EPA/600/2-89/043
July 1989
SOURCE CHARACTERIZATION AND CONTROL
TECUNOLOGY ASSESSMENT OF METHYLENE CHLORIDE EMISSIONS
FROM EASTMAN KODAK COMPANY, ROCHESTER, NY
by
Stephen A. Walata III
and
Richard M. Rehm
Alliance Technologies Corporation
Chapel Hill, NC 27514
EPA Contract No. 68-02-4396
Work Assignment 13
Project Officer
Charles U. Darvin
Air and Energy Engineering Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
AIR AND ENERGY ENGINEERING RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NC 27711

-------
NOTICE
This document has been reviewed in accordance with
U.S. Environmental Protection Agency policy and
approved for publication. Mention of trade names
or commercial products does not constitute endorse-
ment or recommendation for use.
ii

-------
PREFACE
The Control Air Toxics (CAT) program was funded as a cooperative project
by EPA's Control Technology Center (CTC) and the New York Department of
Environmental Conservation.
The CTC was established by EPA's Office of Research and Development (ORD)
and Office of Air Quality Planning and Standards (OAQPS) to provide technical
assistance to State and Local air pollution control agencies. Three levels of
assistance can be accessed 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.
The technical guidance projects, such as this one, focus on topics of
national or regional interest that are identified through contact with State
and Local agencies. In this case, the CTC assisted the State of New York
Department of Environmental Conservation, identify the scope and nature of
methylene chloride emissions from the Kodak film manufacturing facility in
Rochester, NY. Possible emissions control systems and strategies were also
evaluated. The document discusses control technology for methylene chloride
emission streams similar to those defined at the Kodak facility.
ACKNOWLEDGMENT
The Kodak, Rochester, NY, plant assessment study was prepared for EPA's
Control Technology Center (CTC) and the State of New York Department of
Environmental Conservation (NYDEC) by Stephen A. Walata III and Richard M. Rehm
of Alliance Technologies Corporation. The project officers were Charles H.
Darvin of EPA's Air and Energy Engineering Research Laboratory (AEERL), and
Matthew J. Reis of NYDEC. Also participating in the project was Fred Dimmick
of EPA's Office of Air Quality Planning and Standards (OAQPS).
i i i i

-------
CONTENTS
Page
Preface	 : iii'
Acknowledgment	 ; iii
Figures		vi .
Tables		vi
Metric Conversion Factors		vii :
1.0 EXECUTIVE SUMMARY		1-1
2.0 INTRODUCTION		2-1
3.0 BACKGROUND		3-1
References		3-2
4.0 EMISSIONS		4-1
Introduction		4-1
Registered Emission Points		4-1
Secondary Emissions		4-10
Fugitive Emissions		4-11
References		4-12
5.0 PROCESS CHANGES PLANNED BY KODAK		5-1
6.0 CONTROL TECHNOLOGY ASSESSMENT		6-1
Conclusions		6-3
References				6-16
APPENDICES
A.	Cellulose Triacetate Coated from Methylene
Chloride Solvent Systems		A-l
B.	CAT Analysis of Emissions from Four Points
Using Thermal Incineration and Carbon
Adsorption		B-l
C.	Development of Costs for Using Control Devices
for Vents from Buildings 53 and 20		C-l
D.	Comparison of Technology Assessment with
Kodak's BACT Report		D-l
V i

-------
Number
4-1
FIGURES
Title
Diagram of Roll Coating Machine..
Page
4-6
TABLES
Number Title	Page
1-1	Summary of Expected Emissions Reduction from Possible
Control Technologies		1-4
4-1 DCM Emission Points at Kodak Park Facility		4-2
6-1 Possible Control Technologies for DCM Emissions		6-2
6-2 Control Devices at Kodak Park		6-6
6-3 Sample Calculation for Effluent Concentrations		6-7
6-4 Fugitive Emission Sources at Kodak Park		6-10
6-5 Emissions from Ventilation Sources at Kodak Park		6-11
6-6	Natural Gas Cost for Thermal Incineration
Control Device		6-13
VI

-------
Metric Conversion Factors
Where the metric
following metric conve
FROM
BTU
cu ft
cu ft per min (cfm)
feet (ft)
gallons (gal)
in. H20 (60°F)
ran Hg (32°F)
pounds (lb)
pounds/sq in. (psi)
sq feet (ft2)
TO CONVERT
Degree Fahrenheit (°F)
equivalent is not given in 1
sion factors may be used.
TO
Joule
cu meters (m3)
cu meters per sec ( (m3/sec)
meters (m)
liters (1)
pascal (Pa)
pascal (Pa)
Kilogram (kg)
kilopascal (kPa)
sq meter (m2)
IS
Degree Celsius (°C)
text of this report, the
MULTIPLY BY
1054
0 . 028
0.00047
0 . 305
3.785
248 .8
1.33
0 . 454
6.894
0.093
MULTIPLY
t°c = (t°F - 32 ) / 1. 8
vi i

-------
1.0 EXECUTIVE SUMMARY
This report characterizes emissions and control technologies for reducing
emissions of methylene chloride (also known as dichloromethane, or DCM) at
Eastman Kodak Company's Kodak Park facility in Rochester, NY. From data
provided by Kodak, DCM emissions at the Kodak Park facility total
9,200,000 pounds per year, the largest of any source in the United States.
Kodak uses DCM in the manufacture of cellulose triacetate film support.
The assessment of control technologies for DCM emission sources at
Kodak Park was initiated by New York State's Department of Environmental
Conservation (DEC) in order to bring the facility into compliance with
Title 6, Chapter III, Part 212 of New York State's air pollution regulations.
This was due in part to New York reducing the acceptable ambient level for
methylene chloride from 1167 to 0.37 |J.g/m3, and Kodak's plans to increase
cellulose triacetate film production. DEC requested assistance from EPA's
Control Technology Center (CTC) to independently evaluate control technologies
which might be applied to DCM emissions at Kodak Park. Alliance Technologies
Corporation was contracted by CTC to assist in this evaluation.
This report provides an introduction and background to this project,
describes DCM emissions from Kodak Park, explains process changes indicated to
be planned by Kodak to reduce DCM emissions, and assesses possible control
technologies for reducing DCM emissions.
Work on this project focused on the evaluation of category 1 and
category 2 emission points. According to Kodak, category 1 sources are those
emitting greater than 100,000 pounds of DCM per year, while category 2 sources
emit between 8,000 and 100,000 pounds per year. Of the 181 registered emission
points at Kodak Park, 26 (15 percent) are classified as category 1 or 2. These
sources, however, emit approximately 8,400,000 pound of DCM, or greater than
90 percent of all DCM emissions. During the control technology assessment it
was determined that a substantial number of emission sources had emission
estimates with a low confidence level. Of the 26 existing category 1 or 2
sources, emissions from 11 of the points were estimated by best engineering
judgment. The accuracy of such estimates can be held suspect. Before serious
consideration is given to applying a control device to any of the emission
1-1,

-------
points which were estimated using best engineering judgment, better emission
estimates need to be obtained.
By far, the largest source of DCM emissions at Kodak Park comes from the
production of cellulose triacetate film. In this process, triacetate pellets
are dissolved in methylene chloride and other solvents to form "dope." The
dope is then extruded onto a polished surface to form a thin sheet or web. The
web is then dried at elevated temperatures, driving off the methylene chloride
and other solvents. This process happens within roll coating machines which
are enclosed. While Kodak recycles greater than 95 percent of the DCM used in
this process, 7,380,000 pounds, or over 80 percent of total DCM emissions to
the atmosphere occur from this process. Other sources of DCM emissions include
the Dope Department, where triacetate pellets are dissolved in DCM, the
Distilling Department, where DCM is distilled and recovered, fugitive emissions
from pumps, valves, seals, flanges, etc. within Kodak Park, and secondary
losses from wastewater.
By far, the assessment indicated that the greatest potential for emission
reduction is by controlling leaks from the roll coating machines. Kodak has
proposed to remedy this situation by changing latching devices and gasket
seals, covering bearing casings, and installing solid pipe bulkhead fittings on
the machine casing. These changes were projected by Kodak to reduce DCM
emissions by 3,000,000 pounds per year. This assessment found no reason that
this projection cannot be met. Kodak projects that work in this area will be
completed in 1992. It is believed that this projected work schedule, however,
can be significantly accelerated.
The remaining category 1 and 2 sources at Kodak Park can be divided into
two groups. The first group consists of emission points which are already
controlled, while the second group consists of uncontrolled sources. Emission
controls used by Kodak include carbon adsorbers, dual water/methanol scrubbers,
and condensers. A review of available data indicate that the scrubbers and
condensers are not being operated efficiently, and significant emission
reduction can be achieved by more efficient operations.
The majority of uncontrolled emission points have high flow rates and low
DCM concentrations, making control difficult and expensive. Several points in
this group, however, present situations where Kodak could recover DCM. These

-------
include combining emission sources and adding a scrubber or carbon adsorber.
In addition, emission reductions can also be achieved by controlling solvent
loss from ultrasonic cleaning operations, and institution of a leak detection
and repair program for valves, flanges, pumps, seals, etc. in DCM services.
Table 1-1 provides a summary of potential emissions reduction of emission
sources examined by this report.
f '
\ 1-3
		

-------
TABLE 1-1. SUMMARY OF EXPECTED EMISSIONS REDUCTION FROM POSSIBLE CONTROL TECHNOLOGIES
Emissions
Point (s)
Description
Current	Emissions After
Emissions,	Control,	Percent
lb/yr	lb/yr	Reduction
Reasonable Control
Technology
53-85,53-38,
and 20-68
53-22
142-1
120-7
54-15
52-37	and
54-29
53-32	and
53-96
49-53
Fugitives
Machine Room Exhaust 7,380,000
4,380,000
C.A. for Machine Air
Draw-Off
Solvent Recovery Sys.
Vent Scrubber
78,500	45,700
14,000 Cannot Determine
Still System Vent Scrubber 8,700 Cannot Determine
Building 54 Vent System 23,350	2,350
237,835	23,784
Batch Mixers
Felt Wash Process
Hopper Cleaning
Storage Vessel Vents
Ultrasonic Cleaner
Equipment Leaks
41,900
10,000
650,000
2,095
4,000
390,000
40.6	Improving Seals on
Roll Casting Machines
41.7	Improved Operations
Improved Operations
Improved Operations
Improved Operations
89.9
90
95
60
/ 40
Carbon Adsorber or
Scrubber
Inclusion with Flows for
the 18,000 cfm Carbon
Adsorber
Proper Freeboard Ratio
Freeboard Chiller or
C.A.
s
Leak Detection and
Repair Program

-------
2.0 INTRODUCTION
This report presents the results of an assessment of potential control
technologies for methylene chloride (also known as dichloromethane, or DCM)
emission sources at Eastman Kodak Company's Kodak Park facility in Rochester,
NY. DCM is a solvent used by Kodak in the manufacturing of cellulose
triacetate film support.
The State of New York, Department of Environmental Conservation, requested
EPA's Control Technology Center (CTC) assistance in the evaluation of control
technologies which might be applied to DCM emissions sources at Kodak Park.
CTC is responsible for supporting State and local air pollution control
agencies in the implementation of their programs. Alliance Technologies
Corporation was contracted by CTC to assist in this evaluation. Work has
involved: 1) a plant visit where major DCM emission sources were inspected,
and 2) evaluation of current and potential control technologies for the DCM
emission sources. This report contains information gathered during the plant
visit to the Kodak Park facility. Included are emission estimates determined
by Kodak of all emission points greater than 8,000 pounds of DCM per year, as
well as a description of each point observed during the visit. Also included
in this report is an evaluation of control technologies which might be applied
to the major emission sources. A cost analysis of different add-on control
devices is provided for four of the uncontrolled emission points.
2-1

-------
3.0 BACKGROUND
The assessment of control technologies for DCM emission sources at Kodak
Park was initiated by New York State in order to bring the facility into
compliance with Title 6, Chapter III, Part 212 of New York State's air
pollution regulations. This was in part due to New York reducing the
acceptable ambient level (AAL) for DCM from 1167 ug/m to 0.37 ug/m . Another
factor which lead to this study is Kodak's plans to increase acetate film
support production. New York State's Department of Environmental Conservation
(DEC) needed a means of independently evaluating possible control technologies
which would reduce the DCM emissions from Kodak Park. To this end, the DEC
requested the assistance of EPA's Control Technologies Center in the evaluation
of potential control devices. The goal of this process is to provide the DEC
with independently developed control scenarios for reducing the overall DCM
emission from the Kodak Park facility.
Data for this study was collected during a visit to the Kodik Park
facility on June 15-16, 1988. During the course of these two days, Alliance
personnel were provided with a brief history of film support manufacturing,
review of the processes which involve the use of DCM and emissions data
pertaining to the DCM emission sources. Kodak provided a document in support
of their use of DCM in the manufacturing of cellulose triacetate film support.
This document can be found in Appendix A. Kodak provided schematic diagrams
for the major processes which use DCM. These diagrams, however, will not be
presented in this study due to confidential nature of the information the
diagrams contain. The diagrams did provide Alliance personnel with some
insight as to where in the process emission points were located. The emissions
data provided by Kodak were for all category 1 and 2 emission points.
According to Kodak, category 1 emission points emit greater than 100,000 pounds
of DCM per year and category 2 emission points emit in excess of 8,000 pounds
of DCM per year. Of the 181 registered emission points at Kodak Park in 1987,
26 (15 percent) are classified as category 1 or 2. The total amount of DCM
emitted by the category 1 and 2 sources in 1987 was about 8,400,000 pounds.
This accounts for approximately 90 percent of the 9,200,000 pounds of DCM
released into the atmosphere by the Kodak Park facility in 1987.^ Kodak
provided the emission estimate for each point and background information as to
how each estimate was calculated. A summary of the emission points and amounts

-------
emitted at Kodak Park are listed in the next section. For the most part,
overall DCM emissions from Kodak Park are calculated by means of a material
balance, while the DCM emissions for many specific processes are determined
from best engineering judgement.
Alliance personnel were taken on a tour of the processes which use DCM.
During the plant tour, direct observation of 19 emission points was possible.
This accounts for 82 percent of the total category 1 and 2 emissions.
Information gathered during this part of the plant visit was helpful in the
determination of space constraints surrounding each emission point and provided
a better understanding of how emissions are related to the processes. This
information was valuable in the selection of potential control technologies.
REFERENCES
1. Data supplied by J.D. Mathews, Eastman Kodak Company, to R.M. Rehm and
S.A. Walata, Alliance Technologies Corporation, during the plant visit at
the Kodak Park facility. June 15 and 16, 1988.
3-2

-------
4.0 EMISSIONS
INTRODUCTION
The purpose of this section is to provide a brief description of DCM
emissions at the Kodak Park facility. DCM emissions occurring at this facility
come from either registered emissions points, fugitive losses from equipment
leaks, or secondary emissions from wastewater. The numbers presented in this
section comprise a partial inventory of DCM emissions generated from these
sources.
REGISTERED EMISSIONS POINTS
DCM emissions generated through registered emissions points are from
processes involved with the manufacturing of cellulose triacetate film support.
In keeping with the scope of this study, only the category 1 and 2 emissions
points will be dealt with in this section. As defined earlier, category 1
emission points emit greater than 100,000 pounds of DCM per year, and
category 2 emission points emit in excess of 8,000 pounds of DCM per year. A
list of the emission points consider in this section is presented in Table 4-1.
At the Kodak Park facility, three departments (Dope, Roll Coating and
Distilling) are the main users of DCM. The following is the emission points
within each department.
Dope Department
The Dope Department is where cellulose triacetate pellets are dissolved
into a blended solvent, the major constituent of which is DCM, to form a liquid
polymer referred to as "dope." The dope is then processed for further use in
the Roll Coating department. This department has seven emission points which
are of interest to this study. The largest of these points is the emissions
resulting from the charging of batch mixers (point 52-37) with cellulose
triacetate. In past years, all of the cellulose triacetate was dissolved in
this fashion, but since the advent of continuous mixers, batch mixers account
for only a small percentage of dope generation. Emissions of DCM of the
category 1 level occur when the batch mixers are being charged. The solvent is
4-1

-------
TABLE 4-1.
DCM EMISSION POINTS AT KODAK PARK FACILITY
STACK FLOW EMISSIONS (Ib/yr)
SOURCE DESCRIPTION CATAGORY
DEPT. HEIGHT
LOCATION (ft)
RATE
(CFM)
ACTUAL
YEARLY
AVG.
HR.
PEAK
HR
METHOD OF
ESTIMATION
53-85
EXISTING MACHINE ROOM EXHAUST
1
ROLL COATING
80
250000
4700000
537
700
Monitoring

53-38
EXISTING MACHINE ROOM EXHAUST
1
ROLL COATING
80
125000
1300000
150
200
Monitoring

20-68
EXISTING MACHINE ROOM EXHAUST
1
ROLL COATING
60
150000
1380000
158
420
Best Engineering
Estimate
52-37
BATCH MIXERS
1
DOPE DEPT.
70
1110
193680
66.3
66 3
Best Engineering
Estimate
53-K1
FLOOR SWEEPS FOR SOLVENT RECOVERY ROOM
1
ROLL COATING
35
20000
150000
17.2
20 4
Monitoring

53-88
FLOOR SWEEPS - DOPE DEPT.
2
DOPE DEPT.
85
34400
86724
9 9
10 2
Best Engineering
Estimate
53-22
EXISTING C.A. FOR MACHINE AIR DRAW-OFF
2
ROLL COATING
100
4000
78500
9
9 6
Monitoring

53-08
FILTER PRESS CHANGING
2
DOPE DEPT.
83
15000
75680
104
104
Monitoring

329-2
2ND DRYER TO CARBON ADSORBER
2
ROLL COATING
61
27500
45786
35 3
177
Material Balance

54-29
FELT WASH PROCESS (5 FILTERS)
2
DOPE DEPT.
35
1
44155
40.3
42.5
Best Engineering
Estimate
317S0 305 MACHINE CARBON ADSORBER EXHAUST
2
ROLL COATING
82
18000
42670
119
11.9
Material Balance

53-K2
EAST END FLOOR SWEEPS
2
ROLL CASTING
85
9000
36500
4.2
5.0
Monitoring

21-12
KADY MILL EXHAUST
2
ROLL COATING
62
5500
35580
106
201.5
Best Engineering
Estimate
322-4
KPM SOLVENT RECOVERY REFIG. CONDENSER
2
DISTILL
30
15
32000
3 7
97.6
Monitoring

329K2 VESSEL CLEANING EXHAUST
2
ROLL COATING
46
4100
27450
18.3
18 3
Best Engineering
Estimate

* Estimate based on the monitoring results for 53-85 and 53-38





(Continued)


-------
TABLE 4-1. (CONTINUED)
DCM EMISSION POINTS AT KODAK PARK FACILITY
STACK FLOW EMISSIONS (Ib/yr)



DEPT. HEIGHT
RATE
ACTUAL
AVG.
PEAK
METHOD OF
SOURCE DESCRIPTION CATAGORY
LOCATION
(ft)
(CFM)
YEARLY
HR.
HR
ESTIMATION
D63-5
STEAMER CHARGING EXHAUST
2
DISTILL
2B
3550
26280
3
3
Best Engineering Estimate
54-15
BUILDING 54 VENT SYSTEM
2
DOPE DEPT.
70
5 5
23350
2 67
25
Best Engineering Estimate
53-96
STORAGE VESSEL VENTS
2
ROLL COATING
75
1
22900
2.6
4 8
Best Engineering Estimate
53-32
HOPPER CLEANING AND FLOOR SWEEPS
2
ROLL COATING
77
11200
19000
2 1
110
Monitoring
52-M2
WEST WALL EXHAUST FAN - 1ST FLOOR
2
DOPE DEPT
B
6000
17800
2
2
Best Engineering Estimate
53-92
SOLVENT DYE MIXING (FLOOR SWEEP SYSTEM)
2
ROLL COATING
80
12000
15000
1.7
6 4
Best Engineering Estimate
52-M3
WEST WALL EXHAUST FAN - 2ND FLOOR
2
DOPE DEPT.
30
5300
14500
1.7
17
Best Engineering Estimate
142-1
FILM SOLVENT RECOVERY SYS. VENT SCRUBBER
2
DISTILL
48
9
14000
1.9
1.9
Monitoring
317S2
306 MACHINE CARBON ADSORBER EXHAUST
2
ROLL COATING
88
40000
10000
27.8
27.8
Material Balance
49-53
ULTRASONIC CLEANER
2
ROLL COATING
65
1746
10000
1.14
11
Material Balance
120-7
STILL SYSTEM VENT SCRUBBER
2
DISTILL
91
3
8700
1.2
1.2
Monitoring
TOTAL 8410255
Material Balance The known flows of DCM into and out of a process are subtracted from one another to yield an estimate of DCM air emissions.
Best Engineering Estimate Calculations are mode on physical/chemical properties of the materials present, and standard engineering equations. If insufficient
Information Is available then engineering judgment and experience with similar processes are used to estimate DCM emissions.
Monitoring Samples are taken and analyzed for DCM concentration The concentration Is combined with air flow measurements to yeild an
estimate of DCM emissions.

-------
first loaded into the mixer, then the top is opened and the cellulose
triacetate is poured in. Solvent vapors are released when the solids are added
to the fluids. The charging ports are fitted with a cooling jacket to condense
solvent vapors and reduce possible emissions during the loading operation. The
vapors that escape from the charging ports are then discharged into the
atmosphere after passing through a cyclone to remove particulate matter. The
equations used by Kodak to estimate the DCM emissions from this operation were
developed some fifty years ago when the batch mixers were utilized more often.
These emissions estimates have not been revised although the use of the batch
mixers has decreased. Kodak personnel believe that the emissions estimates
they have reported are greater than current actual annual DCM emissions. Given
the reduced use of the batch mixers and that the estimation equation was
developed without monitoring data to confirm its accuracy, Alliance agrees with
the assessment that the reported emissions are probably an overestimation.
Two other emission points of interest are the result of two different
filtering operations during the dope refining process. The emissions from the
filtering operations occur when the filter elements are being changed. The
largest of these two sources is the filter press changing operation
(point 53-08). A series of filter presses are located in Building 53 and when
the elements are changed any DCM vapors are force vented to the atmosphere.
This is to keep the ambient level of DCM in the building at a safe level for
the workers. The other filtering operation which results in DCM emissions is
the felt wash process (point 54-29). The felt wash process removes solid
matter from the excess solvent before the solvent is placed in a storage tank
to await transport to the Distilling Department. The felt washers are located
on a catwalk outside of Building 54. When the filter elements are changed, the
DCM vapors escape directly into the atmosphere. Kodak personnel estimate that
two felt wash tanks are changed per day.
The remaining four emission points in the Dope Department deal with the
general ventilation of the buildings housing the department and the venting of
storage tanks. The largest of these points is the floor sweeps in the Dope
Department (point 53-88). The floor sweeps are ducted vents on floor level
which draw off floor level concentrations of DCM vapors. These vapors are then
vented into the atmosphere through a vent on the roof. The floor sweeps in the
Dope Department are located around the filter presses and pumps. Another
emission point is the vent system which receives emissions from ten storage
r
4-4

-------
tanks, each of which has a capacity of ten thousand gallons (point 54-15). The
control equipment on this emission point includes a condenser and a nitrogen
blanket which are used to recover DCM vapors caused by the emptying and filling
of the tanks. The remaining two emission points in the Dope Department are
exhaust fans in the west wall on the first and second floors of Building 52
(points 52-M2 and 52-M3). The exhaust fans are used for general ventilation
purposes and emit DCM directly to the atmosphere.
Roll Coating
The Roll Coating department accounts for 93.6 percent (7,873,386 lb/yr)
of the category 1 and 2 DCM emissions at Kodak Park through 16 registered
emission points. Kodak personnel estimate that 7,702,350 lb/yr of DCM
emissions are associated with the process of film base casting. Film base
casting is the operation where the cellulose triacetate dope is extruded on a
polished surface to form a thin sheet or web. The web is then dried at
elevated temperatures. Three registered emission points associated with this
process (points 53-85, 53-38 and 20-68) accounts for over 80 percent
(7,380,000 lb/yr) of the total DCM emissions at Kodak Park each year. These
emission points are the exhaust stacks for the ventilation system for the film
base casting rooms in Buildings 53 and 20. Kodak employs a simple method to
ventilate the rooms which contain the roll casting machines. Air is forced-
into the base of the building and is vented to the atmosphere along with the
DCM through the roof. Powered exhaust fans are present in Building 53
(point 53-85) which assist in ventilating the building. The air flow is aided
in its upward motion by the "chimney effects" which occur between the casting
machines due to the high temperatures. Information of the ventilation system-
of each casting room is presented in Table 4-1. The large volumes of air being
pumped through each casting room serves two purposes: 1) to maintain an
ambient level of DCM within OSHA guidelines, and 2) to provide comfortable
working conditions around the film base casting machines.
The DCM emissions in the ventilation exhaust are what Kodak terms
"captive fugitives." These are fugitive emissions from the casting machines
which are released into the room before they are vented to the atmosphere
through the emission points. Figure 4-1 shows the general design of a roll
casting machine. Each machine has various areas where DCM could be released
4-5

-------
Webbing
Observation
Port
	J
Wind —up
Casting Wheel
Figure 4—1. Diagram of Roll Coating Machine

-------
into the casting room. The cumulative emissions from the casting machines in
Building 53 constitutes the single largest source of DCM emissions at Kodak
Park. The area labeled "A" in Figure 4-1 is the hopper compartment where the
dope is extruded on and later stripped off the polished surface or casting
wheel. This is where the DCM vapors are the most concentrated since almost all
of the DCM in the dope is being vaporized in the formation of the web. There
are vents located in this hopper where the web is stripped off of the casting
wheel. These vents capture DCM vapors for solvent recovery. Seals around the
observation and access ports as well as openings in the machine's casing for
hoses and tubes are a source of DCM emissions in this hopper of the machine.
This area of the machine may very well be the source of a majority of the DCM
fugitive emissions generated from the process. Areas labeled "B"/'C", and "D"
are hoppers in which application solutions are applied to the web. These
solutions may or may not contain DCM. There may be DCM emissions from the
observation and access ports but not on the same level as the first hopper
since most of the DCM should be vaporized when the web was formed. The areas
labeled "E" are the bearing housings for the idler rollers located throughout
the casting machine. This area is not sealed and Kodak personnel feel that DCM
can and does escape through the bearings. Alliance personnel concur that there
is a potential for fugitive emissions to occur through the bearings since they
are not sealed. The areas labeled "F" are the locations of pressure relief
vents which are needed in the unlikely event of an explosion inside the
machine. These panels are secured to the machine housing with double backed
tape. Due to the nature of this design, it is not uncommon to have leaks from
these panels from time to time. The area labeled "G" is the wind-up of the web
into a roll for storage purposes. This area is effectively a large opening in
the end of the machine where Kodak personnel feel that some DCM does escape
from the machine. Alliance believes that this area is a minor source of DCM
emissions from the casting machine.
In each of the film base casting machines are vents which capture DCM
vapors for solvent recovery purposes. Although the casting machines lose what
seem to be a large amount in fugitive emissions, most of the DCM used in the
cellulose triacetate film manufacturing is captured for reuse. From the data
provided by Kodak, Alliance estimates that approximately 95 percent of the DCM
used is recycled.^ The solvent recovery system for film base casting is a
closed system which circulates air as a carrier gas from the casting machine to
4-7

-------
a brine condenser and back to the casting machine. The brine condenser
operates at a temperature of -88°F. This causes the DCM in the air to return
to the liquid phase for transportation purposes to the Distilling Department.
After the brine condenser on several of the casting machines, a side stream of
air containing DCM is removed from the air flow and passed through a carbon
adsorber. The reason for removing the side stream is to allow pressure
balancing in the casting machines. The reason for passing the side stream
through the carbon adsorber is to recover the DCM in the side stream. Kodak
personnel reported that the carbon adsorber is 95 percent efficient in the
removal of DCM from the air flow. Thus, the exhaust from the carbon, adsorber
contains DCM and is considered an emission point (point 53-22). There are also
DCM emissions which are caused by equipment leaks around the brine condensers.
These emissions are collected by floor sweeps in the west end and east end of
Building 53 (points 53-K1 and 53-K2). These floor sweeps are vented directly
to the atmosphere. The amount of DCM discharged by 53-K1 and 53-K2 are 150,000
and 36,500 pounds per year respectively. These estimates were derived from a
method of estimation which used monitor data.
There are three other emission points which are directly associated with
the casting process. The first is the vents system on storage tanks
containing surface coating solutions (point 53-96). The solutions are prepared
in the morning and used during the course of the working day and may not always
contain DCM. The vent on each of the tanks is a "U" shaped duct which is
vented through the roof to the atmosphere. The second point consists of the
vents and floor sweeps in the hopper cleaning area (point 53-32). In this area
the parts which extrude the dope onto the casting wheel are cleaned by soaking
in a bath of DCM. There are vents which surround each bath and floor sweeps
to collect and DCM vapors at floor level. The DCM collected by these vents is
then exhausted through the roof to the atmosphere. The last emission point is
associated with the area for solvent dye mixing (point 53-92). This area
consists of several 500 gallon tanks where the dye is prepared for use on the
web. The ventilation in this area is a localized exhaust which is vented
through the roof to the atmosphere.
There are several other emission points which are due to processes which
use DCM. Three of these points are the exhaust from carbon adsorbers. Two of
the carbon adsorbers are located in Building 317 (points 317SO and 317S2) and
the third is located in Building 329 (point 329-2). These carbon adsorbers are
h-8

-------
used as the solvent recovery system in surface coating processes. In a surface
coating process, application solutions are applied to rolls of various film
substrate. Air is used as a carrier gas for the DCM as it vaporizes from the
web. For some film products, DCM is not used in the application solution.
Kodak personnel reported that the carbon adsorbers are 95 percent efficient in
removing DCM from the air flow and thus will discharge DCM into the
2
atmosphere. Another point located in Building 329 is the exhaust from the
vessel cleaning area (point 329K2). DCM is used to clean out tanks which
contain the application solution. The emissions of DCM only occur when the
vessel is open for inspection. These emissions are collected by a localized
venting system and discharged to the atmosphere.
Other emission points associated with manufacturing operations are the
Kady Mill exhaust (point 21-12) and the ultrasonic cleaner (point 49-53). The
Kady Mill is a grinding operation which uses DCM in the grinding process. DCM
is also used to clean the vessel. The DCM emissions from the Kady Mill occur
when the vessel is opened for inspection. The DCM emissions are then collected
by a "ll" shape vent at the rim of the vessel, channeled through the roof and
discharged into the atmosphere. Kodak personnel informed us that the Kady Mill
in recent years has only operated for a two week period each year. They feel
that the emissions reported from this operation are an over estimation of
current operations. The ultrasonic cleaner uses DCM to remove particulate
matter from photo-sensitive glass plates. The operation of this process is
analogous to a degreasing operation. Current controls for DCM emissions on the
Ultrasonic cleaner include a primary condenser to condense DCM vapors, and
draining the unit when not in use. A collection hood is present above the
cleaner to capture emissions and is combined with floor sweeps from the room to
a common duct and vented to the atmosphere through the roof.
Distilling Department
The Distilling Department is the area at Kodak which refines used DCM
into a product which is reusable in the acetate film manufacturing. DCM
emissions from the four points in this department are all of the category 2
classification. The DCM refining is accomplished, as the name implies, by the
distillation of the used DCM to separate it from the other liquid components
which may be present. The Distilling Department has two areas which are in
4-9

-------
DCM service, the KPW site and the KPM site. The KPW site handles the bulk of
the DCM processed in a given year because it handles the used DCM from the
acetate film base manufacturing areas. The KPM site handles the used DCM from
the several surface coating processes from Buildings 317 and 329. Each site
has a tank farm for storage of the used and refined DCM and the distillation
columns. The storage tanks and distillation columns are vented to
methanol/water scrubber systems. The methanol scrubber is used to remove the
DCM present in the air flow, while the water scrubber removes light, water
soluble solvents or methanol remaining in the air flow. Two of the scrubber
systems are located at the KPW site. One system handles the vent, flows from
the tank farms located there and the distillation columns located in
Building 120 (point 120-7). The other system is for the vent flows from the
distillation columns and the DCM containing process tanks located just adjacent
to Building 120 (point 142-1). A refrigerated vent condenser system is located
at the KPM site where it handles the vent flows from the storage tanks and
distillation columns located there. Kodak estimates that these vent control
devices are 90, 94 and 86 percent effective respectively in the removal of DCM
from the vent flows.
The one other DCM emission point worth noting in the Distilling
Department is the steamer charging exhaust (point D63-5). At th-! Kodak Park
facility, liquid streams containing DCM and solids are treated in the steamer
in order to recover the DCM. These streams are transported from their point of
origin in portable tanks and placed in a steamer pot where steam removes the
DCM for further treatment in the distillation process. The remaining solids
and water are then transported to Kings Landing Wastewater Treatment facility
for further treatment before discharge to the environment. When the solids are
transferred from the portable tanks to the steamer pot, DCM emissions occur;
These emissions are collected by a localized venting system and discharged into
the atmosphere.
SECONDARY EMISSIONS
The Kodak Park facility has its own wastewater treatment facility located
on the banks of the Genesee River at Kings Landing. This facility handles 30
million gallons of industrial wastewater per day from Kodak Park. The system
4-10

-------
is acclimated for chlorinated organics, and a study done in 1986 using an on-
line gas chromatograph showed an 80 percent degradation of DCM from the
influent concentration. There are, however, certain areas that have a
potential for generating secondary emissions from the treatment facility. The
two areas with the greatest potential are the grit chamber and the distribution
section to the primary settling tanks. These are two areas where turbulent
water is exposed the atmosphere prior to any biological degradation. Kodak
estimates the secondary emissions from Kings Landing are 66,000 lb/yr. Kodak
personnel at the time of the inspection could not provide detailed information
on the sources of DCM in the wastewater streams.
FUGITIVE EMISSIONS
At the Kodak Park facility, fugitive emissions of DCM are divided into
two categories. This first category is designated by Kodak as "captive"
fugitive emissions. These emissions, as defined above, are those generated
from equipment located in a building and are contained by the building before
being vented to the atmosphere through a registered emission points. The
emission estimates from this type of fugitive emissions were discussed in the
sections above.
The other type of fugitive emissions at Kodak Park are of a more
"standard" nature. These emissions are the result of leaks from pumps, valves
flanges, seals, and other equipment which is in DCM service and are discharged
directly to the atmosphere. Kodak estimates that 650,000 lb/yr of DCM are
emitted to the environment through this type of fugitive emissions. Kodak made
this estimation through a plant level material balance of DCM. Kodak personnel
however have not made a count of the number of pumps, valves, flanges, seals,
etc. in DCM service in order to characterize the fugitive DCM emissions
occurring from Kodak Park.
4-11

-------
REFERENCES
1.	Data supplied by J.D. Mathews, Eastman Kodak Company, to R.M. Rehm and
S.A. Walata, Alliance Technologies Corporation, during the plant visit of
the Kodak Park facility. June 15 and 16, 1988.
2.	Letter from B.M. Wirsig, Eastman Kodak Company, to S.A. Walata, Alliance
Technologies Corporation. September 12, 1988.
4-12

-------
5.0 PROCESS CHANCES PLANNED BY KODAK
During the course of the plant visit, Kodak personnel made a presentation
of process changes which were being planned. All of the process changes
discussed were associated with the film base casting process. As described in
Section 4, emissions from the film base casting machines account for 7,380,000
pounds per year, or over 80 percent of entire DCM emissions from Kodak Park.
Changes include modification to the film base casting machines and the
replacement of the carbon adsorber used to pressure balance several machines
with a large capacity unit.
One of problem areas in the roll casting machines is the DCM leaking from
the observation and access ports due to the lack of air tight seals. Kodak is
proposing to remedy this problem by changing the latching device and gasket
seals. In the areas of the machine which have high concentrations of DCM
vapors, Kodak is planning on replacing the current latches with a spring
loaded type which has a leverage feature built in. This will allow for the
port to be cranked down tight. The gaskets in these areas will be replaced
with a bladder type gasket which will allow for a pressurized seal; The
pressurized seal would be achieved by inflating the bladder type gasket with
air. This would result in a tight, custom fit between the port and machine
housing. Latches in the other areas will be replaced with screw type latches
for a tight fit. Another problem area is the DCM leakage through the bearing
casings of the idler rollers. Kodak plans to fit each bearing casing with a
"bubble cap" which will seal each bearing. Kodak also plans to install solid
pipe bulkhead fittings on the machine casing. This will seal up the openings
in the machine's casing which are required for the various tubes and pipes
which supply the application solutions for the web. Also discussed were
possible changes which could be made for the pressure relief vents. Kodak is
in the process of reviewing the type of tape used to attach the panels for the
vents to the machine casing as well as the installation technique used. Kodak
is also looking at other means of securing these panels. Another possible
modification to the casting machine design would be to build a buffer between
the last curing hopper and the web wind up using a hot air purge system. Kodak
plans to design this feature into their new casting machine but it is unclear
at this time if the older machines would be retrofitted with this feature.
5-1

-------
Kodak personnel also discussed plans for the replacement of the carbon
adsorber used in the Roll Coating Department to pressure balance several of the
film base casting machines in Building 53. The current carbon adsorber has a
capacity of 4,000 cfm. This allows Kodak to pressure balance only six of the
nineteen machines located in Building 53*s casting room. The new carbon
adsorber would have a capacity of 18,000 cfm and would allow all the casting
machines to be pressure balanced. This increased flow capacity would allow
Kodak the latitude to control emissions from currently uncontrolled sources.
During the visit to the hopper cleaning area in Building 53, Kodak personnel
indicated that there were plans to include the vents from the hopper cleaning
baths in the flows to the new carbon adsorber.
Kodak estimates that DCM emissions will be reduced by 3,000,000 pounds per
year by controlling fugitive losses from the film base casting machines and by
installation of the new carbon adsorber. According to Kodak's p.esent
timetable, all film base casting machines will be retrofitted with the
equipment described above by 1992. The new carbon adsorber will be on-line by
1991.
5-2

-------
6.0 CONTROL TECHNOLOGIES ASSESSMENT
The purpose of this study is to provide a broad analysis of the control
technologies used to reduce DCM emissions and the feasibility of their use at
Kodak Park. A cost analysis is also provided for the controls on four emission
points at Kodak Park. This analysis makes use of the EPA Handbook, "Control
Technologies for Hazardous Air Pollutants."^ The handbook describes two types
of devices which can be used to control DCM emissions: combustion and solvent
recovery.
The main advantage of using combustion control devices is that they can
provide a high control efficiency of DCM emissions. Combustion type control
devices cause the destruction of a compound through oxidation of the molecular
bonds. A list of control devices of this type is presented in Table 6-1. When
hydrocarbons are oxidized, the resulting compounds are ideally water and carbon
dioxide. This, however, is not the case when DCM is oxidized due to the fact
DCM is a halogenated hydrocarbon. Basic stoichiometry of the oxidation reveals
the chlorine atoms present in DCM will bond with free hydrogen atoms to form
hydrogen chloride (HC1). The formation of HC1 would require the addition of an
extra control device to the original control device, since HC1 is also
considered a potentially hazardous air pollutant. The corrosive nature of HC1,
when in the presence of water, would also require the control devices to be
constructed of high grade materials which are resistant to corrosion. These
two requirements would increase the initial capital cost for the control
device. Another problem with the combustion of DCM is the energy requirements
to cause its oxidation. One reason for DCM being the solvent of choice in the
manufacturing of acetate film is its non-flammable nature. This property
causes the need for a fuel to be added to the air stream to facilitate
combustion for thermal incineration. The cost of the fuel for combustion or
heat will add to the annual operating cost of the control devices. On top of
this the cost of replacing the lost DCM must also be taken into account. This
cost would normally be incurred for operations in their present state of not
having controls. Since this analysis is reviewing the options of whether best
control technology for DCM emissions is through combustion or recovery devices,
the cost of replacement DCM must also be considered.
Recovery type control devices are those which physically remove a
compound, such as DCM, from an emission stream and convert the compound into a
6-1

-------
TABLE 6-1. POSSIBLE CONTROL TECHNOLOGIES FOR DCM EMISSIONS
COMBUSTION TYPE CONTROL DEVICES
INCINERATION
Thermal
Catalytic
FLARE
BOILER/PROCESS HEATER
RECOVERY TYPE CONTROL DEVICES
CARBON ADSORBER
SCRUBBER
CONDENSER
VAPOR RETURN
EQUIPMENT MODIFICATIONS
MODIFICATIONS TO OPERATING PROCEDURES
SPRING LOADED LATCHES
BLADDER GASKETS
CAPPED BEARINGS
SEALING TO PRESSURE RELIEF VENTS
6-2
¦ "	.y

-------
form for future use, usually in a liquid form. A list of control devices of
this type is also presented in Table 6-1. The means of physically removing DCM
can be accomplished by adsorption, absorption or through a phase change which
occurs when using a condenser. Each of these devices requires energy at some
point of their operations. Absorption type control devices tend to be energy
passive while condensers are more energy intensive. The main problem with
control devices of this type is the recovery of other compounds other than DCM.
Since the DCM recovered is to be used again, it must be separated from the
other compounds. This step requires its own pieces of equipment and their
associated costs in initial investment and operation. If such separation
equipment already exists on site, then consideration must be given to whether
the stream from this control equipment would exceed the capacity of this
equipment. Also, consideration must be given to whether this new stream
containing DCM would increase the current cost of refining DCM. The main
advantage of using recovery control devices is in the fact that every pound of
DCM recovery means one less pound which would have to be purchased to replace
that which is lost in the emission stream.
When considering a control device for any of the category 1 and 2 emission
points, one must pay special attention to the space requirement of the control
device along with the other engineering considerations. The buildings where
the majority of DCM usage takes place (Buildings 20, 52, 53 and 54) are in the
older section of Kodak Park. Subsequent urban growth by the City of Rochester
has greatly reduced the amount of available land in that area for further
construction. This means that any control device selected should be housed
either in an existing facility or a facility which can be constructed on land
already owned by Kodak.
CONCLUSIONS
It is difficult to perform a BACT analysis on the DCM emission points at
Kodak Park. The first problem encountered is a substantial amount of emission
data with a low confidence level. Of the 26 existing category 1 and 2 emission
points, 11 of these points had their emissions estimated by best engineering
judgement. The accuracy of such estimates can be held suspect. There were at
least two instances (emission points 52-37 and 21-12) where Kodak personnel
suggested that the emission estimates provided for each point was higher than
6-3

-------
the actual emissions. Because total DCM emissions from Kodak Park are based on
mass balance, this raises the question if the emissions from points 52-37 and
21-12 are actually lower than what Kodak says, are there points for which the
actual emissions are greater than the estimates report? This provides the
possibility that the analysis process suggest a control device for an emission
source which may be impractical to control while ignoring another emission
source that deserves consideration for control. Before serious consideration
of applying a control device to any of the emission points which were estimated
by best engineering judgement, better emission estimates need to be obtained.
This effort should not be too difficult since all but three of these points are
the result of captured uncontrolled fugitives being vented to the atmosphere
through a duct. A monitoring program can be established for each point on a
short-term basis to develop a better means of estimating emissions from each
point. Kodak should also should evaluate the methods used to estimate the
emissions from category 3 emission points. To do this would ensure that no DCM
emission point is misclassified, and that all DCM emissions estimates are
determined by the best possible means.
Another complication to the BACT analysis of the DCM emission points is
the fact that over 80 percent of the total DCM emitted at Kodak Park is the
result of three emission points (points 53-85, 53-38 and 20-68). These are the
emission points associated with the ventilation of the film base casting rooms
in Buildings 53 and 20. The flowrates through these rooms are such that any
control device applied to controlling DCM emissions would have to be large
enough to require its own building. To lower the air flowrates through these
rooms in order to use a smaller control device would put Kodak personnel's
health in jeopardy. A control device for these emission sources will be
considered later in this section as part of an overall control strategy for
Buildings 53 and 20. A more effective way of controlling the DCM emissions
from these two rooms would be to control the emissions at the source, the film
base casting machines. As discussed earlier Section 5, Kodak has made plans
for modification in the sealing of the film base casting machines. Kodak
personnel estimate the reduction in DCM emissions as a result of these
modifications would be approximately 3,000,000 pounds of DCM per year. This
would represent a AO.6 percent reduction in the DCM emissions from the three
points and an overall reduction of DCM emissions from Kodak Park of 33 percent.
Alliance believes that the current schedule for modifications can be
6-4

-------
accelerated and recommends that Kodak complete this project at the earliest
possible date. Kodak should also continue in their efforts to find more
efficient ways of sealing the film base casting machines concentrating on the
observation and access ports, pressure relief vents, bearing seals and seams in
the machine casing. The goal of this program should be to provide an air tight
casing for the film base casting process.
Alliance personnel lack the expertise of Kodak personnel regarding the
film base casting machines and thus can not accurately provide an independent
confirmation of the reduction estimates. Common sense, however, dictates that
if the emissions in question are caused by leaks in the film base casting
machine casing, then sealing the leaks will cause a reduction in the emissions.
Kodak will have the means to verify the success of their modification program
and should do so. As of this study, Kodak is installing a continuous
monitoring system in the attic and vent ports of the casting room in
Building 53. They have already used monitoring data from a prototype system in
this area to estimate the emissions from this room. Once installed, a
continuous monitoring system should provide Kodak with the ability of
determining whether modification made to the film base casting machines result
in the reduction of emissions. A similar system should be installed in the
attic space of the film base casting room in Building 20. Since the emission
from the casting rooms in Building 53 and 20 constitute a majority of DCM
emission at Kodak Park, the emission estimates made by Kodak should have the
highest degree of confidence. The remaining 23 category 1 and 2 DCM
emissions points at Kodak Park can be divided into two groups. The first group
consists of emission points which are the effluent streams from recovery
devices already in place. A list of these emission points is presented in
Table 6-2 along with the concentration of DCM in the effluent stream. Four
emission points are effluent streams from carbon adsorbers (points 53-88, 329-
2, 317S0 and 317S2), two are effluent streams from packed bed scrubbers (points
142-1 and 120-7) and two are the effluent stream from a condenser (points 322-4
and 54-15). The concentration of DCM in the effluent streams were calculated
from the yearly emission and flowrate data provided by Kodak during the plant
visit. Each device was assumed to be in operation year round, 24-hours a day.
A sample calculation for the effluent concentration is presented in Table 6-3.
The concentration values presented for the scrubbers and condensers can only be
considered approximate values for reasons described later in this section.
6-5

-------
TABLE 6-2. CONTROL DEVICES AT KODAK PARK
Source
Description
Flowrate,
cfm
Actual
Emissions,
lbs/yr
Cone.,
PPm
53-22
EXISTING C.A. FOR MACHINE AIR DRAW-OFF
4,000
78,500
172
329-2
2ND DRYER TO CARBON ADSORBER
27,500
45,786
15
317S0
305 MACHINE CARBON ADSORBER EXHAUST
18,000
42,670
21
322-4
KPM SOLVENT RECOVERY REFIG. CONDENSER
15
32,000
18,714
54-15
BUILDING 54 VENT SYSTEM
6
23,350
34,139
142-1
FILM SOLVENT RECOVERY SYS. VENT SCRUBBER
9
14,000
13,646
317S2
306 MACHINE CARBON ADSORBER EXHAUST
40,000
10,000
2
120-7
STILL SYSTEM VENT SCRUBBER
1
8,700
76,319

-------
TABLE 6-3. SAMPLE CALCULATION FOR EFFLUENT CONCENTRATIONS
SAMPLE EMISSION POINT 53-22
78,500 lb/yr = 0.149 lb/min
CONCENTRATION
(0.149 lb/min) / (4,000 ft3/min) = 3.733E-5 lb/ft3
3.733E-5 lb/ft3 = 0.598 g/m3
0.598 g/m3 x 288" = 172.16 ppm
* See Reference 4.
6-7

-------
While these emission points are controlled, this does not mean further emission
reductions cannot be achieved. The carbon adsorbers, with the exception of the
one, have effluent concentrations which are within the achievable concentration
defined by the EPA. This effluent concentration is defined by EPA as being
less than 100 ppm. The carbon adsorber which has an effluent concentration
greater than 100 ppm (point 53-22) is associated with the side stream being
pulled off of six roll casting machines in Building 53 as a means to pressure
balance these machines. During the plant visit, Kodak personnel informed us of
plans to replace this carbon adsorber with another which will have a flow
capacity of 18,000 cfm (the present carbon adsorber has a flow capacity of
4,000 cfm). This will give Kodak enough capacity to pressure balance all the
casting machines located in Building 53. Since this new carbon adsorber will
not be on line for several years, Kodak personnel should examine the operating
parameters of the currently operating carbon adsorber to see if the effluent
concentration can be reduced to at least 100 ppm. If the effluent
concentration can be reduced from 172 ppm to 100 ppm, over the course of a year
DCM emissions from this point will be reduced by 32,800 pounds.
The data supplied for the packed bed scrubbers operating in the Distilling
Department indicate that their current use is not resulting in the maximum
reduction DCM from the respective emission streams. The first indication can
be found in the effluent concentration from each scrubber system. From the
data provided by Kodak, the concentration of DCM in the emission streams of
points 142-1 and 120-7 following control are 24,574 and 76,354 ppm
respectively. The manual that EPA provides for the sizing of control devices
suggests that a scrubber system best works when the influent concentration is
no greater than 10,000 ppm. Since the effluent concentration from the
scrubber is over 10,000 ppm, then one must assume that the influent
concentration is over 10,000 ppm. Another problem with the scrubber systems
can be found in the fact that each is being used as a control device for vent
streams which can have highly fluctuating concentrations of DCM. Such is the
case as with the scrubber associated with emission point 120-7. This scrubber
controls the vent flows from the storage tanks around Building 120. The
concentration of DCM in these vent flows can approach the saturation point
(this concentration is defined as being 550,000 ppm).^ A concentration of this
magnitude could easily overwhelm the scrubber's ability to effectively reduce
the DCM concentration from the influent stream. This problem can be corrected

-------
by adding a dilution stream to the influent stream. This would allow the
influent concentration to be reduced to the level recommended by EPA.
The condenser associated with emission point 54-15 is another control
device currently operating at Kodak Park which we feel can achieve a better
removal efficiency. According to data provided by Kodak, this condenser has a
removal efficiency of 50 percent."* From the yearly emission and flowrate data
provided during the plant trip and assuming 24-hours a day year round
operation, the concentration of DCM in the condenser's effluent would be
approximately 37,242 ppm. Based on the removal efficiency provided by Kodak,
the influent concentration to the condenser is estimated to be 74,485 ppm. The
EPA's handbook for sizing control devices states that a properly sized
condenser working with a influent concentration greater than 5,000 ppm should
have a removal efficiency of 95 percent.^ Alliance recommends that Kodak
personnel make the proper adjustment in the condenser's operation so that a
removal efficiency of 95 percent may be achieved. This increase in the removal
efficiency would result in a reduction of DCM from this emission source of
about 21,000 pounds per year.
The category 1 and 2 emission points which are remaining can be placed in
the second group. These emission points can be characterized as uncontrolled
fugitive emissions generated by various processes in the acetate film
manufacturing loop and other areas. A list of these emission points is
presented in Table 6-4 along with the concentration of DCM in each stream.
These emission points emit approximately 785,249 pounds of DCM into the
atmosphere per year. The emission points in this group are the ones to which
add-on control technologies may be applied. There are some problems when
applying a control device to many of these points. A majority of these points
are associated with floor sweeps and ventilation of Buildings 52, 53 and 54.
These emission points generally have large flowrates with low DCM
concentrations. A list of emission points which fall into this category is
presented in Table 6-5. One type of control device which can effectively
control such low concentrations is a thermal incinerator. There are, however,
problems associated with thermal incinerating of DCM. The first, as mentioned
earlier, is the formation of HC1. A caustic scrubber can easily remove the HCl
from the air flow but Kodak would then need to properly dispose of the
scrubber's effluent. Another problem has to do with the amount of supplemental
fuel in order to accomplish the incineration. Table 6-6 provides a list of
6-9

-------
TABLE 6-4. FUGITIVE EMISSION SOURCES AT KODAK PARK
Source
Description
Flowrate,
cfm
Actual
Emissions,
lbs/yr
Cone.,
ppm
52-37
BATCH MIXERS
1,110
193,680
1,531
53-K1
FLOOR SWEEPS FOR SOLVENT RECOVERY ROOM
22,000
110,000
44
53-88
FLOOR SWEEPS - DOPE DEPT.
34,400
86,724
22
53-08
FILTER PRESS CHANGING
15,000
75,680
44
54-29
FELT WASH PROCESS (5 FILTERS)
1
44,155
387,341
53-K2
EAST END FLOOR SWEEPS
9,000
36,500
36
21-12
KADY MILL EXHAUST
5,500
35,580
57
329K2
VESSEL CLEANING EXHAUST
4,100
27,450
59
D63-5
STEAMER CHARGING EXHAUST
3,550
26,280
65
53-96
STORAGE VESSEL VENTS
1
22,900
200,886
53-32
HOPPER CLEANING AND FLOOR SWEEPS
11,200
19,000
15
52-M2
WEST WALL EXHAUST FAN - 1ST FLOOR
6,000
17,800
26
53-92
SOLVENT DYE MIXING (FLOOR SWEEP SYSTEM)
12,000
15,000
11
52-M3
WEST WALL EXHAUST FAN - 2ND FLOOR
5,300
14,500
24
49-53
ULTRASONIC CLEANER
1,746
10,000
50
6-10

-------
TABLE 6-5. EMISSIONS FROM VENTILATION SOURCES AT KODAK PARK
Source
Description
Flowrate,
cfm
Actual
Emissions,
lbs/yr
Cone. ,
ppm
52-37
BATCH MIXERS
1,110
193,680
1,531
53-K1
FLOOR SWEEPS FOR SOLVENT RECOVERY ROOM
22,000
110,000
44
53-88
FLOOR SWEEPS - DOPE DEPT.
34,400
86,724
22
53-08
FILTER PRESS CHANGING
15,000
75,680
44
53-K2
EAST END FLOOR SWEEPS
9,000
36,500
36
53-32
HOPPER CLEANING AND FLOOR SWEEPS
11,200
19,000
15
52-M2
WEST WALL EXHAUST FAN - 1ST FLOOR
6,000
17,800
26
53-92
SOLVENT DYE MIXING (FLOOR SWEEP SYSTEM)
12,000
15,000
11
52-M3
WEST WALL EXHAUST FAN - 2ND FLOOR
5,300
14,500
24
6-11

-------
fuel costs for several emission points. The yearly expenditures for fuel shows
that thermal incineration would be an expensive means of controlling DCM
emissions. Recovery of DCM through a carbon adsorber from these emission
points also presents a problems. The area of the carbon bed needed to provide
an acceptable velocity through the bed would cause the carbon adsorber to be a
rather large. The amount of DCM recovered from each of the points in question
may not offset the cost of operating the recovery devices. A brief cost
analysis for both types of control devices applied to a single source can be
found in Appendix B. The cost estimates generated for this analysis are from a
computer program, Controlling Air Toxics (CAT), developed for the CTC.
Another approach for a control strategy would be to combine emission
streams to create a single source. A cost analysis of potential analysis of
potential control devices for combined sources can be found in Appendix C. A
control device of each type was proposed to reduce the emissions from Building
53 and 20. The emission sources under consideration for the Building 53
control device include the fugitive emission sources in Building 53 (see Table
6-4) and the emissions from the machine room exhaust. The only emission source
under consideration for the Building 20 control device is the machine room
exhaust.
Several of the points in this second group, however, present situations in
which Kodak could recover DCM. The first situation involves the combining of
emission points 52-37 (batch mixers) and 54-29 (felt wash process). As
described earlier, the emissions from the emission point 54-29 occur when the
filter elements from the felt wash process are being changed. Up to 5 gallons
of DCM evaporates directly to the atmosphere for each change. A shroud could
be fashioned to surround the immediate area around each felt wash canister
during the changing procedure. A duct would be connected to the top of the
shroud with sufficient flow to draw the DCM vapors toward it during the
element changing. The flow from this duct would then be combined with the flow
from the batch mixer ventilation system. The DCM then could be recovered by
either a carbon adsorber or a dual scrubber system. The carbon adsorber for
the recovery process could be the 4,000 cfm carbon adsorber currently being
used to pressure balance the roll casting machines. This carbon adsorber will
become available once the 18,000 cfm carbon adsorber replaces it. The flowrate
of 4,000 cfm should be sufficient for both emission points. The cost to Kodak
would be moving the carbon adsorber to a proper location, the duct work, and
6-12

-------
TABLE 6-6. NATURAL GAS COST FOR THERMAL INCINERATION CONTROL DEVICE
Fuel Costs
Source	Description	(June 1985 Dollars)
53-K1	FLOOR SWEEPS FOR SOLVENT RECOVERY	ROOM $ 1,565,802
53-88	FLOOR SWEEPS - DOPE DEPT.	$ 2,447,388
53-08	FILTER PRESS CHANGING	$ 1,067,991
53-K2	EAST END FLOOR SWEEPS	$ 640,356
6-13

-------
operations. A dual scrubber system would also be an effective mean of
recovering DCM from these combined emission points. Provided the scrubber is
operated correctly, the advantage of using the dual scrubber is the small
amount of space necessary to house it. If one assumes that the capture
efficiency of the vents around the batch mixers and felt washer is 100 percent
and that the carbon adsorber or scrubber operates at an efficiency of
90 percent, the reduction of DCM emissions from these two points would be
214,051 pounds per year.
Other points in this second group from which Kodak can easily control the
emissions include points 53-32 (hopper cleaning and floor sweeps), 53-96
(storage vessel vents) and 49-53 (ultrasonic cleaner). During the plant visit,
Kodak personnel indicated there were plans for connecting the ve.its surrounding
the hopper cleaning baths to the 18,000 cfm carbon adsorber when it is
installed. Such a move would greatly reduce the emissions from-this point
since the emissions would only consist of the floor sweeps being vented to the
atmosphere. The storage vessel vents has an effluent concentration which can
easily be handled by a recovery device. An easy means of reducing the
emissions would be to connect the vent to the carbon adsorber in Building 53.
Such a connection would only be a small fraction of the carbon adsorber's
flowrate and the cost to Kodak would only be for the duct to connect the vent
to the carbon adsorber. One could expect a reduction of 39,805 pounds of DCM
per year if these points are included in the flows for the 18,000 cfm carbon
adsorber. The ultrasonic cleaner (emission point 49-53) as described earlier,
operates analogous to a degreasing operation. Controls for this source are
described in Title 6, Chapter III, Part 226 of New York State's air pollution
control regulations. Included are general requirements, equipment
specifications, and operating requirements. Equipment specifications would
include a cover when not in use, a freeboard ratio greater than or equal to
0.75, and use of either a refrigerated freeboard chiller, or carbon adsorber.
One emission point which proved difficult to assess a control technology
for was the Kady Mill exhaust (point 21-21). The difficultly was not so much
in analyzing a potential control device to reduce the emissions from this
source, but in the data on the process provided by Kodak. During the plant
visit, Kodak personnel stated that the Kady Mill was in operation for only two
weeks out of the year. This means that the process emits 106 pounds of DCM per
hour for it to have yearly DCM emissions of 35,580 pounds. Since the tank in
6-14

-------
which the Kady Mill process takes place is only open during charging and
cleaning, Alliance feels that this emissions estimate is rather high. Kodak
needs to develop a better method of estimating the DCM emissions from this
point other than a best engineering estimate. If the revised emissions
estimate shows that the same amount of DCM is being lost to the atmosphere,
then Kodak should decide whether to continue the Kady Mill process, given the
large loss of DCM in a relatively short period of time.
Fugitive emissions from equipment and pipe lines in DCM service currently
discharge 650,000 pounds of DCM per year into the environment. This amount
represents approximately 8 percent of the total yearly DCM emissions at Kodak
Park. This estimate is based on best engineering judgement, since Kodak has
never counted the pumps, valves, flanges, seals, or open-ended lines in DCM
service. At the present time, Kodak does not have a program to detect and
repair leaks in this equipment on a regular basis. An option to reduce these
emissions is the institution of a leak detection and repair (LDAR)^ program
for the equipment in DCM service. The recommended detection procedure for
Q
fugitive DCM emissions is EPA Reference Method 21. This method incorporates
the use of a portable analyzer to detect the presence of volatile organic
vapors at the surface of the interface where direct leakage to the atmosphere
can occur. This technique assumes that if a DCM leak exists, there will be an
increased vapor concentration in the vicinity of the leak. By observing the
changes in the concentration levels, the location and extent of the leak can
then be determined. Once the severity of a leak has been defined, Kodak
personnel could then take the appropriate actions for remediation of the
emission source. At this time, Alliance is unable to estimate the amount of
time necessary to complete a thorough inspection of all the potential fugitive
emission sources. This is due to Kodak's inability to provide Alliance
personnel with detailed information regarding the equipment typej and number
which are in DCM service. The cost for the LDAR program would come mostly from
the man-hours required to perform the inspection for DCM leaks. The reduction
in fugitive emissions resulting from the institution of the LDAR program can
not be accurately determined at this time due to the same lack of detailed
information. Generally speaking, a LDAR program has the potential of reducing
q
fugitive emissions by 60 percent. Once an LDAR program is in place, Kodak
could expect fugitive DCM emission to be lowered by 390,000 pounds per year or
a A percent reduction in overall emissions.
&-15

-------
REFERENCES
1.	U.S. Environmental Protection Agency. Handbook: Control Technologies for
Hazardous Air Pollutants. EPA/625/6-86/014. Air and Energy Engineering
Research Laboratory. Research Triangle Park, NC. September 1986.
2.	Reference 1. P 26.
3.	Reference 1. P 24.
4.	U.S. Environmental Protection Agency. Health Assessment Document for
Dichloromethane (Methylene Chloride). EPA/600/8-82/004f. Office of
Health and Environmental Assessment. Washington, DC. February 1985.
P 3-3.
5.	Letter from Brian M. Wirsig, Eastman Kodak Company, to Stephen A. Walata,
Alliance Technologies Corporation, dated September 12, 1988, in response
to information requested by Alliance.
6.	Reference 1. P 24.
7.	U.S. Environmental Protection Agency. VOC Fugitive Emissions in Synthetic
Organic Chemicals Manufacturing Industry - Background Information for
Proposed Standards. EPA-450/3-80-033a (NTIS PB81-152167). Office of Air
Quality Planning and Standards. Research Triangle Park, NC. November
1980. P 4-1
8.	EPA Regulation on Standards of Performance for New Stationary Sources (40
CFR 60) Appendix A. July 1,1987
\
9.	Reference 7. P 7-7.
5—16

-------
APPENDIX A
Cellulose Triacetate Coated from Methylene Chloride
Solvent System
Docket # H-71 OSHA in response to 51FR42257
Submitted Feburary 19, 1987
by
R. Brothers - Director of Regulatory Affairs

-------
Photographic Film Support
Cellulose Triacetate Coated From
Methylene Chloride Solvent Systems
Eastman Kodak Company
February 1987
A-2

-------
—2
Cellulose triacetate is an essential ingredient of many of the
photographic systems in use today. The unique properties of
cellulose triacetate cannot be easily duplicated with other
polymeric materials. This need for cellulose triacetate can best be
appreciated by reviewing the requirements for photographic film
supports.
Film Support Requirements
The requirements for a photographic film support are very exacting.
Optically, it must be transparent, colorless, and free from haze and
visible imperfections. Chemically, it must be stable over long
periods of time, inert to highly sensitive emulsions, and allow for
proper adhesion of the emulsion layers. A low volatile content to
prevent dimensional change on processing and storage is also
necessary. Physical requirements include strength and toughness,
hardness without brittleness, stiffness with flexibility, tear
resistance, and freedom from curl. Thermally, a film support must
have a high softening temperature and be slow burning.*
In addition to these general requirements, there are special needs
in each product area. For example, successful operation of
photographic equipment for 35mm, 110, and 126 amateur roll films and
8mm and Super 8 amateur movie films requires a support material
which will retain the curvature of the core on which the film is
wound during manufacture. This retained curvature must then be
removed during photographic processing to yield film negatives for
printing which are nearly flat and to allow the proper curvature for
automatic threading to be induced in movie films. For the
professional motion picture market, camera original, laboratory
intermediate, and release films require a strong, tough support with
good wearing qualities. Many graphic arts applications require a
film which can be easily cut into sections for the preparation of
advertising layouts.
Film Supports
The above requirements are so rigid that relatively few materials
have proved to be practical as supports for photographic films. The
physical properties of various cellulose film support materials are
listed in Table 1.2»^ The first flexible film support, introduced
in 1889, was manufactured from cellulose nitrate. It had excellent
physical properties, but suffered from poor chemical stability and
was a severe fire hazard because of Its explosive nature and the
toxicity of its fumes.
A—3

-------
—3
In 1922, amateur film was introduced on cellulose diacetate support
because its slow burning feature was considered essential for this
product. Mixed esters of cellulose acetate butyrate or cellulose
acetate propionate possessed Improved physical properties and
gradually replaced cellulose diacetate starting around 1940.
However, none of these materials had the strength, toughness, and
heat resistance required for professional motion picture films.
These products remained on cellulose nitrate support.
Cellulose triacetate, although it will support combustion like most
organic polymers, is considerably less flammable. To further
enhance this property, flame retardant plasticizers have been an
integral component of cellulose triacetate film support since its
inception. The major obstacle to development and use of cellulose
triacetate film supports was its extremely limited solubility.
Ethylene dichloride and propylene dichloride vere initially used as
principal film support casting solvents but concerns over the
toxicity of these solvents prompted a search for a less toxic, more
volatile solvent. Methylene chloride uniquely meets these
requirements. Not only is it an essentially nonflammable solvent
with a low order of chemical reactivity, but its fast evaporation
rate provides a cost efficient method of producing film support at
reasonable speeds. It may also be efficiently recovered for reuse.
The introduction of cellulose triacetate film base in the 1940's
provided for the first time a slow-burning support for
motion-picture film use, permitting the manufacture of cellulose
nitrate to be discontinued. Cellulose triacetate eventually became
the support of choice for roll and amateur movie films and those
sheet films requiring ease of cutting.
Polyethylene terephthalate, which became available as a film support
about 1955, is very moisture resistant as compared with the
cellulosic supports, and also has higher strength, stiffness, and
tear resistance. (See Table 2).5»6,7,8 Polyethylene
terephthalate is used In sheet films, aerial films and industrial
films. It has not been used extensively, however, in the amateur
roll and movie film areas because of its low plastic flow and high
moisture resistance. Professional motion picture films are
currently available on polyethylene terephthalate, but these
products have not been widely used in the trade. Most motion
picture film is manufactured on cellulose triacetate so that solvent
splices may be made. Also, many graphic arts film users prefer
cellulose triacetate because of its ease of cutting.
A—4

-------
Since the advent of cellulose triacetate and polyethylene
terephthalate, no new film support has come into widespread use in
the photographic industry.
Cellulose Triacetate
Cellulose triacetate is ideally suited for the amateur roll and
movie film areas because of its plastic flow characteristics. Core
set represents an important manifestation of plastic flow in
photographic systems because films are normally available as rolls
wound on cores. When a film is held in such a curved configuration,
plastic flow occurs and the support retains a portion of the
curvature of the core on which it was wound. This is called "core
set". Figure l8 illustrates this core set, with the degree of
curvature measured in ANSI curl units (100 divided by the radius of
curvature in inches) so that higher numbers mean greater
curvature.7*9
Host of the amateur photographic equipment in use today, Including
cameras, projectors, magazines, and cartridges, has been designed
for the core set characteristics of cellulose triacetate. The
transport mechanisms and exposure assemblies for all of these
devices usually require a curvature similar to that of cellulose
triacetate films for best performance. Use of film support material
with core set characteristics significantly different from those of
cellulose triacetate will result in decreased reliability and
possible inoperability for many consumer cameras and projectors and
for photographic processing equipment.
Absorption of water in aqueous processing solutions by cellulose
triacetate enables its plastic flow or core set to relax, so that
films on this support are relatively flat after processing. Many
films, such as 35mm, 110, and 126 fortaats, require flat negatives to
facilitate the making of photographic prints and to avoid bulky
customer shipping envelopes. Removal of core set in processing
solutions is also important for many amateur movie film projectors
in which automatic threading depends upon the film core set being
removed in processing and core set being induced in the opposite
direction by storage on processed film reels. With this core set
configuration, the unsupported film naturally takes a path which
leads into the threading mechanism.

-------
--5
Film Support Manufacturing
To better understand the requirements of a film support casting
solvent, a brief overview of the solution making and casting
procedures is necessary. Cellulose triacetate is combined with
methylene chloride, plasticizers and minor amounts of co-solvents
and mixed with heat to produce a viscous concentrated polymer
solution or dope. The dope is then filtered to remove any
impurities, further concentrated and cast onto a polished metal
surface. When sufficient solvent has evaporated to allow the film
to be self-supporting, it is stripped off the metal surface and
conveyed into dryers. The evaporated solvents are condensed,
purified and recycled to the dope-making stage.
Solubility of Cellulose Triacetate
The solubility of a polymer is dependent upon its molecular weight,
the degree and nature of substitution in the polymer chain, and its
degree of crystallinity. The influence of these properties is quite
evident for cellulose and its derivatives. Pure cellulose is
reported to be soluble only In solutions such as Schweitzer's
reagent (copper hydroxide in ammonia) or zinc chloride, which are
capable of forming complexes with it.** The insolubility of
cellulose in common organic solvents is attributed to its
crystallinity and intermolecular hydrogen bonding.
The solubility of cellulose acetate esters is further dependent upon
the degree of acetylation (or the degree of hydrolysis).
Far-hydrolyzed cellulose acetate (18-26% acetic acid content) can be
made water-soluble. The original inherent crystallinity is largely
destroyed and the presence of many hydroxy1 groups allow solvation.
Increasing the acetyl content to a range corresponding to the
diacetate (44-52% acetic acid) decreases the water solubility, since
the number of hydroxyl groups decreases, and the polymer now becomes
soluble in ketone and ester type solvents. At this point the
cellulose acetate molecules exhibit characteristics of both Lewis
acids and bases and hence are soluble in both acidic and basic
solvents. An almost fully acetylated triacetate (61-63% acetic acid
content), Buch as that required for photographic film support, is
reported by one source to be soluble only in acid type solvents
capable of forming hydrogen bonds.I1
A—6

-------
—6
A method useful in predicting the solubility of polymers,
particularly cellulosics, was introduced in 1966 by James D. Crowley
and colleagues of Eastman Chemical Products, Inc.** His concepts
have gained wide commercial acceptance and have been utilized during
the last twenty years to characterize the solubility of cellulose
derivatives and predict potential alternate solvents. In essence,
his work identified three parameters which can be used to uniquely
characterize solvents. These are the solubility parameter, hydrogen
bonding and dipole moment. Using the solubility law of "like
dissolves like," these three parameters or ranges can also be
assigned to solutes.
Using a modification of the ASTM D3132 test procedure for
determining polymer solubility ranges, the profile of cellulose
triacetate has been empirically determined to be:
Solubility Parameter (J): 9.7-11.1
Hydrogen Bonding (V): 1.5-6.3
Dipole Moment (/rf): 1.4-1.6
Table 3 lists (by increasing £) the known, and many predicted
solvents for cellulose triacetate. It is apparent that, among the
solvents listed, there exist anomolies even using Crowley's
three-dimensional approach.
Many of the compounds rieferred to in the literature^^-»^3 as being
solvents for cellulose triacetate do not have practical or
commercial value due to their relatively poor solvent
characteristics or toxicity. The substances evaluated as
potentially practical solvents for cellulose triacetate have been
relisted in Table 4 (by increasing boiling point). Also Included
are the evaporation rates relative to methylene chloride and flash
points which offer a guide to their utility as the major constituent
in a casting formulation.
N-Methyl-2-pyrrolidone, with a boiling point exceeding 200°C, is
not a practical primary casting solvent. Its use would need to be
restricted to less than 10% of the solvent formulation to allow a
film to be stripped from the metal surface. It is also unlikely
that residual solvent could be reduced to a level whereby the film
would meet established physical requirements.
Dioxane and dioxolane, in addition to being flammable and explosive,
also have a tendency to form explosive peroxides and the volumes
required for the manufacture of film support would present extreme
safety risks.
A-7

-------
—7
The use of 2,3-butanedione, also flammable and explosive, as a
primary casting solvent would have a major Impact on manufacturing
efficiency. Having an evaporation rate which is less than half that
of methylene chloride would mean that the casting, stripping and
drying operations would be restricted to half the speed. In
addition, this solvent has an inherent yellow color which remains in
the film support even after drying two hours at 150°C, making it
impractical for producing a clear, transparent support.
Conclusions
Cellulose triacetate is essential to the photographic industry
because many current photographic systems have been designed to take
advantage of the unique plastic flow and moisture absorption
characteristics of this material. These characteristics are not
duplicated by other available polymeric materials, including
cellulose diacetate and mixed cellulose esters. Methylene chloride
is necessary for cellulose triacetate film support production
because no other safe practical coating solvent is available.
A-8

-------
—8
References
1.	C. B. Neblette, Photography. Its Materials and Processes. Sixth
Ed., Van Nostrand Reinhold Company, Hew York, 14621, pp. 161-62.
2.	J. M. Calhoun, The Physical Properties and Dimensional Stability
of Safety Aerographic Film. Photogrammetric Engineering, 13,
June 1947, pp. 162-221.
3.	C. B. Neblette, Photography. Its Materials and Processes. Fifth
Ed., D. Van Nostrand, Inc., New York 14521, p. 153.
4.	C. R. Fordyce, Improved Safety Motion Picture Film Support.
Journal of Society of Motion Picture Engineers, 51, October
1948, pp. 331-350.
5.	J. M. Sturge, Neblette's Handbook of Photography and
Reprography. Seventh Ed., Van Nostrand Reinhold Company, New
York, 1977, p.134.
6.	Modern Plastics Encyclopedia. 62, October 1985, pp. 481, 483.
7.	Properties of Kodak Materials for Aerial Photographic Systems.
Vol. 11. Physical Properties of Kodak Aerial Films. Eastman
Kodak Company, Rochester, New York, 1972.
8.	W. Thomas, SPSE Handbook of Photographic Science and
Engineering. John Wiley and Sons, New York, 1973, p. 481
9.	American National Standard Method for Determining the Curl of
Photographic Film. PHI.29.
10.	J. M. Calhoun, The Physical Properties and Dimensional Behavior
of Motion Picture Film. Journal of the SMPTE, 43, October 1944,
pp. 227-266.
11.	P. Howard and R. S. Parikh, Solution Properties of Cellulose
Triacetate. Journal of Polymer Science, 6, 1968, p. 537.
12.	J. D. Crowley et al, A Three-Dlmenslonal Approach to Solubility.
Journal of Paint Technology, 38, 1966, pp. 269, 496.
13.	C. J. Malm et al, Aliphatic Acid Esters of Cellulose.
Industrial and Engineering Chemistry, 43, Vol. 3, 1951, p. 688.
A—9

-------
APPENDIX B
CAT Analysis of Emission Points
Emission Point	Page
53-K1	Using Thermal Incineration		B-2
53-K1	Using Carbon Adsorption		B-5
53-K2	Using Thermal Incineration		B-9
53-K2	Using Carbon Adsorption		B-13
53-08	Using Thermal Incineration				B-17
53-08	Using Carbon Adsorption		B-21
53-88	Using Thermal Incineration		B-25
53-88	Using Carbon Adsorption		B-29
B-l

-------
CAT Analysis of Emission Point
53-K1 Using Thermal Incineration
B-2

-------
Plant:
Eastman Kodak Company
1669 Lake Avenue
Rochester, NY 14652-4201
Contact: Jeffery Mathews, P.E.
Phone: (716) 722-0692 Ext.
Agency contact:
Emission Stream:
Applicant Calculation Checked
Maximum flow rate (scfm)
22000
22000
Pressure (mmHg)

760
Temperature (degF)
72
72
Heat content (Btu/scf)

1
Oxygen content (%)

21
Moisture content (%)

5
Relative humidity (%)
50
50
halogenated organics present? (Y/N)
Y
Y
Are metals present? (Y/N)
N
N
Hazardous Air Pollutant: 75-09-2 METHYLENE CHLORIDE
Applicant Calculation Checked
Inlet HAP concentration (ppmv) 43.8821 43.8821
Molecular weight (lb/lb-mole)	84.93
Specific heat equation constant A
Specific heat equation constant B
Specific heat equation constant C
Antoine equation constant A
Antoine equation constant B
Antoine equation constant C
Heat of vaporization (Btu/lb-mole)
Applicant Calculation Checked
DESIGN RELATED PARAMETERS	
Destruction efficiency (%)	99
Combustion temperature (degF)	1800
Residence time (sec)	2
Is a heat exchanger used? (Y/N)	Y
Emission stream temp, after preheat (degF)	600
Excess air (%)	25
Area to Qe ratio	1
Heating value of supplement, fuel (Btu/scf)	882
Reference temperature (degF)	70
1 B-3

-------
COST RELATED PARAMETERS

Duct cost ($/linear ft)
15

Length of duct (ft)
150

Total pressure drop (in. H20)
6

Average equipment life (yr)
10
Operator
labor requirements (hr/shift)
.5
Maintenance
labor requirements (hr/shift)
.5
Review of Thermal Incinerator:
METHYLENE CHLORIDE:
The HAP inlet concentration is too low for a high destruction efficiency.
If the HAP concentration in the emission stream exceeds 25% of
the lower explosive limit, then dilution of the emission stream will
be required.
This device is not well suited to emission streams with highly variable
flow rates.
Intermediate Results:
Cpair(600) = 0.0185
Cpair(1800) = 0.0203
Qc = 0
deltaTLM = 1200
Permit Evaluation:
Supplementary heat requirement (Btu/min)	629710
Supplementary fuel flow rate (scfm)	714
Flue gas flow rate (scfm)	22714
Combustion chamber volume (f13)	3390
Heat exchanger surface area (ft2)	2711
Total Capital Investment (June 1985 Dollars):
348475
Direct Operating Costs
Natural Gas	1565802
Electricity	58974
Operator Labor	6197
Operator Supervision	930
Maintenance Labor	6197
Maintenance Materials	6197
B-4

-------
Indirect Operating Costs
Overhead	10659
Property Tax	3485
Insurance	3485
Administrative	6970
Capital Recovery	56801
Net Annualized Cost (June 1985 Dollars):
1725697

-------
CAT Analysis of Emission Point
53-K1 Using Carbon Adsorption
B-6

-------
Plant:
Eastman Kodak Company
1669 Lake Avenue
Rochester, NY 14652-4201
Contact: Jeffery Mathews, P.E.
Phone: (716) 722-0692 Ext.
Agency contact:
Emission Stream:
Applicant Calculation Checked
Maximum flow rate (scfm)	22000	22000
Pressure (mmHg)
Temperature (degF)	72	72
Heat content (Btu/scf)
Oxygen content (%)
Moisture content (%)
Relative humidity (%)	50	50
Are halogenated organics present? (Y/N)	Y	Y
Are metals present? (Y/N)	N	N
Hazardous Air Pollutant: 75-09-2 METHYLENE CHLORIDE
Inlet HAP concentration (ppmv)
Molecular weight (lb/lb-mole)
Specific heat equation constant A
Specific heat equation constant B
Specific heat equation constant C
Antoine equation constant A
Antoine equation constant B
Antoine equation constant C
Heat of vaporization (Btu/lb-mole)
Applicant Calculation Checked
43.8821 43.8821
84.93
Applicant Calculation Checked
DESIGN RELATED PARAMETERS	
Removal efficiency (%)

90
Adsorptive capacity (lb HAP/100 lb carbon)

3
Number of beds

3
Cycle time for adsorption (hr)

4
Cycle time for regeneration (hr)

3.5
Stream velocity through the bed (ft/min)
65
65
Steam ratio (lb steam/lb carbon)
0.4
0.4
Steam inlet temperature (degF)

212
Condensed steam outlet temperature (degF)

100
Cooling water inlet temperature (degF)

50
Cooling water outlet temperature (degF)

100
Carbon bed density (lb/ft3)

30
Cycle time for drying and cooling (hr)

.5
Latent heat of vaporization (Btu/lb)

970
Avg. specific heat of water (Btu/lb-degF)

1
Overall heat trans coef (Btu/hr-ft2-degF)

150
: B—7

-------
COST RELATED PARAMETERS	
Value of recovered product ($/lb)	.2 .2
Duct cost ($/linear ft)	15 15
Stack capital cost ($)	20000 20000
Length of duct (ft)	150 150
Total pressure drop (in. H20)	6
Average equipment life (yr)	10
Operator labor requirements (hr/shift)	.5
Maintenance labor requirements (hr/shift)	.5
Intermediate Results for METHYLENE CHLORIDE:
HAPo	=4.39
Abed	= 340
Vcarbon	= 102
Hload	= 1.45E+006
deltaTLM	= 76.9
Qfg	= 2.2E+004
Permit Evaluation:
Carbon requirement (lb)	9150
Bed diameter (ft)	21
Bed depth (ft)	0
Steam flow rate (lb/min)	20
Condenser surface area (ft2)	126
Cooling water rate (gal/min)	58
Recovered product (lb/hr)	11
Total Capital Investment (June 1985 Dollars):
92151
Direct Operating Costs
Electricity	13446
Steam	52013
Water	8978
Operator Labor	6197
Operator Supervision	930
Maintenance Labor	6197
Maintenance Materials	6197
Replacement Labor	3514
Replacement Parts	3514
Indirect Operating Costs
Overhead	13470
Property Tax	922
Insurance	922
Administrative	1843
Capital Recovery	15021
B-8

-------
Credits
Sale of Product	2
Net Annualized Cost (June 1985 Dollars):
133161
B-9

-------
CAT Analysis of Emission Point
53-K2 Using Thermal Incineration
B-10

-------
Plant:
Eastman Kodak Company
1669 Lake Avenue
Rochester, NY 14652-4201
Contact: Jeffery Mathews, P.E.
Phone: (716) 722-0692 Ext.
Agency contact:
Emission Stream:
Calculation Checked
9000
760
72
1
21
5
50
Y
N
Applicant
Maximum flow rate (scfm)	9000
Pressure (mmHg)
Temperature (degF)
Heat content (Btu/scf)
Oxygen content (%)
Moisture content (%)
Relative humidity (%)
Are halogenated organics present? (Y/N)	Y
Are metals present? (Y/N)	N
Hazardous Air Pollutant: 75-09-2 METHYLENE CHLORIDE
Inlet HAP concentration (ppmv)
Molecular weight (lb/lb-mole)
Specific heat equation constant A
Specific heat equation constant B
Specific heat equation constant C
Antoine equation constant A
Antoine equation constant B
Antoine equation constant C
Heat of vaporization (Btu/lb-mole)
Applicant Calculation Checked
35.5932
84.93
Applicant Calculation Checked
DESIGN RELATED PARAMETERS	
Destruction efficiency (%)	99
Combustion temperature (degF)	1800
Residence time (sec)	2
Is a heat exchanger used? (Y/N)	Y
Emission stream temp, after preheat (degF)	600
Excess air (2)	25
Area to Qe ratio
Heating value of supplement, fuel (Btu/scf)	882
Reference temperature (degF)	70
B-ll

-------
COST RELATED PARAMETERS
Duct cost ($/linear ft)
Length of duct (ft)
Total pressure drop (in. H20)
Average equipment life (yr)
Operator labor requirements (hr/shift)
Maintenance labor requirements (hr/shift)
15
150
6
10
.5
.5
Review of Thermal Incinerator:
METHYLENE CHLORIDE:
The HAP inlet concentration is too low for a high destruction efficiency.
If the HAP concentration in the emission stream exceeds 25% of
the'lower explosive limit, then dilution of the emission stream will
be required.
This device is not well suited to emission streams with highly variable
flow rates.
Intermediate Results:
Cpair(600) = 0.0185
Cpair(1800) = 0.0203
Qc = 0
deltaTLM = 1200
Permit Evaluation:
Supplementary heat requirement (Btu/min)
Supplementary fuel flow rate (scfm)
Flue gas flow rate (scfm)
Combustion chamber volume (ft3)
Heat exchanger surface area (f12)
257608
292
9292
1387
1109
Total Capital Investment (June 1985 Dollars):
246853
Direct Operating Costs
Natural Gas
Electricity
Operator Labor
Operator Supervision
Maintenance Labor
Maintenance Materials
640356
24125
6197
930
6197
6197
3-12

-------
Indirect Operating Costs
Overhead	10659
Property Tax	2469
Insurance	2469
Administrative	4937
Capital Recovery	40237
Net Annualized Cost (June 1985 Dollars):
744774
B-13

-------
CAT Analysis of Emission Point
53-K2 Using Carbon Adsorption
B-14

-------
Plant:
Eastman Kodak Company
1669 Lake Avenue
Rochester, NY 14652-4201
Contact: Jeffery Mathews, P.E.
Phone: (716) 722-0692 Ext.
Agency contact:
Emission Stream:
Applicant Calculation Checked
Maximum flow rate (scfm)	9000	9000
Pressure (mmHg)
Temperature (degF)	72
Heat content (Btu/scf)
Oxygen content (%)
Moisture content (%)
Relative humidity (%)	50
Are halogenated organics present? (Y/N)	Y	Y
Are metals present? (Y/N)	N	N
Hazardous Air Pollutant: 75-09-2 METHYLENE CHLORIDE
Inlet HAP concentration (ppmv)
Molecular weight (lb/lb-mole)
Specific heat equation constant A
Specific heat equation constant B
Specific heat equation constant C
Antoine equation constant A
Antoine equation constant B
Antoine equation constant C
Heat of vaporization (Btu/lb-mole)
Applicant Calculation Checked
35.5932
84.93
Applicant Calculation Checked
DESIGN RELATED PARAMETERS		
Removal efficiency (%)	90
Adsorptive capacity (lb HAP/100 lb carbon)	3 3
Number of beds	2
Cycle time for adsorption (hr)	4
Cycle time for regeneration (hr)	3.5
Stream velocity through the bed (ft/min)	65 65
Steam ratio (lb steam/lb carbon)	.4 .4
Steam inlet temperature (degF)	212
Condensed steam outlet temperature (degF)	100
Cooling water inlet temperature (degF)	50
Cooling water outlet temperature (degF)	100
Carbon bed density (lb/ft3)	30
Cycle time for drying and cooling (hr)	.5
Latent heat of vaporization (Btu/lb)	970
Avg. specific heat of water (Btu/lb-degF)	1
Overall heat trans coef (Btu/hr-ft2-degF)	150
1 B-15
V_	

-------
COST RELATED PARAMETERS		
Value of recovered product ($/1b)
Duct cost ($/linear ft)
Stack, capital cost ($)
Length of duct (ft)
Total pressure drop (in. H20)
Average equipment life (yr)
Operator labor requirements (hr/shift)
Maintenance labor requirements (hr/shift)
.2
15
20000
150
6
10
.5
.5
Intermediate Results for METHYLENE CHLORIDE:
HAPo
Abed
Vcarbon
Hload
deltaTLM
Qfg
3.56
139
33.7
3.21E+005
76.9
9000
Permit Evaluation:
Carbon requirement (lb)
Bed diameter (ft)
Bed depth (ft)
Steam flow rate (lb/min)
Condenser surface area (f12)
Cooling water rate (gal/min)
Recovered product (lb/hr)
2024
13
0
4
28
13
4
Total Capital Investment (June 1985 Dollars):
44622
Direct Operating Costs
Electricity	5501
Steam	10403
Water	2012
Operator Labor	6197
Operator Supervision	930
Maintenance Labor	6197
Maintenance Materials	6197
Replacement Labor	777
Replacement Parts	777
Indirect Operating Costs
Overhead	11281
Property Tax	446
Insurance	446
Administrative	892
Capital Recovery	7273
B-l 6

-------
Credits
Sale of Product	1
Net Annualized Cost (June 1985 Dollars):
59330
B-17

-------
CAT Analysis of Emission Point
53-08 Using Thermal Incineration
B-13

-------
Plant:
Eastman Kodak Company
1669 Lake Avenue
Rochester, NY 14652-4201
Contact: Jeffery Mathews, P.E.
Phone: (716) 722-0692 Ext.
Agency contact:
Emission Stream:
Applicant Calculation Checked
Maximum flow rate (scfm) 15000	15000
Pressure (mmHg)	760
Temperature (degF)	72
Heat content (Btu/scf)	1
Oxygen content (%)	21
Moisture content (%)	5
Relative humidity (%)	50
Are halogenated organics present? (Y/N)	Y Y
Are metals present? (Y/N)	N N
Hazardous Air Pollutant: 75-09-2 METHYLENE CHLORIDE
Inlet HAP concentration (ppmv)
Molecular weight (lb/lb-mole)
Specific heat equation constant A
Specific heat equation constant B
Specific heat equation constant C
Antoine equation constant A
Antoine equation constant B
Antoine equation constant C
Heat of vaporization (Btu/lb-mole)
Applicant Calculation Checked
44.2799
84.93
Applicant Calculation Checked
DESIGN RELATED PARAMETERS	
Destruction efficiency (%)	99
Combustion temperature (degF)	1800
Residence time (sec)	2
Is a heat exchanger used? (Y/N)	Y
Emission stream temp, after preheat (degF)	600
Excess air (%)	25
Area to Qe ratio
Heating value of supplement, fuel (Btu/scf)	882
Reference temperature (degF)	70
B-19

-------
COST RELATED PARAMETERS	
Duct cost ($/linear ft)	15
Length of duct (ft)	150
Total pressure drop (in. H20)	6
Average equipment life (yr)	10
Operator labor requirements (hr/shift)	.5
Maintenance labor requirements (hr/shift)	.5
Review of Thermal Incinerator:
METHYLENE CHLORIDE:
The HAP inlet concentration is too low for a high destruction efficiency.
If the HAP concentration in the emission stream exceeds 25% of
the lower explosive limit, then dilution of the emission stream will
be required.
This device is not well suited to emission streams with highly variable
flow rates.
Intermediate Results:
Cpair(600) = 0.0185
Cpair(1800) = 0.0203
Qc = 0
deltaTLM = 1200
Permit Evaluation:
Supplementary heat requirement (Btu/min)	429347
Supplementary fuel flow rate (scfm)	487
Flue gas flow rate (scfm)	15487
Combustion chamber volume (ft3)	2311
Heat exchanger surface area (ft2)	1849
Total Capital Investment (June 1985 Dollars):
295667
Direct Operating Costs
Natural Gas	1067991
Electricity	40210
Operator Labor	6197
Operator Supervision	930
Maintenance Labor	6197
Maintenance Materials	6197
B-20
V 	

-------
Indirect Operating Costs
Overhead	10659
Property Tax	2957
Insurance	2957
Administrative	5913
Capital Recovery	48194
Net Annualized Cost (June	1985 Dollars):
1198402
B-21

-------
CAT Analysis of Emission Point
53-08 Using Carbon Adsorption
s 	
. B—22

-------
Plant:
Eastman Kodak Company
1669 Lake Avenue
Rochester, NY 14652-4201
Contact: Jeffery Mathews, P.E.
Phone: (716) 722-0692 Ext.
Agency contact:
Emission Stream:
Applicant Calculation Checked
15000	15000
72
50
Y	Y
N	N
Maximum flow rate (scfm)
Pressure (mmHg)
Temperature (degF)
Heat content (Btu/scf)
Oxygen content (%)
Moisture content (%)
Relative humidity (%)
Are halogenated organics present? (Y/N)
Are metals present? (Y/N)
Hazardous Air Pollutant: 75-09-2 METHYLENE CHLORIDE
Applicant Calculation Checked
Inlet HAP concentration (ppmv)	44.2799
Molecular weight (lb/lb-mole)	84.93
Specific heat equation constant A
Specific heat equation constant B
Specific heat equation constant C
Antoine equation constant A
Antoine equation constant B
Antoine equation constant C
Heat of vaporization (Btu/lb-mole)
Applicant Calculation Checked
DESIGN RELATED PARAMETERS	
Removal efficiency (%)	90
Adsorptive capacity (lb HAP/100 lb carbon)	3 3
Number of beds	2
Cycle time for adsorption (hr)	4
Cycle time for regeneration (hr)	3.5
Stream velocity through the bed (ft/min)	65 65
Steam ratio (lb steam/lb carbon)	.4 .4
Steam inlet temperature (degF)	212
Condensed steam outlet temperature (degF)	100
Cooling water inlet temperature (degF)	50
Cooling water outlet temperature (degF)	100
Carbon bed density (lb/ft3)	30
Cycle time for drying and cooling (hr)	.5
Latent heat of vaporization (Btu/lb)	970
Avg. specific heat of water (Btu/lb-degF)	1
Overall heat trans coef (Btu/hr-ft2-degF)	150
B-23

-------
COST RELATED PARAMETERS	
Value of recovered product ($/lb)
Duct cost ($/linear ft)
Stack capital cost ($)
Length of duct (ft)
Total pressure drop (in. H20)
Average equipment life (yr)
Operator labor requirements (hr/shift)
Maintenance labor requirements (hr/shift)
.2
15
20000
150
6
10
.5
.5
Intermediate Results for METHYLENE CHLORIDE:
HAPo
Abed
Vcarbon
Hload
deltaTLM
Qfg
4.43
232
69.9
6.66E+005
76.9
1. 5E+004
Permit Evaluation:
Carbon requirement (lb)
Bed diameter (ft)
Bed depth (ft)
Steam flow rate (lb/min)
Condenser surface area (ft2)
Cooling water rate (gal/min)
Recovered product (lb/hr)
4197
17
0
9
58
27
8
Total Capital Investment (June 1985 Dollars):
59724
Direct Operating Costs
Electricity	9168
Steam	23406
Water	4180
Operator Labor	6197
Operator Supervision	930
Maintenance Labor	6197
Maintenance Materials	6197
Replacement Labor	1612
Replacement Parts	1612
Indirect Operating Costs
Overhead	11949
Property Tax	597
Insurance	597
Administrative	1194
Capital Recovery	9735
B-24

-------
Credits
Sale of Product	2
Net Annualized Cost (June 1985 Dollars):
83569
B-25

-------
CAT Analysis of Emission Point
53-88 Using Thermal Incineration
B-26

-------
Plant:
Eastman Kodak Company
1669 Lake Avenue
Rochester, NY 14652-4201
Contact: Jeffery Mathews, P.E.
Phone: (716) 722-0692 Ext.
Agency contact:
Emission Stream:
Applicant Calculation Checked.
Maximum flow rate (scfm) 34400	34400
Pressure (mmHg)	760
Temperature (degF)	72
Heat content (Btu/scf)	1
Oxygen content (%)	21
Moisture content (%)	5
Relative humidity (%)	50
Are halogenated organics present? (Y/N)	Y Y
Are metals present? (Y/N)	N N
Hazardous Air Pollutant: 75-09-2 METHYLENE CHLORIDE
Inlet HAP concentration (ppmv)
Molecular weight (lb/lb-mole)
Specific heat equation constant A
Specific heat equation constant B
Specific heat equation constant C
Antoine equation constant A
Antoine equation constant B
Antoine equation constant C
Heat of vaporization (Btu/lb-mole)
Applicant Calculation Checked
22.1257
84.93
Applicant Calculation Checked
DESIGN RELATED PARAMETERS	
Destruction efficiency (%)	99
Combustion temperature (degF)	1800
Residence time (sec)	2
Is a heat exchanger used? (Y/N)	Y
Emission stream temp, after preheat (degF)	600
Excess air (%)	25
Area to Qe ratio	1
Heating value of supplement, fuel (Btu/scf)	882
Reference temperature (degF)	70
B—27

-------
COST RELATED PARAMETERS

Duct cost ($/linear ft)
15

Length of duct (ft)
150

Total pressure drop (in. H20)
6

Average equipment life (yr)
10
Operator
labor requirements (hr/shift)
.5
Maintenance
labor requirements (hr/shift)
.5
Review of Thermal Incinerator:
METHYLENE CHLORIDE:
The HAP inlet concentration is too low for a high destruction efficiency.
If the HAP concentration in the emission stream exceeds 25% of
the lower explosive limit, then dilution of the emission stream will
be required.
This device is not well suited to emission streams with highly variable
flow rates.
Intermediate Results:
Cpair(600) = 0.0185
Cpair(1800) = 0.0203
Qc = 0
deltaTLM = 1200
Permit Evaluation:
Supplementary heat requirement (Btu/min)	984637
Supplementary fuel flow rate (scfm)	1116
Flue gas flow rate (scfm)	35516
Combustion chamber volume (ft3)	5301
Heat exchanger surface area (f12)	4240
Total Capital Investment (June 1985 Dollars):
435725
Direct Operating Costs
Natural Gas	2447388
Electricity	92212
Operator Labor	6197
Operator Supervision	930
Maintenance Labor	6197
Maintenance Materials	6197
B-28

-------
Indirect Operating Costs
Overhead	10659
Property Tax	4357
Insurance	4357
Administrative	8715
Capital Recovery	71023
Net Annualized Cost (June 1985 Dollars):
2658234
B-29

-------
CAT Analysis of Emission Point
53-88 Using Carbon Adsorption
B-30

-------
Plant:
Eastman Kodak Company
1669 Lake Avenue
Rochester, NY 14652-4201
Contact: Jeffery Mathews, P.E.
Phone: (716) 722-0692 Ext.
Agency contact:
Emission Stream:
Applicant Calculation Checked
Maximum flow rate (scfm)	34400	34400
Pressure (mmHg)
Temperature (degF)	72
Heat content (Btu/scf)
Oxygen content (%)
Moisture content (%)
Relative humidity (%)	50
Are halogenated organics present? (Y/N)	Y	Y
Are metals present? (Y/N)	N	N
Hazardous Air Pollutant: 75-09-2 METHYLENE CHLORIDE
Inlet HAP concentration (ppmv)
Molecular weight (lb/lb-mole)
Specific heat equation constant A
Specific heat equation constant B
Specific heat equation constant C
Antoine equation constant A
Antoine equation constant B
Antoine equation constant C
Heat of vaporization (Btu/lb-mole)
Applicant Calculation Checked
22.1257
84.93
Applicant Calculation Checked
DESIGN RELATED PARAMETERS	
Removal efficiency (%)	90
Adsorptive capacity (lb HAP/100 lb carbon)	3 3
Number of beds	3
Cycle time for adsorption (hr)	4
Cycle time for regeneration (hr)	3.5
Stream velocity through the bed (ft/min)	65 40
Steam ratio (lb steam/lb carbon)	.4 .4
Steam inlet temperature (degF)	212
Condensed steam outlet temperature (degF)	100
Cooling water inlet temperature (degF)	50
Cooling water outlet temperature (degF)	100
Carbon bed density (lb/ft3)	30
Cycle time for drying and cooling (hr)	.5
Latent heat of vaporization (Btu/lb)	970
Avg. specific heat of water (Btu/lb-degF)	1
Overall heat trans coef (Btu/hr-ft2-degF)	150
B-31

-------
COST RELATED PARAMETERS	
Value of recovered product ($/1b)
Duct cost ($/linear ft)
Stack capital cost ($)
Length of duct (ft)
Total pressure drop (in. H20)
Average equipment life (yr)
Operator labor requirements (hr/shift)
Maintenance labor requirements (hr/shift)
.2
15
20000
150
6
10
.5
.5
Intermediate Results for METHYLENE CHLORIDE:
HAPo
Abed
Vcarbon
Hload
deltaTLM
Qfg
2.21
863
80.2
1.14E+006
76.9
3.44E+004
Permit Evaluation:
Carbon requirement (lb)
Bed diameter (ft)
Bed depth (ft)
Steam flow rate (lb/min)
Condenser surface area (ft2)
Cooling water rate (gal/min)
Recovered product (lb/hr)
7214
33
0
16
99
46
9
Total Capital Investment (June 1985 Dollars):
103251
Direct Operating Costs
Electricity	21025
Steam	41610
Water	7121
Operator Labor	6197
Operator Supervision	930
Maintenance Labor	6197
Maintenance Materials	6197
Replacement Labor	2770
Replacement Parts	2770
Indirect Operating Costs
Overhead	12876
Property Tax	1033
Insurance	1033
Administrative	2065
Capital Recovery	16830
B-32

-------
Credi ts
Sale of Product	2
Net Annualized Cost (June 1985 Dollars):
128651
B-33

-------
APPENDIX C
Development of Costs for Using Control Devices for
Vents from Buildings 53 and 20
Supplemental Task to Work Assignment No. 13
Contract No. 68-02-4396
C-l

-------
CONTENTS
Page
INTRODUCTION		C-3
CONTROL DEVICES				C-4
Carbon Adsorber		C-4
Thermal Incinerators		C-7
CONCLUSION		C-10
REFERENCES		C-10
C-2

-------
INTRODUCTION
As part of this work assignment, Alliance was asked to develop a more
detailed cost analysis for a control device to reduce the methylene chloride
(dichloromethane or DCM) emissions from Building 53 and 20. This will be an
addendum to the original report issued under this work assignment.
The two control devices under consideration in this analysis:
Carbon Adsorber System
Thermal Incineration System
The emission points under consideration in Building 53 (1987 emission
data):


Flowrate
Emissions

Point ID/Description
(cfm)
(lbs/yr)
53-85
Existing Machine Room Exhaust
250,000
4,700,000
53-38
Existing Machine Room Exhaust
125,000
1,300,000
53-K1
Floor Sweeps for Solvent Recovery Room
20,000
150,000
53-88
Floor Sweeps - Dope Department
34,400
86,724
53-08
Filter Press Changing
15,000
75,680
53-K2
East End Floor Sweeps
9,000
36,500
53-96
Storage Vessel Vents
1
22,900
53-32
Hopper Cleaning and Floor Sweeps
11,200
19,000
53-92
Solvent Dye Mixing (Floor Sweeps System)
12,000
15,000

Total
476,600
6,405,805
These emission points are vented through the roof of Building 53 and thus can
be combined for common control device. This analysis will treat these 9
emission sources as a single stream with a flowrate of 500,000 cfm and
6,400,000 pounds of yearly DCM emissions. It should be noted that one category
2 emission source in Building 53 was omitted from the list of points under
consideration. This point, 53-22, is the exhaust from the carbon adsorber for
the machine air draw-off. Placing a secondary control device for the same
compound on a emission source from a control device is rarely cost effective
and does not represent sound engineering practices.
The emission point under consideration in Building 20 (1987	emissions data)
Flowrate Emissions
Point ID/Description	(cfm)	(lbs/yr)
20-68 Existing Machine Room Exhaust	150,000 1,380,000
Kodak is currently undertaking an emission reduction program in its
acetate film base manufacturing loop. The main program of this effort will be
the machine integrity program. The focus of this program will be to modify the
machine casings to improve the "seal quality." Kodak has a target reduction of
DCM emissions due to this program of 3,000,000 pounds per year. Kodak
personnel anticipate the completion of this program by 1990. For the purpose
of this task, we will assume that Kodak completes the emission reduction
program on time and achieves the targeted DCM emission reduction. The next
problem would be how to assign the anticipated emissions reduction to each film
base casting room. A simple solution would be to assume that the amount of
C—3

-------
emissions reduction in each machine room is related to the percent each
machine room contributes to the total emissions caused by the casting machines.
From the 1987 data, the total emissions caused by the film base casting
machines were 7,380,000 pounds per year. The machine room in Building 53
accounted for 6,000,000 pounds or 81.3 percent while the machine room in
Building 20 emitted 1,380,000 pounds or 18.7 percent. Using these percentages
to assign the amount of DCM emissions reduced, the emissions from Building 53
machine room can be expected to be lowered by 2,439,000 pounds and the
emissions from Building 20 machine room would be decreased by 561,000 pounds.
Thus, the yearly emissions expected from Building 53 are 4,000,000 pounds and
from Building 20 machine room 819,000 pounds of emissions would be expected
each year.
One of the parameters which characterizes the emission stream is the
flowrate. The flowrate will govern how big a control device is required. The
greater the flowrate, the larger the control device will be needed to handle
the flow. The flowrate also determine the concentration of the compound in the
stream. According to the 1987 data from Kodak, the total flowrate from the
category 1 and 2 emission sources in Building 53 was approximately 500,000 cfm.
The flowrate from the film base casting machine room in Building 20 was
150,000 cfm.
The concentration of DCM in the emission stream was determined in the same
manner as in Section 6 (see Table 6-3). Using the yearly emissions estimate
and flowrate discussed previously, the DCM concentration in the combined
emission stream from Building 53 was determined to be approximately 70 ppm.
The concentration of DCM in machine room vent flows from Building 20 was
calculated to be 48 ppm.
Other characteristics of the emission streams which will be assumed for
this analysis:
o the temperature is a consistent 100 °F
o the relative humidity is less than 50 percent
o the moisture content is 5 percent
o the oxygen content is 21 percent
o other chemical compounds present will not have any discernable effect
of the control device's ability to reduce the stream's DCM emissions.
CONTROL DEVICES
Carbon Adsorber
Alliance personnel contacted several manufactures of carbon adsorbers
seeking information concerning the purchase cost of a system for each stream.
One manufacturer stated that for the flowrates in question, anticipate a
carbon adsorber system costing $14 per cfm.^ This would mean that the purchase
cost of a carbon adsorber for Building 53 would $7,000,000 and one for
Building 20 would cost $2,100,000. Alliance personnel contact with another
carbon adsorber manufacturer confirmed these purchase costs.^ The carbon
adsorbers would be constructed using 904 stainless steel for the tanks and
associated equipment. The recovery efficiency of the carbon adsorber system
would be 90 to 95 percent for single pass air. One manufacturer stated that
this efficiency could be improved to the high 90*s if at least half the air

-------
could be recycled.^ The determination of annual costs for each carbon adsorber
for each stream is as follows using factors and utilities costs from the EPA
Handbook: Control Technologies for Hazardous Air Pollutants:^
Building 53 Carbon Adsorber
Purchase Cost for a 500,000 cfm flowrate at $14/cfm
500,000 x 14 = $7,000,000.
Total Capital expenditure (factoring in other direct costs and indirect
costs)
$7,000,000 x 1.63 = $11,410,000.
Assume the recovery efficiency of the carbon adsorber system is 90
percent. Thus, from emissions of 4,000,000 lbs/yr, 3,600,000 lbs/yr will
be recovered.
Utilities Required
Steam required (Assume 4 lbs of steam required for each lb of product
recovered.)
4 x 3,600,000 = 14,400,000 lbs of steam/year.
Cooling water required (Assume 12 gallons of cooling water required per
100 lbs of steam.)
12 x (14,400,000/100) = 1,728,000 gals of water/year.
Fan electricity required (Assume a fan efficiency of 65 percent and a
pressure drop of 7 in. H20 across the control system.)
2.0x10 4 x Flowrate x Press. Drop x Hrs =
Electricity required per year (kWh)
2.0xl0"4 x 500,000 x 7 x 8760 = 6,132,000 kWh/year.
Utilities Cost (June 1985 dollars)
Steam cost (Based on $0.00504 per lb.)
14,400,000 x 0.00504 = $72,576/year.
Water cost (Based on $0.0003 per gallon.)
1,728,000 x 0.0003 = $518/yr.
Electricity cost (Based on $0,059 per kWh.)
6,132,000 x 0.059 = $361,788/yr.
Total Utility cost = $434,882/year or
$500,000/year (Sept. 1988 dollars)
C-5

-------
Operating Labor Costs (June 1985 dollars)
Operator Labor (Assume 0.5 hrs/shift or 547.5 hrs/yr and a rate of $11.53
per hr.)
547.5 x 11.53 = $6,312/year.
$6,510/year (Sept. 1988 dollars)
Supervision (Assume 15 percent of Operator Labor.)
0.15 x 6312 = $947/year
$977/year (Sept. 1988 dollars)
Total Labor Cost = $7,487/year
Maintenance Cost
Labor (Assume 0.5 hrs/shift of 547.5 hrs/year and a rate of $11.53 per
hour.)
547.5 x 11.53 = $6,312/year.
$6,510/year (Sept. 1988 dollars)
Material (Assume 100 percent of Maintenance Labor.)
$6,510/year (Sept. 1988 dollars)
Total Mantenance Cost = $13,020/year
Indirect Operating Costs
Overhead	(80 percent of Operator, Supervisor and Maintenance Labor)
$11,200/year.
Property tax (1 percent of Total Capital Cost)
$114,100/year.
Insurance (1 percent of Total Capital Cost)
$114,100/year.
Administration (2 percent of Total Capital Cost)
$228,200/year.
Capital Recovery (16.3 percent of Total Capital Cost assuming equipment
life of 10 years.)
$1,859,830/yr
Credit	(Based on $0.2 per pound of methylene chloride)
$720,000/year.
Total Direct Operating Costs $ 520,507/year
Total Indirect Operating Costs $2,327,430/year
Credit	$ 720,000/year
Annual Cost	$2,128,000/year or
$ 1,200/ton DCM recovered.
C-6

-------
It should be noted that the annual cost does not reflect the cost incurred
for replacement carbon.
A similar break down of cost can be prepared for a carbon	adsorber for the
emission stream from Building 20. Assuming a similar recovery	efficiency,
similar factors and costs, a carbon adsorber system would have	the following
costs:
Total Direct Operating Costs $	147,869/year
Total Indirect Operating Costs $	706,069/year
Credit $	147,420/year
Annual Cost $	706,518/year or
$	2,000/ton DCM recovered.
The cost for the Building 20 carbon adsorber also does not reflect the cost
incurred for replacement carbon.
Thermal Incinerators
Alliance personnel contacted several manufacturers of thermal incinerator
systems regarding the cost of a system for the emission streams in question.
Alliance, however, was unable to contact an incinerator vendor who could
provide the cost of equipment for this type of application. The best estimate
that any vendor could make was that an incinerator for the Building 53 emission
stream (500,000 cfm) would cost $7,000,000 and one for Building 20
(150,000 cfm) would cost $2,300,000. The metal parts of each system would be
fabricated with stainless steel. This cost does not include the cost of a
caustic scrubber to deal with the HC1 formation which may occur during the
incineration process. During the conversations with the vendors, Alliance was
told that it was unusual to be incinerating emissions streams containing DCM .
and with such high flowrates. One of the major problems with this
application is due to the low heat content of the emission stream, a
supplemental fuel would be required to maintain the proper temperature for the
destruction of DCM.
A break down of costs similar to those for the carbon adsorber can be
prepared for each emission stream. For this application, Alliance assumed that
the heat content of both emission streams is 1.0 BTU/ft"*. The cost of an
incinerator for the Building 53 emission stream is as follows:
Purchase Cost for a 500,000 cfm flowrate incinerator.^1
$7,000,000.
Total Capital expenditure (factoring in other direct costs and indirect
costs)
$7,000,000 x 1.63 = $11,410,000.
Assume the destruction efficiency of the incinerator system is 99 percent
(Combustion temperature of 2,200 °F and a residence time of 1 second).
Thus, from emissions of 4,000,000 lbs/yr, 3,960,000 lbs/yr will be
destroyed.
C—7

-------
Utilities Required
Gas required (Assume the heat content of methane is 892 BTU/ft and the
heat recovered by the exchanger in the preheater system is 70 percent.)
Supplementary Heat required (Using Eq. 4.2-1 in the Handbook. )
Hf = 8.433xl06 BTU/min.
Gas Flowrate
8.A33x106/892 = 9,454 scfm.
Yearly Gas requirement = 4.969x10^ ft^/yr.
Fan electricity required (Assume a fan efficiency of 65 percent and a
pressure drop of 7 in. H20 across the control system.)
2.0x10 ^ x Flowrate x Press. Drop x Hrs =
Electricity required per year (kWh)
2.0x10 ^ x 500,000 x 7 x 8760 = 6,132,000 kWh/year.
Utilities Cost (June 1985 dollars)
Gas cost (Based on $0.00425 per ft^.)
4.969xl09 x 0.00425 = $21,118,000/yr.
Electricity cost (Based on $0,059 per kWh.)
6,132,000 x 0.059 = $361,788/yr.
Total Utility cost = $21,479,788/year or
$22,151,000/year (Sept. 1988 dollars)
Operating Labor Costs (June 1985 dollars)
Operator Labor (Assume 0.5 hrs/shift or 547.5 hrs/yr and a -ate of $11.53
per hr.)
547.5 x 11.53 = $6,312/year.
$6,510/year (Sept. 1988 dollars)
Supervision (Assume 15 percent of Operator Labor.)
0.15 x 6312 = $947/year
$977/year (Sept. 1988 dollars)
Total Labor Cost = $7,487/year
Maintenance Cost
Labor (Assume 0.5 hrs/shift of 547.5 hrs/year and a rate of $11.53 per
hour.)
547.5 x 11.53 = $6,312/year.
$6,510/year (Sept. 1988 dollars)
C-8

-------
Material (Assume 100 percent of Maintenance Labor.)
$6,510/year (Sept. 1988 dollars)
Total Maintenance Cost = $13,020/year
Indirect Operating Costs
Overhead	(80 percent of Operator, Supervisor and Maintenance Labor)
$ll,200/year.
Property tax (1 percent of Total Capital Cost)
$114,100/year.
Insurance (1 percent of Total Capital Cost)
$114,100/year.
Administration (2 percent of Total Capital Cost)
$228,200/year.
Capital Recovery (16.3 percent of Total Capital Cost assuming equipment
life of 10 years.)
$1,859,830/yr
Total Direct Operating Costs $22 ,171 ,507/year
Total Indirect Operating Costs $ 2,327,430/year
Annual Cost	$24,499,000/year or
$ 12,400/ton DCM destroyed.
As noted above, the annual cost presented does not included the cost of a
caustic scrubber to remove HC1 formed during the incineration process. Thus,
this annual cost estimate should be considered low.
A similar break down of cost can be prepared for a incinerator for the
emission stream from Building 20. Assuming a similar destruction efficiency
and heat recovery by the preheater system, and similar factors and costs, a
incinerator system would have the following costs:
Total Direct Operating Costs $6,656,507/year
Total Indirect Operating Costs $ 772,247/year
Annual Cost	$7 ,428,754/year or
$ 18,300/ton DCM destroyed.
This estimate also does not include the cost of a caustic scrubber. It should
be noted that because of the low heat content of the emission stream, the
majority of the cost for the thermal incinerator is the purchase of
supplemental fuel to maintain the proper combustion temperature.
C-9

-------
CONCLUSION
For the applications presented above, thermal incineration has been shown
to be the least cost effective means of the two systems examined for removing
DCM from an emission stream. This is due the large amounts of supplemental
fuel required to maintain the combustion chamber at the proper temperature.
REFERENCES
1.	Personal Communication between Stephen Walata, Alliance Technologies
Corporation, and Robert Saxer, Amcec Corporation, Oak Brook, IL.
December 13, 1988.
2.	Personal Communication between Stephen Walata, Alliance Technologies
Corporation, and Tim Cannon, VIC Division of Waltron Inc., Minneapolis,
MN. December 13, 1988.
3.	U.S. Environmental Protection Agency. Handbook: Control Technologies for
Hazardous Air Pollutants. EPA/625/6-86/014. Air and Energy Engineering
Research Laboratory. Research Triangle Park, NC. September 1986.
A. Personal Communication between Stephen Walata, Alliance Technologies
Corporation, and Ray Elsman, Huntington Energy Systems Inc., Union, NJ.
December 19, 1988.
5. Personal Communication between Stephen Walata, Alliance Technologies
Corporation, and Felix DaVarzo, Huntington Energy Systems Inc., Atlanta,
GA. December 16, 1988.
C-10

-------
APPENDIX D
Comparison of Technology Assessment with Kodak's BACT Report
D-l

-------
^ ALLIANCE
Technologies Corporation
December 21, 1988
TO:
Fred Dimmick, EPA/OAQPS
Charles Darvin, EPA/AEERL
FROM:
Richard Rehm
Stephen Walata^Au	»
SUBJECT: Comparison of Technology Assessment with Kodak's BACT Report
This memo is to fulfill Task 3 of Work Assignment 13 of Contract No. 68-
02-4396. Alliance personnel reviewed Kodak's Best Available Control Technology
Analysis for Dichloromethane Air Emission Sources at Kodak Park which will be
referred to as the Kodak report. The following is a comparison between this
report and Alliance's Source Characterization and Control Technology Assessment
of Methylene Chloride Emissions from Eastman Kodak Company, Rochester, NY (to
be referred to as the Alliance report) which was generated for Task 2 of this
work assignment. Each report's goal was to provide an assessment of control
technologies to reduce dichloromethane (methylene chloride or DCM) emissions
from the Kodak Park facility in Rochester, NY.
Both reports provided background information of the uses of DCM at Kodak
Park. The Kodak report provided a more detailed review of the processes which
use DCM including schematic diagrams of the processes. The Alliance report
chose to exclude a discussion of the processes to avoid the possibility of
revealing confidential information.
Both reports discussed the methods of emission estimates. Once again, the
Kodak report provided a more detailed review of the techniques used for each
estimation method. The Kodak report, however, did not address the possibility
that several emission estimates may be erroneous. Alliance discovered this
problem during the plant visit when Kodak personnel indicated that the emission
estimates of emission points 52-37 (Batch Mixers) and 21-12 (Kady Mill Exhaust)
were overestimations. For its part, Kodak assesses the accuracy of the
emission estimates by stating that on a mass basis, estimates for over 78
percent of the total amount of DCM emitted annually were made using monitoring
data. Still, there is no mention in the Kodak report of the possibility of
improving overall emission estimates.
Included in both reports are detailed descriptions of the category 1 and 2
emission sources at Kodak Park. The Kodak report included the projected
emissions which will result from the installation of the proposed acetate film
base casting machine. These emission estimates include the contained
emissions from the ventilation system of the new room which will house the new
machine and the emissions associated with the planned 18,000 cfm carbon
adsorber used for process air draw-off. The Alliance report included neither
of these estimates. During the plant visit by Alliance personnel, Kodak
1 D-2
500 Eastowne
Drive, Chapel Hill. North Carolina 27514 919-489-6550
a tjt r.nmnanv

-------
Fred Dimraick
Charles Darvin
December 21, 1988
Page 2
personnel indicated that the design plans for the new casting machine and new
carbon adsorber were not finalized. Thus, the projected emissions for the new
casting machine could not be accounted for with any degree of certainty. The
emissions projected for the new carbon adsorber had yet to be developed at the
time of the plant visit. What was included in the Alliance report were the
emissions from the 4,000 cfm carbon adsorber currently being used for process
air draw-off.
Both reports included process changes which are being planned by Kodak to
reduce DCM emissions. Kodak calls these process changes their Emission
Reduction Program. The process changes regarding the improvement of the
existing machine's seal integrity (called the Machine.Integrity Program by
Kodak) are similar in both reports. This is because the Alliance report
contains a verbatim account of the changes discussed during the plant visit.
Kodak also proposed to reduce slightly the positive pressure occurring in the
casting hopper of some machines. This will be possible because of the
improvement in the seal quality. The Alliance report misstated that the
anticipated completion date of the Emission Reduction Program was 1992 instead
of 1990 as reported by Kodak.
The Kodak report contains several sections which were not included in the
Alliance report. The subjects covered by these sections are:
A)	The decision criteria used by Kodak in the determination of the
best available control technology (BACT). Kodak defined what was
felt to constitute BACT through the cost effectiveness of the
proposed control. The cost effectiveness level which Kodak believes
is reasonable for a control technology is a cost of $2,000 to
3,000/ton of pollutant removed.
B)	The methodology used by Kodak to evaluate BACT for an individual
emission source. This included consideration used to evaluate
possible combining of emission sources.
C)	A statement of Kodak's manufacturing concerns regarding the
acetate film base. Included were general and worker safety,
equipment corrosion and product quality considerations.
The Alliance report excluded the discussion of these subjects since the
approach of the report was to provide potential control technologies which were
reasonable for Kodak to apply to the individual emission sources.
Kodak is planning to install a total hydrocarbon measurement system in
Building 20. The Alliance report concurs with this action if the proposed
measurement system will operate on a continuous basis. The Kodak report did
not indicate whether this would be the case.
-D-3

-------
Fred Dimmick
Charles Darvin
December 21, 1988
Page 3
Each report provided a discussion on control alternatives. The Alliance
report did not address the possibilities of process changes or material
substitution as in the Kodak report. Alliance personnel felt that the lack of
expertise in these areas would hamper efforts to suggest viable alternatives
and thus concentrated on add-on control devices. The Alliance report
discussed add-on control devices on a general level (i.e. oxidation and
recovery type devices). The Kodak report's discussion of add-on control
devices covered each individual type (i.e. condensers, absorbers, etc.). The
Kodak report also discussed several types of carbon adsorbers such as granular
beds, carbon fiber, combination systems and portable granular beds. The Kodak
report, for the most part, defended the assessment of the add-on control
devices by providing physical data either in the text or Appendix B. There is,
however, one exception to the supporting of assessments and this dealt with
absorption systems. Kodak provided vapor-liquid equilibrium data to support
the claim that water is a poor material to use in a scrubber. Yet when Kodak
discusses the use of alcohol in a multiple stage absorber, there are no data
provided to support Kodak's claim that such a system is a feasible means of
reducing DCM emissions. The Kodak report concludes that at the present time,
there is no viable solvent which can be substituted for DCM.
The Kodak report discusses the new casting machine, its projected
emissions and the application of a BACT analysis for it. The majority of the
discussion was concerned with changes made to the new machine's design, as
opposed to the existing casting machines, to provide what Kodak believes is the
best machine design. The discussion also included an evaluation of placing an
add-on control device (carbon adsorber and thermal incinerator) on the new
machine room exhaust and the exhaust from the machine air draw-off source (the
proposed 18,000 cfm carbon adsorber was assumed to be the base case.) Kodak
concluded that neither device would be a cost effective means of reducing DCM
emissions. Thus, the design of the new machine was chosen as BACT for the new
machine room exhaust and the base case was BACT for the machine air draw-off
source. The Alliance report did not address these issues regarding the new
casting machine.
Concerning the control alternatives evaluated for the existing emission
sources, Kodak has selected the base case as BACT for all except the following
emission points. For the existing machine room exhaust (emission points 53-85,
53-38 and 20-68), Kodak has selected as BACT the Machine Integrity Program. A
carbon adsorber was the control device selected as BACT for reducing the
emissions from the batch mixers (emission point 52-32). Kodak also plans to
reroute the vent flows from the hopper cleaning baths into the proposed
replacement carbon adsorber (18,000 cfm) and reroute the remaining floor sweeps
to be discharged with emission point 53-92. Kodak selected as BACT for the
storage vessel vent (emission point 53-96) and the floor sweeps (emission point
53-32/53-92) the replacement of level sensors; however, there is no description
in the Kodak report regarding this control alternative.
The Alliance report agrees with Kodak that the Machine Integrity Program
is a reasonable control alternative for reducing the DCM emissions from the
D-4

-------
Fred Dimmick
Charles Darvin
December 21, 1988
Page 4
existing machine room exhaust. The reports are in agreement about controlling
the batch mixer emissions with a carbon adsorber, however, the Alliance report
recommends Kodak consider the inclusion of the emissions from the felt wash
process (emission point 54-29) to this control device. Alliance concurs with
Kodak's plan to reroute the vent flows from the hopper cleaning bath to be
included with the flows to the proposed replacement carbon adsorber.
The Alliance report disagrees with Kodak's assessment of control
alternatives for several emission points. The first emission point of
contention involves the storage vessel vents (emission point 53-96). Kodak,
argues that even though the concentration of DCM may be high when the tanks are
being filled, the average concentration will be low due to infrequent use of
the system. Thus, a control device would not be a cost effective alternative
for reducing DCM emissions from this point. Kodak bases this assessment on the
premise that the control device would be dedicated to the emission source.
Alliance believes that including the vent flows from this emission source with
the flows to the proposed 18,000 cfm carbon adsorber would be a reasonable
control alternative. This source emits a greater amount of DCM per year than
the hopper cleaning bath vent flows. The average concentration of DCM in the
emission stream was calculated by Alliance (based on a flowrate of 1 cfm) to be
200,886 ppm. The variability of the flowrate from this source would be
negligible when compared to the total flowrate to the carbon adsorber. The
characteristics of this emission stream meet Kodak's criteria for combining
sources.
Another point of disagreement between the reports concerns the condenser
controlling the emissions from the Building 54 vent system (point 54-15).
Kodak contends that the only cooling fluid available for the condenser is 9°F
brine, consequently the condenser is only 50 percent efficient. Alliance
considers a properly operated condenser should have a removal efficiency of 95
percent for the DCM concentration present in this stream. If changes in the
operating parameters to achieve this removal efficiency cannot be made, then
Kodak should give serious consideration to replacing the current condenser with
a more efficient model.
The next point of disagreement between the two reports involves the
scrubbers (points 120-7 and 142-1) being used at the Kodak Park facility.
Kodak selected as BACT the base case. An EPA document cited by Alliance
suggests that a scrubber system works best when the influent concentration is
no greater than 10,000 ppm.* Currently the effluent concentrations from the
scrubbers are in excess of 10,000 ppm. Subsequently, Alliance recommended that
a dilution stream be applied to the influent stream of each scrubber.
The Kodak report provided a BACT analysis and potential emission reduction
plan for category 3 emission sources. This group of emission sources was
beyond the scope of the work assignment and were not included in the Alliance
report.
Both reports discussed the possibility of reducing fugitive (non-point
source according to Kodak) emissions at Kodak Park. Current efforts presently
employed at Kodak Park, for the most part, rely on visual inspections to detect
D-5

-------
Fred Dimmick
Charles Darvin
December 21, 1988
Page 5
leaks which are the source of fugitive emissions. The Kodak report presented
a three step program which Kodak considers as BACT for these emission sources.
The first step is to improve estimates and characterizations of fugitive
emission sources. The next step would be to take advantage of reduction
opportunities when warranted by cost, time, and reduction potential. The last
step would be to monitor reduction activities to quantify the progress being
made. The Kodak report does not make a projection as to the reduction in DCM
emissions as a result of this program. The Alliance report recommended the
implementation of a leak detection and repair (LDAR) program using EPA
Reference Method 21. EPA documents indicate that the institution of a LDAR
program can reduce fugitive emissions by 60 percent.
The Kodak report provided a detailde break down of costs for a carbon
adsorption system and a thermal incineration system in Appendix E. These,
systems were designed to control the total emissions from the machine room
exhaust in Building 53 (points 53-85 and 53-38). Additionally, Kodak provided
annual cost estimates for control systems (carbon adsorber and thermal
incinerator) for each individual emission source as part of their review of
different control alternatives in Appendix G. Alliance provided an addendum
to the original report in which the annual costs were estimated for the same
two control systems. Alliance combined all the uncontrolled emission streams
in Building 53 into one source to be processed by a control system. Cost
estimates for the control systems were developed for this source and the
machine room exhaust in Building 20.
The annual cost estimates developed by each report differ greatly, with
the Kodak report supplying the greater estimates. This difference is due to
the assumptions and decisions made by each report. The Kodak proposed systems
have in-line back-up capabilities which add to the overall cost of the system.
The Alliance systems are based on the equipment necessary to perform the task,
thereby minimizing the purchase cost of the system. In addition, the
materials-of-construction can add significantly to the cost of the system.
During the conversations with carbon adsorber system vendors, Alliance was told
that the use of titanium in constructing a system, as proposed by Kodak, would
effectively double the cost of that system. The material currently used in
carbon adsorber systems recovering DCM is 904 stainless steel. The Kodak
report does not fully explain the rationale behind having an adsorption cycle
for each tank of approximately 90 hrs. In carbon adsorber design equations,
the amount of carbon the system requires is proportional to the length of the
adsorption cycle.^ The longer the cycle, the more carbon would be needed in
the system. By choosing a shorter adsorption cycle, Kodak could reduce the
amount of carbon required by the system and therefore reduce overall costs.
Each report uses different costs for labor and utilities. The differences in
these values will contribute to the differences in the annual costs. The Kodak
report did not provide the assumptions or equations used to estimate the
required labor and utilities for each control system. Consequently, Alliance
was unable to determine the validity of these cost estimates. There are
factors which contribute to the lower annual cost estimates presented by the
Alliance report. The annual cost incurred for the replacement carbon in the
D-6

-------
Fred Dimmick
Charles Darvin
December 21, 1988
Page 6
adsorber system nor the cost of a caustic scrubber to remove HC1 produced
during thermal incineration were included in the final estimate. The latter
was due to Alliance's inability to contact a vendor of thermal incinerator
systems who was experience in the incineration of DCM. There is, however, one
point that both reports agree. That is for the applications presented in
either report, thermal incineration is the least cost effective means of
controlling a DCM emission source.
REFERENCES
1.	U.S. Environmental Protection Agency. Handbook: Control Technologies for
Hazardous Air Pollutants. EPA/625/6-86/014. Air and Energy Engineering
Research Laboratory. Research Triangle Park, NC. September 1986.
2.	U.S. Environmental Protection Agency. VOC Fugitive Emissions in Synthetic
Organic Chemicals Manufacturing Industry - Background Information for
Proposed Standards. EPA-450/3-80-033a.* Office of Air Quality Planning
and Standards. Research Triangle Park, NC. November 1980.
3.	Personal Communication between Stephen Walata, Alliance Technologies
Corporation, and Tim Cannon, VIC Division of Waltron Inc., Minneapolis,
MN. December 13,1988.
(*) NTIS No. PB81-152167.
D-7

-------