EPA
United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EMB Report 86-CAD-2
November 1986
Air
Cadmium
Screening
Study Test Report
Cadmium Sulfide
Pigments
Harshaw/Filtrol
Partnership
Louisville,
Kentucky
-------
MESHAP DEVELOPMENT
CADMIUM MISSION TESTING AT THE
HARSHAW/FILTROL PARTNERSHIP
LOUISVILLE, KENTUCKY
JULY 1986
EPA Contract No. 86-02-4337
ESED Project No. 80/42
Work Assignment No. 1
CAD-2 Louisville, KY
TRC Project No. 3497-E81-90
Prepared for:
Michael Toney, EPA/EMB
Task Manager
Prepared By:
John H. Powell
Work Assignment Manager
Richard A. Pirolli
Project Engineer
November 1986
-------
TABLE OF CONTENTS
SECTION PAGE
PREFACE vii
1.0 INTRODUCTION 1
1.1 Background 1
1.2 Summary of Process 2
1.3 Applicability of EPA Reference Test Methods .... 2
1.3.1 EPA Method 5 Configuration Modified for Cadmium . 2
1.4 Measurement Program Summary 3
1.4.1 Spray Tower and Venturi 3
1.5 Report Sections 4
2.0 SUMMARY AND DISCUSSION OF RESULTS 5
2.1 Spray Tower 5
2.1.1 Cadmium Emissions 5
2.1.2 Visible Emissions 5
2.1.3 Stack Conditions 9
2.2 Venturi Scrubber 9
2.2.1 Cadmium Emissions 9
2.2.2 Visible Emissions . 15
2.2.3 Stack Conditions 15
2.3 Trace Metals 15
3.0 PROCESS AND OPERATIONS 21
3.1 Process Equipment 21
3.2 Control Equipment 21
4.0 SCOPE OF SAMPLING PROGRAM 22
4.1 Belt Dryer 22
4.2 Rotary Calciner and Vacuum Pan Dryer 22
4.3 Visible Emission Observations (Method 9) 26
5.0 SAMPLING AND ANALYTICAL PROCEDURES 29
5.1 EPA Method 5 Configuration Modified for Cadmium . . 29
5.2 Sample Recovery 31
5.3 Sample Analyses 33
5.4 Visible Emissions Observation 34
6.0 QUALITY ASSURANCE 36
-111-
-------
APPENDICES
A EXAMPLE EQUATIONS AND CALCULATIONS
A.I Cadmium
B FIELD DATA SP ETS
B.I EPA Method 5
B.I.I Spray Tower
B.I.2 Venturi Scrubber
C SAMPLING LOGS
C.I Daily Summary Log
C.2 Sampling Task Logs
C.3 Sample ID Log
C.4 Chain of Custody Forms
C.5 Crew Chief Field Operation Log
C.6 Recovery Sample Sheet
C.7 Train Operators Field Log
D VISIBLE EMISSIONS
D.I Observer Certification
D.2 Visible Emissions Field Data Sheets
D.2.1 Spray Tower Visible Emissions
D.2.2 Venturi Scrubber Visible Emissions
E CALIBRATION DATA
E.I Summary of Equipment Used During Testing
E.2 Orifices, Dry Gas Meter
E.3 Pitot Tubes
E.4 Nozzles
F SAMPLING AND ANALYSIS PROCEDURES
F.I EPA Method 5 Modified for Cadmium
F.2 EPA Method 9 (Visible Emissions)
G LABORATORY ANALYSIS DATA
G.I Cadmium
G.2 Trace Metals
H PARTICULATE
-iv-
-------
LIST OF FIGURES
FIGURE PAGE
2-1 Summary of Visible Emissions, Spray Tower 103, Harshaw/
Filtrol Partnership 10
2-2 Summary of Visible Emissions, Spray Tower 103, Harshaw/
Filtrol Partnership 11
2-3 Summary of Visible Emissions, Spray Tower 103, Harshaw/
Filtrol Partnership 12
2-4 Summary of Visible Emissions, Venturi Scrubber 201, Harshaw/
Filtrol Partnership 17
2-5 Summary of Visible Emissions, Venturi Scrubber 201, Harshaw/
Filtrol Partnership 18
2-6 Summary of Visible Emissions, Venturi Scrubber 301, Harshaw/
Filtrol Partnership 19
4-1 Belt Dryer Venturi Scrubber Outlet 23
4-2 EPA Method 1, Sample and Velocity Traverse for Stationary
Sources 24
4-3 Calciner and Pan Dryer Spray Tower Outlet . 25
4-4 EPA Method 1, Sample and Velocity Traverse for Stationary
Sources 27
4-5 Observers Location Visible Emission, Spray Tower and Venturi
Scrubber, Harshaw/Filtrol 28
5-1 Modified EPA Cadmium Sampling Train (August 18, 1977 Federal
Register) 30
5-2 Schematic of Sample Analyses 35
-v-
-------
LIST OF TABLES
TABLE PAGE
2-la Summary of Controlled Cadmium Emissions, Spray Tower 103,
Harshaw/Filtrol Partnership, July 1986 6
2-lb Summary of Controlled Cadmium Emissions, Spray Tower 103,
Harshaw/Filtrol Partnership, July 1986
2-2 Summary of Visible Emissions, Spray Tower 103, Hawshaw/
Filtrol Partnership, July 1986
2-3a Summary of Controlled Cadmium Emissions, Venturi Scrubber
201, Harshaw/Filtrol Partnership, July 1986 13
2-3b Summary of Controlled Cadmium Emissions, Venturi Scrubber
201, Harshaw/Filtrol Partnership, July 1986 14
2-4 Summary of Visible Emissions, Venturi Scrubber 201, Harshaw/
Filtrol Partnership, July 1986 ] 6
2-5 Trace Metal Analysis 20
-vi-
-------
PREFACE
The work described herein was conducted by personnel from TRC
Environmental Consultants, Inc., Midwest Research Institute (MRI), the United
States Environmental Protection Agency Emission Measurement Branch (EPA/EMB),
and Harshaw/Filtrol Partnership in Louisville, Kentucky.
The scope of work was issued under EPA Contract 68-02-4337, Work
Assignment 1. The work was performed under the supervision of John H. Powell,
TRC Work Assignment Manager, and Richard A. Pirolli, TRC Field Team Leader.
Mark Turner of MRI monitored process operations. MRI was responsible for
preparing Section 3 of this report, which deals with process descriptions and
operations. Sam Narvarte, Technical Superintendent of Harshaw/Filtrol,
provided invaluable assistance and guidance to TRC, EPA, and MRI in the
performance of the test program. Michael Toney, Office of Air Quality
Planning and Standards (OAQPS), Emission Measurement Branch, EPA, served as
Task Manager and was responsible for coordinating the test program.
-vi i-
-------
1.0 INTRODUCTION
1.1 Background
Section 112 of the Clean Air Act of 1977 charges the administrator of the
United States Environmental Protection Agency with the responsibility of
establishing National Emission Standards for Hazardous Air Pollutants (NESHAP)
from sources that may significantly contribute to air pollution. When
promulgated, these standards of performance are to reflect the degree of
emission limitation achievable through application of the best demonstrated
emission control technology. Emission data collected from controlled sources
in the cadmium industry may provide a portion of the data base used by EPA to
develop a NESHAP.
EPA Industrial Studies Branch (ISB) selected Hawshaw/Filtrol Partnership
in Louisville, Kentucky, as a site for an emission test program because there
is no data currently available to quantify cadmium emissions accurately from
belt dryers, rotary calciners or vacuum pan dryers used in pigment
manufacturing plants. The test program was designed to develop controlled
emission factors for cadmium sulfide pigment industries.
TRC Environmental Consultants, Inc. was retained by the EPA Emission
Measurement Branch (EMB) to perform emission measurements at the
Harshaw/Filtrol Partnership in Louisville, Kentucky. Testing was performed on
the red and yellow belt dryer venturi scrubber stacks (SD-301 & 201) and the
rotary calciners/vacuum pan dryer spray tower stack (S-103). This report has
been prepared in accordance with EPA Contract No. 68-02-4337 under the
provisions of Work Assignment No. 1.
Midwest Research Institute (MRI), the NESHAP contractor, was responsible
for coordinating the overall test program with Harshaw/Filtrol personnel and
for assuring that process and control equipment operating conditions were
suitable for testing. Related process data were monitored and recorded by MRI.
-1-
-------
1.2 Summary of Process
1.3 Applicability of EPA Reference Test Methods
EPA is required to publish a national reference test method for each
regulated source category and pollutant for which a NESHAP is established.
Reference test methods are usually specified by a State regulatory agency
during the State implementation planning process and may be different from
national reference test methods.
The purpose of establishing a national reference test method is to ensure
that emission data collected from a specific source is representative of that
source and comparable to data collected at other designated sources. The
primary purpose of this test program was to collect emission data using
standardized test methods that will allow the data to be evaluated to develop
a NESHAP. One test method was selected by EPA to measure emissions from
cadmium pigment manufacturers. This method is briefly described in the
following subsection and is described in detail in Section 5.
1.3.1 EPA Method 5 Configuration Modified for Cadmium
Cadmium sampling was performed by using a modification of the standard EPA
Method 5 train.' This modification consisted of placing flexible Teflon
tubing between the filter and the impingers and the use of 5% Nitric acid
-2-
-------
solution in the first two impingers, teflon tubing was used due to the
particularly difficult sampling locations.
Five percent Nitric acid is used as the impinger solution because metals
are readily soluble in Nitric acid and also Nitric acid increases capture
efficiency.
1.4 Measurement Program Summary
The measurement program was conducted at the Harshaw/Filtrol Partnership
in Louisville, Kentucky during the week of July 14, 1986. Tests were
performed at the red and yellow belt dryer venturi scrubber stacks and the
rotary calciners/vacuum pan dryer spray tower stack.
All emission testing was performed by TRC. MRI personnel monitored
process operating conditions. Michael Toney of EMB observed the test program.
1.4.1 Spray Tower and Venturi
Preliminary Measurements
Preliminary testing was performed on July 14, 1986 to determine volumetric
flowrate and stack gas moisture content. An integrated gas sample was also
taken to determine concentrations of COz, Oz, and CO, which were found to
be ambient. Stack diameter and the sampling port configuration were confirmed
at this time.
Method 5 Modified For Cadmium
Six Method 5 tests modified for cadmium were performed, two tests on the
yellow belt dryer venturi scrubber (SD-201), one test on the red belt dryer
venturi scrubber stack (SD-301) and three tests on the rotary calciner/vacuum
1Code of Federal Regulations 40, Part 60 Appendix A, July, 1981.
-3-
-------
pan dryer spray tower (S-103). Testing at these locations was performed on
July 15, 16 and 17, 1986.
Method 9 - Visible Emissions
Visible emissions from the spray tower stack and the venturi scrubber
stacks were observed concurrently with each Method 5 modified for cadmium test.
1.5 Report Sections
The remaining sections of this report present the Slummary and Discussion
of Results (Section 2), Process Description and Operations (Section 3),
Description of Sampling Locations (Section 4), Sampling and Analytical
Procedures (Section 5), and Quality Assurance (Section 6). Methods and
procedures, field and laboratory data, and calculations are presented in
various appendices as noted in the Table of Contents.
-4-
-------
2.0 SUMMARY AND DISCUSSION OF RESULTS
A summary of all emission measurements and collected data is presented in
this section. Section 2.1 presents cadmium and visible emission results
collected at spray tower 103 and a complete breakdown and discussion of
parameters. Cadmium, and visible emission results collected at venturi
scrubber 201 and 301 are presented in Section 2.2. Section 2.3 presents
results of the trace metals.
2.1 Spray Tower
A total of three tests were conducted at spray tower 103 using a Method 5
sampling train modified for cadmium. All tests were acceptable with a leak
rate of <0.02cfm and isokinetics of 100 ± 10%.
2.1.1 Cadmium Emissions
Tables 2-la (English Units) and 2-lb (Metric Units) present a summary of
measured cadmium emissions from spray tower 103.
The average total cadmium emission rate was 4.04 x 10"3 Ib/hr (5.09 x
10" 4 g/sec) and ranged from 2.75 x 10"3 Ib/hr (3.46 x 10"4 g/sec) to
4.97 x 10"3 Ib/hr (6.26 x 10~4 g/sec). The average total cadmium
concentration was 1.98 x 10~4 gr/DSCF and ranged from 1.07 x 10~4 gr/DSCF
to 2.44 x 10~4 gr/DSCF. All significant cadmium concentrations and emission
rates were collected in the front half of the sampling train.
2.1.2 Visible Emissions
A summary of visible emission observations from the spray tower is
presented in Table 2-2. Average opacities are presented for 6 minute time
periods during each two hour test. The average opacity was 1% for tests 1 and
-5-
-------
TABLE 2-la
(ENGLISH UNITS)
SUMMARY OF CONTROLLED CADMIUM EMISSIONS
SPRAY TOWER 103
HARSHAW/FILTROL PARTNERSHIP
JULY 1986
TEST NO
DATE
TIME
1
7/15/86
1056-1500
2
7/16/86
0832-1245
3
7/17/86
0846-1251
AVG
Sample Conditions
Volume (DSCF)1
Front Half Catch (pg)
Back Half Catch (>ig)
Total Catch (pg)
Isokinesis (%)
Stack Conditions
Flowrate (DSCFM)
Temperature (°F)
Moisture (%)
Oxygen (%)
Opacity (%)
Process Conditions
Production Rate (ton/hr)
Cadmium Emissions
Front
Half
Concentration
Mass
Back
Emission
Half
Concentration
Mass
Total
Emission
Concentration
Mass
Emission
(gr/DSCF)
Rate (Ib/hr)
(Ib/ton)
(gr/DSCF)
Rate (Ib/hr)
(Ib/ton)
(gr/DSCF)
Rate (Ib/hr)
(Ib/ton)
107.65
1680
13.6
1694
101.7
2120
101
11.4
20.9
1
0
2
4
2
1
3
2
2
4
2
.15
.41x10""
.38xlO"3
.92xlO"2
.95xlO"6
.54xlO"s
.36x10""
.43x10""
.41xlO~3
.94xlO"2
118.29
1860
7.2
1867
96.2
2380
99
3.2
20.9
1
0
2
4
3
9
1
1
2
4
3
.15
.43x10""
.95xlO~3
.30xlO~2
.39xlO"7
,92xlO"5
.28x10""
.44x10""
. 97x10" 3
.31xlO"2
150.66
1030
10.2
1040
97.0
3010
103
5.5
20.9
0
0
1
2
1
1
2
1
1
2
1
.15
.06x10""
.72xlO"3
.SlxlO"2
.OSxlO"6
.70xlO"s
.80x10""
.07x10""
.75xlO"3
.83xlO"2
2500
101
6.7
20.9
1
0.
1.
4.
2.
1.
2.
1.
1.
4.
2.
15
97x10"
02x10"
68x10"
31x10"
72x10"
81x10"
98x10"
04x10"
69x10"
4
3
2
b
S
2
4
3
2
Standard Conditions: 29.92 in. Hg @ 68°F
-6-
-------
TABLE 2-lb
(METRIC UNITS)
SUMMARY OF CONTROLLED CADMIUM EMISSIONS
SPRAY TOWER 103
HARSHAW/FILTROL PARTNERSHIP
JULY 1986
TEST NO 1
DATE 7/15/86
TIME 1056-1500
Sample Conditions
Volume (NM3)1
Front Half Catch (ug)
Back Half Catch (ug)
Total Catch (ug)
Isokinesis (%)
Stack Conditions
Flowrate (NM3/MIN)
Temperature (°C)
Moisture (%)
Oxygen (%)
Opacity (%)
Process Conditions
Production Rate (ton/hr)
Cadmium Emissions
Front Half
Concentration (Mg/NM3)
Mass Emission Rate (g/sec)
(g/ton)
Back Half
Concentration (Mg/NM3)
Mass Emission Rate (g/sec)
(g/ton)
Total
Concentration (Mg/NM3)
Mass Emission Rate (g/sec)
(g/ton)
3.05
1680
13.6
1694
101.7
60.04
38
11.4
20.9
1
0.15
551
S.SlxlO'4
13.22
4.46
4.46xlO~6
0.107
555
5.55xlO"4
13.32
2
7/16/86
0832-1245
3.35
1860
7.2
1867
96.2
67.40
37
3.2
20.9
1
0.15
555
6.24xlO'4
14.98
2.15
2.41xlO"6
0.058
557
6.26x10""
15.02
3
7/17/86
OEI46-1251
4U27
1030
10.2
1040
97.0
85.24
39
5.5
20.9
0
0.15
241
3.43xlO~4
8.23
2.39
3. 39x10" 6
0.081
244
3.46xlO"4
8.30
AVG
—
—
—
—
—
70.89
38
6.7
20.9
1
0.15
449
5.06xlO"4
12.14
3.00
3.42xlO~6
0.082
452
5.09x10""
12.21
Standard Conditions: 760 mm in. Hg @ 20°C
-7-
-------
TABLE 2-2
SUMMARY OF VISIBLE EMISSIONS
SPRAY TOWER 103
HARSHAW/FILTROL PARTNERSHIP
JULY 1986
TEST NO 1
7/15/86
1040-1413
SIX MINUTE
TIME PERIOD
1040-1046
1046-1052
1052-1058
1058-1104
1104-1110
1110-1116
1116-1122
1122-1128
1128-1134
1134-1140
STOP TEST
1313-1319
1319-1325
1325-1331
1331-1337
1337-1343
1343-1349
1349-1355
1355-1401
1401-1407
1407-1413
AVG
AVERAGE
OPACITY (%)
1
2
3
2
3
2
1
0
0
0
0
0
0
0
0
1
0
0
0
1
1
TEST NO 2
7/16/86
0824-1135
SIX MINUTE
TIME PERIOD
0824-0830
0830-0836
0836-0842
0842-0848
0848-0854
0854-0900
0900-0906
0906-0912
0912-0918
0918-0924
STOP TEST
1035-1041
1041-1047
1047-1053
1053-1059
1059-1105
1105-1111
1111-1117
1117-1123
1123-1129
1129-1135
AVG
AVERAGE
OPACITY (%)
1
1
0
0
0
0
0
0
1
1
1
0
2
0
0
1
0
1
1
0
1
TEST NO 3
7/17/86
0745-1114
SIX MINUTE
TIME PERIOD
0745-0751
0751-0757
0757-0803
0803-0809
0809-0815
0815-0821
0821-0827
0827-0833
0833-0839
0839-0845
STOP TEST
1014-1020
1020-1026
1026-1032
1032-1038
1038-1044
1044-1050
1050-1056
1056-1102
1102-1108
1180-1114
AVG
AVERAGE
OPACITY
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
-8-
-------
2 and 0% for test 3. These 6-minute average opa.ci.ties are presented
graphically in Figures 2-1 through 2-3.
2.1.3 Stack Conditions
An average volumetric flowrate of 2500 DSCFM was measured exiting the
spray tower stack at 101°F, 6.7% moisture and 20.9% oxygen. For the first
test, the volumetric flowrate was 2120 DSCFM at 101°F. 11.4 % moisture and
20.9% oxygen. This volumetric flowrate was 30% lower than test 3. This may
be accounted for by the high moisture seen in the first test. Test 3 had a
volumetric flowrate of 3010 DSCFM at 103°F, 5.5% moisture and 20.9% oxygen.
2.2 Venturi Scrubber
Two tests were performed at venturi scrubber stack 201 and one test on
venturi scrubber 301 using a Method 5 sampling train modified for cadmium.
All tests were acceptable with a leak rate of <0.02cfrn and 100± 10 percent
isokinetics.
2.2.1 Cadmium Emissions
Tables 2-3a (English Units) and 2-3b (Metric Units) present a summary of
measured cadmium emissions from the venturi scrubber. The average total
cadmium emission rate was 4.85 x 10"3 Ib/hr (6.10 x 10"4 g/sec) and ranged
from 4.45 x 10'3 Ib/hr (5.61 x 10'4 g/sec) to 5.16 x 10"3 Ib/hr (6.50 x
10~4 g/sec).
The average total cadmium concentration was 5.17 x 10"4 gr/DSCF and
ranged from 4.49 x 10~4 gr/DSCF to 6.18 x 10"4 gr/DSCF. All significant
cadmium concentrations and emission rates were collected in the front half of
the sampling train.
-9-
-------
25-
20-
TEST 1 7/15/86
15-
•a:
Q.
O
C3
•f.
Of
10-
5 -
STOP
TEST
P.
STOP
TEST
T
1040
1058
1116
—I—
1313
T
1134 1313 1331
6 MINUTE TIME RECORD
1349
1407
Figure 2-1. Summary of Visible Emissions
Spray Tower 103
Harshaw/Filtrol Partnership
-10-
-------
25 -
20 -
TEST 2 7/16/86
O-
o
ts
2 10
UJ
s»
«t
5 _
STOP
TEST
n
STOP
TEST
ru
0824 0842 0900 0918 1035 1053
6 MINUTE TIME RECORD
1111
1129
Figure 2-2. Summary of Visible Emissions
Spray Tower 103
Harshaw/Filtrol Partnership
-11-
-------
25-
20-
TEST 3 7/17/86
15-
Q.
O
10-
C£
UJ
-------
TABLE 2-3a
(ENGLISH UNITS)
SUMMARY OF CONTROLLED CADMIUM EMISSIONS
VENTURI SCRUBBER 201
HARSHAW/FILTROL PARTNERSHIP
JULY 1986
TEST NO
DATE
TIME
Sample Conditions
Volume (DSCF)Z
Front Half Catch (ug)
Back Half Catch (ug)
Total Catch (ug)
Isokinesis (%)
Stack Conditions
Flowrate (DSCFM)
Temperature (°F)
Moisture (%)
Oxygen (%)
Opacity (%)
Process Conditions
1
7/15/86
1058-1515
96.05
3010
11.0
3021
97.0
1070
130
6.6
20.9
0
2
7/16/86
0820-1227
85.72
3420
11.8
3432
99.8
930
138
9.4
20.9
0
31
7/17/86
0754-1200
120.03
3460
33.5
3494
96.8
1340
140
7.3
20.9
4
AVG
—
—
—
—
—
1110
136
7.8
20.9
1
Production Rate (ton/hr) 0.063
Cadmium Emissions
0.063
0.063
Test No. 3 on Venturi Scrubber 301
Standard Conditions: 29.92 in Hg @ 68°F
0.063
Front Half
Concentration (gr/DSCF)
Mass Emission Rate (Ib/hr)
(Ib/ton)
Back Half
Concentration (gr/DSCF)
Mass Emission Rate (Ib/hr)
(Ib/ton)
Total
Concentration (gr/DSCF)
Mass Emission Rate (Ib/hr)
(Ib/ton)
4
4
7
1
1
2
4
4
7
.84x10"
.44x10"
.05x10"
.77x10"
.62x10"
.57x10"
.85x10"
.45x10"
.06x10"
4
3
2
E
5
4
4
3
2
6
4
7
2
1
2
6
4
7
.16x10"
.91x10"
.79x10"
.13x10"
.69x10"
.68x10"
.18x10"
.93x10"
.83x10"
4
3
2
6
S
4
4
3
2
4
5
8
4
4
7
4
5
8
.45x10"
.11x10"
.11x10"
.31x10"
.95x10"
.86x10"
.49x10"
.16x10"
.19x10"
4
3
2
6
S
4
4
3
2
5.
4.
7.
2.
2.
4.
5.
4.
7.
15xlO"4
82xlO"3
65xlO"2
74xlO"6
75xlO"s
37xlO"4
17xlO"4
85xlO"3
69xlO"2
-13-
-------
TABLE 2-3b
(METRIC UNITS)
SUMMARY OF CONTROLLED CADMIUM EMISSIONS
VENTURI SCRUBBER 201
HARSHAW/FILTROL PARTNERSHIP
JULY 1986
TEST NO 1
DATE 7/15/86
TIME . 1058-1515
Sample Conditions
Volume (MM3)2
Front Half Catch (ug)
Back Half Catch (ug)
Total Catch (ug)
Isokinesis (%)
Stack Conditions
Flowrate (NM3/MIN))
Temperature (°C)
Moisture (%)
Oxygen (%)
Opacity (%)
Process Conditions
Production Rate (ton/hr)
Cadmium Emissions
Front Half
Concentration (ug/NM3)
Mass Emission Rate (g/sec)
(g/ton)
Back Half
Concentration (ug/NM3)
Mass Emission Rate (g/sec)
(g/ton)
Total
Concentration (ug/NM3)
Mass Emission Rate (g/sec)
(g/ton)
2.72
3010
11.0
3021
97.0
30.30
54
6.6
20.9
0
0.063
1107
5.59x10""
31.94
4.04
2. 04x10" 6
0.117
1111
5.61x10""
32.06
2
7/16/86
0820-1227
2.43
3420
11.8
3432
99.8
26.34
59
9.4
20.9
0
0.063
1407
6.18x10""
35.31
4.86
2.13xlO'6
0.122
1412
6.20x10""
35.43
31
7/17/86
0754-1200
3.40
3460
33.5
3494
9S.8
37.95
60
7.3
20.9
4
0.063
1018
6.44x10""
36.80
9.85
6.23xlO"s
0.356
1028
6.50x10'"
37.14
AVG
—
—
—
—
—
31.53
58
7.8
20.9
1
0.063
1177
6.07x10""
34.68
6.25
3.47xlO"6
0.198
1184
6.10x10'"
34.88
Test No. 3 on Venturi Scrubber 301
Standard Conditions: 760 mm Hg @ 20°C
-14-
-------
2.2.2 Visible Emissions
A summary of visible emission observations from the venturi scrubber
stacks is presented in Table 2-4. Average opacities are presented for
6-minute time periods during each two hour test. The average opacity was 0%
for tests 1 and 2 and 4% for test 3. Opacity observations were conducted on
venturi scrubber 301 for test 3. The opacity was high (12% to 0%) for the
first hour of observation. During the second hour of observation, the average
opacity was 0%. These 6-minute average opacities are presented graphically in
Figures 2-4 through 2-6.
2.2.3 Stack Conditions
The volumetric flowrate exiting venturi scrubber stack 201 was 10700 DSCFM
at 130°F, 6.6% moisture and 20.9% oxygen for Test 1 and 930 DSC '4 at 138°F,
9.4% moisture and 20.9% oxygen for Test 2. Test 3 was conducted at venturi
scrubber stack 301 and the volumetric flowrate was 1340 DSCFM at 140°F, 7.3%
moisture and 20.9% oxygen.
2.3 Trace Metals
One sample and blank per set of tests were analyzed for trace metals. The
trace metals analyzed for were barium, chromium, copper, iron, lead,
magnesium, manganese, nickel and zinc. Mercury was not analyzed due to the
different digestion procedure. Table 2-5 presents a summary of trace metals.
Sample train results were corrected for field and/or laboratory blanks. Trace
metals in the field blanks were high. This may be caused from the nitric acid
rinse of the nozzles. The sample train results are representative due to the
significant difference between blanks and runs.
-15-
-------
TABLE 2-4
SUMMARY OF VISIBLE EMISSIONS
VENTURI SCRUBBER 201
HARSHAW/FILTROL PARTNERSHIP
JULY 1986
TEST NO
7/15/86
1
1157-1513
SIX MINUTE
TIME PERIOD
1157-1203
1203-1209
1209-1215
1215-1221
1221-1227
1227-1233
1233-1239
1239-1245
1245-1251
1251-1257
STOP TEST
1413-1419
1419-1425
1425-1431
1431-1437
1437-1443
1443-1449
1449-1455
1455-1501
1501-1507
1507-1513
AVG
AVERAGE
OPACITY (%)
0
1
0
0
0
1
1
0
0
0
0
0
0
0
0
0
0
0
1
0
0
TEST NO 2
7/16/86
0930-1244
SIX MINUTE
TIME PERIOD
0930-0936
0936-0942
0942-0948
0948-0954
0954-1000
1000-1006
1006-1012
1012-1018
1018-1024
1024-1030
STOP TEST
1144-1150
1150-1156
1156-1202
1202-1208
1208-1214
1214-1220
1220-1226
1226-1232
1232-1238
1238-1244
AVG
TEST NO
7/17/86
3*
0851-1217
AVERAGE
OPACITY (%)
0
0
0
0
0
1
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
SIX MINUTE
TIME PERIOD
0851-0857
0857-0903
0903-0909
0909-0915
0915-0921
0921-0927
0927-0933
0933-0939
0939-0945
0945-0951
STOP TEST
1117-1123
1123-1129
1129-1135
1135-1141
1141-1147
1147-1153
1153-1159
1159-1205
1205-1211
1211-1217
AVG
AVERAGE
OPACITY
11
12
7
11
10
9
9
7
6
3
0
0
0
0
0
0
0
0
0
0
4
* Opacity Observations at scrubber 301
-16-
-------
25-
20-
TEST 1 7/15/86
15-
Q.
O
CJ
-------
25-
20-
TEST 2 7/16/86
15-
•f.
o.
o
o
-------
25-
20-
15-
-------
TABLE 2-5
SUMMARY OF TRACE METALS
HARSHAW/FILTROL PARTNERSHIP
JULY 1986
Sample1
Location
Barium (ug)
Chromium (ug)
Copper (ug)
Iron (ug)
Lead (ug)
M Magnesium (ug)
o
Manganese (ug)
Nickel (ug)
Zinc (ug)
Blank 1
Venturi
201
nd2<50
253
9
1010
nd<46
115
22
94
90
Run 1
Venturi
201
nd<50
1280
39
6800
nd<46
153
183
676
6320
Blank 1 Run 1
Spray Tower Spray Tower
103 103
nd<50
53
15
282
' nd<46
104
7
44
110
nd<50
100
37
892
nd<46
174
13
126
4050
HN03
Blank
nd<25
nd<5
nd<3
8
nd<23
1.4
nd<4
nd<7
nd
-------
3.0 PROCESS AND OPERATIONS (Provided by MRI)
3.1 Process Equipment
3.2 Control Equipment
-21-
-------
4.0 SCOPE OF SAMPLING PROGRAM
Sampling train operators maintained a detailed log of their test times for
each run. Data sheets present all necessary information in concise legible
form.
Method 5 test data such as: percent isokinetics, moisture, flowrate
(DSCFM) and stack temperatures was calculated daily. Results were recorded on
a daily summary sheet.
4.1 Belt Dryer
Prior to emission testing, the flowrate of the stack was measured in
accordance with EPA Methods 1-2. Duct measurements were verified at this
time. Stack gas moisture content was determined using wet bulb/dry bulb
thermometers.
Cadmium samples were collected from a 10 inch circular duct. Two 3-inch
sample ports were placed 90° apart in the duct. Figure 4-1 presents the
sample port configuration. Sample point A indicates the Method 5 sampling
location. According to EPA Method 1, 24 traverse points are required. Each
traverse point was sampled for ten minutes for a total test time of 4 hours.
An "EPA Method 1, Sampling and Velocity Traverse" sheet is presented in Figure
4-2.
4.2 Rotary Calciner and Vacuum Pan Dryer
Prior to emission testing, routine preliminary measurements were conducted
as described in Section 4.1.
Cadmium samples were collected in two ports positioned 90° apart and
located 26 inches (2 diameters) downstream from the fan and 14 inches (1
diameter) upstream from the top of the stack. The location of the sampling
ports are presented in Figure 4-3.
-22-
-------
SAMPLE POINT B
SAMPLE POINT A-
10"
T
19"
I
20"
i
NOT TO SCALE
FAN
Figure 4-1. Belt Dryer Venturi Scrubber Outle"
-23-
-------
Environmental
Consultants, Inc.
EPA Method 1 Sample and Velocity Traverse
Figure 4-2 for Stationary Sources
"
-------
SAMPLE POINT B
SAMPLE POINT A
NOT TO SCALE
FAN
T
14"
26
Figure 4-3. Calciner and Pan Dryer Spray Tower Outlet
-25-
-------
According to EPA Method 1, 24 traverse points were sampled at this
location for ten minutes per point for a total test time of 4 hours. Sample
port and traverse point locations are presented in Figure 4-4.
4.3 Visible Emission Observations (Method 9)
Visible emission observations were conducted in accordance with EPA
Method 9. Observations were performed at the venturi scrubber stack and the
spray tower stack separately. Each location was observed for two hours during
each cadmium emission test. A diagram showing the observers location can be
found in Figure 4-5.
-26-
-------
Environmental
Consultants, Inc.
EPA Method 1 Sample and Velocity Traverse
Figure 4-4 for Stationary Sources
-r«Hnn Ventrui Scrubber Outlet Number of Ports
Diameters Upstream—J Points Per Port
Total Traverse Points Required.
24
12
Diameters Downstream,
Traverse ( Horizontal or Vertical).
rMINIMUM NUMBER OF TRAVERSE POINTS FOR PARTICULATE
AND NONPART1CULATE TRAVERSES
Duct Diameters Upstream from Flow Dlsturoance
(Distance A)
0.5 1.0 - 1.5 2.0
50
40
30
"
1°
i i i i i i i
H'OHift Nuutift it ton
HIC7AMOUUAH VTACU OH OUCTI
* 0
.*- PARTICIPATE
24 or 28 f
L"
T
A
• •
\
rf| AXUff |wl VT
OinUMANCI
^ 1
*•
-
1Q I 16 TACK OIAMITIR> 0.11 «C4UJ
' 4 II 12
^NONPARTICULATE
8 or 9
tTACK DIAMITtA -CJO TO O.n m (1J-J4 UJ
1 1 1 1 I 1 1
23 4 5 6 7 8 9 10
Duct Diameters Downstream from Flow Disturbance
(Distance 6)
vLOCATION OF TRAVERSE POINTS IN CIRCULAR STACKS
Point
Number
On A
Diameter '
. 1
2
3
4
5
6
7
8
9.
10
11
12 ;
(Percent of stack diameter from
Inside wall to traverse point)
Number Of Traverse Points On A Diameter
46 8 10 12
6
25
75
93
.7
.0
.0
.3
4.
14.
29.
70.
85.
95.
4
6
6
4
4
6
3.
10.
19.
32.
to.
80.
89.
96.
2
5
4'
3
7
6
5
8
2
8
14
22
34
65
77
85
91
97
•6
.2
.6
.6
.2
.8
.4
.4
.8
.4
2
6
11
17
25
35
64
75
82
88
93
97
.1
.7
.8
.7
.0
.6
.4
.0
.3
.2
.3
.9
Sample
Port "
9"
20"
i
2LW
eq
1
CROSS-SECTIONAL LAYOUT
FOR RECTANGULAR STACKS
Total
Traverse Points Matrix
9 3x3-
12 4x3
16 .4x4
20 5x4 '
25 5x5
'TRAVERSE POINT LOCATIONS
No.
1
2
3
4
5
6
7
8
9
10
11
12
Distance
From Wall
0.2"
0.7"
1.2"
1.8" i
2.5" ...
3.6"
6.4"
7.5" •
8.2"
8.«"
9.3"
9 . 8"
Nipple
Size
•
.
• Total
Distance
-------
301 201 103
PIGMENT BUILDING
NJ
00
I
TEST 2
SUN
TEST 1 and 3
Figure 4-5. Observers Location
Visible Emissions
Spray Tower and Venturi Scrubber
Harshaw/Filtrol Partnership
July 1986
-------
5.0 SAMPLING AND ANALYTICAL PROCEDURES
This section presents descriptions of sampling and analyses procedures
which were employed during the emission testing conducted at Harshaw/Filtrol
Partnership facility in Louisville, Kentucky.
A combination of EPA Methods 1, 2, 3, 4, 5 (modified for cadmium) and 9
were used to measure cadmium, visible and trace metals emissions from the
venturi scrubber stack and the spray tower stack.
5.1 EPA Method 5 Configuration Modified for Cadmium
Sampling was performed using a modification to the standard EPA Method 5
sampling train1. It is shown schematically in Figure 5-1 and consists of a
nozzle, probe, filter with glass frit, a flexible Teflon umbilical line, four
impingers, vacuum pump, dry gas meter, and an orifice flow meter. The
modification of the standard EPA Method 5 train consisted of placing flexible
Teflon tubing between the filter and the impingers. This modification made
the sampling equipment much easier to handle. A second modification was the
use of 5% nitric acid in first two impingers. The sampling train was
calibrated before and after this test program at TRC. This method is based on
the proposed methodology presented in the Appendix.
A nozzle was attached to a stainless steel glass-lined probe which was
heated to prevent condensation. Whatman EPM-1000 fiberglass filter paper
supported in a 4-1/2 inch glass filter holder with a glass frit was used as a
particulate collection medium. Filters were desiccated and pre weighed. A
visible inspection for irregularities in the filter material was conducted.
The filter assembly was enclosed in a heated box to keep the filter
temperature at approximately 150° F (±10). A thermocouple, located inside
Code of Federal Regulations 40, Part 60 Appendix A, July 81
-29-
-------
i
to
O
I
\c
16
I
©
14
16
A^__, ! J
STACK. WALL-
13
LEGEND
1 - NOZZLE
2 - PROBE
3 - FILTER HOLDER
4 - HEATED FILTER BOX
5 - IMPINGER ICE BATH
6 - UMBILICAL CORD
7 - VACUUM GAUGE
8 - MAIN VALVE TO PUMP
9 - PUMP
10 - BYPASS VALVE
11 - DRY GAS METER
12 - ORIFICE AND MANOMETER
13 - PITOT TUBE AND MANOMETER
14 - THERMOCOUPLE READOUT
15 - FLEXIBLE TEFLON SAMPLE LINE
16 - THERMOCOUPLES
17 - IMPINGER OUTLET TEMPERATURE
Figure 5-1. Modified EPA Method 5 cadmium sampling train
(August 18, 1977 Federal Register)
-------
the back half of the filter holder, monitored the gas stream temperature to
ensure proper filter temperature.
Four impingers immersed in an ice bath were attached to the back end of
the filter holder with a flexible Teflon tube. The first two impingers each
contained 100 ml of 5% nitric acid, the third was empty and the fourth
contained 200 grams of silica gel to remove any remaining moisture. Impinger
outlet temperatures were kept between 50°F and 80°F.
Flexible tubing, vacuum gauge, needle valve, leakless vacuum pump, bypass
valve, dry gas meter, calibrated orifice and inclined manometer completed the
sampling train. A check valve was not used in the TRC sampling train.
A nomograph was used to quickly determine the orifice pressure drop
required for any pitot velocity pressure and stack temperature in order to
maintain isokinetic sampling conditions. Sampling flow was adjusted by means
of the bypass valve. Before and after each particulatei test run as well as
port changes, the sampling train was leak checked. Sample time was 4 hours
per run.
Test data recorded at ten minute intervals for each sampling point
included; test time, sampling duration at each traverse point, pitot pressure,
stack temperature, dry gas meter volume and inlet-outlet temperatures, probe
temperature, and orifice pressure drop.
5.2 Sample Recovery
Sample recovery was performed in a clean, wind-free area to avoid sample
contamination. The TRC mobile environmental laboratory was adjacent to the
test site and utilized for sample recovery. Working surfaces were covered
with fresh Kimwipes prior to each sample recovery to mitigate contamination.
Personnel wore disposable gloves when recovering samples. Brushes and
tweezers were kept in plastic bags when not in use. Each sample probe had its
-31-
-------
own brush to lessen cross contamination. All sampling equipment had no
exposed openings before and after sampling* The following sample fractions
were recovered:
Container No. 1 Glass fiber filter.
Container No. 2 Probe, nozzle, and front half of the filter
housing HMOs rinses.
Container No. 3 Impinger solution. Teflon line, impingers and
back-half of the filter housing HMOs rinses.
Container No. 4 Silica gel.
The probe, nozzle and front half of the filter were rinsed and brushed
three times with 5% HN03. A 200 ml flask with 28/12 socket joint was
attached to the ball end of the probe. Approximately 50 ml of HNOs were
rinsed through the probe and collected in the flask. A blunt end brush
constructed of nylon bristles and a stainless steel shaft was then passed 3
times each with a fresh immersion of 5% HN03 solution held in the flask.
The brush was then removed and rinsed into the sample jar. Recovery personnel
sealed the opposite end of the probe with a gloved finger and inverted twice
allowing the HNOs to pass along the inside of the probe. A final flush of
fresh 5% HNOs through the probe was made into the sample container rotating
the probe so that HN03 made contact with all parts of the glass liner. This
sample was labeled and weighed for shipping.
The contents of the first three impingers were weighed, recorded, and
deposited in Container No. 3. The back half of the filter housing, the
flexible Teflon line, and the first three impingers and their connecting
glassware were rinsed 3 times with 5% HNOs solution and combined with the
impinger contents. A 200 ml flask with 28/12 socket joint was attached to one
end of the flexible Teflon line. Approximately 50 ml of 5% HNOs was passed
-32-
-------
through the line and collected three times with fresh 5% HNOs. The final
volume was weighed and recorded. Silica gel moisture gain was weighed on a
triple beam balance and recorded.
The glass fiber filter was removed, placed into a contamination free petri
dish, sealed with parafilm and labeled.
The sample container was labeled with the following information:
1. Sample I.D. (determined by location, run number and content)
2. Project Name
3. Date of Sampling
4. Shipping weight
5. Comments
Sample recovery data sheets were used to record: volumes recovered, filter
ID numbers, silica gel weights and visual descriptions of the samples.
Completed forms can be found in the Appendix.
A sample log sheet was completed recording: date, run number, sample ID,
sample description and remarks. Completed copies of this; form is presented in
the Appendix.
5.3 Sample Analyses
Sample analyses was performed in TRC Environmental Laboratories in East
Hartford, CT. The sample fractions were analyzed as follows:
Container No. 1 The filter was desiccated and weighed for particulate
measurement. The filter and probe wash was digested,
extracted and analyzed for cadmium by flame atomic
absorption. One set of samples was analyzed for
trace metals by flame atomic absorption.
Container No. 2 Nozzle and probe rinse were dried and weighed; then
brought back to solution with HNOs. The solution
was extracted and analyzed with Container No. 1.
Container No. 3 The solution was analyzed separately as Container
No. 2.
-33-
-------
A flow diagram of the sample analysis procedure is presented in Figure 5-2.
Trace metals standards purchased from J.T. Baker Standards, which are MBS
traceable, was used for all trace metals analyses. Samples were concentrated
or diluted in order to bring the analysis range into the center of the
linearity curve. All samples were analyzed and three readings taken as a
measure of precision. A complete description of the analytical procedure can
be found in the Appendix.
The trace metals to be analyzed for were:
Barium
Chromium
Copper
Iron
Lead
Magnesium
Manganese
Nickel
Zinc
5.4 Visible Emissions Observation
A certified visible emissions observer read exhaust opacity at the outlets
of the venturi scrubber arid spray tower. Sample Point B in Figures 4-1 and
4-2 depicts the exhaust points.
The observer's location was determined on-site in accordance with EPA
Method 9 and considering obstructions and traffic problems in the area.
The Method 9 data reduction was calculated by averaging 24 consecutive
observations recorded at 15-second intervals to calculate an average for
observation for a 6 minute time period.
-34-
-------
METHOD 5
CADMIUM TRAIN
FRONT HALF
(PROBE RINSE & FILTER)
PROBE AND FILTER
HOLDER RINSE
FILTER
EVAPORATE HN03
WEIGH
DESICCATE
WEIGH
RESULTS (mg) | | RESULTS (mg)
DISSOLVE RESIDUE
WITH HN03
PLACE FILTER
WITH PROBE
RESIDUE
ADD HN03 AND HEAT FOR 10 MINUTES |
ADD CONCENTRATED HNOs
AND REFLUX FOR 30 MINUTES
COOL SOLUTION
| ADD HC1 AND HEAT FOR 10 MINUTES |
COOL SOLUTION
DILUTE TO KNOWN VOLUME
ADJUST SAMPLE FOR LINEARITY CURVE
ANALYZE FOR Cd BY FLAME AA
I
Cd
BACK HALF
(IMPINGER SOLUTION)
ADD CONCENTRATED
HN03 AND
REFLUX FOR 30 MIN.
COOL SOLUTION
ALIQUOT
(1 SAMPLE ONLY)
ADD HC1 AND HEAT
FOR 10 MINUTES
DILUTE TO
KNOWN VOLUME
COOL SOLUTION
ADJUST SAMPLE FOR
LINEARITY CURVE
DILUTE TO
KNOWN VOLUME
ANALYZE FOR TRACE
METALS BY FLAME
OR GRAPHITE AA
ADJUST SAMPLE FOR
LINEARITY CURVE
ANALYZE FOR Cd
FLAME AA
RESULTS
-------
6.0 QUALITY ASSURANCE
TRC's quality assurance program for source emission measurement is
designed so that the work is done by competent, experienced individuals using
properly calibrated equipment and approved procedures for sample collection,
recovery and analysis with proper documentation.
Specific details of TRC's quality assurance program for stationary air
pollution sources may be found in the Quality Assurance Handbook for Air
Pollution Measurement Systems, Volume III (EPA-600/4-7-027b).
At the beginning of each day, an organizational meeting was held to orient
personnel to the day's activities, to discuss results from the previous day,
and to determine if any special considerations are appropriate for the day's
work.
Method 5
TRC's measurement devices, pitot tubes, dry gas meters, thermocouples,
probes and nozzles are uniquely identified and calibrated with documented
procedures and acceptance criteria before and after each field effort.
Records of all calibration data are maintained in TRC files. Samples of these
Calibration forms for equipment used during the test program are presented in
the Appendix.
All Method 5 sampling was 100 ± 10 percent isokinetic. Filter outlet
temperatures were maintained at 150±10°F.
Clean-up evaluations were performed on each initial set of glassware prior
to collecting field samples. The evaluation tests were performed at
Harshaw/Filtrol Partnership facility.
Each Method 5 sampling train had a clean-up sequence as follows:
-36-
-------
Blank 1
Method 5 Run 1
Method 5 Run 2
Method 5 Run 3
Blank 2
All blank sample recovery was conducted identically to sample recovery
procedures outlined in Section 5.
Blanks of all solution used during the test program (HNOs) were
collected and analyzed.
In summary, the evaluation tests were designed to precondition the sample
collectors, to establish blank background values, and to educate the clean-up
personnel in specific sample recovery procedures.
All sample recovery were performed by a one person clean-up crew.
Appropriate sample recovery data were recorded on the sample identification
log, sample recovery, chain of custody form, and analytical data forms as
presented in the Appendix.
Recovered samples were secured in padlocked, shock-proof, steel containers
for storage and shipment for analysis.
All preparation and analysis of Method 5 samples were performed by TRC.
TRC adhered to the standards of quality assurance as set forth in Quality
Assurance Handbook for Air Pollution Measurement Systems, Volume III
(EPA-600/4-7-027b) and the Handbook for Analytical Quality Control in Water
and Wastewater Laboratories. (EPA-600/4-79-019, March 1979).
Analytical Quality Control
The analytical quality control procedures consisted of analyzing duplicate
spikes, and blanks on 10% of the samples as described in TRC Technical
Standard T/S-902 section E & F. Balances were calibrated with four class S
rates daily. Since stack samples (filters, probe wash and impingers) cannot
-37-
-------
be split before the metals digestion, duplicates were not done. Each type of
sample was analyzed for cadmium by the method of standard addition to
determine if there are any matrix interferences. If this result differed by
more than 5% from the routine analysis, all the samples were analyzed by
standard addition. The atomic absorption spectrophotometer was calibrated
with four standards. Three measurements were taken on «sach sample for flame
analysis and two for furnace analyses. Refer to Appendix A for sample
calculations.
The laboratory participates in the following inter-laboratory quality
control programs:
• State of Connecticut, Department of Health Services, Laboratory
Division, Laboratory Improvement Program (annually).
• EPA, Environmental Monitoring and Support Laboratory, Quality
Assurance Branch, Water Pollution Laboratory Performance
Evaluation and Water Supply Performance Evaluation (voluntary).
• NIOSH, Proficiency Analytical Testing Program (quarterly).
• EPA Environmental Monitoring Systems Laboratory Quality Assurance
Division. Stationary Source and Ambient Air, Inter-Laboratory
Studies (semi-annually).
Method 9
The TRC emissions observer was certified within the past 6 months to
perform visible emission evaluations. Documentation verifying the observers
certification is provided in Appendix D.
-38-
------- |