United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EMB Report 83-CDR-12
October 1983
Air
Calciners and
Dryers

Emission Test
Report
Burgess Pigment
Company
Sandersville,
Georgia

-------
          STANDARDS DEVELOPMENT FOR
             CALCINERS AND DRYERS
    PAR.TICULATE AND PARTICLE SIZE TESTING
           BURGESS PIGMENT COMPANY
           Contract No. 68-02-3541
            Work Assignment No. 8
              EMB No. 83-CDR-12
                 Submitted to

     UiS. Environmental Protection Agency
         Emission Measurement Branch
                 Mail Drop 19
Research Triangle Park, North Carolina  27711
                February 1984
                9142.00/25A-N
                 Submitted by

             Engineering-Science
            10521 Rosehaven Street
           Fairfax, Virginia  22030

-------
                            TABLE OF CONTENTS
CHAPTER 1    INTRODUCTION

CHAPTER 2    SUMMARY OF TEST PROGRAM AND RESULTS

CHAPTER 3    SAMPLE SITES AND METHOD REQUIREMENTS

CHAPTER 4    PROCESS DESCRIPTION

               Introduction
               Pollutants/Sampling Points
               Process Description
               Process Conditions During Testing

CHAPTER 5    SAMPLING AND ANALYTICAL PROCEDURES
             DATA REDUCTION PROCEDURES

               Method 5
               Particle Sizing
               Ancillary Tests
               Data Reduction and Calibration

CHAPTER 6    QUALITY ASSURANCE
1-1

2-1

3-1

4-1

4-1
4-1
4-2
4-8


5-1

5-1
5-4
5-8
5-9

6-1
APPENDIX A   COMPUTER PRINTOUT  SUMMARY SHEETS,  METHOD 5 SAMPLING

APPENDIX B   COMPUTER PRINTOUT  OF PARTICLE  SIZE DISTRIBUTION AND
             PADRE DATA ENTRY PRINTOUTS

APPENDIX C   FIELD DATA SHEETS  FOR METHOD 5 SAMPLING

APPENDIX D   FIELD DATA SHEETS  FOR PARTICLE SIZE SAMPLING

APPENDIX E   FIELD DATA SHEETS  FOR FUGITIVE AND VISIBLE EMISSIONS
             OBSERVATIONS

APPENDIX F   LABORATORY DATA SHEETS

APPENDIX G   EQUIPMENT CALIBRATION DATA

APPENDIX H   PROCESS SAMPLE ANALYSIS
                                    11

-------
                             LIST OF FIGURES
2.1           Burgess Pigment Company, Particle Size Testing          2-16
               Test Site:  HFBI
2.2          Burgess Pigment Company, Particle Size Testing          2-17
               Test Site:  HFBO
2.3          Burgess Pigment Company, Particle Size Testing          2-18
               Test Site:  HFSI
2.4          Burgess Pigment Company, Particle Size Testing          2-19
               Test Site:  FCBI
2.5          Burgess Pigment Company, Particle Size Testing          2-20
               Test Site:  FCBO

3.1 a         Herreschoff Furnace Scrubber Inlet                      3-3
3.1b         Sampling Point Location, Herreschoff Furnace,
               Scrubber Inlet                                        3-4
3.2a         Herreschoff Furnace Scrubber Outlet                     3-6
3.2b         Sampling Point Location, Herreschoff Furnace,
               Scrubber Outlet                                       3-7
3.3a         Herreschoff Furnace Baghouse Inlet                      3-8
3.3b         Sampling Point Location, Herreschoff Furnace,
               Baghouse Inlet                                        3-9
3.4a         Herreschoff Furnace Baghouse Outlet                     3-10
3.4b         Sampling Point Location, Herreschoff Furnace,
               Baghouse Outlet                                       3-1 1
3.5a         Flash Calciner Baghouse Inlet                           3-13
3.5b         Sampling Point Location, Flash Calciner, Baghouse
               Inlet                                                 3-14
3.6a         Flash Calciner Baghouse Outlet                          3-15
3.6b         Sampling Point Location, Flash Calciner, Baghouse
               Outlet                                                3-16

4.1           Simplified Process Flow Diagram for Burgess Pigment
               Company, Sandersville, Georgia                        4-4
4.2          Flow Diagram for Emission Control Systems for
               Calciners at Burgess Pigment Company, Sanders-
               ville, Georgia                                        4-5

5.1           Particulate Sampling Train                              5-2
5.2          Particle Size Sampling Train                            5-6
                                    111

-------
                              LIST OF TABLES
2.1           Burgess Pigment Company, Summary of Method 5 Test
               Results,  Herreschoff Furnace - Baghouse Inlet         2-3
2.2           Burgess Pigment Company, Summary of Method 5 Test
               Results,  Herreschoff Furnace - Baghouse Outlet        2-4
2.3           Burgess Pigment Company, Summary of Method 5 Test
               Results,  Herreschoff Furnace - Scrubber Inlet         2-5
2.4           Burgess Pigment Company, Summary of Method 5 Test
               Results,  Herreschoff Furnace - Scrubber Outlet        2-6
2.5           Burgess Pigment Company, Summary of Method 5 Test
               Results,  Flash Calciner - Baghouse Inlet              2-7
2.6           Burgess Pigment Company, Summary of Method 5 Test
               Results,  Flash Calciner - Baghouse Outlet             2-8
2.7           Burgess Pigment Company, Summary of Particle Sizing
               Test Results, Herreschoff Furnace - Baghouse Inlet    2-9
2.8           Burgess Pigment Company, Summary of Particle Sizing
               Test Results, Herreschoff Furnace - Baghouse Outlet   2-10
2.9           Burgess Pigment Company, Summary of Particle Sizing
               Test Results, Herreschoff Furnace - Scrubber Inlet    2-11
2.10         Burgess Pigment Company, Summary of Particle Sizing
               Test Results, Flash Calciner - Baghouse Inlet         2-12
2.11         Burgess Pigment Company, Summary of Particle Sizing
               Test Results, Flash Calciner - Baghouse Outlet        2-13
2.12         Burgess Pigment Company, Herreschoff Furnace
               Baghouse  Performance                                  2-21
2.13         Burgess Pigment Company, Herreschoff Furnace
               Scrubber Performance                                  2-22
2.14         Burgess Pigment Company, Flash Calciner Baghouse
               Performance                                           2-23
2.15         Burgess Pigment Company, Summary of Visible Emis-
               sions, Herreschoff Furnace - Baghouse Outlet          2-25
2.16         Burgess Pigment Company, Summary of Fugitive Emis-
               sions, Herreschoff Furnace - Process Outlet           2-26
2.17         Burgess Pigment Company, Summary of Visible Emis-
               sions, Herreschoff Furnace - Scrubber Outlet          2-27
2.18         Burgess Pigment Company, Summary of Visible Emis-
               sions, Flash Calciner - Baghouse Outlet               2-28
2.19         Burgess Pigment Company, Summary of Fugitive Emis-
               sions, Flash Calciner - Process Outlet                2-29

3.1           Stack Gas Composition and Conditions                    3-2
                                    IV

-------
4.1

4.2
4.3
4.7
4.8
4.9
4.13
4.14
4.15
Emission Tests Conducted at Burgess
  Company
Summary of Calciner Data
Summary of Control Equipment Data
Herreschoff Control Equipment Data,
Herreschoff Control Equipment Data,
Herreschoff Control Equipment Data,
Plash Calciner Baghouse Data, Run 1
Flash Calciner Baghouse Data, Run 2
Flash Calciner Baghouse Data, Run 3
Pigment
Run 1  (9/26/83)
Run 2 (9/27/83)
Run 3 (9/27/83)
(9/28/83)
(9/29/83)
(9/29/83)
4-3
4-6
4-7
4-9
4-10
4-11
4-13
4-14
4-15

-------
                                CHAPTER 1

                               INTRODUCTION
     The U.S. Environmental Protection Agency (EPA) is in the process of
developing New Source Performance Standards (NSPS) for the Calciners and
Dryers industry.  The focus of these standards will be in the control of
particulate matter, with potential emphasis on controlling particles less
then 10 microns, referred to as PM10.

     Through Contract 68-02-3541, Work Assignment 8, Technical Directive
No. 14, Engineering-Science (ES) was instructed to conduct particulate
and particle size tests at the Burgess Pigment Company facility located
in Sandersville, Georgia.  During the week of September 26, 1983, testing
was conducted on the baghouse and scrubber inlets and outlets of the No.
2 Herreschoff Furnace, and the baghouse inlet and outlet of the No. 4
Flash Calciner.

     The results of the sampling program are presented in Chapter 2 with
a summary discussing the results.  Chapter 3 describes the testing loca-
tions in detail for each of the processes.  Chapter 4 describes the manu-
facturing processes tested and their associated air pollution control
equipment.  Process data collected by Midwest Research Institute (MRI)
are also presented.  Chapter 5 presents sampling and analytical proce-
dures employed.

     The 13-man ES test crew was headed by W. Keith Felts, Jr.  Mr. Den-
nis Holzschuh was present for the U.S. EPA Emissions Measurement Branch
(EMB); Mr. Marc Sack was present for Burgess Pigment.  Mr. William Neuffer
represented the Industrials Studies Branch (ISB) of the U.S. EPA and Mr.
Yogesh Doshi was pr«jsent for MRI.  MRI and ISB were responsible for mon-
itoring the processes and air pollution control equipment, and ensuring
that they were satisfactory for testing.  MRI is serving as the overall
coordinator for the testing program.

     ES was also responsible for the collection of process inlet and
outlet samples, and for the observation of fugitive emissions from the
process outlet of the Herreschoff Furnace and the process outlet of the
Flash Calciner.  Visible Emissions observations were recorded for the
Herreschoff Furnace baghouse and scrubber stacks, and the baghouse stack
of the Flash Calciner.

     Because of constraints of non-technical nature, the original test
plan to conduct a particle size sampling traverse with each Method 5
particulate run was modified to collect a single particle size sample
per Method 5 run.

                                    1-1

-------
                                CHAPTER 2

                   SUMMARY OF TEST PROGRAM AND RESULTS
     This chapter presents the test results and a summary of the sampling
program conducted at the Burgess Pigment Company's Sandersville, Georgia
facility.

     From September 26 to 30, 1983, ES had a test crew of as many as 13
members responsible for particulate, particle sizing, fugitive emissions,
and visible emissions observations testing of the No. 2 Herreschoff Fur-
nace and the No. 4 Flash Calciner.  A total of six sites were tested for
particulate matter using an EPA Method 5 sampling train.  Five of those
sites were also tested for particle size distribution using an Andersen,
six-stage cascade impactor.  Fugitive emissions observations were made
on the Herreschoff Furnace process outlet and the Flash Calciner process
outlet.  Visible emissions observations were made on the Herreschoff Fur-
nace scrubber discharge outlet and baghouse outlet, and the Flash Calci-
ner baghouse outlet.

     Before presenting the results of the test program in this chapter,
an explanation of the sample identification code is in order.  Eighteen
Method 5 test runs a.nd 16 particle sizing test runs were performed at
the Burgess Pigment facility.  Each sample is identified by type and
test code.  The code identifies the source, sample type, and run number.
The codes are listed below.

     o  Herreschoff Furnace:   Baghouse Inlet        HF-BI
                               Baghouse Outlet       HF-BO
                               Wet Scrubber Inlet    HF-SI
                               Wet Scrubber Outlet   HF-SO

     o  Flash Calciner:        Baghouse Inlet        FC-BI
                               Baghouse Outlet       FC-BO

     o  Sample Type:           Particle Size         PS
                               Method 5              M5

     Runs are numbered sequentially for each site.  For example, the
first particle size run at the Flash Calciner baghouse outlet was coded
FC-BO-PS-1.  The first Method 5 run on the Herreschoff Furnace scrubber
inlet has the following run code:  HF-SI-M5-1.  The probe/nozzle/front-
half filter holder acetone rinses were labeled as F. H. Acetone.
                                    2-1

-------
     Visible emissions run sheets include the location of said emissions
and the corresponding Method 5 run number.  VEOs on the stacks are labeled
M9; fugitive emissions observations are labeled as M22.

     Four sites at the Herreschoff Furnace were tested for particulate
matter:  the scrubber inlet and outlet, and the baghouse inlet and outlet.
When reviewing emission data for the baghouse inlet, the reviewer should
bear in mind that the baghouse inlet duct is the pneumatic conveyor for
the calcined product, and the baghouse is the product recovery system.
Tables 2.1 through 2.4 present the Method 5 particulate matter run data
for the Herreschoff Furnace sites.  Data within tests are consistent for
all sites.  The volumetric flow rates  (dry normal cubic meters per minute
- DNm^/roi11) measured at the baghouse inlet and outlet vary considerably
(571 versus 811 DNm3/min).  The flow differences are the result of clean
air injection which is used to clean the bags in the baghouse.  The scrub-
ber inlet and outlet flow rates are not consistent, as well.  According
to Mr. Marc Sack of Burgess Pigment, there are at least two sources of
air in-leakage in the vicinity of the  scrubber, the significant source
being the seal around the variable throat of the scrubber.

     The isokinetic ratios were within the Method 5 criterion for all
Herreschoff Fv.rnace runs with the exception of Run HF-SI-M5-1 .  After
discussing this with Mr. Dennis Holzschuh of EPA/EMB, it was determined
that this run would be acceptable as long as the next two runs at that
site were within isokinetic specifications.

     EPA Method 5 tests were conducted on the Flash Calciner baghouse
inlet and outlet (see Tables 2.5 and 2.6).  All test results are consis-
tent for these sites, with the exception of concentration and emission
rate for Run No. FC-BO-M5-1.  After completion of this run, an inspec-
tion of the baghouse was conducted by  plant personnel; a bag was found
to have slipped off its mounting  (bottom ring) allowing a gap to form
through which a small portion of the gas stream vented.  After correc-
tion of this malfunction, concentrations and emission rates for the sub-
sequent runs dropped to a much lower level.  Note that the volumetric
flows are consistent between the inlet and outlet.  Unlike the Herre-
schoff Furnace baghouse, the Flash Calciner baghouse employs recycled
process air for bag cleaning.

     Tables 2.7 through 2.11 present the particle sizing test results.
The "percent recovery" line presented  in "emission data" of these summary
tables refers to the ratio of the particle sizing particulate concentra-
tion to the Method 5 particulate concentration.  These ratios offer some
indication of the effectiveness of the particle sizing sample recovery
and the representativeness of a given  particle size sample compared to
concurrent Method 5 run.

     The line item on these charts labeled as "concentration < 10 microns
(mg/DNm3 - from PS runs)" is derived from the total particulate matter
concentration of a given particle size sample multiplied by the cumula-
tive percent mass < D$Q of 10 microns  for that specific particle size
sample.  For the concentration <  10 microns from the Method 5 runs, the
cumulative percent mass < 050 of  10 microns was multiplied by the total
particulate matter concentration as determined by the Method 5 runs.
                                     2-2

-------
                                                          TABLE  2.1

                                                   BURGESS PIGMENT COMPANY
                                               SUMMARY OF METHOD 5 TEST RESULTS
                                             HERRESCHOFF FURNACE - BAGHOUSE  INLET
ro
U)
Run Number

SAMPLING DATA
Date
Clock Time
Total Sample Time (min)
Total Sample Volume (DNm3)
Isokinetic Ratio (%)
STACK DATA
Temperature (°C)
Moisture (%)
Velocity (m/min)
Volume Flow Rate (m^/min)
Volume Flow Rate (DNm^/min)
EMISSION DATA
Concentration (mg/DNm3)
Emission Rate (kg/hr)
HF-BI-M5-1
9/26/83
1452-1906
72
0.760
101.8
79
0.7
1638
747
597

87,464.9
3137.9
HF-BI-M5-2
9/27/83
1119-1315
36
0.363
106.4
78
1.8
1516
691
547

83,911.7
2754.6
HF-BI-M5-3 HF-BI-Avg
Q/77 /ft"*
1 A^ri— 1 fi9^
oo _
0-117
i no ^ - —

81 79
3.4 2.0
1613 1589
735 724
568 571

101,430.6 90,935.7
3458.6 3117.0

-------
                                                          TABLE  2.2

                                                   BURGESS  PIGMENT COMPANY
                                               SUMMARY OF METHOD 5 TEST  RESULTS
                                            HERRESCHOFF FURNACE  - BAGHOUSE  OUTLET
to
I
Run Number

SAMPLING DATA
Date
Clock Time
Total Sample Time (min)
Total Sample Volume (DNm3)
Isokinetic Ratio (%)
STACK DATA
Temperature (°C)
Moisture (%)
Velocity (m/min)
Volume Flow Rate (m^/min)
Volume Flow Rate (DNm^/min)
EMISSION DATA
Concentration (mg/DNm3)
Emission Rate (kg/hr)
HF-BO-M5-1
9/26/83
1453-1828
72
2.132
99.8
64
0.9
2080
956
813
2.6
0.1
HF-BO-M5-2
9/27/83
1102-1335
120
3.580
99.9
58
0.3
2047
941
819
1.5
0.1
HF-BO-M5-3
9/27/83
1449-1700
120
3.529
100.5
67
1.0
2073
953
802
0.5
<0.1
HF-BO-Avg





63
0.7
2067
950
811
1.5
0.1

-------
                                                           TABLE 2.3


                                                   BURGESS PIGMENT COMPANY

                                               SUMMARY  OF  METHOD 5 TEST RESULTS

                                             HERRESCHOFF FURNACE - SCRUBBER INLET
to
I
Ul
Run Number

SAMPLING DATA
Date
Clock Time
Total Sample Time (min)
Total Sample Volume (DNm3)
Isokinetic Ratio (%)
STACK DATA
Temperature (°C)
Moisture (%)
Velocity (m/min)
Volume Flow Rate (m3/min)
Volume Flow Rate (DNm3/min)
EMISSION DATA
Concentration (mg/DNm3)
Emission Rate (kg/hr)
HF-SI-M5-1
9/26/83
1640-1825
72
0.559
111.7
301
17.1
1025
299
125

8435.5
63.2
HF-SI-M5-2
9/27/83
1123-1311
72
0.543
103.0
294
17.2
1070
312
132

8151.2
64.4
HF-SI-M5-3
9/27/83
1510-1627
72
0.554
101.3
290
15.5
1080
315
137

9140.8
74.9
HF-SI-Avg






295
16.6
1058
309
131

8575.8
67.5

-------
                                                          TABLE 2.4

                                                   BURGESS PIGMENT COMPANY
                                               SUMMARY OF METHOD 5 TEST RESULTS
                                            HERRESCHOFF FURNACE - SCRUBBER OUTLET
ro
Run Number

SAMPLING DATA
Date
Clock Time
Total Sample Time (min)
Total Sample Volume (DNm3)
Isokinetic Ratio (%)
STACK DATA
Temperature ( °C )
Moisture (%)
Velocity (m/min)
Volume Flow Rate (m^/min)
Volume Flow Rate (DNm^/min)
EMISSION DATA
Concentration (mg/DNm3)
Emission Rate (kg/hr)
HF-SO-M5-1
9/26/83
1455-1826
64
0.795
109.5
56
18.9
958
216
155

56.0
0.5
HF-SO-M5-2
9/27/83
1120-1257
64
0.844
107.0
55
17.6
1022
231
168

38.5
0.4
HF-SO-M5-3
9/27/83
1510-1626
72
0.913
105.3
56
15.7
976
220
164

46.9
0.5
HF-SO-Avg






56
17.4
985
222
162

47.1
0.5

-------
                                                          TABLE 2.5

                                                   BURGESS PIGMENT COMPANY
                                               SUMMARY OF METHOD 5 TEST RESULTS
                                               FLASH CALCINER - BAGHOUSE INLET
to
I
-4
Run Number

SAMPLING DATA
Date
Clock Time
Total Sample Time (min)
Total Sample Volume (DNm3)
Isokinetic Ratio (%)
STACK DATA
Tempe ra tur e ( °C )
Moisture (%)
Velocity (m/min)
Volume Flow Rate (m3/min)
Volume Flow Rate (DNm3 /min)
EMISSION DATA
Concentration (mg/DNm3)
Emission Rate (kg/hr)
FC-BI-M5-1
9/28/83
1636-1917
72
0.705
101.7
130
7.1
1843
840
555

49,478.4
1646.9
FC-BI-M5-2
9/29/83
1021-1228
36
0.353
101 .6
123
8.9
1852
845
556

49,128.9
1638.8
FC-BI-M5-3
9/29/83
1430-1656
36
0.337
100.0
128
5.5
1761
803
539

48,903.3
1582.4
FC-BI-Avg






127
7.2
1819
829
550

49,170.2
1622.7

-------
                                                          TABLE 2.6



                                                   BURGESS PIGMENT COMPANY

                                               SUMMARY OF METHOD 5 TEST RESULTS

                                               PLASH CALCINER - BAGHOUSE OUTLET
to
I
oo
Run Number

SAMPLING DATA
Date
Clock Time
Total Sample Time (min)
Total Sample Volume (DNm3)
Isokinetic Ratio (%)
STACK DATA
Temperature ( °C )
Moisture (%)
Velocity (m/min)
Volume Flow Rate (m3/min)
Volume Flow Rate (DNm3/min)
EMISSION DATA
Concentration (mg/DNm3)
Emission Rate (kg/hr)
FC-BO-M5-1
9/28/83
1635-1859
128
2.708
101.7
91
5.9
1687
744
553
35.8
1.2
FC-BO-M5-2
9/29/83
1017-1317
176
3.960
102.5
77
5.7
1696
748
584
2.7
0.1
FC-BO-M5-3
9/29/83
1450-1730
176
3.963
101.0
83
5.1
1742
768
593
2.2
0.1
FC-BO-Avg





84
5.6
1708
753
577
13.6
0.5

-------
                                                          TABLE 2.7


                                                   BURGESS PIGMENT COMPANY
                                           SUMMARY OF PARTICLE SIZING TEST RESULTS
                                             HERRESCHOFF FURNACE - BAGHOUSE INLET
N)
I
ID
Run Number

SAMPLING DATA
Date
Clock Time
Impactor Flow Rate (1pm)
Isokinetic Ratio (%)
STACK DATA
Temperature (°C)
Moisture (%)
Velocity (m/min)
EMISSION DATA
Concentration (mg/DNm^)
Percent Recovery (PS cone/
M-5 cone)
Cumulative % Mass < 059
of 10 microns
Concentration < 1 0 microns
(mg/DNm-* - from PS runs )
Concentration < 10 microns
(mg/DNm^ - from M-5 runs)
Emission Rate < 10 microns
(kg/hr - from M-5 runs)
HF-BI-PS-1
9/26/83
1752-1752:15
22.480
180.1
76
0.7
1699

48,000
54.9
49.51
23,765
43,304
1553.6
HF-BI-PS-2
9/27/83
1318-1318:20
11 .608
91.6
78
1.8
1704

99,170
118.2
25.29
25,080
21,221
696.6
HF-BI-PS-3
9/27/83
1535-1535:15
10.617
82.1
82
3.4
1715

114,100
88.9
31.27
35,679
31,717
1081 .5
HF-BI-Avg




-
79
2.0
1706

87,090
87.3
35.36
28,200
32,047
1110.6

-------
               TABLE 2.8

        BURGESS PIGMENT COMPANY
SUMMARY OF PARTICLE SIZING TEST RESULTS
 HERRESCHOFF FURNACE - BAGHOUSE OUTLET
Run Number

SAMPLING DATA
Date
Clock Time
Impactor Flow Rate (1pm)
Isokinetic Ratio (%)
STACK DATA
Temperature (°C)
Moisture (%)
Velocity (m/min)
EMISSION DATA
Concentration (mg/DNm3)
Percent Recovery (PS cone/
M-5 cone)
Cumulative % Mass < 059
of 10 microns
Concentration < 10 microns
(mg/DNm3 - from PS runs)
Concentration < 10 microns
(mg/DNm3 - from M-5 runs)
Emission Rate < 10 microns
(kg/hr - from M-5 runs)
HF-BO-PS-1
9/26/83
1457-1557
15.855
99.6
63
0.9
2017

0.9
34.6
53.76
0.5
1.4
<0.1
HF-BO-PS-2
9/27/83
1104-1304
15.996
101.4
58
0.3
1999

0.3
20.0
90.63
0.3
1.4
<0.1
HF-BO-PS-3
9/27/83
1448-1648
16.619
103.5
67
1.0
2026

0.6
120.0
11.76
0.1
<0.1
HF-BO-Avg





63
0.7
2014

0.6
58.2
52.05
0.3
1.0
<0.1

-------
                                                          TABLE 2.9



                                                   BURGESS PIGMENT COMPANY

                                           SUMMARY OF PARTICLE SIZING TEST RESULTS

                                             HERRESCHOFF FURNACE - SCRUBBER  INLET
to
I
Run Number

SAMPLING DATA
Date
f* "\ f\f*]c T*T Tno
l*X(Jl*Jx J. -Llllc
Impactor Flow Rate (1pm)
T Q nV i n o t- 1 r» P a i- 1 r» ( 9t\
J-OUJ^XlltZ l~-LL> fxd I.X (J 1^7
STACK DATA
Temperature (°C)
Moisture (%)
Velocity (m/min)
EMISSION DATA
Concentration (mg/DNm3)
Percent Recovery (PS cone/
M-5 cone)
Cumulative % Mass < 059
of 10 microns
Concentration < 1 0 microns
(mg/DNm^ - from PS runs)
Concentration < 10 microns
(mg/DNm3 - from M-5 runs)
Emission Rate < 10 microns
(kg/hr - from M-5 runs)
HF-SI-PS-1

9/26/83
1 "7^fi — 1 "7^ft
I / JO I / -JO
9.513
117 1
i i / • t
310
17.1
997

5919

70.2

42.36

2507

3573

26.8
HF-SI-PS-2

9/27/83
1 99fl — 1 991
1 £f£*\J \ £• & \
9.201
1 79 fi
1 £tfM . D
296
17.2
1057

6142

75.4

46.88

2879.4

3821

30.2
HF-SI-PS-3

9/27/83
1 ^^Q— 1 ^^n
1 J J_7~ 1 ,J**\J
10.900
141 9
i *« i . ^
275
15.5
1050

5864

64.2

52.42

3074

4792

39.3
HF-SI-Avg








294
16.6
1035

5975

69.9

47.22

2820

4062

32.1

-------
                                                           TABLE 2.10

                                                    BURGESS PIGMENT COMPANY
                                            SUMMARY OF PARTICLE SIZING  TEST  RESULTS
                                                FLASH CALCINER - BAGHOUSE  INLET
I
to

SAMPLING DATA
Date
Clock Time
Impactor Flow Rate (1pm)
Isokinetic Ratio (%)
STACK DATA
Temperature (°C)
Moisture (%)
Velocity (m/min)
EMISSION DATA
Concentration (mg/DNm3)
Percent Recovery (PS cone/
M-5 cone)
Cumulative % Mass < 059
of 10 microns
Concentration < 1 0 microns
(mg/DNm3 - from PS runs)
Concentration < 1 0 microns
(mg/DNm3 - from M-5 runs)
Emission Rate < 10 microns
(kg/hr - from M-5 runs)

FC-BI-PS-1
9/28/83
1744-1744:10
13.250
100.1
132
7.1
1774

33,280
67.3
65.23
21,709
32,275
1074.3

FC-BI-PS-2
9/29/83
1044-1044:05
10.928
77.5
123
8.9
1770

43,310
88.2
51.79
22,430
25,444
848.7
Run Number
FC-BI-PS-3
9/29/83
1530-1530:03
5.804
45.3
136
5:5
1779

83,060
169.8
31.31
26,006
15,312
495.4

FC-BI-PS-4
9/29/83
1705-1705:02
2.548
18.8
133
5.5
1779

92,410
189.0
55.16
50,973
26,975
872.9

FC-BI-Avg





131
6.8
1776

63,015
128.6
50.87
30,300
25,002
822.8

-------
                                                          TABLE  2.11

                                                   BURGESS PIGMENT COMPANY
                                           SUMMARY OF PARTICLE SIZING  TEST RESULTS
                                               FLASH CALCINER -  BAGHOUSE  OUTLET
to
I
u>
Run Number

SAMPLING DATA
Date
Clock Time
Impactor Flow Rate (1pm)
Isokinetic Ratio (%)
STACK DATA
Temperature (°C)
Moisture (%)
Velocity (m/min)
EMISSION DATA
Concentration (mg/DNm3)
Percent Recovery (PS cone/
M-5 cone)
Cumulative % Mass < 050
of 10 microns
Concentration < 10 microns
(mg/DNm3 - from PS runs )
Concentration < 10 microns
(mg/DNm3 - from M-5 runs)
Emission Rate < 10 microns
(kg/hr - from M-5 runs)
FC-BO-PS-1
9/28/83
1635-1843
15.515
104.7
91
5.9
1699

24.1
67.3
95.18
22.9
34.1
1.1
FC-BO-PS-2
9/29/83
1017-1313
14.496
103.7
76
5.7
1659

1.0
37.0
79.85
0.8
2.2
<0.1
FC-BO-PS-3
9/29/83
1430-1726
15.034
102.2
83
5.1
1673

1.9
86.4
89.90
1.7
2.0
<0.1
FC-BO-Avg





83
5.6
1677

9.0
63.6
88.31
8.5
12.8
0.4

-------
(Similarly for emission rate < 10 microns, the Method 5 calculated rate
was used.)

     The particle size summaries for the Herreschoff Furnace scrubber in-
let and the baghouse inlet and outlet are presented in Tables 2.7 through
2.9.  As explained elsewhere in this report, particle size testing of the
scrubber outlet could not be performed, due to the presence of entrained
water droplets.  Entrained water droplets prevent adequate particle size
sampling because, if auxiliary heating is employed to vaporize the drop-
lets, the gas volume through the impactor is altered in an undefined man-
ner.  Also, these entrained water droplets may be carrying suspended or
dissolved particulate: matter.  Upon vaporization, the droplets will leave
behind this material, in positions in the impactor which do not necessar-
ily represent the true particle size distribution in the gas stream.

     Referring to the percent recovery line on these tables, it will be
observed that the percent recovery varied considerably at all sites.  As
careful laboratory technique was practiced by both ES laboratory techni-
cians responsible for particle size sample recovery, it can be assumed
that the variability is the result of other factors.  For the baghouse
inlet and scrubber inlet, sample run times were short because of the high
grain loading present at these two sites.  Notice, also, that for these
sites, the isokinetic ratios were not within specification for all runs.
(90-110% per EPA/EMB correspondence and 80-120% per the IERL guidelines,
"Procedures for Cascade Impactor Calibration and Operation in Process
Stream - Revised 1979.)  This, too, may be attributed to the short run
times required at these sites.  Factors which prevent accurate measure-
ment of gas volumes under such conditions, and, hence, hinder isokinetic
sampling and acceptable particle size sample recovery, are listed below:

     o  The orifice/dry gas meter system in an EPA type Method 5 meter
        box is not designed to accurately measure small gas volumes.  For
        example, the total gas volume collected during Run No. HF-BI-PS-1
        was 0.00467 rP (0.165 ft3), which is equivalent to less than 2
        revolutions of the dry gas meter needle.

     o  While the impactor gas flow rate was preset at the meter box prior
        to sampling, time is required for the vacuum pump to stabilize at
        the preset flow rate after pump activation.  While this takes but
        a few seconds, and has a negligible effect during a Method 5 run,
        the effect can be substantial if the run duration is 10 to 15 sec-
        onds .

For the reasons stated above, the particle size data for the Herreschoff
Furnace baghouse inlet should be considered circumspect.  Note, also,
that the baghouse inlet particle size runs' substrates were all over-
loaded, based on both visual observation and weight gains.  Substrate
loading was satisfactory for the scrubber inlet.  While the percent re-
coveries for the scrubber inlet runs were low (average 69.9%), it was
consistent between runs.  The cumulative percent mass less than the Dgg
of 10 microns was consistent between runs, as well.  Thus, while the
percent recovery was low and the isokinetic ratios were not within al-
lowed criteria for all runs, the scrubber inlet particle size runs of-
fer an indication of the particle size distribution.
                                    2-14

-------
     Percent recoveries were not consistent for the Herreschoff Furnace
baghouse outlet.  At this site, while sample times were sufficient to pro-
vide accurate gas volume measurements (refer to the isokinetic ratios),
the grain loading in the exhaust gas was so low that insufficient mate-
rial was collected on the substrates for accurate weighing.  Sample time
was expanded from 1 hour to 2 hours in an attempt to accommodate this
condition.  However, for all three runs at this site, weight gains for
the majority of the substrates were lower than the precision established
for the determination of a constant weight (i.e., less than 0.05 mg).
This situation was discussed with Mr. Dennis Holzschuh; an agreement was
reached between he and ES that 2 hours sampling time could be considered
satisfactory.  The observed low grain loadings would require sample times
to be expanded beyond acceptable limits.

     Tables 2.10 and 2.11 present particle size test summaries for the
Flash Calciner baghouse inlet and outlet.  As for the Herreschoff Furnace,
the particle size data generated for these sites should be considered of
minimal validity, for similar reasons.  The sample time for the baghouse
inlet was decreased to as low as 2 seconds in attempts to minimize sub-
strate overloading.  As previously described, a baghouse malfunction ef-
fected the high grain loading measured during Run No. FC-BO-PS-1.  The
loading for Run Nos. FC-BO-PS-2 and 3 was higher than that measured at
the Herreschoff Furnace baghouse outlet, but still too low for satisfac-
tory weight measurement  (average substrate weight gains for Run No. FC-
BO-PS-2 and 3 was 0.42 mg, the highest weight gain was 1.14 mg).  Sample
time was expanded from 2 hours to 3 hours at the outlet site in an at-
tempt to recover sufficient material for weighing.

     Another indicator of particle size sample representativeness is the
overall particle size distribution.  Figures 2.1 through 2.5 present the
particle size distribution for the runs conducted on the Herreschoff Fur-
nace and Flash Calciner sites.  Referring to the previous discussion re-
garding particle size sampling, these figures graphically display the
apparent variations of size distributions measured at each site.

     The data for these curves compare the cumulative percent mass small-
er than a given size versus standard particle size diameters in microns.
The size is the diameter of the particle collected at 50% efficiency on
a given impactor stacre and is referred to as 050.  Distributions were cal-
culated according to Mercer's definition, whereby aerodynamic impaction
is used to quantify particle size.  Unit density was used for particulate
matter density.  Five standard deviatons were used to establish acceptable
particle size data.

     Data for differential size distribution on a mass basis (dM/d (logD)
- mg/DNm^) is supplied in Appendix B of this report.  Plots are not sup-
plied for these data.

     Tables 2.12 through 2.14 present control device performance summaries.
Baghouse efficiencies were 99.997% and 99.97% for the Herreschoff Furnace
and Flash Calciner baghouses, respectively.  The Flash Calciner baghouse
efficiency is lower because of the anomalous first run at the baghouse out-
let (refer to discussion regarding baghouse malfunction).  If the average
                                    2-15

-------
                                         PTCTTRT? 9-1
  BURGESS PIGMENT  COMPANY

      PARTICLE SIZE TESTING
SANDERSVILLE, GA

TEST SITE:  HFBI       RUN NO.:   PS-1,2,3       DENSITY
RUN DATE:  9/26-27/83   START TIME: 1752,1318,1535


O  Run 1
A  Run 2
O  Run 3
J.UU
90
80
70
60
50
40
30
20
10
§ 6
s ^
4
H
1 '
1 *
0
m
Q
1
.9
.8
.7
.6
.5
.4
.3
.2
.1
0.

































































































































f























i


H






















:R



'7)





















El!



















p^p



r
vl





















1
^


1
^





















(i
^
j

























)






































































































.

























































































































































































01 0.10.51 2 5 10 20304050607080 9095 9899 99.8 99,
0.050.2 99.9
CUMULATIVE % LESS THAN
99
                    2-16
                                ENGINEERING-SCIENCE

-------
                                         FIGURE 2.2
  BURGESS PIGMENT  COMPANY

      PARTICLE SIZE TESTING
SANDERSVILLE, GA

TEST SITE: HFBO      RUN NO.:   PS-1,2,3       DENSITY
RUN DATE:  9/26-27/83   START TIME: 1457,1104,1448

O  Run 1
A  Run 2
n  Run 3
J.UU
90
80
70
60
50
40
30
20
10
1
1 7
g 5
4
H
2 0
M 3
o
H
3 2
0
o
Q
1
.9
.8
.7
.6
. .5
,4
.3
.2
.1
0.





































































































































































I
r




&





1
1


















n
M


D




























fc-*


A













F








A







































































(Tl


























.







A

















































A
^^^





















A
•^A

























4


























Lw4

























































































































































01 0.10.51 2 5 10 20304050607080 9095 9899 99.8 99.
0.050.2 99,9
CUMULATIVE % LESS THAN
99
                    2-17
ENGINEERING-SCIENCE

-------
                                         FIGURE 2.3
  BURGESS PIGMENT  COMPANY
      PARTICLE SIZE TESTING
 SANDERSVILLE, GA
 TEST SITE: HFSI      RUN NO.:  PS-1,2,3      DENSITY
 RUN DATE:  9/26-27/83   START TIME: 1756,1220,1539
O Run 1
A Run 2
D Run 3
J.UU
90
80
70
60
50
40
30
20
10
9
2 8
o 6
3 5
4
H
L— i 3
o
1 2
O
u-i
O
1
.9
,8
.7
.6
.5
.4
.3
.2
.1
0.





































































































































































































m
OH v






























@
i

























&













































i
¥6
as







ito
S



tti
Bl






































































































































































































































































































































01 0.10.51 2 5 10 20304050607080 9095 9899 99,8 99.
0.050.2 99.9
CUMULATIVE % LESS THAN
99
                     2-18
ENGINEERING-SCIENCE

-------
(
t
100
90
80
70
60
50
40
30
20
10
1
s I
0 6
4
H
5 3
o
P-
S 2
CJ
o
tn
O
1
.9
.8
.7
.6
.5
.4
.3
,2
.1
0.
BURGESS PIGMENT COMPANY
PARTICLE SIZE TESTING
SANDERSVILLE, GA
TEST SITE: FCBI RUN NO.: PS-1,2,3,4 DENSITY
RUN DATE: 9/28-29/83 START TIME: 1744,1044,1530,1705
D Run 1
& Run 2
D Run 3
O Run 4














































































































































































































S>

























EE>



















,







t


















3







2£






















flS
t=>


>


















&



f.
V!




















®>
£j>
•> (
' \*S



)





















l)





































































































































































































































































01 0.10.51 2 5 10 20304050607080 9095 9899 99.8 99.
0.050,2 99.9
CUMULATIVE % LESS THAN
99
2-19
ENGINEERING-SCIENCE

-------
                                       FIGURE 2.5
BURGESS PIGMENT COMPANY
    PARTICLE SIZE TESTING
SANDERSVILLE„ GA

TEST SITE: FCBO
RUN DATE:  9/28-29/83
              RUN NO.:   PS-1,2,3
              START TIME: 1635,1017,1430
DENSITY = lgm/cm3
 Run 1
 Run 2
O

D  Run 3
xuu
90
80
70
60
50
40
30
20
10
9
£ 8
s 1
CJ 0
S ^
4
H
2 -
M 3
o
PM
1 ' *
O
in
Q
1
.9
.8
.7
.6
.5
.4
.3
.2
.1
0.


















































































































































































































[7


























(•;






















(i



A





















r«
v£


























i
a



















































3



















C«]


'fin
3U




















i
c




®.




















V
A



























hw/

























«





















































•



































































































01 0.10.51 2 5 10 20304050607080 9095 9899 99.8 99.
0.050.2 99.9
CUMULATIVE % LESS THAN
)
99
                  2-20
                                 ENGINEERING-SCIENCE

-------
                                TABLE 2.12

                         BURGESS PIGMENT COMPANY
                 HERRESCHOFF FURNACE BAGHOUSE PERFORMANCE
                                   Baghouse Inlet
                   Baghouse Outlet
Emissions
  (kg/hr)

Emissions < 10 microns
  (kg/hr)

Concentration
  (mg/DNm3)

Concentration < 10 microns3
  (mg/DNm3)

Stack Gas Moisture
  3117.0
  1110.6
90,935.7
  32,047
     2.0
0.1
1.5
0.3
0.7
Stack Gas Temperature
      79
 63
Volumetric Flow Rate
  (DNm3)

Baghouse Efficiency
  - Total (%)

Baghouse Efficiency
  - < 10 microns (%)
     571
811
                        99.997
                        99.991
a  From particle size runs.
                                    2-21

-------
                                TABLE 2.13

                         BURGESS PIGMENT COMPANY
                 HERRESCHOFF FURNACE SCRUBBER PERFORMANCE
                                   Scrubber Inlet
                  Scrubber Outlet
Emissions
  (kg/hr)

Emissions < 10 microns
  (kg/hr)

Concentration
  (mg/DNm3)

Concentration < 10 microns3
  (mg/DNm3)

Stack Gas Moisture
Stack Gas Temperature
  67.5
  32.1
8575.8
  2820
  16.6
   295
 0.5
                       	b
46.9
                       	b
15.7
  56
Volumetric Flow Rate
  (DNm3)

Scrubber Efficiency
  - Total (%)
   131
 164
                       99.3
a  From particle size runs.
b  No particle size runs conducted at this site due to the presence of
   entrained water droplets.
                                    2-22

-------
                                TABLE 2.14

                         BURGESS PIGMENT COMPANY
                   FLASH CALCINER BAGHOUSE PERFORMANCE
                                   Baghouse Inlet
                   Baghouse Outlet
Emissions
  (kg/hr)

Emissions < 10 microns
  (kg/hr)

Concentration
  (mg/DNm3)

Concentration < 10 micronsa
  (mg/DNm3)

Stack Gas Moisture
Stack Gas Temperature
  1622.7


   822.8


49,170.2


  30,300


     7.2


     127
 0.5
 0.4
13.6
 8.5
 5.6
  84
Volumetric Flow Rate
  (DNm3)

Baghouse Efficiency
  - Total (%)

Baghouse Efficiency
  - < 10 microns (%)
     550
 577
                        99.97
                        99.95
a  From particle size runs.
                                    2-23

-------
of Run Nos. FC-BO-M5-2 and 3 is used in the determination of baghouse ef-
ficiency, the recalculated efficiency is 99.994%.  The Herreschoff Furnace
scrubber efficiency was 99.3% for the series of runs conducted by ES.

     Tables 2.15 through 2.19 present summaries of visible and fugitive
emissions observations at the various observation sites at the Burgess
Pigment Company.  No visible emissions were observed at the Herreschoff
Furnace sites - the baghouse and scrubber stacks and the process outlet.
Visible emissions were observed for the first run, only, at the Flash
Calciner baghouse outlet.  Infrequent fugitive emissions caused by an
equipment malfunction were observed at the Flash Calciner process out-
let.  These events occurred an average rate of once per Method 5 run.
                                    2-24

-------
                      TABLE 2.15

               BURGESS PIGMENT COMPANY
             SUMMARY OF VISIBLE EMISSIONS
        HEEIRESCHOFF FURNACE - BAGHOUSE OUTLET
RUN NO. HF--BO-M9-1

        No visible emissions observed from this stack
        (9/26/83 - 1446 to 1833)
RUN NO. HF-BO-M9-2

        No visible emissions observed from this stack
        (9/27/83 - 1100 to 1345)
RUN NO. HF-BO-M9-3

        No visible emissions observed from this stack
        (9/27/83 - 1455 to 1659)
                          2-25

-------
                       TABLE 2.16

                BURGESS PIGMENT COMPANY
             SUMMARY OF FUGITIVE EMISSIONS
          HEIRRESCHOFF FURNACE - PROCESS OUTLET
RUN NO. HF-OUTLET-M22-1

        No fugitive emissions observed at this location
        (9/26/83 - 1505 to 1826)
RUN NO. HF-OUTLET-M22-2

        No fugitive emissions observed at this location
        (9/27/83 - 1109 to 1301)
RUN NO. HF-OUTLET-M22-3

        No fugitive emissions observed at this location
        (9/27/83 - 1530 to 1617)
                           2-26

-------
                      TABLE 2.17

               BURGESS PIGMENT COMPANY
             SUMMARY OF VISIBLE EMISSIONS
        HERRESCHOFF FURNACE - SCRUBBER OUTLET
RUN NO. HF-SO-M9-1

        No visible emissions observed from this stack
        (9/26/83 - 1450 to 1834)
RUN NO. HF-SO-M9-2

        No visible emissions observed from this stack
        (9/27/83 - 1102 to 1345)


RUN NO. HF-SO-M9-3

        No visible emissions observed from this stack
        (9/27/83 - 1455 to 1734)
                          2-27

-------
                            TABLE 2.18

                     BURGESS PIGMENT COMPANY
                   SUMMARY OF VISIBLE EMISSIONS
                 FLASH CALCINER - BAGHOUSE OUTLET
RUN NO. FC-BO-M9-1  (9/28/83)
Set
No.
1
2
3
4
5
6
7
8

Start
1646
1652
1658
1704
1713
1725
1734
1749

End
1652
1658
1704
1710
1725
1734
1747
1756
Average
% Opacity
0.63
0.21
0.21
0
0
0
0
0
Set
No.
9
10
11
12
13
14
15


Start
1756
1805
1820
1828
1839
1848
1857


End
1805
1818
1828
1839
1848
1857
1908

Average
% Opacity
0
0
0
0
0
0
0

RUN NO. FOBO-M9-2
        No visible emissions observed from this stack
        (9/29/83 - 1015 to 1323)
RUN NO. FC-BO-M9-3
        No visible emissions observed from this stack
        (9/29/83 - 1445 to 1729)
                                2-28

-------
                                TABLE 2.19

                         BURGESS PIGMENT COMPANY
                      SUMMARY OF FUGITIVE EMISSIONS
                     FLASH CALCINER - PROCESS OUTLET
RUN
Set
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
NO. FC-PO-M22-1 (9/28/83)

Start
1635
1641
1647
1653
1659
1705
171 1
1717
1723
1729
1735
1741
1747

End
1641
1647
1653
1659
1705
1711
1717
1723
1729
1735
1741
1747
1753
Emission
Freq . %
0
0
0
0
0
12.5
0
0
0
0
0
0
0
Avg. %
Opacity
0
0
0
0
0
9
0
0
0
0
0
0
0
Set
No.
14
15
16
17
18
19
20
21
22
23
24
25


Start
1753
1759
1805
1811
1817
1823
1834
1840
1846
1852
1858
1904


End
1759
1805
181 1
1817
1823
1833
1840
1846
1852
1858
1904
1910

Emission
Freq . %
. 0
0
0
8.3
0
0
0
0
0
0
0
0

Avg. %
Opacity
0
0
0
4.8
0
0
0
0
0
0
0
0

Emission frequency for entire run - 0.8%
Opacity range for entire run - 0-95%
RUN NO. FC-PO-M22-2 (9/29/83)

Set               Emission  Avg. %   Set               Emission  Avg. %
No.  Start  End   Freq. %   Opacity  No.  Start  End   Freq. %   Opacity
1
2
3
4
5
6
7
8
9
10
11
12
13
14
1019
1026
1032
1038
1044
1050
1056
1102
1108
1114
1120
1126
1132
1138
1025
1032
1038
1044
1050
1056
1102
1108
1114
1120
1126
1132
1138
1146
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
9
0
0
0
0
0
0
0
0
15
16
17
18
19
20
21
22
23
24
25
26
27
28
1150
1156
1202
1208
1214
1220
1226
1232
1238
1244
1250
1256
1302
1308
1156
1202
1208
1214
1220
1226
1232
1238
1244
1250
1256
1302
1308
1314
0
0
0
0
0
0
8.3
0
0
0
0
0
0
0
0
0
0
0
0
0
3.5
0
0
0
0
0
0
0
Emission frequency for entire run - 0.3%
Opacity range for entire run - 0-65%
                                    2-29

-------
TABLE 2.19~Continued
RUN
Set
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
NO. FC-PO-M22-3 (9/29/83)
Emission Avg. %
Start End
1433
1439
1445
1451
1457
1503
1509
1515
1521
1527
1533
1539
1545
1551
1557
Emission
1439
1445
1451
1457
1503
1509
1515
1521
1527
1533
1539
1545
1551
1557
1603
frequency
Opacity range for
Set
Freq. % Opacity No.
0
0
0
0
0
0
0
0
0
0
0
0
0
0
8.3
for entire
entire run
0
0
0
0
0
0
0
0
0
0
0
0
0
0
4.4
run
- 0-1
16
17
18
19
20
21
22
23
24
25
26
27
28
29

- 0.3%
00%
Emission
Start
1603
1612
1618
1624
1630
1636
1642
1648
1654
1700
1706
1712
1718
1724



End Freq . %
1612
1618
1624
1630
1636
1642
1648
1654
1700
1706
1712
1718
1724
1730



0
0
0
0
0
0
0
0
0
0
0
0
0
0



Avg. %
Opacity
0
0
0
0
0
0
0
0
0
0
0
0
0
0



          2-30

-------
                                CHAPTER 3

                   SAMPLE SITES AND METHOD REQUIREMENTS
     This section describes the specific sampling sites tested and the
sampling methodology employed at these sites.  The physical configuration,
number of sample points, time of sampling, and special considerations are
described, and sketches of the locations are also provided.  A summary of
this information is provided in Table 3.1.

     A total of six sites were sampled for particulate matter at the
Burgess Pigment Company, four on the Herreschoff Furnace, and two on the
Flash Calciner.  The four Herreschoff furnace sites consist of the wet
scrubber inlet and outlet and the baghouse inlet and outlet; the two Flash
Calciner sites are the baghouse inlet and outlet.  The four Herreschoff
furnace sites were tested simultaneously, as were the two Flash Calciner
sites.  Observations were made for fugitive dust emissions at product
transfer points for both processes.  Visible emissions observations were
made at the three exhaust gas discharge points (stacks).  Raw and finished
product grab samples were collected during each Method 5 run for subse-
quent moisture and sieve analysis.

     Five of the aforementioned sites were also tested for particle size
distribution using a. cascade impactor.  The scrubber outlet was not tested
because of entrained water droplets, which prevent satisfactory particle
size sampling.  Particle size samples were collected at points of average
stack gas velocity, concurrent to each Method 5 particulate run.

Herreschoff Furnace

     Wet Scrubber Inlet

     Figures 3.1 a and 3.1b present the wet scrubber inlet sampling site
and sampling point configuration.  The nearest upstream disturbance, a
bend in the duct, is approximately 96 inches, or 4 diameters; the nearest
downstream disturbance is greater than 8 diameters.  To meet EPA Method
1 criteria, a total of 24 traverse points were sampled.  Sampling time
at each traverse point was 3 minutes, for a total sampling time of 72
minutes per Method 5 run.  Due to the high grain loading, the duration
of each particle size run was 2 minutes for the first run, and 1 minute
for Runs 2 and 3.  Particle size samples were collected at different
traverse points (see; Figures 3.1b); all particle size sample points were
points of average velocity.  Samples were collected through two, 4 inch
ports, positioned 90° to each other.
                                    3-1

-------
                                                            TABLE 3.1





                                               STACK GAS COMPOSITION AND CONDITIONS
UJ
Duct Dimension
Source (inches)
Herrescnoff Furnace
Scrubber Inlet 24 diam. I.D.
Scrubber Outlet 21-1/8 diam. I.D.
Baghouse Inlet 30-1/8 diam. I.D.
Baghouse Outlet 30 diam. I.D.
Flash Calciner
Baghouse Inlet 30 diam. I.D.
Baghouse Outlet 29-1/2 diam. I.D.
Average
Average Average Volumetric
Temp. Moist. Flow Rate
(°C) (%) (DNm3)
295 16.6 131
56 15.7 164
79 2.0 571
63 0.7 811
127 7.2 550
84 5.6 577
Method 5
Sampling Time
Combus- (minutes)
tion Per
Gases Point Total3
Yes 3 72
Yes 8, 9 64, 64,
72
No 2 72, 36,
32
No 3, 5 72, 120,
120
Yes 2 72, 36,
36
Yes 8, 11 128, 176,
176
            a  Various times, see text.

-------
                                                  FIGURE 3.1 a
HERRESCHOFF FURNACE SCRUBBER INLET
                 HERRESCHOFF FURNACE ENCLOSURE
                                   4" PORTS, 90° TO EACH OTHER
                                    (PIPE NIPPLE WITH CAP)
                            I
          GUARD RAIL-
      CATWALK
                       4 PORT-
                        DIRECTION
                        OF GAS \
                        FLOW  1
i
i
  HERRESCHOFF
  FURNACE
  ENCLOSURE
I
      >80TO
       DOWNSTREAM
       DISTURBANCE
                                      V
                 BAGHOUSE
                 ROOF LINE
                        ELEVATION
                            3-3
    ENGINEERING-SCIENCE

-------
                                   FIGURE 3.1b
    SAMPLING  POINT LOCATION
      HERRESCHOFF FURNACE
        SCRUBBER  INLET
 4 PORTS
M-5 SAMPLING
POINT
"A" & "B" PORTS
1
2
3
4
5
6
DISTANCE TO
OUTER EDGE. OF
PORTS (INCHES)
26.50
25.39
24.17
22.75
21.00
18.46
M-5 SAMPLING
POINT
"A" & "B" PORTS
7*
8
9
10
11
12
DISTANCE TO
OUTER EDGE OF
PORTS (INCHES)
11.54
9.00
7.25
5.83
4.61
3.5
a  Particle Size Sampling Point - "A" & "B" Ports
                  3-4
ENGINEERING-SCIENCE

-------
     Wet Scrubber Outlet

     Figures 3.2a and 3.2b present the sample site location and sampling
point configuration for the scrubber outlet.  Two, 4 inch ports were in-
stalled for the collection of Method 5 samples.  No particle size samples
were collected, due to the presence of entrained water droplets.*  The
nearest upstream disturbance, a transition piece from the induced draft
fan, is greater than 8 diameters from the sample ports.  The nearest
downsream disturbance, the stack exit, is greater than 8 diameters from
the sample ports.  These distances dictate a total of 8 sample points for
the Method 5 sampling matrix.  Sampling time at each point was 8 minutes
for Runs 1 and 2, and 9 minutes for Run 3.  The run time was increased
for Run 3 to effect a larger sample volume.

     Baghouse Inlet

     Figures 3.3a and 3.3b present the sampling site and sampling point
locations for the baghouse inlet.  The upstream and downstream flow dis-
turbances, 3.5 diameters and approximately 8 diameters, respectively,
dictate a sampling matrix of 36 sampling points.  Method 5 and particle
size sampling was conducted through two, 4 inch ports, positioned 90° to
each other.  Method 5 sampling time was 2 minutes per point, for a total
sample time of 72 minutes.  However, for Runs 2 and 3, the sample time
was shortened to a total of 36 and 32 minutes, respectively, by sampling
a single traverse, instead of the two traverses normally required.  This
action was necessitated by the excessively high grain loading present in
the duct, which effected high pressure drops across the Method 5 filters,
and required the replacement of several filters.  The decision was reached,
after a discussion with Mr. Ed McCarley of EPA/EMB, that this would be a
satisfactory sampling alternative because of the difficulty encountered
in sampling at this site and the fact that it is the inlet to a control
device, and hence, does not represent a direct atmospheric discharge.
Run 3 was shortened an additional 4 minutes because a filter support col-
lapsed during a leak check prior to the filter change.  As this happened
during the leak check, no sample was lost.  However, the decision was
made to terminate the run at that time rather than complete the final 4
minutes of the traverse with a new filter.

     Sampling time for the particle size runs was 15 seconds for Run 1
and Run 3 and 20 seconds for Run 2.  These short sampling times were re-
quired because of the high grain loadings experienced at this site.  Par-
ticle size sampling points are presented in Figure 3.3b; all points are
points of average velocity.

     Baghouse Outlet.

     Figures 3.4a and 3.4b depict the baghouse outlet sampling site con-
figuration and sampling point locations.  The sampling ports, two 4 inch
pipe nipples, positioned 90° to each other, are located 2.4 diameters from
the stack exit and 8 diameters to the nearest upstream flow disturbance.
These distances dictate a total of 12 sampling points for the Method 5
sampling matrix.  However, in the desire to collect a more representative
sample, this number was doubled to 24 points.  Sampling time was 3 minutes
                                    3-5

-------
                                                   FIGURE 3.2a
HERRESCHOFF FURNACE SCRUBBER OUTLET
                       BAGHOUSE ENCLOSURE
                                :^^^^^\v^^^^^^*^^*^^s^^x
    STACK EXIT
    4 PORT-
      STACK-
                                    4 PORTS, 90° TO EACH OTHER
                                      (PIPE NIPPLE WITH CAP)
                           PLAN
>20 TO
DOWNSTREAM
DISTURBANCE
^                                GUARD
                                RAIL
                               V CATWALK
                    >80 TO
                    UPSTREAM
                    DISTURBANCE
                        BAGHOUSE
                        ENCLOSURE
                          ELEVATION
                            3-6
                       ENGINEERING-SCIENCE

-------
                              FIGURE 3.2b
  SAMPLING POINT LOCATION
   HERRESCHOFF FURNACE
      SCRUBBER OUTLET
4 PORTS
M-5 SAMPLING
POINT
"A" ,& "BM PORTS
1
2
3
4
DISTANCE TO
OUTER EDGE OF
PORTS (INCHES)
23
19
8-1/2
4-5/8
               3-7
ENGINEERING-SCIENCE

-------
                                           FIGURE 3.3a
HERRESCHOFF FURNACE BAGHOUSE INLET
          STEEL SUPERSTRUCTURE
         4 PORTS,
  90°TO EACH OTHER

     GUARD RAIL
CORNER POST
                      PLAN
      (~80 TO
\ 1
I
1C


t
)5"
t
5"
.AM i | |
YCE) T U


A


v- — 'IN KU
-^-COOLING DUC
> — 4" PORTS
1 DIRECTION
GAS FLOW
T
Lh*- STEEL SUI
                     ELEVATION
                       3-8
  ENGINEERING-SCIENCE

-------
                               FIGURE 3.3b
  SAMPLING POINT LOCATION
   HERRESCHOFF FURNACE
      BAGHOUSE INLET
4 PORTS


II









a
b
M-5 SAMPLING
POINT
A" & "B" PORTS
1
2
3
4a
5
6
7
8^
9
Particle Size
Particle Size
DISTANCE TO
OUTER EDGE OF
PORTS (INCHES)
4.0
4.3
5.2
6.3
7.4
8.6
10.1
11.9
14.5
Sampling Point -
Sampling Point -
M-5 SAMPLING
POINT
"A" & "B" PORTS
10
11
12
13
14
15
16D
17
18
"A" Port
"B" Port
DISTANCE TO
OUTER EDGE OF
PORTS (INCHES)
21.5
24.1
25.9
27.4
28.6
29.7
30.8
31.7
32.0


                3-9
ENGINEERING-SCIENCE

-------
                                           FIGURE 3.4a
HERRESCHOFF FURNACE BAGHOUSE OUTLET
                                 PORTS
                               90° TO EACH OTHER
                                 STACK
                      PLAN
                               •STACK EXIT
       (CATWALK)

4" PORTS^^


.ATFORM— '


o o

- — - — 1 — - —
>
7

1
>'


80 7
DISTUF
^v
                               80 TO UPSTREAM
                    ELEVATION
                        3-10
ENGINEERING-SCIENCE

-------
                                    FIGURE 3.4b
    SAMPLING POINT  LOCATION
     HERRESCHOFF FURNACE
        BAGHOUSE OUTLET
 4 PORTS
M-5 SAMPLING
POINT
"A" & "B" PORTS
1
. 2
3
4
5
6
DISTANCE TO
OUTER EDGE OF
PORTS (INCHES)
32-1/2
31-3/8
29-5/8
27-7/8
25-5/8
22-3/8
M-5 SAMPLING
POINT
"A" & "B" PORTS
7b
8*
9b
10
11
12
DISTANCE TO
OUTER EDGE OF
PORTS (INCHES)
13-3/4
10-1/2
8-3/8
6-5/8
5
3-1/8
a  Particle Size Sampling Point - "A" Port
b  Particle Size Sampling Point - "BM Port
                   3-11
ENGINEERING-SCIENCE

-------
per point for Run 1, and 5 minutes per point for Runs 2 and 3.  The sam-
pling time was increased for Runs 2 and 3 to increase the total sample
volume, which was necessary because of the low grain loading present at
this site.  Likewise for the particle sizing runs, the first run was 60
minutes long, and Runs 2 and 3 were each 2 hours long.  The particle size
sampling points are shown in Figure 3.4b; all points are points of aver-
age velocity.

Flash Calciner

     Baghouse Inlet

     The baghouse inlet of the Flash Calciner is presented in Figure 3.5a;
the sampling locations for this site are presented in Figures 3.5b.  The
nearest upstream flow disturbance, a bend in the ductwork, is located 120
inches, or 4 diameters from the sampling ports.  The nearest downstrem
disturbance is approximately 4 diameters from the ports.  These distances
dictate a total number of 36 points in order to meet EPA Method 1 criteria.
Samples were collected through two, 4 inch ports, positioned 90° to each
other.  Sampling time was 2 minutes per point, for a total sampling time
of 72 minutes.  However, for Runs 2 and 3 this time was shortened to 36
minutes by sampling every other point.  This action was taken because of
the unusually high grain loading present at this site.  The baghouse in-
let duct from the Flash Calciner is, in fact, the pneumatic conveyor for
the calcined product, and the baghouse is the product recovery system
(likewise for the Herreschoff Furnace baghouse).  Since this site is the
inlet to a control device, it was decided that sampling at every other
traverse point would provide a sufficiently representative sample, while,
at the same time, minimizing Method 5 filter changes and the resultant
time loss.

     Sample time for the particle size runs was 10 seconds for the first
run, 5 seconds for the second run, 3 seconds for the third run, and 2
seconds for the fourth run.  Run 4 was an additional particle size run
collected during the third Method 5 run and was performed to provide an
extra datum point, since the particle size run times were of short dura-
tion at this site.  Particle size sampling points are presented in Figure
3.5b; all points are of an average velocity.

     Baghouse Outlet

     Figures 3.6a and 3.6b present the sampling site configuration and
sampling point locations for the baghouse outlet.  Two, 4 inch ports, 90°
to each other, were installed at a point 2 diameters upstream and 7.7
diameters downstream from the nearest flow disturbances.  This sampling
location necessitates a total of 16 traverse ponts to complete the Method
5 sampling.  Sampling time was 8 minutes per point for the first Method
5 run, and 11 minutes per point for Runs 2 and 3; this increase was in-
corporated to increase the total sample volume for each run, since the
grain loading was low at this, site.  Likewise for the particle size runs
on the baghouse outlet, the first run was 128 minutes long, while Runs 2
and 3 were each 176 minutes of duration.  Particle size sampling points
are shown in Figure 3.6b; all sampling points are of an average velocity.
                                    3-12

-------
                                           FIGURE 3.5a
FLASH CALCINER BAGHOUSE  INLET
^N
t.
12
i
j
12
i
0
1
k 25"
mi:
1
r
1

"BU^."A"


•

•^ —
^

1 |

                             COOLING DUCT
                             - 4" PORTS
                             90° TO EACH OTHER
                           I—-STEEL SUPERSTRUCTURE
                             DIRECTION OF
                             GAS FLOW
                      3-13
ENGINEERING-SCIENCE

-------
                               FIGURE 3.5b
  SAMPLING POINT LOCATION
       FLASH CALCINER
      BAGHOUSE INLET
4 PORTS


II









a
b
M-5 SAMPLING.
POINT
A" & "B" PORTS
1
2
3
4
5b
6
7
8
9
Particle Size
Particle Size
DISTANCE TO
OUTER EDGE OF
PORTS (INCHES)
3.42
4.32
5.25
6.27
7.38
8.64
10.08
11.88
14.46
Sampling Point -
Sampling Point -
M-5 SAMPLING
POINT
"A" & "B" PORTS
10
11
12
13
14
15
16a>b
17 .
18
"A" Port
"B" Port
DISTANCE TO
OUTER EDGE OF
PORTS (INCHES)
21.54
24.12
25.92
27.36
28.62
29.73
30.75
31.68
32.58


               3-14
ENGINEERING-SCIENCE

-------
                                              FIGURE 3.6a
FLASH  CALCINER BAGHOUSE OUTLET
                      >£CATWALK
                                 4 PORTS, 90° TO
                                 EACH OTHER
                                 PIPE NIPPLE AND CAP
                      PLAN
           STACK EXIT
           4 PORTS
         CATWALK
                          O
                      "A"  "B"
                            59"
                            A
•GUARD RAIL
                              7.70
                              TO UPSTREAM
                              DISTURBANCE
                     ELEVATION
                         3-15
   ENGINEERING-SCIENCE

-------
                                     FIGURE 3.6b
    SAMPLING  POINT LOCATION
         FLASH  CALCINER

        BAGHOUSE OUTLET
  4 PORTS
M-5 SAMPLING
POINT
"A" & "B" PORTS
1
2
3
4
DISTANCE TO
OUTER EDGE OF
PORTS (INCHES)
32
30
27-1/4
23-1/2
M-5 SAMPLING
POINT
"A" & "B" PORTS
5
6a»b
7
8
DISTANCE TO
OUTER EDGE OF
PORTS (INCHES)
13
9-1/4
6-5/8
4-3/8
a  Particle Size Sampling Point
b  Particle Size Sampling Point
"A" Port
"B" Port
                   3-16
          .ENGINEERING-SCIENCE

-------
                                CHAPTER 4

                           PROCESS DESCRIPTION
INTRODUCTION

     Source emission tests were conducted on Herreschoff Furnace 2 and
Flash Calciner 4 at the Burgess Pigment Company, Sandersville, Georgia,
from September 26-29, 1983.  The tests were conducted by ES test crew,
headed by Keith Felts.  The process was monitored by Yogesh Doshi of MRI
and William Neuffer of ISB/EPA.  Mark Sack of Burgess Pigment coordinated
testing with the plant personnel and provided process information.  Dennis
Holzschuh of EMB/EPA was present during testing of Flash Calciner 4 to ob-
serve emission testing procedures.
POLLUTANTS/SAMPLING JOINTS

Herreschoff Furnace 2

     Tests on Herreschoff Furnace 2 were completed at the following loca-
tions:

     o  Particulate matter (PM) concentrations, PM mass emission rates,
        and particle size determinations were performed at the inlet of
        the venturi scrubber and the inlet and outlet of the baghouse.
        Only PM concentrations and PM mass emission rate determinations
        were performed at the outlet of the venturi scrubber.  Particle
        size determinations were not possible at this location because
        of the large amount of water droplets in the exhaust gases.

     o  Visible emission (VE) observations were taken at the venturi
        scrubber exhaust stack, the baghouse exhaust stack, and the
        Herreschoff product discharge into the pneumatic conveying sys-
        tem.  The feed discharge on the top of the Herreschoff Furnace
        is totally enclosed, so no observations were made at this point.

     o  Grab samples of the feed and product material were, taken at the
        Herreschoff Furnace inlet and outlet for particle sieve and mois-
        ture content analysis.

Flash Calciner 4

     Tests on Flash Calciner 4 were completed at the following locations;
                                    4-1

-------
     o  PM concentrations, PM mass emission rates, and particle size de-
        terminations were performed at the bagouse inlet and outlet on
        the Flash Calciner.

     o  VE observations were taken at the cooler (serpentine pipe) inlet
        and the baghouse exhaust stack.

     o  Grab samples of the feed material were collected for particle
        size sieve and moisture analysis. • Grab samples of the product
        material (a mixture of material collected by cyclones and a bag-
        house) were also collected for particle size sieve and moisture
        analysis.

     Table 4.1 presents the tests conducted at this plant.
PROCESS DESCRIPTION

     This plant operates continuously except for breakdowns, 24 hours per
day and 7 days per week.  A simplified process schematic is presented in
Figure 4.1.  There are four separate process lines which produce calcined
kaolin.  Lines 1 and 2 use multiple hearth (Herreschoff) furnaces, while
lines 3 and 4 use Flash Calciners.  Raw material is withdrawn from feed
silos and pneumatically conveyed to feed tanks.  Kaolin is conveyed from
the feed tanks and used as feed material for each of the four processing
lines.

Herreschoff Furnace 2

     A simplified flow diagram that shows the emission control systems
for the Herreschoff Furnace lines is shown in Figure 4.2.  The feed ma-
terial is discharged to the top of the Herreschoff by an enclosed screw
conveyor.  The exhaust gas from the Herreschoff is controlled by a ven-
turi scrubber.  The cleaned gas stream leaving the scrubber passes through
a fan to a vent stack.  Calcined kaolin is discharged at the bottom of the
furnace into a pneumatic conveying line (serpentine cooler) that cools the
material.  This conveying stream leads through a baghouse and a fan to a
stack.  All of the calcined product is collected in this baghouse.  From
the baghouse the calcined kaolin is conveyed to finishing operation (see
Figure 4.1), which include screening, size reduction, bulk loading, bag-
ging, and bag loading.

     Herreschoff Furnace 2 was manufactured by Nichols.  This calciner
is fired on natural gas, and the maximum calcining temperature is 1093°C
(2000°F).'  The calciner has a retention time of about 30 minutes.  The
kaolin processed contains less than 1% surface moisture before calcining.
The calciner removes the small amount of surface moisture and approxi-
mately 14% chemically bound water in the material.  Table 4.2 presents
the design and operating parameters for Herreschoff Furnace 2.  Table
4.3 presents the design and operating parameters for the venturi scrub-
ber and the baghouse.
                                    4-2

-------
                                TABLE 4.1

           EMISSION TESTS CONDUCTED AT BURGESS PGIMENT COMPANY
        Sampling Point
        Test Type
 Test
Method
Herreschoff Furnace 2a

Herreschoff Feed Inlet


Herreschoff Product Outlet



Scrubber Inlet


Scrubber Outlet

Baghouse Inlet


Baghouse Outlet



Flash Calciner 4b

Flash Calciner Feed Inlet


Flash Calciner Product0


Cooler (Serpentine Pipe) Inlet

Baghouse Inlet


Baghouse Outlet
Moisture Content              ASTM
Particle Size Sieve           ASTM

Visible Emissions             EPA-22
Moisture Content              ASTM
Particle Size Sieve           ASTM

Particulate Concentration     EPA-5
Particle Size                 Andersen

Particulate Concentration     EPA-5

Particulate Concentration     EPA-5
Particle Size                 Andersen

Visible Emissions             EPA-9
Particulate Concentration     EPA-5
Particle Size                 Andersen
Moisture Content              ASTM
Particle Size Sieve           ASTM

Moisture Content              ASTM
Particle Size Sieve           ASTM

Visible Emissions             EPA-22

Particle Concentration        EPA-5
Particle Size                 Andersen

Visible Emissions             EPA-9
Particle Concentration        EPA-5
Particle Size                 Andersen
a  Tests consisted of three runs.  All tests at each location were per-
   formed simultaneously during each run.
b  Test consisted of two good runs.  All tests at each location were per-
   formed simultaneously during each run.
c  Flash Calciner product samples are mixtures of material collected by
   cyclones and a baghouse.
                                    4-3

-------
                                                                          FIGURE 4.1

                                                                        To Srack
                                                                          i


Cyclone
'.
— *j Sagnouie
r ?
L-€>
Ou,r ^
                                                                                 3
                                                                                 5
                                                                                <5
                                                                                 3k
                                                                                Ib.
                                                                                 T
                                                                              Product
                                                                              Shipment?
Nofe: P.C. 3 Pneumafically Conveyed
                  Simplified process flow diagram  for  Burgess Pigment
                       Company,  Sandersville, Georgia.
                                     4-4
ENGINEERING-SCIENCE, INC.

-------
                                                  TO VENT
    FEED
INDUCED

AMBIENT

  AIR
                              EXHAUST GAS

                              I	
                            BLOWER
                PRODUCT
                                WET

                            SfRIIRRFR
                                                     I
                                                  I	1
                                        WASTEWATER

                                        TO SETTLING

                                          POND
                                                                                     TO VENT
                                                                     COOLER
                                                                                                    BLOWER


                                                                                                       PRODUCT
*•

Ul
m
z
«
rn
m
5

z
o

C/)
o

m
z
o
m

z
o
_^ PROCESS FLOW


> AIR  FLOW
             FEED —W
                                              COOLED AIR TO

                                              ATMOSPHERE
 NATURAL GAS
CYCLONE
COLLECTED OUST
c
400°F
i-^{-
BLOTJ
INDUCED
AMBIENT
•"N| r^; '"•
li'1 : '
1 M ] | i
. ! 1 i I I BAG
--W-' ^-' -' — > 27
) T COOLER
ER i
_.J
AIR
                                     AIR BLOWER  AIR BLOWER
                       FLOW DIAGRAM FOR  EMISSION CONTROL SYSTEMS FOR CALCINERS AT

                       BURGESS PIGMENT COMPANY,  SANDERSVILLE,  GEORGIA.
                                                                                                            BLOWER
                                                                                                           PRODUCT
                                                                                                                      o
                                                                                                                      G

-------
                                TABLE 4.2
                         SUMMARY OF CALCINER DATA
                          Herreschoff Furnace 2
                            Flash Calciner 4
Normal Feed Rate, tph

Hours of Operation

Retention Time

Maximum Calcining
Temperature, °F

Heat Application Method

Fuel

Feed Moisture, %

Product Moisture, %
5

24/day

30 minutes


2000

Countercurrent

Natural gas, kerosenea

1% surface, 14% bound

0
4

24/day

0.5-2 seconds


2400

Cocurrent

Natural gas, kerosenea

1% surface, 14% bound

0
a  Both calciners are usually fired with natural gas.
                                    4-6

-------
                                TABLE 4.3
                    SUMMARY OF CONTROL EQUIPMENT DATA

Manufacturer
Model
Inlet Gas Volume, acfm
Inlet Gas Temp., °F
Design Pressure Drop, in. w.c.
Bag Material
Air-to-Cloth Ratio
Number of Compartments
Cleaning Method
Stack Height, ft
Stack Diameter, ft
Herreschoff
Furnace
Venturi
Scrubber
AAF
Kinpactor
size 12
7,100
140
12
NAa
NA
NA
NA
•^30
^2
Herreschoff
Product
Cooler
Baghouse
Norblo
Unknown
26,000
200-275
10-17
Polyester
Felt
1.1:1
16
Reverse-air
^30
^2
Flash
Calciner
Baghouse
AAF
Amer therm
29-112-9660
28,000
240
Unknown
Polyester
Felt
1 .6 to 1 .1 :1
6
Reverse-air
33
2.8
a  NA = not applicable.
                                    4-7

-------
Flash Calciner 4

     A simplified flow diagram that shows the emission control systems
for the Flash Calciner lines is shown in Figure 4.2.  The feed material
is pneumatically conveyed to the calciner.  Exhaust gas from the calci-
ner is passed through two heat exchangers (gas-to-gas) to a cylcone that
removes a portion of the product.  Heat exchanger 2 was not operating
during the tests.  Following the cyclone, the gas stream passes through
a pneumatic conveying line (serpentine cooler) to a baghouse and an in-
duced draft fan and is then vented through a discharge stack.  Calcined
product, collected in both the cyclone and the baghouse, is then con-
veyed to finishing operations.

     The normal feed rate for Flash Calciner 4 is 3.6 Mg (4 tons) per
hour.  The calciner is fired on natural gas and the maximum calcining
temperature is 1371°C (2500°F).  The calciner has a retention time of
0.5 to 2 seconds.  The kaolin processed contains about 1% surface mois-
ture before calcining.  The calcincer removes the small amount of sur-
face moisture and about 14% of the chemically bound moisture in the
material.  Tables 4.2 and 4.3 present the design and operating param-
eters for the Flash Calciner and its baghouse, respectively.
PROCESS CONDITIONS DURING TESTING

     Note:  Tables 4.4, 4.5, and 4.6 and Tables 4.10, 4.1'i, and 4.12,
process data for the Herreschoff Furnace 2 and Flash Calciner 4, respec-
tively, have been deleted from this report for confidentiality purposes,
at Burgess Pigment's request, and will be stored under lock and key pend-
ing a review by the Administrator as to the legitimacy of Burgess1 claim
on confidentiality.

Herreschoff Furnace 2

     All process conditions were normal during the testing, except the
Herreschoff was shut down for a period of 8 minutes during Run 2 due to
problems with the product screw conveyor.  Testing was discontinued for
this period.  Temperatures at various hearths in the Herreschoff were
monitored from a control panel.  The fuel consumption was monitored by
reading a natural gas meter.  The feed rate was measured several times
during each test by manually collecting the feed in a bag for 10 seconds
and then weighing the collected feed.  Tables 4.4, 4.5, and 4.6 present
the process data recorded during the tests (see Confidential Addendum).
A shown in these tables, the process parameters were faily uniform dur-
ing the tests, and therefore process conditions were normal.  The feed
rate and temperatures were close to design.

     The venturi scrubber operation was monitored by reading the pressure
drop across the venturi, the pump inlet vacuum, the pump outlet pressure,
and the pump power consumption.  A flow meter was installed on September
27, 1983, to monitor scrubber water flow rate during Runs 2 and 3.  The
baghouse operation was monitored by reading pressure drop across the bag-
house and observing pressure columns that reflect cleaning mechanisms of
individual compartments.  Tables 4.7, 4.8 and 4.9 present the control
equipment data recorded during the test.

                                    4-8

-------
            TABLE 4.7

HERRESCHOFF CONTROL EQUIPMENT DATA
         RUN 1  (9/26/83)
Scrubber
Time
15:30
15:45
16:00
16:15
16:30
16:45
17:00
17:15
17:30
17:45
18:00
18:15
18:30
18:45
AP,
in.
w.c.

13.4
13.4
13.0
12.8
13.0
13.0
12.8
13.0
12.8
13.0
13.0
12.8
13.0
Pump Outlet
Pressure,
psig
26
26
26
26
26
26
26
26
26
26
26
26
26
26
Pump Inlet
Vacuum,
in. Hg
4
4
4
4
4
4
4
4
4
4
4
4
4
4
Pump
Power ,
kw
3.6
3.6
3.6
3.6
3.6
3.6
3.6
3.6
3.5
3.5
3.5
3.5
3.5
3.5
Baghouse
AP,
in.
w.c .
6.0
6.8
5.2
5.2
5.5
5.4
5.2
5.4
5.4
5.2
5.2
5.2
5.2
5.2
Compartment
Cleaning
O.K.
O.K.
O.K.
O.K.
O.K.
O.K.
O.K.
O.K.
O.K.
O.K.
O.K.
O.K.
O.K.
O.K.
                  4-9

-------
                                TABLE 4.8

                    HERRESCHOFF CONTROL EQUIPMENT DATA
                             RUN 2a (9/27/83)
Time
10:00
10:15
10:30
11 :00
11 :15
11 :30
1 1 :45
12:00
12:15
12:30
12:45
13:00
13:15
13:30

AP,
in.
w.c.
13.2
12.6
12.6
12.6
12.4
12.4
12.2
12.4
12.4
12.2
12.4
12.0
11 .6
11 .6

Water
Meter,
gal
5240
6110
6900
8640
9620
10390
1 1 260
1 2420
1 3020
1 3770
1 4670
1 5640
16370
1 71 40
Scrubber
Pump
Outlet
Pressure,
psig
36
36
36
36
36
36
36
36
36
36
36
36
36
36

Pump
Inlet
Vacuum,
in. Hg
2
2
2
2
2
2
2
2
2
2
2
2
2
2

Pump
Power,
kw
3.2
3.2
3.2
3.2
3.2
3.2
3.2
3.2
3.2
3.2
3.2
3.2
3.2
3.2
Baghouse
AP,
in.
w.c.

•

5.2
5.6
6.0
5.0
5.0
6.0
5.0
5.4
5.2
5.2
5.4
Compartment
Cleaning
O.K.
O.K.
O.K.
O.K.
O.K.
O.K.
O.K.
O.K.
O.K.
O.K.
O.K.
O.K.
O.K.
O.K.
a  Herreschoff was shut down from 11:50 to 11:58 due to problems with
   product screw conveyor.
                                    4-10

-------
            TABLE 4.9

HERRESCHOFF CONTROL EQUIPMENT DATA
         RUN 3 (9/27/83)
Time
14:45
15:00
15:15
15:30
15:45
16:00
16:15
16:30
16:45
17:00

P,
in.
w.c.
13.2
13.2
13.0
12.4
12.6
12.2
12.4
12.4
12.2
12.2

Water
Meter,
gal
21010
21930
22690
23480
24270
25090
25890
26660
27460
28240
Scrubber
Pump
Outlet
Pressure,
psig
35
35
35
35
35
35
35
35
35
35

Pump
Inlet
Vacuum,
in. Hg
2
2
2
2
2
2
2
2
2
2

Pump
Power,
kw
3.2
3.2
3.2
3.2
3.2
.3.2
3.2
3.2
3.2
3.2
Baghouse
P,
in.
w.c.
5.0
5.0
5.2
5.2
5.2
5.5
5.2
5.2
5.0
5.2
Compartment
Cleaning
O.K.
O.K.
O.K.
O.K.
O.K.
O.K.
O.K.
O.K.
O.K.
O.K.
                4-11

-------
Flash Calciner 4

     The No. 2 heat exchanger is old and will be replaced in 1984.  The
product was pneumatically conveyed through the No. 2 heat exchanger even
though it was not operating during all three runs.  The product quality
and the overall process were not affected because heat exchanger 1 was
operating during all three runs.

     During Run 1,  excessive visible emissions were noted by ground ob-
servers from the baghouse exhaust stack.  The certified recorder read
predominantly 0% from the top of a silo.  Plant personnel felt that the
baghouse was not operating correctly and volunteered to shut down the
process and to inspect the baghouse prior to Run 2.  The baghouse was
checked the next morning and two bags were found loose.  The problem
in the baghouse was corrected prior to Runs 2 and 3.  All processes
were operated normally during Runs 2 and 3.

     The calciner operation was monitored from a control panel that
contains temperature gauges for precombustion air, calciner inlet, cal-
ciner outlet, No. 1 heat exchanger outlet, No. 2 heat exchanger outlet,
and baghouse inlet..  Production rate was monitored by measuring product
silo levels.  Feed rotary valve rpm and feel blower pressure were moni-
tored from the control panel.  The fuel consumption was monitored by
reading a natural gas meter.  Tables 4.10, 4.11, and 4.12 present the
process data recorded during the tests (see Confidential Addendum).

     The baghouse operation was monitored by measuring pressures at main
fan intake and serpentine pipe outlet.  The difference in these pressures
indicate pressure drop across the baghouse.  Tables 4.13, 4.14, and 4.15
present the control equipment data recorded during the tests.
                                    4-1 2

-------
                               TABLE 4.13

                      FLASH CALCINER BAGHOUSE DATA
                            RUN 1a (9/28/83)
Main Fan Intake
Time Pressure, in. w.c.
15:30
15:45
16:00
16:15
16:30
16:45
17:00
17:15
17:30
17:45
18:00
18:15
18:30
18:45
19:00
19:15
9
9
8.5
8.5
8.5
8.5
8.5
9.0
8.5
8.5
8.5
8.5
8.5
8.5
8.5
8.5
Serpentine Pipe
Outlet, in. w.c.
5.5
5.5
5.5
5.0
5.0
5.0
5.5
5.5
5.5
5.0
5.5
5.5
5.5
5.5
5.5
5.5
Pressure Drop,
in. w.c.
3.5
3.5
3.0
3.5
3.5
3.5
3.0
3.5
3.0
3.5
3.0
3.0
3.0
3.0
3.0
3.0
a  There were two loose bags in the baghouse during the run.
                                    4-13

-------
         TABLE 4.14

FLASH CALCINER BAGHOUSE DATA
      RUN 2 (9/29/83)
Main Fan Intake
Time Pressure, in. w.c.
10:00
10:15
10:30
10:45
11 :00
11:15
11:30
11 :45
12:00
12:15
12:30
12:45
13:00
13:15
13:30
9.0
8.5
9.0
9.0
8.5
8.5
8.5
8.5
8.5
8.5
8.5
8.5
8.5
3.5
8.5
Serpentine Pipe
Outlet, in. w.c.
5.5
5.5
5.5
5.5
5.0
5.0
5.5
5.0
5.5
5.5
5.0
5.5
5.5
4.5
5.0
Pressure Drop,
in. w.c.
3.5
3.0
3.5
3.5
3.5
3.5
3.0
3.5
3.0
3.0
3.5
3.0
3.0
4.0
3.5
              4-14

-------
         TABLE 4.15

FLASH CALCINER BAGHOUSE DATA
      RUN 3 (9/29/83)
Main Fan Intake
Time Pressure, in. w.c.
14:30
14:45
15:00
15:15
15:30
15:45
16:00
16:15
16:30
16:45
17:00
17:15
17:30
9.0
8.5
8.5
8.5
8.5
8.5
8.5
8.5
8.5
8.5
8.5
8.5
8.5
Serpentine Pipe
Outlet, in. w.c.
5.5
4.5
5.5
5.5
5.0
5.5
5.5
5.0
5.0
5.0
5.0
5.5
5.5
Pressure Drop,
in. w.c.
3.5
4.0
3.0
3.0
3.5
3.0
3.0
3.5
3.5
3.5
3.5
3.0
3.0
              4-15

-------
                                CHAPTER 5

                    SAMPLING AND ANALYTICAL PROCEDURES
                        DATA REDUCTION PROCEDURES
METHOD 5

Sampling Equipment and Procedures

     The procedure and apparatus employed for Method 5 testing were as
defined in EPA Reference Methods 1 through 5.  Figure 5.1 illustrates
the Method 5 sampling train employed.

     All sampling probes consisted of a stainless steel outer sheath with
a heated borosilicate glass liner.  Each probe was equipped with a type
"S" pitot tube for measuring gas velocity and a type "K" thermocouple for
determining flue gas temperature.  Probe assemblies were also equipped
with a 1/4-inch diameter gas sampling probe for collection of integrated
gas samples used in the determination of %02/ %C02/ and dry gas molecular
weight.  Type "S" pitot tubes were examined prior to conducting the tests
to verify proper alignment of the face openings and conformance with the
dimensional criteria specified under EPA Reference Method 2.  Only pitot
tubes meeting the specified criteria were used and these were assigned
a pitot tube coefficient of 0.84.  Velocity head pressures were measured
using an oil manometer having an inclined scale of 0 to 1 inches H2O and
a full-scale of 10 inches H20.

     Stack gas carbon dioxide, oxygen, and nitrogen content were deter-
mined using an Orsat apparatus.  Integrated Orsat samples were collected,
per EPA Reference Method 3 procedures, using a pump which exhausted into
a Tedlar® bag.

     Leak checks were conducted in accordance with procedures described
in EPA Methods 3 and 5 for Orsat and particulate sampling apparatus, re-
spectively.  Leak checks were performed on the Orsat analyzer before each
analysis.  The Method 5 sampling train was leak checked before and after
each run and before and after each filter change.  The pitot tube and
lines were leak checked prior to and at the conclusion of each test.
All leak checks were successful.

     The sampling train incorporated a thermocouple for measurement of
the temperature of the gas stream inside the filter holder.  This temper-
ature was recorded a.s the sample box temperature on the field data sheet.
Probe heat and sample box heat were regulated manually using variacs in
order to maintain temperatures within the range of 120 + 14°C.
                                    5-1

-------
                      Particulate  Sampling  Train
                                                                               CYCLONE
THERMOCOUPLE
PI TOT TUBE
PROBE 	
THERMOCOUPLE
PROBE 	
PITOT TUBE
STACK WALL
PITOT MANOMETER
HEATED AREA
THERMOMETERS
ORIFICE
 FILTER HOLDER
  THERMOCOUPLE
IMPINGER TRAIN
DRY  GAS METER
     IMPINGER
     ICE BATH
   VACUUM LINE
  VACUUM GUAGE
   MAIN VALVE
 BY-PASS VALVE
AIR-TIGHT PUMP

-------
     Thermometers, nozzles, and meter boxes were calibrated before test-
ing according to procedures outlined in EPA Method 5.  At the conclusion
of the testing after returning equipment from the field, a post-test
meter box calibration was performed.  Calibration data is presented in
Appendix G.

Sample Recovery

     At the conclusion of the test after the final leak check, the probe
was disconnected from the rest of the train and capped.  Filter and im-
pinger inlets and outlets were also capped, and the probe and impinger
box were returned to the sample recovery area for clean up.

     The probe and nozzle were brushed and rinsed a minimum of three
times with acetone.  Two team members were designated for the remainder
of the clean up.  The filter and remaining glassware were removed from
the box and replaced with a pre-prepared second set to minimize sample
train turnaround time.  Individuals so designated recovered the rest of
the samples and prepared for the next runs.

     The filter was removed from the glass housing and placed in its
correspondingly numbered glass petri dish.  The front half of the filter
holder was brushed, and rinsed with acetone and the rinse was placed into
the appropriate probe wash container.  The contents of the first three
impingers were transferred to a graduated cylinder and the volume mea-
sured and noted.  The impingers, back half of the filter holder, and
connecting glassware were rinsed with distilled water into the graduated
cylinder.  The contents of the graduated cylinder were then transferred
into a sample bottle.  All glass sample containers were labeled identi-
fying source, date of sampling, type of sample (i.e., front half probe
rinse, back half), and run number.  Silica gel from the fourth impinger
was transferred to the original plastic bottle(s) and weighed on a tri-
ple beam balance to the nearest 0.1 g.

Laboratory Procedures

     Each run generated one acetone probe rinse, one filter or more if
the pressure drop across the filter was too high, and one impinger con-
tents and rinse sample for potential metals analyses.  At the conclusion
of the testing, all samples were taken by ES personnel to the ES labora-
tory for gravimetric: analysis.  Analytical methods and laboratory proce-
dures are discussed below.

     Probe and Front Half Filter Holder Acetone Rinse

     Clean, numbered 250 ml beakers are desiccated for a minimum of 24
hours and weighed to a constant weight.  Beakers are stored in a desic-
cator prior to use.

     The volume of each acetone rinse is determined gravimetrically,
using a triple beam balance and the following equation:

                        - Wf)
                        n= Volume of Acetone
                                    5-3

-------
where:  W^ = weight of sample bottle and rinse,  (g)
        Wf = weight of sample bottle empty  (after acetone rinse has been
             transferred to beaker), (g)
        D  = density of acetone, (g/ml)

     The contents of each acetone rinse bottle is transferred to a pre-
viously tared and numbered 250 ml beaker and evaporated to dryness.  Each
bottle is rinsed with a known quantity of acetone, preferably from the
same lot as the sample.  If acetone from another lot is used, this neces-
sitates the determination of another blank  residue value.  The dry beakers
are desiccated for at least 24 hours and weighed to a constant weight.  Re-
sults are reported to the nearest 0.1 mg.   Acetone used is reagent grade.

     A sample of acetone is collected at each test site from each lot
used during sampling.  These are labeled "Acetone Blank" and are handled
in the same manner as the acetone rinses.   Acetone Blank values are re-
ported to the nearest 0.1 mg.

     Filters
     Filters, labeled on the back side near the edge, were visually
checked for flaws, irregularities, or pinhole leaks.  A series of glass
petri dishes were marked with the same numbers as the filters and the
respective filters placed in the dishes.  Each filter/petri dish was
thereafter handled a.s a unit.

     The filter/petri dish sets were desiccated at 20 +5.6°C and ambient
pressure for at least 24 hours and weighed at six-hour intervals to a
constant weight.  Constant weight is defined as a difference of 0.5 mg
or a 1 % difference between the values obtained by subtracting the tare
from the gross weight, for two consecutive weighings with at least six
hours of desiccation between weighings.  The greater value thusly ob-
tained is reported as the constant weight.  After sample collection, the
filters were desiccated and weighed in the same manner as above and re-
sults reported to the nearest 0.1 mg.  Filters and dishes were handled
only with forceps or with latex gloves.

     Impinger Solutions

     The condensate and back half water rinse were returned to the Fair-
fax Laboratory for metals screening.
PARTICLE SIZING

     For all particle size sampling and analysis, the procedures pre-
scribed in the IERL document, "Procedures for Cascade Impactor Calibra-
tion and Operation in Process Streams - Revised  1979", and Mr. Peter
Westlin's memo dated August 11, 1983, were adhered to, with the excep-
tion that the particle size samples were not collected at the four point
traverse defined in the guidelines but were collected at a single point
of average velocity.  This deviation had the effect of changing the num-
ber of particle size samples per particle size run from as many as four
to one.
                                    5-4

-------
Sampling Equipment and Procedures

     Particle size determinations were conducted using the Andersen Mark
III high temperature, six-stage, inertial cascade impactor.  The Andersen
is a multi-stage, multi-jet impactor for stack particulate effluents.  It
aerodynamically and automatically classifies particulate into multiple
size ranges and accounts for all their physical properties; namely size,
shape, and density.  ES had available six Flow Sensor preseparators for
use as a precutter with the impactor.  Straight nozzles were used for
sampling; nozzle sizes were selected to insure that impactor flow rates
were in the range of 0.8 to 1.6 liters per minute (1pm).

     Samples were extracted with a standard Method 5 meter box, backed up
with an impinger assembly to condense moisture and protect the meter box
and pump assembly  (see Figure 5.2).  Velocity determinations were made
prior to each series of particle size runs using an "S" type pitot tube
and inclined monometer.  Ap readings were made at the traverse points of
average velocity (sampling points).  The sampling rate for the impactor
was calculated using a pre-programmed Texas Instruments 58C calculator,
in lieu of a Method 5 type nomograph.  The sampling rate was calculated
to be isokinetic to the velocity and temperature conditions at the spe-
cific sampling point.  Initial estimates for the sampling time was based
on preliminary runs conducted at each site.  This sampling time was used
for the first day Scimples and was "fine tuned" with each subsequent sam-
ple.

     Positioning of the impactor in the duct at the specified sampling
points was accomplished with an iron pipe to which the impactor was at-
tached.  Incorporateid in this probe assembly were two thermocouplers for
meauring the stack gas temperature and impactor temperature.  Impactor
temperature is a measure of the sample gas as it passes through the final
filter of the impactor.  Note that a right angle preseparator was used
to insure that no right angle bends were effected upon the sample gas
until after being drawn through the impactor.  This insures no secondary
deposition of particles.  In those cases where no auxiliary heating was
required, sampling was not initiated until the impactor temperature had
equilibrated to within 3°C of the stack gas temperature..  During this ini-
tial impactor heat-up, the impactor was positioned so that the nozzle was
not facing directly into the stack gas flow.  In those cases requiring
auxiliary heating  (the Flash Calciner baghouse outlet), sampling was not
initiated until the impactor temperature was 15 to 30°C higher than the
flue gas temperature.  This procedure was followed at this site because
a portion of the impactor extended outside the stack during sampling.

     Prior to sampling, the particle size sampling assembly was leak
checked at 5 inches of mercury with an allowed leak rate not to exceed
0.02 acfm.  No post-test leak checks were performed, as such action might
cause disturbance a:nd possible reentrainment of particles collected.  All
threaded joints were sealed with Teflon® tape prior to leak checks.

     A single pretest leak check was conducted during this field test
for each impactor sample, in lieu of the two leak checks specified in
the IERL/EMB guidelines.  It was felt that the low stack temperatures
did not warrant the additional leak check after the impactor had heated
in the duct.

                                    5-5

-------
                                     FIGURE 5.2
ARTICLE SIZE SAMPLING TRAIN
   PROBE TUBE
             IMPACTOR
                                DRY GAS METER
                      NOZZLE-
                  5-6
                                ENGINEEKING-SCISNCls

-------
     Moisture and dry gas molecular composition were obtained from the
Method 5 and gas composition analyses.  Stack gas velocity was calculated
from these data along with the stack temperatures and Ap's recorded at
the impactor sampling points.

     Incorporated in the IERL post-test procedure is a purge of the im-
pactor with clean, dry ambient air to prevent condensation of moisture
in the impactor prior to substrate recovery.  Purging was not conducted
at any of the Burgesss Pigment sites because of the unusually high ambient
air dust levels and the concern that this might contaminate the samples.

Sample Recovery

     After collection of a particle size sample, the impactor was trans-
ported to the field laboratory for recovery.  The recovery was conducted
according to the procedures delineated in the IERL/EMB particle sizing
documents.

     The impactor was disassembled and the collection substrates folded
and placed into the tared aluminum foil liners and stored in individual,
pre-labeled petri dishes.  Particulate matter adhering to the crossbar
gaskets and stage plates were brushed onto the appropriate substrate.
Any material found on the upstream side of an impactor plate was included
with the substrate mounted on that plate.  Material found on the down-
stream side of the plate was included with the following substrate.  The
nozzle, inlet cone, and preseparator were brushed with a fine camel's
hair brush.  This catch was combined with the zero stage catch and hence,
handled as a single weight.

     As a part of the pretest measurements, a single impactor run was
conducted at each site with a pre-filter attached to the front of the
impactor.  The purpose of this run was to evaluate potential substrate
weight gains due to acid gas uptake.  This prefilter was tare weighed
prior to sampling, and was weighed again following sampling.  The in-
formation gained thusly allowed calculation of the grain loading at
each site.  This information was used, in turn, to aid the selection
of the proper particle sizing sampling time.

     An additional part of the pretest measurements was the performance
of a handling blank.  For these samples, the impactor was loaded with
substrates, and the loaded impactor was carried to the sampling site.
There, the impactor was mounted on the probe and was then inserted in
the stack and allowed to come to temperature.  Afterwards, the impactor
was disconnected from the probe and returned to the field lab for sub-
strate recovery.  Ha.ndling blanks were performed to evaluate potential
substrate weight gains or losses due to the physical manipulations of
the substrates when collecting particle sizing samples.

Laboratory Analysis

     The glass fiber collection substrates were pre-weighed at the ES
Fairfax Laboratory using a Cahn® Model 21 electrobalance.  Circular,
aluminum foil discs were cut for each filter and included in the initial
tare.  Filters were desiccated and weighed until two successive weights
                                    5-7

-------
were obtained with no more than 0.05 mg difference (constant weight).
The minimum time between successive weighings was six hours.

     All exposed substrates, excepting blanks, were post-weighed at the
ES Fairfax Laboratory.  All laboratory weighings were conducted to ob-
tain a constant weight (i.e., equal or less than 0.05 mg weight differ-
ence between successive weighings, with a six-hour desiccation time be-
tween weighings).  All weighings were made with a Cahn® balance capable
of precision to 0.01 mg.

     Each substrate and final filter was transported to and from the Bur-
gess Pigment facility in a numbered plastic petri dish.  The aluminum
foil disc was used to form an envelope to retain any particulate matter
collected on that substrate during sampling.

     Whatman® 934-AH substrates were used for this test.  This material
has displayed the minimum reaction to acid gas exposure and the resultant
weight gains.  These substrates were not pretreated per the procedures
defined in the IERL document "Procedures for Cascade Impactor Calibration
and Operation in Process Streams - Revised 1979."  However, blanks were
run to investigate potential weight gains or losses.  Subsequent analysis
of these blank substrates indicated that they were not reactive.  Ambi-
polyvinyl gloves were used when handling the substrates and filters.
ANCILLARY TESTS

Visible and Fugitive Emissions Observations

     As a requirement of the test program, visible emissions observations
(VEO's) were made at. three control device discharge points  (stack) as well
as the finished product transfer points  (pneumatic conveyors).  Method 9
was used to determine opacity levels from the three stacks.  VEO's were
made for the duration of the concurrent Method 5 run.  The  proposed EPA
Method for fugitive dust, Method 22, was used for all product transfer
fugitive dust observations.  Length of observations were as for the stack
VEO's.  In those instances where fugitive emissions were observed, an es-
timate of the opacity was provided by the observer.

     Opacity readings were recorded every 15 seconds, except during in-
terruptions in production or testing, or when it was necessary for the
observers to rest their eyes.  The observers locations were selected to
provide clear views of the emissions without interference from the sun,
and a good line of vision approximately perpendicular to the plume di-
rection with a good background for observation.  Wind shifts at times
prevented continuous; observation and the data sheets so indicate.  Ad-
verse conditions were not serious enough to preclude making sufficient
observations for each run.

Product Sampling

     Grab samples of raw and finished products were collected during each
Method 5 run.  Samples were collected of these materials for moisture and
                                    5-8

-------
sieve analysis.  Moisture content analysis was performed at the ES Fair-
fax Laboratory.  Sieve analysis was performed by an outside laboratory.
DATA REDUCTION AND CALIBRATION

Data Reduction

     Method 5 data reduction was accomplished per the equations presented
in the Method 5 procedures.  ES has on file a computer program which per-
formed the computations necessary to supply grain loading, emission rates,
etc., for each run.  Isokinetic calculations were performed for each run
at the completion of sample recovery to insure that sampling was performed
within the limits imposed by Method 5.

     Particle size data reduction was performed using the PADRE computer
program, as requested by EPA.

Calibration

     The temperature recording devices, dry gas meters, and nozzles were
calibrated per EPA Methods 2 and 5 procedures.  Calibration data can be
found in Appendix G.

     The impactor plates were checked for hole diameter prior to the test
using an optical microscope and above stage, fiber optic, light source.
All impactor plates were cleaned using acetone and an ultrasonic cleaner
prior to checking the hole diameters.  The results of this check can be
found in Appendix G of of this report.  The criterion for tolerance was
+_ 10% of hole area.  All plates met the tolerance criterion.

     As part of the standard weighing procedure, a Class "S" standard
weight is check weighed prior to weighing each set of substrates.  In
this way, subsequent, weighings can be corrected for balance drift.
                                    5-9

-------
                                CHAPTER 6

                            QUALITY ASSURANCE
     Engineering-Science (ES) has incorporated in all its source sampling
work a quality assurance (QA) program based on the EPA "Quality Assurance
Handbook for Air Pollution Measurement Systems," Volume III (EPA 600/4-7-
027b).  Trained, qualified personnel were employed for the performance of
all sampling and analysis associated with this project.

EPA Method 1-5 (Gross Particulate Matter)
     All particulate and gaseous sampling equipment was calibrated prior
to and at the completion of the field test.  Thermometers and thermocou-
ples, thermocouple readouts, pitot tubes, and dry gas meters were cali-
brated.  The wet test meter employed for calibrating the dry gas meters
is periodically checked using a spirometer.  Calibration data are pro-
vided in Appendix H.

     ES regularly engages in the EPA QA audit program for EPA Method 5
dry gas meter calibrcitions.  The analytical balance used for filter/res-
idue weighings is semi-annually checked and overhauled (if needed) by
a certified technician.  The balance is checked with class "S" standard
weights before each weighing session.  Other procedures employed by ES
to promote QA include:

     o  Conduct all tests in triplicate (three runs).
     o  Collect integrated samples for Orsat analysis.
     o  Maintain probe and filter temperature at 1 20 +^ 14°C.
     o  Conduct two to three hour tests for each Method 5 run.
     o  Conduct sampling 100 +_ 10% of isokinetic.
     o  Perform leak checks on all sampling equipment as prescribed in
        the appropricite EPA method.

Particle Size Sampling

     As with the particulate matter sampling equipment, the particle size
sampling equipment was calibrated prior to and at the completion of the
field test.  Additionally, the impactor plates were visually checked for
hole diameter using an optical microscope and calibrated reticules.  These
calibration data are included in Appendix H.  Additionally, ES incorpor-
ates the following QA procedures in the field and laboratory:

     o  Conduct a blank particle size run for potential acid gas uptake.
                                    6-1

-------
     o  Conduct a particle size handling blank run as a check on lab/
        field technique.

     o  Weigh first ran particle size substrates as a check on proper
        substrate loading/sampling time.

     o  Add auxiliary heat to impacts,  if needed, to prevent moisture
        condensation.

     o  Conduct particle  sizing runs during Method 5 sampling, providing
        a cross check on  grain loadings.

     o  Preheat impactor  either in the duct or with auxiliary heat to
        prevent condensation and assure consistent gas viscosity during
        sampling.

     o  Conduct sampling  at 1 00 +_ 20% isokinetic.

     o  Check weigh substrates to assure constant weights.

     o  Check balance with standard weights.

     Personnel were assigned to a given sampling site for the duration of
the test program.  Two laboratory technicians were assigned to conduct all
field sample recovery and analysis.  Assignment of personnel to specific
locations/duties assures  consistent performance by each team member.  All
visible emissions observers had been certified within the past six months
of the field test.

     Chain of custody of  samples is documented by sample number.  Except
for process samples, all samples were kept in ES custody for the duration
of the testing.

     All filters, substrates, probes, and sampling trains were capped
while in transit between  the field lab and the sample site  to prevent
potential contamination.
                                    6-2

-------