oEPA
            United States
            Environmental Protection
            Agency
           Office of Air Quality
           Planning and Standards
           Research Triangle Park NC 27711
EMB Report 80-IBR-11
November 1980
            Air
Industrial Boilers

Emission Test Report
General  Motors
Corporation
Parma, Ohio

-------
              INDUSTRIAL BOILERS
             Emission Test Report
          General Motors Corporation
                  Parma, Ohio
             November 10-17,  1980
           Technical Directive No. 4
                 Prepared for

        Environmental Protection Agency
 Office of Air Quality Planning and Standards
          Emission Measurement Branch
            Research Triangle Park,
             North Carolina  27711
                      by

       C.  L.  Cornett,  Jr.,  and R.  G.  Beer

Contract No.  68-02-3547, Work Assignment No. 2
                 (ESED 80/33)
                 November 1980
         MONSANTO RESEARCH CORPORATION
               DAYTON LABORATORY
              Dayton, Ohio 45407

-------
                            CONTENTS


                                                            Page

Figures	      iv
Tables 	       v
1.  Introduction
2.   Summary of Results	       3
      Description of Monitoring	       3
      Summary of Results 	       5

3.   Process Description and Monitoring 	      25
      Process Description	      25
      Monitoring of Plant Processes During Testing ...      29
      Comments on Tests	      38

4.   Location of Sampling Points	      41
      Scrubber Inlet 	      41
      Scrubber Exhaust	      43
      Coal Sampling Site	      43

5.   Description of Sampling Trains 	      45
      Particulate Sampling Trains	      45
      Particulate Size Distribution Sampling Apparatus .      45

6.   Sampling and Analytical Procedures 	      49
      Summary	      49
      Field Sampling	      49
      Sample Handling	      50
      Chemical Analysis	      50
      Coal Analysis	      51
      Data Reduction	      52
      Quality Assurance	      53
LCC/D                          iii

-------
                      CONTENTS (continued)


                                                            Page

Appendices (continued)
  E.   Analytical Methods for Sulfuric Acid and
         Sulfate Determinations	     E.I
  F.   Analytical Data	     F.I
  G.   Project Participants	     G.I
                               IV

-------
                             FIGURES
Number                                                      Page

  1       Cumulative size distribution for first inlet
            Andersen run at General Motors Chevrolet Plant
            in Parma,  Ohio	     18

  2       Cumulative size distribution for second inlet
            Andersen run at General Motors Chevrolet Plant
            in Parma,  Ohio	     19

  3.      Cumulative size distribution for scrubber exhaust
            Andersen run at General Motors Chevrolet Plant
            in Parma,  Ohio	     20

  4       Schematic diagram of Boiler No.  1 and associated
            equipment at General Motors Chevrolet Plant in
            Parma,  Ohio	     26

  5       Flow diagram of double-alkali scrubbing unit. .     30

  6       Scrubber inlet sampling site at Parma,  Ohio
            GM Plant	     42

  7       Scrubber exhaust sampling site at Parma,  Ohio
            GM Plant	     44

  8       Particulate sampling train - EPA Methods 5
            and 5B	     46

  9       Particulate size distribution sampling
            apparatus - Andersen 2000, Inc	     47

-------
                             TABLES
Number                                                       Page

   1     Source Sampling and Analyses at Chevrolet Plant
           in Parma,  Ohio (November 10-17,  1980)	       4

   2     Emission Data and Stack Gas Parameters, General
           Motors Chevrolet Plant,  Parma,  Ohio,
           November 10-17, 1980 (Metric Units)	       6

   3     Emission Data and Stack Gas Parameters, General
           Motors Chevrolet Plant,  Parma,  Ohio,
           November 10-17, 1980 (English Units)  	       9

   4     Summary of Scrubber Efficiency Measurements
           and Differences between Method 5 and  Method 5B
           Results at General Motors Chevrolet Plant in
           Parma, Ohio (November 10-17, 1980)  	      13

   5     Summary of Duration of Sampling,  Stack  Temperature,
           Stack Flow Rate,  Sample Volume,  and Water Content
           of Methods 5 and 5B Stack Samples at  the General
           Motors Chevrolet Plant,  Parma,  Ohio
           (November 10-17,  1980) 	      15

   6     Summary of Integrated Gas Analysis Results at
           the General Motors Chevrolet Plant, Parma,
           Ohio (November 10-17,  1980)	      16

   7     Summary of Andersen Particle Sizing Results at
           the General Motors Chevrolet Plant, Parma,
           Ohio (November 10-17,  1980)	      17

   8     Summary of Plume Opacity Observations at the
           General Motors Chevrolet Plant in Parma,  Ohio
           (November 10-17,  1980) 	      21

   9     Summary of Coal Analysis at the General Motors
           Chevrolet Plant,  Parma,  Ohio (November 10-17,
           1980)	      23

  10     Design, Operating,  and Performance Characteristics
           of Double-Alkali  Scrubbing System	      27
                                VI

-------
                       TABLES (continued)

Number                                                      Page

  11      Power Plant Operating Parameters (11/12/80,
            Run SI)	      32

  12      Power Plant Operating Parameters (11/13/80,
            Run 1)	      33

  13      Power Plant Operating Parameters (11/13/80,
            Run 2)	      34

  14      Power Plant Operating Parameters (11/14/80,
            Run 3R)	      35
                                VI1

-------
                            SECTION 1
                          INTRODUCTION

Emissions from Boiler No. 1 at the General Motors Chevrolet Plant,
on Brookpark Road in Parma, Ohio,  were tested November 11-14,  1980,
by Monsanto Research Corporation (MRC).  This work was performed
for the Emission Measurement Branch of the U.S. Environmental
Protection Agency (EPA) under Contract No. 68-02-3547, Work Assign-
ment No. 2.  The boiler tested was a 12.6 kg/s (100,000 Ifo/hr)
steam capacity waste oil and coal-fired spreader stoker with an
economizer, a multiclone, and a double-alkali scrubber.

The objective of the sampling program is to provide background
information on well-controlled sources for the development of new
source performance standards.  Within this framework,  the objec-
tives of the sampling at the Parma, Ohio, General Motors Plant
were:

   •  To determine the effect of raising the temperature of the
     filter and probe on an EPA Method 5 train from 120°C (248°F)
     to 177°C (350°F) on the amounts of particulate, sulfate,  and
     sulfuric acid emissions measured downstream of the scrubber
     (Method 5 testing, with the filter and probe at 177 ± 14°C
     [350 ± 25°F] is called Method 5B testing);

   •  To determine the effect of the double-alkali scrubber on par-
     ticulate, sulfate, and sulfuric acid emissions, as measured
     by EPA Method 5B.

-------
The field test was monitored by Dennis Holzschuh, Emission Measure-
ment Branch, EPA.  The sampling was directed by Richard G. Beer,
MRC team leader.

The monitoring performed at this plant consisted of three simul-
taneous runs of Methods 5 and 5B at the scrubber outlet performed
simultaneously with Method 5B sampling at the scrubber inlet.
Separate stack test trains, traversing on perpendicular diameters
of the stack, were used for the tests at the scrubber outlet.  One
additional single-point run of Methods 5 and 5B was performed at
the scrubber outlet with both sampling systems operating simulta-
neously at essentially the same point in the stack and with no
testing at the scrubber inlet.  The purpose of the run of Methods
5 and 5B at a single point in the stack was to obtain an indica-
tion of whether or not the differences between the results of
simultaneous Method 5 and 5B sampling are significantly different
when both methods were used at a single point versus when they
were used with two stack test trains traversing the same stack
independently.

The boiler was running under steady state conditions (no soot
blowing, ash unloading,  etc.) at a minimum of 87% capacity during
these tests, and the coal burned was considered to be representa-
tive of normal feed.

The filters and dried acetone and water rinses of the Method 5 and
5B runs were weighed and analyzed for sulfuric acid and sulfates
using procedures described in the Sampling and Analytical Proce-
dures section of this report.

The plume opacity was measured during all Method 5 and 5B tests
by EPA Method 9, when possible.   In addition,  particle size dis-
tributions were measured twice at the inlet to the scrubber and
once at the outlet, using Andersen cascade impactors.

-------
                            SECTION 2
                       SUMMARY OF RESULTS

DESCRIPTION OF MONITORING

Table 1 summarizes the monitoring that was performed at this
plant.  It consisted of three simultaneous runs of Methods 5 and
5B at the scrubber outlet performed simultaneously with Method 5B
sampling at the scrubber inlet.  Separate stack test trains,
traversing on perpendicular diameters of the stack, were used for
the tests at the scrubber outlet.  Sample volumes of at least
2.4 dry m3 (85 dry ft3) were taken during these runs.  One addi-
tional single-point run of Methods 5 and 5B was performed at the
scrubber outlet, with both sampling systems operating simultaneous-
ly at essentially the same point in the stack,  with no testing at
the scrubber inlet.  About 1.9 dry m3 (68 ft3)  was sampled by each
sampling train during the single point sampling run.  The boiler
was running under steady state conditions (no soot blowing, ash
unloading, etc.) at a minimum of 87% of its full capacity during
these tests.   The coal burned was considered to be representative
of normal feed.

One additional run of Method 5B at the inlet and Methods 5 and 5B
at the outlet occurred, but it was terminated before the traverses
were completed when it became evident that the coal being burned
contained unusually high amounts of coal dust.

Distilled water was used to remove deposits from the inside of
the Method 5 and 5B probes, nozzles, and filter holders, after
the usual acetone rinses.  This was done because, during previous

-------
                TABLE 1.  SOURCE SAMPLING AND ANALYSES AT CHEVROLET
                          PLANT IN PARMA, OHIO (NOVEMBER 10-17, 1980)
SAMPLING AND ANALYSIS REQUIREMENTS
MRC Job No. 101.1221
Total
no. of
samples
3*
34
3a
1
1
2
1
4
26b



Sample
type
articulate & ORSAT
crubber InTet
Scrubbet Bxhaust
articulate t ORSAT
crubber Exhaust
articulate & ORSAT
crubber Exhaust
articulate £, ORSAT
crubber Exhaust
ORSAT Scrubber
Inlet
ORSAT Scrubber
Exhaust
Plune Opacity
Coal



Sampling
method
SB
5B
5
iingle Poin
5-
iingle Poin
tndersen
Lnderaen
9C
Crab



Contract No. : bs-02-3547
Company Name: GENERAL MOTORS
Industry: INDUSTRIAL BOILER
Sample
collected
by












Minimum
sampling
time
120 min
20 min
s multaneoi
* th above
0 min
i multaneoi
i th above
60 min
60 min
iimultaneoi
rith above
)epends on
.oading
tepends on
wading
iimultaneoi
ilth Methoi
S i. SB
Before,
during anc
Method S
tests


Assignment Number: WA ,*2 Technical Directive: 2
CHEVROLET Company Location: PARMA, OHIO
Process: COAL t WASTE OIL-FIRED
SPREADER STOKER
Minimum
volume gas
sampled ft3
60
s 60
s 60
60
s 60
Depends on
Loading
Depends on
Loading
s
Not
Applicable
Not
Applicable



Initial Analysis
Type Method By
Velocity
temp, flow
Oa li. C0a
Velocity
temp, flow
Oa t, COa
Velocity
temp, flow
Oa t COa
Velocity
temp, flow
Oj S. COa
Velocity
temp, flow
Oa ' CO2
Velocity
temp.
Oa ' COa
Velocity
temp.
Oa ' COa
Visual
Observatio
Split and
Prepare 8"
aggregate
Samples fo
runs Rl,
R2, R3
R3 retest
L AI-R2,
(PO RS t.
R6)
1-4
1-4
1-4
1-4
1-4
2-4
2-4
9C
iSTH D201
172)



MRC












Control Equipment: MULTICLONES
t DUAL-ALKALI SCRUBBER
Final Analysis
Type Method By
Particulate
50»-J,
HaSO.
Particulate
SO,-",
HaSO«
'articulate
HaSO,
Particulate
H,SO.
HaSO,
'article
iize
pistributio
'article
iice
>istributio
tot
applicable
Iltl
BTU,
mate,
Size



Method S
> 10/10/80
Procedure
Method S
t 10/10/80
Procedure
Method S
t 10/10/80
'rocedure
Method 5
> 10/10/80
Procedure
Method S
f i 0/1 0/80
Procedure
kndersen
n
uidersen
tot
applicable
fkSTM D3176
ASTM D2015
60, 1721
ISTM D410-
38
Except use
ivailable
lample 14,
16, 30, 20
Sieves 6
1" screen.
MRC











 One  additional run occurred but it was terminated prior to completing the traverse.

 19 of these  samples were split and combined into aggregate samples  for  all  four
 Method 5/5B  traverse tests, one of the single point tests and one of the Andersen
 tests.
•»
'Not  all of these measurements were strictly Method 9 because of an  interfering steam
 plume.

-------
testing on boilers,  yellow films had been observed on the probe
linings of Method 5 trains after acetone rinses.   Sulfate deposits
were a suspected cause of the film.

The filters and dried acetone and water washes of the Method 5
and 5B runs were weighed and analyzed for sulfuric acid and
sulfates.

Methods 1 through 4 were used during all Method 5 sampling runs,
as in typical compliance monitoring.

The plume opacity was measured during all Method 5 and 5B tests
by EPA Method 9 when possible.  However, a steam plume from
another stack prevented valid Method 9 readings from being made
during much of the emission testing.  In addition, particle size
distributions were measured twice at the inlet to the scrubber
and once at the outlet, using Andersen cascade impactors.

SUMMARY OF RESULTS

The particulate, sulfuric acid, and sulfate emissions measured by
Methods 5 and 5B are summarized in Tables 2 and 3.  In addition,
Table 3 shows the percents of the concentrations measured by
Method 5 at the outlet represented by the concentrations measured
by Method 5B simultaneously.  Run 3 is the partial traverse run,
and run SI is the run using Methods 5 and 5B at a single point in
the scrubber exhaust.  The emission rates in Tables 2 and 3 do
not include the water washes of the Method 5 and 5B trains.  Par-
ticulate emissions represent all emissions measured by weighing
the Method 5 or 5B samples (including sulfuric acid and sulfates).

Particulate concentrations measured by Method 5B traverses were
50% to 84% of those measured by Method 5.  On the single point
run, Method 5B particulate concentrations were 56% of particulates
measured by Method 5.  Sulfuric acid concentrations measured on

-------
TABLE 2.  EMISSION DATA AND STACK GAS PARAMETERS,  GENERAL MOTORS
          CHEVROLET PLANT, PARMA, OHIO, NOVEMBER  10-17,  1980
                          (METRIC UNITS)

Emissions
Run Sampling
number Date Location method Pollutant
1 11-13-80 Inlet
Outlet
Outlet
Inlet
Outlet
Outlet
Inlet
Outlet
Outlet
2 11-13-80 Inlet
Outlet
Outlet
Inlet
Outlet
Outlet
Inlet
Outlet
Outlet
5B
5
5B
5B
5
5B
5B
5
5B
5B
5
5B
5B
5
5B
5B
5
5B
Particulate
Particulate
Particulate
Sulfuric acid
Sulfuric acid
Sulfuric acid
Sulfates
Sulfates
Sulfates
Particulate
Particulate
Particulate
Sulfuric acid
Sulfuric acid
Sulfuric acid
Sulfates
Sulfates
Sulfates
Actual
g/dscm
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.5165
.0968
.0489
.0164
.0333
.0029
.0264
.0132
.0129
.7981
.1154
.0976
.0159
.0362
.0176
.0236
.0122
.0180
kg/hr
25.8
4.7
2.4
0.8
1.6
0.1
1.3
0.6
0.6
41.1
5.6
4.9
0.8
1.8
0.9
1.2
0.6
0.9
ng/J
256
48
23
8
16
I
13
6
6
385
57
48
1
18
8
11
6
8
.4
.9
.8
.1
.8
.4
.1
.7
.3
.8
.8
.5
.7
.1
.7
.4
.1
.9
Cor-
rec-
ted
to 12%
C02,
g/dscm
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.6324
.1223
.0593
.0201
.0421
.0035
.0323
.0167
.0156
.958
.1473
.1195
.0191
.0462
.0216
.0283
.0156
.0220
Iso-
ki-
netic
% Notes
100
103
99
100
103
99
100
103
99
98
101
98
98
101
98
98
101
98
.5
.0
.2
.5
.0
.2
.5
.0
.2
.9
.3
.4
.9
.3
.4
.9
.3
.4
                                                                  (continued)

-------
TABLE 2  (continued)

Emissions
Run
number Date Location
3 11-14-80 Inlet
Outlet
Outlet

Inlet
Outlet
Outlet

Inlet
Outlet
Outlet
3R 11-14-80 Inlet
Outlet
Outlet
Inlet
Outlet
Outlet
Inlet
Outlet
Outlet
Sampling
method Pollutant
5B
5
5B

5B
5
5B

5B
5
5B
5B
5
5B
5B
5
5B
5B
5
5B
Particulate
Particulate
Particulate

Sulfuric acid
Sulfuric acid
Sulfuric acid

Sulfates
Sulfates
Sulfates
Particulate
Particulate
Particulate
Sulfuric acid
Sulfuric acid
Sulfuric acid
Sulfates
Sulfates
Sulfates
Actual
g/dscm
0
0
0

0
0
0

0
0
0
0
0
0
0
0
0
0
0
0
.3336
.1990
.2897

.0062
.0260
.0095

.0197
.0198
.0274
.5646
.1016
.0598
.0020
.0325
.0068
.0293
.0153
.0141
kg/hr
12
9
13

0
1
0

0
0
1
6
4
2
0
1
0
1
0
0
.7
.3
.4

.2
.2
.4

.8
.9
.3
.8
.8
.8
.1
.5
.3
.4
.7
.7
ng/J
162
98
147

3
12
4

9
9
14
261
49
31
0
15
3
13
7
7
.7
.8
.8

.0
.9
.8

.6
.8
.0
.5
.6
.1
.9
.9
.5
.6
.5
.3
Cor-
rec-
ted
to 12%
C02,
g/dscm
0.4170
0.2437
0.3698

0.0078
0.0318
0.0121

0.0246
0.0242
0.0350
0.6392
0.1219
0.0772
0.0023
0.0390
0.0088
0.0332
0.0184
0.0182
Iso-
ki-
netic
%
105.4
99.5
102.0

105.4
99.5
102.0

105.4
99.5
102.0
101.3
99.4
100.1
101.3
99.4
100.1
101.3
99.4
100.1
Notes
Run terminated
prior to com-
pleting trav-
erses because
excessive
amounts of
coal dust
being burned












                                            (continued)

-------
                                      TABLE 2  (continued)

Emissions
Run
number Date Location
SI 11-12-80 Outlet
Outlet
Outlet
Outlet
Outlet
Outlet
Sampling
method Pollutant
5
5B
5
5B
5
5B
Particulate
Particulate
Sulfuric acid
Sulfuric acid
Sulfates
Sulfates
Acutal
g/dscm
0
0
0
0
0
0
.0955
.0536
.0292
.0026
.0152
.0033
kg/hr
5
2
1
0
0
0
.0
.6
.5
.1
.8
.2
ng/J
48
25
14
1
7
1
.3
.7
.8
.2
.7
.6
Cor-
rec-
ted
to 12%
C02/
g/dscm
0.1169
0.0624
0.0358
0.0033
0.0186
0.0038
Iso-
ki-
netic
% Notes
96
91
96
91
96
91
.5
.9
.5
.9
.5
.9
00

-------
TABLE 3.  EMISSION DATA AND  STACK GAS PARAMETERS, GENERAL MOTORS
          CHEVROLET PLANT, PARMA,  OHIO NOVEMBER 10-17, 1980
                         (ENGLISH UNITS)

Emissions
Run
number Date Location
1 11-13-80 Inlet
Outlet
Outlet
Inlet
Outlet
Outlet
Inlet
Outlet
Outlet
2 11-13-80 Inlet
Outlet
Outlet
Inlet
Outlet
Outlet
Inlet
Outlet
Outlet
Sampling
method Pollutant
5B
5
5B
5B
5
5B
5B
5
5B
5B
5
5B
5B
5
5B
5B
5
5B
Particulate
Particulate
Particulate
Sulfuric acid
Sulfuric acid
Sulfuric acid
Sulfates
Sulfates
Sulfates
Particulate
Particulate
Particulate
Sulfuric acid
Sulfuric acid
Sulfuric acid
Sulfates
Sulfates
Sulfates
Actual
gr/dscf
0.2256
0.0423
0.0213
0.0072
0.0145
0.0013
0.0115
0.0058
0.0057
0.3487
0.0504
0.0426
0.0069
0.0158
0.0077
0.0103
0.0053
0.0079
Ib/hr
56.8
10.4
5.3
1.8
3.6
0.3
2.9
1.4
1.4
90.6
12.4
10.8
1.8
3.9
1.9
2.7
1.3
2.0
Ib/mm
Btu
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.596
.114
.055
.019
.039
.003
.030
.016
.015
.897
.134
.113
.018
.042
.020
.026
.014
.021
Correc-
ted to
12% C02,
gr/dscf
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.2762
.0534
.0258
.0088
.0183
.0016
.0141
.0073
.0069
.4184
.0643
.0522
.0083
.0202
.0094
.0124
.0068
.0097
Percent of
Method 5
Iso- g/dscf repre-
ki- sented by
netic Method 5B g/
J% dscf at outlet
100.
103.
99.
100.
103.
99.
100.
103.
99.
98.
101.
98.
98.
101.
98.
98.
101.
98.
5
0
2
5
0
2
5
0
2
9
3
4
9
3
4
9
3
4

50.4


9.0


98.3


84.5


48.7


149.1

                                                                     (continued)

-------
TABLE 3  (continued)

Emissions
Run
number Date Location
3 11-14-80 Inlet
Outlet
Outlet
Inlet
Outlet
Outlet
Inlet
Outlet
Outlet
3R 11-14-80 Inlet
Outlet
Outlet
Inlet
Outlet
Outlet
Inlet
Outlet
Outlet
Sampling Actual
method Pollutant gr/dscf
5B
5
5B
5B
5
5B
5B
5
5B
5B
5
5B
5B
5
5B
5B
5
5B
Particulate 0.14573
Particulate 0.08703
Particulate 0.1266a
Sulfuric acid 0.0027
Sulfuric acid 0.0114
Sulfuric acid 0.0041
Sulfates 0.0086
Sulfates 0.0087
Sulfates 0.0120
Particulate 0.2467
Particulate 0.0444
Particulate 0.0261
Sulfuric acid 0.0009
Sulfuric acid 0.0142
Sulfuric acid 0.0030
Sulfates 0.0128
Sulfates 0.0067
Sulfates 0.0061
Ib/mm
Ib/hr Btu
0.3783
0.230a
0.3443
0.007
0.030
0.071
0.022
0.023
0.033
59.1 0.608
10.6 0.115
6.3 0.072
0.2 0.002
3.4 0.037
0.7 0.008
3.1 0.032
1.6 0.017
1.5 0.017
Correc-
ted to
12% C02,
gr/dscf
0.18213
0.10653
0.16163
0.0034
0.0140
0.0052
0.0108
0.0107
0.0153
0.2793
0.0533
0.0337
0.0010
0.0170
0.0039
0.0145
0.0080
0.0079
Percent of
Method 5
Iso- g/dscf repre-
ki- sented by
netic Method 5B g/
% dscf at outlet
105.4
99.5
102.0
105.4
99.5
102.0
105.4
99.5
102.0
101.3
99.4
100.1
101.3
99.4
100.1
101.3
99.4
100.1
a

36.0


137.9


58.8


21.1


91.0

                                                  (continued)

-------
                                    TABLE  3 (continued)

Emissions
Run
number Date Location
SI 11-12-80 Outlet
Outlet
Outlet
Outlet
Outlet
Outlet
Sampling
method Pollutant
5
5B
5
5B
5
5B
Particulate
Particulate
Sulfuric acid
Sulfuric acid
Sulfates
Sulfates
Actual
gr/dscf
0.0417
0.0234
0.0128
0.0011
0.0066
0.0014
Ib/hr
11.0
5.8
3.4
0.3
1.7
0.4
Ib/mm
Btu
0
0
0
0
0
0
.112
.060
.034
.003
.018
.004
Correc-
ted to
12% C02,
gr/dscf
0
0
0
0
0
0
.0511
.0273
.0157
.0013
.0081
.0016
Percent of
Method 5
Iso- g/dscf repre-
ki- sented by
netic Method 5B g/
% dscf at outlet
96
91
96
91
96
91
.5
.9
.5
.9
.5
.9
56.1

8.6

21.2


Run terminated prior to completing traverses because excessive amounts of coal dust being burned.

-------
the Method 5B traverses were 9% to 49% of the sulfuric acid con-
centrations on the Method 5 samples (it was 9% for the single
point run).  Sulfate concentrations measured by Method 5B tra-
verses were 91% to 121% of the concentrations by Method 5 (it
was 21% for the single point run).

The weights of particulate collected in the water washes were
less than 3% of the weights of the particulate in the filters and
acetone rinses.  Appendix A. 3 shows the weights of particulate,
sulfuric acid, and sulfates in the filters, acetone rinses and
water rinses.  Appendix A.3 also cross references the run numbers
in this section with the run numbers in the appendices and on the
computer printouts.  Computer printouts of the particulate,
sulfuric acid, and sulfate emission rates associated with the
filters, acetone rinses and water rinses are provided in
Appendix B.

Table 4 summarizes the particulate, sulfate, and sulfuric acid
collection efficiency of the scrubber, as measured by EPA Method
5B.  The differences between emissions measured by Method 5 and
5B are also summarized in this table.  Finally, this table
summarizes what percent of the differences between Method 5 and
5B particulate emissions can be accounted for by differences
in the amounts of sulfuric acid (and sulfuric acid and sulfates
combined) measured by Method 5 versus 5B.  Run 3 is not included
(because it was terminated prior to completing the traverse).

Forty-nine to 125% of the amount of difference in particulate
emissions measured by Method 5 versus 5B traverses are accounted
for by differences in the amounts of measured sulfuric acid emis-
sions.  For the single point run,, sulfuric acid accounted for 60%
of the particulates.  Sulfuric acid and sulfates combined account
for 65% to 82% of the difference in particulate emissions between
measurements by Method 5 and Method 5B traverses.  For the single
                              12

-------
TABLE 4.  SUMMARY  OF  SCRUBBER EFFICIENCY MEASUREMENTS  AND DIFFERENCES BETWEEN METHOD
          AND METHOD  5B RESULTS AT GENERAL MOTORS CHEVROLET PLANT IN PARMA, OHIO
          (NOVEMBER 10-17,  1980)

Run
number
1


2


3R


SI






Percent of difference
Scrubber collection Method 5 in particulate emis-
efficiency, % minus sions kg/hr (Ib/hr)
based Method SB represented by dif-
on kg/hr (Ib/hr) emissions ference in sulfuric
Pollutant emissions kg/hr Ib/hr acid emissions
Particulate
Sulfuric acid
Sulfates
Sulfuric acid
and sulfates
Particulate
Sulfuric acid
Sulfates
Sulfuric acid
and sulfates
Particulate
Sulfuric acid
Sulfates
Sulfuric acid
and sulfates
Particulate
Sulfuric acid
Sulfates
91%
83%
52%

88%
6%
26%

89%
-250%
52%




2.3
1.5
0.0
1.5

0.7
0.9
-0.3
0.6

2.0
1.2
0.0
1.2

2.4
1.4
0.6
5.1
3.3 64.7
0.0
3.3

1.6
2.0 125.0
-0.7
1.3

4.3
2.1 48.8
0.1
2.8

5.2
3.1 59.6
1.3
Percent of difference
in particulate emis-
sions kg/hr (Ib/hr)
represented by dif-
ference in sulfuric
acid and sulfates
combined

64.7


81.3


65.1




       Sulfuric acid
         and sulfates
2.0
4.4
84.6

-------
point run, sulfuric acid and sulfates combined accounted for 85%
of the particulates.

Table 5 summarizes the duration of the Method 5 and 5B sampling,
the sample volumes, the stack temperatures, the stack flow rates,
and the water content of the flue gas.  Table 6 summarizes the
flue gas carbon dioxide, carbon monoxide, oxygen and nitrogen
measurements associated with the Method 5 and 5B samples.  Differ-
ences between the results associated with the Method 5 and 5B
sampling at the outlet in Tables 5 and 6 are indicative of the
precision of the measurements.

Table 7 summarizes the results of the Andersen cascade impactor
sampling.  The cumulative aerodynamic size distributions are
graphed in Figures 1,  2, and 3.

Table 8 summarizes the results of the plume opacity observations
by EPA Method 9.  This data should be viewed with caution since the
opacity observations  were taken about 15 m (50 ft) downwind of the
mouth of the stack where the scrubber steam plume dissipated; and
visible plumes from other stacks sometimes mixed with the plume from
Boiler No. 1 before the steam dissipated, preventing the opacity
from being measured during portions of the observation periods.

Table 9 summarizes the results of the coal sampling and analysis.
To obtain an indication of the precision of the coal analysis
data, half of the raw coal sample taken during the second Andersen
run at the inlet at Parma was analyzed along with a set of coal
samples from an emission test in Parkersburg,  West Virginia.   The
last column of Table  9 shows the difference in the results between
the analyses of different halves of the same sample.  The most
significant differences were in the reported sulfur content (2.62%
versus 1.35% dry) and in the ash content (8.05% versus 3.92% dry).
Portions of the same  half-samples discussed above were then anal-
yzed a second time; the differences in sulfur content were between
                              14

-------
TABLE 5.  SUMMARY OF DURATION OF SAMPLING, STACK TEMPERATURE,  STACK FLOW RATE,  SAMPLE
          VOLUME, AND WATER CONTENT OF METHODS  5 AND  5B  STACK  SAMPLES  AT THE GENERAL
          MOTORS CHEVROLET PLANT, PARMA, OHIO  (NOVEMBER  10-17,  1980)

Run
1


2


3


3R


SI

Location
Inlet
Outlet
Outlet
Inlet
Outlet
Outlet
Inlet
Outlet
Outlet
Inlet
Outlet
Outlet
Outlet
Outlet
Sampling
method
5B
5
5B
5B
5
5B
5B
5
5B
5B
5
5B
5
5B
Duration
of
sampling,
min
141
144
144
144
144
144
60
60
60
144
144
144
108
108
Measured
stack
temperature
°C °F
167
47
48
176
47
48
159
47
48
165
47
47
48
48
332
117
118
349
117
118
318
117
118
328
116
117
118
118
Stack flow rate
dscm/
min
832
813
827
859
815
834
636
778
774
792
789
792
872
815
dscf/
min
29,361
28,696
29,185
30,332
28,780
29,438
22,443
27,469
27,326
27,966
27,843
27,953
30,802
28,792
Sample
dscm
2.52
2.57
2.50
2.62
2.54
2.50
0.86
0.99
1.00
2.47
2.41
2.41
1.94
1.71
volume
dscf
89.03
90.86
88.23
92.50
89.56
88.32
30.37
34.99
35.38
87.34
85.07
85.25
68.55
68.46
Water
content,
%
6.04
11.68
11.69
5.93
14.31
13.08
11.13
12.12
12.67
6.84
12.04
12.56
11.69
12.18

-------
TABLE 6.  SUMMARY OF INTEGRATED GAS ANALYSIS RESULTS AT THE
          GENERAL MOTORS CHEVROLET PLANT,  PARMA, OHIO
          (NOVEMBER 10-17, 1980)

Run
1


2


3


3R


SI

Location
Inlet
Outlet
Outlet
Inlet
Outlet
Outlet
Inlet
Outlet
Outlet
Inlet
Outlet
Outlet
Outlet
Outlet
Sampling
method
before
Method 3
train
5B
5
5B
5B
5
5B
5B
5
5B
5B
5
5B
5
5B
C02/ %
9.8
9.5
9.9
10.0
9.4
9.8
9.6
9.8
9.4
10.6
10.0
9.3
9.8
10.3
CO, % 02, %
<0.1 9.8
<0.1 10.0
<0.1 9.6
<0.1 9.5
<0.1 9.9
<0.1 9.8
<0.1 9.6
<0.1 9.8
<0.1 10.1
<0.1 9.0
<0.1 9.6
<0.1 10.3
<0.1 10.0
<0.1 9.4
N2, %
80.4
80.5
80.5
80.5
80.7
80.4
80.8
80.4
80.5
80.4
80.4
80.4
80.2
80.3
Dry
molecular
weight,
kg/kg mole
(Ib/lb mole)
30.0
29.9
30.0
30.0
29.9
30.0
29.9
30.0
29.9
30.1
30.0
29.9
30.0
30.0
                              16

-------
TABLE 7.  SUMMARY OF ANDERSEN PARTICLE SIZING RESULTS AT THE GENERAL
          MOTORS CHEVROLET PLANT, PARMA, OHIO (NOVEMBER 10-17, 1980)

Flow rate Percent
Location Run acm/min acf/min isokinetic Stage
Inlet 1 0.011 0.40 103.4 0
1
2
3
4
5
6
7
Final filter
Inlet 2 0.012 0.41 106.0 0
1
2
3
4
5
6
7
Final filter
Outlet 1 0.017 0.60 101.2 0
1
2
3
4
5
6
•~i
i
Final filter
Size range,
pm
>15
10.6 -
7.3 -
4.9 -
3.15 -
1.56 -
0.917 -
0.67 -
0 -
>15
10.5 -
7.1 -
4.8 -
3.05 -
1.55 -
0.95 -
0.66 -
0 -
>11
7.7 -
5.2 -
3.6 -
2.25 -
1.12 -
0.70 -
0.47 -
0 -
.6
15.6
10.6
7.3
4.9
3.15
1.56
0.97
0.67
.4
15.4
10.5
7.1
4.8
3.05
1.55
0.95
0.66
.5
11.5
7.7
5.2
3.6
2.25
1.12
0.70
0.47
Percent
in size
range
78.8
3.52
0.70
0
1.41
10.56
1.41
0.70
2.82
76.92
1.22
6.88
0
0.81
4.86
0
3.64
5.67
8.76
0
0
4.38
4.38
24.08
16.06
21.90
20.44
Cumulative
percent,

-------
00
                                   0      t0   30  y.o r.o »•  o l.o


                           PARTICLE AERODYNAMIC DIAMETER - pm
16.
                  Figure  1.   Cumulative size  distribution  for  first inlet Andersen
                              run at General Motors Chevrolet Plant in Parma,  Ohio
                              (November 12, 1980).

-------
                       L.»   3 0  y.o f.o ».« 10 '••*'„

                     PARTICLE AERODYNAMIC DIAMETER
Figure 2.   Cumulative size  distribution  for second inlet  Andersen
            run at General Motors Chevrolet Plant in Parma,  Ohio
            (November 13, 1980).

-------
                  PARTICLE AERODYNAMIC DIAMETER - \im
Figure 3.  Cumulative size distribution  for  scrubber exhaust Andersen
           run  at  General Motors Chevrolet Plant in Parma, Ohio
           (November 14,  1980).

-------
TABLE  8.    SUMMARY  OF PLUME  OPACITY  OBSERVATIONS AT  THE
               GENERAL  MOTORS  CHEVROLET  PLANT  IN PARMA,  OHIO
               (NOVEMBER  10-17,  1980)
           Date:   11/11/80
           Type of Discharge:  Stack
           Height of Point of Discharge:   100 ft
           Wind Direction:  N
           Color of Plume:  White
           Observer Name:  C Clark
           Distance from Observer
             to Discharge Point:   250 ft
           Direction of Observer
             from Discharge Point: SW
           Height of Observation  Point:  Ground Level
           Description of Background:  Water tower
           Distance Downwind of Stack where
             Steam Plume Dissipated and
             Plume Opacity Measured:  50 ft
Type of Plant:  Industrial Boiler
Location of
  Discharge:  Scrubber exhaust
Description of Sky:  Overcast
Wind Velocity:  7-10 mph
Duration of Observation:  100 min
                  o
                  
13:
14:
14:

:30
:36
:42
:48
•54
;00
:06
:12
•IB
?4
•30
;3ft
:42
•4fi
:54
:00
:06

End
12:
12:
12:
12:
1?;
13:
13:
13;
13'
13;
13'
13;
13:
13'
13:
14:
14:

:35
:41
;47
:53
;59
:OS
:11
:17
?3
?9
;35
:41
:47
•S3
:59
:05
:09

Sum
260
275
220
165
240
200
195
120
145
190
220
95
285
240
200
?6B
180

Opacity
Average


















all sets
10.
11,
11 .
10,
10
10,
10,
10,
10
10
10
10
1,3
1?
12,
11
12.

11
.8
•s..
.od
.3d
.0
V
,ftd
,od
7d
6d
Od
.6"
,0d
nd
.5"
,n
.9d

.1%
                           Interference from a steam plume  from
                           another stack prevented the opacity
                           from being measured during portions of
                           this six minute time interval.
                                     1             2

                                       TIME, hours
            Date:  11/12/80
             A plume  from Stack #4 was mixed with the plume from the source being
             tested and prevented obtaining valid opacity readings.
                                         21

-------
                        TABLE  8   (continued)
      Date:  11/34/80
      Type of Discharge:   Stack
      Height of Point of  Discharge:  100 ft
      Wind Direction:  SW
      Color of Plume:  White, black for
                       last 41 min
      Distance from Observer
        to Discharge Point:   1,050 ft
      Direction of Observer
        from Discharge Point:  SE
      Height of Observation Point:  100 ft
      Description of Background:  Water tower
      Distance Downwind of Stack where
        Steam Flume Dissipated and
        Plume Opacity Measured:  50 ft
                         Type of Plant:  Industrial Boiler
                         Location of
                           Discharge:  Scrubber exhaust
                         Description of Sky:  Partly cloudy
                         Wind Velocity:  10-12 mph
                         Duration of Observation:  251 min
Summary of average opacity
Set

Time
number Start
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
9;
9:
9:
9:
9:
9:
Q ,
9:
9:
9:
10:
10:
10:
10;
10:
10:
10:
10:
10:
10:
11;
:00
:06
;12

24
:30
:36
:42
48
:54
:00
;06

:18
^24
:30
:36
:42
:48
:54
:00

End
9:
9:
9;
9:
9:
9:
9:
9;
9:
9:
10;
10;
10:
10
10:
10:
10:
10
10
10:
11
;05
: 11
:17
123
:29
:35

^47
:53
:59
:05

: 17
^23
:29
:35

^47
:53
:59
:05
Opacity
Sum
170a
120a
185a
80a
200f
210a
200a
225a
135a
225a
215a
195a
115a
130a
205a
120a
195a
180a
230a
160a
170a
Average
9.
10,
9.
10.
9.
10,
10.
9.
10.
11.
9.
9.
7.
9.
9.
7.
9,
9,
10,
10,
7,
4a
oa
',7a
,oa
,5a
•°a
,0a
,8a
,4a
.3a
,8a
,8a
.7a
,3a
,8a
1
'.8*
.5a
.oa
,0a
,7a
Summary of average opacity
Set
number
22
23
24
25
26
27
28
29
30
31
32
33
34

36b
37
38
39
40
41
42C
Time
Start
11;
11:
11:
11:
11:
11:
11;
11:
11:
12
12:
12;
12:
12:
12;
12:
12
12
12
13:
13
:06
. 12
:18
:24
;30
:36
:42
:48
:54
:00
:06

: 18
•24
;3ob
:36
:42
:48
:54
:00
:06
End
111
Hi
11:
11:
11;
11:
11:
11:
11:
12:
12:
12:
12:
12:
12:
12:
12:
12:
12;
13;
13:
ill
: 17
:23'
:29
:35
;41
i47
: 53
:59
:05
: 11
:17
:23

:35
:41
:47
:53
:59
:05

Opacity
Sum
60a
130a
150a
160a
120a
250a
220a
165a
60a
90a
145a
130a
135a
165a
545a
325a
420a
395a
380a
710a
545C
Average
5,
5,
6.
7,
5,
12
12,
9,
10
11
12.
10,
11
11
23
16
23
24
17
30
27
a
. 9a
,5a

•7a
.5a
.2a
•7a

.3a
la
:sa
,3a
.8a
*7
'.3a
.3a
.7a
.3a
.9a
.3C
      Average, all sets
                                    11.£
    35

    30
    20
    10
       Interference from a steam  plume  from another stack prevented the  opacity
       from being measured during portions of this six minute time  interval.

       Plume color changed to black at  this time.
      cFinal set from a five minute observation period.
              RUN T-3
t
I—I
o
      0
1
                                      TIME,  hours
      Date;  11/14/80
        Plumes from two other stacks were mixed with the plume from the  source being
        tested and prevented obtaining of valid opacity readings.
                                       22

-------
KJ
00
                  TABLE  9.   SUMMARY  OF COAL ANALYSIS AT  THE GENERAL
                                PARMA,  OHIO  (NOVEMBER  10-17,  1980)
MOTORS  CHEVROLET  PLANT,
Parameter
Total mo i stut'e, %
Heating value J/g (Btu/lb)
Dry basis

As received

Ash, %
Dry basis

As received

Sulfur, %
Dry basis

As received

Carbon. %
Dry basis
Hydrogen, %
Dry basis
As received
Oxygen, %
Dry basis
As received
Nitrogen, %
Dry basis
As received
Sieve sample weight, g (Ib)

Percent by weight collected
25. 4 mm sieve (1.0 in.)
«4 sieve, 4.75 era (0.187 in.)
816 sieve, 1.18 nun (0.046 in.)
#30 sieve, 0.60 mm (0.024 in.)
S200 sieve, 0.075 mm (0.003 in.)
Passing through all sieves
Cumulative percent by weight
less than sieve sizes
25 .4 nun (I in. sieve )
«4 sieve. 4.75 mm (0.187 in.)
HIA sieve, 1.19 ss {0 04* ;" '
»30 sieve, 0.60 mm (0.024 in.)
H200 sieve, 0.075 mm (0.003 in.)
Ruti 1
3.

32,
(14,
31,
(13,

5.

4.


1.

1.


79.
75 ,
6.
5.

8.
8.

1.
1.
15

900
100)
800
700)

13

97


09

05


6

.0
6

.6
1

.59
50
2,019.3
(4.

22.
73.
2.
0.
1.
0.


77.
4.
i
1.
0.
45)

46
01
49
41
14
49


54
53
nji
.63
.49
Run 2
3.

33,
(14,
31,
(13,

4.

4


1.

1.


79.
74
5
5

9
8

1
I
66

000
200)
800
700)

.43

.26


.48

.43


,9
. 9
.6
.2

.1
.6

.37
.29
1,981. B
(4

11
79.
5
l!
2.
0


88
8.
3
2
0
.37)

.40
.79
.14
.10
.55
.0


.60
.79
. 55
.55
.0
Run 3
4.

31.
(13,
30,
(13.

7.

7.


1.

1.


76.
70 .
5.
5.

8.
8.

1.
1.
63

BOO
700)
400
100)

40

06


91

82


7
9
5
1

7
1

36
26
1,200.7
(2.

4.
75.
11.
3.
4.
0.


95.
19.
T
4.
0.
65)

61
55
95
23
48
18


39
84
oa
66
18
Run 3R
3.74

33.000
(14,200)
31,700
(13.700)

3.51

3.38


1.43

1.38


81.3
75.9
6.0
5.6

8.5
7.9

1.58
1.47
2,487.5
(5.48)

14.58
82.46
2.03
0.26
0.67
0.20


85.42
3.16
1 11

-------
0.02% and 0.06% (dry) and the differences in ash content were
between -0.18% (dry) and 0.15% (dry).  This suggests that the poor
reproducibility of the coal sulfur and ash content results in
Table 9 was caused by real differences in the composition of coal
from one piece to another and the small amounts of sample that
were split in half, prepared, and analyzed.

Appendix F.4 contains a summary of available information on the
precision and accuracy of the sulfate and sulfuric acid analysis.

Two quality control samples of sulfuric acid were prepared from
0.02 N sulfuric acid and analyzed along with the Parma stack
samples.  The measured amounts of sulfuric acid were 101% and 121%
of the reported "true" values.  The results of the analysis of two
EPA sulfate quality control filter strip samples (with more than
2 mg of sulfate) showed the measured values to be 87.5% and 100%
of the "true" amounts of sulfate.  For two filter strips with
less than 1.0 mg/sample, the percentages were 181% and 195%.  The
possibility exists of differences in the physical and chemical
composition of the stack samples (compared to those of the qual-
ity control samples) resulting in differences in the accuracy of
analysis of the stack samples from that reported for the quality
control samples.

Appendix C contains the Method 9 plume opacity measurements.  These
readings were taken downwind of the mouth of the stack at locations
after the scrubber water plume dissipated.
                              24

-------
                            SECTION 3
               PROCESS DESCRIPTION AND MONITORING

PROCESS DESCRIPTION

Boiler System

The steam plant at CMC contains four boilers rated at a total
combined steam generating capacity of 145,000 kg/hr (320,000 Ib/hr),
two with a nominal capacity of 45,000 kg/hr (100,000 Ib/hr) and
two with a nominal capacity of 27,000 kg/hr (60,000 Ib/hr).
Figure 4 shows a flow diagram of a typical boiler and controls.
Each is fired by a spreader stoker with traveling grates and opera-
tes with variable excess air rates in the 100% range.   The larger
boilers (1 and 2) are equipped with economizers that lower flue
gas temperature to 135°C (275°F),  and the smaller boilers (3 and 4)
operate at a flue gas temperature of 302°C (575°F).  Each boiler
is fitted with mechanical dust collectors (multiclones) for primary
particulate control.  Normal burning of medium- to high-sulfur
(2% to 3%) eastern coal plus occasional firing of low-sulfur waste
oil results in flue gas generally containing 500 ppm to 1,300 ppm
S02 by volume.  Boiler No.  1, with a capacity of 45,000 kg/hr
(100,000 Ib/hr) of steam, was tested according to the scenario
shown in Table 1, page 4.

Scrubber System

Each boiler is equipped with a regenerable liquor, double-alkali
scrubbing unit.  Table 10 lists design, operating, and performance
characteristics of the scrubbers.
                              25

-------
                                                                                .,.-   REMAINING  -Jf'
                                                                            ^,  '- PARTI CULATE>LUME
                                                                           STEAM PLUME
COAL WEIGHING
   DEVICE
                                                                     27 - 38°C
                                                                    (80 - 100°F)
                                                  SCRUBBER INLET
                                                  SAMPLING SITE
                                                                               SCRUBBER EXHAUST
                                                                                SAMPLING SITE
                                       SPREADER
                                      STOKER BOILER
                                         WITH
                                      ECONOMIZER
DUAL ALKALI
 SCRUBBER
       Figure 4.   Schematic diagram  of Boiler No.  1  and  associated equipment
                    at  General Motors  Chevrolet Plant  in Parma,  Ohio.

-------
   TABLE  10.   DESIGN,  OPERATING, AND PERFORMANCE CHARACTERISTICS
                OF DOUBLE-ALKALI SCRUBBING SYSTEM
           Application
      Four coal-fired stoker boilers
Fuel characteristics
System design


Process mode

Pressure drop

Status

Startup date

Inlet gas conditions
  Flow rate
  S02
  02
  Particulate

S02 outlet concentration

S02 removal efficiency

Scrubbing solution characteristics
  PH
  Total sodium
  Active sodium
  Calcium ion

Soda ash makeup

Lime utilization

Filter cake solids

Filter cake disposal
Coal:   25,600  to  31,400 J/g  (11,000 to
  13,500 Btu/lb),  1.5% to 3.0% sulfur
  (waste oil also burned, but in small
  quantities compared with coal)

Four three-stage  multiventuri flexitray
  scrubber modules

Dilute active  alkali

25 to 33 cm H20 (10  to 13 in. H20)

Operational

March 1974
30.9 ms/s at 27°C (65,500  acfm at 80°F)
800 to 1,300 ppm
Not available
0.7 g/m3 (0.3 gr/scf)  (dry)

20 to 130 ppm
90% to 99%
5.5 to 7.5
0.58 to 0.96 molar
0.087 to 0.13 molar
305 to 490 ppm

>0.1  mole/mole S02 removed

1.32 to 1.90 mole/mole S  in cake
40% to 55%
Offsite landfill
 Nonsteady-state operations  result in makeup rates of 0.028 to 0.05  mole
 N02 per mole S02 removed.
                                      27

-------
The primary function of the scrubber is to reduce S02 emissions in
the boiler flue gas by reaction with an alkali scrubbing liquor.
A secondary function is the simultaneous removal of particulate
matter.  The functions of the liquor regeneration system are to
precipitate the reacted sulfur (in the form of sodium-sulfur com-
pounds) and vacuum filter the solids for disposal while simultane-
ously regenerating the scrubber liquor via an ion exchange
mechanism.

Sulfur dioxide is removed from the flue gas in the scrubber by
reaction with the alkali liquor, 0.1N NaOH.  S02 is combined
chemically with the NaOH to form sodium bisulfite by the reaction
mechanism shown below.

                  2NaOH + S02 - *-Na2S03 + H20
                  Na2S03 + S02 + H20 - *-2NaHS03

Simultaneously some sodium sulfite reacts with the oxygen in the
flue gas to produce sodium sulfate:

                   Na2S03 + 1/202 - *» Na2S04

Scrubber liquor regeneration is achieved by reacting calcium
hydroxide (lime slurry) with the soluble sodium-sulfur compounds
formed in the scrubber.  Calcium sulfite and calcium sulfate
precipitate out by the mechanism shown below while the sodium
remains in solution with the hydroxide, thus regenerating the
scrubber liquor.
          2NaHS03 + Ca(OH)2 - ^Na2S03 + CaS03   + H20
           Na2S03 + Ca(OH)2 - >-CaS03 1 + 2NaOH
           Na2S04 + Ca(OH)2 - *-CaS04 | + 2NaOH
                              28

-------
Figure 5 shows a flow diagram of the scrubber system.  Flue gases
enter through a prequench section at the bottom of each unit and
then flow in a countercurrent direction through an aqueous; sodium
hydroxide sulfite-bisulfite solution.  The scrubber is a two-tray,
impingement-type unit with feed and recycle streams entering at
the top.  It was modified from the original three-tray design in
an effort to reduce scaling.  The absorption trays (Koch) have
movable self-adjusting bubble caps that respond to variations in
gas flow.  Pressure drop through each unit is typically 19 cm
(7.5 in.) H20, and the maximum liquid-to-gas ratio is 2.7 L/m3
(20 gal/1,000 ft3).  For control of entrained liquor, each unit
is equipped with a high-efficiency mist eliminator.  Acting as
an impingement separator, the mist eliminator is composed of corru-
gated profile plates assembled with phase separating chambers.

The spent scrubbing liquor, containing absorbed sulfur compounds
and fly ash, is bled from the scrubber recycle tank to the chemical
mix tank where lime (CaO) is added for sodium hydroxide regenera-
tion.  After a final clarification step to remove solids (calcium
sulfur compounds and fly ash) and a softening step to remove excess
calcium, the liquor is ready to be fed back to the scrubber.  Vacuum
filters are used to dewater the sludge from the clarifier bottoms
after washing to remove NaOH.

MONITORING OF PLANT PROCESSES DURING TESTING

The boiler and scrubber processes were monitored during testing
for the following reasons:

1.  To insure that the boiler was operating near capacity and at a
    relatively steady state;

2.  To insure that the boiler was operating within normal tolerances;
                              29

-------
                                                     NaOH
U)
o
                                                                                             FILTRATE
                                                                                              PUMP

                                                                                            VACUUM FILTERS

                                                                                               *»
                         CAUSTIC
                       STORAGE
                                                                                        SODA ASH
                                                                                        FEED PUMP
                                      CARBON DIOXIDE
                                      STORAGE TANK
                         Figure 5.   Flow  diagram  of  double-alkali scrubbing unit.

-------
3.  To insure that the scrubber was operating normally at or near
    the design pressure drop and that no liquor feed abnormalities
    occurred;

4.  To record process data used in determining emission rates
    (pounds emitted/106 Btu); and

5.  To record process data used as a qualitative guideline to
    document normal boiler operation.

Eight process parameters were monitored during emissions testing.
Tables 11 through 14 are compilations of these data.  All process
parameters recorded are described below.

Steam Production Rate

The steam production rate was monitored by two instruments on the
boiler control panel.  The instantaneous values in Table 11 through
14 were recorded by a continuous monitor while the average values
at the bottom of the tables were obtained from a steam integrator
monitored over the entire test period.  Fluctuating steam demand
on the power plant caused pressure variations in the steam lines
that slightly affect the accuracy of all steam production monitor-
ing equipment.  The degree to which these fluctuations can affect
the accuracy is unknown.

Flue Gas Oxygen

The percent oxygen was sampled on a continuous basis at the flue
gas inlet to the economizer.  The accuracy of this instrument was
considered poor by company personnel.  It was monitored for unusual
fluctuations which might indicate abnormal boiler operation.

The flue gas oxygen meter in conjunction with the opacity meter
provided useful information for the boiler engineer.  These
readings allowed the engineer to minimize smoke production by

                              31

-------
TABLE 11.  POWER PLANT OPERATING PARAMETERS (11/12/80,  RUN SI)

Time
5:34
5:35
5:50
6:12
6:24
7:02
7:19
7:34
7:50
8:00
8:02
Average
Average
Flue
Steam gas
production, oxygen,
Ib/hr %
(START)
93,500 5.2
89,500 5.7
93,000 4.9
Test interrupted in
92,500 5.2
95,000 4.9
92,000 5.1
92,000 5.0
92,500 5.0
(FINISH)
coal feed
steam production
Mechanical
Scrubber Inlet collector
pressure scrubber gas pressure
drop, Scrubber temperature, drop,
in. H20 pH °F in. H20

7.5
7.4
7.6

6.8 239 3.4
6.8 239 2.6
6.8 239 2.6
Gas Smoke
temperature, density,
°F %

502 2
500 1
504 1
order to change sampling train filters.
7.5
7.6
7.6
7.7
7.2

7,
90,
6.8 241 2.7
6.7 243 2.9
6.6 242 2.7
6.6 243 2.7
6.6 243 .2.8

415 Ib/hr
385 Ib/hr
509 1
510 1
505 1
508 1
508 1




-------
                 TABLE  12.   POWER PLANT OPERATING PARAMETERS (11/13/80,  RUN 1)
LO

Time
9:22
9:22
9:37
9:54
10:09
10:35
10:50
11:14
11:38
11:35
12:02
12:23
12:35
12:36
Average
Average
Flue
Steam gas
production, oxygen,
Ib/hr %
(START)
98,500
93,000
96,000
89,000
94,000
92,000
93,500
95,000
Outlet filter
94,000
94,000
99,000
(FINISH)
coal feed
steam production

5
5
5
5
5
5
5
5

.3
.2
.4
.1
.5
.4
.0
.1
change
5
5
5



.7
.5
.1



Scrubber
pressure
drop,
in. H20

8.2
8.2
8.3
8.1
8.2
7.5
7.4
7.5
reguired
7.7
7.8
7.7

7,500
92,850
Scrubber
PH

6.
6.
6.
6.
6.
6.
6.
6.
due to
6.
6.
6.

Ib/hr
Ib/hr

8
8
8
8
8
8
8
8
high
8
8
8



Inlet
scrubber gas
temperature,
op

255
255
255
253
254
253
251
253
loading.
256
260
263



Mechanical
collector
pressure
drop,
in. H20

3
3
3
3
3
3
3
3

3
3
3




.2
.2
.2
.2
.1
.2
.1
.2

.3
.4
.3



Gas
temperature ,
op

519
515
512
512
510
502
510
510

520
520
525



Smoke
density,
o,
o

3
2
1
1
2
1
1
8

4
5
25




-------
TABLE 13.   POWER PLANT OPERATING PARAMETERS (11/13/80, RUN 2)

Time
6:51
6:52
7:09
8:10
8:31
8:55
9:15
10:26
10:57
11:15
11:15
Average
Average
Steam
production,
Ib/hr
(START)
93,000
93,500
95,500
94,000
95,000
97,000
95,000
94,500
96,000
(FINISH)
coal feed
Flue
gas
oxygen ,

5.7
5.3
5.6
6.0
5.3
5.8
5.8
5.6
5.7


steam production
Scrubber
pressure
drop,
in. H20

8.2
8.2
8.1
7.8
7.9
8.2
7.6
7.7
7.6

8,196
95,524
Scrubber
PH

6.9
7.0
7.1
6.6
6.8
6.8
7.2
7.8
7.4

Ib/hr
Ib/hr
Inlet
scrubber gas
temperature,
°F

264
270
260
265
265
270
268
270
270



Mechanical
collector
pressure
drop,
in. H20

3.5
3.5
3.6
3.4
3.5
3.5
3.6
3.6
3.6



Gas
temperature,

520
532
530
523
521
560
525
530
530



Smoke
density.

3
2
2
4
4
2
6
8
2




-------
TABLE 14.  POWER PLANT OPERATING PARAMETERS (11/14/80,  RUN 3R)

Time
7:08
7:08
7:28
7:55
8:21
8:57
9:26
9:47
10:14
10:14
Average
Average
Steam
production,
Ib/hr
(START)
87,000
92,000
93,500
94,000
96,000
94,000
93,000
91,500
(FINISH)
coal feed
Flue
gas
oxygen,

5.8
5.5
5.4
5.0
4.8
5.2
5.3
5.2


steam production
Scrubber
pressure
drop,
in. H20

7.4
7.4
7.2
7.2
7.3
7.4
7.3
7.3

7,677
93,065
Scrubber
PH

6.8
6.8
6.8
6.8
6.8
6.8
6.8
6.9

Ib/hr
Ib/hr
Inlet
scrubber gas
temperature,
Op

248
246
247
247
249
250
249
248



Mechanical
collector
pressure
drop,
in. H20

2.9
3.0
2.9
3.1
3.0
3.0
3.0 .
3.0



Gas
temperature ,
op

508
508
509
510
510
510
507
505



Smoke
density,

2
1
1
1
1
1
1
2




-------
adjusting the excess air flow to the boiler (as indicated by the
oxygen meter) until the optimum flow was achieved (as indicated
by a subsequent decrease in opacity).  The oxygen meter also
indicated that if opacity increased without a corresponding
decrease in flue gas oxygen, other problems such as poor coal
feed stock, improper grate speed, or a combination of both could
be responsible for the increase in opacity.

Scrubber Pressure Drop

The scrubber pressure drop was measured across the inlet and outlet
on a continuous basis.  These readings served as an indication of
plugging on the trays.  An increase in pressure drop would indicate
that plugging has occurred in the scrubber.  It should be noted
that the pressure drop will vary slightly with boiler load.  If
plugging is thought to have occurred, comparisons must be made
with pressure drops recorded when the scrubber was known to be
unrestricted and operating under the same boiler load conditions.

Scrubber pH

The pH was monitored at the top tray of the two-tray countercurrent
flow scrubber.  The pH was used as a primary means of controlling
the entire SO2 scrubbing system by controlling the amount of
scrubber feed liquor added to the recycle line.   In turn,  the
total volume of this flow determines how much bleed liquor is
sent to the chemical mix tanks,  how much lime is added for re-
generation, how much sludge is formed in the clarifier,  and how
much soda ash is needed for softening.

The original scrubber design called for a pH of 6.0 at the bottom
tray.  Modifications to the original design,  discussed in the
"Scrubber System" section,  resulted in monitoring the top tray
pH and maintaining it at a level of 6.8.
                              36

-------
Scrubber Inlet Gas Temperature

The inlet gas temperature is monitored at the gas prequench section
of the scrubber.  A high inlet temperature would indicate that the
quench section is not operating normally, resulting in a flue
gas temperature to the scrubber that is not within the normal
design range.

Mechanical Collector Pressure Drop

The pressure drop across the multiclone collector was monitored to
insure that it operated normally.  A high pressure drop would
indicate that flow through the collector was restricted, possibly
due to obstructions in the cyclones or to hopper overloading.

It should be noted that boiler load affects the pressure drop
across the collector.  When plugging was thought to have occurred,
the procedure described under "Scrubber Pressure Drop" was followed.

Gas Temperature

The flue gas temperature was monitored at the boiler outlet (inlet
to the economizer) on a continuous basis.  The flue gas temperature
typically varies with boiler load.

Smoke Density

The smoke density was monitored at the mechanical collector outlet
by an outdated in-stack Baily opacity meter.  Plant personnel
commented that the instrument's accuracy and reliability were
marginal.   For this reason these data should not be used as a
quantitative measure of the unscrubbed flue gas smoke density.
However, the data were useful as a qualitative gauge, indicating
how well the boiler responded to adjustments made in an effort to
reduce smoke production.
                              37

-------
Average Coal and Steam Values

The average coal feed rates and steam production rates reported at
the bottom of Tables 11 through 14 were measured by integrators.
The steam integrator was located on the boiler control panel and
was subject to the same pressure fluctuations discussed under
"Steam Production Rate."  The coal integrator was located on the
boiler feed hopper.  This integrator measured how many batch
loads of coal [in 91-kg (200-lb) increments] were fed into the
boiler over an entire test period.

COMMENTS ON TESTS

Boiler Steam Production Rate

Throughout all of the tests, boiler load was maintained at a
relatively constant value.  Although demand on the power plant as
a whole was fluctuating with time, steam demand on Boiler No. 1
was held steady by creating a false load (venting steam to the
atmosphere) when demand was low.  This procedure allowed boiler
engineers to maintain a steady steam demand above 41,000 kg/hr
(90,000 Ib/hr; 90% capacity) for Boiler No. 1.

Coal Feedstock

Plant personnel reported that higher than normal amounts of coal
fines were in the coal feedstock during Runs 2 and 3.  As a result,
smoke production was increased and adjustments normally used to
control smoke (underfire air, overfire air, and bed adjustments)
were ineffective.  Observation of the coal bed revealed that a
significant portion of the bed,  consisting of these coal fines,
would blow-off and become entrained in the flue gases to the
scrubber.  Smoke density readings, as monitored by the in-stack
opacity meter, ranged between 3% and 8% throughout test No. 2.  It
                              38

-------
should be noted that momentarily spikes up to as high as 25%
occurred during the final minutes of the test.  The fines became
so high during Run 3 that the run was terminated prior to com-
pleting the traverse.
                              39

-------
                           SECTION 4

                   LOCATION OF SAMPLING POINTS

Two locations along the exhaust path of Boiler No. 1 were tested - a
site between the multiclones and the scrubber (the scrubber inlet)
and a site on the stack after the scrubber (the scrubber exhaust).
In addition, the coal being fed into the boiler was sampled.  The
sampling sites are shown in Figure 4 (Page 26).

SCRUBBER INLET

Diagrams of this site are shown in Figure 6.   It is located on a
1.8 mx 1.2 m (6 ft x 4 ft) ID rectangular duct located below a
damper and elbow coming from the multiclones  and the boiler.  Three
7.6 cm (3 in.) ID sampling ports are located  on this duct,  about
2.7 m (9 ft) below the damper and about 0.8 m (2 1/2 feet)  above
a widening of the ductwork leading to ID fans and the scrubber.

The ports are about two equivalent duct diameters downstream of a
flow disturbance and half a diameter upstream.  A 49-point traverse
in a 7 x 7 matrix would be needed for Method  5B testing by the full
EPA reference method traverse procedures.  Such a traverse would
make the installation of additional sampling  ports necessary.  Be-
cause EPA needed results quickly, a decision  was made to sample
sixteen points along each of three diameters  in line with the
existing ports,  instead of installing new ports for the Method 5B
samples.

The Andersen particle size samples were taken through the center
port, 51 cm (20.1 in.) from the inside wall of the duct.  This
                              41

-------
                        SIDE VIEW
                                 DETAIL A
                             1/2" PIPE
       DAMPER (DISTURBANCE)

        1/2 PIPE (PRESSURETAP)
         WHICH MAY INTERFERE
                                                                    GRATING
                                                             TO SCRUBBER
                                                        3"PORTS. TYPICAL FEMALE
                                                        THREADS ~ 3" DEEP WITH
                                                        MALE PLUG TO SEAL
                                         1/2" PI PECAN BE
                                            REMOVED
                                      -4'
                                                                                   DETAIL "A
             SECTION A-A

              TOP VIEW


NOTE:-8ft. IN FRONT OF PORTS TO
    ANOTHER DUCT
NOTES: (1)~ 4" INSULATION ON ALL SIDES OF DUCT
      (2) 3rd LEVEL, SAMPLE LOCATION INSIDE BUILDING, POWER
         ~ 50 FEET AWAY (SPECIAL ADAPTERS WILL BE FURNISHED)
      (3) GOOD AREA FOR CLEAN-UP
         Figure 6.    Scrubber  inlet  sampling site  at  Parma,  Ohio GM Plant.

-------
location has a velocity approximately equal to the arithmetic mean
of the velocities at the 48 traverse points.

SCRUBBER EXHAUST

Figure 7 provides a side view and a top view of the site on the
roof of the boiler house,  where scrubber exhaust was sampled.  Three,
7.6 cm (3 in. ID) sampling ports were located on three sides of a
1.3 m (53 in. ID) stack at this site.  These ports were about seven
stack diameters above a fitting connecting the scrubber to the stack
and about four stack diameters below the mouth of the stack.

During full Method 5 and 5B traverses,  12 points were sampled along
each of two perpendicular diameters of the stack.

The Andersen particle size sample was taken at a location having a
velocity approximately equal to the arithmetic mean of the veloc-
ities measured at the 24 Method 5 and 5B traverse points.  This
location is 47.2 cm (18 5/8 in.) from the inside wall of the stack,
from port B (shown in Figure 7).

For the Method 5 and 5B sampling at a single point, the nozzles
were initially placed next to each other at the center of the stack.
Before the sampling began, the probes and nozzles were then moved
back far enough apart so that there was no evidence (from the pitot
tube pressure drops) of either train interfering with the other
train's pitot tube velocity measurements.  The nozzles are estimated
to have been within 7.6 cm (3 in.) of each other during the single
point tests.

COAL SAMPLING SITE

Coal was sampled with a shovel from the middle port of the coal
feed chute leading directly into the boiler coal feeder overthrow
rotor.  This site is located after the coal weighing device.
                              43

-------
           N
       -H	1—
        LADDER
            SHED
          WITH TABLE
                                          DISTURBANCE
                                                     18'
                                                 ROOF
                           PORT B
PORT A
                  H	h
                      LADDER
                                                                  •4'-6"
                                                                    O
                                                                                      3" PORTS
                                                                                  (COUPLINGS.  2.5"
                                                                                    TO INSIDE WALL)
                                                  31'
\
                                                                                   *- DISTURBANCE
            TOP VIEW
                                                                 SIDE VIEW
                            NOTE: TWO CONSOLES CAN FIT INSIDE. POWER
                                70ft. AWAY -100ft. TO GROUND.
                                EQUIPMENT MAY BE BROUGHT UP BY ROPE.
Figure  T.   Scrubber exhaust  sampling site  at Parma,  Ohio  GM Plant.

-------
                            SECTION 5
                 DESCRIPTION OF SAMPLING TRAINS

PARTICULATE SAMPLING TRAINS

EPA Method 5 sampling trains were used for determining Method 5
and Method 5B particulate mass emission rates (see Figure 8).
The temperature control system settings on the trains used for
Method 5B sampling were set to maintain filter temperatures of
177 ±14°C (350 ±25°F) instead of the 120 ±14°C (298 ±25°C) used
for Method 5 sampling.  Heated glass-lined probes and Reeve Angel
Type 934 AH filters were used for the Method 5 and 5B testing.
The filter temperatures were monitored using thermocouples in-
stalled on the back halves of the filter holders of each sampling
train.  A heated glass cyclone was used between the sampling
probe and the filter holder to minimize filter changes.

PARTICLE SIZE DISTRIBUTION SAMPLING APPARATUS

Sampling for particle size was performed using an Andersen cas-
cade impactor with seven stages and a back-up filter.  Figure 9
is a schematic drawing of the sampling train.  It consisted of
the following equipment, listed in order of the air flow:  a
0.47 cm (0.187 in.) diameter probe tip; a standard Andersen head;
a 1.2 m (4 ft) stainless steel probe; a Smith-Greenburg impinger
with water,  then an impinger charged with color-indicating silica
gel; an EPA-5 console equipped with a dry gas meter, a digital
electronic thermometer, and an inclined manometer.  Reeve Angel
Type 934 AH substrate was used on each stage of the Andersen
sampler.
                              45

-------
            .TEMPERATURE
              SENSOR
0 75 TO 1 in
THERMOMETER
  CHECK VALVE
                                                                                     THERMOMETERS
                                                                     VACUUM LINE

                                                                        ACUUM GUAGE
                                          HEATED AREA
                                               FILTER HOLDER
  NOZZl E/HEATED -
      PROBE     £
    REVERSE TYPE
     PI TOT TUBE
               PITOT MANOMETER
                                   IMPINGERS
                                                                                        ORIFICE
                                                                                        AND
                                                                                        MANOMETER
                                                             ICE BATH
         Figure 8.   Particulate  sampling train -  EPA  Methods 5  and 5B.

-------
             Cascade
             Impaclor
Nozzle •
                             Stack Wall
 ,

L
                                      Probe
                t
        Reverse-Type
         Pilol Tube
T^X
                                                                              Thermometer
                                                          Vacuum
                                                           Tubing
                                                      Check Valve
Ice
                                                          Water
             Ull
             ~J f

                                          .Silica
                                           Gel
                       Pilot Manometer/ Thermometers   BVPass Valve       Impingers

                          Orifice.
                                                                       Vacuum Gauge
                                                             Air Tight Pump
                                                                                                Vacuum
                                                                                                 Line
          Figure  9.   Particle size distribution  sampling apparatus - Andersen  2000,  Inc.

-------
                            SECTION 6
                SAMPLING AND ANALYTICAL PROCEDURES

SUMMARY

The general sampling and analytical procedures used and the number
of tests performed are described in Section 2 of this report.
Particulate emissions downstream of the scrubber were monitored
by EPA Methods 5 and 5B.  Particulate emissions upstream were
measured by Method 5B.  The flue gas velocity, temperature, flow
rate, oxygen content, and carbon dioxide content were measured by
EPA Methods 1-4.  The sulfuric acid and sulfate content of the
particulates collected was measured by techniques supplied to MRC
by EPA October 10,  1980.  The plume opacity was measured by EPA
Method 9.  Grab samples of coal were taken before, during, and
after every Method 5 and 5B emission test and analyzed by ASTM D3176
ultimate analysis,  ASTM D2015-66(72) heating value analysis, and a
modified version of ASTM D410-38 sieve size analysis.  Further
details on the sampling, analysis,  and quality assurance procedures
are shown below.

FIELD SAMPLING

The Method 5 and 5B testing during each run was approximately
simultaneous.  A decision was made by EPA not to shut down all
sampling trains when filters were changed on one of the trains.
This resulted in small deviations from true simultaneous sampling.
The exact times that samples were taken, filters changed, etc. are
provided in the field data sheets in Appendix A.4.
                              49

-------
Each Andersen sampler run was conducted for 5-8 minutes under iso-
kinetic conditions.  After the completion of each run, the Andersen
heads were opened and the substrates inspected and stored for weigh-
ing at MRC's laboratory.

Coal samples were taken with a shovel through the middle sampling
port and stored in 1.9 x 10~3 m3 (1/2 gallon) polypropylene
containers.

SAMPLE HANDLING

Method 5, Method 5B, and Andersen impactor filters were trans-
ferred to closed containers after sampling.  Deposits on the inside
of the sampling equipment were removed with acetone and distilled
water and bottled.  The acetone rinses were bottled separately from
the water rinses.  Access to the samples was limited in the field by
storing them in locked MRC vehicles, except when they were being
handled by authorized individuals.   The samples were shipped to
the MRC Dayton Laboratory for analysis in the vehicles used for
storing them in the field.  MRC's usual security procedures were
used to limit access to the samples while they were being analyzed.
Records of the chain-of-custody of the samples were maintained.

CHEMICAL ANALYSIS

The detailed procedure used for analyzing the sulfate and sulfuric
acid content of the Method 5 and 5B stack samples are summarized
in Appendix E, along with the results of tests of the procedure.
The procedure used is based on the procedure supplied to MRC by
EPA at a meeting October 10, 1980.

After the probe rinses and filters were dried and weighed,  room
temperature isopropanol was added to each sample; the samples
soaked for at least 12 hours; and the filter in isopropanol was
                              50

-------
ultrasonically extracted for 30 minutes.  The extracts were filter-
ed and analyzed for sulfuric acid, using the barium-thoriri titration.
This isopropanol extraction and analysis procedure was repeated twice
on every sample.  Those portions of the extracts that were not
titrated were bottled and retained.

After the sulfuric acid extraction, water was added to the filter
and solid residue; they soaked for at least 12 hours; and were
ultrasonically extracted for 30 minutes.  The extract was filtered,
passed through a Rexyn 101 ion exchange column,  and analyzed for
sulfates (using barium-thorin titration).  Four Method 5 or 5B
filters were extracted twice with water and analyzed twice for
sulfates, as an indicator of the effectiveness of the extraction
procedure.  The remaining water extracts were also retained.

The solid residues from these extractions were dried and saved under
dry nitrogen in a refrigerator, along with the untitrated extracts.
These samples were saved until May 12, 1981.

Blank filters,  residue from the evaporation of clean acetone and
residue from the evaporation of clean water (the solvents used for
probe washing)  were also analyzed when the stack sample were
analyzed, along with standards and quality assurance audit samples
of sulfate and sulfuric acid.

No chemical analysis was performed for the Andersen cascade impactor
samples (only particulate masses were determined).

COAL ANALYSIS

Grab samples of the coal being fed into the boiler were taken
before, during, and after every Method 5 and 5B stack sample.
These coal samples were split in half using ASTM D2013 riffle
sampler protocol,  and appropriate half samples were combined to
make one aggregate coal sample for every Method 5/5B sample run.
                              51

-------
The carbon, hydrogen, nitrogen, oxygen, sulfur, ash, and moisture
content of the coal for each Method 5 and 5B sampling run was
measured using ASTM D3176 ultimate analysis.  The energy content
(Btu) was measured by ASTM D2015-66(72) bomb calorimetry, and the
size distribution of each aggregate sample was measured using a
modified version of ASTM D410-38 sieve analysis.  The coal samples
were not large enough to meet all applicable ASTM sieve analysis
requirements for all size ranges; and a 2.54 cm (1 in.) screen was
used to separate large pieces of coal from the finer material,
which was sifted through #4, #16, #30, and #200 sieves.

DATA REDUCTION

MRC's computer and programmable calculators were used to reduce the
analytical and field data to determine results.  The "F" values
used to determine lb/106 Btu emissions were taken from the boiler
emission regulations in 40 CFR 60.45(f)(4)(ii).

Appendix A contains copies of all raw field data sheets (except
for opacity monitoring) and coding sheets for data processing.

Appendix B contains complete printouts of the results of the
sampling.

Appendix C contains the opacity monitoring results.

Appendix D contains boiler monitoring data taken during the testing.

Appendix E contains the detailed analytical methods  used and the
results of tests of the methods.

Appendix F contains the detailed analytical data.

Appendix G identifies the people performing the sampling,  analysis,
and data reduction.
                              52

-------
QUALITY ASSURANCE

Quality assurance and control measures included all applicable
procedures specified in the Federal Register for EPA Methods 1
through 5 and the procedures specified in the EPA Guidelines for
the Development of Quality Assurance Programs for these methods.
Particle deposits on the stages of the Andersen Sampler were
checked to assure that the stages remained tight during sampling.
A certified smoke inspector was used for opacity monitoring by
EPA Method 9.   Standard ASTM procedures were used for the coal
analysis.

MRC's Quality Assurance and Quality Control (QA/QC) Supervisor
obtained sulfate quality assurance audit filters from the Quality
Assurance Division of the U.S. Environmental Protection Agency.
Sulfuric acid audit samples were prepared by the MRC Laboratory
under the supervision of the MRC QA/QC Supervisor.

Three sulfate filters and one sulfuric acid audit sample were run
blind by the analysts prior to the receipt of the samples from
Parma to test the procedures used, to study recovery efficiencies,
and to study the precision of the sulfate analysis method at the
MRC Laboratory.

Additional audit samples as well as blank filters, clean acetone
samples, clean water samples, and calibration standards were run
along with the Parma stack samples.  All blanks and quality assur-
ance audit samples were handled in the same manner as the field
samples.

Standard ASTM procedures were used for the coal analysis.  To
obtain an indication of the precision of the coal analysis data,
half of the raw coal sample taken during the second Andersen run
at the inlet was relabeled as a sample from Parkersburg, West
Virginia and analyzed along with the coal samples for an emission
                              53

-------
test in Parkersburg.  Because it was a leftover half sample,  the
sample was not split in half prior to the second analysis.  This
was the only change from the procedure used for the actual
Parkersburg samples.  The chemists analyzing the coal samples were
not told that this sample was part of a sample analyzed earlier.
After the results of the analysis of the other half of the raw
coal sample previously analyzed were reported, the analysts were
informed of the fact that this sample was analyzed previously.
Portions of the coal sample halves that were homogenized,  pul-
verized, and prepared for the first analysis were then analyzed a
second time for sulfur content and ash content.  The purpose of
this was to help determine if the small sample size and sample
preparation procedures may have been a significant cause of the
poor reproducibility of the sulfur and ash content results
observed, or if this was caused mainly by error in the chemical
analysis.  The results (reported in Table 9 of Section 2)  sug-
gested that the small sizes of the samples and real differences
in the composition of coal from one piece to another were  the main
causes of the poor reproducibility of the results of the analysis
of different "halves" of the same raw coal samples.

The data used in computerized data processing were checked by
comparing the printout of the data used to calculate results with
the raw field data used to code the computer input.
                              54

-------