TEST NUMBER 73-ROC-l
INTERNATIONAL MINERALS
AND CHEMICAL CORP.
KINGSFORD, FLORIDA
PEDCo ENVIRONMENTAL
-------
PEDCo-EN VI RON MENTAL
SUITE 8 - ATKINSON SQUARE
CINCINNATI. OHIO 45 3 4 G
513 1-7-7 1-4330
TEST NUMBER 73-ROC-l
INTERNATIONAL MINERALS
AND CHEMICAL CORP.
KINGSFORD, FLORIDA
February, 1973
Prepared By
Norman Kulujian, P.E.
Richard W. Gerstle, P.E.
Contract No. 68-02-0237
Task No. 12
-------
I. TABLE OF CONTENTS
Subject Page
II. INTRODUCTION 1
III. SUMMARY OF RESULTS 3
IV. PROCESS DESCRIPTION • 10
V. PROCESS OPERATION 12
VI. LOCATION OF SAMPLING POINTS 14
VII. SAMPLING AND ANALYTICAL PROCEDURES 20
VIII. APPENDIX 27
A. Complete Particulate Results with
Example Calculations.
B. Complete Fluoride Results with
Example .Calculations.
C. Process Operational Log.
D. Field Data.
E. Standard Sampling Procedures.
F. Laboratory Report.
G. Sample Number Log.
H. Test Log.
I. Sampling Handling Log.
J. Project Participants and Titles.
K. Presurvey Report.
-------
II. INTRODUCTION . •' ' .."/. " -'.;;; . •;••'•••.'.
Under the Clean Air Act of. 1970, as 'amended, the Environ-
mental Protection Agency is 'charged with the establishment
of performance standards for stationary sources which may
contribute significantly to air pollution. A performance
standard is based on the best emission reduction systems
which have been shown to be technically and economically
feasible. .. . ''.-.' . ,
In order to set realistic performance standards, accurate
data on pollutant emissions must be gathered from the stationary
t '
source category under consideration.
Atmospheric emissions of particuJate and fluorides from
the International Minerals and Chemical Corporation (IMC)
phosphorus plant dryer in Kingsford, Florida were sampled to
establish a guide for New Source. Performance Standards as
authorized by the Clean Air Act of 1970. Triplicate tests were
made to determine particulate and fluoride concentrations at the
inlet and outlet of the dryer scrubber. The six tests were made
February 12-15, 1973. ;'•.'
-------
Wet phosphate rock enters the IMC dryer where the
particles are dried by hot furnace gases. Dust entrainment
in the exit flue gases are scrubbed in a vertical spray
chamber scrubber. Fluoride emissions in the phosphate rock
are driven off to the exit gases by the heat in the dryer.
- 2 -
-------
III. SUMMARY OF RESULTS
Table 3.1 presents the overall test summary of particulate
and fluoride concentrations in U.S. and metric units. Three
i
particulate and three fluoride tests were averaged to obtain
values in Table 3.1. Individual test results are shown as
follows in subsequent Tables:
Table Description
3.2 Particulate Analysis Summary
3.3 Fluoride Analysis Summary
3.4 Scrubber Water Analysis Summary
3.5 Rock Material Analysis Summary
In all cases, except particulate Test 3, outlet flow
rates were slightly higher than volumes sampled at the inlet.
This may have been caused by a cyclonic flow distribution, but
since the maximum difference is only three percent, other
factors may have contributed to the variation.
Individual particulate test data (Table 3.2) do not vary
appreciably for this type of process. Outlet loadings from Run 1
are approximately twice the values from the last two runs, but
this could be due to the particle size distribution in the type
of feed. Types of feed are discussed in Section V.
The inlet combustion gas analysis for the particulate runs
indicates arise in 009 and a decrease in O for the three tests.
- 3 -
-------
Table 3.1 OVERALL SUMMARY OF RESULTS FROM PHOSPHATE ROCK DRYER
Pollutant
Measurement
System
Volume of Gas Sampled
Percent Moisture by Volume
Average Stack Temperature
Dry Stack Volumetric Flow Rate
Actual Stack Volumetric Flow Rate
Percent Isokinetic
Feed Rate
Partial Catcha
Weight
Concentration/Dry Volume
Concentration/Actual Volume
Concentration/Time
Lb/ton, Feed Concentration/
Input Feed Rate
Total Catch
Weight
Concentration/Dry Volume
Concentration/Actual Volume
Concentration/Time
Lb/ton. Feed Concentration/
Input Feed Rate
Percent Impinger Catch
Percent Insoluable Fluoride
Units
U.S.
DSCFb
%
op
DSCFM0
ACFMd
Ton/hr
mg
gr/DSCF
gr/ACF
Ib/hr
lb/tort
mg
gr/DSCF
gr/ACF
Ib/hr
Ib/ton
%
'
Metric
DNm3
%
°C
M3/ g
'see
Mton/hr
mg
mg/DNm3
kg/hr
kg/Mton
mg
mg/m
kgAr
kg/Mton
%
•'
PARTICIPATE
U.S.
Inlet
51.608
31.63
164
65597
110262
107.6
333
5621
1.639
0.9783
923.8
2.741
5746
1.677
0.9922
945.5
2.810
2.55
Outlet
108.212
27.19
151
65040
101562
105.0
333
302
0.04275
0.02741
23.80
0.076
411
0.05832
0.03740
32.51
0.102
27.6
Metric
Inlet
1.4614
31.63
73.3
30.958
52.038
107.6
302
5621
3751
2239
419.0
1.371
5746
3837
226.2
428.9
1.405
2.55
Outlet
3.0642
27.19
66.1
30.696
47.932
105.0
302
302
97.83
62.72
10.80
0.0380
411
133.4
85.57
14.75
0.0510
27.6
FLUORIDE
U.S.
Inlet
65.795
31.89
164
65011
109554
107.8
295
3.8
0.00089
0.00053
0.50
0.0017
166.2
0.03902
0.02319
21.76
0.0732
97.7
Outlet
106.414
26.00
151
66754
102868
100.5
295
4.7
0.00068
0.00044
0.39
0.0014
27
0.0040
0.0026
2.3
0.0076
63
Metric
Inlet
1.8609
31.89
73.3
30.682
51.704
107.6
268
3.8
2.0
1.2
0.23
0.00085
166.2
89.28
53.06
9.870
0.0366
97.7
Outlet
3.0133
26.00
66.1
31.505
48.549
105.0
268
4.7
1.6
1.0
0.18
0.00070
27
9.2
5.9
1.0
0.0038
63
a) For particulate, catch includes probe, cyclone and filter. For fluoride, catch includes the water soluble portion.
b) Dry standard cubic feet at 70°F, 29.92 in. Hg.
c) Dry standard cubic feet per minute at 70°F, 29.92 in. Hg.
d) Actual cubic feet per minute
e) Dry normal cubic meters at 21.1°C, 760 mm Hg.
f) Dry normal cubic meters per second at 21.1°C, 760 mm Hg.
g) Actual cubic meters per second
-------
Table 3.2 PARTICULATE EMISSION DATA SUMMARY
IMC Corp. , Kingsford, Florida - Rock Dryer Scrubber
Date
Volume of Gas Sampled-DSCFa
Percent Moisture by Volume
.Average Stack Temperature- °F
Inlet
1
2-12-73
43.182
31.06
164
Stack. Volumetric Flow Rate- 64,172
DSCFM
Stack Volumetric Flow Rate-
ACFMC
Percent Isckinetic
Percent CO2
Percent 0_
Percent CO
Feed Rate-ton/hr
Particulates-probe , cyclone ,
and filter catch
106,284
107.5
2.93
17.93
0.00
270
mg ' j 3170
gr/DSCF 1.133
gr/ACF 1 0.6840
i
Ib/hr j 623.2
Ib/ton feed
Particulates-total catch
mg
gr/DSCF
gr/ACF
Ib/hr
Ib/ton feed
Percent impinger catch
2.307
2
2-13-73
56.089
32.43
164
65,578
110,850
109.3
3.20
17.50
0.03
' 350
7566
2.082
1.231
1170
3.343
i
3315
1.185
0.7153
651.7
2.413
4.37
7698
2.118
1.253
1191
3.401
1.71
3
2-13-73
55.553
31.41
164
67,040
113,652
105.9
5.00 .
15.70
0.07
380
6128
1.702
1.020
978.2
2.574
6226
1.729
1.036
993.8
2.615
1.57
Outlet
1 j 2
2-12-73
111.796
26.94
151
64,707
100,604
108.9
5.00
15.43
0.03
270
*
438
0.06046
0.03892
33.53
0.124
537
0.07412
0.04772
41.11
0.152
18.4
2-13-73
10_6.232
27.15
150
66,569
103,573
100.6
3
2-13-73
106.817
27.48
152
63,843
I
100,508
105.5
4.03 5.47
15.13
0.00
15.57
0.03
i
350 380
i i
I
229
0.03326
0.02138
239
0.03452
0.02193
18.98
0.054
376
0 .05462
13.90
0.049
320
0.04623
i
0.03511 0.02936
31.12
0.089
39.1
25.30
0.066
25.3
i
a) Dry standard cubic feet at 70°F, 29.92 in. Kg.
b) Dry standard cubic feet per minute at 70°F, 29.92 in. Hg.
c) Actual cubic feet per minute.
-------
The Orsat apparatus did not have a water jacket around the
gas tube which may have varied the gas temperature and, in turn,
the gas analysis.
The insoluble portion of the outlet fluoride catch (Run 5,
Table 3.3) is approximately an order of magnitude greater than
values from the other two tests. Duplicate laboratory analysis
produced the same results. A 60% pebble - 40% concentrate was
being fed to the dryer during this test. The fluoride analysis
of the feed (Table 3.5) confirms the rock was similar to other
runs. Therefore it appears that the scrubber efficiency was
lower during this run.
After each of the three fluoride runs, 100 ml of inlet
and outlet sample was removed and given to IMC personnel for
their analysis. This was considered when EPA analyzed the
samples for fluoride concentration.
No major testing problems occurred during sampling at the
Kingsford site. A pump in one of the outlet meter boxes acted
up at times, so another meter box was used to complete the tests,
A cyclone bypass connector on the outlet train broke after one
test was complete, so the sample stayed intact. During the last
fluoride test, the probe contacted the mud on the inside of the
duct creating a vacuum in the sample train. The impinger water
- 6 -
-------
Table 3.3 FLUORIDE EMISSION DATA SUMMARY
IMC Corp., Kingsford, Florida - Rock Dryer Scrubber
'
Date
Volume of Gas Sampled-DSCFa
Percent Moisture by Volume
Average Stack. Temperature-°F
Stack Volumetric Flow Rate-
DSCFMb
Stack Volumetric Flow Rate-
ACFMC
Percent Isokinetic
Feed Rate-ton/hr
Fluoride-water soluble
mg
gr/DSCF
gr/ACF
Ib/hr
Ib/ton feed
Fluoride -total
mg
gr/DSCF
gr/ACF
Ib/hr
Ib/ton feed
Percent insoluble fluoride
Inlet
4
2-14-73
66.354
33.19
164
64,884
111,213
108.9
270
3.1
0.00072
0.00042
0.400
0.0015
137.2
0.03190
0.01861
17.75
0.0654
97.7
5
2-14-73
65.280
32.27
164
65,641
111,127
105.9
300
5.7
0.00134
0.00079
0.758
0.0025
189.6
0.04482
0.02647
25.22
0.083S
97.0
6
2-15-73
65.751
30.22
164
64,528
106,321
108.5
315
2.6
0.00061
0.00037
0.337
0.0011
171.9
0.04034
0.02448
22.30
0.0705
93.5
Outlet |
4
2-14-73
105.672
25.45
151
66,251
101,226
100.6
270
5.6
0.00082
0.00053
0.464
0.0017
7.9
0.00115
0.00075
0.655
0.0024
29.1
5
2-14-73
105.913
24.92
150
67,700
102,610
98.6
300
6.3
0.00091
0.00060
0.532
0.0018
67.4
0.00982
0.00648
5.69S
0.0187
90.7
6
2-15-73
107.658
27.64
151
66,312
104,767
102.4
315
2.2 :
i
0.00031
0.00019
0.179
0.0006
6.9
0.00098
0.00062
0.562
0.0018
68.1
a) Dry standard cubic feet at 70°F, 29.92 in. Hg.
b) Dry standard cubic feet per minute at 70°F, 29.92 in. Hg.
c)-Actual cubic feet per minute
-------
Table 3.4. SCRUBBER WATER ANALYSIS SUMMARY
Item
Total Fluorides, mg/1
P205, mg/1
PH
Test Number i
Scrubber Inlet
4
0.4
0.1
7.77
5
0.4
0.1
7.93
6
0.4
0.1
7.81
Avg.
0.4
0.1
7.84
Scrubber Outlet
4
1.3
1.4
7.60
5
1.2
1.2
7.89
6
1.1
1.4
7.66
Avg_.
1.2
1.3
7.72
Table 3.5. ROCK MATERIAL ANALYSIS SUMMARY
Test
1
2
3
4
5
6
Total Fluoride, mg/gm
Raw Feed
38.9
37.8
38.5
38.6
36.9
38.5
Dry Product
36.2
39.9
39.2
38.5
37.1
39.4
Total P20sr %
Raw Feed
31.1
30.5
33.0
31.8
31.5
32.4
Dry Product
31.8
32.4
33.1
30.9
30.7
33.8
Sample collection times can be found in Appendix H.
- 8 -
-------
backed up and wet the filter. The sampling team replaced the
filter without losing any of the sample water.
- 9 -
-------
IV. PROCESS DESCRIPTION
A phosphate rock plant flow diagram, shown in Figure 4.1,
illustrates the steps of the process in relation to the drying
operation. After the phosphate rock is mined, a flotation
process separates the impurities by chemically separating the
impurities from the phosphate sand. The washed rock is then
conveyed to storage bins.
The wet rock contains between 7 and 20 percent moisture
while being transported to the drying site. All grades of rock
are normally dried to 3 percent or less in the drying operation,
The IMC.rock dryer at Kingsford, Florida is a 270 TPH
fluid bed unit. The rock is dried by heat supplied by natural
gas, with No. 5 residual fuel oil as a standby.
From the drier, combustion gases enter a cyclone
separator where the majority of the particulate is removed
and recycled back to the product conveyor belt. The gases
then flow to a cyclonic scrubber using fresh water as the
scrubbing media. The clean gas flow leaves the scrubber and
exits through a ninety foot stack.
- 10 -
-------
HIGH PRESSURE
WATER
DRAGLINE
~7 \ RECOVERY PLANT
CI
WASHER
>, r^x^ ^
-—-». • * \ *•' V N/ V V
-XV \V ' / V VX'N NX
, ' /-- '< ^ A/V v v\X M ! M r
(>^S7^^^ M'Nl-
'.x^xx x vyy v v \2i2iiX
PUMP
777-
SURGE
HOPPER
DRY ROCK
STORAGE
DRY ROCK
SHIPPING
GROUND ROCK
J SH PPING
WET
ROCK
SHIPPING
WET ROCK STORAGE
AND RECOVERY
Figure 4.1. Phosphate rock flow sheet.
-------
V. PROCESS OPERATION*
The operations at the Kingsford rock drying plant were
monitored from the systems control room by Mr. J. Peoples
of EPA. From this vantage point, Mr. Peoples was able to
collect data and communicate any upsets or abnormalities
during the testing periods to the test crew. Data recorded
included recording times, feed and production rates, and other
additional information related to the fluid bed dryer and
scrubber cleaning operation (Appendix C tabulates this informa-
tion along with related comments).
Particulate and fluoride tests were conducted at the
rock drying plant during the week of February 12, 1973. On
the first day, Feb. 12, 1973, no process or sampling problems
were encountered while conducting Run 1. However, fuel oil
to furnish heat for the dryer was substituted for natural gas
due to the fuel shortage.
The raw materials fed to the dryer consisted of a mixture
of 60% pebble - 40% concentrate. For the second particulate
feed-in materials consisted of 70% pebble - 30% concentrate.
One process shutdown occurred during the second run and delayed
testing from 11:00 A.M. until 12:50 P.M. due to conveyor belt
trouble. The dryer feed for the third run was 100% concentrate.
*Written and supplied by EPA.
- 12
-------
On Wednesday, Feb. 14, at 9:10 A.M. the fluoride tests
were begun at Kingsford. A mixture of 60% pebble - 40% concen-
trate was being processed during runs one (1) and two (2).
Test three (3) was completed on Thursday with 100% concentrate
being processed under normal conditions. No process upsets
interfered with these tests.
The opacity of the Kingsford dryer effluent was difficult
to estimate because of the high moisture content.
- 13 -
-------
VI LOCATION OF SAMPLING POINTS
A. Inlet Location
The configuration of the ductwork downstream of the drier
as shown in Figure 6.1 made it impossible to choose a section
completely free of turbulent effects. The best possible
sampling location was chosen at a point nine inches upstream
from two vertical water sprays and approximately twelve feet
downstream of an induced draft fan.
Since this chosen sampling location was less than 0.5 diameters
from the water sprays, the maximum number of traverse points
was chosen as required by the Federal Register, Vol. 36, No. 247.
Figure 1.1, Dimensions of the duct are 79 inches high by 56.5
inches wide. Five ports were decided upon with ten traverse
points per port for a total of fifty (50) points as illustrated
in Figure 6.2. Each sampling area would then theoretically
be 15.8 inches long by 5.65 inches wide, resulting in an aspect
ratio of 2.66, Increasing the sampling ports would reduce the
aspect ratio but the existing framework around the bypass
stack made this impossible, since a minimum of 14 inches was
needed from the existing framework to the sampling port.
Actual sampling areas in Figure 6.2 are somewhat smaller
than theoretical areas due to a 6 to 7 inch layer of mud on
the bottom of the duct. This reduction in area was considered in
- 14 -
-------
Figure 6.1 Dryer Inlet Location
Gas Flow
cyclone
scrubber
and existing
stack
24" / water sprays
-
_ j
Top View
78" ID
water spray line
sampling platform
A
B
C
sampling
port
Elevation
- 15 -
-------
Figure 6.2 Inlet Traverse Points
.-H 4"
I
7.5"
o- o
14.5"
14"
14"
14"
15"
Al A2 A3 A4 A5 A6 A7 A8 A9 A10
Bl B2 B3 B4 B5 B6 B7 B8 B9 BIO
OOOOOOOSOO
Cl C2 C3 C4 C5 C6 C7 C8 C9 CIO
Dl D2 D3 D4 D5 D6 D7 D8 D9 D10
El E2 E3 E4 E5 E6 E7 E8 E9 E10
OOOO OGOOOO
"6 to 7 inch mud layer
79"
56.5"
Looking Down Stream
- 16 -
-------
determining air flow rates. A bypass stack without control
equipment is less than one duct diameter upstream from the
chosen sampling location and was blocked off by a damper
which eliminated air stream turbulence.
B. Outlet Location
The rock drying operation is equipped with a stack
approximately 90 feet high. Two existing ports were found
to be satisfactory for emission testing. A diagram of the
stack is shown in Figure 6.3 with sampling port and water
spray locations. Two sampling locations, 90.degrees apart
are needed for vertical stack sampling (Federal Register,
Vol. 36,,No. 247, Page 24882). When dimensions of the stack
were taken, it was noticed the ports were only 80 to 85
degrees apart, which was assumed to be satisfactory.
The stack is 83 inches in diameter at the sampling
location. Ports are located 6 ft. below the top of the stack,
and 10 ft. above the closest obstruction. Forty-eight sampling
points (24 along each diameter) were chosen to satisfy traverse
point requirements as specified by the Federal Register. A cross
section of the stack is shown in Figure 6.4 with the chosen
sampling points.
A portion of the existing railing had to be removed to
allow for the sample box traverse. A support platform was
designed so the sample box could be overhung over the walk,
since the needed sampling width was much greater than the
catwalk width of 30 inches.
- 17 -
-------
Figure 6.3 Dryer Outlet Location
-83"ID-H
no:
Outlet Ports
\
T
17'10"
•c
11...
Water Sprays
Water Manifold
10'
5'6'
Gas Flow
51'9'
Inlet Duct
78"
\
Water Exit
Elevation
- 18 -
-------
Figure 6.4 Outlet Traverse Points
Traverse Point Locations are Tabulated in Appendix D
— Port B
Looking Down
- 19 -
-------
VII. SAMPLING PROCEDURES f
All particulate and fluoride sampling procedures were
selected and approved by EPA prior to field sampling. The
contractor performed all testing and sample recovery operations;
collected samples were analyzed by EPA personnel. The inlet
and outlet sampling crew consisted of two man teams which
included a meter and probe technician.
At the inlet site, there was a relatively high positive
pressure combined with high humidity air at approximately
165°F. Water condensed in the first two impingers which had
to be replaced midway through the test. After each inlet test
the total water volume was determined and properly stored for
analysis.
The outlet meter box was positioned approximately fifty
feet below the sampling site. Since the diameter of the
stack was much greater than the width of the catwalk, two
different probe lengths reduced the sample box overhang.
Preliminary Traverse and Moisture
A preliminary velocity traverse at the inlet and outlet
location determined approximate nozzle sizes and isokinetic
sampling conditions. A 0.180 inch I.D. nozzle was used for
subsequent inlet testing while the outlet was sampled with a
0.277 inch I.D. nozzle.
- 20 -
-------
The inlet stack gas moisture was determined by a train
similar to Figure 4.2, Federal Register, Vol. 36, No. 247,
Page 24887, Tests on inlet flue gas indicated values were close
to saturation; the outlet gas was assumed to be saturated.
Gas Velocity and Temperature
Velocities were measured at each sampling point across the
stack diameter to determine an average value according to pro-
cedures described in the Federal Register. Flow rates were
calculated from velocities at inlet and outlet stations from
which flow continuity was checked. Gas flow temperatures were
measured by long stem dial thermometers.
Molecular Weight and Gas Analysis
An integrated sample of the stack gases was collected
at the inlet and outlet during each particulate run by pumping
flue gas into a Mylar bag at the rate of approximately 0.5 liter
per minute. This bag sample was then analyzed on-site with an
Orsat apparatus for CO?, 0_, and CO. Prior to the first test
all chemical reagents were changed to assure accurate readings.
The molecular weights of the particulate combustion gas ranged
from 29.19 to 29.50. A molecular weight of 29.00 was assumed
for all fluoride test calculations.
- 21 -
-------
Particulate Sampling
Particulate matter was isokinetically sampled from the
drying operation with a train shown in Figure 7.1 Although
the Federal Register stipulates a minimum sampling time of 5
minutes for each sampling point, shorter times were used due
to the number of traverse points selected.
The train consisted of a stainless steel nozzle, a heated
glass probe (a 6 ft. probe was used at the inlet while testing
at the outlet was done with 4 ft. and 8 ft. lengths), a heated
glass fiber filter, and four impingers connected in series
with glass ball joint fittings. The first two impingers
contained 100 ml of water each, the third impinger was left
empty and approximately 200 grams of preweighed silica gel
were placed in the fourth impinger.
In all cases sampling was conducted under isokinetic
conditions by continually monitoring the velocity with a pitot
tube and adjusting the sampling rate accordingly.
At the inlet site, the first impinger initially containing
100 ml. of water had to be replaced midway during the tests
due to the large volume of water which condensed in the
impinger. The stack team was very careful so as not to contact
the nozzle with the muddy ports and stack walls when changing
ports.
- 22 -
-------
FILTER HOLDER.
STACK WALL
THERMOMETER
U)
I
REVERSE-TYPE
PITOT TUBE
DRY TEST METER VACUUM PUMP
Figure 7.1. Particulate sampling train,
-------
The train cleanup procedure consisted of measuring the
water collected and weighing the silica gel to determine
moisture content. The water was then poured into a glass
jar. The filter was removed and placed in a marked container.
The probe and front half of the filter holder were then rinsed
with analytical reagent grade acetone and the washings placed in
a glass container. The rear half of the train consisting of
filter holder, impingers, and connectors was rinsed with
distilled water and this water added to the impinger contents.
The rear half of the train was then rinsed with acetone and
placed in a third sample jar. A portion of the acetone and
distilled water used in the sample recovery were set aside and
used as blanks for analysis.
Fluoride Sampling
Particulate and gaseous fluorides were withdrawn iso-
kinetically from the phosphate drying operation with the train
shown in Figure 7.2.
The design and contents of the four impingers are identical
to the particulate train. An unheated filter holder containing
a Whatman No. 41 paper filter was placed between the third and
fourth impingers to trap particulate fluorides. Contents of
the first three impingers, water wash of probe, nozzle, and
filter holder were placed in the same container. No acetone
wash was required for the fluoride clean up.
- 24 -
-------
NJ
Ul
PROSE
fc=l
£
U.
REVERSE-TYPE
PITOT TU8E
STAC KW ALL
THtRMOMEfEP.
FILTER HOLDER
\
CRY TEST METER
VAC'JU.'/ PU.UP
Figure 7.2 Fluoride Sampling Train
-------
During the first fluoride test, the filter exhibited a
darkish brown color. Since the Whatman filter was unheated,
the exhaust gas constituents must have accounted for the color
change. As explained earlier in Section 3, the original pump
used at the outlet site sounded as if the vanes were sticking.
On the first outlet fluoride test, the flow rate was difficult
to maintain for isokinetic sampling (see Appendix D). At high
velocities either the filter was starting to plug up or the
pump connections could have been loose. In either case, another
pump was used for subsequent outlet testing.
Scrubber water along with raw and finished product rock
samples were taken during each test. Three scrubber water
samples during each test were measured for pH and temperature
and identified on the individual bottles.
Sample Storage
All samples were placed in 1000 ml glass and polyethylene
containers and marked with EPA identification tags (see
Appendix G). The bottles were then put in wooden boxes with
styrofoam separators and hand delivered to EPA, North Carolina,
after all tests were completed.
- 26 -
------- |