EPA
United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EMB Report 79-GRA-1
March 1979
Air
Graphic Arts
Emission Test Report
Meredith-Burda
Lynchburg, Virginia
-------
GRAPHIC ARTS EMISSION TEST REPORT
Plant Sampled
MEREDITH-BURDA, INC.
4201 Murray Place
Lynchburg, Virginia 24506
Project No.: 79-GRA-l
April 4, 1979
Contract 68-02-2818, Work Assignment No. 16
Prepared for
Environmental Protection Agency
Emission Measurement Branch
Research Triangle Park, N.C. 27711
by
William R. Feairheller
MONSANTO RESEARCH CORPORATION
DAYTON LABORATORY
Dayton, Ohio 45418
-------
TABLE OF CONTENTS
-Section Page
I Introduction 1
II Summary and Discussion of Results 3
III Process Description and Operation 17
A. The Printing Operation 17
B. Fugitive Emissions 18
C. The Emission Control System 18
D. Operating Conditions 20
IV Location of Sampling Points 21
V Sampling and Analytical Procedures 29
111
-------
LIST OF TABLES
Table Page
1 Summary of Sampling and Analytical Methods 4
2 Summary of Results 5
3 Species Analysis Data 14
Run 1 12/14/78
4 Species Analysis Data 15
Run 2 12/15/78
5 Species Analysis Data 16
Run 3 12/16/78
IV
-------
LIST OF FIGURES
Figure Page
1 Plot of the variation in inlet, outlet and vapor 11
control enclosure concentrations and press
operating conditions with time - Run 1
December 14, 1978
2 Plot of the variation in inlet, outlet and vapor 12
control enclosure concentrations and press
operating conditions with time - Run 2
December 15, 1978
3 Plot of the variation in inlet, outlet and vapor 13
control enclosure concentrations and press
operating conditions with time - Run 3
December 16, 1978
4 Meredith/Bur da fume collection system 19
5 Process diagram showing sampling locations 22
6 Atmospheric exhaust from carbon adsorbers 23
(o,93°F, 62,000 CFM) Site 1
7 Diagram of duct between manifold and filter 24
62,000 CFM) Site 2
8 Modification of the duct work at Site 2 25
9 Diagram of sampling location for Sites 8 and 9, 27
manifold from the vapor control enclosures
of Press 505 and 506
-------
I. INTRODUCTION
The Meredith/Burda Inc. plant located at 4201 Murray Place,
Lynchburg, Virginia, was sampled during the week of Dec. 11-16,
1978. The objective of the sampling program was to obtain vol-
atile organic compound emission data from a well-controlled
plant in order to provide background data for the establishment
of new source performance standards. This plant, representing
the Graphic Arts Industry category, employs toluene as a solvent
for the printing ink and emission of organic vapors are con-
trolled by a Lurgi designed carbon adsorption system. The sam-
pling and on-site analysis was conducted by a Monsanto Research
Corporation team consisting of W. R. Feairheller (team leader),
W. McCurley, W. Meyer, L. Cox, and C. Clark, and was observed by
Mr. Frank Clay of the Emission Measurement Branch of EPA.
Mr. Richard Reich of Radian Corporation was on-site to obtain
process design information.
Total hydrocarbon measurements were made on a semicontinuous
basis at both the inlet and outlet of the carbon adsorbers.
Grab samples were collected at the inlet and outlet adsorber
sites and at the ventilation ducts from vapor control enclosures
around both press units (505 and 506). The grab samples were
analyzed on-site by gas chromatography to determine both identi-
fication and concentration of the components of the gas streams.
Samples of diluted ink were obtained from each of the eight feed
tanks on each press for determination of the toluene content.
In addition, samples of boiler feed water and the water from the
toluene/water decanter (separator) were also collected for
toluene content analysis.
-------
The test program consisted of three sampling periods. Each test
period was defined as the time required for all of the three
carbon adsorbers to complete a sorption-desorption cycle. The
test periods ranged from about 8*5 to 9 hours during which each
adsorber remained in the adsorption cycle for 160-180 minutes
followed by a 50 minute steam desorption and conditioning, cool-
ing cycles. In the event that breakthrough of the solvent
vapors occurred on a particular adsorber, the adsorption cycle
for that adsorber was terminated early and the desorption cycle
began.
During the test program, velocity data and liquid samples were
collected by Robert Oppenheimer of the Gravure Research Insti-
tute, located in Port Washington, N.Y., 11050. No information
is available on these results. A test engineer from Lurgi was
on-site during the test program in order to monitor the Opera-
tion of the adsorber system.
-------
II. SUMMARY AND DISCUSSION OF RESULTS
A summary of the sampling and analytical procedures employed at
each site is given in Table 1. The procedures used followed the
EPA test request with one exception. The grab samples of ink
could not be obtained directly; however, samples of the diluted
ink ready for use, containing ink (one of 4 colors), two types
of varnish and diluent toluene, were obtained from each of the
eight feed tanks on each press. Separate meters on each of the
eight ink feed systems provided the amount of ink and varnish
added to the mix tank. The overall amount of toluene used as a
diluent for each press was obtained from the recovery meters.
Each of the diluted ink samples was analyzed for toluene content.
The total amount of toluene used by each press during the sam-
pling period was calculated from the various amounts of each com-
ponent and the amount of toluene in each.
Sites 8 and 9, the uptake ducts from the vapor control enclosures
on presses 505 and 506, were added to the program as additional
sites. It was originally intended that Century System OVA ana-
lyzers would be used at these locations, however, it was found
that the high negative static pressure at these locations pre-
vented their use. Grab samples were collected on an hourly basis
from each site for analysis by gas chromatography- Periodic
velocity measurements were made on both sites during the test
period on 12/16.
A summary of the results, including operation parameters, veloc-
ity measurements and hydrocarbon data, is presented in Table 2.
Additional clarification of the items in Table 2 is given below:
-------
Table 1. SUMMARY OF SAMPLING AND ANALYTICAL METHODS
Location
1. Outlet from Adsorber
Parameter
2.
3.
4.
4a.
Inlet to Adsorber
Decanter
Press Ink Supply
Meter
Press Varnish
Meter
5. Recovery Meter
6. Decanter
6a. Steam Orifice Meter
7. Boiler Room
8,9. (Optional)
Enclosure Uptakes
Total hydrocarbon
Organic species
Temperature
Moisture
Velocity
Static pressure
Total hydrocarbons
Organic species
Toluene recovered
Amount of ink used
Sample diluted ink
Amount of varnish added
(2 types)
Toluene added to ink and
varnish
Decanter water
Steam used for desorption
Boiler feed water
Velocity
Organic species
Sampling Method
Direct coupling
Grab sampling
Type K thermocouple
Method 4 (12/14 only)
Wet bulb/dry bulb
(12/15, 16)
Type S pitot tube
Type S pitot tube
Direct coupling
Grab sampling
Read meter
Read meter
8-grab samples/press
Read meter
Read meter
Grab sample
Read meter
Grab sample
Type S pitot tube (12/16)
Grab sample
Analytical Method
FID
GC (on-site)
Digital thermometer
Gravimetric
Pyschometric chart
Inclined manometer
Slack tube manometer
FID
GC (on-site)
Calculation
Calculation
% toluene in ink
Calculation
Calculation
ppm toluene in water
Calculate H20 volume
ppm toluene in water
Inclined manometer
GC (on-site)
-------
Table 2. SUMMARY OF RESULTS
Parameter (Units)
Operational Parameters
Date
Start time (24 hour basis)
Finish time (24 hour basis)
Elapsed time (minutes)
Press run time 505 (minutes)
Press run time 506 (minutes)
Volume of diluted ink (liters)
Total toluene used (liters)
Total ink used (liters)
Total varnish used (liters)
Toluene recovered (liters)
Toluene in decanter H20 (ppm)
Steam rate (Ib/hr)
Total desorption time (minutes)
Condensed water (liters)
Toluene/dil. ink ratio
Varnish/ink ratio
Toluene recovered/tol. used ratio
Toluene in feed water (ppm)
Velocity Parameters
Inlet
Area (ft2)
Average temperature (°F)
Moisture (%)
Molecular weight gas
Velocity (ft/sec)
Vol. flow rate (SCFM)
Toluene Concentrations
Inlet
Both press running (ppm)
Only 505 running (ppm)
Only 506 running (ppm)
Both down
Outlet
Normal operation (ppm)
Breakthrough max. (ppm)
Vapor Enclosure 505
Press down (ppm)
Press running (ppm)
Run 1
Run 2
Run 3
12/14
1251
2120
509
418
288
5030.3
4186.7
1623.5
513.6
3975
1985
11000
150
12500
0.83
0.32
0.95
<3
12/15
940
1840
540
463
382
6267.6
5253.1
1857.5
751.9
4945
720
11000
150
12500
0.84
0.40
0.94
<3
12/16
803
1633
510
457
323
5516.5
4605.7
1735.0
571.1
4705
715
11000
150
12500
0.83
0.33
1.02
<3
33.82
84.3
1.33
28.70
26.07
47140
33.82
88.0
1.76
28.64
28.64
49910
33.82
(A)
1.23
28.72
(A)
(A)
1450
600
750-1000
220
18
(B)
110
430-510
1750
775
1275
290
12
169
240
500
1850
850
1225
275
15
235
140
500
-------
Table 2 (continued). SUMMARY OF RESULTS
Parameter (Units)
Run 1
Run 2
Toluene Concentrations - Continued
Vapor Enclosure 506
Press down (ppm)
Press running (ppm)
150
660
140-210
760-870
Run 3
150
620-760
NOTES: (A) Separate measurements were made on these parameters during
Run 3. Average values are as follows:
Inlet - presses down
Inlet - 505 running
Inlet - 506 running
Inlet - both running
Vapor enclosure 505 running
Vapor enclosure 506 down
Vapor enclosure 506 running
Temp
(°F)
86
89
96
94
99
92
102
Velocity
(ft/sec)
19.14
22.48
25.35
27.55
19.77
17.99
18.12
Vol. Flow Rate
(SCFM)
31710
39340
43430
48840
7454
6885
6897
(B) No breakthrough occurred during this run.
-------
(1) The volume of diluted ink includes the ink,
toluene and varnish in the diluted ready-to-
use form. The diluted ink contains 73 to
81.8% toluene.
(2) Total toluene used includes diluent toluene
and the amount of toluene in the supplied ink
(53.9 to 58.8% depending on the color) and the
amount of toluene in the two varnish formulations
(71.9 and 76.6%).
(3) The total ink and total varnish used includes
both the pigment and toluene solvent as received
and were measured by the sum of the individual
press meters.
(4) Toluene recovered is obtained from the decanter
meter which indicates the amount of toluene col-
lected from the adsorber and returned to storage
tanks.
(5) The toluene in decanter water was obtained by
GC analysis on grab water samples. The sample
from Run 1 contained toluene droplets and thus
gave an abnormally high value. Under typical
conditions, this value would not exceed the
maximum amount of toluene that was soluble in
' water at about 70°P.
(6) The steam rate was determined by an orifice
meter in the steam supply line. Steam was sup-
plied at a rate of 11,000 Ib/hr for each 50
minute desorption period, a total of 150 minutes
for each sampling period.
-------
(7) The toluene recovered to toluene used ratio on
Run 3 exceeded 1.0. It would appear that the
final readings of the decanter meter were taken
during a time when toluene was being delivered
to the storage tank. It is our understanding
that toluene is pumped to storage only when the
toluene in the decanter reaches a specific level,
and thus the flow is not continuous.
(8) Velocity measurements on 12/14 and 12/15 were
obtained just prior to the start of Runs 1 and
2. During Run 3 periodic measurements were made
at both the inlet and at the vapor control en-
closure ducts and the data given in Note A indi-
cates the changes in temperature, flow rate,
and velocity with changes in the press operation.
(9) The toluene concentration in the inlet varies
with the operating conditions of the presses.
Semicontinuous monitoring of the total hydro-
carbon concentration with the FID makes it pos-
sible to determine the changes in concentration
for the various operating modes. The concentra-
tion differences shown by presses 505 and 506
are the result of different types of material
being printed. Press 506 was printing a mag-
azine section with high ink and varnish coverage,
whereas press 505 was being used for an adver-
tisement with lower ink and varnish coverage.
These differences are also reflected in the
toluene levels determined from the vapor enclo-
sure ducts.
-------
(10) The total hydrocarbon concentration at the out-
let is independent of the press operation pro-
vided the capacity of the adsorber and the time
of adsorption is such that all of the toluene
from the press can be adsorbed. Normally, at
any one time at least two of the adsorbers
should be in the adsorption cycle. During
Runs 2 and 3, adsorber 1 showed evidence of
breakthrough, with greatly increased toluene
emissions. The Lurgi engineer indicated that
for some reason that adsorber was not being fully
stripped, breakthrough occurred and the vapor
detector (Solvamet) started the desorption cycle
long before the normal time period. Based on the
adsorber control panel strip chart, it would ap-
pear that Adsorber 1 should begin a desorption
cycle before Unit 3 had completed its desorption
cycle and thus Unit 1 remained in the adsorption
period while it was saturated, thus increasing
the outlet emissions.
(11) Based on the organic species analysis of inlet,
outlet, and vapor control enclosure duct sam-
ples, the organic vapors in the inlet and en-
closure consist of more than 99% toluene, how-
ever, the outlet contains toluene, benzene and
another unidentified material at about a 1:1:1
ratio. Thus, the assumption that the total hy-
drocarbon values at the outlet are due solely
to toluene is erroneous and the results are,
therefore, misleading.
The variation in inlet and outlet total hydrocarbon concentra-
tions, the toluene content of the control enclosure duct samples
-------
and press operation during each of the sampling runs is shown in
Figures 1, 2 and 3 for Runs 1, 2 and 3, respectively. The var-
iations of inlet and control enclosure concentrations are clearly
shown to be directly influenced by the press operation.
Summaries of the species analysis for each sampling run are given
in Tables 3, 4 and 5. Toluene concentrations were obtained on-
site. The benzene concentrations given were obtained using a
laboratory calibration curve as no benzene standards were avail-
able in the field, and thus must be considered as approximate.
The data shows that the benzene levels are not significant in
the inlet and vapor control enclosure samples. This is not true
at the outlet. An estimation of the ethanol, benzene and toluene
content was attempted for a number of the outlet samples and the
total of the three component concentrations agreed in most cases
with the total hydrocarbon values obtained by the FID analyzer
on-site. During the presurvey, the plant personnel indicated
that only toluene should be present at all sites and thus we
were not prepared to analyze for the quantities of benzene and,
in several samples, ethanol that were observed. It would have
been desirable to collect bag samples of the outlet gases during
the presurvey so that GC/MS analysis prior to the field trip
would have indicated the need to analyze for these other com-
ponents on-site. The benzene occurs as a trace contaminant in
the toluene and based on limited data, it appears that the ad-
sorbers do not efficiently trap this material. Ethanol, used to
a limited extent in cylinder making, was detected in an inlet
sample analyzed by GC/MS after the field trip.
10
-------
IIM
10 PH
TIME
Figure 1. Plot of the variation in inlet, outlet and vapor control enclosure concentrations and
press operating conditions with time - Run 1 December 14, 1978
-------
259
Figure 2. Plot of the variation in inlet, outlet and vapor control enclosure concentrations and
press operating conditions with time - Run 2 December 15, 1978
-------
1S8
Figure 3. Plot of the variation in inlet, outlet and vapor control enclosure concentrations and
press operating conditions with time - Run 3 December 16, 1978
-------
Table 3. SPECIES ANALYSIS DATA
Run 1 12/14/78
Concentration of Toluene (ppm)
Location
Cone.
Benzene
Time
Inlet
Outlet
505
506
506
505
Inlet
505
506
505
506
Inlet
505
506
Outlet
505
506
505
506
Inlet
505
506
Outlet
1230
1235
1300
1308
1400
1405
1415
1510
1515
1605
1615
1630
1725
1735
1745
1830
1835
1935
1940
1950
2030
2035
2055
Sample 1
Sample 2
Sample 3 Average
954
2.8
221
101
575
326
1188
365
134
514
668
1322
506
229
1.6
530
676
439
157
726
446
148
1.9
979
2.6
217
105
582
307
1180
349
140
518
660
1322
506
226
1.6
530
666
434
148
726
425
150
1.7
954
2.9
205
101
581
319
1207
356
130
512
685
1322
489
231
1.6
482
668
427
150
711
435
142
1.8
962
2.
214
102
579
317
1192
357
135
515
671
1322
500
229
1.
514
670
433
151
721
435
147
1.
8
6
8
2.7
3.7
5.3
2.8
3.0
14
-------
Table 4. SPECIES ANALYSIS DATA
Run 2 12/15/78
Concentration of
Location
505
506
Inlet
Outlet
505
506
Inlet
Outlet
505
506
Inlet
506
505
Outlet
Inlet
505
506
Inlet
505
506
Outlet
506
505
Inlet
Outlet
506
505
Inlet
506
505
Outlet
Time
940
945
1020
1025
1040
1045
1120
1135
1142
1147
1220
1250
1254
1332
1325
1342
1348
1425
1450
1455
1535
1605
1600
1620
1650
1710
1700
1730
1815
1820
1750
Sample 1
553
763
1516
8.1
571
207
325
7.3
464
790
1561
903
519
3.5
837
517
820
1524
247
792
1.5
807
498
1308
2.5
708
431
1500
141
475
1.6
Sample 2
565
759
1530
7.5
569
203
813
7.4
479
763
1522
865
519
3.4
837
498
800
1482
242
773
1.4
751
476
1337
1.9
429
1520
138
468
1.2
Toluene (ppm)
Sample 3
565
755
1532
7.6
567
201
843
7.5
475
730
1565
830
514
3.3
837
498
804
1502
247
748
1.5
745
476
1288
1.7
1467
136
451
1.2
Average
561
759
1526
7.7
569
204
827
7.4
473
761
1549
866
517
3.4
837
504
808
1503
245
771
1.5
768
483
1311
2.0
708
430
1496
138
465
1.3
Cone.
RPTI 5^^rtP
CCl 1 £
-------
Table 5. SPECIES ANALYSIS DATA
Run 3 12/16/78
Concentration of
Location
506
505
Inlet
505
506
Outlet
506
Inlet
505
Outlet
506
505
Inlet
506
505
Inlet
Outlet
506
505
Inlet
Outlet
506
505
Outlet
506
505
Inlet
Time
820
815
830
910
915
900
1012
935
1005
1110
1135
1130
1200
1240
1235
1255
1310
1410
1403
1435
1450
1508
1503
1548
1615
1610
1620
Sample 1
710
139
1553
520
154
1.5
350
1437
524
1.2
706
463
183
725
473
1585
7.6
722
552
1460
2.7
637
527
1.2
675
495
1182
Sample 2
726
138
1572
520
150
1.3
340
1484
524
1.1
463
183
725
459
1545
7.4
676
549
1413
2.6
626
515
1.4
645
476
1215
Toluene (ppm)
Sample 3
694
134
1552
148
1.6
334
1477
513
.9
463
695
462
1559
725
544
1426
2.8
601
526
1.4
645
468
1182
Average
710
137
1552
520
151
1.5
341
1466
520
1.1
706
463
183
715
465
1563
7.5
708
548
1433
2.7
621
523
1.3
655
480
1193
Cone.
Benzene
(PPJ"1.)
26.1
11.4
3.0
9.G
10.5
20.7
3.3
21.6
5.9
11.4
22.0
10.0
3.7
9.6
3.8
9.1
21.1
2.7
16
-------
III. PROCESS DESCRIPTION AND OPERATION
The Meredith/Burda Inc. plant operates a total of six rotogravure
publication presses. The two newest presses, press 505 and 506,
were the only presses monitored during the course of this test.
These two presses, located in a separate press room, are con-
trolled by a carbon adsorption solvent recovery system.
A. THE PRINTING OPERATION
Meredith/Burda uses a pure toluene solvent in all their printing
operations. Press 505 and 506 are new, 8-unit Cerutti presses
with a two meter web capacity. The engraved cylinders used were
approximately 34 inches in circumference. Each printing unit is
equipped with a dual uptake dryer. These dryers function as two
separate drying systems, by applying heated air to both sides of
the moving web. This thorough drying action minimizes the re-
tained solvent in the final product, thus increasing capture ef-
ficiency. Each dryer is designed to concentrate the solvent in
the drying airstream by internal exhaust recirculation. The ex-
haust from each dryer is drawn into a collection header contain-
ing the fugitive emissions exhaust stream. The exhaust flow rate
from each dryer varies depending upon the press speed, ink color,
percent ink coverage, and desired solvent exhaust concentration.
Under most circumstances, the exhaust flow will be about 1000 CFM
for each uptake, or a total of 2000 CFM per unit.
17
-------
B. FUGITIVE EMISSIONS
A unique system is utilized in this press room to capture fugi-
tive hydrocarbon emissions. A sheet metal enclosure (called a
cabin) surrounds the upper area of each press line. A schematic
diagram of this fume collection system is illustrated in Figure 4.
The cabin encloses the top one-third of each printing unit and
extends several feet above, for the length of the press line.
The cabin area is constructed with rectangular fume pickup noz-
zles located at the bottom. These nozzles, positioned between
and around each printing unit, draw solvent laden air (SLA) into
the cabin. This air is removed from the cabin by an exhaust
header system as shown in Figure 4. Each cabin exhaust system
was designed to remove about 10,000 CFM, of 100°F air at about a
200 ppmv toluene level. In cold winter months, this airstream is
passed through a "heat wheel" heat exchanger. In this manner,
waste heat is used to heat and humidify the fresh outside air be-
fore it enters the press room ventilation system. The "heat
wheels" do not affect the emission control system.
C. THE EMISSION CONTROL SYSTEM
Presses 505 and 506 are controlled by a new Lurgi "Supersorbon"
carbon adsorption system. A schematic diagram of the equipment
layout is presented in Figure 5 of Section IV. In essence, the
system consists of three adsorption vessels containing activated
carbon. Two vessels adsorb simultaneously while the third vessel
is stripped using countercurrent live steam injection. The re-
covered solvent/steam mixture is condensed, cooled, and separated,
The dewatered solvent, from the decanter, is sent to the recycle
solvent storage tank. The water layer (the condensed steam) is
sent to a condensate stripper, where it is contacted countercur-
rently with air, and then reused as boiler feed water. The
toluene laden airstream is then recycled back into the adsorber
induction system.
18
-------
Dryer Exhaust
Cabin Exhaust
-Fume Pickup Monies
Outside Air
•Dual Uptska Dryer Outlet*
Press Exhaust to Booster Fin
Figure 4. Meredith/Burda fume collection system
-------
Each adsorption vessel is rated for a flow of 50,000 cubic meters
per hour (about 30,000 CFM) . The solvent laden air from the
dryers and cabins is drawn through an induction header by two
250 HP fans. A third 250 HP fan acts as a spare. After filter-
ing and cooling, the airstream is forced into the adsorption ves-
sels. The air leaving the adsorption system is discharged into
the atmosphere.
The regeneration of the adsorption vessels is initiated automati-
cally when a detector senses solvent breakthrough. A timer serves
as a backup, triggering regeneration automatically after a fixed
period of slightly more than two hours. The 50 minute steaming
cycle is followed by a cooling cycle.
D. OPERATING CONDITIONS
The plant operated in a normal fashion during the three, nine-
hour tests conducted on December 14, 15, and 16. Normal opera-
tion includes periodic press shutdowns due to paper web tearing,
poor color reproduction, and other mechanical problems. Press
505 printed a furniture advertising booklet using a 50-inch web.
Press 506 printed a color feature section of a magazine using a
78 3/8-inch web.
In general, the test ran very well. The actual press room print-
ing load was less than maximum capacity, but certainly represented
typical conditions. Since the press room solvent loading was
lower than the maximum level, the regeneration cycle was usually
triggered by the timer. This condition increases the adsorption
efficiency while decreasing the steam utilization efficiency.
This situation would not appreciably affect the results of the
test.
20
-------
IV. LOCATION OF SAMPLING POINTS
A schematic diagram of the process indicating the sampling loca-
tions is shown in Figure 5. Site 1 was located at the absorber's
outlet to the atmosphere. Details of the duct work at this site
are shown in Figure 6. Site 2 was located on the duct work be-
tween the headers from the press ventilation system and the
filters and fans. Details of the duct work at this site are
shown in Figure 7. Prior to sampling, the duct work at Site 2
was modified as shown in Figure 8. Site 3 required no sampling,
however, a meter on the line between the decanter and the sol-
vent storage tank provided an indication of the amount of
toluene (in liters) delivered to the storage tanks. Sites 4
and 4a involve meters that read the amount of ink used in the
diluted ink mixture (4), and the varnish used - 2 meters (4a).
These meters are part of the control system that meters the
proper amount of each component into the diluted ink tank, of
which there are eight on each press. An additional meter at
this location provides a measure of the total ink and varnish
delivered to the mix tank. As there was some disagreement be-
tween the individual component meters and the total meter, the
individual meter readings were used to analyze the data. Sam-
ples of the diluted ink ready for use were collected from the
eight ink feed tanks on both presses during each run.
Reading of the recovery meter, Site 5, indicates the total
amount of toluene, delivered to each press, required to dilute
the concentrated ink and varnish mixtures. The meter readings
at the beginning and end of the test period were recorded. A
21
-------
to
I I I I I I I I
y I '"" It I i V 4 i •
f'TI'T'l i i
pow-s
Htnc*
iu,
I/UK 14IMW* T>»*>K.
srres - /, 31,4, 6 ,1
3(f»» *!*,
Figure 5. Process diagram showing sampling locations
-------
T
9'
c*s
NJ
CO
Figure 6. Atmospheric exhaust from carbon adsorbers
, 62,000 CFM) Site 1
-------
1
C****£
,
Jo
y.e f *
/
rf *;«g -p, * "*
i A
Figure 7. Diagram of duct between manifold and filter
(M.10-115°F, 62,000 CFM) Site 2
-------
tf B«Ttl
I
I
vi f fie
/
T08R
/?
U1
NOTES: 3" half nipples - cut 8" long 3" pipe nipple in half and weld unthreaded end to plate and duct.
(std. pipe threads)
Close off nipples with 3" caps before and after sampling.
1" half nipple - cut 6-8" long 1" pipe nipple in half, cut hole and weld in place.
Figure 8. Modification of the duct work at Site 2
-------
grab sample of the lower water layer was collected during each
run at Site 6, the decanter which separates the toluene and
water after desorption. A valve was provided in the lower sec-
tion of the decanter for this purpose. A grab sample of the
boiler feed water was collected from the outlet end of the
stripper during each run (Site 7). A valve in the feed water
line was provided for sampling. Sites 8 and 9 were added after
the presurvey- These sites were located on the manifold duct
from the vapor control enclosures on presses 505 and 506 prior
to the heat wheel and the manifold that received the individual
uptake ducts from the press area outside of the vapor control
enclosure. The two identical sites, one of which is shown in
Figure 9, are not ideal from a velocity measurement viewpoint,
however,- no other location was found that would provide more
reliable data.
26
-------
to
Figure 9. Diagram of sampling location for sites 8 and 9, manifold from the
vapor control enclosures of Press 505 and 506.
-------
V. SAMPLING AND ANALYTICAL PROCEDURES
The procedures for the determination of the temperature, static
pressure, velocity, moisture and volumetric flow rate followed
the procedures given in Methods 1, 2, and 4 as presented in the
Federal Register (August 18, 1977).
The total hydrocarbon measurements were obtained by directly
coupling a flame ionization analyzer to both Sites 1 and 2 using
heated sampling lines. The sample probe consisted of a suitable
length of V 316 stainless steel tubing. At the exit of the
duct, a stainless steel filter assembly containing a sintered
metal filter was installed to act as a flame arrester. The
filter was attached to a valve and then a tee, one leg of which
was used to inject calibration gases. The other leg was con-
nected to a heated teflon sample line which connected to the
inlet of the FID instrument. An AID Model 511 chromatograph,
operated in the total hydrocarbon mode, was used at Site 2. At
Site 1, a Varian 1400 gas chromatograph was used as a total
hydrocarbon analyzer by removing the column and substituting a
minimum length of 1/8" stainless tubing between the sample loop
and the detector. The outlet of the gas sampling valve was
connected through a tee and an isolation valve, to an integrated
gas sample type pump (Method 3, Federal Register). The valves
were operated in a manner so that the sample was drawn into the
sample loop, the pump and stack were isolated from the loop,
the gas in the loop allowed to equilibrate at atmospheric pres-
sure and then injected into the instrument. Standards and
source gas were sampled identically-
29
-------
Grab gas samples from Sites 1, 2, 8 and 9 were collected using
a hand operated sampling pump to fill 12 x 12" teflon sampling
bags. Separate bags were used for each site and were flushed
and leak checked with nitrogen between each run. The collected
samples were analyzed on-site using a Hewlett-Packard Model 5750
gas chromatograph equipped with a heated gas sampling valve, a
1/8" x 6' stainless steel column containing 5% SP1200/1.75%
Bentone 34 on 100/200 Supelcoport and a flame ionization detector.
The chromatograph was operated in an isothermal mode at 80°C.
Ink samples were collected in 45 ml vials with teflon backed
septum caps of the type recommended for priority pollutant water
analyzis. The bottles were filled to the top (no void space),
capped and maintained in an ice chest (0°C) prior to analysis.
These samples were analyzed in the laboratory by injecting
1 u liter of the liquid into a Perkin Elmer 3920 gas chromatograph
equipped with a 1/8" x 20" 10% FFAP column and a flame ionization
detector. The column was operated isothermally at 90°C.
Water samples were collected in 1 liter Wheaton bottles (with
teflon lined caps), previously cleaned and heated in an oven to
remove organic compounds. The bottles were filled to the top to
eliminate void space, and held at 0°C prior to analysis. The
samples were analyzed by direct injection on a Perkin Elmer 3920
gas chromatograph equipped with a 1/8" x 10' Poropak P column,
operated at 170°C, and a flame ionization detector.
The total hydrocarbon analyzer and the HP 5750 were calibrated
using commercial toluene in nitrogen mixtures and site-prepared
standards using the Ford Motor Company "Solvent Standard Prepa-
ration Procedures", May 18, 1977.
30
------- |