet
concentration. Adsorber outlet
ppm concentration, ppm
53.08
78.11
59.94
72.8
5.8
24
24
5.8
NR
NR
NR
NR
0.3
0.9
NRd
NR
Incinerator outlet
concentration, ppm
1.63
1.43
ND
ND
2.6
6.4
NR"
NR
Overall removal
efficiency (%)
96.9°
98.2"
>99.9"
>99.9"
90"
95b
95
90
Source
Paint spray booth
Toyobo, Ltd. (Japan)
Paint spray booth
Toyobo, Ltd. (Japan)
Paint spray booth
Toyobo, Ltd. (Japan)
Paint spray booth
Toyobo, Ltd. (Japan)
Aerospace plant
paint booth
Aerospace plant
paint booth
Aerospace plant
paint booth
Aerospace plant
paint booth
Reference
Kenson, 198593
Kenson, 198594
Kenson, 1 98595
Kenson, 198596
Kenson and
Jackson, 1 98897
Kenson and
Jackson, 198898
Kenson, 199099
Kenson, 1990'°°
ND = Not detected.
NR=Not reported.
•Based on incinerator only, this does not account for losses through the rotary carousel which would make the number somewhat lower.
bAs reported by the author, a higher removal efficiency value is suggested by reported mass emission rates at this site.
Tests 7 and 8 appear to be performance tests at the same site reported for tests 5 and 6, respectively.
dlnlet OV loading is reported as 23.04 Ib/hr for test 7 and 3.668 Ib/hr for test 8 and the combined OV emissions from both the rotary
carousel and incinerator outlets total 1.197 Ib/hr for test 7 and 3.623 Ib/hr for test 8, giving an overall efficiency of 95% and 90%, respectively.
-------
Overall control efficiency for the KPR system normally exceeds 90 percent,
including the capture efficiency of the rotary carousel and destruction efficiency of
the catalytic oxidation (or recovery) system. The individual efficiencies of the rotary
carousel unit and catalytic unit are reported to be 95 to 98 percent each.
3.3.4 CADRE (Calgon, Inc.)
The CADRE process (Figure 3-4) uses two fixed beds of activated carbon
(one on-line, one off-line) to adsorb dilute organics. The carbon beds function to
concentrate the 0V, producing upon regeneration a higher concentration, lower
flow-rate gas stream that can be incinerated more economically compared to a
thermal incinerator designed for the dilute inlet gas stream. Concentrating 0V
results in both lower capital costs (i.e., a smaller incinerator) and lower operating
costs (i.e., less auxiliary fuel required for oxidation and the heat produced in the
oxidation step used for regeneration of the carbon). This process was developed
specifically for the control of gases with concentrations in the range of interest
here. As with other adsorption based processes, particulates and high boiling
compounds (>200°F) must be removed before the gas contacts the carbon beds.
The CADRE system can be applied to air-stripper offgases, surface coating
operations, and a wide range of other manufacturing processes.101 However, only
seven actual installations of the CADRE system exist as of September 1991. The
CADRE system is designed for 1,000 to 50,000 scfm but most installations to date
have been for inlet gas concentrations above 100 ppm and for flow rates near
50,000 scfm. Results of a pilot unit test feeding a gas containing 230 to 350 ppm
0V,102 and tests results for gases below 100 ppm are shown in Table 3-5. Calgon
recently has installed a new CADRE system controlling high flow, low concentration
0V streams in the U.S. This system is installed on a 320,000 cfm paint booth at
the Saturn Corporation in Spring Hill, TN. This system has reportedly undergone
warranty testing, and is scheduled for an agency compliance test as soon as
methods are approved by EPA.103
58
-------
Clean Air
Regeneration
Gas Blower
Combustion
Air Blower
Natural Gas
Figure 3-4. CADRE Adsorption-Regeneration Process
59
-------
Table 3-5. Summary of Field Studies of CADRE Adsorption/Incineration System for Gases Containing less than
100 ppm Inlet OV Concentration
Vendoi
Calgon
JCADRES
Calgon
(CADRE)
Calgon
SCADREI
O> Calgon
0 (CADRE)
Calgon
(CADRE)
NR = Not
Adsorber Inlet
Gas flow concentration
r Site (scfm) (ppm)
Western 2,650- 31
Processing 3,170
Purex, (Nassau 9,000
County, NY)
Occidental 3,000 ~ 1 5
Chemical
(Ashtabula,
OH)
Steelcase 77,000 73
(Grand Rapids,
Ml)
Calgon pilot 1,500 60-270
test
reported
Thermal Oxidizer
Outlet Destruction/
concentration removal
(ppm) efficiency (%) Comments
NR 17-71
NR" NR" CADRE system
used for
periodic control
of air-stripper
offgases
NR 97 OV from
furniture coating
<0.1b >99C Chlorinated C,,
C2 compounds
References
PEI, 1989104
NETAC, 1991105
NETAC, 1991106
Calgon, 1991107
SenGupta and
Schuliger,
undated108
"Because of the high variability in inlet concentration, reliable data are not available.
bThe CADRE process has two outlet gases, one from the carbon adsorber and one from the thermal oxidizer. The 0.1 ppm is the outlet from the
oxidizer; the out concentration from the adsorber is not reported.
This is the removal efficiency of the adsorber. The removal efficiency of the oxidizer is >99.8% based on 60 ppm inlet and a reported detection
limit of 0.1 ppm.
-------
SenGupta gives costs for a case study comparing a regenerative thermal
oxidizer with 90 percent heat recovery and a CADRE system (consisting of
multiple units) for a 110,000 scfm flow rate on an 0V inlet concentration of
42 ppm.109 Fuel costs for the regenerative thermal oxidizer are roughly an order
of magnitude greater than for the CADRE system, making CADRE more
economical, at least for this case study.
3.3.5 Catalytica
This process is similar in principle to other adsorption/incineration system
such as the Met-Pro KPR process. The 0V are adsorbed in a first stage,
thermally desorbed, and then catalytically oxidized. Little technical detail of the
technology is given by Catalytica. For example, the adsorbent is not disclosed
nor is the contacting scheme (rotary carousel, fixed bed, etc.). A key to the
process is said to be the "way in which the adsorbent and catalyst are
heated.110" A technical brochure, apparently elaborating on this, states that
"Heat is supplied only during the oxidation position of the cycles. . ." and "The
adsorption and oxidation systems are heated directly ..." Their process also is
said to be highly automated and not to require dedicated operating labor.
Catalytica claims its process is most advantageous for gases in the
100 to 20,000 scfm flow range with concentrations between 50 and
1,000 ppm. They report what is presumed to be a lab demonstration giving
96 percent removal of 250 ppm methyl ethyl ketone in air.
Catalytica has no adsorption based control systems operating on low
concentration, high flow OV streams in the U.S.111; therefore, the Catalytica
system has not yet been field tested for the concentrations and flow rates of
interest, though a working prototype was being developed under an EPA
Phase II Small Business Innovative Research (SBIR) Grant at the time
information was gathered for this report.
61
-------
3.3.6 Munters Zeol
The Munters Zeol system (Figure 3-5} consists of a rotary adsorber,
similar in principle to the Met Pro KPR unit, differing in that it rotates on a
horizontal rather than vertical axis and that it is made from a hydrophobic zeolite
rather than activated carbon fiber. As in the Met-Pro system, the dilute
contaminant-bearing air flows through one sector of the carousel while,
simultaneously, hot desorption air removes the contaminants in another sector,
at a much higher concentration. As with other adsorption systems, both
particulate filters and some adsorbent for removing high boiling compounds are
provided. An interesting feature of the Munters system are zeolite beds located
upstream of the rotary carousel. They are used to minimize rapid changes in
OV concentration from the process by adsorbing them when their concentration
is high and desorbing them into the inlet gas to the rotary carousel when the
concentration drops. Though Munters Zeol does not manufacture either the
incinerator or carbon recovery units for treating the concentrated desorbed gas,
the company provides them for their customers.
Munters also makes a fixed-bed version using the same hydrophobic
zeolite for applications where there are "reactive or high boiling solvents." The
high thermal stability of the zeolite allows the zeolite to be simply heated in air
to high temperatures which either desorb the OV or burn it off.
Munters claims that their hydrophobic zeolite has two advantages over
activated carbon. The first is a higher capacity at lower solvent concentrations
(Figure 3-6). The second is a higher capacity at relative humidities above
50 percent (Figure 3-7).
Munters has provided a list showing installations using the hydrophobic
zeolite system, in Europe, the USA, and Japan (Table 3-6). Inlet concentrations
for these units vary from 20 to 150 ppm. No specific removal efficiencies are
available for individual field installations, but meeting the European regulations
would require 95 percent efficiency for an inlet concentration of 90 ppm or
62
-------
Prefilters
(Zeolite Beds)
Zeolite Adsorption
Cylinder
Fan
Organic Laden
Air
Incenerator
Recovery
Heat Exchanger
Solvent (Return to Process)
Hot
Desorbing
Air
Figure 3-5. Munters Zeol System
63
-------
Inlet Solvent Concentration (ppm)
Zeolites have greater capacity than carbon at low inlet concentrations.
At high concentrations, carbon has more capacity.
Figure 3-6. Munters1 Hydrophobic Zeolite Showing Inlet Solvent
Concentration Versus Adsorption Capacity
31.
c
o <
w ^
50 100
Relative Humidity (%)
Hydrophobic zeolites adsorb very little water until the relative
humidity exceeds 90%.
Figure 3-7, Munters' Hydrophobic Zeolite Showing Relative Humidity
Versus Adsorption Capacity
64
-------
table 3-6. OV Abatement Systems Using Munters Hydrophobic Zeolites"
o>
Company
Installations in United
GM Linden
Read Rite
Sillconix
Incorporated
Sillconix
Incorporated
Location
States:
New Jersey
Fremont, California
Milpitas, California
Milpitas, California
Digital Equipment Shrewsbury,
Corporation Massachusetts
Worthington
Plastics
Munters
Corporation
YDK America,
Incorporated
Installations in Europe
Tetra Pak
Becker Acronma
ABB Flakt
AlS/Citroen
Daimier Benz
Mason, Ohio
Jackson, GA
Canton, Georgia
Berlin, Germany
Marsta, Sweden
Va'xjo, Sweden
Rennes, France
Bremen, Germany
Size (cfm)
1 40,000
30,000
10,000
10,000
20,000
64,000
15,000
30,000
26,000
1 1 ,000
59,000
18,000
1 40,000
Plant type
Concentrator
Concentrator &
Thermal Oxidizer
Concentrator &
Thermal Oxidizer
Concentrator &
Thermal Oxidizer
Concentrator &
Thermal Oxidizer
Concentrator
Concentrator &
Catalytic Oxidizer
Concentrator &
Catalytic Oxidizer
Fixed Bed
Concentrator
Concentrator
Concentrator
Concentrator
Year of
installation
1993
1993
1993
1993
1993
1992
1992
1991
1991
1991
1991
1991
1991
Type of pollutant
Automotive
Spray Painting
Mixed Solvents
Mixed Solvents
Mixed Solvents
Mixed Solvents
Paint Solvents
MEK/Toluene
Paint Solvents
Plastic Fumes
Paint Solvents
Paint Solvents
Paint Solvents
Paint Solvents
Industry
Automotive
Semiconductor
Semiconductor
Semiconductor
Semiconductor
Automotive
Plastic Parts
PVC Glueing
Computer
Plastic Parts
Packaging
Paint/ Resins
Manufacturing
Auto Paint Pilot
Plant
Automotive
Automotive
(continued)
-------
TABLE 3-6 (continued)
Company
!BM
Hone
Tetra Pak
Soab
Volvo
Voivo
o>
°> Volvo
Termoregulator
Tetra Pak
Saab
Daimler Benz
Nusec
AGA
Tetra Pak
Dalmter Benz
Tetra Pak
Location
Paris, France
Dorfprozelten,
Germany
Madrid, Spain
Molndal, Sweden
Torslanda, Sweden
Umea, Sweden
UmeS, Sweden
Motala, Sweden
Forshaga, Sweden
Lulea, Sweden
Bremen, Germany
Hamburg, Germany
Knivsta, Sweden
Lund, Sweden
Sindel Figen,
Germany
Lund, Sweden
Size (cfm)
30,000
1 00,000
24,000
1 3,000
34,000
10,600
75,000
1 8,000
27,000
18,000
24,000
1,200
1 1 ,000
27,000
1,000
3,000
Plant type
Concentrator &
Thermal Oxidizer
Concentrator Wet
Electrostatic
Precipitator
Fixed Bed
Concentrator &
Thermal Oxidizer
Concentrator
Concentrator &
Catalytic Oxidizer
Concentrator &
Catalytic Oxidizer
Concentrator
Fixed Bed
Concentrator &
Catalytic Oxidizer
Concentrator
Concentrator
Fixed Bed
Fixed Bed
Concentrator
Fixed Bed
Year of
installation
1991
1991
1990
1990
1990
1990
1990
1990
1990
1990
1989
1989
1989
1988
1988
1987
Type of pollutant
Acetone, NMP
Paint Solvents
Solvents
Solvents
Solvents
Solvents
Solvents
Solvents
Solvents
Solvents
Solvents
Petroleum Comp
Solvents
Plastic Fumes
Solvents
Plastic Fumes
Industry
Semiconductor
Automotive
Plastic Parts
Packaging
Metal
Fabrications
Automotive
Automotive
(Truck)
Auto-Truck
Metal
Fabricating
Packaging
Automotive
(Truck)
Automotive
Confidential
Gas Cylinders
Packaging
Automotive
Pilot Plant
Packaging Pilot
(continued)
-------
TABLE 3-6 (continued)
Company Location
Volvo Torslanda, Sweden
Installations in Japan:
Toyo Can Yokohama, Japan
Hitachi Zosen Japan
Size (cfm)
3,500
7,000
17,700
Plant type
Fixed Bed
Concentrator &
Catalytic Incinerator
Concentrator &
Catalytic Incinerator
Year of
installation
1987
1993
1993
Type of pollutant
Solvents
Solvents
Solvents
Industry
Packaging Pilot
Can Coating
Ship Building
"Inlet concentrations for these 32 installations vary from 20 to 150 ppm. Outlet concentrations were reported to meet regulations requiring
less than 20 mg/Nm3 which, for a compound of molecular weight 100 would be 4.5 ppm.
-------
higher. European regulations, specifically the German "TA Luft," require a
specific outlet concentration of 20 mg/Nm3 which, for a compound of molecular
weight 100, would be 4.5 ppm. Flow rates for these installations range from
1,000 to 140,000 cfm and many applications are for control of painting
emissions with typical (inlet) concentrations of 25 to 75 ppm 0V; control
efficiencies for these applications are reported by Munters to be generally above
95 percent.
Munters Zeol has installed one new control system in the U.S. It is a
135,000 cfm concentrator on a refinishing operation at Letterkenny Army
Depot/ABB Paint in Chambersburg, PA. No source test data or permit
information has been obtained for this unit.
3.3.7 Amcec
The Amcec HYBRID process uses multiple fixed beds of activated carborr
to adsorb the 0V (Figure 3-8). These beds are then cyclically regenerated using
steam and the desorbed gas, containing a higher concentration of OV, is
thermally oxidized. The heat from the oxidation step is used to produce the low
pressure regeneration steam.
Amcec has no modified adsorption based control systems operating on
low concentration, high flow OV streams in the U.S.112; though in mid-1991,
they described a project involving two Amcec systems with a common oxidizer
as being under construction. The size and inlet concentration were not given.
3.3.8 Dedert/Lurgi Cyclosorbon
Dedert markets a conventional dual fixed-bed carbon adsorber called
Supersorbon" ,113 The beds are regenerated with steam and the OV-steam
mixture is condensed and gravity separated. A schematic is shown in
Figure 3-9. As with other systems using steam regeneration, OV that reacts
with steam or is miscible with water cannot be recovered in this system.
However, common hydrocarbon solvents such as toluene, hexane, carbon
tetrachloride, acetone, and methylene chloride can be recovered. There are
68
-------
Process
Vent Air
Clean Air
Exported Clean
Steam Air
Waste Heat
Boiler
Oxidizer
Air
Figure 3-8. Amcec HYBRID Adsorption/Oxidizer Process
69
-------
Clean
Exhaust
Flow While Steaming
Adsorber 2
Subcooler
5»M
^)
/
t
.—
Recovered T
i
^ L
)eca
•^fi^
T Water
Solvent
Figure 3-9. Supersorbon Solvent Recovery Plant
70
-------
plans to install this type of system on a waste wood-fired boiler at an installed
cost of $300,000 for a 20,000 ft3/min system ($1 5/ft3/min).
In addition, Dedert/Lurgi has recently developed a modified adsorption
process, the Cyclosorbon, that uses multiple cells of pelletized activated carbon
or zeolite as the adsorbate. The concentrated desorption gas is then thermally
incinerated. This system which is designed specifically for low organic
concentration gas streams can process gas flows from 5.7 to 133.1 cms
(12,000 to 282,000 scfm) by increasing the number of adsorber cells from one
up to nine, as needed. The manufacturer notes that in the Cyclosorbon system
(Figure 3-10) the adsorption cells are individually valved to allow operation in
either adsorption or desorption mode. No performance or cost data are
currently available from the manufacturer on this system; however, the
manufacturer reports that the Cyclosorbon is competitively priced and offers
advantages over the carousel (or rotary wheel) adsorber; for example, no
rotating face seals are required, a wide selection of compatible adsorbent media
are available, and the use of conventional adsorbents results in low cost
replacement of adsorbent media when its useful life is finished. It is said to be
applicable to OV concentrations from 50 to 500 ppm.114 Installed capital costs
range from about $25/ft3/min (for low flow rates) to 1 5/ft3/min (for
100,000 ft3/min or larger).115 Dedert/Lurgi Cyclosorbon has no systems
installed in the U.S. on high flow, low concentration OV streams at this time.116
3.3.9 HONEYDACS" System (Daikin Industries)
This system is based on a rotary carbon wheel, similar in principle to that
used in the Met-Pro KPR" system previously described in this report. The
OV-containing gas flows through one section of the wheel while hot desorption
gas flows through another (see Figure 3-3). The desorbed gas is simply
discharged outdoors in most odor control applications. However, Daikin
Industries does provide both catalytic incineration and recovery options for the
71
-------
_^. Clean Air to
Atmosphere
o|
00
of
o-g of
O-j/j
Clean Air to
Atmosphere
Hot Desorbing Air
Incinerator Air Heater
Concentrated Solvent
Laden Air (Low Volume)
L.3 g
Solvent Laden Air
(High Volume)
Fresh Air
Figure 3-10. Dedert/Lurgi Cyclosorbon Modified Adsorption System
72
-------
desorbed gas. They suggest that catalytic incineration be used when the 0V
concentration is 800 to 1,000 ppm, and that recovery be used for chlorinated
0V. Table 3-7 shows the performance of the adsorption system on a
hydrocarbon OV-containing gas from a paint booth with an inlet concentration
of 172 ppm. Specifications are given in their product literature for various unit
sizes with solvent concentrations of 100 ppm, suggesting that this system can
be applied to the gas streams of interest here. No costs were available.
3.3.10 Diirr Industries System
Diirr manufactures various control systems including adsorption based
rotary concentrators as well as incinerators with regenerative heat exchangers.
DCirr also teams with other vendors such as Anguil to provide a variety of
complete, integrated control systems. Crompton and Gupta describe a Durr
rotary adsorption system that can be either carbon or zeolite (Figure 3-11).117
The rotary wheel is protected by a fixed carbon "guard bed" which removes
high molecular weight compounds. The desorbed 0V can be thermally or
catalytically incinerated or recovered by condensation or a second-stage
adsorber. Results are reported on a pilot unit (of unspecified size) for a ternary
mixture of methanol, xylene, and methyl n-amyl ketone (MAK). Total 0V inlet
concentrations of these compounds were 80, 172, and 275 ppm, respectively.
Results shown in Figure 3-12 correspond to removal efficiencies in the adsorber
of about 93 percent for the three inlet concentrations for carbon. There will be
some emissions from the thermal oxidizer as well. Somewhat lower removal
removals (83 to 93 percent) were observed for the zeolite adsorbent. Rotary
wheels containing both carbon and zeolite had higher removals than either
single material (96 to 98 percent). Blocki presents test results from in-house
studies of the Durr system at 0V concentrations of 70 and 85 ppm
(Table 3-8).118 Overall removals are 91 and 95 percent for two simulated
solvent mixtures containing polar and nonpolar organics.
73
-------
Table 3-7. Composition of Organic Solvents Versus Efficiency for the
HONEYDACS™ System
Exhaust gas
Solvent concentration (ppm)
fl-Hexane
Acetone
Benzene
Toluene
/7-Butanol
p-Xylene
m-Xylene
o-Xylene
Total
2.2
1.2
5.4
104.2
11.4
22.0
18.8
6.4
171.6
Purified gas
concentration (ppm)
0.6
0.2
0.4
3.2
0.3
0.5
0.4
0.1
5.7
Deodorizing
efficiency (%)
72.7
83.3
92.6
96.9
97.4
97.7
97.9
98.4
96.7
Precleaning
Unit
Exhaust Air Fan
t
Solvent-laden
Exhaust Air
Cleaned
Rotor Unit Exhaust Air
Desorption Air J Fresh Air
Heat Exchanger >^^x Fan
Thermal
Inceneration
Cleaned Exhaust Air
Heat
Exchanger
Stack
Hot Air
(Desorption Air)
Hot Air Fan
Combustion
Chamber
Figure 3-11. Schematic Diagram of a Durr Industries' Rotory Concentrator
74
-------
CJ1
Figure 3-12. Outlet Concentration Profiles for Carbon Honeycomb Blocks for Solvent Mixture
-------
Table 3-8. Results of Test of Diirr Industries System
Test Conditions
Inlet gas temperature (°C)
Inlet gas RH (%)
Face velocity (m/s)
Reactivation
Concentration ratio
32 °C
60
1.8
160 °C
10
Concentration
(ppm)
Solvent
Solvent Compostion One
Solvesso 100
Ethanol
MAK
MEK
MIBK
n-Butyl acetate
Xylene
Octyl acetate
Solvent Compostion Two
Butyl Cellosolve Acetate
Xylene
I PA
Ethyl acetate
Butyl cellosolve
MW
--
46.1
114.2
72.1
100,2
116.2
106
172
160
106
60
88
118
vol%
20
21.9
16.1
8.3
7.6
9.9
9.7
6.5
wt'd avg.
5.0
80.0
8.0
3.0
4.0
wt'd avg.
BP (°C)
160
78
150
80
118
126
138
200
127
192
138
66
77
168
134
In
14
15.3
11.3
5.9
5.3
6.9
6.7
4.6
70
4.3
68.0
6.8
2.6
3.4
85.0
Out
1.72
3.37
N.D.
0.05
N.D.
N.D.
1.15
N.D.
6.29
0
3.3
0.8
0
0
4.1
Efficiency
87.7
78
100
99.1
100
100
82.8
100
91
100
95.2
88.0
100
100
95.2
76
-------
Figure 3-13 shows that, for the Durr systems, catalytic and regenerative
thermal oxidation and concentration (by adsorption, followed by recovery or
oxidation) are applicable for the concentrations that are of interest here. This
figure also characterizes 0V concentration of less than 100 ppm as typically
being malodorous, not necessarily hydrocarbons only. Durr has seven control
systems installed in the U.S. on high flow, low concentration organic vapor
streams.119 All use rotary zeolite or activated carbon adsorbers followed by
incinerators. Six are installed on automobile painting lines, and one is installed
on a semi-conductor facility.
Annual electricity and fuel costs are given (Table 3-9), but no allowance
is made for low 0V concentrations. The costs given, however, show the clear
advantage of an adsorption concentrator. Table 3-10 presents a comparison of
the Durr system to a regenerative thermal oxidizer and a carbon-based rotary
adsorption system, showing the lower utility costs of the latter two
concentrators and a somewhat lower capital cost of the hydrophobic zeolite
concentrator.
3.3.10.1 Permit Conditions
Review of the draft permit (dated, July 18, 1994) issued by the Air
Pollution Control Bureau of the New Mexico Environment Department to the
Intel Corporation does indicate that permit conditions are placed on the three
Durr VOC control systems in use at this facility.120 Condition 1 .c specifies
allowable emission rates prior to control resulting from the Durr thermal oxidizer.
Condition 3.a establishes operating requirements for the units, and notes an
oxidizer efficiency of 90%. Condition 3.c specifies that each thermal oxidizer
unit shall achieve and maintain a VOC reduction efficiency of at lest 90% on an
hourly basis for all VOC's except methanol (60%). Various other conditions are
specified in Section C of the permit; these deal with requirements for the
thermal oxidizer units relating to maintenance, restrictions on halogenated
compounds, fuel use, firing rate, and incinerator combustion temperature. No
77
-------
C
.o
"«*3
2
•+-»
CD
O
O
O
100
10
0.1
Chemical/pharmaceutical Industry
500
Printing and Coating of
Foils, Paper, Te>
Paint Bake Ovens and
Paint Spray Booths
5,000
50,000
Exhaust Air Volume (nm3/h)
Note: 1g/Nm3 = 224 ppm for a compound of MW =100
5,000 Nm3/h = 2,940 stdfP/min
Figure 3-13. Diirr Industries Concentration Versus
Flow Rate Application Chart
78
-------
Table 3-9. Comparative Operating Costs for Diirr Systems
Abatement options
30,000 std ft3/min
Concentrator/thermal oxidizer
Regenerative thermal oxidizer
Recuperative thermal oxidizer
Catalytic thermal oxidizer
Gas
$18,854
66,568
320,309
172,931
Annual operating cost
Electricity
$9,951
77,942
28,169
28,169
Total
$28,805
144,510
348,478
201,100
1. Solvent credit for 50 Ib/h input at 1 5,000 Btu/lb.
2. Gas cost = $3.00/1,000 ft3
3. Electric power cost = $0.05/kWh
4. Annual operating hours = 8,000/yr.
-------
Table 3-10. Durr Industries Comparative Costs
Case One
Investment cost
(base =100)
Fuel gas
Electricity
Dry filters
Total utility cost
Carbon rotor Hydrophobic
Regenerative concentration/ zeolite
thermal thermal concentration/
oxidation oxidation thermal oxidation Basis
100
$145,300 $45,
$105,200 $25,
$ 19,300" $19,
$269,800 $90,
102.8 91.7
800 $45,800
800 $24,800
300 $19,300
900 $90,900
68,000 cfm
$4.00/MMBtu
$0.05/kWh
1 .0 g/K stdft3
inlet
Case Two
Complete cost
Fuel
Electricity
Dry filters
Total annual cost
Direct regenerative
thermal oxidation
$8,200,000
$574,600
$496,100
$0a
$1,070,700
Hydrophobic zeolite
concentration/thermal
oxidation
$7,800,000
$79,200
$134,900
$60,200
$274,300
Basis
327,000 cfm
$4.00/MMBtu
$0.06/kWh
1.5 g/K stdft3 inlet
Dry filters are not mandatory for the regenerative thermal oxidation option. Without filters a
bake-out feature is typically purchased. When operated, fuel gas expense will be higher
than stated above.
permit conditions relate to the gas stream characteristics, such as flow rate or
VOC concentration, for the Durr systems at the facility.
Review of the engineering evaluation included in the New United Motor
Manufacturing, Inc., (NUMMI) facility in Fremont, California application for an air
permit (Application Number 3611, Plant Number 1438) indicated that Durr
rotary carbon adsorption/incineration systems were proposed a BACT for control
of VOC emission from paint shop spray booths.121 The estimated capture
efficiency of these systems is reported by NUMMI as 85% and the minimum
destruction efficiency of the incinerator is 95%. NUMM! estimates that the
cost of VOC control for the spray booth will be approximately $20,000 per ton,
using a 10 year annualization.
80
-------
3.3.10.2 Source Testing Data
Review of the source test report for the stack testing conducted at the
Ford Motor Company's Twin Cities Assembly Plant in April 1993 indicates that
the Diirr concentrator/oxidizer system achieved an overall VOC removal
efficiency of 95.8 percent.122 The total inlet gas flow rate was calculated to
average about 370,000 dscfm during the test runs. The inlet VOC
concentration was well below 100 ppmv, averaging about 50.4 ppmv on the
uncontrolled stream (measured as propane). The carbon wheel removal
efficiency was calculated at an average of 97 percent. The incinerator outlet
was measured to be 39,565 dscfm with a VOC concentration of 10 ppm, for an
overall emission rate of 2.78 Ib/hr from the incinerator exhaust. Interpretation
of the available source test summaries was not straight forward however; a
diagram of the source and control device would aide in further interpreting the
test results in relation to the overall performance of the control system.
VOC destruction efficiency and emissions testing were conducted in
January, February, and March, 1992, on the incinerators located at the New
United Motor Manufacturing, Inc. (NUMMI) facility in Fremont, California.123 At
the NUMMI automotive production facility, fumes from the truck plant coating
ovens are vented to four Diirr thermal incinerators; fumes from the truck plant
spray booths are vented to three combination carbon adsorption and incineration
units. The incinerators were tested by simultaneously monitoring the inlet duct
of the incinerator and the outlet exhaust stack for VOC concentration using
BAAQMD method St-7; tests were conducted at a variety of incinerator
operating temperatures, flow rates, inlet VOC mass loadings, and VOC
concentrations. This was done in order to ascertain the effect of incinerator
temperature on controlled emissions. Gas stream VOC inlet concentrations
varied from as low as 30 ppm to as high as 1500 ppmv. Flow rates for the
incinerator test were quite low; most units showed inlet flows of less than
10,000 scfm. No flows were reported prior to the rotary concentrator. The
81
-------
VOC destruction efficiencies determined under the variety of operating
conditions were, in nearly all cases, greater than 95%.
3.3.11 Eisenmann System
Eisenmann markets a rotary adsorption system that can be coupled with
either a catalytic or thermal incinerator (for low value, multicomponent 0V) or a
condenser (for high value, relatively pure 0V). Figure 3-14 shows the rotary
system coupled to a thermal oxidizer for 0V destruction and Figure 3-1 5 shows
the rotary system coupled to a condenser for 0V recovery. Table 3-11
summarizes field installations reported by Eisenmann for 0V concentrations
approaching 100 ppm.
3.3.12 Purus System
Purus has developed the PADRE™ system that uses a hydrophobic
polymer sorbent developed by Dow. A schematic is shown in Figure 3-16. The
system consists of dual fixed beds which are alternately on-line and off-line.
The 0V is condensed and recycled. One advantage claimed by Purus is that the
Dow sorbent has high 0V capacity even at high humidities. This appears to
have led to a number of applications of this system for air stripping and soil
venting.
One field installation in California is operating on a small (14 ft3/min) gas
flow with inlet OV concentration of 330 ppm of C2 chlorinated solvents.
Adsorption isotherms are provided by Purus for OV concentrations down to less
than 10 ppm, suggesting that the sorbent is capable of removing practical levels
of OV at inlet concentrations that are of interest here. No costs were available.
3.3.13 Kelco System
The VAPOREX™ system is similar in principle to the Dedert Supersorbon
system, it consists of dual fixed carbon beds with steam regeneration. The OV
is condensed and either recycled or disposed (Figure 3-17), Field applications of
82
-------
Air Purified By
Absorption
. Paniculate
4 Filter
Conditioner
J Chambers With
Conditioned Easily
Process Replaced
Exhaust Adsorbent
Process Exhaust
With Low Solvent
Concentration
Thermal
Oxidizer
Adsorption Wheel
Exhaust With
Concentrated Solvent
Cooling Air Inlet
Cooling Air Outlet
Heated
Desorption
Air
Thermally
Purified
Exhaust
Desorption Air -^
Heat Exchanger
Figure 3-14. The Eisenmann Rotary Adsorber
83
-------
Purified Air
t
Chambers With
Paniculate Conditioned Easily
Filter Process Replaced
Conditioner Exhaust Adsorbent
Process Exhaust
With Low Solvent
Concentration
Condenser
Adsorption Wheel
Exhaust With
Concentrated Solvent
Cooling Air Inlet
Cooling Air Outlet
Solvent
Tank
Drying Unit
Desorption Air Cooler
Desorption
Air Heater
Cooling Water
Figure 3-15. The Eisenmann Rotary Adsorber Coupled
with Condensation System for Solvent Recovery
84
-------
oo
CJl
Table 3-11. Eisenmann Adsorption System Field Installations
OV source
Paint spray booth
Low temperature paint ovens
Coil coating
Inlet cone.
(ppm)
160a
125-150
160
Inlet flow
(stdft3/min)
35,300
17,600
15,300
Desorbed OV treatment
Condensation
Thermal oxidiation
Returned to process oven
OV consist of toluenes, methyl isobutyl ketone, methoxy 2 propyl acetate, ethyl glycol, and butanol.
-------
t
Soil Vapor
Inlet
Ground water
Inlet
To
Recycle
Figure 3-16. PADRE™ Schematic: Soil or Water
Remediation Vapor Treatment System
86
-------
Contaminated
Air In
Water
Chiller
Water
Out
Clean Air Out Product
Out
Figure 3-17. Kelco VAPOREX™ System
87
-------
the system have been low flow rate gases (<600 scfm) from air stripping or
soil extraction. Inlet OV concentrations in these applications often decrease
dramatically with time, sometimes from levels as high as 104 ppm. In principle,
however, this system would be applicable to low OV concentrations.
Equipment costs for a 600 ft3/min are $35,000.124
3.3.14 EC&C System
EC&C developed the EcoBAC" system, which is a fluid-bed carbon
adsorption system (Figure 3-18). The principle of operation relies on the same
adsorption/desorption process common to all adsorption systems. The
difference is that the process is carried out continuously in one vessel as the
carbon itself moves from the top (adsorption) section to the bottom (desorption
or "stripping") section. The "stripping agent" in this case is an inert hot gas
which removes the OV from the carbon, after which the OV is condensed and
recovered. The stripping agent can be steam, hot nitrogen, or even hot ambient
air in some cases.
Vendor literature states that for unspecified "low" OV concentrations and
high air flows, this system is preferable to fixed-bed systems because, in fixed
beds, the small working capacity of the carbon requires large bed volumes and
long regeneration times. Table 3-12 gives some actual field data, showing
applications with inlet OV concentrations as low as 30 ppm. This system is
generally most cost effective for high flow rates; the largest reported application
for this technology is 145,000 scfm. There are 1,000 installations worldwide.
Table 3-13 gives a summary of applications.
3.4 COSTS FOR ADSORPTION SYSTEMS
Table 3-14 shows total annualized costs and corresponding cost
effectiveness for nonregenerable and regenerable carbon-based adsorption
processes. Available information was insufficient to evaluate the modified
regenerable systems, These costs are calculated for 10,000 scfm gas flow,
other assumptions are described in Appendix B.
-------
Treated Air Outlet
BAG Returning by
Way of the Air Lift
Removing
Solvent from Air
by Adsorption
Solvent
Laden Air
Inlet
Desorption by use
of Heat and
Stripping Gas
Recovered,
Reusable solvent
t'
t
Heating Medium
Stripping Agent
bcaacaca'c
t
«
Air for Lifting BAG
X
V
V
6
6
\7
Condenser
Cooling
Water
Figure 3-18. ECC EcoBAC System
89
-------
Table 3-12. Field Data for EcoBAC™ System
Application
VOC makeup
Inlet
(ppmv)
Outlet
(ppmv)
Removal
Ink production
Semiconductor
Semiconductor
Semiconductor
LSI
Magnetic tape
production
Toluene, 350
MEK, IPA, etc.
Phenol 200
Phenol, 100
solvent blend
Naphthalene, 100
mixed solvents
Phenol, DCB 250
Terpenes, 30
mixed solvents
10
0.05
0.05
0.05
10
0.01
98.2
>99
>99
>99
96
>99
90
-------
Table 3-13. Application of the EC&C System by Industry Type and Materials
Treated
Fabric washing Perchloroethylene
Car wash Kerosene
Auto parts degreasing Trichloroethylene
Chemical production Carbon tetrachloride
PVC resin manufacturing plant Vinyl chloride
Feed processing Odor control
Film coating Toluene
Paint booth Thinner
Animal lab Odor control
Lacquering of film Toluene, ethyl acetate, MEK
Gravure printing Toluene, xylene
Medicine manufacturing plant Butanol
Electronics parts cleaning Trichloroethylene, toluene, etc.
Adhesive tape manufacturing Toluene
Iron casting Phenol, formaldehyde, ammonia
Film laminating Toluene, ethyl acetate, n-hexane
Magnetic tape Chlorinated hydrocarbon
Film coating MEK, methyl cellosolve, etc.
Electronics manufacturing plant DMF, MEK, etc.
Aluminum casting Phenol, formaldehyde, ammonia
Wastewater aeration Ordor control
Gravure printing Mixed solvents
Landfill Odor control
Ceramic condenser a-Terpineol
Atomic power plant Styrene
Medicine manufacturing plant Methylene chloride
Brewery Odor control
Sand paper manufacturing Toluene, xylene, ethyl acetate
Bakery Odor control
* Lubrication manufacturing Mixed solvent odor control
Chemical production Toluene, higher m.w. alcohols
Coating process Acrylate odor
Printing ink manufacturing Mixed solvents
Film coating Tetrahydrofuran
Rubber coating Toluene
Resin manufacturing n-Hexane
Agricultural products Organic acid & ammonia odor
Confectionery Odor control
Gravure printing n-Propanol, n-propyl acetate
Rubber vulcanizing Odor control
Electronic parts cleaning 1,1,1-trichloroethane
Resin plant Methylene chloride
Silk screen printing Xylene, mixed solvents
Cellophane coating Toluene, ethyl acetate, butyl acetate
Dye production Perchloroethylene
LSI manufacturing plant Phenol, mixed solvents
Electronics manufacturing plant Trifluorotrichloroethane
LSI manufacturing plant Acetone, methanol, etc.
91
-------
Table 3-14. Cost Effectiveness of Adsorption Processes
CO
ro
Benzene concentration Tetrachloroethylene concentration
Nonregenerable
Total annualized costs, "
$/yr
Cost effectiveness,8 $/ton
OV removed
Regenerable fixed-bed total
annualized costs
Cost Effectiveness8
$/ton OV removed
100 ppm
1,285,000
25,900
98,900
2,000
10 ppm
241,000
48,700
61,100
12,300
100 ppm
991,000
20,000
88,200
1,800
10 ppm
171,900
34,700
58,700
11,900
"All costs are in 1991 dollars rounded to nearest $100; estimates are based on 10,000 scfm flow
rate. Other assumptions discussed in text and Appendix B.
-------
SECTION 4
ABSORPTION
Absorption is a commonly applied operation in chemical processing that is
used as a raw material and/or a product recovery technique in separation and
purification of gaseous streams containing high concentrations of organics (e.g.,
in natural gas purification and coke by-product operations). In absorption, the
organics in the gas stream are dissolved in a liquid. The contact between the
absorbing liquid and the vent gas is accomplished in counter current spray
towers, scrubbers, or packed or plate columns. This emission control technique
is much more commonly employed for inorganic vapors (e.g., hydrogen sulfide,
chlorides) than for organic vapors.
The use of absorption as the primary control technique for organic vapors
is subject to several limitations and problems. One problem is the availability of
a suitable solvent. The 0V must be soluble in the absorbing liquid and even
then, for any given absorbent liquid, only 0V that are soluble can be removed.
Some common solvents that may be useful for volatile organics include water,
mineral oils, or other nonvolatile petroleum oils. Another factor that affects the
suitability of absorption for organic emissions control is the availability of
vapor/liquid equilibrium data for the specific organic/solvent system in question.
Such data are necessary for the design of absorber systems; however, they are
not readily available for uncommon organic compounds. Another consideration
in the application of absorption as a control technique is the treatment or
disposal of the material removed from the absorber. In most cases, the
scrubbing liquid containing the 0V is regenerated in an operation known as
stripping, in which the OV is desorbed from the absorbent liquid, typically at
elevated temperatures and/or under vacuum; the OV is then recovered as a
liquid by a condenser. In addition, the low outlet concentrations typically
required in organic air pollution control applications often lead to impractically
tall absorption towers, long contact times, and high liquid-gas ratios that may
not be economically viable.125 Nevertheless, for many organics, absorption can
be used to achieve extremely low outlet concentrations.
Only one commercial system was identified that is directly applicable to
low concentration (i.e., less than 100 ppm) organic gas streams, although the
93
-------
use of absorption processes at higher organic concentrations and for organic
removal coupled with sulfur dioxide (SO2) and hydrochloric acid (HCI) removal is
reported.126'127*128
4.1 QVF GLASTECHNIK
4.1.1 Principle of Operation
QVF (Weisbaden, Germany) has developed an absorption process
specifically for low concentration 0V removal. The OV-containing gas is
brought into contact with a liquid at ambient temperature and pressure in a
countercurrent absorber into which the 0V dissolves. The contaminant-
containing liquid is then regenerated by steam in a stripping column at 100 to
1 30°C temperatures under vacuum (around 50 mbar or 38 mmHg). The
regenerated liquid is then returned to the absorber column. The process is
shown in Figure 4-1.
Because the 0V must be soluble in the absorbing liquid, the choice of the
liquid is critical. The QVF system uses tetraethyleneglycol dimethylether. The
process is limited to OV with boiling points greater than 30°C. High humidities
are said not to adversely affect the process, though water vapor is absorbed
and later desorbed in the stripping column.
Because the system recovers the OV (with no ultimate disposal), this
process is limited to those gases containing compounds with boiling points
above 30 °C and that have some value when recovered. Otherwise,
subsequent disposal is needed. This system should thus be compared to
regenerable carbon adsorption processes. However, carbon systems frequently
are not able to recover low concentration OV at the 95 percent efficiency level
in a consistent and practicable way. Thus, the QVF system may have some
technical or cost advantage when both low concentrations and high removal
efficiencies are required.
The systems are designed in accordance with the German "TA-Luft
regulations," which place limits on both mass flow and concentration,
depending on the defined "class" of the emission:
94
-------
Absorption
Stripping
Organic-containing
Emissions
Figure 4-1. QVF Process
95
-------
Class
1
II
III
Mass flow
>0.1 kg/h
> 2.0 kg/h
> 3.0 kg/h
Concentration
20 mg/Nm3
100mg/Nm3
150 mg/Nm3
Typical solvents
dichloromethane
acetaldehyde
dichlorobenzene
ethylbenzene
toluene
styrene
acetone
alcohols
ethers
4.1.2 Applications
No currently documented control system for low concentration, high flow
0V streams in the U.S. used QVF Glastechnik devices. One initial press release
on QVF Glastechnik process129 states that it is designed to be used in
applications where suitably low exit concentrations cannot be achieved with
carbon adsorption. QVF has 13 plants either installed or under construction as
of mid-1 991 ; however, no data are yet available on these sites. A mobile pilot
unit (20 scfm) has been built and tested on a gas stream containing between
2 and 45 ppm methylene chloride, but no results are reported by QVF.
4.1.3 Costs
Costs for the QVF absorption process are based solely on information
supplied by the vendor. Capital costs for a 10,000 m3/h (5,883 scfm) unit are
given130 as $1.1 million. This $1.1 million is assumed to be the total capital
investment, including direct and indirect installation costs. This cost is scaled
up to the 10,000 scfm model gas stream flow rate using an exponential factor
of 0.6, i.e.,
Capital cost for 10,000 »cfm unit
10 000
° '
5,883
- 41.51 million
i
.1 million]
Utility usage rates are scaled linearly from the following values given for
the 5,883 scfm unit:
96
-------
Utility
Electricity
Cooling water
Size
5,883 scfm
11 kW
22^pm
10,000 scfm
18.7 kW
37.4 gpm
The nature of the absorption process is that the operating costs are
essentially independent of concentration in the range of interest here. This is
because the energy needed for circulation and heating of the absorbent depends
primarily on the volume of absorbent, which is constant for the cases
considered here. The energy and cooling water are assumed to be independent
of concentration and are also, for the purposes of this report, assumed to be the
same for benzene and tetrachloroethylene.
Costs and cost-effectiveness values are shown in Table 4-1.
Table 4-1. Cost Effectiveness for QVF Absorption Process
Benzene concentration Tetrachloroethylene concentration
Total annualized costs,'
$/yr
Cost effectiveness,"
$/ton 0V removed
100 ppm
331,400
6,700
10 ppm
331,400
66,900
100 ppm
331,400
3,300
10 ppm
331,400
32,800
"All costs are in 1991 dollars rounded to nearest $100; estimates are based on 10,000 scfm
flow rate. Other assumptions discussed in text.
4.2 QUAD
4.2.1 Principle of Operation
The QUAD system for 0V control is an absorption technology developed
and patented by QUAD Environmental Technologies Corporation as the
Chemtact system. In principle, it is simply a once-through (non-regenerable)
absorption process in which the absorbing liquid is finely atomized. No liquid
stripping section is provided (as in the case of the QVF process). This
technology was traditionally used for odor control applications where the
contaminant concentration may be only a few ppm and the mass of organics
involved was not considered to be a significant air pollution problem. The
QUAD system has only recently been employed for removal of toxic and
97
-------
nontoxic 0V generated in ventilated air at wastewater facilities; in this and
other types of air pollution applications, the mass of organics would be greater
and their ultimate removal or destruction must be considered.131'132
The QUAD system consists of a concurrent gas-liquid absorption
chamber. Figure 4-2 shows the schematic diagram of such an atomized mist
system. The contaminated air enters from the top and the OV-free air is
discharged from the bottom of the tower. The crucial component of this
system is a patented liquid atomizing nozzle located at the inlet of the reaction
chamber. The nozzle is designed to continuously spray liquid droplets as small
as 10 //m into the reaction tower. This provides a very high surface area
between the gas and liquid interface. The 0V molecules are absorbed from the
gas phase into these fine liquid droplets. The clean air is vented to the
atmosphere and the atomized mist containing the 0V is coalesced and removed
from the base of the reaction chamber as a liquid for further treatment. OV
removal efficiency has been reported to be as high 90 to 95 percent by this
method.133'134
The liquid absorbent solution consists of softened water mixed with
sodium hypochlorite. Depending on the type and concentration of the OV, the
pH of the scrubbing solution is adjusted with an alkali such as sodium
hydroxide. The high pH of the solution helps in capturing acid gases. Different
chemical additives can be used to enhance the OV removal efficiency.
Every QUAD Chemtact system can be custom designed to handle varying
OV concentrations, type of OV, destruction efficiency, and total gas flow rates.
By varying the tower diameter from 6 to 12 ft, the system can be designed to
handle air flows ranging from 500 to 70,000 cfm. Depending on the desired
destruction efficiency, the design gas-liquid contact times can be varied from
10 to 60 seconds by varying the tower height from 10 to 70 ft.
One advantage of this technology over the conventional countercurrent
packed tower is the complete use of the scrubbing solution and therefore no
liquid recycle. This decreases the consumption of chemicals and water and also
reduces the electrical power requirement needed to circulate the absorbent
liquid between the absorption and stripping towers in a conventional system
(e.g., Figure 4-1), Furthermore, since the reaction chamber in the QUAD
98
-------
Inlet Air
^^*
Coalesced Droplets <\
\
Spent Absorbent
- *
^ r
4fr
Atomized
Mist Nozzle
i
^ Contact (
i Chamber
I
j (
i
i
Compressed Air
Chemical Feed
Water
^
^-
Wastewater
Figure 4-2. Schematic Diagram of an Atomized Mist
System - QUAD System
99
-------
system is empty, it provides low pressure drop and does not plug up.
However, with a once-through nonregenerable absorption process, a water
treatment step is necessary for ultimate removal or destruction of the organics.
4.2.2 Applications
This system is frequently used for odor control at wastewater treatment
plants and rendering plants; the manufacturer reported that more than
30 systems were installed in 1992 and 1993 for this purpose.135 QUAD also
has installed several systems controlling low concentration 0V streams in the
U.S., but the flows do not exceed 75,000 cfm. Table 4-2 shows information
from a few of the sites at which this technology has been applied. Tables 4-3
to 4-6 present the 0V removal efficiencies from various sites using the QUAD
system. In all cases, the overall removals were in the range of 60 to
98 percent, most frequently over 80 percent. In general, the removal efficiency
for benzene was effectively 100 percent. However, the toluene removal
efficiency was in the range of 50 to 93 percent.
4.2.3 Permit Conditions
The permit issued to Valley Proteins, Inc., by the Maryland Air
Management Administration (Permit No. 02-00023, Issued October 1, 1992)
lists a QUAD system, two stage fog air scrubber, Model 11-11-1, in the
permit's source description.136 The spray scrubber serves as control equipment
for the Dupps Continuous Rendering Cookers. The State reported in their cover
letter that there are no source test information or data available on this facility,
but, the company reported that the control equipment had been stack tested in
July 1 992. No gas stream flow rate or concentration data were contained in
the permit. The only terms and conditions in the permit were a requirement for
proper maintenance and operation of the QUAD scrubbing system. No limits or
restrictions are placed on the emission rate, gas stream characteristics, or
overall performance of the device as a part of the operating permit.
4.2.4 Costs
Although no cost data were available at the time information was
gathered for this report, the vendor believes this technology to be significantly
cost competitive compared to carbon adsorption. Ullinsky et al. have compared
the cost for treatment of 90,000 scfm air flow at Los Angeles - Glendale water
100
-------
Table 4-2. Summary of Field Studies of QUAD System for Gases Containing less than 100
ppm Inlet 0V Concentrations
Technology Vendor
Mist scrubbing QUAD
Site
Site 1
Site 2
Site 3
Site 4
Gas
flow
(scfm)
NR
NR
NR
NR
Concentration, ppm
Inlet
3.61"
10.75"
0.49C
1.25"
Outlet
1.17
1.65
0.083
0.027
Removal
efficiency
(%)
67.6
84.6
83.2
97.8
Reference
Rafson
(1991)137
NR = Not reported
•Mixture of benzene, toluene and an unknown. Feedstream from wastewater plant.
bMixture of benzene and toluene. Feedstream from wastewater plant.
GMixture of benzene, toluene and an unknown. Feedstream from compost facility.
"Mixture of benzene, toluene and an unknown. Feedstream from dewatering facility.
-------
Table 4-3. Results of OV Removal from Waste water Plant Using the QUAD
System - Site 1
Concentration (ppm)
OV
Benzene
Toluene
Unknown 1
Four more unknowns
Total
"ND = less than 0.005 ppm
Table 4-4. Results of
OV
Benzene
Toluene
Total
aND = less than 0,005 ppm
Table 4-5. Results of
OV
Unknown
Benzene
Toluene
Total
Inlet
0.135
2.75
0.55
0.175
3.61
Outlet
NDa
0.97
0.2
ND
1.17
Removal efficiency (%)
>96.3
64.7
63.6
>97.1
67.6
OV Removal from Wastewater Plant Using the QUAD
System - Site 2
Concentration
Inlet
0.05
JILL
10.75
(ppm)
Outlet
NDa
1.65
1.65
OV Removal from a Compost
System
Concentration
Inlet
0.005
0.279
0.206
0.493
(ppm)
Outlet
NDa
ND
0.083
0.083
Removal efficiency (%)
>90
84.6
84.6
Facility Using the QUAD
Removal efficiency (%)
>37.5
>97.1
59.7
83.16
aND = Less than 0.005 ppm
102
-------
Table 4-6. Results of OV Removal from a Dewatering Facility Using the QUAD
System
OV
Unknown
Benzene
Toluene
Total
Concentration (ppm)
Inlet Outlet
0.863 NDa
0.005 ND
0.378 0.027
1.248 0.027
Removal efficiency (%)
>99.4
>28.6
92.8
97.84
aND = Less than 0.005 ppm
reclamation plant (LAGWRP).138 The cost comparison is shown in Table 4-7.
Dunson, referring specifically to wastewater plant odor control, states that this
type of technology is less expensive than others for concentrations below 100
ppm.139 It is not clear whether the cost of subsequent treatment of the saturated
absorbent liquid is included.
4.3 DAVIS PROCESS SYSTEM
A number of different contacting systems are available for this absorption
based technology; all are based on countercurrent packed towers (Figure 4-3). The
OV-containing gas (typically containing H2S or mercaptans), at inlet concentrations
of 60 to 100 ppm, is fed into the towers. The scrubbing solution removes the OV
and the spent solution is simply returned to the wastewater plant or discharged.
No recovery is attempted because of the dilute concentrations of contaminant in
the scrubbing liquid.
Table 4-7. ODOR/OV Emission Control Systems Costs'
(FOR EXPANDED 50 mgd PLAN, 90,000 scfm AIR FLOW)
Capital cost" Annual cost"
Mist scrubber (QUAD) $1,304,000 $251,000
Granular carbon adsorption 2,723,000 1,472,000
Mist scrubber/incineration 3,394,000 1,635,000
•All costs are in 1991 dollars.
"Includes 10 percent contingencies and 15 percent allowance for engineering, legal, and administrative.
'Includes amortized capital cost (20 years, 8-5/8 percent), power, labor, chemical, and material costs.
103
-------
Air Flow Path
To Atmosphere
A
Feed Pump
Blower
<#-
Scrubbing
Chemical
Storage
Scrubbing
Solution Flow
Path
Water Supply
Module 3
Wet Well
Figure 4-3. Typical Triplex™ Scrubber System Operational Flow Diagram
104
-------
SECTION 5
OTHER COMMERCIAL TECHNOLOGIES
In addition to the technologies described in Sections 2.0 through 4.0,
there are two other commercially available processes that differ in principle from
those described earlier and have been applied to low concentration gases.
Table 5-1 summarizes the results reported on these processes.
5.1 ULTROX D-TOX SYSTEM
5.1.1 Principle of Operation
Ultrox International (Santa Ana, CA) has developed a catalytic oxidation
system called D-TOX Process for destroying 0V in humid air.140 This system
uses small amounts of ozone to oxidize the organic compounds to carbon
dioxide and water. A patented oxidizing catalyst is used for this purpose. The
use of ozone enables the oxidation to be conducted at a relatively low
temperature, 160 to 220 °F. Any unreacted ozone in the effluent stream is
converted back to oxygen using a second catalyst. An adsorption bed placed
downstream of the catalytic reactor is used to capture any residuals or acids.
The adsorbent is made up of a mixture of bases and is replaced about every
3 months. The catalyst life is about 2 years.
The schematic of the Ultrox system is shown in Figure 5-1,141 [No
process schematic was provided by Ultrox for the D-TOX system. It is assumed
that the system shown in Figure 5-1 also represents the'D-TOX process.] No
exhaust gas stream is shown in Figure 5-1 by Palazzolo et al. (1986, p. 14). It
is assumed that the exhaust gas to the atmosphere is located just downstream
of the degasser. The OV-containing air stream is introduced upstream of the
gas-phase UV-catalytic reactor shown in Figure 5-1. An air-stripping unit,
which is part of the commercial Ultrox system and which is located between
the ozone generator and reactor in Figure 5-1, was not used in the tests
reported by Palazzofo et al. and is therefore not included in Figure 5-1. The
major parts of this system are the ozone generator and a gas phase UV-catalytic
105
-------
Table 5-1. Summary of Field Studies of Other Commercial Technologies for Gases
Containing less than 100 ppm Inlet OV Concentrations
0
O)
Technology
Enhanced carbon
adsorption
UV/oxidation
Vendor
Terr-Aqua
Enviro Systems
Ultrox (D-Tox)
Site
General Dynamics
Pomona Division, CA
Pilot test
Gas
flow
(scfm)
29,523
20
Concentration, ppm
Inlet
31 ab
52a.b
2.8-10.6
Outlet
0.31
0.21
1.9-7.2°
Destruction
efficiency {%)
99
99.6
19-64
Reference
Jackson
(1991)142
Palazzolo, et
al. (1986)143
aOV include acetone, butyl acetate, ethyl acetate, toluene, and xylene.
"Concentration converted to ppm from Ib/h summing an average molecular weight of 80.
"Calculated from reported destruction efficiencies.
-------
Recycled Air
Ozone
Generator
Air/Ozone
Polluted Air
Air/Ozone/
Hydrocarbons
UV-Catalytic
Reactor
Compressor
Makeup Air
Figure 5-1. Schematic Diagram of Pilot-Scale Ultrox
D-TOX UV-Catalytic Test System
107
-------
reactor. The OV-laden air is mixed with small amounts of ozone and passed
over the catalyst in the reactor. The reactor contains UV lamps with a specific
wavelength. The heat generated in the reactor by UV lamps is removed by a
precooled stream of air or nitrogen
5.1.2 Applications
To date, the D-TOX Process has been installed at three locations although
none are documented for low concentration, high flow OV streams in the U.S.
Performance data from these three locations and from several pilot plant studies
indicate 95 to 99 percent OV destruction are achievable with feed concentration
ranging from <2 ppm to 200 ppm.
In a parametric test of the Ultrox D-TOX process on a 20 scfm pilot
plant,144 the effect of inlet D-TOX OV concentration was investigated by using
two feed streams: one with about 15 ppm concentration, and the other with
about 3 ppm. Both streams contained trichloroethylene and
1,2-dichloroethylene. A series of experiments were conducted to study the
effect of UV, ozone, space velocity and humidity on the OV destruction
efficiency.
The presence of ozone in the feed stream was found to be important in
achieving high destruction efficiency. Thus, destruction efficiencies of only
16 to 67 percent were achieved in the absence of ozone over a wide range of
space velocities. Table 5-2 summarizes the results, in absence of ozone, for
three different space velocities, two levels of humidity and two different OV
concentrations. The single most important parameter affecting destruction
efficiency was space velocity, i.e., the relationship between feed rate and
reactor volume in a flow process is defined as the volume per unit time per unit
volume of reactor (e.g., hours(h)1). The highest destruction efficiency
(64 percent) was obtained at 200 h'1.
The presence of small quantities of ozone (140 to 440 ppm) increased
the destruction efficiency of trichloroethylene and 1,2-dichloroethylene to as
high as 99 percent, as shown in Table 5-3. However, three products of
incomplete combustion were formed in the presence of ozone. Two of these
were identified as methyl formate and methyl acetate. The total concentration
108
-------
Table 5-2. Summary of Test Results on LMtrox D-TOX System Without Ozone*
o
to
Experiment
No.
1
2
3
4
5
Compound
dichloroethylene
trichloroethylene
Total
dichloroethylene
trichloroethylene
Total
dichloroethylene
trichloroethylene
Total
dichloroethylene
trichloroethylene
Total
dichloroethylene
trichloroethylene
Total
Inlet
concentration
(ppmv)
5.99
3.82
9.92
6.43
4x12
10.60
5.18
3.28
8.46
1.91
0.95
2.86
1.80
1.03
2.83
Space
velocity (h 1)
3,000
800
800
800
200
Humidity level
(Ib H20/lb dry air)
Ambient
(0.0133)
Ambient
(0.0138)
High
(0.0144)
High
(0.0229)
High ,
(0.0150)
Destruction
Efficiency
(%)
16
22.
19
29
2S.
32
30
44
35
36
25
32
67
££
64
Outlet CO
concentration
(ppmv)
1.3
3.0
2.3
0.5
3.0
'All tests conducted with a catalyst temperature between 80 and 96 °F.
-------
Table 5-3. Summary of Test Results on Uitrox D-TOX System with Ozone'
Experiment
No.
6
7
8
Inlet
Compound concentration
(retention time) (ppmv)
Dichloroethylene
Trichloroethylene
Unknown (1.04)°
Unknown (1.59)
Unknown (1.90)
TOTAL
Dichloroethylene
Trichloroethylene
Unknown (1.0)
Unknown (0.34)
Unknown (0.52)
TOTAL
Dichloroethylene
Trichloroethylene
Unknown (1.1)
Unknown (0.32)
Unknown (0.60)
TOTAL
4.66
2.88
-
-
-
7.54
4.83
2.99
-
-
-
7.82
4.83
2.88
-
-
-
7.71
Space Outlet
velocity concentration
(h-1) (ppmv)
80O NDb
ND
1.2
0.36
0.43
1.99
3,000 ND
ND
1.0
0.34
0.52
1.86
3,000 ND
ND
1.1
0.32
0.60
2.02
Humidity level Destruction
(Ib H2O/lb dry efficiency
air) (%)
0.0150 99 +
99 -f
-
-
-
74
0.0150 99 +
99 +
-
-
-
76
0.0150 99 +
99 +
-
-
-
74
UV lights
(on - off)
on
on
off
"All tests conducted with catalyst operating temperature of 88 °F.
"ND = not detected at 0.03 ppmv detection limit.
cUnidentified compound quantitated as trichloroethylene. GC column retention time given in parentheses.
-------
of these incomplete combustion products was 2 ppm (quantitated as
trichloroethylene). If the products of incomplete combustion are accounted for,
the total 0V destruction efficiency was only about 75 percent for tests
conducted at space velocities of 800 to 3,000 h'1.
5.2 ENHANCED CARBON ADSORPTION
5.2.1 Principle of Operation
The Terr-Aqua Enviro Systems' Air Pollution Control System combines
the industrially proven technologies of wet scrubbing, carbon adsorption, and
ozone reaction.145-146 The schematic of this system is shown in Figure 5-2. The
system uses various stages for the collection and elimination of the 0V. Stage
one is a 2-step prefilter to collect particulates from the air stream. The pre-filter
is designed to collect up to 99 percent of particulates down to a nominal 1 /vm
in size.
The organic-laden air then enters the photolytic reactor where it is
exposed to ultraviolet light and mixed with activated oxygen/ozone. At this
stage, partial destruction of the 0V begins. The air then enters a countercurrent
ozonated water scrubber, called the Aqua Reactor, where the 0V from the gas
phase is transferred to the liquid phase. The water is then heavily oxidized in
the reactor recycle tank for an extended period of time. The 0V present in the
water is oxidized to C02 and H2O, and, presumably, HCI if chlorine atoms are
present.
After the Aqua reactor, the effluent air stream enters a coalescer to
remove /vm level water droplets and wetted particles entrained in the air stream.
The air stream then enters one of two activated carbon beds which remove any
remaining 0V that did not dissolve in water. These carbon beds are alternated
every 24 hours. At any time, one of the beds is on-line to collect the 0V while
the other is sealed and fed oxidant to regenerate the carbon. During this
regeneration, the 0V is converted to C02 and H20.
Terr-Aqua Enviro Systems has no U.S. control systems for low
concentration, high flow 0V streams. They did, however submit permits and
test reports for a 18,000 cfm system at the Northrop Corporation B-2 Division
111
-------
Paint Spray
Booth
Bake Oven
Mixing Room
Hood Vent
UVOX
Main
System
Blower
to
24.000
scfm
!
1
1
1
1
1
1
!
0-/JP-0
Clean Air
to
Atmosphere
Legend
UVOX System i
_P£pcessCpntrler1
Air Stream - VOC
Oxidant/ozone Flow
Oxidant-regeneration
Mode
Control Monitors
Flow Control Valve
Figure 5-2. Schematic of Terr-Aqua Enviro Systems' Air
Pollution Control System
112
-------
in Pico Riviera, CA. The test was on a paint booth and indicated 99 percent
removal efficiency. Test reports and permit information were also submitted for
2 additional units located at General Dynamics facilities. Summaries of the
information and data for these systems are provided below.147
5.2.2 Source Test Results
5.2.2.1 Northrop Corporation, Pico Rivera, California
On January 25, 1990, personnel of VOC Testing, Inc. and Horizon Air
Measurement Services performed emissions testing of a Terr-Aqua UV-AO
Enhanced Carbon Adsorption Treatment System controlling organic emissions from
an automated paint spray booth at the Northrop Corporation Plant in Pico Rivera,
California.148 The test program included the continuous monitoring of volatile
organic concentrations at the inlet and exhaust of the control device during the
spraying of solvent compounds in the paint spray booth, using EPA's Method 25A,
and composite inlet and outlet sampling in accordance with SCAQMD
Method 25.1.
The results of the continuous monitoring performed in accordance with
EPA's Method 25a at the inlet and outlet (exhaust) of the control device are
summarized in Table 5-4.
5.2.2.2 General Dynamics, Pomona, California
Aqua Enviro Systems, Inc. contracted York Research Consultants to perform
a VOC study on their Terr Aqua Enhanced Carbon Treatment System in October,
1988.149 The VOC control unit is located at the U.S. Naval Weapons Development
Facility and is operated by General Dynamics Pamona Divison. The control system
is designed to remove OV from paint spray booths and ovens at that location. The
objectives of this testing program were to demonstrate a minimum recovery
efficiency of 91 % for the total VOC and to show a 91 % or greater recovery
efficiency for each of the solvents used.150 A spray gun was used to simulate the
OV emissions from a spray booth. The solvents chosen for the test were based on
the actual use rate in the facility on a weekly basis. EPA's Reference Method 25A
was performed to determine the total hydrocarbon concentration (THC) of the
exhaust stack. One test was performed on each of the two carbon beds in the
system. The average THC for carbon beds #1 and #2 was determined to be 12.35
ppmv (as propane) and 12.89 ppmv (as propane) respectively. The emission rate
113
-------
Table 5-4. Continuous Monitoring Results—Terr-Aqua Unit at Northrop
Corporation151
Test
Number8
1
2
3
4
5
6
Hydrocarbon Concentration
(ppmv as propane)
Inlet
644
719
564
282
232
211
Outlet
2.2
2.2
1.0
1.0
1.0
1.0
Flow Rate
• (ACFM,
Wet)
18,143
17,935
18,000
18,140
18,000
18,060
Control
Efficiency
99.7
99.7
99.8
99.6
99.6
99.5
a During tests 1 and 4 the solvent used was xylene. During tests 2 and 5 the solvent sprayed
was methyl isobutyl ketone (MIBK). During tests 3 and 6 the solvent sprayed was VM & P
Naptha (60%), toluene (20%) and cellosolve/glycol ether (20%).
(controlled) for the carbon beds was calculated to be 0.252 Ib/hr and
0.263 Ibs/hr, The removal efficiency for the system was calculated to be
97.9% for carbon bed #1 and 98.7 for bed #2 (on a mass basis). The
calculated emission rate (uncontrolled) of the solvent mixture at the control
device inlet was 11.97 Ibs/hr for carbon bed #1 and 20.34 Ibs/hr for bed #2.
Inlet gas flow rates and concentration information were included in the
materials received in the test reports.152 Table 5-5 summarizes the type of 0V
and their concentration used in the test by individual chemical constituent.
During the testing of bed #2, the feed 0V concentration was nearly doubled.
The total flow rate was 29,523 cfm. The destruction efficiency for each 0V
was 98 to 99 percent (on a concentration basis) with the overall destruction
efficiency also being above roughly 99 percent.
5.2.2.3 Genera! Dynamics, Rancho Cucamonga, California
The Terr-Aqua system at the General Dynamics Valley Systems Division
Facility was tested in September 1990 to determine hydrocarbon removal and
collection efficiency.153 Test runs were conducted simultaneously at the
control system outlet and at each of the two inlet ducts, one inlet duct venting
a touch-up spray booth and the second venting 3 spray booths. Testing of the
coater system collection and destruction efficiency shows a collection of
98.6% and a destruction efficiency of 99.4%, for an overall efficiency of
114
-------
Table 5-5. Results of Terr-Aqua Enviro Systems' Air Pollution Control
Equipment at General Dynamics, Pomona Division
CJI
ov
Bed#1
Isopropyl alcohol
Methyl ethyl ketone
Cyclohexanone
Toluene
Butyl cellosolve
m-Xylene
Bed #2
Isopropyl alcohol
Methyl ethyl ketone
Cyclohexanone
Toluene
Butyl cellosolve
/77-Xylene
Concentration
Inlet
9.09
9.60
0.79
1.29
2.78
6.16
29.71
15.47
16.33
1.34
2.16
4.72
10.45
50.47
(ppm)
Outlet
0.14
0.08
0.01
0.01
0.03
0.08
0.35
0.03
0.06
0.01
0.01
0.02
0.10
0.23
Efficiency (%)
98.5
99.2
98.7
99.2
98.9
98.7
98.8
99.8
99.6
99.3
99.5
99.6
99.0
99.5
-------
98.0 percent. Stack gas characteristics were also reported in the test results;
the average inlet gas flow for the 3 spray booths was 19,789 acfm
(17,981 dscfm) and the average for the touch-up booth was 2,143 acfm
(1,947 dscfm). The total non-Methane hydrocarbon concentrations were
438 ppm and 2,496 ppm, respectively for the 2 flows at the system inlet.
5.2.3 Permit Conditions
5.2.3.1 Northrop Corporation, Pico Rivera, California
The permit issued by the South Coast Air Quality Management District,
Permit No. D34532, A/N 175368f contains a number of permit conditions that
apply directly to the Terr-Aqua control device in use at this facility.154 Permit
Condition No. 3 states that the collection efficiency of the system shall not be
less than 90% by weight of emissions generated. Condition No. 4 states that
the destruction efficiency of the system shall not be less than 95% by weight
of emissions it collects. Condition No. 5 limits the total quantity of VOC
emissions vented to this equipment to not more than 210 Ibs in any one day.
5.2.3.2 General Dynamics, Pomona, California
The permit issued by the South Coast Air Quality Management District,
Permit No. D37050, A/N21442, contains a number of permit conditions that
apply directly to the Terr-Aqua control device in operation at this General
Dynamics facility.155 The total amount of VOC emissions vented to the control
system from the emission sources is limited to 25 Ibs per day for each of
3 spray booths and 14 Ibs per day for the remaining spray booth (Permit
Conditions 4, 5, 6, and 7). A filtering system for prefiltering the gases and a
VOC monitoring system to indicate carbon breakthrough are required by the
permit (Permit Conditions 3 and 8),
5.2.3.3 General Dynamics, Rancho Cucamonga, California
The permit issued by the South Coast Air Quality Management District,
Permit No. D 39603, A/N236597, contains a limit on the total amount of VOC
emissions that the source and control device can discharge to the atmosphere,
i.e., 1 Ib per day.'56 No other restruction relevant to the control device are
contained in the permit.
116
-------
5.2.4 Costs
No cost data were provided by the vendor. However, the reported
advantages of Terr-Aqua system include low operating and maintenance costs
and no secondary waste generation.
5.3 CONDENSATION
Condensation has not generally been considered applicable to gases
with OV concentrations that are of interest here.157 This is because
condensation is a simple vapor-liquid equilibrium process and the temperatures
needed to condense OV at levels below several thousand ppm have been
impractical. As an example, the vapor pressure of methylene chloride, a
common solvent, is 1 mm Hg (corresponding to a concentration of 1,316 ppm
at 1 atm pressure) at -70 °C. Much lower temperatures would be needed to
even begin to condense inlet concentrations of 100 ppm.
Nevertheless, condensation processes, largely based on liquid nitrogen
(-196 °C), have been developed and are claimed to be applicable for low
OV concentrations. Of course, these systems are best for low flow rate
gases, such as working and breathing losses from tank and containers.
Because there is no physical contact between the OV-containing gas and the
coolant, recovered solvents are not contaminated with water, as would be the
case if steam is used in adsorption-based systems (recall, however, that inert
nitrogen can be used to overcome this problem). Systems based on cooling by
means other than liquid nitrogen (e.g., the Brayton cooling cycle) have also
been developed.
5.3.1 Liquid Nitrogen Systems
5.3.1.1 Airco Gases Systems
The Kryoclean" system uses liquid nitrogen to cool the incoming gas in
the system shown in Figure 5-3. It has been used in the pharmaceutical
industry on low flow rate gases (500 to 1,000 stdft3/min158). The operating
cost benefit for this system depends on the pre-existence of a liquid nitrogen
storage system at the site, which is true at about half the U.S. pharmaceutical
plants. For a 500 stdft3/min system, the capital cost for the heat exchanger
117
-------
To Vapor
Condenser
To Vapor
Condenser
To Vapor
Condenser
To Vapor
Condenser
\ Defrost
1 Fluid
Precooier
Fluid
Chiller
Refrigeration
Unit
^ fe
^ w
Refrigeration
Unit
Liquid Nitrogen
Unit
oo
I I
High
Low
1,1 sta9e i sta9e
Defrost . Precooier . .
Section Section Cascade Refrigeration
p,
1
J
Final Stage
Liquid
Nitrogen
Figure 5-3. Vapor Recovery System: Refrigeration Liquid Nitrogen Sections
-------
and control system is $650,000 to $1.5 million. Two units have been
ordered for pharmaceutical plants. Inlet 0V concentrations are not known.
5.3.1.2 Edwards Engineering System
This is actually a hybrid system that couples liquid nitrogen cooling with
a conventional Rankine refrigeration cycle. A mechanical refrigeration unit
cools the gases to about -70 °C and then liquid nitrogen cools the gas further
to -185 °C (Figure 5-4). The optimum flow rate for the process is
<5,000 ft3/min.169 Several hundred such systems are installed in the field.
Vapors include solvents, gasoline, chlorocarbons, and alcohols. Inlet 0V
concentrations are not reported and no costs were available.
5.3.2 NUCON System
The Braysorb™ system combines carbon adsorption with a reverse
Brayton thermodynamic refrigeration cycle to cool OV-containing desorbed
gases. This cooling step is simply a variant on conventional mechanical
refrigeration cycles and involves the compression and expansion of a
refrigerant gas.160'161-162
A schematic is shown in Figure 5-5. The system consists of two fixed
carbon beds using nitrogen as the desorbing gas. One variation is the use of a
vacuum during desorption to remove strongly adsorbed 0V. Desorbed gas at
310 °F is cooled in a series of steps to -44 °C, although lower temperatures
are possible.
One installation is reported treating 8,000 to 10,000 stdft3/min at a 3M
plant in Greenville, South Carolina. The capital cost was $1.45 million and
annual operating costs (including depreciation) are $397,000 per year. Inlet
0V concentrations are not reported.
5.4 FLAMELESS THERMAL OXIDATION
Flameless thermal oxidation is the name given to the oxidation of
gaseous contaminants or fuels by contacting them with air (and an auxiliary
fuel for the low concentrations that are of interest here) in a hot inert ceramic
matrix which provides the necessary heat for complete reaction to take place.
There is no visible flame in these systems. Once combustion is initiated, heat
is transferred by convection and radiation from the ceramic to the incoming
119
-------
DiSCHARGE FROM UNIT
MAIN CONDENSING COIL
bTRIPPING COIL
LIQUID
GASOLINE
DEFROST
FLUID
PRESSURIZATION-K7
AIR INTAKE [
1
1 . CONDENSER
• . AIR
' : -f T -
I 1
LOW TEMPERATURE
REFRIGERATION
SYSTEM
(
J
j
:ONDENSER AIR
(PRECOOLER)
^ ^
V
PRECOOLER
REFRIGERATION
UNIT'
• •
•V
AIR VAPOR INLET
*Heat transfer fluids supplied by others
Figure 5-4. Refrigeration Vapor Recovery System: Components
-------
Turbo
Compressor
Expander
Cooler Low
Temperature
Chiller
Filter
SLA Blower
Figure 5-5. The Braysorb® Process Regeneration Flow Diagram
121
-------
gas mixture, raising its temperature to the ignition point. Heat released in this
reaction is, in part, transferred back to the ceramic. This process results in
low NOX and CO levels compared to flame-based systems which operate at
comparable temperatures. The economic feasibility of this type of system of
inlet 0V concentrations less than 100 ppm is somewhat questionable, though
field applications approaching these concentrations are reported.
5.4.1 Thermatrix System
Figure 5-6 shows the Thermatrix system, which operates on the
principle described above. Other contacting patterns are also available.
Thermatrix reports control systems on low concentration but not high flow 0V
streams.163 Test results show the use of this system on an inlet gas
containing 400 ppm isopropanol, but most results appear to be for
concentrations in the 103 to 10s ppm range. In principle, 0V concentrations of
100 ppm could be destroyed, but would require supplemental gaseous fuels to
reach a required inlet gas enthalpy of 7 to 1 5 Btu/stdft3.164 [An inlet 0V
concentration of 100 ppm corresponds to about 0.5 Btu/stdft3 for C6 to
C8 hydrocarbon solvents.]
5.4.2 Alzeta System
Alzeta manufactures a broad line of air pollution control devices,
including the Alzeta Adiabatic Radiant Burner, which is an inward firing
incinerator that produces much iess oxides of nitrogen compared to
conventional burners (see Figure 5-7). They market this incinerator with a
zeolite concentrator wheel from Munters. However, no systems installed on
low concentration, high flow 0V streams in the U.S. are documented.165
In control systems using the Alzeta Adiabatic Radiant Burner, the
OV-laden air is inducted to a blower and directed through a paper element filter
to eliminate dust and entrained droplets. After filtration, the stream passes
through a recuperator, an optional flame arrester, and then enters a mixer
where natural gas is added if necessary. The air stream is then passed
through -i perforated support screen coated with a porous ceramic or metallic
fiber mat. The mat has been treated and bonded to permit stable operation on
its surface with no flashback at ;nlet temperatures exceeding 800 °F. Typical
122
-------
Support
Screen
Incandescent
Surface
Porous
Ceramic
Matrix
Premixed Air, VOC, and Fuel Center
Figure 5-6. Thermatrix System's Porous Surface Radiant Burner
123
-------
Exhaust
VOC-iaden
stream
Preheated
Bypass
Flow
Recuperator
Figure 5-7. Alzeta Adiabatic Radiant Burner
124
-------
oxidizer temperatures are between 1,600 to 1,800 °F. As the mixture passes
through the mat, ignition and most of the combustion occur. A substantial fraction
of the combustion heat dissapates to the external surface of the burner causing it
to glow uniformly without a visible flame.
This burner is similar in principle to the Thermatrix system (Figure 5-6), but
uses different ceramic geometry and can be inwardly and outwardly fired (air/fuel
can flow inside-out or outside-in). Results are presented by Bertz and Barone for
gasoline vapors with concentrations as low as 335 ppm and for chlorobenzene at
210 ppm.166 Destruction efficiencies are >99% in both cases. Alzeta also markets
their burner coupled with an upstream rotary adsorption wheel to concentrate the
0V (Figure 5-8), identical in principle to those units described in Sections 3.3.9 to
3.3.11.
5.5 BIOFILTRATION
Biofiltration is a relatively recent air pollution control technology in which
off-gases containing biodegradable organic compounds are vented, under controlled
temperature and humidity, through a biologically active material (Figure 5-9). The
microorganisms contained in the bed of compost-like material digest or biodegrade
the organics to C02 and water. This technology has been successfully applied in
Germany and The Netherlands in many full-scale applications to control odors,
VOC, and air toxic emission from a wide range of industrial and public sector
sources, though the process is limited to organic concentrations of approximately
1,000 ppm or less. Control efficiencies of more than 90 percent have been
achieved for many common air pollutants.167 Information on capital and operating
cost for various biofilter systems installed in Europe and the USA has been
reported; however, cost-effectiveness values were not calculated for biofiltration in
this study. The literature reports that, due to lower operating costs (i.e., $0.60 to
$1.50 per 100,000 cubic feet of off-gas), biofiltration can provide significant
economic advantages over other air pollution control technologies if applied to
off-gases that contain readily biodegradable pollutants in low concentrations.168
Environmental benefits include low energy requirements and the avoidance of
cross-media transfer of pollutants. No currently documented control system for
low concentration, high flow 0V streams in the U.S. uses biofiltration.
125
-------
Thermal Processor
"I ,
^v
J_
1
t
ILL
i
1
t
i
-T
\
1
t
~1
\
1
t
1
,
1
t
1
VOC-laden Air
sSteam
A
1"
t
Clean Exhaust
V Heat
Exchanger
Natural Gas
Concentrated | Hot Air for Wheel
VOCs
Regeneration
Cleaned
Process Air
Li/
VOC Concentrator Wheel
Figure 5-8, Alzeta VOC Flameless Thermal Oxidizer
126
-------
Clean Gas
Ducting
* Blower
Raw Gas
A
t
Filter Material
^XXXXXXXXXX
>yWWW9Wy
Air Distribution System
Humidifier Drainage
Biofilter
Figure 5-9. Schematic of an Open Single-Bed Biofilter System
127
-------
This page is intentionally left blank.
128
-------
SECTION 6
EMERGING TECHNOLOGIES
In addition to the commercially available processes described in earlier
sections, there are two technologies currently under development that appear
to be applicable to low concentration gases for which technical information
was obtained. Neither has been applied in the field, though both offer some
possible advantages over current systems.
6.1 CORONA DISCHARGE PROCESSES
Corona discharge processes use a high voltage/low current electrical
discharge to destroy a wide range of molecules in an OV-containing gas
stream. Although no currently documented control system for low
concentration, high flow 0V streams in the U.S. uses corona discharge
processes, the use of these processes for low concentration OV destruction is
described by Yamamoto.169 This technology is being evaluated by RTI and
EPA/AEERL for conditions of interest here. Several process and electrical
configurations are possible. Although the so-called silent corona (which uses
radio frequency energy) and the direct current corona have been evaluated,
both have been found to require too large an energy input to be practical. Two
recent developments to overcome this limitation are the dielectric packed-bed
reactor and nanosecond pulse corona. The packed bed system is shown in
Figure 6-1.17° The OV-containing gas simply flows through a bed of dielectric
beads (such as the perovskite BaTi03). At both the entrance and exit of the
bed an electrode is connected to a high voltage AC power supply. The beads
act as small capacitors and charge and discharge out of phase with the applied
field. The intra-bead discharge generates electrons that react with the OV to
destroy it. The nanosecond pulse corona uses a wire centered in an unfilled
tube through which the OV-containing gas flows (Figure 6-2). A novel power
supply discharges a capacitor through a spark gap to generate a high voltage
pulse. The advantage of such a configuration is thought to be the generation
of free electrons without excessive generation of ions.
129
-------
Gas Flow
Teflon
AC Power Supply
Glass Tube
Fine Mesh Screen
High Dielectric
Pellets
Figure 6-1. Schematic of AC Packed-Bed Corona Reactor
130
-------
Corona Wire
Glass Tube
Teflon
Stainless Steel
Tube
O-Ring
Pulsed High
Voltage
Gas
Flow
L-™,
r=
++*ti X
,'
If I . (^ —
•\ 1 1
1 — 1
/- lvv<
W^Wf
1 —
1
Figure 6-2. Schematic of Pulsed Corona Reactor
131
-------
Results are shown in Figure 6-3 for 48 ppm toluene in air for the
nanosecond pulse corona. Though high destructive efficiencies are possible,
generation of ozone, NOX, and partial reaction products is possible (about
500 ppm ozone was generated at a voltage corresponding to near 100%
destruction efficiency in Figure 6-3). Figure 6-4 shows comparable results for
the packed-bed corona.
Further work is focused on reducing power consumption and byproduct
formation, and on scaleup to commercial application.
6.2 HETEROGENEOUS PHOTOCATALYSIS
Peral and Ollis at N.C. State University report research on the use of
near-ultraviolet light to continuously activate a semiconductor (such as
Ti02).'71 The activated surface of the semiconductor then acts as a catalyst
for the oxidation of 0V in air. A schematic of the heterogeneous
photocatalysis system is shown in Figure 6-5. This process is closely related
to the Ultrox process (Section 5.1) except that ozone is not generated
upstream of the catalyst bed. There may also be differences in the
wavelength and/or intensity of the UV light as well as the- catalyst itself, but
insufficient information is available on the Ultrox process to make this
determination. Results for formaldehyde oxidation with inlet concentrations
between 4 and 72 ppm showed destruction efficiencies between 54 and 98
1 7*3
percent."2
This technology offers the possibility of ambient temperature operation
and high oxidation activity for a wide range of organics. Possible limitations
include incomplete reactions at all but dilute concentrations (perhaps even at
100 ppm), the development of a contacting pattern to allow UV illumination of
the entire catalyst surface, and possibly slower rates at high humidities. No
currently documented control system for low concentration, high flow 0V
streams in the U.S. uses heterogeneous catalysis devices.
!32
-------
esi
Efficiency (%)
to
1
O)
CO
o.
c
»
CD
D
(D
5
I
5
Qj
55'
o
o
TJ
5'
-» to
O O
u
o
01
o
O)
o
00
o
CO
o
CJ1
NJ
O
0)
CQ
(D
Ul
00
•o
•o
CO
o
c
(D
3
(D
CO
Ul
-------
100
CO
u
0)
O
Organic inlet concentration = 50 ppmv
Gas Flowrate = 0.85 L/min
11
Voltage (kV)
2.5
.. 2
a
.. 1.5 S
01
I
.. 1
.. 0.5
17
Destruction __*_ Power
Figure 6-4 Effect of Increasing Voltage on Power Usage and Destruction Efficiency for the Packed-Bed Corona
System
-------
Gas in
UV
Light
Source
Catalyst Bed
Gas out
Photoreactor
TiO2 Beads
Figure 6-5. Experimental Heterogeneous Photocatalysis System Used
for Experiments at NCSU
135
-------
This page is intentionally left blank.
136
-------
SECTION 7
REFERENCES
1. Leson, G., and A.M. Winer,. J. Air Waste Mgt. Assoc., 41 (8): 1045-1054.
1991.
2. U.S. EPA. "OAQPS Control Cost Manual," Fourth Ed. U.S. EPA. Research
Triangle Park, NC. Publication No. 450/3-90-006. January 1990.
3. Vatavuk, W. M. Chemical Engineering, Special Supplement: Environmental
Engineering. June 1994. Page 17.
4. Rodberg, J.A., J.F. Miller, G.E. Keller, and J.E. Woods. A novel Technique to
Permanently Remove Indoor Air Pollutants," ASHRAE, Healthy Building, IAQ.
1991.
5. Dombrowski, C. (Ed.), Air Techn News, (Silver Spring, MD). pp. 31-32.
September 1991.
6. Telecon. C. W. Sanford, Research Triangle Institute (RTI), and Dan
Banks, John Zink Company. April 12, 1994. Installed air pollution
control devices.
7. Renko, R.J., 1990, Chem Eng. Prog., 47-49, October.
8. Spivey, J.J., 1990, "Catalytic Incineration of Gaseous Wastes," in Thermal
Process, H.M. Freeman (ed.), Technomic Publishing (Cincinnati, OH),
p. 95-108.
9. Palazzolo, M.A., C.L. Jamgochian, J.I. Steinmetz, and D.L Lewis, 1986,
"Destruction of Chlorinated Hydrocarbons by Catalytic Oxidation," EPA
Contract Report EPA 600/2-86-079.
10. Hylton, T.D., 1990, "Interim Report on the Performance Evaluation of the TCE
Catalytic Oxidation Unit at Wurtsmith AFB," for HQ AFESC, Tyndall AFB.
11. See Reference 10.
12. Ritts, D.,C. Garretson, Ch. Hyde, J. Lorelli, and C.D. Wolbach, 1990,
"Evaluation of Innovative Volatile Organic Compound and Hazardous Air
Pollutant control tecnologies for U.S. AF Paint Spray Booths," Draft Final
Report, EPA Contract No. 68-02-4285.
13. Telecon. C. W. Sanford, RTI, and Gary Nagle, Wheelabrator. April 22,
1994. Incinerators.
14. Cenci, Carole J., Permit Section, Air Qualtiy Division, Minnesota Pollution
Control Agency, letter and attachments to Jeffrey Muffat, 3M Environmental
Engineering and Pollution Contro. March 9, 1993.
15. See Reference 2.
137
-------
16. See Reference 9.
17. See Reference 10.
18. See Reference 12.
19. Fiedler, M.E , Anguil Environmental Systems, Inc., letter to C. W. Sanford,
RTI,June29, 1994.
20, Telecon. C. W. Sanford, RTI, and Gene Anguil. June 27, 1994.
Incinerators.
21. Patterson, Steve, Monsanto Enviro-Chem, letter and attachments to C. W.
Sanford, RTL June 7, 1994.
22. Joseph Haggin. Catalytic Oxidation Process Cleans Volatile Organics
From Exhaust. Chemical & Engineering News. June 27, 1994. Page 42.
23. Telecon. C. W. Sanford, RTI, and Paul Holland, CSM. April 12, 1994.
24. Telecon. C. W. Sanford, RTI, and Robert Saxer, Amcec. April 12, 1994.
Incinerators.
25. Telecon. C. W. Sanford, RTI, and R. Moreno, Alzeta. May 5, 1994.
Control devices.
26. Telecon. C. W, Sanford, RTI, and Tom Sanden, Thermo Electron. April
14, 1994. Control devices.
27. Telecon. C. W, Sanford, RTI, and Sophia Block, Catalytica. April 14,
1994. Control devices.
28. See Reference 2.
29. See Reference 2
30. See Reference 2, p. 3-52.
31. See Reference 2, p. 3-48.
32. Martin st al. 1993.
33. Smith Engineering Brochure, Smith Engineering Systems, Ontario, CA, 1990.
34. Telecon. C. W. Sanford, RTI, and E. Biedell, Smith Engineering Co. June
28, 1994. Incinerators.
35. Mcllwee, Roy, Smith Engineering Co., letter and attachments to
C. W. Sanford, RTI, July 14, 1994
36. See Reference 33,
37. See Reference 33.
138
-------
38. NETAC, 1991 (National Env. Tech. Appl. Corp., Pittsburgh, PA),
Environmental Product Profiles, "Re-Therm Air Toxics/VOC Control System,"
April.
39. Pennington, R.L., 1991, Reeco (Morris Plains, NJ), letter to S.K. Agarwal,
Research Triangle Institute, November 4.
40. See Reference 38.
41. See Reference 39.
42. Golson, Glen, Air Division, Alabama Department of Environmental
Management, letter and attachments to R. A. Zerbonia, RTI.
Septembers, 1994.
43. Interpoll Laboratories, Report Number 9-2726, submitted to Louisiana Pacific
Corporation, Hayward, Wisconsin. March 28, 1989.
44. Division of Qir Quality, Minnesota Pollution Control Agency, letter and
attachments to C. W. Sanford, RTI. August 1994.
45. Facsimile Transmission to C. W. Sanford, RTI, from Keith Jordan, Air Quality
Division, Louisiana Department of Environmental Quality. July 19, 1994.
Emission Test Results—Louisiana Pacific Corporation.
46. Air Quality Division, Louisiana Department of Enviornmental Quality, letter
and attachments to R. A. Zerbonia, RTI. August 11, 1994.
47. Crabtree, T.A., 1991, Smith Engineering Company, letter to RTI.
48. See Reference 42.
49. See Reference 45.
50. See Reference 46.
51. See Reference 47.
52. See Reference 42.
53. See Reference 42.
54. See Reference 38.
55. See Reference 39.
56. Telecon. C. W. Sanford, RTI, and Jack Clark, Reeco. April 14, 1994.
Incinerators.
57. Precision Environmental, St. Paul Tape Emissions Monitoring Report,
submitted to Gary Hippie, 3M/Environmental Engineering and Pollution
Control. June 1993.
58. See Reference 14.
139
-------
59. Carol Lee, M. K., Bay Area Quality Management District, letter and
attachments to C. W. Sanford, RTI.
60. Somary, G., Eisenmann, Inc., 1993, (Crystal Lake, IL), letter to J.J. Spivey,
Research Triangle Institute, August 11, 3pp.
61. See Reference 3.
62. General Control Device Requirements, 40 Code of Federal Regulations
§60.18(1993).
63. U.S. EPA, 1990, "Alternative Control Technology Document-Organic Waste
Process Vents," EPA-450/3-91-007, Office of Air Quality Planning and
Standards, Research Triangle Park, NC, December.
64. See Reference 63.
65. Spivey, J.J., Env. Prog., 7(1):31-40. February 1988.
66. Kovach, J.L., 1979, In Separation Processes for Chemical Engineers, PA.
Schweitzer, ed., McGraw-Hill, p. 3-1.
67. Ruthven, D.M., 1984, Principles of Adsorption and Adsorption Processes,
ASHRAE, Health Buildings, IAQ
68. U.S. EPA. Carbon Adsorption for Control of VOC Emissions: Theory and Full
Scale System Performance, Publication No. EPA-450/3-88-012, EPA
Contract 68-02-4378, Work Assignment 20. June 1988.
69. See Reference 68.
70. U.S. EPA. Control Technologies for Hazardous Air Pollutants (Handbook).
U.S. EPA, Office of Research and Development, Cincinnati, OH. Publication
No. EPA-625/6-91-014. June 1991.
71. Saunders, G. Comparisons of Air Stripper Simulations and Field
Performance Data. U.S. EPA, Office of Air Quality Planning and Standards,
Research Triangle Park, NC. Publication No. EPA 450/11-90-002. March
1990.
72. Durham, J. B. Chemical Scrubbing for control of Air Emissions, presented at
1991 AlChE Summer National Meeting, Pittsburgh, PA. August 1991.
73. Murphy, A.J., Dr. C.B. Moyer, and J. Ayer. VOC Control Effectiveness.
U.S. EPA, EPA Contract No. 68-02-4285, Work Assignment 1/022. February
28, 1989.
74. SenGupta, U., and W.G. Schuliger. Carbon Adsorption for Control of VOCs,
presented at AWMA meeting, Atlanta, GA. November 4-7, 1991.
75. Stenzel, M.H., and R.J. Bourdeau. Granular Activated Carbon Adsorption
with Air Stripping for Groundwater Treatement. Calgon Carbon Corp.,
!40
-------
Pittsburgh, PA. Undated. 13pp.
76. Stenzel, M.H., and U. SenGupta. APCA J. 35(12): 1304-1309. December
1985.
77. Byers, W.D. Env. Pray.. 7(1):17-21. 1988.
78. See Reference 74.
79. See Reference 65.
80. See Reference 77.
81. PEL Summary of Comparisons of Air Stripper Simulations and Field
Performance Data. U.S. EPA, Office of Air Quality Planning and Standards,
Research Triangle Park, NC, EPA/450-1-90-002, February 1990.
82. Vancil, M.A., R.H. Howie, D.J. Herndon, and S.A. Shareet. Air Stripper
Emissions and Controls. Final Report for U.S. EPA Contract 68-02-03816.
May8, 1987.
83. Schuliger, W.G. Controlling Low Concentrations of Volatile Organic
Compounds Using Granular Activated Carbon, presented at Bulk
Pharmaceutical Chemicals Spring Meeting, Kalamazoo, Ml.
April 24-26, 1983.
84. Urbanic, J.E., and W.d. Lovett. The Use of Activated Carbon for Control of
Paint Bake Oven Emissions, presented at the 67th APCA meeting, June
1974.
85. Lubozynski, FT., R.A. Ashworth, and R.W. Coutant. In Proc. 5th Nat. Conf.
On Haz. Wastes and Haz. Mails. Las Vagas, NV, Haz. Matl. Cont. Res. Inst.
(Silver Spring, MD), April 19-21, 1988. pp. 445-449.
86. See Reference 76.
87. Ayer, J., and C.D. Wolbach. Solvent Emissions Reduction Study and Newark
AFB, Ohio. U.S. Air Force Report ESL-TR-89-27 (Tyndall AFB, FL). May
1990.
88. See Reference 74.
89. See Reference 74.
90. See Reference 74.
91. SenGupta, U. Granular Activated Carbon-Thermal Regeneration Process for
Control of VOC Emissions from Surface Coating Operations, paper 88-84.4,
presented at 81st Annual APCA Meeting, Dallas, TX. June 19-24, 1988.
92. Telecon. C. W. Sanford, RTI, and Richard Kenson, Met-Pro KPR. April
12, 1994. Control devices.
141
-------
93. Kenson, R.E. KPR Systems for VOC Emission Control from Paint Spray
Booths, presented at 78th APCA meeting, Detroit, Ml. June 16-21, 1985.
94. See Reference 93.
95. See Reference 93.
96. See Reference 93.
97. Kenson, R.E., and J.F. Jackson. Paper FC88-566, Society of Manufacturing
Engineers, Deerborn, Ml. 1988.
98. See Reference 97.
99. Kenson, R.E., and H. Fernback. Market Study of Paint Spray Booth Solvent
Emission Control System, paper FC90-631, presented at Finishing West '90,
Anaheim, CA, Society Manufacturing Engineers (Dearborn, Ml).
September 25-27, 1990.
100. See Reference 99.
101. See Reference 91.
102. See Reference 91.
103. Telecon. C. W. Sanford, RTI, and Ram Gupta, Calgon Corp., May 19,
1994. Control devices.
104. See Reference 81.
105. NETAC (National Env. Tech. Appl. Corp., Pittsburgh, PA). Environmental
Product Profiles—CADRE™ VOC Control Process. July 1991.
106. See Reference 105.
107. Calgon. Letter from C. Thomas Calgon, Inc. (Charlotte, NC) to J. Spivey,
Research Triangle Institute including Sales and Installation Report
(May 1, 1990). 1991.
108. SenGupta, U. And W.G. Schuliger. Carbon Adsorption Technology for the
Control of Low Level Chlorinated Hydrocarbon Emissions, Calgon Corp.,
Pittsburgh, PA. Undated
109. See Reference 91.
110. Chemical Engineering, March 1991, p. 25.
111 Telecon. C, W. Sanford, RTI, and Sophia Block, Catalytica. April 14,
1994 Control devices,
112, See Reference 24.
113 Air Tech News, January 1993, p.2.
142
-------
114. Ruhl, J. Dedert Corporation (Olympia Fields, IL). Letter and accompanying
literature to J.J. Spivey, Research Triangle Institute. August 11, 1993.
115. See Reference 114.
116. Telecon. C. W. Sanford, RTI, and John Ruhl, Dedert. April 15, 1994.
Control devices.
117. Crompton, D., and A. Gupta. Removal of Air toxicis: A Comparison of the
Adsorption Characteristics of Activated Carbon andZeolotes, paper
93-TP-31B.06, presented at 86th Annual AWMA Meeting, Denver, CO.
June 13-18, 1993.
118. Blocki, S.W. Env. Prog. 12(3):226-260. August 1993.
119. Telecon. C. W. Sanford, RTI, and Jason Valia, DURR. June 27, 1994.
Incinerators.
120. Blankenship, Bill, Air Pollution Control Bureau, New Mexico Environment
Department, letter and attachments (draft Air Quality Permit No. 325-M-7), to
Angela Boggs, Intel Corporation. July 18, 1994.
121. See Reference 59.
122. Division of Air Quality, Minnesota Pollution Control Agency, letter and
attachments to C. W. Sanford, RTI. August 1994.
123. See Reference 59.
124. Haas, J.E. Kelco Group, Inc. (Raynham, MA). Letter and Attachments to
J.J. Spivey, Research Triangle Institute. August 10, 1993.
125. See Reference 70.
126. NETAC (National Env. Tech. Appl. Corp., Pittsburgh, PA). Environmental
Product Profiles-Procedair. July 1991.
127. Armand, B.L., H.B. Uddholm, and P.T. Vikstrom. Ind Eng. Chem. Res.
29:436-439. 1990.
128. Heisel, M.P., and A.E. Velloni. Gas Sep. Purif. 5:111-113. 1991.
129. Chemical Engineering, March 1989, p. 17-19.
130. Chemical Engineering, March 1989, p. 191.
131. Rafson, H.J.. Removing VOC's with a Mist Scrubber and Comparison to
Alternate Technologies, presented at AlChE Meeting, San Diego, CA. 1990.
132. Rafson, H.J. QUAD Environmental Technologies Corporation. Letter to RTI.
1991.
133. See Reference 131.
143
-------
134. See Reference 132
135. Rafson, Harold J., QUAD Technologies, Inc., Chicago, IL, letter and
attachments to C.W. Sanford, RTI, April 25, 1994.
136. Daniel, Laramie, Maryland Department of Environment, letter and
attachments to C. W. Sanford, RTI, August 12, 1994.
137. See Reference 132,
138. Ullinskey, J.D., D.M. Metts, and W.A. Ambrose. Air Toxics Considerations in
Wastewater Treatment Plant Design. Undated.
139. Dunson, J.B. Chemical Scrubbing for Control of Air Emissions, presented at
1991 AlChE Summer National Meeting, Pittsburgh, PA. August 1991.
140. Ultrox Report. Description of the Ultrox® D-Tox™ Process. Ultrox, Inc.,
Santa Ana, CA. 1991.
141. See Reference 9.
142. Jackson, T.E. Terr-Aqua Enviro Systems, Inc., private communications.
1991.
143. See Reference 9.
144. See Reference 9.
145. York Research Consultants. Report for South Coast Air Quality Management
District Volatile Organic Compound Compliance Test at General Dynamics,
Pomona Division. 1988.
146. See Reference 142.
147. Shugarman, Lynn, Terr-Aqua Enviro Systems, Inc., letter and attachments to
C.W. Sanford, RTI, May 20, 1994.
148. See Reference 147.
149. See Reference 147.
150. See Reference 145.
151 See Reference 147.
152, See Reference 145.
153, See Reference 147.
154. See Reference 147.
155, See Reference 147.
156, See Reference 147,
!44
-------
157. See Reference 65.
158. Chemical Engineering. 27. April 1992.
159. Air Tech News, September 1992. p. 122.
160. Air Pollution Consultant, March/April 1993. p. 1.14-1.17.
161. Jain, N. Brayton Cycle Solvent Recovery, presented at 85th Annual AWMA
Meeting, Kansas City MO. June 21-26, 1992.
162. Air Tech News, January 1993. p. 1-2.
163. Rick Martin, Thermatrix. Letter and attachments to C.W. Sanford, RTI.
April 20, 1994.
164. Woods, K.B., and U.T. Schofield. Control of Toxic Air Emissions with a
Flameless Thermal Oxidizer, paper 93-WP-94.06, presented at 86th Annual
AWMA Meeting, Denver, CO. June 13-18, 1993.
165. Telecon. C. W. Sanford, RTI, and R. Moreno, Alzeta. May 5, 1994.
Control Devices.
166. Bartz, D.F., and S.P. Barone. Ultra-High VOC Desturction with Low NOX and
CO in Adiabatic Radiant Combustors, presented at HAZMACON '92, Long
Beach, CA. March 30-April 2, 1992.
167. See Reference 1.
168. See Reference 1.
169. Yamamoto, T., K. Ramanathen, M.K. Owen, D.S. Ensor, and J.J. Spivey.
Corona Desturction and Other Technologies for VOC Control, report to U.S.
EPA:AEERL, Cooperative Agreement CR815169. November 1990.
170. U.S. EPA. Corona Desturction Program Peer Review. U.S. EPA:AEERL.
Septebmer 10, 1991.
171. Peral, J., and D.F. Ollis. Photocatalysis for Emission and Air Quality Control,
presented at 203rd ACS National Meeting, San Francisco, CA. April 1992.
172. See Reference 171.
145
-------
-------
APPENDIX A
ORGANIZATIONS CONTACTED FOR CONTROL TECHNOLOGIES FOR
GASES CONTAINING LESS THAN 100 ppm INLET OV
CONCENTRATION
-------
-------
APPENDIX A. ORGANIZATIONS CONTACTED FOR CONTROL TECHNOLOGIES FOR GASES
CONTAINING LESS THAN 100 ppm INLET OV CONCENTRATION
Organization
Person Contacted
Phone/FAX
Comments
Allied Signal Palatine, IL George Lester (708) 391-3314
Amcec Oak Brook, IL
Robert Saxer
Anguil Environmental Gene Anguil
Systems, Inc. Milwaukee, Wl
ARI Technologies, Inc.
Palatino, IL
Bay Area Quality
Management District San
Francisco, CA
Calgon
Charlotte, NC
New Jersey
Pittsburgh, PA
Orlando, FL
Carus Chemical Company
Ottawa, IL
Ed Dowd
Alex Saschin
Carol Thomas
Kim Freidman
Alan Roy
Utpal SenGupta
Nirmal Singh
(708) 954-1515
(708) 954-4077
(414) 332-0230
(414) 332-4375
(708) 359-7810
(708) 359-3700
(415) 749-4713
(704) 527-7580
(704) 523-3550
(908) 526-4646
(412) 787-6700
(412) 787-6713
(407) 567-1320
(815) 224-6818
(815) 433-9075
Allied only makes catalysts for OV oxidation; however, they do not
market complete oxidation systems for OV control.
Amcec provides both carbon adsorption systems and a hybrid
system for low concentrations (20-300 ppm).
120 oxidation systems in the market. Use noble metal catalysts
from Engelhard, Johnson-Matthey and Allied Signal. Some of these
system are operating on less than 100 ppm OV streams.
ARI has several catalytic oxidation systems in the market using
chromia-alumina-based catalyst. Radian has tested a low
concentration OV stream using ARI's pilot plant. ARI has also
developed an adsorption-catalytic oxidation system, especially
suited for low concentration OV stream.
Conducted tests using catalytic oxidation at higher OV
concentrations than are of interest here. A request for information
to possibly review low concentration sites has not been answered
as of December 1991.
Calgon provides nonregenerable, regenerable, and CADRE modified
adsorption systems. There are a number of installations, including
three CADRE installations, with inlet concentrations less than 100
ppm.
Cams only makes catalysts for OV oxidation and does not market
complete oxidation systems for OV control. Carus catalyst is used
by Anguil Environmental, Demptrol, and M&W Industries.
(continued)
-------
Organization
Person Contacted
Phone/FAX
Comments
to
Catalytica Mountain View, Thomas Duffy
CA
(415) 960-3000
(415) 960-0127
CSM Environmental Thomas Otchby (718)522-7000
Systems, inc. Brooklyn, NY Walt Taiboi (718) 852-1686
Dedert Topsoe
Oiympia Fields, IL
Demtroi Hartlano, W!
Engelhard Iselin, NJ
KSE, Inc. Amherst, MA
John Ruhl
Robert Hablewitz
Kenneth Burns
Jim Kittrell
McGill Environmental Paul Kennedy
Systems, Inc. Tulsa, OK
M & W Industries.. Inc. Rural Denny Clodfelter
Hall, NC
Met-Pro Harleysville, PA
Robert Kenson
(708) 747-7000
(708) 755-8815
t414) 367-7548
(414) 367-0831
(908) 205-6640
(908) 205-6146
(413) 549-5506
(918) 445-2431
(919) 969-9526
(215) 723-6751
Catalytica is developing an adsorption/catalytic oxidation system.
is currently in the developmental stage.
It
CSM has been working in ihe area of catalytic oxidation for the last
20 years, especially with low concentration OV contro!. They did
not provide any technical/cost information.
Subsidiary of Haldor Topsoe. Several catalytic systems in the
market; however, all are operating at high OV concentration. The
lowest concentration for which their system has been used is 1
g/m3 (corresponds to about 290 ppm for benzene).
Use Carus catalyst in their catalyst oxidation systems.
Engelhard only makes catalysts for OV oxidation and does not
market complete oxidation systems for OV control. Their catalyst is
used by SCM, Anguil, McGill Environmental, and Temprite
Industries.
KSE has developed their own oxidation catalyst. This catalyst has
been tested below 100 ppm and as low as 1 ppm. Some of these
data are proprietary and cannot be released. KSE constructs small
systems (100 to 200 scfm) in-house, larger systems are made by
other vendors using KSE technology.
Most of their systems are operating at high OV concentration.
Several oxidation units in the market designed for 4,000 to 30,000
scfm and operating at feed OV concentrations up to as high as 10
percent. Recently introduced a RE-GENSORB system consisting of a
carbon adsorption bed in series with a thermal oxidizer. This is
especially suited for low OV concentration stream.
Met-Pro provides adsorption/thermal incineration systems, most of
which are used for control of paint spray booths.
(continued)
-------
Organization
Person Contacted
Phone/FAX
Comments
MTR, Inc. San Francisco, CA Vicki Simmons
(415) 328-2228
Munters Zeol Amesbury, MA Jasper Gronvaldt (508) 388-2666
(508) 388-0292
Nichimen of America F. Kuma
Occidental Petroleum Jim Taylor
Ashtabula, OH
On-Demand Environmental Rick Hamilton
Systems San Jose, CA
Procedair Cedar Knolls, NJ
Purex Nassau County, NY Mark Whitney
QUAD Environmental Harold Rafson
Technologies Corporation
Northbrook, IL
QVF Glastechnik Weisbaden, H. Blanke
Germany
Reeco Morris Plains, NJ
Rod Pennington
(212) 719-1000
(212) 536-0549
(216) 992-3200
(408) 764-9104
(201) 455-8821
(516) 222-0955
(708) 564-5070
(708) 564-5606
(49) 611-2650
(49) 611-265108
(201) 538-8585
(201) 538-0407
MTR has developed a membrane-based process for OV control. This
technology is best suited for stream with OV concentration in the
0.5 to 20 percent range and is not economical when OV
concentration is less than 100 ppm.
Munters provides a rotary carousel for adsorption and several
options for downstream treatment. They have numerous
installations in Europe, the United States, and Japan.
Nichimen provides a rotary hydrophobic adsorber, but has not yet
responded to a request for information.
Occidental operates a CADRE system for groundwater remediation
via air stripping. Inlet concentrations are less than 100 ppm.
Most of their systems are used for high concentration OV and in-
batch operation. For such operations heat exchangers are not used
to recover the heat.
Procedair provides an absorption/stripping process which, in
principle, could be applied to low concentrations but, so far, has
not.
Purex operates a CADRE system for groundwater remediation via air
stripping. Inlet concentrations are less than 100 ppm.
QUAD makes an absorption-based control technology which
transfers contaminants from gas phase to the liquid phase. The
company does not provide any technology for treating the
contaminated liquid. Several existing commercial technologies can
be used for this purpose.
QVF provides an absorption/stripping process designed to meet
emission limits not possible with carbon adsorption.
Market regenerative thermal incinerators. Some of these systems
are being used for low concentration OV oxidation.
(continued)
-------
Organization
Person Contacted
Phone/FAX
Comments
Seibu Giken Fukuoka, Japan
(92) 947-4311
(92) 947-4314
Smith Engineering Systems John Kirkwood (714)923-3331
Ontario, CA Joe Steiwart
TEC Systems De Pere, Wi Richard Carman (414) 336-5715
U.S. Air Force Tyndall AFB, Capt. Ed Marchand (904) 283-6023
FL (904) 283-6499
j> Ultrox International Santa Jerry Barich
.p. Ana, CA Jack Zeff
(714) 545-5557
(714) 557-5396
VIC Nantucket, MA
Minneapolis,
Nate Shaw
Robert Cannon
(508) 228-3464
(508) 228-4293
(612) 781-6601
(612) 781-8559
Seibu Giken provides zeolite/inorganic adsorbent processes for
recovery of organic vapors. Contacts with the company have not
yet been answered and it is not known if any installations treating
low concentrations exist.
Smith has been in existence since 1925. Several hundred oxidation
units in the market; mostly thermal, some catalytic. Most of these
units operate above 100 ppm 0V concentration. Use noble metal
catalysts by Johnson-Matthey.
This is a division of W.R. Grace & Company. Most of their systems
are operating at 300 to 400 ppm inlet 0V concentration.
Wurtsmith AFB is using a catalytic oxidation unit to control OV
emissions from air strippers. The feed OV concentration is very low
(about 1 ppm). Capt. Marchand sent an interim report of
performance tests conducted on the catalytic unit at Eglin AFB on
an air stripper.
Ultrox makes a UV-catalytic oxidation system. Economically good
for low concentration OV streams. Only three small commercial
units in the market. Still working on
commercialization/development. Little hesitant to release any
information at this time because it has recently been bought by
$7 billion construction engineering company.
VIC provides regenerable carbon adsorption systems, but has no
installation with inlet concentrations below 100 ppm.
-------
APPENDIX B
COST TABLES
-------
-------
Cost tables are presented here for the following
technologies:
• catalytic incineration
• regenerative thermal incineration
• nonregenerable carbon adsorption
• regenerable fixed-bed carbon adsorption
• absorption/stripping.
Costs were developed using the methodology given in the
OAQPS Control Cost Manual (Reference 2); all costs are in 1991
dollars, unless otherwise noted in the table. All costs
presented here are calculated from factored estimates, with
the exception of absorption/stripping. Costs were developed
for four cases:
• 100 ppm benzene
• 10 ppm benzene
• 100 ppm tetrachloroethylene
• 10 ppm tetrachloroethylene.
All these cases are for continuous streams and
• OV in clean air
• 10,000 scfm
• 70 °F inlet temperature
• 70 percent relative humidity
• 70 percent heat recovery for the unit (where
appropriate)
• 95 percent destruction efficiency1
destruction/removal efficiency of 95 percent was
selected to represent the lower end of the range of control
efficiencies required for the control of organic vapors by EPA
regulations. In many cases, EPA requires higher control
efficiencies especially in those situations where incineration is
the technology serving as the basis of the standard. The
incineration-based technolgies discussed in this document have
demonstrated control efficiencies of 98 percent or higher and
therefore are applicable when a higher performance standard
(e.g., 98%) is required by regulation. Conducting the analysis
at 95 percent as opposed to 98 percent also can impact the
cost-effectiveness calculation because in many cases the
additional organics control can be achieved at little or no cost.
(continued...)
B-1
-------
• 8,000 h/yr operation.
For absorption/stripping, insufficient information was
available to distinguish any difference in capital or
operating costs among the four streams and therefore the total
annualized costs are independent of both the type of 0V and
concentration.
( ...continued)
Cost-effectiveness values would therefore be lower at this higher
control efficiency,
B-2
-------
TABLE B-1. TOTAL ANNUALIZED COSTS FOR CATALYTIC INCINERATION
FOR MODEL GAS STREAMS
(100 ppm BENZENE)
Cost item
Suggested factor
Cost
Cost/unit
Direct annualized
costs. DC
Op. Labor
Operator
Supervisor
Maintenance
Labor
Materials
Catalyst
replacement
Utilities
Natural gas
Electricity
Total DC
Indirect annualized
costs. 1C
Overhead
Admin.
Prop, taxes
Insurance
Capital recovery
Total 1C
Total annualized cost
(rounded)
Cost effectiveness
($/ton OV removed)
0.5 h/shift
15% op. labor
0.5 h/shift
100%maint. labor
100% replacement
every 2 years
12.96
14.26
650
3.30
0.06
$/h
$/h
S/kft3
S/kWh
60% of sum of
op., supv., and
maintenance labor
and maintenance
materials
2% TCI
1%TCI
1%TCI
(10%/10 years, or
16.275% of TCI)
4,630
695
5,090
5,090
13,000
135,921
13,104
177,530
9,303
8,394
4,197
4,197
68,303
94,393
271,923
5,489
B-3
-------
TABLE B-2. TOTAL ANNUALIZED COSTS FOR CATALYTIC INCINERATION
FOR MODEL GAS STREAMS
(10 ppm BENZENE)
Cost item
Suggested factor
Cost
Cost/unit
Direct annualized
costs. DC
Op. Labor
Operator
Supervisor
Maintenance
Labor
Materials
Catalyst
replacement
Utilities
Natural gas
Electricity
Total DC
Indirect annualized
costs. 1C
Overhead
Admin.
Prop, taxes
Insurance
Capital recovery
Total 1C
Total annualized cosi
(rounded)
Cost effectiveness
($/ton OV removed)
0.5 h/shift
15% op. labor
0.5 h/shift
100%maint. labor
100% replacement
every 2 years
12.96
14.26
650
3.30
0.06
$/h
$/h
S/kft3
$/kWh
60% of sum of
op., supv., and
maintenance labor
and maintenance
materials
2% TCI
1%TCI
1%TCI
(10%/10 years, or
16.275% of TCI)
4,630
695
5,090
5,090
13,000
138,374
13,104
179,983
9,303
8,394
4,197
4,197
68,303
94,393
274,376
55,381
8-4
-------
TABLE B-3. TOTAL ANNUALIZED COSTS FOR CATALYTIC INCINERATION
FOR MODEL GAS STREAMS
(100 ppm TETRACHLOROETHYLENE)
Cost item
Suggested factor
Cost Cost/unit
Direct annualized
costs. DC
Op. Labor
Operator
Supervisor
Maintenance
Labor
Materials
Catalyst
replacement
Utilities
Natural gas
Electricity
Total DC
Indirect annualized
costs. 1C
Overhead
Admin.
Prop, taxes
Insurance
Capital recovery
Total 1C
Total annualized cost
(rounded)
Cost effectiveness
($/ton OV removed)
0.5 h/shift
15% op. labor
0.5 h/shift
100% maint. labor
100% replacement
every 2 years
12.96
14.26
650
3.30
0.06
$/h
$/h
S/ft3
S/kft3
$/kWh
60% of sum of
op., supv., and
maintenance labor
and maintenance
materials
2% TCI
1%TCI
1%TCI
(10%/10 years, or
16.275% of TCI)
4,630
695
5,090
5,090
19,500
148,204
13,104
196,312
9,303
8,394
4,197
4,197
68,303
94,393
290,705
5,868
B-5
-------
TABLE B-4, TOTAL ANNUALIZED COSTS FOR CATALYTIC INCINERATION
FOR MODEL GAS STREAMS
(10 ppm TETRACHLOROETHYLENE)
Cost item
Suggested factor
Cost
Cost/unit
Direct annualized
costs. DC
Op. Labor
Operator
Supervisor
Maintenance
Labor
Materials
Catalyst
replacement
Utilities
Natural gas
Electricity
Total DC
Indirect annualized
costs. 1C
Overhead
Admin
Prop.taxes
Insurance
Capital recovery
Total 1C
Total annualized cost
(rounded)
Cost effectiveness
($/ton OV removed)
0.5 h/shift
15% op. labor
0.5 h/shift
100%maint. labor
100% replacement
every 2 years
60% of sum of
op., supv., and
maintenance labor
and maintenance
materials
2% TCI
1%TCI
1%TCI
(10%/10 years, or
16.275% of TCI)
12.96
14.26
650
3.30
0.06
$/h
$/h
$/fP
S/kft3
$/kWh
4,630
695
5,090
5,090
19,500
148,960
13,104
197,069
9,303
8,394
4,197
4,197
68,303
94,393
291,462
58,829
B-6
-------
TABLE B-5. TOTAL ANNUALIZED COSTS FOR REGENERATIVE THERMAL
INCINERATION FOR MODEL GAS STREAMS
(100 ppm BENZENE)*
Cost item
Suggested factor
Cost
Cost/unit
Direct annualized
costs. DC
Op. Labor
Operator
Supervisor
Maintenance
Labor
Materials
Utilities
Fuel"
Electricity"
Total DC
Indirect annualized
costs. 1C
Overhead
Admin.
Prop, taxes
Insurance
Capital recovery
Total 1C
Total annualized cost
(rounded)
Cost effectiveness
($/ton OV removed)
0.5 h/shift
15% op. labor
0.5 h/shift
100% maint. labor
60% of sum of
op., supv., and
maintenance labor
and maintenance
materials
2% TCIe
1 % TCI
1%TCI
(10%/10 years, or
16.275% of TCI)
12.96 $/h
14.26 $/h
6.0C $/h
3.36C $/h
4,630
695
5,090
5,090
48,000
26,880
90,385
9,303
8,020
4,010
4,010
65,263
90,606
180,991
3,653
8 Thermal energy recovery is 95 percent for benzene.
b Based on $4/106Btu.
c Fuel and electrical costs provided by vendor (Pennington, 1991) in units of $/h. Costs based
on 8,000 h/yr operation.
" Based on $0.06/kWh.
9 TCI given by vendor (Pennington, 1991) as $401,000.
B-7
-------
TABLE B 6. TOTAL ANNUALIZED COSTS FOR REGENERATIVE THERMAL
INCINERATION FOR MODEL GAS STREAMS
(10 ppm BENZENE)1
Cost item
Suggested factor
Cost Cost/unit
Direct annualized
costs. DC
Op. Labor
Operator
Supervisor
Maintenance
Labor
Materials
Utilities
Fuelb
Electricity"*
Total DC
Indirect annualized
costs. 1C
Overhead
Admin.
Prop, taxes
Insurance
Capital recovery
Total 1C
Total annualized cost
(rounded)
Cost effectiveness
($/ton OV removed)
0.5 h/shift
15% op. labor
0.5 h/shift
100%maint. labor
60% of sum of
op., supv., and
maintenance labor
and maintenance
materials
2% TCIe
1%TCI
1%TCI
(10%/10 years, or
16.275% of TCI)
12.96 $/h
14.26 $/h
7.0° $/h
3.36C $/h
4,630
695
5,090
5,090
56,000
26,880
98,385
9,303
8,020
4,010
4,010
65,263
90,606
180,991
38,147
a Thermal energy recovery is 95 percent for benzene.
b Based on $4/10s Btu.
c Fuel and electrical costs provided by vendor (Pennington, 1991) in units of $/h Costs based
on 8,000 h/yr operation.
" Based on $0.06/kWh.
9 TCI given by vendor (Pennington, 199!) -ss $401,000.
B-8
-------
TABLE B-7. TOTAL ANNUALIZED COSTS FOR REGENERATIVE THERMAL
INCINERATION FOR MODEL GAS STREAMS
(100 ppm TETRACHLOROETHYLENE)'
Cost item
Suggested factor
Cost Cost/unit
Direct annualized
costs. DC
Op. Labor
Operator
Supervisor
Maintenance
Labor
Materials
Utilities
Fuelb
Electricity01
Total DC
Indirect annualized
costs. 1C
Overhead
Admin.
Prop, taxes
Insurance
Capital recovery
Total 1C
Total annualized cost
(rounded)
Cost effectiveness
($/ton OV removed)
0.5 h/shift
15% op. labor
0.5 h/shift
100% maint. labor
12.96 $/h
14.26 $/h
13.0C $/h
2.22° $/h
60% of sum of
op., supv., and
maintenance labor
and maintenance
materials
2% TCI8
1%TCI
1%TCI
(10%/10 years, or
16.275% of TCI)
4,630
695
5,090
5,090
104,000
17,760
137,265
9,303
9,800
4,900
4,900
79,748
108,651
245,916
4,964
a Thermal energy recovery is 88 percent. This is lower than for benzene in order to maintain the
exhaust gas above 300 °F to minimize HCI condensation and subsequent corrosion problems.
b Based on $4/106 Btu.
c Fuel and electrical costs provided by vendor (Pennington, 1991) in units of $/h. Costs based
on 8,000 h/yr operation.
" Based on $0.06/kWh.
e Given by vendor (Pennington, 1991) as $490,000.
B-9
-------
TABLE B-8. TOTAL ANNUALIZED COSTS FOR REGENERATIVE THERMAL
INCINERATION FOR MODEL GAS STREAMS
(10 ppm TETRACHLOROETHYLENE)*
Cost item
Admin.
Prop, taxes
Insurance
Capital recovery
Total 1C
Total armualized cost
(rounded)
Cost effectiveness
($/ton OV removed)
Suggested factor
Cost
Cost/unit
Direct annualized
costs. DC
Op. Labor
Operator
Supervisor
Maintenance
Labor
Materials
Utilities
Fuel"
Electricity"
Total DC
Indirect annualized
costs. 1C
Overhead
0.5 h/shift 12.96 $/h
15% op. labor -
0.5 h/shift 14.26 $/h
100% maint. labor
13.8° $/h
2.2C $/h
60% of sum of
4,630
695
5,090
5,090
110,400
17,760
143,665
9,303
op., supv., and
maintenance labor
and maintenance
materials
2% TCI"
1%TCI
1%TCI
(10%/10 years, or
16.275% of TCI)
9,800
4,900
4,900
79,748
108,651
252,316
50,928
a Thermal energy recovery is 88 percent. This is lower than for benzene in order to maintain the
exhaust gas above 300 "F to minimize HCI condensation and subsequent corrosion problems.
b Based on $4/106Btu.
c Fuel and electrical costs provided by vendor (Pennington, 1991) in units of $/h. Costs based
on 8,000 h/yr operation.
" Based on $0.06/kWh.
6 Given by vendor (Pennington, 1991) as $490,000.
B-10
-------
TABLE B-9. TOTAL ANNUALIZED COSTS FOR NONREGENERABLE
CARBON ADSORPTION FOR MODEL GAS STREAMS
(100 ppm BENZENE)
Cost item
Suggested factor
Cost
Cost/unit
Direct annualized
costs. DC
Op. Labor
Operator
Supervisor
Maintenance
Labor
Materials
Carbon regeneration
Electricity
0.5 h/shift
15% op. labor
0.5 h/shift
100% maint. labor
1.5E6lb/yr
13.5kW
12.96
14.26
$/h
$/h
0.80 $/lb
0.06 $/kWh
4,630
695
5,090
5,090
1,233,800
6,500
Total DC
Indirect annualized
costs. 1C
Overhead
Admin.
Prop, taxes
Insurance
Capital recovery
Total 1C
Total annualized cost
(rounded)
Cost effectiveness
($/ton OV removed)
60% of sum of
op., supv., and
maintenance labor
and maintenance
materials
2% TCI
1%TCI
1%TCI
(10%/10 years, or
16.275% of TCI)
1,255,805
9,303
1,952
976
976
15,884
29,091
1,284,896
25,935
Based on 7 in. H2O pressure drop for fan. Electricity for the fan is the only cost accounted for here.
B-11
-------
TABLE B-10. TOTAL ANNUALIZED COSTS FOR NONREGENERABLE
CARBON ADSORPTION FOR MODEL GAS STREAMS
(10 ppm BENZENE)
Cost item
Suggested factor
Cost
Cost/unit
Direct annualized
costs. DC
Op. Labor
Operator
Supervisor
Maintenance
Labor
Materials
Carbon regeneration
Electricity
0.5 h/shift
15% op. labor
0.5 h/shift
100% maint. labor
2.5 E5 Ib/yr
5.6 kW
12.96
14.26
0.80
0.06
$/h
$/h
$/lb
$/kWh
4,630
695
5,090
5,090
197,000
2,700
Total DC
Indirect annualized
costs. 1C
Overhead
60% of sum of
op., supv., and
maintenance labor
and maintenance
materials
215,205
9,303
Admin.
Prop, taxes
Insurance
Capital recovery
Total 1C
Total annualized cost
(rounded)
Cost effectiveness
($/ton OV removed)
2% TCI
1%TCI
1%TCI
(10%/10 years, or
16.275% of TCI)
1,630
815
815
13,264
25,827
241,031
48,650
a Based on 2.9 in H2O pressure drop for fan. Electricity for the fan is the only cost accounted for
hers.
8-12
-------
TABLE B-11. TOTAL ANNUALIZED COSTS FOR NONREGENERABLE
CARBON ADSORPTION FOR MODEL GAS STREAMS
(100 ppm TETRACHLOROETHYLENE)
Cost item
Suggested factor
Cost
Cost/unit
Direct annualized
costs. DC
Op. Labor
Operator
Supervisor
Maintenance
Labor
Materials
Carbon regeneration
Electricity
0.5 h/shift
15% op. labor
0.5 h/shift
100% maint. labor
1.2E6lb/yr
10.2 kW
12.96
14.26
0.80
0.06
$/h
$/h
$/lb
$/kWh
4,630
695
5,090
5,090
942,000
5,400
Total DC
Indirect annualized
costs. 1C
Overhead
Admin.
Prop, taxes
Insurance
Capital recovery
Total 1C
Total annualized cost
(rounded)
Cost effectiveness
($/ton OV removed)
60% of sum of
op., supv., and
maintenance labor
and maintenance
materials
2% TCI
1%TCI
1%TCI
(10%/10 years, or
16.275% of TCI)
962,905
9,303
1,856
928
928
15,103
28,118
991,022
20,003
• Based on 5.8 in. H2O pressure drop for fan. Electricity for the fan is the only cost accounted for
here.
B-13
-------
TABLE B-12. TOTAL ANNUALIZED COSTS FOR NONREGENERABLE
CARBON ADSORPTION FOR MODEL GAS STREAMS
(10 ppm TETRACHLOROETHYLENE)
Cost item
Suggested factor
Cost
Cost/unit
Direct annualized
costs. DC
Op. Labor
Operator
Supervisor
Maintenance
Labor
Materials
Carbon regeneration
Electricity
0.5 h/shift
15% op. labor
0.5 h/shift
100% maint. labor
1.6E5lb/yr
5.2 kW
12.96
14.26
$/h
$/h
0.80 $/lb
0.06 $/kWh
4,630
695
5,090
5,090
128,300
2,500
Total DC
Indirect annualized
costs. 1C
Overhead
Admin.
Prop, taxes
Insurance
Capital recovery
Total 1C
Total annualized cost
(rounded)
Cost effectiveness
($/ton OV removed)
60% of sum of
op., supv., and
maintenance labor
and maintenance
materials
2% TCI
1%TCI
1%TCI
(10%/10 years, or
16.275% of TCI)
146,305
9,303
1,612
806
9806
13,118
25,644
171,949
34,707
Based on 2.7 in H;O pressure drop for fan. Electricity for the fan is the only cost accounted for
here
B-14
-------
TABLE B-13. TOTAL ANNUALIZED COSTS FOR REGENERABLE FIXED BED
CARBON ADSORPTION FOR MODEL GAS STREAMS
(100 ppm BENZENE)
Cost item
Suggested factor
Cost Cost/unit
Direct annualized
costs. DC
Op. Labor
Operator
Supervisor
Maintenance
Labor
Materials
Carbon regeneration
(Steam + cooling
water)
Carbon replacement
Electricity
0.5 h/shift
15% op. labor
0.5 h/shift
100% maint. labor
1.56E6lb/yr
5 yr carbon bed life.
5% loss in regen.
13.5 kW
12.96
14.26
8.23
2
0.06
$/h
$/h
$/1000lb
S/lb
$/kWh
4,630
695
5,090
5,090
12,700
17,600
6,500
Total DC
Indirect annualized
costs. 1C
Overhead
60% of sum of
op., supv., and
maintenance labor
and maintenance
materials
52,305
9,303
Admin.
Prop, taxes
Insurance
Capital recovery
Total 1C
Total annualized cost
(rounded)
Cost effectiveness
($/ton OV removed)
2% TCI
1%TCI
1%TCI
(10%/10 years, or
16.275% of TCI)
3,676
1,838
1,838
29,913
46,568
98,873
1,996
• Based on 7.0 in. H2O pressure drop for fan. Electricity for the fan is the only cost accounted for
here.
B-15
-------
TABLE B-14. TOTAL ANNUALIZED COSTS FOR REGENERABLE FIXED BED
CARBON ADSORPTION FOR MODEL GAS STREAMS
(10 ppm BENZENE)
Cost item
Direct annualized
costs. DC
Op. Labor
Operator
Supervisor
Maintenance
Labor
Materials
Carbon regeneration
Suggested factor
0.5 h/shift
15% op. labor
0.5 h/shift
100%maint. labor
2.5 E5 !b/yr
Cost
12.96
14.26
8.23
Cost/unit
$/h
$/h
$/1000lb
4,630
695
5,090
5,090
2,000
(Steam + cooling
water)
Carbon replacement
Electricity
5 yr carbon bed life.
5% loss in regen.
5.6 kW
2
0.06
$/lb
$/kWh
2,800
2,700
Total DC
Indirect annualized
costs. 1C
Overhead
60% of sum of
op., supv., and
maintenance labor
and maintenance
materials
23,005
9,303
Admin.
Prop, taxes
Insurance
Capital recovery
Total 1C
Total annualized cost
(rounded)
Cost effectiveness
($/ton OV removed)
2% TCI
1%TCI
1%TC!
(10%/10 years, or
i 6.275% of TCI)
2,840
1,420
1,420
23,111
38,093
61,098
12,332
8 Based on 2.9 in H2O pressure drop for fan. Electricity for the fan is the only cost accounted for
here.
8-16
-------
TABLE B-15. TOTAL ANNUALIZED COSTS FOR REGENERABLE FIXED BED
CARBON ADSORPTION FOR MODEL GAS STREAMS
(100 ppm TETRACHLOROETHYLENE)
Cost item
Suggested factor
Cost Cost/unit
Direct annualized
costs. DC
Op. Labor
Operator
Supervisor
Maintenance
Labor
Materials
Carbon regeneration
(Steam + cooling
water)
Carbon replacement
Electricity
0.5 h/shift
15% op. labor
0.5 h/shift
100%maint. labor
1.2E6lb/yr
5 yr carbon bed life.
5% loss in regen.
10.2kW
12.96
14.26
8.23
2
0.06
$/h
$/h
$/1000lb
$/lb
$/kWh
4,630
695
5,090
5,090
9,700
13,500
5,400
Total DC
Indirect annualized
costs. 1C
Overhead
60% of sum of
op., supv., and
maintenance labor
and maintenance
materials
44,105
9,303
Admin.
Prop, taxes
Insurance
Capital recovery
Total 1C
Total annualized cost
(rounded)
Cost effectiveness
($/ton OV removed)
2% TCI
1%TCI
1%TCI
(10%/10 years, or
16.275% of TCI)
3,432
1,716
1,716
27,928
44,095
88,199
1,780
* Based on 5.8 in. H2O pressure drop for fan. Electricity for the fan is the only cost accounted for
here.
B-17
-------
TABLE B-16. TOTAL ANNUALIZED COSTS FOR REGENERABLE FIXED BED
CARBON ADSORPTION FOR MODEL GAS STREAMS
(10 ppm TETRACHLOROETHYLENE)
Cost item
Suggested factor
Cost Cost/unit
Direct annualized
costs. DC
Op. Labor
Operator
Supervisor
Maintenance
Labor
Materials
Carbon regeneration
(Steam + cooling
water)
Carbon replacement
Electricity
0.5 h/shift
15% op labor
0.5 h/shift
100% maint. labor
1.65E5lb/yr
5 yr carbon bed life.
5% loss in regen.
5.2 kW
12.96
14.26
8.23
2
0.06
$/h
$/h
$/1000lb
$/lb
$/kWh
4,630
695
5,090
5,090
1,300
1,800
2,500
Total DC
Indirect annualized
costs. 1C .
Overhead
60% of sum of
op., supv., and
maintenance labor
and maintenance
materials
21,105
9,303
Admin.
Prop, taxes
Insurance
Capital recovery
Total 1C
Total annualized cost
(rounded)
Cost effectiveness
($/ton OV removed)
2% TCI
1%TCI
1%TCI
( 1 0%/1 0 years, or
16.275% of TCI)
2,792
1,396
1,396
22,720
47,607
58,711
11,850
Based on 2.7 in. H2O pressure drop for fan. Electricity for the fan is the only cost accounted for
here.
B-18
-------
TABLE B-17. TOTAL ANNUALIZED COSTS FOR QVF ABSORPTION PROCESS
FOR MODEL GAS STREAMS
Cost item
Suggested factor
Cost
Cost/unit
Direct annualized
costs. DC
Op. Labor
Operator
Supervisor
Maintenance
Labor
Materials
Utilities
§ oolirig water
lectricity
Total DC
Indirect annualized
costs. 1C
Overhead
Admin.
Prop, taxes
Insurance
Capital recovery
Total 1C
Total annualized cost6
(rounded)
Cost effectiveness
($/ton OV removed)
100 ppm benzene
10 ppm benzene
100 ppm tetrachloroethylene
10 ppm tetrachloroethylene
0.5 h/shift
15% op. labor
0.5 h/shift
100% maint. labor
12.96
14.26
$/h
$/h
$/1,000 ft3
$/kWh
60% of sum of
op., supv., and
maintenance labor
and maintenance
materials
2% TCI"
1%TCI
1%TCI
(10%/10 years, or
16.275% of TCI)
4,630
695
5,090
5,090
960
8,826
25,291
9,303
30,200
15,100
15,100
245,753
306,153
331,400
1,996
6,700
66,900
3,300
32,800
• TCI is taken as $1.51 million as discussed in Section 5.0.
b TAG is the same for all model gas streams.
B-19
-------
-------
Attachment C
Summary Table of Control Devices Installed on High Flow, Low
Concentration Organic Vapor Streams in the U.S.
-------
-------
APPENDIX C
SUMMARY TABLE OF CONTROL DEVICES INSTALLED ON HIGH
FLOW, LOW CONCENTRATION ORGANIC VAPOR STREAMS
-------
-------
Appendix C. Summary Table of Control Devices Installed on High Flow, Low Concentration Organic Vapor Streams in the U.S.
User
Louisiana-Pacific
Louisiana-Pacific
Louisiana-Pacific
Commonwealth
Aluminum
Louisiana-Pacific
Louisiana-Pacific
Louisiana-Pacific
Valley Protein,
Inc.
Ford
Looitlana-Pacfflc
Ford
Vendor
Smith
Smith
Smith
OURR
EC&C
S II nttl
Srnitn
SlTntn
QUAD
DURR
Smith
REECO
DURR
Device
Regenerative
oxidizer
Ragenaratlva
oxkHzar
Regenerative
oxldlzer
Concentrator
end oxldlzer
Epcon
EcoBAC
Regenerative
oxldlzer
Regenerative
oxldlzer
Regenerative
oxldlzer
Mist scrubber
Concentrator
end oxkNzer
Regenerative
oxldlzar
Regenerative
oxldlzer
Concentrator
and oiNflzer
Row
(scfm)
120.000
120.000
120.000
240.000
90.OOO
120.000
120.000
80,000
75.000
400.000
80.0OO
220.00O
800.000
Inlet
(ppmv)
100-300
100-300
100-300
100
10
1OO-30O
100-300
100-300
75
100-300
100
Industry
OSB
OSB
OSB
Automobile
Aluminum
manufacture
OSB
OSB
OSB
Rendering
Automobile
Waferboard
Tape coating
and laminating
Automobile
Pollutant*
Formaldehyde
MDI & VOC
Mix
Formaldehyde
MDI & VOC
Mix
Formaldehyde
MDI S. VOC
Mix
Paint
Kerosene.
Mineral seal oil
Formaldehyde
ft VOC
Formaldehyde
& VOC
Formaldehyde
& VOC
Odor
Paint
VOC Mix
Solvent*
Paint
City
Hanceville
Hanceville
Hanceville
Fremont
Lewlsport
Urania
Urania
Urania
Baltimore
Wlxom
Two Harbors
St. Paul
Minneapolis
State
AL
AL
AL
CA
KY
LA
LA
LA
MD
Ml
MN
MN
MN
Efficiency
99 + IT)
95%IP)
99-KT)
95%(P>
99 + IT)
95%(PI
98-99%(RI
96%(T)
NO
NO
NO
97%(R)
ND
95%IT)
97X|TI
Permit
ye«
yes
yes
yes
yes
yea
Source
Test
yet
yes
yes
yes
yes
yes
yes
yes
yes
yes
Permitting
Contact
Glen Golson
(205) 271-7700
Glen Golson
(205) 271-7700
Glen Golson
(2051 271-7700
Carol Lee
(4151 749-4689
Jerry Goble (5021
573-3382
Barbara
WHIamson
(504) 765-0219
Barbara
Wlllamson
(504) 765-0219
Barbara
Wlllamson
(504) 765-0219
Laramle Daniel
fax:
(410) 631-3202
Tom Julian
(517) 373-7023
Stuart Arkly
(612) 296-7331
Stuart Arkly
(612)296-7331
Stuart Arkly
(61?) 296 7331
o
-------
Appendix C, Summary Table of Control Devices (continued)
User
^
•
Fo>d
General Motor*
Intel
Inland Product*
Qlldden
Ford
Toyote
Letterkenny
Army Depot/ABB
Paint
Saturn
GM
LTV
Vendor
REECO
DURR
Munter*
Zeol
DURR
QUAD
REECO
DURR
DURR
Munters
Zeol
Calgon
CADRE
REECO
Met-Pro
KPR
Device
Regenerative
oxldlzer
Concentrator
and oxidlzer
Concentrator
Concentrator
and oxldlier
Mi*t scrubber
Regenerative
oxldlzer
Concentrator
and oxldlzer
Concentrator
and oxtdlzer
Concentrator
Activated
carbon
Regenerative
oxldlzer
Rotary
ad*orber
Flow
(tcfm)
150,000
250,000
140.000
136,000
70,000
145.000
350.000
280.000
135.000
320.000
500.000
105.000
Inlet
(ppmv)
75-80
90
60
100
100
24
Industry
Printing and
packaging
Automobile
Automobile
Semi-
conductor
Rendering
Pelnt
manufacture
AutomobHe
Automobile
Reflnlshlng
Automobile
Automobile
Aerospace
Pollutants
Solvents
Paint
Paint
Solvent*
Odor
Paint
Paint
Paint
Paint/Solvents
Paint
Paint
Paint
City
Edison
Linden
Rio Rancho
Avon Lake
Georgetown
Chambersburg
Spring Hill
Arlington
Ft. Worth?
State
NC
NJ
NJ
NM
OH
OH
OH
OH
PA
TN
TX
TX?
Efficiency
97%(R)
95%IR)
90% + (P)
96%(R|
98%(R)
93%(P)
95%(R)
Permit
ye*
Source
Test
Permitting
Contact
Laura Butler
(919) 733-3340
Patrick Zidran
(609) 292-6704
Patrick Zidran
(609) 292-6704
Lawrence Alare*
(505) 827-2850
Sara Gary
(614) 644-2270
Sara Gary
(614) 644-2270
Sara Gary
(614) 644-2270
Sara Gary
(614) 644-227O
Rob Fisher
(7171 787-9256
Lacy Hardln
(616) 532-6645
Mike CokJiron
(512) 239-1260
Mike Coldlron
(512) 239-1260
o
S3
OSB = Oriented strand board
P = Required by permit condition
ND = Not determined or not reported
T = Documented in test results
R = Reported by vendor
-------
TECHNICAL REPORT DATA
(Please read Instructions on reverse before completing)
1 REPORT NO
EPA-456/R-95-003
3. RECIPIENT'S ACCESSION NO.
4 TITLE AND SUBTITLE
Survey of Control Technologies for Low Concentration
Organic Vapor Gas Streams
5. REPORT DATE
May 1995
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S) R. Zerbonia, J. Spivey, S. Agarwal, A. Damle, and
C. Sanford, Research Triangle Institute
Research Triangle Institute, NC 27709
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
10. PROGRAM ELEMENT NO.
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards (OAQPS)
Information Transfer & Program Integration Div. (ITPID)
Research Triangle Park, NC 27711
11. CONTRACT/GRANT NO.
Contract No. 68D10118
Work Assignment 114
12. SPONSORING AGENCY NAME AND ADDRESS
Director
Office of Air Quality Planning and Standards (OAQPS)
Office of Air and Radiation (OAR)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA/200/04
15. SUPPLEMENTARY NOTES This report was prepared by EPA's Control Technology Center which is jointly
sponsored by OAQPS/OAR and the Air Pollution Prevention and Control Division, National Risk
Management Research Laboratory, Office of Research and Development. The EPA project lead was
Robert J. Blaszczak (MD-12), CTC/ITPID/OAQPS, RTF, NC 27711 (919/541-5432).
16 ABSTRACT This document presents the results of a series of studies conducted to identify commercially
available control technologies suitable for application to low organic concentration gas streams at high
flow rates. For this report, low concentration means 100 ppm or below, and high flows are assumed to
be those above 100,000 cfm.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b. IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Catalytic Incineration, Regenerative Thermal
Incineration, Adsorption, Absorption, Enhanced
Carbon Adsorption, Condensation, Flameless
Thermal Oxidation, Biofiltration, Corona
Discharge, Heterogeneous Photocatalysis.
Air Pollution control
18. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (Report)
Unclassified
21. NO. OF PAGES
20. SECURITY CLASS (Page)
Unclassified
22. PRICE
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION IS OBSOLETE
------- |