EPA-600/R-92-201
October 1992
ALTERNATE VOC CONTROL TECHNIQUE OPTIONS FOR
SMALL ROTOGRAVURE AND FLEXOGRAPHY FACILITIES
David A. Green
and
Coieen M. Northeim
Center for Environmental Analysis
Research Triangle Institute
P.O. Box 12194
Research Triangle Park, North Carolina 27709
EPA Contract No. 68-D1-0118
Work Assignment 5225-016/D
Project Officer.
Jamie K. Whitfield
U. S. Environmental Protection Agency
Air and Energy Engineering Research Laboratory
Research Triangle Park, North Carolina 27711
Prepared for:
Control Technology Center
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before complet'
REPORT NO. 2.
EPA-600/R-92-201
a. P E9 3- 1223 07
4. TITLE AND SUBTITLE
Alternate VOC Control Technique Options for Small
Rotogravure and Flexography Facilities
5. REPORT DATE
October 1992
6. PERFORMING ORGANIZATION CODE
7. AUTHORES)
David A, Green and Coleen M. Northeim
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Research Triangle Institute
P. 0, Box 12194
Research Triangle Park, North Carolina 27709
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO,
68-D1-0118, Task 5225-016D
12. SPONSORING AGENCY NAME AND AODRESS
EPA, Office of Research and Development
Air and Energy Engineering Research Laboratory
Research Triangle Park, North Carolina 27711
13. TYPE OF REPORT AND PERIOD COVERED
Task Final; 2-6/92
14. SPONSORING AGENCY CODE
EPA/600/13
is. supplementary notes ^eeRL project officer is Jamie K. Whitfield, Mail Drop 61, 919/
541-2509.
16. abstract The report identifies Available Control Techniques (ACTs) for states to use
as a reference when implementing Reasonable Available Control Technology (RA CT)
for graphic arts facilities that are covered by the Control Technologies Guidelines
(CTGs), but emit less than 91 tonnes of volatile organic compounds (VCCs) per year.
The CTGs for the graphic arts industry was published in December 1978. It defined
RACT for VOCs emitted from publication and packaging rotogravure and from pack-
aging flexography. Subsequent EPA guidance limited the applicability of RACT re-
quirements to sources that emit 91 tonnes/yr or more of VOCs. The Clean Air Act
Amendments of 1990 (CAAAs) now require RACT for VOC sources that emit as little
as 9 tonnes/yr in extreme ozone nonattainment areas. Therefore, states are now
required to establish and implement RACT for these smaller sources as well.
17. KEY WORDS AND DOCUMENT ANALYSIS
a, DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Pollution
Flexography
Printing
Organic Compounds
Volatility
Graphic Arts
Pollution Control
Stationary Sources
Rotogravure
Volatile Organic Com-
pounds (VOCs)
13 B
14 E
07C
20 M
05F
18. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
20. SECURITY CLASS (This page}
Unclassified
22. PRICE"
EPA Form 2220-1 (9-73) *
i
-------
EPA REVIEW NOTICE
This report has been reviewed by the U.S. Environmental Protection Agency, and
approved for publication. Approval does not signify that the contents necessarily
reflect the views and policy of the Agency, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
i!
-------
PREFACE
Environmental Protection Agency's (EPA) Control Technology Center (CTC) was established by
EPA's Office of Research and Development and Office of Air Quality Planning and Standards to
provide technical assistance to State and local air pollution control agencies. Three levels of
assistance are available through the CTC. First, a CTC HOTLINE has been established to
provide telephone assistance on matters relating to air pollution control technology. Second,
more in-depth engineering assistance can be provided when appropriate. Third, the CTC can
provide technical guidance through publication of technical guidance documents, development
of personal computer software, and presentation of workshops on control technology matters.
Engineering assistance projects, such as this one, focus on topics of national and regional interest
that are identified through contact with State and local agencies. The CTC has received
numerous calls from State agencies concerning control technologies for graphic arts sources.
This study was undertaken to describe control options for small rotogravure and flexographic
printing facilities.
iii
-------
TABLE OF CONTENTS
Section
Page
Preface iii
Figures and Tables v
Acknowledgement vi
Metric Conversions vi i
1.0 Summary . .
. 1
2.0 Introduction.
3.0 General Description 3
3.1 Description of the Rotogravure Process 4
3.2 Description of Flexography 6
4.0
5.0
7.0
Capture Technologies .
4.1 Total Enclosure .
4.2 Capture Devices
Control Technologies and Pollution Prevention Options .
5.1 Carbon Adsorption
5.2 Thermal Incineration
5.3 Catalytic Incineration
5.4 Water-Borne Inks
5.5 Other Types of Inks
6.0 Description of Model Plants
Costs
7.1
7.2
7.3
7.4
7.5
Costs of Converting to Water-Borne Inks
Total Enclosures for Existing Facilities .
Costs of Thermal Incinerators
Costs of Catalytic Incinerator
Costs of Carbon Adsorption Systems . .
6
8
12
14
15
18
19
21
21
22
22
26
31
References
36
IV
-------
FIGURES AND TABLES
Figure Page
1 Rotogravure printing 5
2 Flexographic printing 7
3 Carbon adsorption 10
4 Thermal incineration 13
5 Catalytic incineration 16
Table Page
1 Cost Effectiveness of Control Technologies for
Small Rotogravure and Flexography Facilities 2
2 Model Plant Specifications 20
3 Permanent Total Enclosure Capital Cost Estimates 23
4 Assumptions for Model Plant Costing 24
5 Thermal Incinerators -- Capital Costs 25
6 Thermal Incinerators -- Annual Operating Costs 27
7 Thermal Incineration -- Cost Effectiveness 28
8 Catalytic Incinerators - Capital Costs 29
9 Catalytic Incinerators - Annual Operating Costs 30
10 Catalytic Incineration -- Cost Effectiveness 32
11 Carbon Adsorption Systems -- Capital Costs 33
12 Carbon Adsorption Systems -- Operating Costs 34
13 Carbon Adsorption -- Cost Effectiveness 35
v
-------
ACKNOWLEDGEMENT
The authors acknowledge the efforts of Bob Blaszczak and Karen Catlett of Environmental
Protection Agency's Office of Air Quality Planning and Standards, both members of the project
review team.
vi
-------
METRIC CONVERSIONS
Nonmetric
Times
Yields Metric
ft
0.3
m
ft2
0.093
..*2
m
in
2.54
cm
in H20
249
Pa
« 2
m
6.45
cm2
lb
0.454
kg
scfm
0.00047
sm3/s
ton
0.907
tonne
vii
-------
1.0 SUMMARY
Emissions of volatile organic compounds (VOC) from rotogravure and
flexographic printing facilities arise from the evaporation of solvents during ink drying.
These emissions can be reduced by conversion of solvent-borne ink systems to water-
borne ink systems, or by capture of the solvent vapors and use of a control device
such as thermal or catalytic incineration systems or carbon adsorption systems. There
are limitations associated with each approach and individual circumstances, including
the type of product produced, the customer base, and the type of ink used, which will
affect the applicability of different technologies.
Conversion to waterborne inks can reduce VOC emissions by approximately
80%. Uncertainties in retrofitting existing presses for water-borne inks exist; required
modifications are site-specific. Drier systems and, in some cases, ink-supply systems
must be modified. Gravure cylinders must be replaced. Water-borne inks can
eliminate the problems of designing and testing capture systems. In some cases, it is
difficult to achieve the same level of gloss with water-borne inks as with solvent-borne
inks. In cases where water-borne inks are suitable, conversion to water-borne inks
may be the most cost-effective solution. Due to the site-specific nature of conversion
costs, no generalized cost estimates can be developed.
Properly operated carbon adsorption systems with total enclosures or
efficient capture systems can reduce VOC emissions by 95 percent. Solvent can be
recovered for reuse on site or sold to a reclaimer. Carbon adsorption systems are
incompatible with certain inks and are most suitable for facilities with a predictable,
long-term production schedule. Facilities using a wide variety of inks to print
numerous small jobs are not likely to be able to use carbon adsorption systems.
Activated carbon has a solvent capacity which varies for different organic components.
Cost estimates have been developed on the basis of toluene as the design solvent. In
some cases, other solvents which are present in some inks may require larger and,
thus, more expensive systems.
1
-------
Property operated catalytic incinerator systems with total enclosures or efficient
capture systems can reduce VOC emissions by up to 98 percent. Solvents are
destroyed in these systems. Catalytic incinerators provide an energy savings over
thermal incinerators, but they are not compatible with all inks. Incinerator
specifications must be written with specific reference to the type of inks to be used.
Small facilities may avoid catalytic incinerators because of higher initial capital costs
than thermal incinerators, and the desire to maintain flexibility to print a wider variety
of jobs.
Properly operated thermal incinerator systems, with total enclosures or efficient
capture systems, can reduce VOC emissions by 98 percent. Thermal incinerators are
compatible with most inks used in rotogravure and flexography, but these systems are
relatively energy intensive. Cost-effectiveness data for these control devices are
summarized in Table 1. The cost of total enclosures must be added to these costs.
Table 1. Cost Effectiveness of Control Technologies for
Small Rotogravure and Flexography Facilities*
Plant Sizeb
(ton/yr)
Cost Effectiveness ($/Ton}
Thermal
Incineration
Catalytic
Incineration
Carbon
Adsorption
10
$3,500 to $4,800
$3,900
$3,500
25
$2,000 to $3,000
$2,500 to $2,800
$1,400
50
$1,200 to $2,400
$960 to $2,000
$760 to $780
100
$850 to $2,000
$1,200 to $1,600
$450 to $460
1000
$170 to $480
$170 to $350
$120
*1991 dollars, exclusive of total enclosure or capture devices. Control efficiencies
assumed to be 95 to 100 percent. Capture efficiencies are assumed to be 100
percent.
'Total solvent use including solvent present in purchased ink and solvent added by
facility.
2
-------
2.0 INTRODUCTION
In 1978, the U.S. Environmental Protection Agency (EPA) published Control
Technology Guidelines (CTG) for the graphic arts industry applicable to emissions of
VOC. This CTG document (1) defined reasonably available control technology
(RACT) for publication and packaging rotogravure and packaging flexography. The
applicability of these guidelines was subsequently limited to plants emitting 100 tons*
of VOC per year or more by EPA guidance (2).
The Clean Air Act Amendments (CAAA) of 1990 required RACT for sources
that emit as little as 10 tons of VOC per year in extreme ozone nonattainment areas.
To meet the ambient air quality standards for ozone, States with "extreme," "severe,"
or "serious" problems were required to establish and implement RACT for these
smaller sources. This document identifies alternate control techniques (ACT) for
States to use as a reference when establishing and implementing RACT for existing
graphic arts facilities with potential uncontrolled VOC emissions of less than 100 tons
per year.
3.0 GENERAL DESCRIPTION
In rotogravure and flexographic printing, ink is applied to a cylindrical image
carrier. The ink is transferred to the print surface as it passes by the rotating cylinder.
Different colors are printed on the print surface (paper, film, foil, etc.) as it passes
successively through multiple print stations. Presses are available with as many as
eight stations. In most cases, the stock to be printed is supplied from a roll or web.
graphic arts facilities with potential uncontrolled VOC emissions of less than 100 tons
per year. The printed stock can be rewound onto a roll, or cut, slit, and/or folded,
depending on the type of press and application. In some cases, stock to be printed is
fed sheet by sheet. Very fluid inks are required for these processes; drying involves
the evaporation of the fluid part of the ink into heated air, which leaves behind the
pigment and a binder. The evaporated material is typically an organic solvent, a
mixture of organic solvents or a mixture of organic solvents and water. Evaporation of
these solvents creates a potential source of VOC emissions.
•Readers more familiar with the metric system may use the conversion factors at the
end of the front matter of this report {page vi).
3
-------
Additional background information on printing technologies is presented in
Vincent and Vatavuk (1), Strauss (3), National Association of Printing Ink
Manufacturers (4), and Gravure Association of America (5).
3,1 Description of the Rotogravure Process
Rotogravure is an intaglio printing process. This means that the image area is
depressed relative to the surface of the printing cylinder. The printing cylinder typically
is composed of chrome-plated copper or steel plated with copper then chrome. The
image is formed by etching depressions (or cells) in the chrome or by etching the
copper prior to chrome plating. The depth of the cells (typically about 0.035 mm)
determines the amount of ink applied to the surface. The density of the cells varies
between about 600 and 32,000 per square inch, with finer quality printing requiring
more cells.
The printing cylinder rotates through a trough of ink (or ink fountain) where ink
is applied to the cells. Excess ink is removed from the surface of the cylinder by a
flexible doctor blade; leaving the cells full of ink. The ink is then transferred to the
surface of the stock as it is pressed against the rotating cylinder. The printed stock, or
web, then passes through one or more driers where the solvent in the ink is
evaporated from the printed image. Solvent-laden air is captured from the driers and
vented to the atmosphere or to a control system. In some cases, the hot air from the
driers is cascaded from station to station. Figure 1 illustrates a typical rotogravure
printing press.
The ink in the fountain is composed of ink supplied by the ink manufacturer (50
percent or more solvent) that has been mixed with additional solvent to maintain the
proper viscosity. After mixing, inks used for rotogravure printing may contain 50 to 85
percent (or more) organic solvent (6). The solvents in rotogravure ink include esters,
alcohols, ketones, and aromatic and aliphatic hydrocarbons (4).
Changing from one job to another involves changing the cylinders and, if
different colors are to be printed, cleaning the ink residue in the fountains. In
packaging rotogravure, the cylinders are stored to permit printing the same job at a
later date.
4
-------
Impression Cylinder
Stock
Doctor Blade
Etched Cylinder
Ink Pump
Ink Tank
T—T
TT
Source; EPA/450/2-78-033.
Figure 1. Rotogravure printing.
5
-------
3.2 Description of Flexoaraphv
Flexography is a relief printing process. This means that the image surface is
raised relative to the surface of the cylinder. The image surface is a flexible rubber or
polymeric plate that is wrapped around the plate cylinder. Ink is applied to the raised
image surface by one of two different methods. In the older method, a rotating
cylindrical "fountain roller" is partially submerged in the ink fountain. This ink is
transferred to a "form roller," which rotates against the fountain roller. Excess ink is
removed from the form roller with a doctor blade. The form roller then transfers the
ink to the raised surfaces of the printing plate, which rotates against it. In the newer
method, the form roller rotates through the ink trough and the fountain roller is omitted.
The ink is applied to the stock as it is pushed against the printing plate by a rotating
impression cylinder. Figure 2 illustrates a typical flexographic printing process.
The ink used in flexographic printing is composed of one or more resins
dissolved in one or more solvents. The solvents, which include alcohols, esters, glycol
ethers, and aliphatic hydrocarbons, must be compatible with the rubber or polymeric
printing plates. Dyes and pigments also are dissolved and dispersed in the ink. The
inks contain approximately 75 percent solvent. Water-based inks are also available.
After printing, the inks are dried by evaporating the solvents into hot air driers.
. Solvent-laden air is captured from the driers and vented to the atmosphere or to a
control system.
Changing from one job to another involves replacing the printing plates on the
plate cylinders and if different colors are to be printed, cleaning the ink residue in the
fountains and on the rollers.
4.0 CAPTURE TECHNOLOGIES
Capture technologies are methods, procedures, or facilities that can be used to
collect and contain VOCs that are emitted from a particular process. The term capture
efficiency is defined as the fraction of all organic vapors generated by a process that
is directed to an abatement or recovery device. Two types of capture technologies
are total enclosures and capture devices.
6
-------
Stock
Plate
Cylinder
Impression
Cylinder
Distribution
Rollers
Source; EPA/450/2-78-033.
Figure 2. Flexographlc printing.
7
-------
4.1 Total Enclosure
Permanent total enclosures are structures that completely surround emissions
sources so that all air leaving the enclosure is exhausted through a control device.
Existing rooms can be modified and tested to confirm compliance with the total
enclosure criteria. New structures can be built around emissions sources to serve as
total enclosures within existing rooms. EPA has specified five criteria that must be
met for an enclosure to be considered a permanent total enclosure. These criteria
have been published in New Source Performance Standards for the magnetic tape
industry and for the coating of polymeric substrates, 40 CFR 60 (7). The criteria are
as follows:
1) Ail VOC emissions must be captured and contained for discharge
through a control device.
2) The total area of all natural draft openings (NDOs) must not exceed 5
percent of the surface area of the enclosure walls, floor, and ceiling.
NDOs include makeup air vents, open doors and windows, cracks under
doors, and other openings.
3) All doors and windows that are not to be considered NDO must be
closed during routine operation of the process. (Doors can be equipped
with automatic closures to establish that they are not NDO.)
4) The average facial velocity of air entering through all NDOs must be at
least 200 ft/min. This is equivalent to a pressure drop of at least 0.004
inches of water across the NDO. (The pressure inside the enclosure
must be negative relative to the outside of the enclosure. The direction
of flow must be into the enclosure.)
5) Any NDO must be at least four equivalent diameters from a source of
VOC. (A window with a 4-ff opening could be located no less than 9 ft
from the nearest source of emissions.)
Any enclosure meeting these criteria will be considered a total enclosure, and
capture efficiency will be assumed to be 100 percent. An existing room might meet
these criteria with some modifications. Alternatively, an enclosure could be
constructed around a press or presses within an existing room.
8
-------
4.2 Capture Devices
Capture devices may be enclosures or rooms that do not meet the criteria for
permanent total enclosures. Alternatively, devices such as fume hoods, "floor
sweeps," or open booths may be used to capture VOC. These devices collect air in
the region where VOC is emitted and duct it to a control device. If a capture device is
used as part of a VOC emissions control system, in other than a permanent total
enclosure, the capture efficiency may be determined through testing. The overall
efficiency of a system that uses a capture device is the product of the capture
efficiency and the efficiency of the control device.
The efficiency of a capture device may be determined using a mass balance
based on gaseous VOC concentrations measured around the partial enclosure. A
temporary total enclosure may be constructed for the purpose of obtaining an accurate
mass balance for use in determining the capture efficiency of a capture device.
5.0 CONTROL TECHNOLOGIES AND POLLUTION PREVENTION OPTIONS
5.1 Carbon Adsorption
Activated carbon is a material with a high surface area that adsorbs many
organics from air streams. When used for gas phase emissions control, granular-
activated carbon is used in fixed-bed adsorbers. Typically, contaminated air is
collected using permanent total enclosures or capture devices and is passed through
two or more beds of carbon. VOC in the air is adsorbed on active sites on the
carbon. A schematic of a typical carbon adsorption system is given in Figure 3.
Carbon has an experimentally determined capacity for specific concentrations of
specific organics. Over time, this capacity is exhausted (essentially all of the active
adsorption sites are occupied) and the organics pass through unadsorbed. Adsorbers
are operated in parallel so that when the capacity in one unit is exhausted, it can be
removed from service and a second adsorber can be put into service. The exhausted
carbon in the first adsorbent can then be replaced or regenerated. A wide variety of
differently manufactured activated carbons are available. These materials have
different capacities for different organic compounds. Where a variety of volatile
organics are present in a contaminated gas stream, the control of the compound that
breaks through first dictates the selection and operating procedure for the adsorbers.
9
-------
Emission Stream
I
Condenser
Low-Pressure
Steam
Source: EPA/625/6-91/014.
Adsorbers
P"'
-¦ecanter ^°*veri5
Water (,or reus®)
(to treatment plant)
Exhaust to
Atmosphere
Figure 3. Carbon adsorption system.
10
-------
Ketones, which are present in many rotogravure inks, have caused problems in
carbon adsorption systems in the past because they polymerize on the surface of the
carbon granules, at times leading to bed fires. Very high recovery efficiencies are
possible if the carbon adsorbers are properly maintained (9).
in contrast to incineration techniques, carbon adsorption does not destroy the
VOC in the contaminated air but merely removes it. This may be an advantage or a
disadvantage. To restore the capacity of the activated carbon for reuse, it must be
regenerated. Activated carbon is regenerated by heating it with steam or hot air to
drive off the adsorbed organics. These desorbed organics must be controlled eventu-
ally. Carbon regeneration can take place on site, or the cartoon can be returned to the
vendor for routine regeneration and replacement.
If the composition and flow rate of the gas stream to the carbon adsorption
system are relatively well known, the required replacement schedule can be predicted
from data provided by the carbon vendor. Where these parameters are uncertain,
organic vapor monitors can be installed to detect breakthrough. At this point, the
adsorber can be removed and regenerated. A new adsorber containing regenerated
carbon can be installed at this time.
Rotogravure and flexographic printing facilities typically buy solvents for diluting
the purchased ink to maintain the proper flow characteristics. At facilities where one
solvent (or one solvent mixture) is used exclusively (or nearly exclusively), the use of
activated carbon for emissions control offers the potential for recovery and reuse of
solvent. Facilities that use a variety of inks, with different solvents and solvent
mixtures, to meet the requirements of numerous small printing jobs will be less likely
to take advantage of this option. A mixture, particularly an unknown mixture of
solvents, is much less useful for blending purposes.
Design and cost estimation procedures for carbon adsorption systems are given
in EPA (8) and Vatavuk (10). Carbon adsorption efficiencies vary depending on the
specific VOC or VOC mixture. VOC removal efficiencies of at least 95 percent can be
achieved, provided: (a) the adsorber is supplied with an adequate quantity of high-
quality activated carbon, (b) the gas stream receives appropriate conditioning (e.g.,
cooling, filtering) before entering the carbon bed, and (c) the carbon beds are
regenerated before breakthrough (11).
11
-------
5,2 Thermal Incineration
Thermal incinerators are control devices in which the contaminated air streams
collected with total enclosures or capture devices are burned. VOCs are converted to
carbon dioxide and water with a high degree of efficiency. Gas streams of Interest in
the graphics arts industry will contain sufficient oxygen to support combustion. Dilute
gas streams, which may be present during process shutdown and make-ready
periods, may require the addition of supplemental fuel to sustain combustion. A
schematic of a thermal incineration system is shown in Figure 4.
In thermal incineration, the contaminated air stream is preheated, ignited, and
combusted. Various designs use different types of combustion devices. In general,
combustion chamber designs must provide high turbulence to mix the VOC, fuel, and
air to ensure essentially complete combustion. The other requirements are a high
enough temperature and a long enough residence time to complete the combustion
process. Temperatures are generally maintained at about 900 °C (1600°F) but will
vary depending on the solvent mixture involved. Properly designed and adjusted
incinerators operating at a maximum of 900 °C (1600°F) and 0.75 second residence
time will achieve at least 98 percent VOC reduction.
Upon ignition, the VOCs in the gas stream are oxidized and destroyed. The
rate of destruction at any given temperature varies depending on the composition of
the contaminated air stream. Generally, heat released from the reaction is sufficient to
heat the contaminated air stream to the ignition temperature. In some cases, heat
exchangers are used to recover heat from the combustion reaction for use in
preheating the contaminated air stream.
Because the incinerator must be in operation at times when VOC emissions are
very low (i.e., when presses are on standby or during changeovers), supplemental fuel
requirements will vary. During these time periods, supplemental fuel in the form of
natural gas is used to sustain combustion at the necessary temperature to ensure
destruction of the VOC in the gas stream.
Incineration systems are supplied with controls to start-up and bring the
combustion chamber to the proper temperature. These controls can provide an
interlock to prevent operation of the press until the incinerator temperature is adequate
to ensure destruction of the VOC. Adjustments in supplementary fuel requirements
12
-------
Emission Source
Exhaust Gas
(to Stack)
ricinerator
Supplementary Fuel
Heat
Exchanger
(Optional)
Figure 4. Thermal Incineration.
13
-------
based on changes in emission rate, air flow rate, and ambient temperature can also
be made automatically.
Design and cost estimation procedures for thermal incinerators are given in
Handbook: Control Technologies for Hazardous Air Pollutants (8) and OAQPS Cost
Control Manual, (10). Incinerators are assumed to be 98 percent effective for VOC
destruction (1).
5.3 Catalytic Incineration
Catalytic incinerators are control devices in which the contaminated air streams
collected with total enclosures or capture devices are burned. VOCs are converted to
carbon dioxide and water with a high degree of efficiency. In the presence of a
catalyst, this reaction will take place at lower temperatures than those required for
thermal incineration. Temperatures between 350 and 500 °C (660 and 930 °F) are
common. The catalysts are metal oxides or precious metals which are supported on
ceramic or metallic substrates.
From an operational standpoint, the lower reaction temperature means that the
requirement for supplemental fuel is reduced or eliminated during normal operation.
Provision for supplemental fuel must still be made for start-up and standby periods.
Gas streams of interest in the graphics arts industry will contain sufficient oxygen to
support combustion. The lower operating temperatures also decrease the formation of
oxides of nitrogen.
The use of a catalyst is inconsistent with certain ink formulations. Chlorinated
solvents and some silicone ink additives are among some of the compounds that can
poison or deactivate catalysts. The specification of a catalytic incinerator for a
particular application must take into account the specific inks to be used. Facilities
dedicated to long-term production for specific packaging applications will be able to
commit to a specific type of ink chemistry more readily than facilities printing
numerous small jobs.
Design of catalytic incinerators varies from manufacturer to manufacturer. The
major differences involve the geometry of the combustion chamber, the type of
catalyst and support material, and the type of contact between the gas and the
catalyst. In fixed-bed designs, gas flows through beds of catalyst beads (or pellets) or
through a monolithic support. In fluidized bed designs, gas flows through an
14
-------
expanded fluidized bed of catalyst pellets. A summary of performance test data is
given by Spivey (12). Selection and cost estimation procedures for catalytic
incinerators are given in Handbook: Control Technologies for Hazardous Air Pollutants
(8) and the OAQPS Control Cost Manual (10). Catalytic incinerators have an
estimated VOC destruction efficiency of 98 percent. A schematic of catalytic
incineration is shown in Figure 5.
5.4 Water-Borne Inks
Problems associated with capture and control of VOC present in rotogravure
and flexographic inks can be reduced substantially by substituting water-borne inks for
the more commonly used solvent-borne inks. Water-borne inks contain a higher
proportion of solids (pigments and resins) than solvent-borne inks. The VOC content
of water-borne inks is approximately 5 to 30 percent as used (1) in contrast to 50 to
85 percent for solvent-borne inks. As a control option, the 1979 EPA Guidelines (2)
specified a limit of 25 percent organic in the volatile fraction of water-borne ink.
Conversion of an existing uncontrolled press to water-borne inks could eliminate the
need for a capture and control system. However, extensive changes in equipment
and operating procedures may be required to convert to water-borne inks.
In general, water-borne inks produce different printing results than organic
solvent-based inks. These results may include lower gloss and lower resolution.
Some types of stock, particularly high-gloss coated paper, may be incompatible with
water-borne inks. For many applications, particularly in packaging, water-borne inks
can produce jobs of adequate quality.
Conversion of a press from solvent-borne to water-borne inks requires changes
in driers, materials, and operation. Rotogravure presses also require differently
engraved cylinders for operation with water-borne inks. Drying involves evaporation of
the liquid portion of the ink from the stock. Water has a higher boiling point, a higher
heat of vaporization, and a lower vapor pressure than the organic solvents used in
rotogravure and flexographic inks. As a result, it dries more slowly than solvent under
any set of conditions. Driers used with water-borne inks require some combination of
more air, warmer air, and more residence time than do driers used for a similar job
printed with solvent-borne inks. More residence time is achieved with a longer drier
and, in some cases, a lower press speed. Some presses do not have enough space
for installation of longer driers without major rebuilding. In addition, lower press
speeds result in reduced output from a given press.
15
-------
Emission Source
Supplementary
Fuel
Catalytic Incinerator
Preheater
Catalyst Bed
i r
Figure 5. Catalytic Incineration.
16
-------
Water is more corrosive to carbon steel components of the presses than are
organic solvents, and driers fabricated from stainless steel are more expensive than
the commonly used carbon steel driers. Increased corrosion of structural elements,
driers, bearings, and gravure cylinders may require expensive modifications when
existing presses are converted to water-borne inks. It is difficult to predict the extent
of this problem.
Rotogravure cylinders have a particular cell geometry consistent with the flow
characteristics of the ink being used. Conversion of a packaging gravure facility to
water-borne inks would mean that stored cylinders, originally designed for use with
organic solvent-borne inks, could no longer be used. This is less of a problem with
publication rotogravure facilities, and not a problem with flexographic facilities.
Conversion to water-borne inks also necessitates modifications in certain
operating procedures. Solvent-borne inks contain dissolved resins; these can be re-
moved from ink fountains, rollers, cylinders and other components with solvent during
setup and changeover periods. Water-borne inks contain dispersed resins, which
harden upon drying. Cleanup operations are more difficult and less flexible with water-
borne inks.
The actual cost of the ink required for a particular job is slightly higher for
water-borne inks. The cost per pound of water-borne inks is considerably higher than
solvent-borne inks; however, less water-borne ink than solvent-borne ink is required
to print a particular job because of its higher solids content.
Direct cost comparisons between conversion to water-borne Inks and
installation of capture and control systems are not possible. If existing equipment can
be adapted to water-borne inks and product quality is acceptable to the customer,
VOC emission reductions of approximately 80 percent are possible. In addition, air
quality in the work place will improve and the potential for fire hazards will decrease.
A significant reduction in fire insurance premiums will result from conversion to water-
borne ink.
Morris (13) describes the following case studies of conversion from solvent-
borne to water-borne inks. Facility A extrudes polyethylene bags and prints them
flexographically. An equivalent or better product was produced after altering the
surface treatment of the polyethylene, replacing the anilox rolls on the press (changing
from 150 to 360 line screen), and increasing the drier air velocity and air volume.
17
-------
Facility C prints clay-coated cardboard by rotogravure. Conversion involved changing
the type of stock printed, retrofitting the driers for a higher volume of lower
temperature air, and changing the cell depth and line count of the cylinders (13).
Friedman and Vaught (14) described the conversion of a flexographic facility
operated by Amko Plastics from solvent-borne to water-borne inks. The conversion
involved redesign of air blowers, plenums, and drier hoods to improve drying. The
anilox rollers, fountain rollers, and printing plates were also modified. As a result of
the conversion, explosion-protected storage rooms were no longer required for the ink.
Other advantages included the elimination of an underground solvent storage tank
with a resulting decrease in liability insurance, more consistent color throughout press
runs, a savings on ink costs (the ink is more expensive per pound but less is
required), and an improvement in air quality in the press room. Disadvantages
included lower gloss on some jobs and higher costs associated with wastewater and
sludge produced during cleanup.
5.5 Other Types of Inks
The need to capture and control VOC emissions from graphic arts facilities
would be reduced or eliminated if the VOC in the ink formulations could be reduced or
eliminated. The potential for replacing VOC with water has been described. Other
approaches involve using inks that have less liquid (organics or water) and more
solids (pigments, binders and resins). Several types of high-solid inks are available for
use in other printing processes (lithography and letter press in particular). At the
current stage of development, however, these inks are unsuitable for rotogravure and
flexographic presses. Improved high-solid inks are continuously under development.
However, no high-solid inks are presently available with viscosities low enough to be
suitable for rotogravure or flexography.
Radiation-curable inks are combinations of monomers that are applied to the
substrate and polymerize in the presence of ultraviolet (UV) light or electron beams.
At present, these inks are available for letter press, lithographic, and screen printing.
Walata and Newman (15) found no current applications of these technologies in
rotogravure or flexographic printing. The limitations of these ink systems result
primarily from their high viscosity. Electron beam curing must take place in an
oxygen-free environment. This is inconsistent with existing rotogravure and
flexographic presses. UV-curable inks are less sensitive to the presence of oxygen
but they require dilution with solvent or water to reduce the ink viscosity to a level
18
-------
suitable for the existing presses. Modifications to existing presses could theoretically
be made to allow these inks to be used, but drier modifications may also be needed to
evaporate the solvent or water. UV systems suitable for retrofit to existing presses are
not presently available.
6.0 DESCRIPTION OF MODEL PLANTS
Model plants have been specified to provide a consistent basis for comparison
among control options. The potential uncontrolled emissions considered were 10, 25,
50,100, and 1,000 tons per year of VOC. For the purposes of this report, potential
uncontrolled emissions are equal to solvent use. This includes the solvent content of
the purchased ink, solvent added to make the ink press-ready, and makeup solvent to
correct for evaporation in the ink fountain.
There are, of course, an infinite number of scenarios that can result in any
given annual ink usage. Model plants have been specified in Table 2 on the basis of
total uncontrolled emissions. The important parameters are the yearly emissions, the
VOC concentration in the drier exhaust, and the hours of operation. Rotogravure ink
would be more fluid (i.e., less viscous as a result of greater solvent dilution) than
flexographic ink. There is enough variation between individual presses and printing
jobs that the distinction is unimportant. The EPA Guidelines Document (1) lumps
rotogravure and flexography, and estimates solvent content at between 50 and 80
percent. These data are presumably on an as-used basis. If it is important to have a
higher as-used solvent content, this parameter could be adjusted from 67 percent to
as high as 80 percent by making corresponding changes in the other specifications.
The fraction of time presses used varies widely depending on the size of
individual jobs. Setup time is not affected by length of the job. Small facilities, such
as those considered in the model plants, may have much more downtime as a result
of frequent changeovers between small jobs. Facilities that operate less than 24
hours per day have the flexibility to schedule additional shifts if needed. A utilization
rate of 60 percent has been specified. This could be adjusted to a lower rate (much
lower if downtime, due to lack of business, is considered) by adjusting other
specifications (e.g., potential operating time).
19
-------
Table 2. Model Plant Specifications a
Potential
Model uncontrolled No. of Anticipated
plant emissions (ton/yr) presses operating schedule (h/yr)
1
10
1
8 h/d, 5 d/wk (2,086 h/yr)
2
25
1
16 h/d, 5 d/wk (4,170 h/yr)
2A
25
2
8 h/d, 5 d/wk (2,086, h/yr)
3
50
1
16 h/d, 5 d/wk (4,170 h/yr)
3A
50
2
8 h/d, 5 d/wk (2,086, h/yr)
4
100
1
16 h/d, 5 d/wk (4,170 h/yr)
4A
100
2
8 h/d, 5 d/wk (2,086 h/yr)
5
1,000
5
16 h/d, 7 d/wk (5,840 h/yr)
Specifications based on the following assumptions:
Uncontrolled emissions are based on total VOC used.
Press utilization: 60 percent.
Setup, make-ready, etc.: 40 percent.
Exhaust flows will be specified at 10 percent and 25 percent of the lower
explosive limit (based on toluene).
Capture efficiencies are assumed to be 100 percent.
20
-------
The number of presses, web width, press speed, and coverage at these
facilities can vary greatly. Within the expected precision of the estimates on
emissions, the primary consideration is the total solvent use (i.e., the total potential
uncontrolled emissions). Width, speed, and coverage, consistent with a particular ink
consumption, will make little difference in emissions control costs.
7.0 COSTS
7.1 Costs of Converting to Water-Borne Inks
The costs of conversion to water-borne inks involve modifications to existing
presses. At a minimum, this will involve installation of higher capacity driers or
modification of existing driers to handle greater volumes of air. A variety of additional
modifications to equipment and operating procedures (e.g., construction material
modifications, operating speeds, and equipment cleaning procedures) will also be
necessary. The costs of these modifications cannot be predicted accurately because
they will depend on the particular jobs being printed. Required changes in operating
procedures (reduced press speed, for example) may impose significant costs.
Another important consideration is the downtime associated with installation and
debugging of retrofit equipment. This downtime in production may make press
modifications unreasonable for small facilities with a single printing press.
The primary costs associated with conversion result from the downtime
necessary during the experimentation to develop satisfactory operating procedures.
For rotogravure facilities, the largest item of capital equipment associated with
conversion to water-based units is the dryer. A typical cost for a 5-feet dryer on a 36-
inch web press is $36,000 + 50 percent (16). An eight-station press might require
eight such driers. Packaging rotogravure plants maintain an inventory of cylinders for
reuse in subsequent press runs, which would be made obsolete by conversion to
water-borne inks. Each cylinder can cost as much as $2,000 to replace. Thus,
replacement of 100 cylinders might cost approximately $200,000.
The overall cost of water-borne inks to print a particular job may be slightly
higher, or slightly lower, than solvent-borne inks. This takes into account both the
higher cost per pound and the fact that water-borne inks have a higher solids content
and, thus, less is required. The conversion of a solvent-borne ink facility to water-
borne inks will result in a significant savings in fire insurance premiums. In addition,
press room air quality will be greatly improved.
21
-------
7.2 Total Enclosures for Existing Facilities
The cost of retrofitting an existing facility with a total enclosure capture system
varies greatly depending on the layout of the facility. Many existing press rooms
already function as total enclosures. Confirmation that an existing structure qualifies
as a total enclosure may be as simple as measuring the dimensions, the face velocity,
and direction of the air flow at all natural draft openings. This could be accomplished
with a tape measure and an anemometer in a few hours.
In the worst case, construction of a total enclosure would necessitate
construction of a new room within the existing structure. A more reasonable situation
might involve using two existing walls and building two new walls (each containing a
door) to create a permanent total enclosure within an existing facility. Sample retrofit
specifications and estimated costs are given in Table 3 (17). It should be noted that
the actual costs to construct a permanent total enclosure can vary from essentially
zero to more than twice the costs listed.
7.3 Costs of Thermal Incinerators
Emissions stream characterization assumptions that are necessary to evaluate
control technology costs are given in Table 4. Control systems are assumed to
operate during downtime at the same flow rate at which they operate during printing
operations. The capital costs of thermal incineration systems are given in Table 5, as
estimated from the EPA Handbook (8). The scale of equipment required is primarily
determined by the flow rate and is much less sensitive to the VOC concentration. For
any given annual quantity of potential emissions, the cost of the equipment will be
affected by the operating hours. Thus, a plant with uncontrolled emissions of 100 tons
annually will need a larger incinerator if it operates 40 hours per week than if the
same mass of VOC is created by a smaller press working three shifts per day.
Similarly, a plant emitting 100 tons per year with a given operating schedule and 50
percent downtime will have higher control costs than a smaller plant printing 75
percent of the operating hours.
22
-------
Table 3. Permanent Total Enclosure Capital Cost Estimates a
Potential Cost
Model uncontrolled Enclosure estimate
plant emissions (ton/yr) specifications (1991 $)
10
Two walls - 15 ft high/12 ft wide (each)
Two doors - 10 ft high/12 ft wide
8 ft high/ 3 ft wide
$ 4,000
25
Two walls - 15 ft high/12 ft wide (each)
Two doors - 10 ft high/ 4 ft wide
8 ft high/ 3 ft wide
4,000
N)
U>
50
Two walls — 15 ft high/20 ft wide (each)
Two doors -- 10 ft high/ 8 ft wide
10 ft high/ 4 ft wide
6,800
100
Two walls --15 ft high/25 ft wide (each)
Two doors -- 10 ft high/ 8 ft wide
10 ft high/ 4 ft wide
6,800
1,000
Two walls -- 18 ft high/50 ft wide (each)
Two doors - 14 ft high/ 8 ft wide (each)
19,000
aThese estimates include materials, labor, contractor's overhead, and contractor's profit for 6-inch thick concrete block
construction. Labor rates are based on average costs in 30 major U.S. cities. Data from R.S. Means Company, Building
Construction Cost Data - 1990, 48m Annual Edition. Costs have been adjusted to fourth quarter 1991 using Marshall and Swift
cost index (18).
-------
Table 4. Assumptions for Model Plant Costing a
Model
plant
Potential
uncontolled
emissions
(ton/yr)
Operating
Time
(h/yr)
25% LEL
Concentration
(ppmv)
Flow rate
(std ft3/min)
10% LEL
Concentration
(ppmv)
Flow rate
(std ft3/min)
1
10
2,086
3,250
500
1,300
965
2
25
4,170
3,250
500
1,300
1,207
2A
25
2,086
3,250
965
1,300
2,413
3
50
4,170
3,250
965
1,300
2,413
3A
50
2,086
3,250
1,930
1,300
4,826
4
100
4,170
3,250
1,930
1,300
4,826
4A
100
2,086
3,250
3,860
1,300
9,652
5
1,000
5,840
3,250
8,239
1,300
20,658
LEL « tower explosive limit (for toluene).
8 The actual flow rates for model plants 1 and 2 are 385 and 483 std ft3/min respectively, but 500-std ft3/min thermal
incinerator costs have been used because smaller sizes may not be available. A permanent total enclosure (100
percent capture) is assumed.
-------
Table 5. Thermal Incinerators - Capital Costs a
Model
plant
Potential
uncontrolled
(ton/yr)
Total ($)
25% LEL
Annualized ($/yr)
10%LEL
Total ($) Annualized ($/yr)
1
10
$ 99,000
$ 16,000
$ 120,000
$ 19,000
2
25
99,000
16,000
120,000
20,000
2A
25
115,000
19,000
140,000
23,000
3
50
120,000
19,000
140,000
23,000
3A
50
130,000
22,000
170,000
28,000
4
100
130,000
22,000
160,000
27,000
4A
100
160,000
26,000
200,000
33,000
5
1,000
360,000
59,000
450,000
74,000
* Costs are derived from U.S. EPA, Handbook: Control Technologies for Hazardous Air Pollutants,
EPA/625/6-91/014,1991. Annualized costs are based on 10-year equipment life at 10 percent
interest Costs for 10- and 25- ton/yr plants at 25 percent LEL are biased high due to unreliable cost
data at flow rates below 500 std ft3/min. Estimates assume 50 percent heat recovery for the 1,000-
ton/yr plants and no heat recovery for all other plants as energy savings do not justify Increased
capital costs. Costs have been escalated to the fourth quarter of 1991 using the Marshall and Swift
Cost Index (18). Costs do not include construction of a permanent total enclosure.
-------
Annual operating costs for thermal incinerators are given in Table 6.
Total costs are very sensitive to the VOC concentration in the emissions
stream for any given model plant size because of the importance of gas
flow rate in the capital cost and because of the natural gas requirement
(included under utilities) to heat the emission stream to approximately
900 °C (1600 °F). Cost effectiveness, or total cost/ton of VOC controlled, is
given in Table 7.
7,4 Costs of Catalytic Incinerators
The cost of catalytic incineration has been estimated based on flow
and concentration assumptions given in Table 4. Catalytic incinerator
systems are rarely used to control streams of less than 2,000 std ft3/min.
No estimates are given for the 10- and 25-ton/yr model plants at 25 percent
lower explosive limit (LEL). An upper bound on these costs would be the
costs for facilities of the same size operating with exhaust streams at 10
percent LEL
Catalytic incineration is not technically feasible for all ink
formulations. The cost estimates are given for facilities using ink
formulations consistent with the available catalysts. Under some
circumstances, it may be possible to change to a different type of ink in
order to produce emissions controllable with catalytic incineration. The
capital cost of catalytic incinerator systems is given in Table 8. As with
thermal incinerators, the flow rate is the most significant factor in sizing the
equipment. A given mass of VOC is much more expensive to control in
dilute form than in more concentrated form.
Total annual costs for catalytic incinerator systems have been
estimated in Table 9. These costs are based on operation of the system
during downtime between press runs. As with thermal incineration systems,
the cost of natural gas to heat the exhaust stream to the operating
temperature is the major operating expense. Catalytic incinerators operate
at lower temperatures than thermal incinerators, which provides a cost
advantage. Catalytic incinerators operating at 500° C (900° F) can achieve
destruction efficiencies equivalent to thermal incinerators operating at
900 °C (1600° F). Cost effectiveness, or total cost per ton of VOC
26
-------
Table 6. Thermal Incinerators - Annual Operating Costs *
plant
Potential
uncontrolled
(ton/yr)
%LEL
Capital
(annualized)
Operating
Labor Maintenance
Utilities
Overhead
TOTAI
1
10
25
$16,000
$ 2,100
$4,000
$ 5,000
$ 7,700
$ 35,000
1
10
10
19,000
2,100
4,000
13,000
9,500
48,000
2
25
25
16,000
4,300
8,100
10,000
11,000
49,000
2A
25
25
19,000
2,100
8,100
10,000
12,000
51,000
2
25
10
20,000
4,300
8,100
32,000
12,000
76,000
2A
25
10
23,000
2,100
8,100
32,000
10,000
75,000
3
50
25
19,000
4,300
8,100
19,000
12,000
62,000
3A
50
25
22,000
2,100
8,100
19,000
12,000
63,000
3
50
10
23,000
4,300
8,100
64,000
13,000
112,000
3A
50
10
28,000
2,100
8,100
64,000
14,000
120,000
4
100
25
22,000
4,300
8,100
38,000
13,000
85,000
4A
100
25
26,000
2,100
8,100
38,000
26,000
100,000
4
100
10
27,000
4,300
8,100
130,000
13,000
180,000
4A
100
10
33,000
2,100
8,100
130,000
30,000
200,000
5
1,000
25
59,000
5,900
11,000
74,000
25,000
170,000
5
1,000
10
74,000
5,900
11,000
370,000
28,000
490,000
a Costs are derived from U.S. EPA, Handbook: Control Technologies for Hazardous Air Pollutants, EPA/625/6-91/014,1991.
Annualized costs are based on 10-year equipment life at 10 percent interest. Costs for 10- and 25-ton/yr plants at 25 percent LEL are
biased high due to unreliable cost data at flow rates below 500 std ft3/min. Estimates assume 50 percent heat recovery for the 1,000-
ton/yr plants and no heat recovery for all other plants as energy savings do not justify increased capital costs. Natural gas cost
estimated at $0.0033/std ft3; electricity at $0.059/kWh. Costs have been escalated to the fourth quarter of 1991 using the Marshall and
Swift Cost Index (18). Annualized capita! costs do not include costs of construction of a permanent total enclosure.
-------
Table 7. Thermal Incineration - Cost Effectiveness*
Potential
Uncontrolled
Emissions
Concentration
Cost Effectiveness
Model Plant
(ton/yr)
(% LEL)
($/ton)
1
10
25
$3,500
1
10
10
4,800
2
25
25
2,000
2A
25
25
2,000
2
25
10
3,000
2A
25
10
3,000
i 3
50
25
1,200
3A
50
25
1,300
3
50
10
2,200
3A
50
10
2,400
4
100
25
850
4A
100
25
1,000
4
100
10
1,800
4A
100
10
2,000
5
1000
25
170
5
1000
10
480
"Cost effectiveness is based on 100 percent capture. If less efficient capture devices are used, cost
effectiveness should be adjusted by dividing listed cost effectiveness by fractional efficiency.
28
-------
Table 8. Catalytic Incinerators - Capital Costs 8
Potential
uncontrolled
Model
emissions
plant
(ton/yr)
25% LEL
10% LEL
Total
Annualized
Total
Annualized
1
10
NA
NA
$ 110,000
$ 17,000
2
25
NA
NA
120,000
20,000
2A
25
NA
NA
170,000
28,000
3
50
$ 110,000
$17,000
180,000
29,000
3A
50
150,000
25,000
250,000
41,000
4
100
150,000
25,000
250,000
41,000
4A
100
220,000
37,000
370,000
61,000
5
1,000
340,000
55,000
690,000
110,000
NA = Not Applicable.
LEL = lower explosive limit (for toluene).
a Costs are derived from U.S. EPA, Handbook: Control Technology for Hazardous
Air Pollutants, EPA/625/6-91/014,1991. Annualized costs are based on 10-year
equipment life at 10 percent interest. Costs for 10- and 25-ton/yr plants at 25
percent LEL are not given because equipment of this size may not be available.
Costs for 10- and 25-ton/yr plants may be biased high because reliable cost data
are not available in this size range. Costs assume 50 percent heat recovery for
the 1,000-ton/yr plant operating at 10 percent LEL and no heat recovery for all
other plants, as energy savings do not justify increased capital costs. Costs have
been escalated to fourth quarter 1991 using the Marshall and Swift Cost Index
(18). Capital costs do not include cost of construction of a permanent total
enclosure.
29
-------
Table 9. Catalytic incinerators-Annual Operating Costs
Potential
uncontrolled
Model
emissions
Capital
Operating
plant
(ton/yr)
%
(annualized)
labor
Maintenance
Utilities
Overhead
Total
LEL
1
10
10
$17,000
$2,100
$ 5,400
$ 5,800
$ 8,900
$ 39,000
2
25
10
20,000
4,300
9,900
15,000
13,000
62,000
2A
25
10
28,000
2,100
9,900
15,000
14,000
69,000
3
50
25
17,000
4,300
9,500
5,400
13,000
49,000
3A
50
25
25,000
2,100
9,500
5,400
13,000
55,000
3
50
10
29,000
4,300
11,000
29,000
16,000
89,000
3A
50
10
41,000
2,100
11,000
29,000
18,000
100,000
4
100
25
25,000
4,300
11,000
65,000
15,000
120,000
4A
100
25
37,000
2,100
11,000
65,000
17,000
130,000
4
100
10
39,000
4,300
16,000
58,000
22,000
140,000
4A
100
10
61,000
2,100
16,000
58,000
26,000
160,000
5
1000
25
51,000
5,900
23,000
65,000
31,000
180,000
5
1000
10
100,000
5,900
41,000
148,000
56,000
350,000
a Costs are derived from U.S. EPA, Handbook: Control Technologies for Hazardous Air Pollutants, EPA/625/6-91/014,
1991. Annualized costs are based on 10-year equipment life at 10 percent interest. Costs for 10- and 25-ton/yr plants
at 25 percent, LEL are not given because equipment of this size may not be available. Costs for 10- and 25-ton/yr
plants may be biased high because reliable cost data are not available in this size range. Costs assume 50 percent
heat recovery for the 1,000-ton/yr plant operating at 10 percent LEL and no heat recovery for all other plants as energy
savings do not justify increased capital costs. Maintenance costs include a change of catalyst every 2 years.
Overhead includes taxes, insurance, and administration. Costs have been escalated to fourth quarter 1991 using the
Marshall and Swift Cost Index (18). Annualized capital costs do not include costs of construction of a permanent total
enclosure.
-------
controlled, is given in Table 10. Note that the decision to use catalytic incineration
implies a commitment to use inks that are compatible with the catalyst.
7.5 Costs of Carbon Adsorption Systems
Carbon adsorption system costs have been estimated based on toluene as the
VCXJ to be controlled. Working capacities of activated carbon applied to the model
plants are assumed to be 0.67 lb toluene/lb carbon for the plants operating at 25
percent LEL and 0.61 lb toluene/lb carbon for the plants operating at 10 percent LEL.
Different ink solvents and solvent mixtures may require more or less carbon.
Regenerable carbon adsorption systems are rarely used for control of flows less
than 2,000 std ft3/min. Carbon canister systems for low flows are available at about
$4/lb. This is equivalent to $13,000 per ton of VOC controlled. An oversized
regenerable carbon system provides an upper bound on the cost of systems for small
facilities. These systems are more economical than canister systems, even if sized
several times larger than necessary.
Capital costs for regenerable carbon adsorption systems are given in Table 11.
The cost of systems for the 50-ton/yr model plants can be applied to smaller plants.
The carbon system costs are much less sensitive to the VOC concentration of the
stream and are primarily influenced by the amount of carbon required and the size of
the vessels required to house it. Costs have been estimated using the procedures
described in Handbook: Control Technologies for Hazardous Air Pollutants (8); how-
ever, more conservative assumptions about cycle time have been used.
Annual operating costs are estimated in Table 12. Cost estimates for the 10-
and 25-ton/yr model plants are extremely conservative but are included because over-
sized regenerable systems will still be much more economical than canister systems.
No credits have been included for recovered solvents; if the solvents are suitable for
reuse, a significant reduction in control costs can be achieved. Cost effectiveness, in
terms of cost per ton of VOC controlled, is given in Table 13. Carbon systems will be
more economical than incineration systems in many cases. Note that ink formulations
used in specific plants may be inconsistent with adsorption systems or may require
different, and perhaps more expensive designs.
31
-------
Table 10. Catalytic Incineration - Cost Effectiveness"
Potential
uncontrolled
Cost
Model
emissions
Concentration
effectiveness
plant
(ton/yr)
{% LEL)
($/ton)
1
10
10
$3,900
2
25
10
2,500
2A
25
10
2,800
3
50
25
980
3A
50
25
1,100
3
50
10
1,800
3A
50
10
2,000
4
100
25
1,200
4A
100
25
1,300
4
100
10
1,400
4A
100
10
1,600
5
1000
25
180
5
1000
10
350
LEL = Lower explosive limit (for toluene).
*Cost effectiveness assumes 100 percent capture. If capture efficiency is less than
100 percent, cost effectiveness should be adjusted by dividing by the fractional
capture efficiency.
32
-------
Table 11. Carbon Adsorption Systems-Capital Costs 8
Model
plant
Potential
uncontrolled
emissions
(ton/yr)
25% LEL
10% LEL
Total
Annualized
Total
Annualized
3, 3A
50
$ 77,000
$13,000
$ 77,000
$13,000
4, 4A
100
$110,000
$17,000
$110,000
$17,000
5
1000
$330,000
$53,000
$330,000
$53,000
LEL = Lower explosive limit (for toluene).
aCosts are derived from, Handbook: Control Technologies for Hazardous Air Pollutants,
EPA/625/6-91/014, 1991. Annualized costs are based on 10-year equipment life at 10 percent
interest. Costs for 10- and 25-ton/yr plants are not given because equipment of this size may not
be available. Costs for 50- and 100-ton/yr model plants are based on two adsorbers each with a
working capacity equivalent to 24 hours of emissions. Costs for the 1,000-ton/yr model plant are
based on three adsorbers each with a capacity equivalent to 12 hours of emissions. Smaller
facilities could be controlled with systems suitable for the 50-ton/yr model plant. Costs for 100- and
1,000-ton/yr model plants may be biased high, as shorter cycle times (and smaller absorbers) may
be adequate. Costs have been escalated to fourth quarter 1991 using the Marshall and Swift Cost
Index (18). Capital costs do not include construction of a permanent total enclosure.
33
-------
Table 12. Carbon Adsorption Systems - Operating Costs a
Model
plant
Potential
uncontrolled
emissions
(ton/yr)
% LEL
Capital
(annualized)
Operating
labor
Maintenance
Utilities
Overhead
Total
1
10
25
$(13,000)
$(2,100)
$(9,000)
$(200)
$(11,000)
$(35,000)
1
10
10
(13,000)
(2,100)
(9,000)
(200)
(11,000)
(35,000)
2, 2A
25
25
(13,000)
(2,100)
(9,000)
(500)
(11,000)
(36,000)
2, 2A
25
10
(13,000)
(2,100)
(9,000)
(650)
(11,000)
(36,000)
3, 3A
50
25
13,000
4,300
9,200
1000
(11,000)
38,000
3, 3A
50
10
13,000
4,300
9,000
1300
(11,000)
39,000
4, 4A
100
25
17,000
4,300
9,000
2000
13,000
45,000
4, 4A
100
10
17,000
4,300
9,000
2600
13,000
46,000
5
1000
25
53,000
5,900
14,000
20,000
25,000
120,000
5
1000
10
53,000
5,900
14,000
26,000
25,000
120,000
LEL - Lower explosive limit (for toluene).
aCosts are derived from Handbook: Control Technologies for Hazardous Air Pollutants, EPA/625/6-91/014,1991.
Annualized costs are based on 10-year equipment life at 10 percent interest. Costs for 10- and 25-ton/yr plants have
been estimated as equivalent to costs for 50-ton/yr plants because of the lack of reliable cost data for flows less than
2,000 std ft3/min and very high economies of scale in equipment costs. Estimates assume no credit for recovery of
solvent. Recovered solvents may be worth up to $200/ton» if the composition is suitable for reuse. Electricity cost is
estimated at $0.059/kWh, Carbon costs estimated at $2/lb with a 5-year life. Overhead costs include taxes, insurance,
and administration. Costs have been escalated to fourth quarter 1991 using the Marshall and Swift Cost Index (18).
Annualized capital costs do not include costs of construction of a permanent total enclosure.
-------
Table 13. Carbon Adsorption - Cost Effectiveness 8
Model
plant
Potential
uncontrolled
emissions
(Ton/year)
Concentration
(% LEL)
Cost effectiveness
($/ton)
1
10
25
$3,500 (13,000)b
1
10
10
3,500 (13,000)b
2, 2A
25
25
1,400 (13,000)b
2, 2A
25
10
1,400 (13,000)b
3, 3A
50
25
760
3, 3A
50
10
780
4, 4A
100
25
450
4, 4A
100
10
460
5
1000
25
120
5
1000
10
120
a Carbon capacity has been assumed at 0.6 lb carbon/lb toluene. Estimates for
regenerable systems for 10- and 25-ton/yr model plants are based on capital costs for
50-ton/yr model plants. Actual cost for these plants will be lower. Costs were
estimated using Handbook: Control Technologies for Hazardous Air Pollutants,
EPA/625/6-91/014,1991, and escalated to fourth quarter 1991 using the Marshall and
Swift Cost Index (18). Cost effectiveness assumes 100 percent capture. If capture
efficiency is less than 100 percent, cost effectiveness should be adjusted by dividing by
the fractional capture efficiency.
b The cost of "throwaway" carbon canisters is estimated at $4/lb total.
35
-------
References
1. Vincent, E. J. and W. M. Vatavuk. Control of Volatile Organic Emissions from
Existing Stationary Sources, Volume VIII, Graphic Arts-Rotogravure and
Flexography. EPA-450/2-78-033 (NTIS PB292490). Office of Air Quality
Planning and Standards, Research Triangle Park, NC. 1978.
2. Capone, S. V. and M. W. Petroccia. Guidance to State and Local Agencies in
Preparing Regulations to Control Volatile Organic Compounds from 10
Stationary Source Categories. EPA-450/2-79-004 (NTIS PB80-157290). Office
of Air Quality Planning and Standards, Research Triangle Park, NC. 1979.
3. Strauss, V. The Printing Industry. Printing Industries of America, Inc.
Washington, DC. 1967.
4. National Association of Printing Ink Manufacturers. Printing Ink Handbook.
Fifth Edition. Harrison, NY. 1988.
5. Gravure Association of America. Gravure Process and Technology. Distributed
by Gravure Association of America, 1220A Scottsville Road, RochesterNV
14624. 1991.
6. International Paper. Pocket Pal -- A Graphic Arts Production Handbook.
Fourteenth Edition. International Paper Company, 6400 Poplar Ave. , Memphis,
TN 38197. 1989.
7. Standards of Performance for Magnetic Tape Coating Facilities. 40 CFR 60,
Subpart SS, July 1990, pp 438-444.
8. Sink, M. Handbook: Control Techniques for Hazardous Air Pollutants.
EPA/625/6-91/014. Center for Environmental Research Information, Cincinnati,
OH. 1991.
9. U.S. Environmental Protection Agency. Carbon Adsorption for Control of VOC
Emissions: Theory and Full-Scale System Performance. EPA-450/3-88-012
(NTIS PB91-182006). 1988.
36
-------
10. Vatavuk, W. M. OAQPS Control Cost Manual, Fourth Edition. EPA-450/3-
90-006 (NTIS PB90-169954). Office of Air Quality Planning and Standards,
Research Triangle Park, NC. 1990.
11. Johnson, Bill. Chemical Applications Section, Chemicals and Petroleum
Branch, U.S. Environmental Protection Agency. Memo to Chemicals and
Petroleum Branch and Standards Development Branch Technical Staffs.
July 19, 1982.
12. Spivey, J. J. Catalytic Incineration of Gaseous Wastes. In: Thermal
Processes. H. M. Freeman (ed.), Technomic Publishing (Cincinnati, OH), pp
95-108. 1990.
13. Morris, J. J. Innovative VOC Compliance in the Flexographic and Rotogravure
Package Printing Industry. Presented at the 78,h Annual APCA Meeting, June
1985.
14. Friedman, B. and C. Vaught. Best Demonstrated Control Technology for
Graphic Arts. EPA-450/3-91-008 (NTIS PB91-168427). Office of Air Quality
Planning and Standards, Research Triangle Park, NC. February 1991.
15. Walata, S. A. and C. R. Newman. Radiation-Curable Coatings. EPA-600/2-
91-035 (NTIS PB91-219550). Air and Energy Engineering Research
Laboratory, Research Triangle Park, NC. 1991.
16. DeCelles, R. Decatech, Inc. Personal communication with David Green,
Research Triangle Institute, Research Triangle Park, NC. April 8, 1992.
1390. R. S. Means Company. Building Construction Cost Data, 48th Annual Edition.
18. Economic indicators. In: Chemical Engineering. 99:206, March 1992.
37
------- |