c/EPA
United States Industrial Environmental Research EPA-600/7-78-099a
Environmental Protection Laboratory June 1978
Agency Research Triangle Park NC 27711
Emission Reduction
on Two Industrial
Boilers with Major
Combustion
Modifications
Interagency
Energy/Environment
R&D Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of, control technologies for energy
systems; and integrated assessments of a wide range of energy-related environ-
mental issues.
EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for publication. Approval does not signify that the contents necessarily reflect
the views and policies of the Government, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/7-78-099a
June 1978
Emission Reduction on Two Industrial Boilers
with Major Combustion Modifications
by
W.A. Carter, HJ. Buening, and S.C. Hunter
KVB, Inc.
17332 Irvine Boulevard
Justin, California 92680
Contract No. 68-02-2144
Program Element No. EHE624A
EPA Project Officer: Robert E. Hall
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460 ]
-------
ACKNOWLEDGMENTS
The authors wish to acknowledge the assistance of Mr. Robert E. Hall,
the EPA Project Officer, whose direction and evaluations were of great
benefit.
It is our pleasure to acknowledge the generous assistance of
William Morton and Harvey Frey of the E. Keeler Company and Regis Laurendeau
of the IBM Corp., who provided the two test boilers and assisted with the
modifications and testing.
ii
-------
CONTENTS
Section
Acknowledgments ii
Figures iv
Tables vii
1.0 SUMMARY 1
1.1 Objective and Scope 1
1.2 Test Procedures 2
1.3 Results at Location 19 2
1.4 Results at Location 38 5
2.0 INSTRUMENTATION AND TEST PROCEDURES 10
2.1 Location 19 Instrumentation 10
2.1.1 Gaseous Emissions 12
2.1.2 Particulate Emissions 19
2.1.3 Trace Species and Organic Emissions 19
2.2 Location 38 Instrumentation 20
2.3 Location 19 Equipment Characteristics 22
2.4 Location 38 Equipment Characteristics 25
3.0 Test Results 27
3.1 Location 19 Test Results 27
3.1.1 Location 19 Baseline Tests 27
3.1.2 Location 19 Combustion Modifications With
#2 Oil 30
3.1.3 Combustion Modifications With #6 Oil 37
3.1.4 Combustion Modifications With Natural Gas 42
3.1.5 Particulate and SOX Testing 53
3.1.6 Trace Species and Organics Emissions (TS&O) 53
3.1.7 Boiler Efficiency 83
3.1.8 Conclusions Prom Location 19 Tests 92
3.2 Location 38 Combustion Modifications 94
3.2.1 Location 38 Baseline Tests 96
3.2.2 Combustion Modifications With #6 Oil 100
3.2.3 Particulate and SOX Testing 104
3.2.4 Combustion Modifications With Natural Gas 110
3.2.5 Boiler Thermal Efficiency 117
3.2.6 Conclusions From Location 38 Tests 117
REFERENCES 123
APPENDIX A - TRACE SPECIES AND ORGANICS SAMPLING AND
ANALYSIS PROCEDURES 125
APPENDIX B - CONVERSION FACTORS \QQ
iii
-------
FIGURES
Number Page
2-1 Instrumentation trailer floor plan and side wall
elevation. 11
2-2 Sulfur oxides analyzer sampling probe configuration. • 18
2-3 Sulfur oxides sample collection apparatus. 18
2-4 Schematic diagram of staged air and flue gas
recirculation system installed at Location 19. 23
2-5 Schematic diagram of staged air system installed
at Location 38. 26
3-1 The effect of excess oxygen on NO emissions (No. 2 oil). 33
X
3-2 NOX emissions as a function of staged air injection
depth and burner air (No. 2 oil). 35
3-3 NOX emissions as a function of percent flue gas
recirculation (No. 2 oil). 36
3-4 NOX emissions as a function of combined flue gas
recirculation and staged air at 1.2 m (No. 2 oil). 38
3-5 NOX emissions as a function of combined flue gas
recirculation rate and staged air at 2.1 m (7 ft.). 39
3-6 The effect of excess oxygen on NO emissions (No. 6 oil). 40
X
3-7 NOX emissions as a function of staged air injection
depth and burner air (No. 6 oil). 41
3-8 The effect of flue gas recirculation on NOX emissions
(No. 6 oil). 43
3-9 The effect of combined flue gas recirculation and staged
combustion air at 1.2 m on NO emissions (No. 6 oil). 44
A.
3-10 The effect of combined flue gas recirculation and staged
combustion air at 2.1 m on NO emissions (No. 6 oil). 45
X
3-11 The effect of excess oxygen on NO emissions (natural
gas; gas gun). x 47
w
-------
FIGURED (continued)
Number Page
3-12 The effect of excess oxygen on CO emissions (natural
gas; gas gun). 48
3-13 NOX emissions as a function of staged air injection
depth (natural gas; gas burner). 49
3-14 The effect of flue gas recirculation rate and low excess
O on NO emissions (natural gas). 51
& X
3-15 The effect of combined flue gas recirculation and
staged combustion air on NO emissions (natural gas). 52
X
3-16a Particulate size distribution for an oil fired steam
boiler (No. 6 oil). 55
3-16b Particulate size distribution for an oil fired steam
boiler (No. 2 oil). 56
3-17 The effect of excess oxygen on boiler thermal efficiency
for a watertube boiler. 84
3-18 The effect of excess oxygen on boiler thermal efficiency
(No. 2 oil). 85
3-19 The effect of excess oxygen on boiler thermal efficiency
(No. 6 oil). 86
3-20 The effect of secondary air insertion depth on boiler
thermal efficiency (No. 2 oil). 88
3-21 The effect of flue gas recirculation rate on boiler
thermal efficiency (No. 2 oil). 89
3-22 The effect of flue gas recirculation rate on boiler
thermal efficiency (No. 2 oil). 90
3-23 The effect of flue gas recirculation rate on boiler
thermal efficiency (natural gas). 91
3-24 Emission traverse while firing No. 6 oil. 97
3-25 Emission traverse while firing No. 6 oil. 98
3-26 The effect of excess oxygen on NO emissions (No. 6 oil). 101
3-27 The effect of windbox temperature on NO emissions
(No. 6 oil). 102
-------
FIGURES (continued)
Number Page
3-28 NO versus SCA - port location for No*. 6 oil fuel. 103
3-29 Variable SCA - NO versus 0-. 105
c 2
3-30 SCA single port variations. 106
3-31 SCA single and multiple port variations. 107
3-32 Particulate size distribution for an oil fired steam
boiler. Ill
3-33 The effect of excess oxygen on NO emissions (natural
gas). 112
3-34 The effect of windbox temperature on NO emissions
(natural gas). 113
3-25 NO versus windbox register setting (natural gas). 115
3-36 NO versus SCA - port location for natural gas fuel. 116
3-37 Variable SCA - NO versus 0,, 118
c 2
3-38 SCA single port, Tests 204-1 through 204-22 119
3-39 SCA multiple port combinations with air heater bypass
100% (^ 140 °F), Tests 204-23 through 204-32 120
-------
TABLES
Number
1-1 Summary of NO Reduction as a Function of Combustion
Modification Techniques for Various Fuels - Location 19 3
1-2 Summary of Method 5 Particulate Measurements for
Location 19 Steam Boiler 4
1-3 Summary of Change in Boiler Efficiency Due to Combustion
Modifications at Location 19 6
1-4 Summary of Total Trace Species and Organics Emissions
For The Modified Boiler at Location 19 Firing No. 6
Fuel Oil 7
1-5 POM Compounds in the XAD-2 Resin Determined by Gas
Chromatograph-Mass Spectrometry, Location 19 8
1-6 Summary of NOX Reduction as a Function of Combustion
Modification Technique for No. 6 Fuel Oil and Natural Gas
Location 38 9
1-7 Summary of Method 5 Particulate Measurements For
Location 38 Steam Boiler with No. 6 Fuel Oil 9
2-1 Trace Species and Organics to be Identified 21
3-1 Summary of Modified Boiler Tests 28
3-2 Summary of Location 19 Fuel Oil Analyses 31
3-3 Location 19 Natural Gas Analysis 32
3-4 Summary of Method 5 Particulate Measurements for
Location 19 Steam Boiler 54
3-5 Trace Species and Organics Sampling Conditions, Watertube
Boiler, Location 19 57
3-6 Summary of Emissions Data at Location 19 During Trace
Species and Organics Tests (TS&O) 59
3-7 Summary of Location 19 Test Fuel Oil Analyses for Trace
Species and Organics Tests 60
vii
-------
TABLES (continued)
Number
«
3-8 General Notes for Trace Species and Organics Data
Tabulations 62
3-9 Trace Species and Organics Emissions, SASS Solids
Section Collection, Test 19-2, Modified Boiler,
Location 19, Baseline Condition 63
3-10 Trace Species and Organic Emissions, SASS Organic
and Liquids Section Collection, Test 19-2, Modified
Boiler, Location 19, Baseline Condition 64
3-11 Trace Species and Organic Emissions, Process Samples
and Mass Balances, Test 19-2, Modified Boiler, Location 19
Baseline Condition 65
3-12 Trace Species Emissions by Spark Source Mass Spectrometry,
Test 19-2, Modified Boiler, Location 19, Baseline Condition 66
3-13 Trace Species and Organic Emissions, SASS Solids Section
Collection, Test 19-3, Modified Boiler, Location 19, Low
NO Condition 69
x
3-14 Trace Species and Organic Emissions, SASS Organic and
Liquids Section Collection, Test 19-3, Modified Boiler,
Location 19, Low NO Condition 70
X
3-15 Trace Species and Organic Emissions, Process Samples and
Mass Balances, Test 19-3, Modified Boiler, Location 19,
Low NO Condition 71
x
3-16 Trace Species Emissions by Spark Source Mass Spectrometry,
Test 19-3, Modified Boiler, Location 19, Low NO Condition 72
X
3-17 Trace Species and Organic Emissions, SASS Solids Section
Collection, Test 19-4, Modified Boiler, Location 19, Low
NO Condition 75
x J
3-18 Trace Species and Organic Emissions, SASS Organic and
Liquids Section Collection, Test 19-4, Modified Boiler,
Location 19, Low NO Condition 76
3-19 Trace Species and Organic Emissions, Process Samples and
Mass Balances, Test 19-4, Modified Boiler, Location 19,
Low NO Condition 77
X
viii
-------
TABLES (continued)
Number
3-20 Summary of Total Trace Species and Organics Emissions
for the Modified Boiler at Location 19 Firing No. 6
Fuel Oil 79
3-21 POM Compounds in the XAD-2 Resin Determined by Gas
Chromatograph-Mass Spectrometry, Location 19 82
3-22 Summary of Changes in Boiler Efficiency Due to
Combustion Modifications 93
3-23 Summary of Location 38 Combustion Modification Test Data 95
3-24 Summary of Location 38 Fuel Oil Analyses 99
3-25 Location 38 Natural Gas Analysis 99
3-26 Summary of Method 5 Particulate Measurements for
Location 38 Steam Boiler Firing No. 6 Oil 108
3-27 SO Summary, Location 38 Firing No. 6 Oil 109
X
ix
-------
SECTION 1.0
SUMMARY
1.1 OBJECTIVE AND SCOPE
The objective of this program was to evaluate the effectiveness of
combustion modifications and operating variable changes as means of improve-
ment in thermal efficiency and for emissions control in industrial size water-
tube boilers. These techniques have previously been shown to be effective
on industrial boilers (Refs. 1, 2) and the purpose of this program was to
evaluate feasibility of implementing each candidate combustion modification
independently and in various combinations.
The program scope provided for tests on two watertube boilers to
evaluate low excess air, variable combustion air preheat, staged combustion
air, and flue gas recirculation while firing natural gas, #2 and #6 oil at
Location 19 and natural gas and #6 oil at Location 38. Emissions to be
measured were NO, NOX, SO , SO , CO, CO , O , gaseous hydrocarbons, particu-
^ J £1 £
lates, particulate size distribution, smoke number and opacity.
On four tests at Location 19,samples were collected for analysis of
trace species and selected organic emissions. Two of these tests were at
baseline operating conditions and two were at low NO conditions.
This is a final report on this test program documenting the test
equipment, a summary of the test data and a discussion of the data in relation
to each type of combustion modification.
-------
1.2 TEST PROCEDURES
Parametric tests were conducted to evaluate the effectiveness of
combustion modifications on emissions reduction and boiler thermal efficiency.
For natural gas, #2 oil and #6 oil, the effect of lowered excess air, staged
combustion air, variable air preheat, flue gas recirculation and combinations
of these, was evaluated.
A government-owned mobile laboratory was used for measuring air
pollutant emissions and unit efficiency parameters at Location 19. The
mobile laboratory contains analytical instrumentation for continuous measure-
ment of NO, NOX, SO , CO, CO , O , and gaseous hydrocarbons. Sulfur oxides
(SO and SO ), total particulate, and particulate size distribution were
measured and analyzed by grab sample techniques. Four tests were conducted
to sample and analyze for trace species and selected organics using the source
assessment sampling system (SASS train).
Gaseous emissions were measured at Location 38 using a KVB mobile
laboratory containing analytical instrumentation for continuous measurement
of NO, NOX, CO, O , CO , and hydrocarbons. Sulfur oxides (SO and SO ) ,
^ ^ & J
total particulate and particulate size distribution were measured using grab
sample techniques.
1.3 RESULTS AT LOCATION 19
Results of combustion modification tests at Location 19 with natural
gas, #2 oil and #6 oil are summarized in Table 1-1 where NOX reduction as a
function of combustion modification technique is tabulated. The greatest
reduction with both #2 and #6 oil was obtained with a combination of all
three techniques - low 0„, flue gas recirculation and staged combustion air.
With natural gas, the greatest NOV reduction was with flue gas recirculation
x
and low O .
Results of particulate measurements obtained using EPA Method 5 are
shown in Table 1-2. The minimum total particulate measurements were obtained
at the low NO condition for each fuel. The same was trua for solid particulate
with the exception of Ilo. 2 oil, for \rhicli minimum particulate was obtained
with FGR and low 0 .
-------
TABLE 1-1. SUMMARY OF NOx REDUCTION AS A FUNCTION OF COMBUSTION
MODIFICATION TECHNIQUES FOR VARIOUS FUELS - LOCATION 19
Fuel Type
Average Baseline NOX, ppm at 3% O?
Combustion Modification Technique
Lowered Excess Air
Staged Combustion Air (SCA) ,
Normal O_
SCA, Low 0
Flue Gas Recirculation (FGR) ,
Normal O«
FGR, Low O?
FGR + SCA, Normal O2
FGR + SCA, Low O
No. 2 Oil No. 6 Oil
(0.008% N) (0.20% N) Natural Gas
114 220 96*.
NO Reduction,
X
92
Percent of Baseline
20 30 19*
30 29 32*
46~
44 42 42*
68 11 77"f ,
f" i
73 40 |79^
69 53 76^
I77| 5
5 %*
*Ring burner
tGas gun burner
iStability limits prevented lowering
Indicates lowest NO mode
-------
TABLE 1-2. SUMMARY OF METHOD 5 PARTICULATE MEASUREMENTS
FOR LOCATION 19 STEAM BOILER
Boiler
Operating
Mode
Baseline
Low O
SCA
FGR, Low O2
FGR + SCA
No. 2 Oil
Total Solid
ng/J ng/J
(Ib/MMBtu) (Ib/MMBtu)
24.24 2.595
(0.0564) (0.0060)
16.29 5.95
(0.0379) (0.0138)
11.6 9.01
(0.0270) (0.0210)
5.84 1.95
(0.0136) (0.0045)
4.16 3.31
(0.0097) (0.0077)1
No. 6 Oil
Total Solid
ng/J ng/J
(Ib/MMBtu) (Ib/MMBtu)
36.21 27.55
(0.084) (0.064)
28.87 25.80
(0.0672) (0.060)
31.8 27.2
(0.0743) (0.0635)
32.33 29.36
(0.075) (0.068)
28.80 9.10
(0.0670) (0.021)
Natural Gas
Total Solid
ng/J ng/J
(Ib/MMBtu) (Ib/MMBtu)
3.68 1.92
(0.0086) (0.0045)
2.63 1.67
(0.0061) (0.0039)
indicates lowest NO mode
-------
Boiler thermal efficiency was calculated using the ASME heat loss
method. The results of these measurements are shown in Table 1-3. The
change in boiler efficiency at the low NO conditions is shown for each
fuel.
Four trace species and organics tests were conducted at Location 19
with the boiler firing No. 6 oil. The first two tests were with the boiler in
*
the normal or baseline conditions and the next two tests were with the boiler
in the low NO operating mode. Samples for the first baseline test were not
X.
analyzed because of a major change in fuel supply. The results of these
tests are summarized in Table 1-4. Additional data on the specific POM
compounds present in the SASS Train XAD-2 adsorbant are shown in Table 1-5.
1.4 RESULTS AT LOCATION 38
A summary of NO reduction as a function of modification technique
X
is presented in Table 1-6 for the tests conducted at Location 38. NO reduc-
X
tions of 43% and 32% with No. 6 oil and natural gas, respectively, were
achieved using a combination of staged combustion air and lowered excess
oxygen. Combined staged combustion and lowered air preheat produced 69%
reduction of NO , compared to baseline NO emissions.
X X
Results of the particulate measurements obtained using EPA Method 5
are presented in Table 1-7 for No. 6 fuel oil. For all modifications the
total particulate emissions were reduced by up to 34% compared to baseline
emissions. Solid particulates were increased slightly (up to 8%) with all
modifications, compared to baseline emissions.
-------
TABLE 1-3. SUMMARY OF CHANGE IN BOILER EFFICIENCY
DUE TO COMBUSTION MODIFICATIONS AT LOCATION 19
Boiler
Operating
Mode
Low O
SCA, Normal O
SCA, Low 0
FGR, Normal 0
FGR, Low O
FGR + SCA, Normal O
FGR + SCA, Low 02
No. 2 Oil
•f 1.5%
+ 0.9%
+ 1.1%
- 1.9%
+ 0.9%
- 1.2%
| - 0.8% |
No. 6 Oil Natural Gas
" 1'2%*
+ -1'5 + 0.9%+
+ 0.1%*
+ °-1% + 0.3%+
+ 0.8% + 0.5%*
- 0.7% - 0.8%+
4- 0.6% | - 0.6%f j
- 0.8% - 0.5%+
| + 0.1% | §
*Ring burner
Gas gun burner
§
Stability limits prevented lowering O,
A
Indicates lowest NO mode
-------
TABLE 1-4. SUMMARY OF TOTAL TRACE SPECIES AND ORGANICS EMISSIONS FOR THE MODIFIED BOILER
AT LOCATION 19 FIRING #6 FUEL OIL
Total Emission Concentre
Atomic Absorption,
Test
Condition
Antimony
Arsenic
Barium
Beryllium
Cadmium
Calcium
Chromium
Cobalt
Copper
Iron
Lead
Manganese
Mercury
Nickel
Selenium
Tellurium
Tin
Titanium
Vanadium
Zinc
Chloride
Fluoride
Nitrates
Sulfates
Total POM
Total PCB
19-2
Baseline
< 380
6.5 < 15
95 < 210
< 6
13
650
750
65 < 130
32
4300
45 < 70
70
< 1.9
1300 <1400
< 12
< 300
< 750
70 < 1600
3200 <3400
370
12000
170 < 180
130
18000
NES
NES
19-3
Low HO*
< 540
59 < 64
640 < 740
< a. 9
4.8 < 12
2000
740
79 < 150
39 < 44
4700
9.9 < 21
99
0.06 < 21
1600
9.9 < 290
< 450
< 1000
120 < 2500
3400 <3600
810
3500
64 < 79
110 < 120
18000
50 < 51
< 7
Jtions by
ag/ia3
19-4
Low 110*
< 350
55
800 < 850
< 6
1.1 < 6
440 < 460
530
18 < 85
95
3100
< 15
65
2
2200
< 11
< 290
< 700
20 < 100
2400
3300
6000
24 < 33
85
21000
NES
NES
Total Emission Concen-
trations by Spark Source
Mass Spectroinetry, pg/
19-2
Baseline
11
6.5
MC
0.055 •: 3
7.5 < 13
2000 < MC
960
8 < MC
49
1300 1.05
0.80
23.00
0.31
0.30
1.35
0.83
0.61
1.20
—
—
—
0.92
1.00
1.30
—
0.52
—
19-3
Low NOv
—
1.7 '
1.90
—
—
0.55
2.90
0.54
1.00
0.99
—
1.66
0.83
1.40
—
—
1.40
0.80
1.60
17.00
0.70
—
19-4
Low NOx
< DL
1.60
2.30
< DL
> 0.05
0.12
3.00
0.10
0.74
2.50
< DL
1.62
1.10
< 0.15
< DL
< DL
> 0.22
0.18
15.00
46.00
0.09
—
See notes on Table 3-8, page 62
-------
TABLE 1-5. POM COMPOUNDS IN THE XAD-2 RESIN DETERMINED BY
GAS CHROMATOGRAPH-MASS SPECTROMETRY, LOCATION 19
POM Component
Anthracene
Phenanthrene
*Methyl Anthracenes
Fluoranthene
Pyrene
*Benzo (c)phenanthren
Chrysene
Benzo Fluoranthenes
*Benz (a)pyrene
Benz (e)pyrene
Total POM
Tegt 19-2,
ng/g
3.2
—
0.2
1.2
0.05
0.002
0.03
0.007
0.004
0.004
4.74
Baseline
ng/m-3
24
—
1.6
9.0
0.4
0.02
0.19
0.05
0.032
0.032
35.5
Test 19-3,
ng/g
0.45
0.02
0.12.
0.13
0.05
—
0.004
0.007
—
—
0.78
Low NO
Jv
ng/m^
3.4
0.1
0.9
0.9
0.4
—
0.03
0.05
—
—
5.8
*Compounds required to be identified
Note: Values in this table are expressed in nanograms (ng), (1 ng = 10 g)
Values in other trace species and organics tables in this report are
expressed in micrograms (yg), (1 pig = 10~6 g).
-------
TABLE 1-6. SUMMARY OF NOx REDUCTION AS A FUNCTION
OF COMBUSTION MODIFICATION TECHNIQUE
FOR NUMBER 6 FUEL OIL AND NATURAL GAS
Location 38
Modified Condition
Lowered excess air
Staged combustion air
Staged combustion air
Variable preheat (max.
Variable preheat (min.
Staged combustion air
No. 6 Oil Natural Gas
20% 14%
(normal 0 ) 36% 23%
(low 00) 431 32%
temp.) -4% -24%
temp.) 18% 24%
& variable preheat — 69%
Indicates lowest NO mode
v
TABLE 1-7. SUMMARY OF METHOD 5 PARTICULATE MEASUREMENTS
FOR LOCATION 38 STEAM BOILER WITH #6 FUEL* OIL
Condition
Total Particulate
ng/J
(Ib/MMBtu)
Solid Particulate
ng/J
(Ib/MMBtu)
Baseline
Low Excess Air
Staged Combustion Air
Variable Air Preheat
(Minimum Temperature)
66.4
(0.154)
47.6
(0.110)
43.7
(0.101)
52.6
(0.122)
62.2
(0.144)
54.4
(0.126)
36.6
(0.085)
38.3
(0.088)
38.7
(0.089)
38.7
(0.090)
39.9
(0.092)
37.6
(0.087)
-------
SECTION 2.0
INSTRUMENTATION AND TEST PROCEDURES
2.1
LOCATION 19 INSTRUMENTATION
The emissions measurements were made using analytical instruments
and equipment contained in a government-furnished mobile instrumentation
laboratory contained in an 2.4 x 9.1 meter (8 x 30 ft) trailer. A plan
and side view of the trailer are shown in Figure 2-1. Gaseous emission
measurements are made with the following analytical instruments:
Emission
Species
Manufacturer
Measurement
Method
Model
No.
Hydrocarbon
Carbon Monoxide
Oxygen
Carbon Dioxide
Nitrogen Oxides
Sulfur Dioxide
Beckman Instruments
Beckman Instruments
Teledyne
Beckman Instruments
Thermo Electron
Dupont Instruments
Flame lonization
IR Spectrometer
Polarographic
IR Spectrometer
Chemiluminescent
UV Photometric
402
865
326A
864
10A
400
Total oxides of sulfur were measured by wet chemistry methods
using the sampling train and analytical procedure of the Shell-Emeryville
method. Total particulate measurements were made using an EPA Method 5
sampling train manufactured by Western Precipitation Div. of Joy Manufac-
turing Co. Particulate size distribution was measured using an Andersen 2000
cascade impactor. Smoke density was measured using an automated Bacharach
smoke spot pump. Samples for trace species and organics analysis were taken
with the source assessment sampling system (SASS train), a high volume
sampling train.
10
-------
Sulfur Oxides Bench
\
Rear
F
ume
Hood
Sample Line Connection
Distribution
Panel
8
PWM
runnn
nnw
•
Left Side Wall
Left Side Participate Bench
Front View of Instrument Panel
Oven
a
a
a
a
a
QQQCT
Transformer
Calibration Gases
Right side f
'Washing and Drying Bench
1
•
::::.
—
— 1
in M i it
•in nil
ill! Ill)
A JL
-h
«
Gas
B
C
^
yii
nde
-y
rs
•p
Rear Wall
Eight Side Wall
Figure 2-1. Instrumentation trailer floor plan and side wall elevation.
-------
2.1.1 Gaseous Emissions
The laboratory is equipped with analytical instruments to continuously
measure concentrations of NO, NO , CO, CO , 0 , SO , and hydrocarbons. The
X £, £ £t
sample gas is delivered to the analyzers at the proper condition and flow
rate through the sampling and conditioning system described in the previous
sections. This section describes the analytical instrumentation.
Total Nitrogen Oxides
The oxides of nitrogen monitoring instrument used is a Thermo
Electron chemiluminescent nitric oxide analyzer. The operational basis
of the instrument is the chemiluminescent reaction of NO and O to form
NO in an excited state. Light emission results when excited NO mole-
cules revert to their ground state. The resulting chemiluminescence is
monitored through an optical filter by a high sensitivity photomultiplier
tube, the output of which is electronically processed so it is linearly
proportional to the NO concentration.
Air for the ozonator is drawn from ambient through an air dryer
and a 10-micron filter element. Flow control for the instrument is
accomplished by means of a small bellows pump mounted on the vent of
the instrument downstream of a separator which insures that no water
collects in the pump.
The basic analyzer is sensitive only to NO molecules. To mea-
sure NOx (i.e., NO + NO2), the NO2 is first converted to NO. This is
accomplished by a converter which is included with the analyzer. The
conversion occurs as the gas passes through a thermally insulated,
resistance heated, stainless steel coil. With the application of heat,
NO2 molecules in the sample gas are reduced to NO molecules, and the
analyzer then reads NOX- NO2 is obtained by the difference in readings
obtained with and without the converter in operation.
12
-------
Specifications
Accuracy: 1% of full scale
Span stability: ;+ 1% of full scale in 24 hours
Zero stability: +_ 1 ppm in 24 hours
Power requirements: 115 +_ 10V, 60 Hz, 1000 watts
Response: 90% of F.S. in 1 sec (NO mode); 0.7 sec (NO mode)
X
Output: 4-20 ma
Sensitivity: 0.5 ppm
Linearity: -^ 1% of full scale
Vacuum detector operation
Range: 2.5, 10, 25, 100, 250, 1000, 2500, 10,000 ppm F.S.
Both the total nitrogen oxides (NO ) and nitric oxide (NO)
X
concentrations are measured directly using a sample line heated to
about 394 K (250 °F) to conduct the gas sample to the analyzer in the
trailer. In addition, the nitric oxide concentration is measured
sequentially using an unheated sample line connected to the same analyzer
in the trailer. Here, the water is first removed from the sample gas by
a drop-out bottle and a refrigerator.
Carbon Monoxide and Carbon Dioxide
Carbon monoxide and carbon dioxide concentrations are measured by
Beckman Model 864 and 865 short pathlength nondispersive infrared analyzers.
These instruments measure the differential in infrared energy absorbed from
energy beams passed through a reference cell (containing a gas selected
to have minimal absorption of infrared energy in the wavelength absorbed
by the gas component of interest) and a sample cell through which the
sample gas flows continuously. The differential absorption appears as a
reading on a scale of 0 to 100% and is then related to the concentration
of the specie of interest by calibration curves supplied with the instru-
ment. A linearizer is supplied with each analyzer to provide a linear out-
put over the range of interest. The operating ranges for the CO analyzer
are 0-100 and 0-2000 ppm, while the ranges for the CO analyzer are 0-5%
and 0-20%.
13
-------
Specifications
Span stability: +_ 1% of full scale in 24 hours
Zero stability: +_ 1 ppm in 24 hours
Ambient temperature range: 273 to 322 K (32 °F to 120 °F)
Line voltage: 115 +_ 15V rms
Response: 90% of F.S. in 0.5 or 2.5 sec
Linearity: Linearizer board installed for one range
Precision: +_ 1% of full scale
Output: 4-20 ma
Oxygen
A Teledyne Model 326A oxygen analyzer is used to automatically
and continuously determine the oxygen content of the flue gas sample.
Oxygen in the flue gas diffuses through a Teflon membrane and is reduced
on the surface of the cathode. A corresponding oxidation occurs at the
anode internally and an electric current is produced that is proportional
to the concentration of oxygen. This current is measured and conditioned
by the instrument's electronic circuitry to give a final output in per-
cent O by volume for operating ranges of 0% to 5%, 0% to 10%, and 0%
to 25%.
Specifications
Precision: +_ 1% of full scale
Response: 90% in less than 40 sec
Sensitivity: 1% of low range
Linearity: +_ 1% of full scale
Ambient temperature range: 273 K to 325 K (32 to 125 °F)
Fuel cell life expectancy: 40,000+ hrs
Power requirement: 115 VAC, 50-60 Hz, 100 watts
Output: 4-20 ma
14
-------
Total Hydrocarbons
Hydrocarbons are measured using a Beckman Model 402 hydrocarbon
analyzer which utilizes the flame ionization method of detection. The
sample is filtered and supplied to the burner by means of a pump and flow
control system. The sensor, which is the burner, has its flame sustained
by regulated flows of hydrogen fuel and air. In the flame, the hydrocarbon
components of the sample undergo a complete ionization that produces elec-
trons and positive ions. Polarized electrodes collect these ions, causing
a small current to flow through an electronic measuring circuit. This
ionization current is proportional to the concentration of hydrocarbon
atoms which enter the burner. The instrument is available with range
selection from 6 ppm to 10% full scale as CH .
Specifications
Full scale sensitivity: adjustable from 5 ppm CH to 10% CH
Ranges: Range multiplier switch has 8 positions: Xl, X5, XlO,
X50, X100, X500, X1000, and X5000. In addition, span
control provides continuously variable adjustment
within a dynamic range of 10:1
Response time: 90% full scale in 0.5 sec
Precision: +_ 1% of full scale
Electronic stability: +_ 1% of full scale per 24 hours with
ambient temperature change of less thar;
10 °F
Reproducibility: +_ 1% of full scale for successive identical
samples
Analysis temperature: ambient
Ambient temperature: 273 K to 317 K (32 °F to 110 °F)
Output: 4-20 ma
Air requirements: 250 to 400 cc/min of clean, hydrocarbon-free
air, supplied at 2.07 x 105 to 1.38 x 10
n/m2 (30 to 200 psig)
Fuel gas requirements: 75 to 80 cc/min of fuel consisting of
100% hydrogen supplied at 2.07 x 105
to 1.38 x 106 n/m2 (30 to 200 psig)
Electric power requirements: 120 V, 60 Hz
Automatic flame indication and fuel shut-off valve
15
-------
Sulfur Dioxide
A Dupont Model 400 photometric analyzer is used for measuring SO2-
This analyzer measures the difference in absorption of two distinct wave-
lengths (ultraviolet) by the sample. The radiation from a selected light
source passes through the sample and then into the photometer unit where
the radiation is split by a semi-transparent mirror into two beams. One
beam is directed to a phototube through a filter which removes all wave-
lengths except the "measuring" wavelength, which is strongly absorbed by
the constituent in the sample. A second beam falls on a reference photo-
tube, after passing through an optical filter which transmits only the
"reference" wavelength. The latter is absorbed only weakly, or not
at all, by the constituent in the sample cell. The phototubes translate
these intensities to proportional electric currents in the amplifier.
In the amplifier, full correction is made for the logarithmic relation-
ships between the ratio of the intensities and concentration or thickness
(in accordance with Beer's Law). The output is therefore linearly pro-
portional, at all times, to the concentration and thickness of the sample.
The instrument has full scale ranges of 0-500 and 0-5000 ppm.
Specifications;
Noise: Less than 1/4%
Drift: Less than 1% full scale in 24 hours
Accuracy: (+_ 1% of analyzer reading)+(+_ 1/4% of full scale range)
Sample cell: 304 stainless steel, quartz windows
Flow rate: 6 CFH
Light source: Either mercury vapor, tungsten, or "Osram"
discharge type lamps
Power rating: 500 watts maximum, 115 V, 60 Hz
Reproducibility: 1/4% of scale
Electronic response: 90% in 1 sec
Sample temperature: 378 K (220 °F)
Output: 4-20 ma d.c.
16
-------
Sulfur Oxides
Measurement of SC>3 concentrations is done by wet chemical analy-
sis using the "Shell-Emeryville" method. In this technique the gas
sample is drawn from the stack through a glass probe (Figure 2-2), con-
taining a quartz wool filter to remove particulate matter, into a system
of three sintered glass plate absorbers (Figure 2-3). The first two
absorbers contain aqueous isopropyl alcohol and remove the sulfur tri-
oxide; the third contains aqueous hydrogen peroxide solution which
absorbs the sulfur dioxide. Some of the sulfur trioxide is removed
by the first absorber, while the remainder, which passes through as a
sulfuric acid mist, is completely removed by the secondary absorber
mounted above the first. After the gas sample has passed through the
absorbers, the gas train is purged with nitrogen to transfer sulfur
dioxide, which has dissolved in the first two absorbers, to the third
absorber to complete the separation of the two components. The iso-
propyl alcohol is used to inhibit the oxidation of sulfur dioxide to
sulfur trioxide before it gets to the third absorber.
The isopropyl alcohol absorber solutions are combined and the
sulfate, resulting from the sulfur trioxide absorption, is titrated
with standard lead perchlorate solution, using Sulfonazo III indicator.
In a similar manner, the hydrogen peroxide solution is titrated for
the sulfate resulting from the sulfur dioxide absorption.
The gas sample is drawn from the flue by a single probe made
of 5 mm ID Vycor glass inserted into the duct approximately one-third
to one-half way. The inlet end of the probe has a section 50 mm long
by 15 mm OD which holds a quartz wool filter to remove particulate
matter. It is important that the entire probe temperature be kept
above the dew point of sulfuric acid during sampling (minimum tempera-
ture of 533 K) . This is accomplished by wrapping the probe with a
heating tape.
17
-------
Flue Wall
End of Opening
15 mm ID
'^zZc, (-V-^l^N-1 •—-
=^^Jb^:Eb==\ •••!••••.:-. ••
Pyrometer
and
Thermocouple
Figure 2-2. Sulfur oxides analyzer sampling probe configuration.
Sintered
Glass
Absorbers
Spray Trap
Dial Thermometer
Pressure Gauge
Volume Indicator
Vopor Trap
Diaphragm Dr* Tost Mefcr
Figure 2-3. Sulfur oxides sample collection apparatus.
18
-------
2.1.2 Particulate Emissions
Particulate samples are taken at the same sample port as the gas
sample using a Joy Manufacturing Company portable effluent sampler. This
system, which meets the EPA design specifications for Test Method 5,
\
(Determination of Particulate Emissions from Stationary Sources, Federal
Register, Volume 36, No. 27, page 124888, December 23, 1971) is used to per-
form both the initial velocity traverse and the particulate sample collection.
Dry particulates are collected in a heated case that contains, first, a cyclone
to separate particles larger than 5 microns and, second, a 100 mm glass-fiber
filter for retention of particles down to 0.3 microns. Condensible particu-
lates are collected in a train of 4 Greenburg-Smith impingers in a chilled
water bath.
2.1.3 Trace Species and Organic Emissions^
Particulate and gaseous samples for analysis of trace species and
organics emissions at Location 19 were taken at the same stack port as the
gas and standard particulate samples. The sampling system is based on a
modified high volume sampling system developed by EPA and is called the "Source
Assessment Sampling System" (SASS train). Dry particulates are collected in a
heated case that contains three cyclones to separate particles larger than
10 urn, 3-10 urn, and 1-3 urn. Particles less than 1 urn are collected on a
142 mm glass-fiber filter. Filtered sample gas is then cooled to 293 K to
333 K (68 to 130 °F) and passed through an organic adsorbent consisting of
XAD-2 chromosorb type adsorbent. Condensate is collected in a trap and the
dried gas passes through an impinger train in a chilled water bath. The
first impinger contains a hydrogen peroxide solution for SC>2 scrubbing and
the second and third impingers contain reagents for volatile trace species
collection. The fourth impinger contains Drierite for final drying prior to
flow control and volume measurement equipment.
19
-------
The SASS samples at a rate of 0.0019 m /s (4 ft /m) . The specific
train used in this program was designed to sample at 0.0019 m /s based on
cyclone inlet conditions at 478 K (400 °F). For later trains the design
was shifted to a sampling rate of 0.0019 m /s based on standard dry conditions.
The rate of sampling has an influence on cyclone size cuts.
During sampling with the SASS, fuel samples were obtained for analysis
so that emissions measured in the stack could be compared with species
entering the boiler in the fuel. All sampling collections on the Location
19 boiler were while firing #6 fuel oil.
Operation of the SASS, sample recovery and handling procedures were
performed in accordance with an EPA document defining sample collection
procedures (Ref. 3 ).
Samples from the SASS train and fuel sampJes were analyzed by atomic
absorption, spark source mass spectrometry and gas chromatography to establish
the emission rates and mass balances for the species listed in Table 2-1.
2.2 LOCATION 38 INSTRUMENTATION
The emissions measurements were made using analytical instruments
and equipment contained in a KVB owned mobile laboratory. Gaseous emission
measurements were made with the following analytical instruments:
Emission
Species
Hydrocarbon
Carbon Monoxide
Oxygen
Carbon Dioxide
Nitrogen Oxides
Manufacturer
Beckman Instruments
Beckman Instruments
Teledyne
Beckman Instruments
Thermo Electron
Measurement
Method
Flame lonization
IR Spectrometer
Polarographic
IR Spectrometer
Chemi lumine s cen t
Model
No.
402
865
326A
864
10A
These instruments were described in detail in the previous paragraphs.
20
-------
TABLE 2-1. TRACE SPECIES AND ORGANICS TO BE IDENTIFIED
Elements
Antimony Cobalt Selenium
Arsenic Copper Tellurium
Barium Fluorine Tin
Beryllium Iron Titanium
Cadmium Lead Vanadium
Calcium Manganese Zinc
Chlorine Mercury
Chromium Nickel
Species
Total sulfates
Total nitrates
Organics
Total polychlorinated biphenyls (PCB)
Total polycyclic organic matter (POM)
Specific POM compounds:
7, 12 - dimethyIbenz (a) anthracene
Dibenz (a,h) anthracene
Benzo (c) phenanthrene
3-Methylcholanthrene
Benzo (a) pyrene
Dibenzo (a,h) pyrene
Dibenzo (a,i) pyrene
Dibenzo (c,g) carbazole
21
-------
Total oxides of sulfur were measured by wet chemistry methods using
the sampling train and analytical procedure of the Shell-Emeryville method.
Sulfur oxides were measured only during the #6 oil tests. Total particulate
measurements were made using an EPA Method 5 sampling train. Particulate
size distribution was made using a cascade impactor manufactured by Meteorology
Research, Inc.
2.3 LOCATION 19 EQUIPMENT CHARACTERISTICS
The test unit used to evaluate the combustion modifications at
Location 19 was a Type DS two drum watertube package steam generator, rated at
a heat output of 5.1 MW (17500 Ib/h of steam flow), designed for pressurized
furnace operation. The unit has been modified to incorporate staged combus-
tion air and flue gas recirculation. A schematic of the boiler is shown in
Figure 2-4.
Staged combustion air was introduced into the furnace through four
3 inch pipe size diameter lances. Insertion depth of the lances was variable
up to 2.1 m (7 ft). Staged combustion air was supplied by a separate blower
and flow rate was adjusted by dampers in each supply line.
A new burner windbox was designed and fabricated for incorporating
flue gas recirculation into the system. Flue gas was extracted from the
base of the stack by means of a blower and ducted to the burner windbox via
an insulated 14" duct. The flue gas was injected into the burner from a
plenum at the rear of the windbox.
The flue gas is recirculated to the burner by withdrawing the gas
from the base of the stack (T ^ 545 K) with a high temperature fan. The
CJciS
high temperature gas is ducted to the burner windbox through an insulated
duct containing a calibrated flat plate orifice for measuring flow. The
burner windbox has been modified to inject the flue gas radially inward
through an adjustable slot at the rear of the burner.
Staged combustion is accomplished by injecting air into the combus-
tion zone through four probes oriented 90 deg apart around the burner. These
staged air probes are 76.2 mm (3 in.) diameter and are adjustable in length
up to 2.1 m (7 ft). The ends of the probes are blocked and slots machined
in the pipe to inject the air radially inward toward the flame centerline.
22
-------
Staged Air Lances
Flue Gas
Recirculation
Duct
Damper
Furnace
Fan
Stack
(a) Top View
Stack
Staged Air Lances
flue Gas Recirculation Duct
(b) Side View
Figure 2-4.
Schematic diagram of staged air and flue gas recirculation
system installed at Location 19.
23
-------
The capacity and design conditions of the boiler are presented in
the following table.
Location 19, Steam Boiler Specifications
Pounds of steam per hour (kg/s) 17,500 (2.206)
Operating pressure, psig (kPa) 160 (1207)
Design pressure, psig (kPa) 250 (1827)
Steam temperature saturated
Feedwater temperature, °F (K) 220 (377)
Total heating surface, sq ft (m ) 1,881 (175)
Radiant heating surface, sq ft (m ) 288 (27)
Convection heating surface, sq ft (m ) 1,593 (148)
Furnace volume, cu ft (m3) 430 (12)
Fuel Type natural gas, #2 oil, #6 oil
Furnace length ft., (m) 9' - 10-13/16" (3.018)
Furnace width ft., (m) 5' - 10" (1.778)
Special Instrumentation Requirements
The Location 19 test boiler was modified t.o incorporate staged
combustion air and flue gas recirculation. As a consequence of these
modifications, additional instrumentation was required to measure the staged
combustion air flow and the recirculated flue gas flow.
Staged combustion air was introduced into the boiler through four
lances on the face of the boiler. To measure the staged air flow, a set
of curves was generated which give secondary air flow as a function of lance
injection pressure. The lance injection pressure was then measured for all
staged air tests.
Recirculated flue gas was measured by installing a calibrated sharp-
edged orifice in the duct which returned the flue gas to the combustion air
plenum. The pressure drop across the orifice was measured with an inclined
manometer as was the static pressure in the duct. Temperature of the flue
gas was measured at the orifice inlet with a high temperature thermometer.
All pertinent data were recorded for each flue gas recirculation test.
24
-------
2.4 LOCATION 38 EQUIPMENT CHARACTERISTICS
The test boiler at Location 38 was a vertical watertube type rated
at 5.67 kg/s (45,000 Ib/hr) of saturated steam flow. The unit was modified
to incorporate sidefire air capability. It had a single burner that fired
either natural gas or #6 fuel oil. The boiler was also equipped with an air
preheater which can raise the combustion air temperature to a maximum of
450 K (350 °F).
The sidefire installation is shown schematically in Figure 2-5. A
36 cm diameter manifold was run along each side of the boiler and was con-
nected to a fan mounted on the floor at the rear of the boiler. Flexible
pipes connected the manifold to the overfire air ports in the furnace side
walls.
The amount of sidefire air going to each downcomer was controlled
by butterfly valves installed in each of the two legs of the manifold and
in the upper section of each downcomer.
Staged combustion air tests were conducted on this unit with both
natural gas and #6 oil firing. During the staged air tests the amount and
location of the injection of the sidefire air addition was systematically
varied while the total amount of combustion air was held constant.
Location 38, Steam Boiler Specifications
Pounds of steam per hour (kg/s) 45,000 (5.67)
Operating pressure, psig (kPa) 140 (1070)
Design pressure, psig (kPa) 250 (1830)
Steam temperature, °F (K) 360 (456)
Feedwater temperature, °F (K) 275 (408)
Furnace volume, cu. ft. (m3) 1,537 (43.5)
Fuel type natural gas, Ho. 6 oil
Furnace length, ft. (m) 10'-6" (3.2)
Furnace width, ft. (m) 12' (3.7)
Maximum air temperature, °F (K) 350 (450)
25
-------
PORT NOS. VgJ7~|^|9^
t
183cm
I
/'
WINDBOX
P
r
FURNACE
o/ici
:. — 166cm -»-
--86— 1
cm 1
ORT 1^16^1 8
JOS. \& &
r-t^
^jll l@| 13,15
II
II
___^J H
||
J II
II
II
II
^JIO ^|I2,I4
*
/ -V
/
f [
DIAMETER
MANIFOLD (a)- TOP VIEW
FAN
320cm
FURNACE
14,15
86cm
80cm I 83cmi6lcm
e—-G—e—e-
89
cm
PORT 6,7 8,9 10,11 12,13
NOS.
WINDBOX
366
cm
X
DIVIDING
WALL
(b) SIDE VIEW
Figure 2-5. Schematic diagram of staged air system installed at Location
38.
26
-------
SECTION 3.0
TEST RESULTS
This section summarizes the emission and efficiency data collected
on the Location 19 and 38 watertube boilers. The Location 19 boiler was
modified to incorporate staged combustion air, flue gas recirculation and
combinations of these. At Location 19 the boiler was tested with natural
gas, #2 and #6 fuel oils. The Location 38 boiler had capability for staged
combustion air and variable combustion air preheat. Tests were conducted
with natural gas and #6 fuel oil. The results presented herein summarize
the gaseous and particulate emissions data, efficiency and conclusions, for
various combustion modifications.
3.1 LOCATION 19 TEST RESULTS
3.1.1 Location 19 Baseline Tests
Baseline measurements were made with the boiler in the "as found"
condition firing #2 oil, #6 oil, and natural gas. Baseline measurements were
made at the start of each group of combustion modification tests and when the
boiler fuel was changed. The boiler load for baseline and all combustion
modification tests was approximately 80% of rated capacity.
Complete baseline emissions data on each fuel are included in Table
3-1, which summarizes the emissions data from those tests during which
particulate, particulate size distribution or SO were measured. This table
includes baseline data and data at each of the modified boiler conditions.
Particulate, particulate size distribution, and SO measurements were made
X
only at the optimum low NO condition for each combustion modification or
combination of modifications. These data were not measured during the
parametric tests for each modification and the emissions data for the
parametric tests are not tabulated in Table 3-1.
27
-------
TABLE 3-1. SUMMARY OF MODIFIED BOILER TESTS
Test Run
19-5
19-7
19-74
19-76
19-83
19-85
19-116
19-117
19-179
19-97
19-99
19-132
19-143
19-159
19-170
19-147
19-171
19-177
19-181
19-184
19-186
19-188
19-190
19-193
a
Date
1977
1-6
1-6
1-19
1-19
1-20
1-20
1-26
1-26
3-3
1-24
1-24
2-17
2-21
2-23
3-1
2-22
3-2
3-2
3-4
3-4
3-4
3-4
3-7
3-8
Nominal
Steam
Load
Mg/hr
6.58
6.35
6.58
6.58
6.58
6.62
6.58
6.62
6.49
*
6.26
6.35
6.44
6.24
6.35
6.49
6.59
6.36
6.82
6.27
6.32
6.59
6.23
6.29
6.49
Fuel
#2
#2
#2
#2
#2
«2
#2
#2
»2
#6
#6
#6
#6
#6
#6
NG
NG
NG
NG
NG
NG
NG
NG
NG
°2
3.20
2.95
1.15
1.08
0.85
0.68
3.10
3.18
3.5
3.00
3.10
0.98
3.1
1.75
4.2
3.03
3.2
2.5
3.2
1.1
2.8
4.8
3.25
3.3
co2
13.5+
13. 2+
14.6+
14. 6+
14. 8+
14. 9+
13.0+
13.0+
13.2
13.6+
13. 6+
15. 2+
13.6+
14. 5+
13.2+
10.2
10.6
. 10.6
11.0
12.0
11.0
9.7
10.3
10.5
NOX*
ng/J
(ppm)
67.4
(120)
64.6
(115)
54.5
(97)
54.5
(97)
17.4
(31)
18.5
(33)
54.0
(96)
55.1
(98)
18. 0
(32)
120.3
(214)
123.1
(219)
84.3
(150)
84.9
(151)
84.3
(150)
90.5
(161)
48.5
(95)
12.2
(24)
11.2
(22)
45.9
(90)
44.4
(R7)
25.0
(49)
11.2
(22)
13.3
(26)
42.3
(83)
NO*
ng/J
(ppm)
67.4
(120)
64.6
(115)
54.5
(97)
54.5
(97)
17.4
(31)
18.5
(33)
54.0
(96)
55.1
;ri>
18.0
(32)
120.3
(214)
123.1
(219)
83.7
(149)
83.7
(149)
64.3
(150)
90.5
(161)
48.5
(95)
12.2
(24)
11.2
(22)
45.9
(90)
44.4
(87)
25.0
(491
11.2
(22)
13.3
(26)
42.3
(83)
HC*
ng/J
(ppm)
0.4
(2)
1.2
(6)
0.6
(3)
0.6
(3)
1.0
(5)
0.8
(4)
1.2
(6)
0
0
2.1
(11)
1.2
(6)
3.7
(19)
1.2
(6)
1.2
(6)
0
0.7
(4)
0.5
(3)
0
0.9
(5)
0
0
0
0
0.2
(1)
CO*
ng/J
(ppm)
1.4
(4)
4.1
(12)
61.7
(181)
61.7
(181)
7.2
(21)
24.2
(71)
11.9
(35)
8.2
(24)
6.1
(18)
1.4
(4)
1.4
(4)
62.4
(183)
8.2
(24)
30.7
(90)
9.2
(27)
1.2
(4)
6.2
(20)
5
(16)
5.9
(19)
308
(995)
90.7
(293)
4.0
(13)
5.0
(16)
7.1
(23)
so2*
ng/J
(ppm)
62.8
(80)
95.8
(113)
70.3
(90)
74.2
(95)
71.1
(91)
65.6
(84)
73.4
(94)
82.8
(106)
76.5
(98)
256.4
(329)
277.2
(355)
305.4
(391)
269.4
(345)
293.6
(376)
289.0
(370)
0
0
0
0
0
0
0
17.7
(25)
:•
Wet Chemistry
SO2*
ng/J
(ppm) j
76-3
(90)
—
9S.2
(127)
—
71.1
(91)
—
67.2
(86)
—
78.9
(101)
245
(314)
316.3
(405)
280.4
(359)
266.9
(329)
279.6
(358)
..
„
33.3
(47)
SBHBS
S03*
ng/J
(ppm)
1
(l.l)
—
3
(3)
—
5
(5)
—
1
(1)
7
(9)
4
(4)
3
(4)
3
(4)
2
(2)
6
(7)
— -
...
..
2
(3)
••••^•mMM
^^HBBBMpBB
28
(continued)
•Corrected to 3* O,
Calculated from fuel analysis
-------
TABLE 3-1 (continued).
Test Run
19-5
19-7
19-74
19-76
19-83
19-85
19-116
19-117
19-179
19-97
19-99
19-132
19-143
19-159
19-170
19-147
19-171
19-177
19-181
19-184
19-186
19-188
19-190
19-193
Fuel
»2
#2
#2
#2
»2
#2
#2
#2
#2
#6
#6
#6
#6
#6
#6
NG
NG
NG
NG
NG
NG
NG
NG
NG
Total
Partic.
ng/J
< Ib/MMB)
24.24
(0.0564)
—
16.29
(0.0379)
5.84
(0.0136)
__
11.6
(0.0270)
4.16
(0.0097)
36.21
(0.084)
28.87
(0.067)
31.80
(0.074)
32.33
(0.075)
28.79
(0.067)
—
—
—
—
—
2.63
(0.0061)
3.68
(0.0086)
Solid
Partic.
ng/J
(Ib/MMB)
2.545
(0.006)
—
5.95
(0.0138)
1.95
(0.0045)
9.01
(0.0210)
3.31
(0.0077)
27.55
(0.064)
25.80
(0.060)
27.20
(0.064)
29.36
(0.068)
9.10
(0.021)
—
—
—
—
—
1.67
(0.0039)
1.92
(0.0045)
Stack
Temp.
K
<°F)
542
(516)
539
(511)
545
(521)
545
(521)
578
(581)
579
(582)
544
(520)
542
(516)
564
(555)
539
(511)
543
(517)
519
(475)
526
(487)
548
(526)
544
(579)
554
(538)
561
(550)
541
(557)
538
(508)
527
(489)
534
(501)
550
(530)
555
(539)
541
(514)
Eff .
%
82.4
82.6
83.4
33.3
81.7
81.8
82.4
82.5
81.3
82.8
82.6
83.8
82.9
82.1
80.7
77.8
77.3
77.5
78.2
79.1
78.8
76.9
77.6
78.3
Smoke
Spot
0
0
5
4.5
0
0
4
5.5
0
, 0.5
0
2.5
2.5
1.5
1.5
—
—
—
—
—
~
~
—
~
FGR
%
—
—
28.4
27.9
—
—
26.3
—
—
—
~
24.7
23.1
0
20.3
19.9
—
—
—
17.8
17.2
0
Opacity
%
0
0
0
0
0
0
13
13
0
0
0
6
22
0
0
0
0
0
0
0
0
0
0
0
*
B
—
__
._
__
1.04
1.04
1.10
—
—
—
1.03
—
1.01
—
—
—
—
—
0.97
0.86
—
—
Lance
Depth
m
(ft)
~
__
__
__
__
1.8
(6)
1.8
(6)
1.2
(4)
—
—
—
1.5
(5)
—
1.2
(4)
—
—
—
—
—
2.1
(7)
2.1
(7)
—
—
Comments
Baseline - particulate
test
Baseline - Cascade
impactor
Low 0^ particulate
2
Low 0 - Cascade
impactor
FGR, Low 0., Particulate
2
FGR, Low O,, Particulate
2
Staged air - Particulate
Staged air - Cascade
impactor
FGR B SA - Low O , Cascade
impactor, Particulate
Baseline - Particulate
Baseline - Cascade
impactor
Low O - Particulate
Staged Air, Particulate
Max. FGR, Particulate
FGR & SA, Particulate
Baseline ring burner
Max. FGR, Gas gun
burner
Max. FGR, Low 02> Gas
gun burner
Baseline, Gas gun
burner
Minimum 0 , Gas gun
burner
Max. SCA, lance at 7'
FGR S SCA - Normal 0^
Max. FGR, Particulate
Baseline - Particulate
29
-------
The measured baseline NO emissions with the boiler firing #2 oil
X
were 65.8 ng/J (117 ppm). With the unit firing #6 oil, baseline NOx emissions
were 121.7 ng/J (216 ppm). Baseline NO emissions with natural gas w-re
X
42.3 ng/J (83 ppm).
Baseline particulate measurements were made with the boiler firing
#2 oil, #6 oil and natural gas. With #2 oil, the total particulate was 24.24
ng/J (0.0564 Ib/MMBtu) and the solid particulate was 2.595 ng/J (0.0060
Ib/MMBtu). Number 6 oil baseline particulate measurements were 36.21 ng/J
(0.084 Ib/MMBtu) total particulate and 27.55 ng/J (0.064 Ib/MMBtu) solid
particulate. When firing natural gas, the total particulate measured
3.68 ng/J (0.0086 Ib/MMBtu) and solid particulate measured 1.92 ng/J
(0.0045 Ib/MMBtu).
During this test program, combustion modifications were evaluated
using all three fuels. Oil samples were taken periodically during the
test program and sent to an independent testing laboratory for analysis.
A natural gas sample was also taken during the testing and submitted for
analysis. A summary of fuel properties is presented in Tables 3-2 and
3-3 for" oil and natural gas respectively.
3.1.2 Location 19 Combustion Modifications With #2 Oil
Combustion modification testing with the unit operating on #2 oil
consisted of excess air variation, staged combustion, flue gas recirculation,
and combinations of staged air and flue gas recirculation.
Excess Oxygen—
The effect of excess oxygen on NO emissions was evaluated by Tests
X
1, 11, 12, and 13. These data are presented in Figure 3-1. Reducing excess
O resulted in a decrease of 2.7 ng/J (0.0063 Ib/MMBtu) per % O
30
-------
TABLE 3-2. SUMMARY OF LOCATION 19 FUEL OIL ANALYSES
Fuel
Date
Laboratory No.
Carbon , %
Hydrogen , %
Nitrogen, %
Sulfur, %
Ash, %
Oxygen, %
API Gravity
HHV, Btu/lb
HHV, kJ/kg
#2 Oil
1/6/77
13520
86.45
13.01
<0.001
0.16
<0.001
0.38
37.8
19,680
45,775
#2 Oil
1/10/77
13520
85.62
12.99
<0.001
0.10
<0.001
1.09
37.7
19,680
45,775
#2 Oil
1/19/77'
14009
86.49
13.17
0.008
0.14
0.001
0.19
37.5
19,610
45,613
#6 Oil
12/15/76
13042
87.55
11.49
0.20
0.56
0.020
0.18
20.7
18,910
43,984
#6 Oil
1/19/77
14009
86.91
11.78
0.20
0.54
0.019
0.55
20.7
19,000
44,194
#6 Oil
2/16/77
14667
87.55
11.40
0.23
0.60
0.034
0.19
17.0
18,780
43,682
#6 Oil
3/2/77
14667
87.30
11.34
0.22
0.60
0.026
0.51
17.3
18,850
43,845
#6 Oil
3/10/77
16554
87.23
11.34
0.23
0.55
0.027
0.62
17.5
18,850
43,845
#6 Oil
3/17/77
16554
86.06
11.11
0.32
1.17
0.025
1.32
14.3
18,670
43,426
#6 Oil
3/21/77
16554
86.25
11.25
0.32
1.18
0.024
0.98
15.7
18,740
43,589
#6 Oil
3/23/77
16554
86.55
11.26
0.30
1.02
0.025
0.85
15.8
18,750
43,613
-------
TABLE 3-3. LOCATION 19 NATURAL GAS ANALYSIS
Date
Laboratory No.
3/7/77
14781
Oxygen, % <0.009
Nitrogen, % 0.29
Carbon dioxide, % 0.54
Methane 95.84
C 2 2.92
C 3 0.22
C 4 0.099
C 5 0.039
C 6 plus 0.054
Heating value, Btu/SCF (dry) 1035
32
-------
150
,600
o
(11)
400 7x
O
oV
ro
o
Location 19
Load = 83% of Rated
Fuel: No. 2 Fuel Oil
Air Atomization
200
o
u
1234
Stack Gas Excess Oxygen, % Dry
Figure 3-1. The effect of excess oxygen on NO emissions
(No. 2 oil)
33
-------
Staged Combustion Air—
For the staged air tests, the depth of the lances which supplied the
staged air was varied from 1.2 to 2.1m (4 to 7 ft) and the ratio of burner-
to-staged combustion air was changed. Total furnace length is 3.0 m (9.9 ft).
The effect of staged combustion air injection point on NO emissions is pre-
X
sented in Figure 3-2 for the unit operating on #2 oil. These data indicate
that the effect of injection distance is very slight beyond about 1.2 m
(4 ft) and the more fuel rich the burner operates, the greater the reduction
in NO . At an injection depth of 1.2 m, the NO was reduced 27% from the
X X
baseline condition and increasing the injection point to 2.1 m resulted in
a reduction of 30% from the baseline condition, with the burner operating
at 91% of the theoretical air ( = 1.1). The symbol (j> is the equivalence ratio,
defined as the ratio of stoichiometric air-fuel ratio to actual air-fuel ratio.
With the burner operating at approximately the theoretical air-fuel ratio, the
reduction is only 15.5% at an injection depth of 2.1 m. The effect of lowering
the overall excess air with staged combustion is also shown in this figure.
A reduction of 44% in NO was measured with the lance at 2.1 m and the
x
burner operating at 91% theoretical air while the overall O2 level was
reduced from 3% O2 to 2.6%.
Flue Gas Recirculation—
The influence of flue gas recirculation on NO emissions was
X
evaluated at two excess O conditions. Tests were conducted at a nominal
O of approximately 3% and a low 0 condition of approximately 1%. The
test results are presented in Figure 3-3. At the nominal O condition
(^ 3%), flue gas recirculation results in a decrease in NO of 68% for
X
the maximum recirculation rate of 23.6%. Even the low recirculation rate
of 14.7% results in a decrease in NOx of 52%. With the boiler operating in
the low 02 condition, the maximum recirculation rate of 26.5% resulted in a
reduction in NO of 71% and the minimum recirculation rate of 15% gave a
58% reduction in NO .
X
34
-------
200
(A/F)
stoic
(A/F)
actual
Location 19
Load: 83% of Rated
Fuel: No. 2 Fuel Oil
Air Atomization
(38)
3.0- 3.1% O
(32s
'(90) (91)
(Test No.)
I
23456
Secondary Air Tube Insertion Depth, ft
I I
Meters
Figure 3-2. NOX emissions as a function of staged air injection depth
and burner air (#2 oil).
35
-------
125
Location 19
Load'= 83% of Rated
Fuel: No. 2 Oil
Air Atomization
(17)
Q2.85%
(21)
QO.9%
(Test No.)
Excess O
10 15 20
Recirculated Flue Gas, %
Figure 3-3.
NOX emissions as a function,of percent flue gas
recirculation (No. 2 oil).
36
-------
Flue Gas Recirculation and Staged Combustion Air—
Tests were conducted to evaluate the effect of flue gas recirculation
in combination with staged combustion air. NO as a function of flue gas
X
recirculation rate is presented in Figures 3-4 and 3-5 for staged air injec-
tion depths of 1.2 m and 2.1 m respectively. NO reductions of 77% and 73%
X
were demonstrated using both maximum flue gas recirculation and staged air at
1.2 m and 2.1 m injection depth respectively with the boiler in the low O
operating mode. This is a slightly greater reduction in NO than realized by
X
flue gas recirculation alone. The combination of FGR and staged combustion
with the boiler in the normal 0 mode resulted in a 69% reduction in NO .
^ X
3.1.3 Combustion Modifications With #6 Oil
Combustion modification testing with the unit operating on #6 oil
consisted of excess air variation, staged combustion, flue gas recirculation
and combinations of flue gas recirculation and staged air.
Excess Oxygen—
The effect of excess air on NO emissions for #6 oil firing is shown
X.
in Figure 3-6. Excess 0- was varied over a range of 0.7% to 5.2%. Lowering
the O« level to 0.7% resulted in a 30% decrease in NO from the baseline
>£ X
condition. The overall effect of O on NO emissions is 11.1 ng/J (0.026
£ X
Ib/MMBtu) decrease in NO per percent decrease in excess 0 .
Staged Combustion Air—
During this test series staged combustion air injection depths were
varied. The results are shown in Figure 3-7. As in the case of #2 oil
combustion, little effect on NO was detected beyond 1.2 m insertion depth
X
but the effect of burner equivalence ratio (((>) is more pronounced. At the
normal O condition (^ 3.1% O ), changing the burner air from slightly air
£-• £
rich (cj> ^ 0.96) to slightly fuel rich (cf> > 1) resulted in a NOx emission
reduction of about 13%. Lowering the overall 0 level to 1.9% while
maintaining the burner in the fuel-rich condition resulted in an additional
decrease of 14%. These reductions are both relative to the staged combustion
condition with the lance depth set at 2.1 m. Relative to the baseline
condition, the reductions are: 19% for normal 02, burner slightly air rich;
29% for normal 0 , burner slightly fuel rich; and 42% for low O2, burner
slightly fuel rich.
37
-------
125
Location 19
Load - 83% of Rated
Fuel: No. 2 Fuel Oil
Air Atomization
Staged Air Injection
Depth = 1.2 m (4 ft)
(Test No.)
10 15 20
Recirculated Flue Gas, %
Figure 3-4. NOX emissions as a function of combined flue gas
recirculation and staged air at 1.2 m (No. 2 oil)
38
-------
125
Location 19
Load: 83% of Rated
Fuel: No. 2 Oil
Air Atomization
Staged Air Injection
Depth = 2.1 m (7 ft)
(103)
2.9%
(56)
2.7%
= 1.05 - 1.1
(Test No.)
Excess O
10 15 20
Recirculated Flue Gas, %
Figure 3-5. NOX emissions as a function of combined flue gas
recirculation rate and staged air at 2.1 m (7 ft)
39
-------
300
300
Location 19
Load: 83% of Rated
Fuel: No. 6 Fuel Oil
Air Atomization
(Test No.)
1234
Stack Gas Excess Oxygen, % Dry
Figure 3-6. The effect of excess oxygen on NO emissions
(No. 6 oil). x
40
-------
** —» v
c
200
^i
p
0^150
OP
•P
id
6 100
ft
^
X
o
S 50
n
1 1 1
(69)
1 1
Normal O.
^ 3.1% A < l <*
^~^ -^"^l ^/U09. (110, -
^^^•8^.22
— <^ /^
3. 05%^ v-
di3p — 7
1.75% /
" /
Low ©2
Location 19 ^> > i
Load - 83% of Rated B
— Fuel: No. 6 Fuel Oil
Air Atomization
1 1 1 1 1
^_j^ 3.07% 3.11%
% § ^= 1 . OT^ (112)
"\ (~\ 1 ] a —
•^(71)3. 3%^7 ^
. : /__n(114)
/ W 1.9%
V^Normal O —
^g > 1
—
1 1
012345678
Secondary Air Tube Insertion
L |
Depth, ft
I
n 1 2
Secondary Air Tube Insertion Depth, m
Figure 3-7. NO emissions as a function of staged air injection depth and
burner air (#6 oil).
41
-------
Flue Gas Recirculation—
Flue gas recirculation tests with #6 oil were conducted with the
boiler operating in the normal and low O2 conditions. The data from these
tests are presented in Figure 3-8. Flue gas recirculation rates were varied
from 13.8% to 23.3% at the normal O condition and from 14.4% to 25.8% at
the low Q condition. With the unit operating at normal O^ levels, flue
gas recirculation resulted in an 11% decrease while at the low O^ conditions,
a 40% reduction in NO was realized. Both of these reductions were accomplished
x
at the maximum recirculation rates of 23.3 and 25.8% respectively.
Flue Gas Recirculation and Staged Combustion Air—
The effect of flue gas recirculation combined with staged combustion
air was determined for the boiler firing #6 oil. The combined operation
tests were conducted at nominal and low excess C2 conditions and at staged
air injection depths of 1.2 and 2.1 m. Figure 3-9 shows the effect of flue
gas recirculation rate on NO emissions at the nominal, high and low excess
X,
O conditions with staged air injection at 1.2 m. The maximum reduction in
£*
NO occurred with the excess O0 at 1.5% and 19.6% flue gas recirculation.
x 2 '
The NO was reduced 50% relative to the normal baseline at this operating
X
condition. Figure 3-10 shows the same data but with an injection depth of
2.1 m. The maximum reduction at this condition is 53% relative to the base-
line condition. The effect of burner equivalence ratio,
-------
250
200
150
100
50
(Test No.)
Excess O
O Normal 0,
^
Q Low On
% FGR
C.A.
I
I
m
.Location 19
Load = 83% of Rated
Fuel: No. 6 Oil
Air Atomized
x 100
fuel
I
I
5 10 15 20 25
Recirculated Flue Gas, % of Total
30
Figure 3-8. The effect of flue gas recirculation on.NO
emissions (No. 6 oil).
43
-------
250
B/L 3.6%,
(133)
232 ppm
50
Location 19
Load = 83% of Rated
Fuel: No. 6 Oil
Air Atomization
= 1.01
= 1.09
1.65%
(134)
= 1.19 1.5% <
1.21
Q High 02
A Normal O,
£.
O Low O0
( ) Test No.
Secondary Air Lance Depth = 1.2 m (4 ft)
10 15 20
Recirculated Flue Gas, %
25
30
Figure 3-9.
The effect of combined flue gas recirculation and staged combustion
air at 1.2 in on NO emissions (#6 oil).
x
-------
250
Ul
3.25%
(1) (128)
200
•o
M 150
O
*>
ro
e
100
i
50
Location 19
Load = 83% of Rated
Fuel: No. 6 Oil
Air Atomization
0.99
0.99
= 1.1
^ Normal O?, A
O Normal O-, d>
2 B
= 0.99
= 1.1
Secondary Air Lance Depth - 2.1 m (7 ft)
( ) Test No.
Excess O_
10 15 20
Recirculated Flue Gas, %
25
30
Figure 3-10. The effect of combined flue gas recirculation and staged combustion air
at 2.1 m on NO emissions (#6 oil).
-------
Excess Oxygen—
Figure 3-11 presents the test results of NO variation versus excess
X
O variation with the natural gas gun burner. These data show that reducing
the excess O level from the baseline condition of 3.2%, to 2.0%, resulted in
an increase in NO emissions of approximately 2%. Diminishing the O level
further, to 1.1%, decreased the NO by 3% from the baseline condition. The
X
lower limit of excess 0 was determined to be approximately 1% based on the
effect of O on CO emissions as shown in Figure 3-12. This figure shows that
below about 2% 0 , the CO emissions increase very rapidly which decreases
boiler efficiency. Increasing the 00 to 6.15% reduced the NO by nearly 9%,
£ X
but with an accompanying decrease in efficiency.
Staged Combustion Air—
Staged combustion with natural gas fuel was evaluated with the gun
burner in Test 185 and 186. Figure 3-13 shows the effect of secondary air
injection depth on NO emissions. As was the case with the ring burner, the
X
effect diminishes beyond 1.2 m (4 ft) injection depth with 85% of the
reduction in the initial 1.2 m (4 ft). NO was reduced by approximately
X
38% by injecting the staged air at 1.2 m (4 ft) whereas at 2.1 m (7 ft),
the NO was reduced by 46%. The burner equivalence ratio for these tests
X
was 0.96.
Flue Gas Recirculation—
The effectiveness of flue gas recirculation as an NO reduction
x
technique was evaluated with the boiler firing natural gas. Tests 172
through 177 were conducted with the flue gas recirculation rate varied up
to a maximum rate of 20%. The initial gas burner configuration - a ring
burner design - was unstable with even small amounts of recirculated flue
gas. A narrow plate was installed in the burner to shield the gas jets from
the flue gas but the combustion was still unstable. A gas gun burner was then
installed and tested in place of the ring burner. A combination of gas
orifices, swirl, gun and diffuser position was found which was stable with
46
-------
100
o
ro
I
o
a
80
60
40
20
(184)
(182)
( ) Test Number
Location 19
Load: 83% of rated
Fuel: Natural Gas
Gas Gun Burner
0 12 345 6
Stack Gas Excess Oxygen, %, Dry
Figure 3-11. The effect of excess oxygen on NOX emissions (natural gas;
gas gun).
-------
1000
03
800
600
o
ro
| 400
<5
u
200
Location 19
Load: 83% of rated
Fuel: Natural Gas
Gas Gun Burner
( ) Test Number
4-
2345
Stack Gas Excess Oxygen, %, Dry
(182)
Figure 3-12. The effect of excess oxygen on CO emissions (natural gas; gas gun)
-------
100
60
o
ro
o, 40
20
Location 19
Load: 83% of rated
Fuel: Natural Gas
Gas Gun Burner
4>B = 0.96
( ) Test Number
Excess Oxygen
2345
Secondary Air Tube Insertion Depth, Ft
Secondary Air Tube Insertion Depth, m
Figure 3-13. NOX emissions as a function of staged air injection depth (natural gas; gas burner)
-------
atgt, '"lue cas recirculation rates. The effect of flue gas recirculatj, >> <:•,>
iv.; ciu-L.,.;! :..-D iri shown in Figure 3-14. The data indicate that even s>a<..i
x
ai..cunts of f^ae gas recirculation (8%) result in relatively large (4;.*)
re tuition _.n NO emissions. At the maximum flue gas recirculation rate
X
as limited by combustion stability, the reduction is 77% with the bo:, it.*
o ... rating at 3.2% 0 and 79% with the boiler at 2.5% excess O2-
Also illustrated in the figure i:. the effect of low O^ operation in
conjunction with flue gas recirculation. The amount of flue gas recirculateu
to the burner was increased until the flame stability limit was reached. Wi*r,
tne gas gun burner., the maximum amount of flue gas which could be re^ircuiat.eo
was approximately 20%. At the maximum recirculation rate, the measured WO
ai.ue was 12.2 ny'/J (24 ppm) at the normal O» condition. At low O~ conditions,
£ "
cue measured WO value was 11.2 ng/J (22 ppm). These values represent reduc-
tions of /7% and 79% from the baseline condition.
Flue Gas Recirculation and Staged Combustion Air—
The effect of flue gas recirculation combined with staged coruhustiqn
air was evaluated for the boiler firing natural gas with the gas gun burners.
The tests were conducted with the maximum amount of staged combustion as
limited by the secondary air flow. The staged air was injected at i.2
(4 ft) and 2.1 m (7 ft) for these tests. The percentage of recirculated
flue gas was varied up to the maximum determined by burner stability iima.*. i
The maximum flue gas recirculation rate was 17.8%. The reduction in NO wrift
x
76% at the maximum recirculation rate of 17.8% and 68% at a recirculation
rate of 11%. The combination of flue gas recirculation and staged combustion
air gave the greatest reduction in NO , but only 5% more than flue ga^
X
recirculation alone. The effect of combined flue gas recirculation ana
staged combustion air is presented in Figure 3-15 for these tests.
50
-------
120
iocr
60
Cvl
o
of
©
a 40
a
§
20
Location 19
Load: 83% of Rated
Fuel: Natural Gas
(Gas Gun)
Stability Limit
3.75%O
(176)
Excess Oxygen
( ) Test Number
10 15 20
Recirculated Flue Gas, %
25
30
Figure 3-14. The effect of flue gas recirculation rate and low excess O
on NO emissions (natural gas).
Jt
51
-------
100
M
T3
Jj
<0
a,
0*
o
z
CN 60
Baseline
(181)
3.2%
20 —
o L
Location 19
Load: 83% of rated
Fuel: Natural gas
Gas Gun Burner
B
(187) (188)
5.0% 4.8%
0.85 0.86
5 10 15
Recirculated Flue Gas, %
20
Figure 3-15. The effect of combined flue gas recirculation and staged
combustion air on NO emissions (natural gas).
X
52
-------
3.1.5 Particulate and SO Testing
x =1
Particulate tests were conducted at baseline conditions with the
boiler firing #2 oil, #6 oil, and natural gas fuel. Particulate
measurements were also made at low 0 conditions, flue gas recirculation
with low O , staged combustion air, and flue gas recirculation in combination
with staged combustion air with #2 oil and #6 oil. A particulate test with
flue gas recirculation and low O was conducted on natural gas fuel. A
summary of all Method 5 particulate measurements is presented in Table 3-4.
With #2 oil, #6 oil, and natural gas all modified tests resulted in lower
total particulate than measured at baseline conditions. Particulate size
distributions are presented in Figure 3-16a for #6 oil and Figure 3-16b
for #2 oil. These data show that for all but one test, between 30 and
50% of the particulate is 3 l_im or less aerodynamic diameter.
Wet chemistry SO was measured firing #? oil, #6 oil, and natural gas
A
fuel at baseline and modified boiler conditions. Total sulfur oxides emissions
for the boiler firing oil ranged from 68 ng/J (87 ppm) with #2 oil (0.13% S)
to as high as 319 ng/J (409 ppm) with #6 oil (0.5 to 1.2% S). The level of
sulfur oxides emissions is dependent solely upon the sulfur content of the
fuel. A sulfur content of 1% in an oil fuel results in approximately 445 ng/J
(580 ppm at 3% O ) of sulfur oxides emissions. Total sulfur oxides emissions
with natural gas were below detection for all tests except one, for which SO
X
emissions were 18 ng/J (25 ppm) by the Dupont analyzer and 33 ng/J (47 ppm)
by wet chemistry. SO varied from 10 to 30 ppm that test period (March 7-8,
X
1977). Because of the severe gas shortage (winter of '77) the natural gas may
have not been of normal pipeline quality, which requires negligible sulfur.
3.1.6 Trace Species and Organics Emissions (TSSO)
Four tests were conducted at Location 19 to sample for trage species
and organics. The sampling and analysis procedures are described in detail in
Appendix A. Table 3-5 presents the sampling conditions for the trace species
and organics tests. Two tests were conducted at baseline conditions (19-1,2)
and two were at the optimum low-NO condition (19-3,4). All tests were
X
conducted with #6 oil. After the initial baseline test, however, a new
load of #6 oil was received by the operator. The new shipment of oil had
53
-------
TABLE 3-4. SUMMARY OF METHOD 5 PARTICULATE MEASUREMENTS
FOR LOCATION 19 STEAM BOILER
un
Boiler
Operating
Mode
Baseline
Low 0_
A
SCA
FGR, Low O
FGR + SCA
No. 2 Oil
Total Solid
ng/J ng/J
(Ib/MMBtu) (Ib/MMBtu)
24.24 2.595
(0.0564) (0.0060)
16.29 5.95
(0.0379) (0.0138)
11.6 9.01
(0.0270) (0.0210)
5.84 1.95
(0.0136) (0.0045)
4.16 3.31
(0.0097) (0.0077)
No. 6
Total
ng/J
(Ib/MMBtu)
36.21
(0.084)
28.87
(0.0672)
31.8
(0.0743)
32.33
(0.075)
28.80
(0.0670)
oil
Solid
ng/J
(Ib/MMBtu)
27.55
(0.064)
25.80
(0.060)
27.2
(0.0635)
29.36
(0.068)
9.10
(0.021)
Natural Gas
Total
ng/J
(Ib/MMBtu)
3.68
(0.0086)
2.63
(0.0061)
Solid
ng/J
(Ib/MMBtu)
1.92
(0.0045)
1.67
(0.0039)
-------
30.0
10.0
o
in
Q
w
H 3.0
a
w
H
g
o
Q
s
W
^ 0.5
0.3
0.1
Mill I I I I I 1 I—I I I |
Test No.
©143 (#6, SA)
O170 (#6, FGR H- SA)
Ql32 (#6, Low 02)
^159 (#6, FGR)
-------
F- 11 i i i I—n—MMI
30.0
10.0
o
in
5.0
3.0
a
u
H
0.5
0.3
0.1
I I I I
Test No.
(7 179 (#2, FGR+SA+Low 02) __
n 85 (#2, Low 0_)
£
O 76 (#2, Low 02)
O 7 (#2, Baseline)
Location 19
I I I I I I I I I I I I I i III I
I
0.01 0.1 0.5 1 2 5 10 20 3040506070 80 90 95 9899
CUMULATIVE PROPORTION OF IMPACTOR CATCH, % Volume
99.9 99.99
Figure 3-16b.
Particulate size distribution for an oil fired steam boiler
(No. 2 oil).
56
-------
TABLE 3-5. TRACE SPECIES AND ORGANICS SAMPLING
CONDITIONS, WATER TUBE BOILER - LOCATION 19
Ul
-o
Test Number
Date
Port Location
Velocity, m/s (f/s)
Stack Temp., K (° F)
Oxygen Content, % Dry
Moisture, %
Sample Time, min.
Cyclone Flow, awm /m (awcfm)
Isokinetic Rate, %
Oven Temp. , K (° F)
XAD-2 Temp., K (° F)
Nozzle Size, mm (in.)
No. of Filters Used
Sample Flow, Dry, DNm /m (scfm)
Volume Collected, Dry, DNm (scf)
Particulate Collected, g
Solid Particulates, ng/J (Ib/MMBtu)
Unit Conditions
Test Time, min.
Steam Flow, Mg/h (Mlb/h)
Fuel
19-1 SASS
3/10/77
stack
7.55 (24.8)
539 (511)
2.9
7.6
300
0.115 (4.07)
99.7
479 (402)
290 (63)
19 (0.75)
1
0.065 (2.31)
19.6 (693.6)
0.4289
6.26 (0.0146)
305
6.45 (14.3)
#6 oil
19-2 SASS
3/17/77
stack
7.23 (23.7)
541 (514)
3.0
8.2
300
0.118 (4.15)
107.4
479 (402)
291 (64)
19 (0.75)
1
0.066 (2.35)
20.0 (704.7)
2.7416
39.5 (0.092)
302
6.62 (14.6)
#6 oil*
19-3 SASS
3/21/77
stack
7.37 (24.2)
564 (556)
1.8
8.4
300
0.120 (4.24)
110.9
480 (405)
290 (62)
19 (0.75)
1
0.67 (2.38)
20.2 (713.3)
4.6981
63.1 (0.147)
302
6.53 (14.4)
#6 oil*
19-4 SASS
3/23/77
stack
6.95 (22.8)
552 (534)
1.5
8.0
300
0.117 (4.15)
114.7
478 (400)
288 (59)
19 (0.75)
1
0.066 (2.35)
19.97 (705.8)
3.8446
51.28 (0.119)
^02
6.42 (14.1)
#6 oil*
*New shipment of #6 oil received
-------
significantly higher sulfur content than the original oil. Samples were
taken and were submitted for analysis. All tests were conducted for five
hours sample time. The samples were prepared by KVB and transmitted to the
laboratory for analysis. Because of the significant difference in the fuel
on Test 19-1, only SASS train samples from Tests 19-2, 3 and 4 were analyzed.
Table 3-6 presents the gaseous and particulate (by SASS) measurements
and efficiency data. Table 3-7 presents the fuel sample analyses. The fuel
analyses for Tests 19-2, 3 and 4 indicate sulfur content was twice that of
the fuel for Test 19-1 and fuel nitrogen content was about 40% higher.
Table 3-6 indicates that NO emissions increased by 15% for Test 19-2 compared
X
with Test 19-1. This indicates that the fuel nitrogen conversion to NO was
X
about 24%. Assuming 24% fuel nitrogen conversion for each test and subtract-
ing the fuel nitrogen NO from total measured NO results in 77 ng/J (139
X X
ppm) of thermal NO for both Test 19-1 and Test 19-2.
X
Tests 19-3 and 19-4 are duplicate runs at the optimum low NO con-
X
dition with reduced excess air and maximum flue gas recirculation. NO
X
emissions were reduced by 28% compared with the Test 19-2 baseline.
Total particulate emissions were significantly higher for Test 19-2
with the higher sulfur fuel compared with Test 19-1. There was a further
increase in particulates of 30 to 60% for the two low NO tests compared with
X
baseline Test 19-2 on the same fuel. These results are in contrast to
previous results presented in Section 3.1.5 for which the comparable low
NO condition produced no change in solid particulates as compared with
X
baseline.
Samples were analyzed by atomic absorption (AA) and spark source mass
spectrometry (SSMS) to determine concentrations of elements. Wet chemistry (WC)
was used for chloride, fluoride, nitrate and sulfate. POM and PCB were analyzed
by gas chromatography (GC). The XAD-2 resins for two tests (19-2, 19-3) were
analyzed by gas chromatography-mass spectrometry (GC-MS) to quantify specific
POM compounds. Appendix A contains the details of the analytical procedures.
58
-------
TABLE 3-6. SUMMARY OF EMISSIONS DATA AT LOCATION 19 DURING
TRACE SPECIES AND ORGANICS TESTS (TS&O)
Test Run
19-1 SASS
19-2 SASS
19-3 SASS
19-4 SASS
Date
1977
3-10
3-17
3-21
3-23
Steam
Load
Ko/h
G.51
6.63
6.54
6.41
Fuel
16
#6
#6
#6
°2
%
2.9
3.0
1.8
1.5
C02
%
13.9
13.8
14.5
14.4
NOX*
ng/J
(ppm)
118
(213)
135
(244)
97
(176)
98
(176)
NO*
ng/J
(ppm)
118
(213)
135
(244)
97
(176)
98
(176)
HC*
ng/J
(ppm)
0.2
(1)
1.3
(7)
0.6
(3)
._
CO*
ng/J
(ppm)
7
(20)
12
(37)
55
(162)
37
(110)
S02*
ng/J
(ppm)
305
(395)
638
(831)
627
(815)
578
(750)
SASS
Solid
Partic.
ng/J
(Ib/KMS)
6.26
(0.0146)
39.5
(0.092)
63.1
(0.147)
51.28
(0.119)
Stack
Temp.
K
(°F)
541
(513)
543
(517)
562
(551)
560
(548)
Eff.
%
82
82
81.8
82
Smoke
Spot
0
0.5
1.0
0.5
%
FGR
0
0
34
35
Opacity
0
2.5
7.5
5
A
?B
—
Lance
Depth
m
(ft)
—
Comments
Baseline TSSO Test
Baseline TSsO Test
New oil delivery
Low 02 , Max. FGR
TSsO Test
Low 02, Max. FGR
TSSO Test
Ul
*Data reported on 3* O , dry basis
-------
TABLE 3-7. SUMMARY OF LOCATION 19 TEST FUEL OIL ANALYSES
FOR TRACE SPECIES AND ORGANICS TESTS
Fuel
Date
Test Number
Carbon , %
Hydrogen , %
Nitrogen, %
Sulfur, %
Ash, %
Oxygen , %
API Gravity
Heat Content kJ/kg
(Btu/lb)
#6 Oil
3-10-77
19-1
87.23
11.34
0.23
0.55
0.027
0.62
17.5
43845
(18850)
#6 Oil
3-17-77
19-2
86.06
11.11
0.32
1.17
0.025
1.32
14.3
43426
(18670)
#6 Oil
3-21-77
19-3
86.25
11.25
0.32
1.18
0.024
0.98
15.7
43589
(18740)
#6 Oil
3-23-77
19-4
86.55
11.26
0.30
1.02
0.025
0.85
15.8
43613
(18750)
60
-------
Results of the trace species and organics analyses are presented in
Tables 3-8 through 3-19. The following list provides a key to these tabula-
tions :
Test Table Results
All 3-8 General Notes for all Tables
19-2 3-9 AA* for solids section
3-10 AA for organic module and impingers
3-11 AA for total emissions, fuel and mass balance
3-12 SSMS* results (3 pages)
19-3 3-13 AA for solids section
3-14 AA for organic module and impingers
3-15 AA for total emissions, fuel and mass balance
3-16 SSMS results (3 pages)
19-4 3-17 AA for solids section
3-18 AA for organic module and impinger
3-19 AA for total emissions, fuel and mass balance
The results are presented for each sample as concentration in the sample
(yg/g) and in the stack flue gas (ug/m ) . For mass balance comparison the
total emission rate and fuel input are presented as a flow rate (yg/s).
Emissions contained in particles less than 3 microns collected by the 1 um
cyclone and filter are also given in Tables 3-11, 3-15, and 3-19 for each
test.
The SSMS results are presented on three pages for each of two tests
(19-2, 19-3). The first page contains the results for all elements also
determined by AA analysis. The second and third pages contain results for
all other elements as determined by SSMS.
All twenty-two inorganic elements specifically sought to be identified
were detected in the SASS samples. Several elements (antimony, beryllium,
mercury, selenium, tellurium and tin) could not be detected by AA analysis.
* AA = Atomic absorption
SSMS = Spark source mass spectrometry
61
-------
TABLE 3-8. GENERAL NOTES FOR TRACE SPECIES AND ORGANICS DATA TABULATIONS
1. All sample data are rounded to two significant digits and corrected
for blanks.
2. Single number indicates all sample concentrations were above detection
limits.
3. Single number preceded by "<" indicates all samples were less than
detection limits. Value shown is maximum amount that could be present
if the sample actually contained an amount equal to the detection limit
value.
4. For two numbers separated by "<", the number on the left of < indicates
the detected amount, and the number on the right indicates the maximum
amount including the detected amount plus the amount that could be
present in samples reported as below detection, if those samples
actually contained an amount equal to the detection limit value.
5. < DL, concentration below detection limits
=B, sample value equals blank, net value assumed zero
< B, sample value less than blank, net value assumed zero
MC, major component, exceeds maximum measureable quantity (about
1000 yg/g for spark source mass spectrometry)
NES, not enough sample for adequate analysis
NR, not reported, results uncertain because of complex sample matrix
composition
6. Species for which either the emission rate or input (or both) were below
detection limits have mass balance values indicated as follows:
< DL, both emission and input below detection limit
> value, input value is below detection limit or emission value is
above detection limit
< value, emission value is less than detection limit.
62
-------
CTi
TABLE 3-9. TRACE SPECIES AND ORGANIC EMISSIONS, SASS SOLIDS SECTION COLLECTION
Test 19-2, Modified Boiler, Location 19, Baseline Condition
Samole Tyoe
Sample Naraber
Sample W»iqht/Vol.
Units
Antimony
Arsenic
Barium
Beryllium
Cadmium
Cslciura
Chromium
Cobalt
Copper
Iror,
Lead
Manganese
Mercury
Nickel
Selenium
Tellurium
Tin
Titanium
Vanadium
Zinc
Chloride
Fluoride
Nitrates
Sulfates
Total POM
Total PCS
Nozzle, Probe,
10 urn Cyclone
Solids
566
1.6620 g
vg/g
< 38
< 1.5
38
< 0.8
0.8
1900
12
120
46
5400
150
51
< 0.03
1900
< 1.5
< 38
< 76
< 460
6300
400
279
205
113
14200
NES
NES
pg/m
< 3.2
< 0.13
3.2
< 0.065
0.065
160
6
10
3.8
450
13
4.3
< 0.0025
160
< 0.13
< 3.2
< 6.3
< 38
520
33
23
17
9.4
1200
NES
NES
3 un Cyclone
Solids
716
0.4443 g
Ug/g
< 210
33
460
< 4.2
< 4.2
1900
140
310
50
7200
< 20
75
< 0.17
2200
< 8
< 210
< 420
<1200
10000
400
NES
134
NES
NES
NES
NES
Ug/raJ
< 4.7
0.73
10
< 0.093
< 0.093
42
3.1
6.7
1.1
160
< 0.44
1.7
< 0.0038
49
0.13
< 4.7
< 9.3
< 27
220
8.9
HES
3
NES
NES
NES
NES
1 Urn Cyclone
Solids
720
0.2196 g
P9/9
< 230
< 10
6000
< 5
< 5
3500
310
940
100
32000
< 25
210
< 0.?.
900^
< 10
< 250
< 500
2500
43000
1300
NES
909
NES
NES
NES
NES
ug/m
< 2.7
< 0.11
66
< 0.055
< 0.055
36
3.4
10
1.1
350
< 0.27
2.3
< 0.0022
99
< 0.11
< 2.7
< 5.5
27
470
14
NES
10
NES
NES
NES
NES
Filters
538
0,4157 g
Ug/g
< 500
200
800
< 10
< 10
4800
240
1600
460
21000
1000
250
< 0;4
30000
< 20
< 500
<1000
<3000
39000
4900
11600
< 2
39.5
457000
NES
NES
ug/m
< 10
4.2
17
< 0.2
< 0.2
100
5
33
9.6
440
21
5.2
< 0.008
620
< 0.42
< 10
< 21
< 62
1800
100
240
< 0.04
0.82
9500
NES
NES
Solid
Section
Wash
19-2A
1605 ml
yg/ml
< 0.5
0.018
< 0.1
< 0.005
0.005
0.49
0.09
< 0.2
0.06
2.6
0.14
0.17
< 0.005
0.77
< 0.01
< 0.3
< 1
< 1
2.5
1.7
2.1
< 0.1
0.24
7.P
NR
NR
1
ug/ra
< 40
1.4
< 8
< 0.4
0.4
39
7.2
< 16
4.8
210
11
14
< 0.4
62
0.8
< 24
'< 80
< 80
200
140
170
< 8
19
560
NR
NR
See notes on Table 3-8.
-------
TABLE 3-10. TRACE SPECIES AND ORGANIC EMISSIONS, SASS ORGANIC AND LIQUIDS SECTION COLLECTION
Test 19-2, Modified Boiler, Location 19, Baseline Condition
Sample Type
Sample Hunter
Sample Neight/Vol.
Units
Antimony
Arsenic
Barium
Beryllium
Cadmium
Calcium
Chrouiom
Cobalt
Copper
Iron
Lead
Manganese
Mercury
Nickel
Selenium
Tellurium
Tin
Titanium
Vanadium
Zinc
Chloride
Fluoride
Nitrates
Sulfates
Total POM
Total FCB
XAD-2
Resin
533
150 g
pq/q
< 25
< 1
< 50
< 0.5
< 1
40 - 10
5.8
0
0
8
< 2
1.5
< 0.02
< 2
< 1
< 25
< 50
< 150
< 1.5
8.4
0
0
3.6
67
0.005
< 1
pg/m3
< 190
< 7.5
< 75
< 3.8
< 7.5
230
44
0
0
60
< 15
11
< 0.15
< 15
< 7.5
< 190
< 380
< 1100
< 75
63
0
0
27
500
0.04
< 7.5
Organic Nodule
Rinse
19-2B
460 ml
lig/ml
< 0.5
< 0.005
< 0.1
< 0.005
0.34
0.16
14
0.3
0.14
69
< 0.05
0.70
< 0.005
8.9
< 0.01
< 0.3
< 1
< 1
0.2
0.03
< 0.5
0.16
0.84
260
NES
NES
pg/n3
< 12
< 0.12
< 2.3
< 0.12
7.8
3.7
320
6.9
3.2
1600
< 1.2
16
< 0.12
200
< 0.23
< 6.9
< 23
< 23
4.6
0.7
< 12
3.7
19
6000
NES
NES
Condensate
19-2C
4702 ml
pg/ml
< 0.5
< 0.005
< 0.1
< 0.005
0.018
0.028
1.4
< 0.2
0.036
4.3
< 0.04
0.074
< 0.005
0.7
< 0.01
< 0.3
< 1
< 1
< 0.1
0.04
49
0.57
0.21
9800 (S02)
< 0.001
< 0.001
pg/m
< 120
< 1.2
< 24
< 1.2
4.2
6.6
330
< 47
8.5
1000
< 9.4
17
< 1.2
160
< 2.4
< 71
< 230
< 230
< 23
9.4
L2000
130
49
2.3E6(S02>
< 0.24
< 0.24
Impinqer Ho. 1
Combined With
Condensate
Hg/ml
gg/m3
Impinqer No. 2
Contained with
Condensate
Vq/ml
pq/m3
ImDinaer No. 3
Combined With
Condensate
pq/ml
pq/m3
See notes on Table 3-8.
-------
TARbE 3-11. TRACE SPECIES AND ORGANIC EMISSIONS, PROCESS SAMPLES AND MASS BALANCES
Test 19-2, Modified Boiler, Location 19, Baseline Condition
Sample Tyce
Sample Number
Sacrole Keight/Vol.
Units
Antimony
Arsenic
Barium
Beryllium
Cadndura
Calcium
Chromium
Cobalt
Copper
Iron
Lead
Manganese
Mercury
Nickel
Selenium
Tellurium
Tin
Titanium
Variadiuaj
Zinc
Chloride
Fluoride
Nitrates
Sulfates
Total PCM
Total PCB
Emission
in Partic.
< 3 urn
720,533
0.6353 g
vg/n
< 13
4
83
< 0.26
< 0.26
140
8.4
43
11
790
21
7.5
< 0.01
720
< 0.53
10 < 13
< 26
27 < 62
2300
110
240
10
0.8
9500
NES
NES
Total
Enission
Conce.-i.
SASS
20 n3
Ua/m
< 380
6.5 < 15
95 < 210
< 6
13
650
750
65 < 130
32
4300
45 < 70
70
< 1.9
1300 <1400
< 12
< 300
< 750
70 < 1600
3200 <3400
370
12000
170 < iao
130
18000
NES
NES
Total
Emission
Rate
1.6 m3/s
Ug/s
< 610
10 < 24
150 < 340
< 9.6
21
1000
1200
100 « 210
51
6900
70 < 110
110
< 3.0
2100 <2200
< 19
< 480
< 1200
110 < 2600
5200 <5400
590
19000
270 < 290
210
29000
!JES
NES
Input
No. 6 Fuel Oil
19-2LF
129 g/s
IJg/g
< 25
< 2
< 5
< 0.3
< 0.3
31
< 5
< 10
< 3
12
< 3
1.4
< 0.1
14
< 1
< 25
< 25
< 250
40
3
< 11.6
45.8
NR
NR
NR
NR
pg/s
< 3200
< 260
< 640
< 39
< 39
4000
< 640
< 1300
< 390
1500
< 390
180
< 13
1800
< 130
< 3200
< 3200
<32000
5200
390
< 1500
5900
—
—
—
Mass
Balance
Emission
Input
< DL
> 0.04
> 0.2
< DL
> 0.5
0.25
> 1.9
> 0.07
> 0.13
4.6
> 0.18
0.61
< DL
1.2
< DL
< DL
< DL
> 0.003
1.0
1.5
> 12
0.05
—
—
._ _
—
See notes on Table 3-8.
-------
TABLE 3-12. TRACE SPECIES EMISSIONS BY SPARK SOURCE MASS SPECTROMETRY
Test 19-2, Modified Boiler, Location 19, Baseline Condition
Sample Type
Sample Number
Sample Weiqht/Vol.
Units
Antimony
Arsenic
Barium
Beryllium
Cadmium
Calcium
Chromium
Cobalt
Copper
Iron
Lead
Manganese
Mercury
Nickel
Selenium
Tellurium
Tin
Titanium
Vanadium
Zinc
Chlorine
Fluorine
Combined
Solids
19-2D
2.7416 g
ug/g
26
18
MC
0.4
0.6
< MC
96
MC
280
MC
380
140
NR
MC
10
0.3
3
330
MC
MC
32
59
ug/m3
3.6
2.5
—
0.055
0.08
—
13
—
38
—
52
19
~
--
1.4
0.04
0.4
45
--
—
4.3
8.1
XAD-2 Resin
533
150 g
ug/g
1
0.4
=B
< 0.4
< 0.7
30
< B
< 0.1
=B
4
< 2
0.1
NR
< B
< 0.4
< 0.4
< 1
1.1
0.1
< B
< B
2
pg/m
7.5
3
0
< 3
< 5
220
0
< 0.8
0
30
< 15
0.8
—
0
< 3
< 3
< 8
a
0.8
0
0
15
Combined
Liquids
19-2F
5162 ml
pg/ml
< 0.001
0.004
=B
< 0.001
0.03
7
3.7
0.027
0.043
5
=B
0.19
NR
0.68
0.003
< 0.001
< B '
< B
0.027
< B
< B
< B
pg/m3
0
1.0
0
< 0.26
7.5
1800
950
7
11
1300
0
49
—
180
0.75
< 0.26
0
0
7
0
0
0
Total
Emission
Concen .
SASS
20 m3
pg/m
11
6.5
MC
0.055 <3
7.5 < 13
2000 < MC
960
8 < MC
49
1300 < MC
52 < 67
69
NR
180 < MC
2 < 5
0.04 < 3
0.4 < 8
53
8 < MC
MC
4.3
23
Total
Emission
Rate
1.6 m3/s
ug/s
18
10
MC
0.09 < 5
12 < 21
3200 < MC
1500
13 < MC
78
2000 < MC
83 < 110
110
—
290 < MC
3 < 8
0.06 < 5
0.6 < 13
85
13 < MC
MC
6.9
37
Input
No. 6 Fuel Oil
19-2U
129 g/s
yg/g
< 0.15
0.3
5
< 0.15
< 0.15
200
0.4
2.5
2
41
0.8
0.3
NR
51
< 0.75
< 0.15
< 0.45
0.9
340
3.5
2
4
Wg/s
< 20
39
65
< 20
< 20
26000
52
320
260
5100
100
39
—
6600
< 97
< 20
< 58
120
44000
450
260
520
S5MS
Mass
Balance
Emission
Input
> 0.90
0'.26
MC
> 0.005
> 0.60
0.12 < MC
29
0.04 < MC
0.30
0.39 < MC
0.83 < 1.10
2.32
—
0.04 < MC
> 0.03
> 0.003
> 0.01
0.71
MC
MC
0.027
0.071
Best
Balance
AA & SS
AA or SS
Emission
Input
—
0.26
2.30
—
> 1.05
0.80
23.00
0.31
0.30
1.35
0.83
0.61
—
1.20
—
~
—
0.92
1.00
1.30
—
0.52
cr>
en
See notes on Table 3-8.
-------
TABLE 3-12. TRACE SPECIES EMISSIONS BY SPARK SOURCE MASS SPECTROMETRY (Continued)
Test 19-2, Modified Boiler, Location 19, Baseline Condition
Sample Type
Sample Number
Sample Weight/Vol.
Units
Aluminum
Bismuth
Boron
Bromine
Cerium
Cesium
Dysprosium
Erbium
Europium
Gadolinium
Gallium
Germanium
Gold
Hafnium
Holmium
Iodine
Iridium
Lanthanum
Lithium
Lutetium
Magnesium
Molybdenum
Neodymiura
Niobium
Osmium
Combined
Solids
19-2D
2.7416 g
ug/g
MC
0.3
18
0.7
0.1
22
1
0.5
0.3
1
12
0.8
< 0.1
0.5
0.7
0.4
< 0.1
190
4
0.1
MC
MC
19
0.4
< 0.1
ug/m
MC
0.04
2.5
0.1
0.01
3
0.14
0.07
0.04
0.14
1.5
0.1
< 0.014
0.07
0.1
0.05
< 0.014
26
0.5
0.01
MC
MC
2.6
0.05
< 0.014
XAD-2 Resin
533
150 g
ug/g
3
< 0.4
•=B
1
< 0.6
< 0.4
< 0.4
< 0.4
< 0.4
< 0.4
< 0.3
< 0.4
< 0.4
< 0.4
< 0.4
0.5
< 0.4
< 0.4
< 0.4
< 0.4
=B
2
< 0.4
< 0.4
< 0.4
vg/m
23
< 3
0
8
< 5
< 3
< 3
< 3
< 3
< 3
< 2
< 3
< 3
< 3
< 3
4
< 3
< 3
< 3
< 3
0
15
< 3
< 3
< 3
Combined
Liauids
19-2F
5162 ml
pg/ml
< B
< 0.001
< B
0.023
< B
0.001
< 0.001
< 0.001
< 0.001
< 0.001
< 0.001
< 0.001
< o.ooi
< 0.001
< 0.001
0.0025
< 0.001
< B
< B
< 0.001
< B
0.13
< 0.001
< o.ooi
< 0.001
ug/m
0
0
0
6
0
0.25
< 0.3
< 0.3
< 0.3
< 0.3
< 0.3
< 0.3
< 0 3
< 0.3
< 0.3
0.65
< 0.3
0
0
< 0.3
0
33
< 0.3
< 0.3
< 0.3
Total
Emission
Concen.
SASS
20 m3
pg/m
23 < MC
0.04 < 3
2.5
14
0.014 <4.5
3.3
0.14 < 3
0.07 < 3
0.04 < 3
0.14 < 3
1.7 < 4
0.1 < 3
< 3
0.07 < 3
0.1 < 3
4.7
< 3
26
0.5 < 3
0.01 < 3
MC
48 < MC
3 < 6
0.05 < 3
< 3
Total
Emission
Rate
1.6 m3/s
Ug/s
37 < MC
0.06 < 5
4
22
0.02 < 7
5.3
0.22 < 5
< 5
< 5
< 5
2.7 < 6
0.2 < 5
< 5
0.1 < 5
0.2 < 5
7.5
< 5
42
< 5
< 5
MC
77 < MC
5 < 10
0.08 < 5
< 5
Input
No. 6 Fuel Oil
19-2LF
129 a/s
vig/g
19
< 0.1
0.6
< 0.1
< 0.6
< 0.1
< 0.1
< 0.1
< 0.1
< 0.1
< 0.1
< 0.1
< 0.1
< 0.1
< 0.1
0.5
< 0.1
< 0.1
< 0.1
< 0.1
24.0
6.0
< 0.1
< 0.1
< 0.1
ug/s
2500
< 13
77
< 13
< 77
< 13
< 13
< 13
< 13
< 13
< 13
< 13
< 13
< 13
< 13
64
< 13
< 13
< 13
< 13
3000
773
< 13
< 13
< 13
SSMS
Mass
Balance
Emission
Input
—
> 0.005
0.05
> 1.70
> 0.0003
> 0.40
> 0.02
< DL
< DL
< DL
> 0.20
> 0.02
< DL
> 0.01
> 0.02
0.12
< DL
> 3.2
< DL
< DL
—
> 0.10
> 0.38
> 0.006
< DL
See note on Table 3-8.
-------
TABLE 3-12. TRACE SPECIES EMISSIONS BY SPARK SOURCE MASS SPECTROMETRY (Continued)
Test 19-2, Modified Boiler, Location 19, Baseline Condition
Sample Type
Sample Number
Sample Weight/Vol.
Units
Palladium
Platinum
Phosphorus
Potassium
Praseodymium
Rhenium
Rhodium
Rubidium
Ruthenium
Samarium
Scandium
Silicon
Silver
Sodium
Sulfur
Strontium
Tantalum
Thallium
Terbium
Thorium
Thulium
Tungsten
Uranium
Ytterbium
Yttrium
Zirconium
Combined
Solids
19-2D
2.7416 g
yg/g
< 0.1
< 0.1
MC
MC
11
< 0.1
< 0.1
1
< 0.1
2
0.2
MC
7
MC
MC
110
< 0.1
< 0.1
0.3
1
0.1
O.B
< 0.7
0.7
1
5
Ug/m
< 0.014
< 0.014
MC •
MC
1.5
< 0.014
< 0.014
0.15
< 0.014
0.3
0.03
MC
1.0
MC
MC
15
< 0.014
< 0.014
0.04
0.15
0.02
0.1
< 0.1
0.1
0.15
0.7
XAD-2 Resin
533
150 g
yg/g
< 0.4
< 0.4
2
2
< 0.4
< 0.4
< 0.4
•0
< 0.4
< 0.4
< 0.1
15
< 0.8
5
93
=B
< 0.4
< 0.4
< 0.4
< 0.4
< 0.4
< 0.4
< 0.4
< 0.4
< 0.4
1
Ug/m3
< 3
< 3
15
15
< 3
< 3
< 3
0
< 3
< 3
< 0.8
110
< 6
38
690
0
< 3
< 3
< 3
< 3
< 3
< 3
< 3
< 3
< 3
8
Combined
Liquids
19-2F
5162 ml
ug/ml
< 0.001
< 0.001
=B
1.9
< O.OO1
< 0.001
< 0.001
< B,DL
< 0.001
< 0.001
< 0.001
0.58
MC
> 0.12
MC
.< B
< 0.001
< 0.001
< 0.001*
< 0.006
< 0.001
< 0.001
< 0.004
< 0.001
< 0.001
•B
yg/m
< 0.3
< 0.3
0
480
< 0.3
< 0.3
< 0.3
0
< 0.3
< 0.3
< 0.3
150
-
> 30
MC
0
< 0.3
< 0.3
< 0.3
< 1.6
< 0.3
< 0.3
< 1.0
< 0.3
< 0.3
0
Total
Emission
Conccn.
SASS
20 m3
yg/m3
< 3
< 3
15 < MC
495 < MC
1.5 < 3
< 3
< 3
0.15
< 3
0.3 < 3
0.03 < 1
260 < MC
1 < MC
38 < MC
690 < MC
15
< 3
< 3
0.04 < 3
0.15 < 4
0.02 < 3
0.1 < 3
< 4
0.1 < 3
0.2 < 3
9
Total
Emission
Rate
1.6 m3/g
ug/s
< S
< 5
24 < MC
800 < MC
2.4 < 5
< 5
< 5
0.24
< 5
0.5 < 5
0.05 < 2
420 < MC
2 < MC
60 < MC
1100 < MC
24
< 5
< 5
0.06 < 5
0.24 < 6
0.03 < 5
0.2 < 5
< 6
0.2 < 5
0.3 < 5
14
Input
No. 6 Fuel Oil
19-2LF
129 g/s
yg/g
< 0.1
< 0.1
29
23
< 0.1
< 0.1
< 0.1
< 0.1
< 0.1
< 0.1
< 0.1
23
0.1
210
MC
1
< 0.1
< 0.1
< 0.1
< 0.1
< 0.1
< 0.1
< 0.1
< 0.1
< 0.1
0.9
ug/s
< 13
< 13
3700
3000
< 13
< 13
< 13
< 13
< 13
< 13
< 13
3000
< 13
27000
MC
130
< 13
< 13
< 13
< 13
< 13
< 13
< 13
< 13
< 13
120
SSMS
Mass
Balance
Emission
Input
< DL
< DL
>. 0.004
> 0.17
> 0.18
< DL
< DL
> 0.02
< DL
> 0.03
> 0.003
> 0.14
> 0.15
> 0.002
—
0.18
< DL
< DL
> 0.005
> 0.02
> 0.002
> 0.02
< DL
> 0.02
> 0.02
0.12
OV
03
See note on Table 3-8.
-------
TABLE 3-13. TRACE SPECIES AND ORGANIC EMISSIONS, SASS SOLIDS SECTION COLLECTION
Test 19-3, Modified Boiler, Location 19, Low NOV Condition
J\.
Sample Tvpe
Sample Number
Sample Weight/Vol.
Units
Antimony
Arsenic
Barium
Beryllium
Cadmium
Calcium
Chromium
Cobalt
Copper
Iron
Lead
Manganese
Mercury
Nickel
Selenium
Tellurium
Tin
Titanium
Vanadium
Zinc
Chloride
Fluoride
Nitrates
Sul fates
Total POM
Total PCB
Nozzle, Probe,
10 ym Cyclone
Solids
722
2.4163 g
yg/g
< 50
4
480
< 1
< 1
1900
69
104
43
5200
NES
49
0.5
970
< 2
< SO
< 100
600
5400
281
< 30
54
43
8910
< 1
< 1
yg/m
< 6
0.5
57
< 0.12
< 0.12
230
8
12
5
620
—
6
0.06
120
0.24
6
12
72
650
34
< 4
6.5
5.1
1100
< 0.1
< 0.1
3 urn Cyclone
Solids
723
1.0724 g
ug/g
2500
< 100
1000
< 50
< 50
3500
950
450
< 50
4400
NES
200
< 2
1700
< 5000
< 2500
< 5000
<15000
7000
250
< 97
NES
67
14000
NES
NES
ug/m
< 130
< 5
< 53
< 3
< 3
190
50
24
< 3
230
—
11
< 0.1
90
< 270
< 130
< 270
< 800
370
13
< 5
—
3.6
740
—
1 yn Cyclone
Solids
726
0.2120 g
ug/g
< 830
< 33
< 330
< 17
< 17
3000
300
300
< 150
6500
NES
117
< 0.67
3000
< 33
< 830
<1700
5000
14000
350
NES
NES
HES
NES
NES
NES
y«/m
< 9
< 0.4
< 4
< 0.2
< 0.2
31
3.1
3.1
< 2
68
—
1.2
< 0.007
31
< 0.4
< 9
< 18
52
150
3.7
—
—
—
—
—
"
Filters
539
0.9974 g
ug/g
< 170
33
730
< 3.3
< 3.3
3100
180
807
200
20000
NES
160
< 0.13
13000
< 6.7
< 170
< 330
<1000
43000
3000
1700
< 1
46
170000
NES
NES
yg/m
< 8
1.6
36
< 0.2
< 0.2
150
8.9
40
9.9
1000
—
7.9
< 0.006
640
< 0.3
< 8
< 16
< 49
2100
150
84
< 0.05
2.3
8400
—
"
Solid
Section
Wash
19- 3A
1839 ml
'ig/ml
< 0.5
0.01
< 0.1
< 0.005
< 0.005
IS
0
< 0.2
0.11
1.8
0.11
0.16
< 0.005
0.5
0.04
< 0.3
< 1
< i
1.4
0.49
< 0.5
< 0.1
0.26
12
NR
NR
yg/m
< 46
o.v-i
< 9
< 0.5
< 0.5
1400
0
< 18
10
160
10
15
< 0.5
46
3.6
< 27
< 91
< 91
130
45
< 46
< 9
24
1100
—
"
CTi
See notes on Table 3-8.
-------
TABLE 3-14. TRACE SPECIES AND ORGANIC EMISSIONS, SASS ORGANIC AND LIQUIDS SECTION COLLECTION
Test 19-3, Modified Boiler, Location 19, Low NO,, Condition
•••pie Type
Sample Number
ilmple Weight/Vol.
Wilts
Antimony
Arsenic
Barium
Beryllium
Cadmium
Calcium
Chromium
Cobalt
Capper
Iron
lead
Manganese
Mercury
Nickel
ielenium
Tellurium
Tin
Titanium
Vanadium
line
Chloride
Fluoride
•itrates
iulfates
Total POM
Total PCS
XAD-2
Resin
537
150 q
ug/g
< 25
7
75
< 0.5
< 0.5
< B
5.3
< B
< B
11
< B
1.5
< 0.02
< 2.5
< 1
< 25
< 50
< 150
< 10
-B
< B
< B
0.2
289
0.0008
< 1
pg/m3
< 186
52
560
< 4
< 4
0
39
0
0
82
0
11
< 0.2
< 19
< 8
< 19O
< 370
<1100
< 74
0
0
0
1.5
2100
0.006
< 7
Organic Module
Rinse
19- 3B
703 ml
< 0.5
0.005
< 0.1
< 0.005
0.015
0.02
8.4
< 0.2
0.11
38
< 0.05
0.57
< 0.005
5.9
0.04
< 0.3
< 1
< 1
< 0.1
-B
1.6
< 0.1
0.88
130
1.45
< 0.001
ug/m3
< 17
0.2
< 3.5
< 0.2
0.52
0.70
290
< 7
3.8
1300
< 2
20
< 0.2
200
1.4
< 10
< 35
< 35
< 3.5
0
56
< 4
31
4500
50
< 0.03
Condensate
19- 3C
4616 ml
ug/ml
< 0.50
0.005
< 0.1
< 0.005
0.018
0
1.5
< 0.2
0.05
5.2
< 0.04
0.12
< 0.005
2.0
0.02
< 0.3
< 1.0
< 1.0
< 0.1
2.5
15
0.3
0.22
5400 (S02>
0.002
< 0.001
ug/m3
< 110
1.1
< 23
< 1
4.1
340
< 50
11
1200
< 9
27
< 1
460
4.6
< 69
< 230
< 230
< 23
570
3400
69
50
0.5
< 0.2
Impinger No. 1
Combined With
Condensate
ug/ml
yg/m3
Impinqer No. 2
Combined With
Conden
uq/ml
sate
uq/m3
Imoinaer No. 3
Combined With
Conden
Ug/ml
ate
ug/»3
•-4
O
See notes on Table 3-8.
-------
TABLE 3-15. TRACE SPECIES AND ORGANIC EMISSIONS, PROCESS SAMPLES AND MASS BALANCES
Test 19-3, Modified Boiler, Location 19, Low NO Condition
2v
Sample Type
Sample Number
Sample Weight/Vol.
Units
Antimony
Arsenic
Barium
Beryllium
Cadmium
Calcium
Chromium
Cobalt
Copper
Iron
Lead
Manganese
Mercury
Ni_kel
Selenium
Tellurium
Tin
Titanium
Vanadium
Zinc
Chloride
Fluoride
Nitrates
Sulfates
Total POM
Total PCB
Emission
in Partic.
< 3 Urn
726,539
1.2094 g
Vig/m3
< 17
1.6 < 2
36 < 40
< 0.34
< 0.34
181
12
43
9.9 < 12
1000
NES
9.1
< 0.013
670
< 0.65
< 17
< 34
52 < 100
2200
150
84
< 0.05
2.3
8400
NES
NES
Total
Emission
Concen.
SASS
20.2 m3
Ua/m
< 540
59 < 64
640 < 740
< 8.9
4.3 < 12
2000
740
79 < 150
39 < 44
4700
9.9 < 21
99
0.06 < 21
1600
9.9 < 290
< 450
< 1000
120 < 2500
3400 <3600
820
3500
64 < 79
110 < 120
18000
50 < 51
< 7
Total
Emission
Rate
1.495 m3/s
IJQ/S
< 810
88 < 96
960 < 1100
< 13
7.2 < 18
3000
1100
120 < 220
58 < 66
7000
15 < 31
150
0.09 < 3.1
2400
15 < 430
< 670
< 1500
180 < 3700
5100 <5400
1200
5200
96 < 120
160 < 180
27000
75 < 76
< 11
Input
No. 6 Fuel Oil
19-3LF
128.1 g/s
vg/g
< 25
< 2
< 5
< 0.3
< 0.3
< 10
< 5
< 10
< 3
7
< 3
< 0.5
< 0.1
< 10
< i
< 25
< 25
< 250
50
5
< 34.4
< 34.6
NR
NR
NR
NR
vg/s
< 3200
< 260
< 640
< 38
< 38
< 1300
< 640
< 1300
< 380
900
< 380
< 64
< 13
< 1300
< 130
< 3200
< 3200
< 32000
6400
640
< 4400
< 4400
—
—
—
—
Mass
Balance
Emission
Input
< DL
> 0.24
> 1.50
< DL
> 0.19
> 2.30
> 1.70
> 0.10
> 0.15
7.8
> 0.04
> 2.30
> 0.007
> 1.80
> 0.12
< DL
< DL
> 0.006
0.80
1.90
> 1.20
0.02
—
—
—
—
Soo notes on Tablo 3-8.
-------
TABLE 3-16. TRACE SPECIES EMISSIONS BY SPARK SOURCE MASS SPECTROMETRY
Test 19-3, Modified Boiler, Location 19, Low NO Condition
Sample Type
Sample Number
Sample Weight/Vol.
Unics
Antimony
Arsenic
Barium
Beryllium
Cadmium
Calcium
Chromium
Cobalt
Copper
Iron
Lead
Manganese
' Mercury
Nickel
Selenium
Tellurium
Tin
Titanium
Vanadium
Zinc
Chlorine
Fluorine
Combined
solids
19- 3D
4.6981 g
y<3/g
20
7
460
0.1
0.4
MC
33
400
70
MC
160
44
NR
MC
3
0.8
5
300
MC
570
160
46
yg/m3
4.7
1.6
110
0.02
0.09
MC
7.7
93
16
MC
37
10
NR
MC
0.7
0.2
1.2
70
MC
130
37
11
XAD-2 Resin
537
150 g
pg/g
17
< 0.7
< B
< 0.2
< 0.3
9
3
< 0.1
2
8
1
0.2
NR
" 1
< 0.4
< 0.2
0.4
1.1
0.5
6
14
22
yg/m
130
< 5
< 0
< 2
< ?
67
22
< 0.7
15
59
7
1.5
NR
7
< 3
< 2
3
8
4
45
100
160
Combined
Liquids
19- 3F
5319 ml
yg/ml
0.1
0.02
2
0.001
< 0.01
MC
1.8
-B
0.4
4
0.04
0.2
NR
0.8
0.4
< 0.006
0.03
1.9
0.02
1.9
7.7
"B
ug/m
26
5
530
0.26
< 3
MC
470
0
100
1000
10
50
NR
210
100
< 2
a
500
5
500
2000
0
Total
Emission
Concen.
SASS
20.2 m3
ug/m3
160
6 < 12
640
0.3 < 2
0.09 < 5
67 < MC
500
94
130
1100
54
64
NR
240 < MC
100 < 110
0.2 < 3
12
580
9 < MC
680
2100
170
Total
Emission
Rate
1.5 m3/s
yg/s
240
9 < 18
960
0.5 < 3
0.1 < 3
100 < MC
750
140
190
1600
81
96
NR
360 < MC
150 < 160
0.3 < 5
18
870
13 < MC
1000
3200
260
Input
No. 6 Fuel Oil
19-3LF
128 g/s
U9/9
< 0.2
0.4
4
< 0.2
< 0.2
43
2
2
1.5
14
< 0.9
0.45
NR
23
0.9
< 0.2
< 0.2
1
150
2.5
1.5
3
pq/s
< 30
51
510
< 30
< 30
5500
260
260
190
1800
< 110
58
NR
2900
110
< 30
< 30
130
19000
320
190
380
SSMS
Mass
Balance
Emission
Input
> 9.00
0.18 < 0.3!
1.90
> 0.02
> 0.004
MC
2.90
0.54
1.00
0.89
> 0.74
1.66
NR
MC
1.40
> 0.01
> 0.70
6.70
MC
3.XO
17.00
0.70
Best
Balance
AA S SS
AA or SS
Emission
Input
—
1.7
1.90
—
—
0.55
2.90
0.54
1.00
0.89
—
1.66
—
0.83
1.40
—
—
1.40
0.80
1.60
17.00
0.70
NJ
See notes on Table 3-8.
-------
TABLE 3-16. TRACE SPECIES EMISSIONS BY SPARK SOURCE MASS SPECTROMETRY (Continued)
Test 19-3, Modified Boiler, Location 19, Low NO Condition
X
Samole Type
Sample Number
Sample Weight/Vol.
Units
Aluminum
Bismuth
Boron
Bromine
Cerium
Cesium
Dysprosium
Erbium
Europium
Gadolinium
Gallium
Germanium
Gold
Hafnium
Holmium
Iodine
Iridium
Lanthanum
Lithium
Lutetium
Magnesium
Molybdenum
Neodymium
Niobium
Osmiusi
Combined
Solids
19-3D
4.6981 g
ug/g
> 130
< 0.1
1
2
19
0.8
0.8
0.2
0.3
0.7
11
0.6
< 0.1
< 0.1
0.4
1
< 0.1
60
2
< 0.1
MC
970
12
0.7
< 0.1
yg/m
> 30
< 0.02
0.23
o. ;e
4.4
0.19
0.19
0.05
0.07
0.16
2.6
0.14
< 0.02
< 0.02
0.09
0.23
< 0.02
14
0.46
< 0.02
MC
230
2.8
0.16
< 0.02
XAD-2 Resin
537
150 g
yg/g
=B
< 0.2
=B
=B
0.2
< 0.1
< 0.2
< 0.2
< 0.2
< 0.2
< 0.1
< 0.2
1
< 0.2
< 0.2
0.3
< 0.2
0.4
< 0.1
< 0.2
10
9
< 0.2
< 0.2
< 0.2
yg/m
0
< 2
0
0
1.5
< 0.7
< 2
< 2
< 2
< 2
< 0.7
< 1.5
7.4
< 2
< 2
2.2
< 2
3
< 0.7
< 2
74
67
< 2
< 2
< 2
Combined
Liquids
19-3F
5319 ml
Ug/ml
> 7
< 0.006
< B
0.28
0.007
0.03
< O.OOG
< 0.006
< 0.006
< 0.006
< O.OP6
< 0.006
< 0.02
< 0.006
< 0.006
0.02
< 0.006
0.024
0.034
< 0.006
MC
0.24
< 0.006
< 0.02
< 0.006
pCT/m
> 1800
< 2
0
74
1.8
8
< 2
< 2
< 2
< 2
< 2
< 2
< j
< 2
< 2
5
< 2
6
9
< 2
HC
63
< 2
5
< 2
Total
Emission
Concen.
SASS
20.2 m3
Ug/m
> 1800
< 4
0.23
74
7.7
8.2 < 9
0.2 < 4
0.05 < 4
0.07 < 4
0.16 < 4
2.6 < 5
0.14 < 4
7.4 < 9
< 4
0.09 < 4
7.4
< 4
23
9.5 < 10
< 4
74 < HC
360
2.8 < 6
5.2 < 7
< 4
Total
Emission
Rate
1.50 m3/s
Va/s
> 2700
< 6
0.35
110
12
12 < 14
0.3 < 6
0.07 < 6
0.11 < 6
0.24 < 6
3.9 < 8
0.21 < 6
11 < 14
< 6
0.13 < 6
11
< 6
35
14 < 15
< 6
110 < HC
540
< 9
< 11
< 6
Input
No. 6 Fuel Oil
19-3LF
128 a/s
ug/g
2
< 0.2
0.7
< 0.5
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.1
< 0.2
27
8
< 0.2
< 0.4
< 0.2
ug/s
260
< 30
90
< 60
< 30
< 30
< 30
< 30
< 30
< 30
< 30
< 30
< 30
< 30
< 30
< 30
< 30
< 30
< 13
< 30
3400
1000
< 30 i
< 50
< 30
SSMS
Mass
Balance
Emission
Input
> 10.00
< DL
0.004
> 1.80
> 0.40
> 0.40
> 0.01
> 0.002
> 0.004
> 0.01
> 0.13
> 0.01
> 0.40
< DL
> 0.004
> 0.40
< DL
> 1.20
> 1.08
< DL
> 0.03
0.54
< DL
< DL
< DL
u>
See note on Table 3-8.
-------
TABLE 3-16. TRACE SPECIES EMISSIONS BY SPARK SOURCE MASS SPECTROMETRY (Continued)
Test 19-3, Modified Boiler, Location 19, Low NOX Condition
Sample Type .
Sample Number
Sample Weight/Vol.
Units
Palladium
Platinum
Phospnorus
Potassium
Praseodymium
Rhenium
Rhodium
Rubidium
Ruthenium
Samarium
Scandium
Silicon
Silver
Sodium
Sulfur
Strontium
Tantalum
Thallium
Terbium
Thorium
Thulium
Tungsten
Uranium
Ytterbium
Yttrium
Zirconium
Combined
Solids
19-3D
4.6981 g
pg/g
< 0.1
< 0.1
MC
MC
7
< 0.1
< 0.1
0.8
< 0.1
2
0.2
MC
5
> 330
> 800
42
< 0.1
< 0.1
0.2
< 1
< 0.1
< 0.7
2
< 0.1
2
7
pg/m3
< 0.02
< 0.02
MC
MC
2
< 0.02
< 0.02
0.2
< 0.02
0.5
0.05
MC
1.2
> 77
>190
9.8
< 0.02
< 0.02
0.05
< 2
< 0.02
< 0.2
0.5
< 0.02
0.5
2
XAD-2 Resin
537
150 g
pg/g
< 0.2
< 0.2
11
-B
< 0.2
< 0.2
< 0.2
< B
< 0.2
< 0.2
< 0.1
15
0.7
39
21
< B
< 0.2
< 0.2
< 0.2
< 1
< 0.2
< 0.2
< 1
< 0.2
< 0.2
< B
ug/m3
< 2
< 2
82
0
< 2
< 2
< 2
0
< 2
< 2
0.7
110
5
290
160
0
< 2
< 2
< 2
< 7
< 2
< 2
< 7
< 2
< 2
0
Combined
Liquids
19-3F
5319 ml
pg/ml
< 0.006
< 0.006
MC
MC
< 0.006
< 0.006
< 0.006
0.002
< 0.006
< 0.006
< 0.009
2.6
MC
MC
MC
1
< 0.006
<• 0.006
< 0.006
< 0.06
< 0.006
< 0.006
0.1
< 0.006
< 0.008
0.02
pg/ra
< 2
< 2
MC
MC
< 2
< 2
< 2
0.5
< 2
< 2
2.4
690
MC
MC
MC
260
< 2
< 2
< 2
< 20
< 2
< 2
26
< 2
< 2
6
Total
Emission
Concen .
SASS
20.2 m3
Ug/m
< 4
< 4
82 < MC
MC
2 < 6
< 4
< 4
0.7
< 4
0.5 < 5
3
800 < MC
6 < MC
360 < MC
350 < MC
270
< 4
< 4
0.05 < 4
< 29
< 4
< 4
27 < 33
< 4
0.5 < 4
8
Total
Emission
Rate
1.50 m3/s
pg/s
< 6
< 6
120 < MC
MC
3 < 8
< 6
< 6
4
< 6
0.8 < 8
5
1200 < MC
9 < MC
550 < MC
530 < MC
410
< 6
< 6
0.8 < 6
< 44
< 6
< 6
41 < 50
< 6
0.8 < 6
12
Input
No. 6 Fuel Oil
19-3LF
128 g
pg/g
< 0.2
< 0.2
11
8
< 0.2
< 0.2
< 0.2
< 0.1
< 0.2
< 0.2
< 0.2
19
2
72
MC
1
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
0.6
s
pg/s
< 30
< 30
1400
1000
< 30
< 30
< 30
< 13
< 30
< 30
< 30
2400
260
9200
MC
130
< 30
< 30
< 30
< 30
< 30
< 30
< 30
< 30
< 30
77
SSMS
Mass
Balance
Emission
Input
< DL
< DL
> 0.09
—
> 0.10
< DL
< DL
> 0.31
< DL
—
> 0.17
> 0.50
> 0.03
> 0.06
—
3.10
< DL
< DL
> 0.03
< DL
< DL
< DL
> 1.40
< DL
> 0.03
0.15
See note on Table 3-8.
-------
Ul
TABLE 3-17. TRACE SPECIES AND ORGANIC EMISSIONS, SASS SOLIDS SECTION COLLECTION
Test 19-4, Modified Boiler, Location 19, Low NOX Condition
Sample Type
Sample Number
Sample Weight/Vol.
Units
Antimony
Arsenic
Barium
Beryllium
Cadmium
Calcium
Chromium
Cobalt
Copper
Iron
Lead
Manganese
Mercury
Nickel
Selenium
Tellurium
Tin
Titanium
Vanadium
Zinc
Chloride
Fluoride
Nitrates
Sulfates
Total POM
Total PCB
Nozzle , Probe ,
10 (Jm Cyclone
Solids
728
1.5802 g
Jiq/9
< 50
20
140
< 1
< 1
2000
37
95
28
2700
NES
30
< 0.04
1000
< 2
< 50
< 100
< 300
< 20
3
< 28
NES
49.5
7890
NES
NES
yg/m
< 4
1.6
'11
< 0.08
< 0.08
160
2.9
7.5
2.2
210
—
2.4
< 0.003
79
< 0.2
< 4
< 8
< 24
< 2
0.24
< 2
—
3.9
620
—
3 \x> Cyclone
Solids
732
0.9878 g
yg/g
< 100
20
2400
< 2
< 2
840
56
110
30
2440
NES
34
< 0.08
1320
< 4
< 100
< 200
< 600
5800
174
< 53
274
35
12700
NES
NES
yg/n
< 5
1
120
< 0.1
< 0.1
42
2.3
5.4
1.5
120
—
1.7
< 0.004
65
< 0.2
< 5
< 10
< 30
290
8.6
< 3
14
1.7
630
—
1 pm Cyclone
Solids
734
0.3263 g
vg/g
< 125
50
625
< 2.5
< 2.5
4700
115
270
33
8600
NES
73
< 0.1
8800
< 5
< 125
< 250
< 750
14000
300
NES
323
NES
NES
NES
NES
ug/n
< 2
0.8
10
< 0.04
< 0.04
77
1.9
4.4
0.54
140
—
1.2
< 0.002
140
< 0.08
< 2
< 4
< 12
230
4.9
--
5.3
—
—
--
"
Filters
540
0.9503 g
ug/g
< 100
46
650
< 2
< 2
<60000
110
< 24
156
11600
NES
114
< 0.08
12600
< 4
< 100
< 250
420
36800
2170
446
< 1
< 20
19900
NES
NES
yg/m
< 5
2.2
31
< 0.1
< 0.1
<3000
5.2
< 1
7.4
550
—
5.4
< 0.004
600
< 0.2
< 5
< 12
20
1800
100
21
< 0.05
< 1
950
—
"
Solid
Section
Wash
19-4A
1763 ml
yg/ml
< 0.5
0.005
< 0.1
< 0.005
< 0.005
0.17
0.05
< 0.2
0.05
1.1
< 0.05
0.16
< 0.005
1.1
< 0.01
< 0.3
< 1
< 1
1.1
2.6
< 0.5
< 0.1
0.12
6.0
NR
NR
lia/is
< 44
0.4
< 9
< 0.4
< 0.4
15
4.4
< 20
4.4
97
< 4.4
14
< 0.4
97
< 0.9
< 30
< 90
< 90
97
230
< 44
< 9
11
530
—
"
See notes on Table 3-3.
-------
TABLE 3-18. TRACE SPECIES AND ORGANIC EMISSIONS, SASS ORGANIC AND LIQUIDS SECTION COLLECTION
Test 19-4, Modified Boiler, Location 19, Low N0v Condition
Sample Type
Sample Number
Simple Weight/Vol.
Units
Antimony
Arsenic
Barium
Beryllium
Cadmium
Calcium
Chromium
Cobalt
Copper
Iron
Lead
Manganese
Mercury
Nickel
Selenium
Tellurium
Tin
Titanium
Vanadium
Zinc
Chloride
Fluoride
Nitrates
Sul fates
Total POM
Total PCB
XAD-2
Resin
535
150 g
vg/g
< 22
6.2
85
< 0.45
< 0.45
< B
3.5
«B
6
9
NES
0.8
< 0.02
< 2.2
< 0.89
< 22
< 45
< 130
< 0.89
3.8
4.5
«B
1.87
449
< 0.1
< 1
pg/o3
< 170
47
640
< 3.4
< 3.4
0
26
0
50
70
~
6
< 0.2
< 17
< 7
< 170
< 340
< 980
< 6.7
29
34
0
14
3400
< 0.8
< 8
Organic Module
Rinse
19-4B
535 ml
pa/ml
< 0.5
< 0.005
< 0.1
' < 0.005
0.04
=B
8.3
< 0.2
0.54
46
< 0.05
0.62
< 0.005
20
< 0.01
< 0.3
< 1
< 1
< 0.1
27
< 0.5
0.17
0.72
210
NES
NES
US/m3
< 13
< 0.1
< 3
< 0.1
1
0
220
< 5
14
1200
< 1
17
< 0.1
540
< 0.3
< 8
< 30
< 30
< 3
720
< 13
5
19
5600
--
——
Condensate
19-4C
4460 ml
pg/ml
< 0.5
< 0.005
< 0.1
< 0.005
< 0.003
0.65
1.2
< 0.2
0.08
3.1
< 0.04
0.08
< 0.005
3.2
< 0.010
< 0.3
< 1
< 1
< 0.1
9.7
27
=B
0.15
9000 (SOj)
NR
NR
pg/m
< 110
< 1
< 23
< 1
< 0.8
150
270
< 45
17
700
< 9
17
< 1
700
< 2
< 70
< 225
< 225
< 22
2200
6000
0
33
2xl06(S02>
~
—
lapinoer No. 1
Combined With
Condensate
pg/ml
pg/m
Inpinqer No. 2
Combined With
Condensate
ug/ml
vg/m
Inroinqer No. 3
Combined With
Condensate
pg/»i
pg/m
See notes on Table 3-8.
-------
TABLE 3-19. TRACE SPECIES AND ORGANIC EMISSIONS, PROCESS SAMPLES AND MASS BALANCES
Test 19-4, Modified Boiler, Location 19, Low NO Condition
Sample Type
Sample Number
Sample Weight/Vol.
Units
Antimony
Arsenic
Barium
Beryllium
Cadmium
Calcium
Chromium
Cobalt
Copper
Iron
Lead
Manganese
Mercury
Nickel
Selenium
Tellurium
Tin
Titanium
Vanadium
Zinc
Chloride
Fluoride
Nitrates
Sul fates
Total POM
Total PCS
Emission
in Partic.
< 3 urn
734,540
1.2766 g
vig/m3
< G
3
41
< 0.14
< 0.14
2900
7.1
4.4 < 6
8
690
NES
6.6
<; 0.006
740
< 0.3
< 7
< 16
20 < 32
2000
100
21
5
< 1
950
NES
NES
Total
Emission
Concen.
SASS
20.0 m3
yg/n>
< 350
55
SOO < b50
< 6
1.1 < 6
440 < 460
530
13 < 85
95
3100
< 15
65
< 2
2200
< 11
< 290
< 700
20 < 1500
2400
3300
6000
24 < 33
85
21000
NES
NES
Total
Emission
Rate
1.45 m3/s
yq/s
< 510
80
1200
< 9
1.6 < 9
640 < S70
770
26 < 120
140
4500
< 22
94
< 3
3200
< 16
< 420
< 1000
29 < 2200
3500
4785
8700
35
123
30000
NES
WES
AA Analysis
Input
No. 6 Fuel Oil
19-4LF
126 g/s
vg/g
< 25
< 2
15
< 0.3
< 0.3
< 10
< 5
< 10
< 3
30
6
< 0.5
< 0.1
< 10
< 1
< 25
< 25
< 250
55
< 5
< 46
41
NR
NR
NR
NR
ug/s
< 3200
< 250
1900
< 38
< 38
< 1300
< 630
< 1300
< 380
3800
750
< 63
< 13
< 1300
< 1 ? J
< 3100
< 3100
<31000
6900
< 630
< 58DO
5200
—
—
—
—
AA
Mass
Balance
Emission
Input
< DL
> 0.30
0.63
< DL
> 0.04
> 0.50
> 1.20
> 0.02
> 0.40
1.30
< 0.03
> 1.50
< DL
> 2.50
< DL
< DL
< DL
> 0.001
0.54
> 8.00
> 1.50
0.01
—
—
—
—
SSMS Analy.
Test 19-3
No. 6 Fuel Oil
19-3LF
126 g/s
ug/s
< 30
51
510
< 30
< 30
5500
260
260
190
1800
< 110
58
NR
2900
110
< 30
< 30
130
19000
320
190
380
—
—
—
—
Mass
Balance
AA Emission
SS Input
< DL
1.60
2.30
< DL
> 0.05
0.12
3.00
0.10
0.74
2.50
< DL
1.62
—
1.10
< 0.15
< DL
< DL
> 0.22
O.IB
15.00
46.00
0.&9
--
--
—
—
See notes on Table 3-8.
-------
However, SSMS analysis gave positive results for all these elements except
mercury which was not reported. Mercury was detected by AA analysis in only
one sample, the 10 Urn cyclone sample on Test 19-3 (Table 3-13).
Analysis of fuel samples by AA produced many results that were below
detection limits so that mass balances were not obtainable based on AA
results alone. However, SSMS detection limits were lower than for AA and
mass balances for most of the elements were obtained by combined use of
both AA and SSMS results.
Duplicate analyses were performed on four SASS samples and all
fuel samples. Results that were above detection limits were evaluated
statistically by a paired t statistic test. This test indicated no statis-
tically significant difference between the duplicate analyses. Out of a total
of 85 pairs of concentration values in excess of detection limits only 15
duplicate result pairs differed by more than a factor of 2.
Table 3-20 compares the total trace species and organics concentra-
tions as measured in the three tests conducted. Results by Atomic Absorption
and Spark Source Mass Spectrometry are shown separately and the composite
mass balances are shown. Arsenic, barium, and zinc emissions appear to be
significantly higher for the two low NO tests by AA analysis as compared
X
with the baseline test by AA. Cadmium, lead, chloride and fluoride emissions
appear to be significantly lower for the low NO conditions. For the other
Ji
elements emissions for all three tests are comparable.
The above conclusions for AA results are not completely consistent
with SSMS results. Arsenic emissions by SSMS were nearly the same for base-
line and low NO condition and are similar to the baseline emissions by AA.
X
Barium and zinc were too high to be detected by SSMS for the baseline test.
Cadmium emissions were lower for the low NO condition by SSMS and results
X
are similar to the AA results. Lead emissions by SSMS were nearly the same
at both conditions and comparable to baseline emissions by AA. Chloride
emissions were higher for the low NO conditions by SSMS and for both tests,
X
chloride emission was lower by SSMS than by AA. In contrast with the AA
results, selenium emissions by SSMS were higher at the low NO condition
A
compared to baseline and significantly higher than AA results.
78
-------
TABLE 3-20.
SUMMARY OF TOTAL TRACE SPECIES AND ORGANICS EMISSIONS FOR THE MODIFIED BOILER
AT LOCATION 19 FIRING #6 FUEL OIL
Total Emission Concentre
Atomic Absorotion .
Test
Condition
Antimony
Arsenic
Barium
Beryllium
Cadmium
Calcium
CHronium
Cobalt
Copper
Iron
Lead
Manganese
Mercury
Nickel
Selenium
Tellurium
Tin
Titanium
Vanadium
Zinc
Chloride
Fluoride
Nitrates
Sulfates
Total POM
Total PCS
19-2
Baseline
< 380
6.5 < 15
95 < 210
< 6
13
650
750
65 < 130
32
4300
45 < 70
70
< 1.9
1300 <1400
< 12
< 300
< 750
70 < 1600
3200 < 3400
370
12000
170 < 180
130
18000
NES
NES
19-3
Low NOV
< 540
59 < 64
640 < 740
< 8.9
4.8 < 12
2000
740
79 < 150
39 < 44
4700
9.9 < 21
99
0.06 < 21
1600
9.9 < 290
< 450
< 1000
120 < 2500
3400 <3600
810
3500
64 < 79
110 < 120
18000
50 < 51
< 7
tions by
q/n?
19-4
Low HOX
< 350
55
800 < 850
< 6
1.1 < 6
440 < 460
530
18 < 85
95
3100
< 15
65
2
2200
< 11
< 290
< 700
20 < 100
2400
3300
6000
24 < 33
85
21000
NES
NES
Total Emission Concen-
trations by Spark Source
Mass Spectrpmetry, Mg/m3
Best Mass Balances
19-2
Baseline
11
6.5
MC
0.055 < 3
7.5 < 13
2000 < MC
960
8 < MC
49
1300 1.05
0.80
23.00
0.31
0.30
1.35
0.83
0.61
1.20
—
—
—
0.92
1.00
1.30
—
0.52
—
19-3
Low NOV
—
1.7
1.90
—
—
0.55
2.90
0.54
1.00
0.89
—
1.66
0.83
1.40
—
—
1.40
0.30
1.60
17.00
0.70
—
19-4
Low KOx
< DL
1.60
2.30
< DL
> 0.05
0.12
3.00
0.10
0.74
2.50
< DL
1.62
1.10
< 0.15
< DL
< DL
> 0.22
0.18
15.00
46.00
0.09
—
See notes on Table 3-8.
-------
Mass balances for the baseline test were within a factor of two for
ten elements. Arsenic, cobalt and copper were underbalanced (emission rate
less than fuel input) by less than a factor of two. Barium and chromium
were overbalanced (emission rate greater than fuel input) by more than a
factor of two. For seven elements (antimony, beryllium, mercury, selenium,
tellurium, tin and chloride) the values in both fuel and emissions were less
than detection limits by both AA and SSMS so no mass balance could be
obtained.
Mass balances for the first low NO test (19-3) were all within a
X
factor of two except for chromium and chloride both of which were overbalanced
in all three tests. Chromium might be expected to be overbalanced because
of the stainless steel used in the SASS train. However nickel proved to be
balanced within +_ 20% for all three tests indicating no contamination by
train stainless materials.
The observed increase in solid particulates, previously mentioned, of
30 to 60% for the low NO condition compared with baseline was reflected in
X.
the measured emissions of barium, chromium, cobalt, copper, iron, manganese,
titanium, and zinc collected in the solid section of the SASS. Comparison
of Test 19-3 (low NO ) with Test 19-2 (baseline) indicates that calcium,
X
chromium, iron, manganese, titanium, and zinc were increased by 20 to 90% in
the solid particulate less than 3 Um. These same elements and also barium,
cobalt and copper were increased by over 20% in the total amount of solid
particulate collected. The other elements that could be detected did not
appear to be increased in the solid particulate collected.
The use of three cyclones and a filter in the SASS train provides
data on the enrichment of species on small particles. Particle surface area
per unit mass increases as particle size decreases. Therefore species that
condense on particle surfaces will be more concentrated on the smaller
particles. Species that showed a definite enrichment include arsenic, cobalt,
copper, iron, manganese, nickel, vanadium, zinc, chloride and sulfates.
Species that showed no definite enrichment were calcium, chromium, fluoride,
and nitrate. The remaining species were not present in sufficient quantities
to allow an assessment.
80
-------
Conclusions with regard to the inorganic species were that operation
of this boiler with the combustion modifications implemented tended to produce
an increase in emissions of certain inorganic species in rough proportion
to the increase in solid particulate. Within the precision of the sampling
methods there was no evidence to suggest any significant increases in
emissions attributable to causes other than increased solid particulate
emissions.
Organic species were difficult to identify. Most samples were of
insufficient size to allow all inorganic and organic analyses to be performed.
No polychlorinated biphenyls (PCB) were identified in any samples that could
be analyzed for organics. With regard to total polycyclic organic matter
(POM) , there was an indication that POM increased at the low NO condition.
X
Comparison of Table 3-14 with Table 3-10 shows that POM was present in the
condensate at 0.5 yg/m for the low NO condition but was below detection in
X
the condensate for the baseline condition. Howevar in both cases the POM was
below detection by GC in the XAD-2 resin. The organic module rinse for the
low NO condition (19-3, Table 3-14) contained the largest amount of total
x 3
POM, 50 yg/m . Unfortunately there was insufficient organic module sample
for the baseline test 19-2 so it is not possible to make a firm conclusion.
Since POM concentrations in the XAD-2 resins were below detection
limits for gas chromatography, the XAD-2 samples were analyzed by gas
chromatography-mass spectrometry (GC-MS). Analysis of the organic module
wash by GC-MS might have also been informative. However, these samples were
entirely consumed in performing other analyses. The results for the XAD-2
samples, Table 3-21, indicate a significantly lower level of total POM for
the low NO test compared with baseline. 0* the eight compounds required
to be identified (Table 2-1), only three were detected as noted in the table.
The fact that POM is lower in the XAD-2 for test 19-3 (Optimum Low NO mode) ,
X
compared to baseline, is in contrast to the results for the organic module
wash and condensate noted above.
81
-------
TABLE 3-21. POM COMPOUNDS IN THE XAD-2 RESIN DETERMINED BY
GAS CHROMATOGRAPH-MASS SPECTROMETRY, LOCATION 19
POM Component
Anthracene
Phenanthrene
*Methyl Anthracenes
Fluoranthene
Pyrene
*Benzo (c)phenanthrene
Chrysene
Benzo Fluoranthenes
*Benz (a)pyrene
Benz (e)pyrene
Total POM
Test 19-2,
ng/g
3.2
—
0.2
1.2
0.05
0.002
0.03
0.007
0.004
0.004
4.74
Baseline
ng/m^
24
—
1.6
9.0
0.4
0.02
0.19
0.05
0.032
0.032
35.5
Test 19-3,
ng/g
0.45
0.02
0.12
0.13
0.05
—
0.004
0.007
—
—
0.78
Low NO
X
ng/m^
3.4
0.1
0.9
0.9
0.4
—
0.03
0.05
—
—
5.8
*Compounds required to be identified for this contract
_Q
Note: Values in this table are expressed in nanograms (ng), (1 ng = 10 g)
Values in other trace species and organics tables in this report are
expressed in micrograms (ug), (1 yg = 10~6 g).
82
-------
3.1.7 Boiler Efficiency
Boiler thermal efficiencies were determined by the ASME Heat Loss
Method using on-site measurements of the fuel and flue gas compositions. The
efficiency of steam generating equipment determined within the scope of the
ASME Code is the gross efficiency and is defined as the ratio of the heat
absorbed by the working fluid to the heat input. This definition disregards
the equivalent heat in the power required by the auxiliary apparatus external
to the envelope. The abbreviated efficiency calculation considers only the
major heat losses and only the chemical heat in the fuel as the input.
Location 19 Boiler Efficiency
Baseline Condition—
Thermal efficiency was measured at baseline conditions with the boiler
firing #2 and #6 oil and natural gas. A baseline measurement was made at the
start of the test series and each day prior to combustion modification tests.
With #2 oil and #6 oil the thermal efficiency at baseline condition was
82.5%. The baseline thermal efficiency of the boiler when firing natural gas
was 79.2%
Effect of Excess Air—
The effect of low excess air firing is shown in Figure 3-17 where
boiler thermal efficiency as a function of stack gas oxygen content is plotted
for natural gas fuel. Two burner configurations, a ring burner and a gas
gun burner, were tested. The effect of excess 0 on efficiency was similar
for both burners, but the gas gun showed higher efficiency over the range of
0 Decreasing the excess oxygen from the baseline condition of 3.2% O to
2.0% resulted in an efficiency increase of 0.8% due to lower stack losses. A
further decrease in O to 1.1% (where CO increased drastically) led to only
an additional 0.1% increase in efficiency.
The effect of excess oxygen on boiler thermal efficiency is illustrated
in Figure 3-18 for #2 oil firing and in Figure 3-19 for #6 oil firing. With
#2 oil, decreasing the excess O from the baseline value of 3.05% to 0.6% O^
83
-------
80
CO
79
0)
•H
U
•H
«W
-------
CO
Ul
84
83
<*>
u
g 82
-H
u
•H
81
0
CQ
80
(12)
(13)
79
I
Location 19
Load: 83% of Rated
Fuel: No. 2 Oil
Air Atomization
( ) Test Number
(10) Baseline
(ID
345
Stack Gas Excess Oxygen, %, Dry
Figure 3-18. The effect of excess oxygen on boiler thermal efficiency (#2 oil).
-------
Ch
85
84
83
82
81
80
79
T
T
(43)
(40) Baseline
Location 19
Load: 83% of Rated
Fuel: No. 6 Oil
Air Atomization
( ) Test Number
(41)
1
I
_L
2345
Stack Gas Exdess Oxygen, %, Dry
Figure 3-19. The effect of excess oxygen on boiler thermal efficiency (#6 oil).
-------
resulted in an increase in thermal efficiency of 1.9%. Decreasing the 0
from 0.9% to 0.6% (the high CO threshold) resulted in a minimal increase
in efficiency of 0.1%. With #6 oil, only 0.2% increase in efficiency was
realized when the 0 decreased from 1.3% to 0.7%.
In general, the efficiency increased 1-2% when the 0 was lowered
to near the CO threshold from the baseline 0 condition for all test fuels.
j^
Peak efficiency for oil fuels was approximately 84% at 0.7% O and 79% at
^
1% o for natural g~is.
£
Effect of Staged Combustion Air—
Boiler thermal efficiency was calculated using measurements taken
during the staged combustion air parametric tests. The depth at which the
secondary air was injected was varied as was the ratio of burner air to
secondary air. When firing #2 oil, a very slight change in efficiency was
noted as the injection point was varied. The influence of injection point
on efficiency is illustrated in Figure 3-20 for #2 oil. These data show an
increase of only 1/2% from the baseline condition as the injection point is
varied to the maximum distance of 7 feet with a burner equivalence ratio of
approximately 1. Slightly greater increases are noted for burner equivalence
ratio of 1.1 (fuel rich). All data are at approximately the same operating
condition of 3% excess O . When firing #6 oil and natural gas, no measurable
change was noted as injection depth was varied. The only change in boiler
thermal efficiency was due to changes in excess 0 .
Effect of Flue Gas Recirculation—
The effect of flue gas recirculation on boiler efficiency is shown in
Figures 3-21 through 3-23 for #2 oil, #6 oil, and natural gas respectively.
Figure 3-21 shows the change in efficiency, normalized to the baseline
efficiency as a function of the percent of flue gas recirculated to the
burner for #2 oil firing. These data indicate that boiler efficiency is more
sensitive to changes in flue gas recirculation when the boiler is operating
at the normal 0 level of 3% than when operating in the low 0 mode. When
2 ^
the recirculated flue gas approaches 25% of the total, the boiler efficiency
drops by approximately 3%. In the low 0 mode (^ 1% excess 0^) , the
degradation in efficiency due to 25% recirculated flue gas is only 1.5%.
87
-------
1.0
oo
oo
o
o
o
9
•H
U
•H
w
0.5
T
T
Location 19
Load: 83% of Rated
Fuel: No. 2 Oil
Air Atomization
Stack Gas O =
( ) Test No.
3%
(32)
(31)
B = 1.10
4>B - 1.02
= (VF) stoic
(A/F)actual
L
2345
Secondary Air Tube Insertion Depth, ft.
Secondary Air Tube Insertion Depth, m
Figure 3-20. The effect of secondary air insertion depth on boiler thermal efficiency (#2 oil).
-------
-3
00
VD
O
o
4-1
m
W
n-i
H
w
-1
I I
Location 19
Load: 83% of Rated
Fuel: No. 2 Oil
Air Atomization
( ) Test Number
VL. @ 3% = 82'7%
1
10 15 20
Recirculated Flue Gas, %
Normal O,
(21)
25
30
Figure 3-21.
The effect of flue gas recirculation rate on boiler thermal efficiency
(#2 oil).
-------
-1.0
O
O
PQ
M-l
w
Jl
M-l
w
M-l
w
-0.5
0
L
(48)
Location 19
Load: 83% of Rated
Fuel: No. 6 Oil
Air Atomized
( ) Test Number
Normal O
3.2%
I
10 15 20
Recirculated Flue Gas, %
25
30
Figure 3-22. The effect of flue gas recirculation rate on boiler thermal efficiency
(#2 oil).
-------
-1. O
O
O
m
4-t
CQ
4J
U
H
w
-0.5
Location 19
Load: 83% of Rated
Fuel: Natural Gas
Gas Gun Burner
10 15 20
Recirculated Flue Gas, %
Figure 3-23.
The effect of flue gas recirculation rate on boiler thermal efficiency
(natural gas).
-------
On an absolute basis, the boiler efficiency is approximately 1% higher
when operating in the low 0 mode with flue gas recirculation than operation
at the normal O level.
Efficiency degradation as a function of percent recirculated flue
gas is shown in Figure 3-22 for #6 oil firing. At the maximum recirculation
rate, boiler efficiency was degraded approximately 0.8% from the baseline
condition.
Figure 3-23 illustrates efficiency degradation as a function of flue
gas recirculation rate for natural gas firing at normal O conditions.
Maximum recirculation rate of 20% results in an efficiency loss of
approximately 1%.
The effects on efficiency of the various combustion modifications and
combinations of the modifications are tabulated in Table 3-22. The values
enclosed in boxes represent the lowest NO condition. With the oil fuels,
X
the low NO condition was with flue gas recirculation, staged combustion air
JC
and low excess air firing. With #2 oil, operation in this mode resulted in
an efficiency degradation of 0.8% and with #6 oil the efficiency increased
0.1%. Natural gas firing in the low NO mode (FGR and low O ) resulted in an
X £+
efficiency penalty of 0.4%.
3.1.8 Conclusions From Location 19 Tests
The test series conducted on the watertube boiler at Location 19 per-
mitted an evaluation of several combustion modification techniques using #2
and #6 oil and natural gas.
Combustion modifications evaluated were lowered excess air, flue gas
recirculation, staged combustion air, and combinations of these. The data
presented previously in Table 1-1 show the reduction in NO from the baseline
condition as a function of combustion modification technique and type of fuel.
These data indicate that for oil fuel, the combination of all three combustion
modification techniques results in the greatest reduction in NO . For natural
gas, the maximum deduction occurs with the combination of flue gas recircula-
tion and low 0_. NO reductions were greatest with #2 oil and natural gas
which showed 77% and 79% reductions respectively. A 53% reduction in NO was
A
achieved while firing #6 oil using FGR, staged air, and low 0 .
£
92
-------
TABLE 3-22. SUMMARY OF CHANGE IN BOILER EFFICIENCY
DUE TO COMBUSTION MODIFICATIONS
Boiler
Operating
Mode
Low 0
£i
SCA, Normal O
SCA, Low 02
FGR, Normal O
fi
FGR, Low O
FGR + SCA, Normal O
FGR + SCA, Low O
No. 2 Oil
+ 1.5%
+ 0.9%
+ 1.1%
- 1.9%
+ 0.9%
- 1.2%
| - 0.8% |
No. 6 Oil
+ 1.5%
+ 0.1%
+ 0.8%
- 0.7%
+ 0.6%
- 0.8%
j + 0.1% |
Natural Gas
+ 0.9%*
+ 0.6%*
+ 0.6%+
- 0.8%+
| - 0.4%+|
- 0.5%*
§
*0ptimized gas gun
^Ring burner
^Stability limits prevented lowering O,,
Indicates lowest NO condition I ]
x
93
-------
Table 1-2 presented the results of particulate measurements using EPA
Method 5 for the boiler under baseline operation in each modified condition
and for each fuel. The lowest particulate emissions also occur at the same
condition as the low NO mode.
x
3.2 LOCATION 38 COMBUSTION MODIFICATIONS
The test program was conducted during the period of May 3 through
June 11, 1976. The test unit was a watertube boiler operating on #6 fuel oil
and natural gas at a nominal load of 40,000 Ib/hr steam flow. The design
capacity was 45,000 Ib of steam flow per hour. The testing consisted of
measuring gaseous and particulate emissions at three different modes of opera-
tion and comparing these emissions to baseline operation. The modified modes
of operation were staged combustion, variable preheater temperature and excess
air variations. Combinations of these variables were also evaluated.
Gaseous emissions were measured for both fuels with the exception of
SO which was only measured with #6 oil. Particulate emissions measurements
x •*
were only performed with #6 fuel oil. Table 3-23 presents a summary of
emission data at modified conditions as well as baseline conditions.
The gaseous emissions were sampled at the boiler exit prior to
the air heater just as was done during baseline testing. Cold line data
was used as the primary NO value since some problems with the heated line
were encountered during the testing.
Excess air variations, variable preheat, windbox register vane
setting variation and staged combustion air tests were conducted.
The side walls of the boiler tested were fitted with a series of
opposing ports at five locations which ranged in distance from less than 100
to more than 300 centimeters from the furnace front (Fig. 2-4). Separate
fan and duct work was provided to allow a fraction of the total boiler air
flow to be admitted at these ports to give staged or secondary combustion.
The relative air flows entering the boiler through the burner and through
the secondary air ports were determined from velocity profiles taken in the
fan inlet ducts using a standard pitot tube. It should also be noted that
94
-------
TABLE 3-23- SUMMARY OF LOCATION 38 COMBUSTION MODIFICATION TEST DATA
Test No.
200G-2
201G-5
204-9
205-4
200G-19
204-18
204/5-30
200-24
201-12
202-4
203-26A
203-26B
201-15
203-15
203-22B
Load
Date kg/s O2
1976 (103 Ib/hr) Fuel (%)
6/2
6/2
6/7
6/8
6/11
6/9
6/11
5/24
5/25
5/25
5/20
5/20
5/27
5/17
5/18
5.04 NG 1.60
(40.0)
4.94 NG 1.25
(39.2)
4.91 NG 2.25
(39.0)
4.98 NG 2.2
(39.5)
4.98 NG 1.98
(39.5)
5.04 NG 2.25
(40.0)
4.91 NG 2.25
(39.0)
4-79 #6 2.9
(38.0)
4-89 #6 1.55
(38.8)
4-28 #6 2.6
(34.0)
4-79 »6 3.3
(38.0)
4-91 »6 3.1
(39.0)
4-89 #6 1.6
(38.8)
4.73 *6 3.0
(37.5)
4.66 #6 2.8
(37.0)
NOX NO
ng/J ng/J
(ppm) (ppm)
* 82.1
(161)
* 70.9
(139)
* 56.6
(111)
* 62.2
(122)
* 66.8
(131)
« 52.5
(103)
* 24.0
(47)
* 145.9
(286)
* 115.3
(226)
* 132.1
(259)
* 97.4
(191)
*
* 128.0
(251)
* 86.0
(153)
* 91.0
(162)
CO HC SO2 SO 3
ng/J CO2 ng/J ng/J ng/J
(ppm) » (ppm) (ppm) (ppm)
43.3 10.5 *
(140)
>619 10.5 *
(>2000)
37.8 10.5 *
(122)
3.1 10.5 *
(10)
10.5 *
85.1 10.5 *
(275)
17.3 10.25 *
(56)
7.5 13.8 * 737.3 21.9
(22) (944) (28)
22.2 15.0 * 740.4 i-,.9
(65) (948) (14)
10.9 14.1 *
(32)
21.1 13.0 * 788.8 10.9
(62) (1010) (14)
34.1 13.0 *
(100)
15.0 * 781.0 28.1
(1000) (36)
39.9 12.1 *
(117)
34.1 12.0 *
(100)
Particulate Stack
Total Solid Temp.
ng/J nq/J K
Ub/MMB) (Ib/MMB) (°F)
497
(434)
494
(430)
497
(435)
514
(465)
492
(425)
506
(450)
564
(556)
66.4 36.6 501
(.154) (.085) (442)
47.6 38.3 493
(.110) (.088) (427)
54.4 37.6 511
(.126) (.087) (550)
52.6 38.7 491
(.122) (.09) (424)
62.2 39.9 492
(-144) (.092) (416)
43.7 38.7 494
(.101) (.089) (430)
498
(436)
508
(455)
Eff.
(%) Comments
81.2 Baseline
81.4 O2 swing, Port 14 & 15
80.8 Staged combustion air
80.2 Variable preheat (VPH)
81.2 Windbox register
adjustment
80.4 Constant SCA, variable
FD fan. Ports 14 & 15
open
78.0 SCA multiple port
variation with APH
bypass. Ports 14 &
15, 6 S 7
84.8 Baseline, particulate.
impactor, SO
85.6 Low O2, particulate.
SO
X
82.0 VPH, particulate, SOX
85.0 SCA, particulate, SO,,
85.0 SCA, particulate
85.6 Low O , particulate
2
84.9 Constant SCA, variable
FD fan, Port 14 & 15
84.6 SCA multiple port
variation 14 & 15
open, 12 & 13 10% open
* Heated line malfunction prevented measurement of NO and HC data.
-------
the secondary air fan delivered essentially ambient (boiler room) temperature
air into the boiler compared to the preheated burner air flow. Thus, the
influence of reduced air temperature as well as staged combustion must be
considered when evaluating the NO trends.
A gaseous emission traverse was conducted at the boiler outlet for
both fuels. Figures 3-24 and 3-25 show the variation in emission values
versus probe insertion depth. The graphs indicated an O2 variation across
the duct that increases on the east and west walls. The 02 variation is
approximately 2% for baseline conditions for either fuel. The high O2
values near the walls indicate furnace air leakage especially since the
NO distribution does not follow the 0 distribution in the duct. The sample
£
probe was installed at the center of the boiler outlet duct where there
was no interference of wall air leakage.
3.2.1 Location 38 Baseline Tests
Baseline emissions measurements were made with the boiler in the
"as found" condition firing #6 fuel oil. Subsequent baseline tests were
made with the boiler firing natural gas. Baseline measurements were made
at the start of each series of combustion modification tests. The boiler
load was constant at approximately 89% of rated load for all tests.
The measured baseline NO emissions when firing #6 oil were 167.5
ng/J (298 ppm). The baseline NO values for natural gas were measured at
X
82.1 ng/J (161 ppm). Baseline particulates were 66.4 ng/J (0.154 Ib/MMBtu).
Particulate size distribution was also measured at baseline conditions.
The baseline size distribution indicated that more than 90% of the particulate
was 3 micrometers diameter or less. Analyses of the #6 fuel oil and natural
gas are presented in Tables 3-24 and 3- 2S
The heated sample line was not operating during the test series so
that only cold line NO data are recorded. The hydrocarbon data are also
not reported because the hot line malfunctioned.
96
-------
CN
O
4J
•H
X
H
0
ffl
0
0
West
#6 Oil
Baseline
Location 38
A
D
O
NO
CO
Boiler Outlet
4 6
Probe Location (ft)
260
O
3
220
60
40 e
%
a
8
0
10
East
Figure 3-24. Emission traverse while firing #6 oil.
97
-------
(N
4J
•H
X
w
•H
0
m
Location 38
Natural Gas
Baseline
0
West
4 6
Probe Location (ft)
220
o
180*
140
40
o
u
0
10
East
Figure 3-25. Emission traverse while firing #6 oil.
98
-------
TABLE 3-24. SUMMARY OF LOCATION 38 FUEL OIL ANALYSES
Laboratory No.
Carbon , %
Hydrogen , %
Nitrogen , %
Sulfur, %
Oxygen , %
Ash, %
API Gravity
HHV, Btu/lb
G3393
86.21
11.22
0.32
1.88
0.34
0.03
15.2
18,449
G3430
86.26
11.20
0.30
1.88
0.29
o.oi
15.2
18,484
TABLE 3-25. LOCATION 38 NATURAL GAS ANALYSIS
Oxygen, % 0.00
Nitrogen, % 0.28
Carbon Dioxide, % 0.60
Methane 96.99
Ethane 1•98
Propane 0.10
Butanes °-04
Pentanes 0.01
Hexane °-00
Heating Value, Btu/SCF (Dry) 1011
99
-------
3.2.2 Combustion Modifications With #6 Oil
o Excess Air Variations—
The effect of excess oxygen on NO emissions is shown in Figure 3-26
for the unit firing #6 oil. The excess oxygen varies from a low value of
1.25% to a high value of 4.4%. Baseline O2 for this test series was found
to be at 2.5% with 221 ppm NO at 3% 02 dry. At the low O2 setting of 1.25%,
CO values of approximately 950 ppm were measured. Increasing the air flow
to 1.6% 02 decreased CO emissions to approximately 80 ppm.
o variable Preheat (VPH)—
Figure 3-27 shows the effect of windbox temperature on NO emissions
for #6 oil. The graph shows a considerable decrease of NO with reduced
combustion air temperature. The high windbox temperature is obtained with
a steam coil air heater located between the FD fan and the normal regenerative
air heater. Windbox temperature is lowered with a bypass duct and damper
which redirect the FD exit flow either partially or totally around the
air/flue gas heat exchanges. The test series was conducted at a load of
85% of rated load with approximately 2.55% excess oxygen. The NO decreased
by 2.75 ppm/10 °F windbox temperature reduction.
o Staged Combustion Air (SCA)—
The data of NO versus staged combustion air port location for oil
X
fuel (at a nominal ratio of SCA to total air flow of 14%) indicates that
the most effective location for staged air is the farthest from the burner,
ports 14 and 15 (Fig. 3-28). Ports 14 and 15 are the same distance from
the burner as ports 12 and 13 but are 40 inches higher. Ports 12 and 13 are
the next most effective and the effectiveness decreases as the distance from
the burner diminishes.
100
-------
300
250
200
ro
(SJ
£150
I
O
a
100
(201-1)
(201-3)
(201-4)
Baseline
(201-8)
Location 38
Load = 89% of rated
Steam Atomization
#6 Oil
@ CO ^ 950 ppm
^ CO ^ 80 ppm
O CO < 20 ppm
( ) test number
Boiler Exit 02 (%) dry
Figure 3-26. The effect of excess oxygen on NO emissions (#6 oil).
101
-------
300
200
I
a
100
(202-3)
Baseline
(200-21)
(202-2)
(202-1)
Location 38
Load = 85% of rated
Steam atomization
#6 Oil
O = 2.55%
( ) test number
100 200
Windbox Temp. (°F)
300
400
Figure 3-27. The effect of windbox temperature on NO emissions (#6 oil)
1D2
-------
•a
200
o
2)
100
2.9
2-7
Q2.7
A
U
2"7
3.6
Q
2.3%0
Nominal 14% SCA Plow
Symbol Port Open
None
6 & 7
8 & 9
O
Q
10 & 11
12 & 13
14 & 15
1
Load = 87% of rated
1
6&7 8&9 10&11
Position No.
I [
12&13
100 200 300
Distance from Furnace Front (cm)
Figure 3-28. NO versus SCA -
port location for #6 oil fuel.
KVB 6004-734
103
-------
A further series of tests was conducted to evaluate the sensitivity of
NO emissions to burner stoichiometry. For these tests both the total boiler
air flow and the fraction of the total air flow entering through the SCA ports
were varied by (1) fixing the forced draft (FD) fan flow and reducing the SCA
flow and (2) by fixing the SCA flow and varying the FD fan flow. The data are
shown in Figure 3-29 for #6 oil fuel. The lowest NO level is obtained at the
minimum burner air flow. This trend is more readily observable when the NO
data are presented as a function of the percent of stoichiometric air at the
burner as shown in Figure 3-30. The oil data in Figure 3-30 correlates the
data from Figure 3-29 and illustrates as well the effect of SCA port location.
j
The multiple port variations of the staged combustion air with the
boiler firing #6 fuel oil (Fig. 3-31) shows a similar trend as the single
port 14 and 15 operation. The only multiple ports tested were combinations
of ports 12 and 13 with ports 14 and 15 and little further reduction was
observed. There appears to be no advantage to running lower than stoichio-
metric air through the burner with these test conditions.
3.2.3 Particulate and SO Testing
Particulate, particulate size, and sulfur oxides measurements were
made with the boiler operating on the #6 fuel oil. Measurements were made
with the boiler operating in the baseline condition and with lowered excess
0„, variable air preheat and staged combustion air.
A summary of the Method 5 particulate measurements is presented in
Table 3-26. These measurements indicate that lowered excess air resulted in
the lowest total particulate emissions. All combustion modifications resulted
in lower total particulate measurements than the baseline condition, but solid
particulate emissions for all combustion modifications were higher than for
baseline.
SO emissions were measured for baseline, low 0_, staged combustion
X £,
air and variable preheat operation using the Shell-Emeryville absorption-
titration method. Three samples were normally taken for each operating condi-
tion. The data obtained for the variable preheat tests are not reported due
to sampling error. The remaining data are presented in Table 3-27. The data
104
-------
300
H-
o
Ul
Location 38
#6 Oil
250
n
tSJ
>i
n
•O
Constant FD/
Variable SCA
200
-H
X
O
-p
•H
150
2 3
Boiler Exit O,
Constant SCA
Variable FD
(No ports
Baseline open)
Port 14 & 15
Shaded Symbol
Indicates CO
of 150 ppm
I
(%), dry
Figure 3-29.
Variable SCA - NO versus O..
c 2
-------
300
* 250
ro
<2>
M
13
g
a
•H
x
o
u
-H
^
-p
-H
200
150
I
Location 38
#6 Oil
O
Symbol Port Open
1
I
80 90 100 110 120
Theoretical Air at Burner, % of Stoichiometric
130
Figure 3-30. SCA single port variations.
106
-------
300
1
Location 38
#6 Oil
250
o
•-j
TJ
-------
TABLE 3-26. SUMMARY OF METHOD 5 PARTICULATE MEASUREMENTS
FOR LOCATION 38 STEAM BOILER FIRING NO. 6 OIL
Condition
Test 'No.
Total Particulate
ng/J
(Ib/MMDtu)
Solid Particulate
ng/J
(Ib/MMBtu)
Baseline
200-24
Low Excess Air 201-12
201-15
Staged Combustion
Air 203-26A
203-26B
Variable Air
Preheat (Min.
Temperature)
202-4
66.4
(0.154)
47.6
(0.110)
43.7
(0.101)
52.6
(0.122)
62.2
(0.144)
54.4
(0.126)
36.6
(0.085)
38.3
(0.088)
38.7
(0.089)
38.7
(0.900)
39.9
(0.920)
37.6
(0.087)
108
-------
TABLE 3-27. SO SUMMARY, LOCATION 38 FIRING NO. 6 OIL
Baseline
200-23
Low 0
201-12*
201-14*
SCA
203-26*
203-27
°2
Boiler
Exit
2.9
1.55
1.5
3.1
3.4
Stack
6.8
5.7
7.72
7.1
S°2 S°3
ppm-Corrected to 3%
02
944
948
1000
Avg. 974
1010
968
Avg. 989
28
13.5
35.9
Avg. 25
14.0
13.9
Avg. 14
*Single sample
109
-------
indicate that the level of total sulfur oxides emissions is dependent only
upon the fuel sulfur content and not upon operating mode. The data indicate
that staged combustion air resulted in a 50% reduction in SO^.
The particulate size distribution for boiler operation with #6 fuel
oil is presented in Figure 3-32 for four operating modes. Particle size
diameter is plotted as a function of cumulative proportion of the impactor
catch for baseline, staged combustion air, low O2 and variable preheat opera-
tion. The data indicate that the particle distribution is not represented
by a log-normal distribution since the data do not plot as a straight line.
The cumulative proportion of impactor catch below 3 urn diameter varied from
65 to 94%, indicating that the particulate catch for all operating modes is
of very small diameter material. For all operating modes, between 60 and 88%
of the catch is below 1 urn diameter. All combustion modifications resulted in
increased particle size, compared with baseline. Low excess O^ operation
produced the largest particle size with only 65% of the particulate below
3 urn.
3.2.4 Combustion Modifications With Natural Gas
Excess Air Variations—
The NO versus 0_ data for natural gas are presented in Figure 3-33.
Baseline measurements show 162 ppm NO at 1.6% O_. The excess oxygen was
varied from a low of 1.25% to a high of 4%. At the low O setting CO values
of > 2000 ppm were measured, while NO went to a low of 139 ppm. The unit
should not be operated at this low O9 level. High CO firing caused efficiency
loss and was a dangerous operating mode. Increasing O_ to approximately 1.5%
decreased CO to approximately 300 ppm. The NO peaked at 175 ppm for
approximately 3% O .
£t
Variable Preheat (VPH)—
The data for variable preheat temperature with natural gas fuel show
a big decrease in NO with reduced windbox temperature (Fig. 3-34). Dropping
windbox temperature from a baseline condition of 284 °F to a low of 145 °F
110
-------
100
I I I I
Q Test No. 200-24 Baseline
Test No. 201-13 Low 0
Test No. 203-27 SCA
Test No. 203-28 SCA
Test No. 202-5 VPH
Location 38
Load = 89% of rated
Fuel: #6 Fuel Oil
30 40 50 60 70 80 90 95 98 99 99.8
Cumulative Proportion of Impactor Catch, % by Mass
Figure 3-32. Particulate size distribution for an oil fired steam boiler.
Ill
-------
150
CM
ro
(2)
§
100
50
Baseline
(200G-2)v
Location 38
Load = 89% of rated
Gas Ring Burner
^ High CO > 2000 ppm
Moderate CO ^ 300 ppm
D Low CO < 150 ppm
( ) test number
Boiler Exit 0
3
(%) dry
Figure 3-33. The effect of excess oxygen on NO emissions (natural gas)
112
-------
200
(205-1)
aseline
(200G-7)
-------
reduced NO emissions from 166 ppm to 122 ppm. All tests in this series were
conducted at a load of 86% of rated load with an excess oxygen level of
approximately 2.3%. The NO decreased by 3.65 ppm/10 °F windbox temperature
reduction.
Windbox Register Vane Setting Variation—
Variation of the windbox register vane angle testing was performed
with natural gas and shows that NO increases significantly as the-vanes
which swirl the air entering the burner throat were closed (Fig. 3-35) . The
increased swirl contributed to better mixing and results in higher NO emis-
sions. Closing the register vanes also decreased the effective area for
air entering the burner which increased the FD fan discharge pressure and
reduced the maximum obtainable air flow.
The FD fan was required to operate at maximum output for register
settings of 54 degrees and less. The nominal baseline operating vanes
setting is 66 degrees open. Even with the FD fans at maximum it was necessary
to increase the furnace draft to obtain sufficient air flow to prevent CO.
To maintain proper operating O0 levels (approximately 2.4%), high negative
*£
furnace pressures are required. The lower NO observed at the 42 degree
vane position required 0.6 in. HO negative furnace draft (compared to
£*
-0.2 in. HO) which would raise the furnace leakage considerably. Thus,
the actual burner excess air is probably lower than would be indicated by
the measured 2.5% 0_.
Staged Combustion Air (SCA)—
The data for staged combustion port location versus NO for natural
gas (Fig. 3-36) shows again as it did for oil fuel, that the most effective
location for staged air are the ports 14 and 15. In addition, ports 6 and 7,
those nearest the burner, exhibited the same effectiveness on NO reduction
as did the previously mentioned ports 14 and 15.
The differences seen with natural gas may result from: (1) the different
geometric relationship between the burner fuel and air flows with the two
fuels (the oil gun being in the center of the burner air flow and the gas
ring being on the outside of the burner air flow and more directly influenced
by the secondary air), and (2) the greater sensitivity of the NO emissions to
combustion air temperature with natural gas.
114
-------
300
200
o
dP
ro
I
o
100
(200G-15)
(200G-16)
Baseline
(200G-8)
(200G-9)
Location 38
Load = 89% of rated
Gas Ring Burner
O
2.4%
( ) test number
20 40 60 80
Register Vane Position (degrees open)
100
Figure 3-25. NO versus windbox register setting (natural gas).
115
-------
200
13
fN
ro
, 100
"
Oi
O
"Z
Base
Base
2.3
2 4
—i r
Nominal 14% SCA Flow
2'4%°
X
2.7
3.0
2.5
Symbol Port Open
0
O
O
D
None
6 & 7
8 & 9
10 & 11
12 & 13
14 & 15
LoacT= 86% of rated
I
6&7
8&9 10&11
Position No.
12&13
14&15
i
1
I
100 200 300
Distance from Furnace Front (cm)
Figure 3-36. NO versus SCA - port location for natural gas fuel.
116
-------
Figure 3-37 shows the effect of variable SCA fan and FD fan versus
NO and 02 for natural gas fuel. The lowest NO levels correspond to the
minimum burner air flows. NO data as a function of the percent of stoichio-
metric air at the burner are presented in Figure 3-38. These data demonstrate
the dominance of burner stoichiometry over total boiler stoichiometry in
regard to NO emissions.
The operation on natural gas using single and multiple port locations
combined with full air heater bypass (Fig. 3-39) indicates that very low
values of NO can be obtained. An NO reduction of 69% relative to the baseline
condition was measured. The three best SCA port locations indicated little
difference in NO emissions with ports 6 and 7 and 14 and 15 being slightly
better than ports 12 and 13. All three port locations indicate some CO emis-
sions with ports 12 and 13 and 6 and 7 showing 200-400 ppm. The full open
bypass requires maximum ID fan capacity to maintain 0.5 kPa (0.2 in. HO)
negative furnace pressure with one pair of ports open and operated at negative
pressures near 0.25 kPa (0.1 in. HO) with 2 pairs of SCA ports fully open.
3.2.5 Boiler Thermal Efficiency
Boiler thermal efficiency was calculated based on the ASME Heat
Loss Method for various operating modes. The calculated efficiency as a
function of boiler operating condition for both natural gas and #6 oil was
presented in Table 3-23.
Very little effect of operating mode on efficiency was exhibited
with both natural gas and #6 oil. With natural gas, a slight increase in
efficiency was shown with lowered excess 02 and the variable preheat test
showed a loss in efficiency of 1%. With #6 oil, the low O^ condition
resulted in slightly higher efficiency (0.8%) than the baseline condition
but the variable preheat test showed nearly a 3% degradation in efficiency.
3.2.6 Conclusions From Location 38 Tests
The tests conducted on the watertube boiler at Location 38 permitted
an evaluation of several combustion modification techniques using #6 oil and
natural gas fuel. The following combustion modifications were evaluated:
lowered excess air, variable preheat temperatura, staged combustion air
117
-------
200
Locationi 38
Natural <3as
150
m
(Si
n
•o
§,100
a
(216 ppm CO)
(66 ppm CO)
(122 ppm CO)
SCA Fan Constant
Variable FD Fan
Q)
T!
•H
X
o
u
•H
S-l
4J
•H
50
Shaded Symbol
Indicates CO
Boiler Exit O (%)
Figure 3-37. Variable SCA - NO versus 0
C £,
118
-------
200
O
cs, 150
•a
100
0
•H
M
-P
•H
S
50
"~l 1
Natural Gas
Location 38
O
- D
f*\ jf
OA-/A
i i
i i
0
0
—
O
Symbol Port Open
0 None
0 6 & 7
Q 8 & 9 ~~
D 10 & 11
V 12 & 13
A 14 & 15
! 1
80
90 100 HO 120
Theoretical Air at Burner, % of Stoichiometric
130
Figure 3-38. SCA single port, Tests 204-1 through 204-22.
119
-------
to
o
150
CM
100
<3i
M
TJ
14 & 15 - 100 6 & 7 - 100
(^ 14 S 15 - 100 6 & 7 - 100
($" 14 S 15 - 30 6 S 7 - 100
1
90 100 no 120 130
Theoretical Air at Burner, % of Stoichiometric
140
Figure 3-39. SCA multiple port combinations with air heater bypass 100% (^ 140 °F) ,
Tests 204-23 through 204-32.
-------
injection at different port locations, windbox register vane setting variations
and combinations of these. The data previously presented in Table 1-5 show
the reduction in NO from the baseline condition as a function of combustion
modification and type of fuel. The data show that the maximum reduction in
NOx while firing natural gas was accomplished with a combination of staged
combustion and lowering the air preheat temperature. This combination
resulted in a 69% reduction in NO from the baseline condition. While firing
#6 oil, the maximum reduction demonstrated was 43% using staged combustion
air and lowered excess 0~.
The excess Q^ data indicate that the effect of 0 on NO for natural
gas is leveling off above approximately 2-1/2% O . The maximum reduction,
^
at 1.25% 02, was approximately 14% less than the nominal condition. With #6
oil a similar trend was exhibited. The reduction was approximately 20%
from the nominal condition of 2.9% O .
The reduction of windbox air temperature showed a large decrease in
NO emissions for both fuels. The windbox temperature variation had a greater
effect on NO with natural gas firing than with #6 oil firing. Nominal values
of NO decreases were 3.65 ppm/10 °F for natural gas and 2.75 ppm/10 °F for
#6 oil. These NO reductions were accompanied by a loss in thermal efficiency.
With natural gas there was a decrease of 2 points in efficiency when air
temperature was lowered to 145 °F, and with #6 oil there was a loss of 3
points when air temperature was lowered to 128 °F. Baseline air temperatures
were 283 and 272 °F for natural gas and #6 oil, respectively.
The data for NO versus SCA port location for oil fuel indicate, that
the most effective injection point is the most distant from the burner. The
data for natural gas firing indicate a similar trend, except that the ports
nearest the burner exhibited the same effectiveness as did the most distant
ports. This may be due to the different geometric relationship between
burner fuel and air flows (center oil gun versus outside ring gas burner)
and greater sensitivity of the NO emissions to combustion air temperature
with natural gas. At the nominal condition of 2.8% 02 and 14% SCA flow,
the NO was reduced 30% for #6 oil with injection at approximately 300 cm
x
121
-------
from the furnace front. Reducing the operating 0_ to 2.3% reduced the NO
£ x
by 42% from the baseline condition. Staged combustion with natural gas
resulted in a reduction of 32% from the baseline condition with the boiler
operating at 2.4% excess O and 14% SCA.
Windbox register adjustment reduced NO on natural gas fuel by 18%
from the baseline condition. For natural gas the maximum reduction occurs
with the combination of air preheater bypass and SCA multiple port combina-
tion. Test 204/5-30 shows a 69% NO reduction, with no apparent side effects.
122
-------
REFERENCES
1. Cato, G. A., et al. , "Field Testing: Application of Combustion
Modifications to Control Pollutant Emissions from Industrial
Boilers - Phase I," EPA 650/2-74-078a, NTIS Order No. PB 238 920,
June 1975.
2. Cato, G. A., et al., "Field Testing: Application of Combustion
Modifications to Control Emissions from Industrial Boilers - Phase
II," EPA 600/2-76-086a, NTIS Order No. PB 253 500, April 1976.
3. Hamersma, J. W., Reynolds, S. L., and Maddalone, R. F., "IERL-RTP
Procedures Manual: Level I Environmental Assessment," EPA
Report 600/2-76-160a, NTIS Order No. PB 257 850, June 1976.
123
-------
BLANK PAGE
124
-------
APPENDIX A
TRACE SPECIES AND ORGAN1CS
SAMPLING AND ANALYSIS PROCEDURES
Table of Contents
Page
A-1.0 INTRODUCTION 127
A-2.0 PREPARATION OF XAD-2 RESIN 130
A-3.0 PREPARATION FOR A SAMPLING RUN 133
A-4.0 SAMPLING PROCEDURES 14°
I
A-5.0 TRAIN DISASSEMBLY AND SAMPLE RECOVERY 144
A-6.0 SUPPLEMENTARY REFERENCE MATERIAL 149
A-7.0 SAMPLE PREPARATION AND ANALYSIS (Calspan Corp.)
A-3.0 SAMPLE PREPARATION AND ANALYSIS (Battelle) 163
Note: Units on values in this Appendix are given in the actual English
or metric units as used or measured on field equipment. Alternate
English to metric, or metric to English conversions are not listed
to avoid confusion, as the Appendix is intended for direct field use.
125
-------
A-1.0 INTRODUCTION
Sampling and analysis procedures for trace species and organics emis-
sions used in the current program were based on procedures developed by the
EPA Industrial Environmental Research Laboratory at Research Triangle Park, NC.
The IERL-RTP procedures are defined in a procedures manual prepared for EPA by
TRW Systems Group (Ref. A-l) that relates the procedures in terms of a multi-
media Level I stream prioritization sampling and a Level II detailed assess-
ment sampling. Although those basic sampling procedures were adapted for the
current program (with modifications noted herein), this program was not formu-
lated in the specific Level I-Level II framework. Level I sampling is intended
to show the presence or absence and, within a factor of j^ 2 to 3, the emission
rates of all inorganic elements, selected inorganic anions and classes of
organic compounds. The current program objective is to obtain qualitative
and quantitative data for a large number of elements (approximately 60) by use
of spark source mass spectrometry and this objective is similar to the Level I
philosophy. A second objective of the current program, more related to the
Level II definition, is to more accurately quantify the emissions of the
elements, species, and organics as shown in Table A-l, and to relate the emis-
sions of these species, by mass balance, to the amounts input with fuel or
process materials. In addition to total quantities, information is
required on the relationship of particulate species emissions to parti-
culate size.
The referenced Level I procedures manual refers to several
multi-media sampling procedures. The current program is more narrowly
concentrated on exhaust emissions from the stacks of industrial com-
bustion devices. Therefore Chapter III "Gaseous Streams Containing
Particulate Matter" of the referenced manual is that portion pertinent
to the current program. That chapter discusses sampling with the use
of a "Source Assessment Sampling System" (SASS). The features of that
127
-------
TABLE A-l. TRACE SPECIES AND ORGANICS TO BE IDENTIFIED
Antimony
Arsenic
Barium
Beryllium
Cadmium
Calcium
Chlorine
Chromium
Elements
Cobalt
Copper
Fluorine
Iron
Lead
Manganese
Mercury
Nickel
Selenium
Tellurium
Tin
Titanium
Vanadium
Zinc
Species
Total sulfates
Total nitrates
Organics
Total polychlorinated biphenyls (PCB)
Total polycyclic organic matter (POM)
Specific POM compounds:
7, 12 - dimethyIbenz (a) anthracene
Dibenz (a,h) anthracene
Benzo (c) phenanthrene
3-Methylcholanthrene
Benzo (a) pyrene
Dibenzo (a,h) pyrene
Dibenzo (a,i) pyrene
Dibenzo (c,g) carbazole
sampling train are presented in the referenced manual and will not be
repeated here. The remainder of this appendix presents the specifics
of the referenced procedures as adopted &r modified for the current
program.
The SASS sampling.train and samples obtained are shown schematical-
ly in Figure A-l. The sample combinations for analysis are somewhat
different than for Level I.
128
-------
BLANKS
(1 sample
each per
test site)
Probe
and Nozzle
Pre-Clean Liquids
Nitric Acid-LB #1
Distilled-H2O-LB #2
Isopropanol-LB #3
Methylene Chloride-LB
Cy-
clone
10-3 ym
Cy-
clone
3-1 pm
cy-
clone
Filter
XAD-2
Absorber
Post-Wash Liquids
50:50 Methanol:
Methylene Chloride-LB #5
50:50 IPA: Dist. Water-LB #6
#4
TRAIN
1
LS #1
LIQUID
SAMPLES
Condensate
LS#2 LS#3
[Probe &
Nozzle
10 y
Wash
3 pm
Wash
1 pm
Wash
Filter Absorber Cond.
Wash
Wasl1
SOLID
SAMPLES
LB =
SB *
LS =
SS =
SS #1
Impingers
1
Reagent
#1
2
3 4
Reagent
#2
Drier-
ite
LB #7
LB 88
D-2
Tl
_ 6
H
r
^
mola
2° 2
0.2 molar(NH ) S O
NCT
0.02 molar Ag
LS#4
LS#5
LS#6
Liquid
#1
Liquid
Liquid
*3
IXAD-2 (SS#5)
Fllter
Cup
Solids
SS #3' . SS #4
1
J
Liquid Blank
Solid Blank
Liquid Sample
Solid Sample
Discard
Drierite
Possible Combined Blanks
Total Samples - Liquids
Solids
10
6
8
2
Figure A-l. SASS train schematic.
-------
A-2.0 PREPARATION OF XAD-2 RESIN
A-2.1 General Procedure
The XAD-2 resin to be used in the SASS train sorbent trap must
be cleaned prior to use. The resin as obtained from the supplier is
soaked with an aqueous salt solution. This salt solution plus residual
monomer and other trace organics must be removed before the resin can
be used for sampling trace organics.
Clean-up is normally achieved in a giant Soxhlet extractor. Any
other continuous extractor working on the same principle of circulating
distilled solvent would be adequate.
The wet XAD-2 resin is charged into the extractor thimble and
extracted in sequence with refluxing solvent as follows:
1. Water, 20-24 hours
2. Methanol, 20-24 hours
3. Anhydrous ether, 8 hours (during day only)
4. Pentane, 20r24 hours
Methanol is used primarily to remove the water from the resin. Ether
removes a substantial portion of the organics—overnight reflux is
acceptable if apparatus is secure to the hazards of ether. Pentane is
used as the final stage because it is the solvent used to extract the
resin after sample collection.
A commercial giant extractor has a dumping volume of 1500 ml and
thus about 2.5 1 of solvent is required in a 3 1 flask.
After the final pentane extraction, the resin is transferred to
a clean flask and dried under vacuum aided by mild heat from a heat lamp.
Care should be taken (traps) to prevent backstreaming from vacuum systems.
A-2.2 Soxhlet Cleaning of XAD-2
Follow the general procedure given above. However, the follow-
ing procedural details may be helpful to those not familiar with operating
the Soxhlet extractors. These recommendations and comments are based on
our recent experience in preparing XAD-2 for EPA SASS tests.
130
-------
1. Quality of solvents*
Water: Arrowhead distilled
Methanol: Anhydrous methyl alcohol, Mallinckrodt, AR grade
Anhydrous ether: anhydrous (ethyl) ether, Mallinckrodt,
AR grade
Pentane: Mallinckrodt, spectr. AR grade
2. The use of paper (cellulose) thimbles was recommended
by ADL. With a soft lead pencil, mark on the outside
of the thimble the desired fill line which corresponds
to tha entrance level of the syphon tube when the thimble
is inserted into the extractor. Handle the thimble with
plastic gloves. (Glass thimbles are now used in Level I.)
3. Fill (i.e., "charge") the thimble with XAD-2 using a
stainless steel spoon. Intermittently moisten the XAD-2
with distilled water (from a plastic wash bottle) to
compact the XAD-2 in the thimble. .Excess water will
flow through the walls of the thimble. In this manner,
add XAD-2 up to the pencil fill line.
**4. Install the charged thimble in the extractor, place
approximately 300 ml of distilled water in the Soxhlet
flask and assemble the Soxhlet extractor. Room temperature
tap water is adequate for the condenser cooling.
When inserting the charged thimble into the Soxhlet, make
a small indent at the bottom of the thimble to avoid
obstructing the inlet to the syphon tube.
5. Bring the water to a boil and allow the extractor to
syphon several times (one hour of operation is adequate) .
Discard the flask contents, refill with fresh distilled
water and continue the extraction. By discarding the
initial water, most of the salt originally contained in
the raw XAD-2 is removed from the system. This will prevent
salt carryover back into the XAD-2 and will "even out"
the boiling.
6. The methanol solvent should also be replaced in a similar
fashion. This assures complete removal of the water.
(Any water remaining during the ether extraction stage
will "plug" the XAD-2 pores thereby interferring with
the ether extraction.) Three hundred to four hundred
ml of methanol in the extraction flask is adequate for all-
night operation. Use room temperature tap water for the
condenser.
*Mention of trade names does not constitute approval by U.S. EPA.
**Soxhlets actually used were not the giant type.
131
-------
7. For the ether and pentane extraction, a circulating ice
bath should be used for condenser cooling to minimize vapor
loss through the top of the condenser. Three hundred to
four hundred ml of solvent is adequate for all-night pentane
operation. To avoid condensing water (from the air)
on the inside of the condenser during startup, operate
the Soxhlet for several minutes without condenser
cooling (until solvent vapors purge out the air) before
turning on the circulating water.
8. Use extreme caution when handling ether and pentane.
Both are extremely volatile and highly flammable. Make
sure all heating mantles, electrical equipment, etc. are
off while containers are open.
9. The Soxhlet reflux rate can be judged by observing the
drip rate from the condenser onto the XAD-2. One to Two
drops per second is desirable. This is accomplished
by adjusting the power to the heating mantle. For this
condition, the water may be boiled vigorously but no
boiling (bubbling) will be observed for the other three
solvents.
10. When changing over from one solvent to another, residual
solvent remaining in the thimble and extractor should be
removed to as high degree as practical(i.e., do not
desiccate or vacuum dry). One approach which works
quite well is to apply suction to the discharge end of the
Soxhlet syphon tube. The use of a plastic "filtering
pump" (an aspirator pump operated by tap water from the
faucet) has proved adequate.
11. While drying the XAD-2 in the vacuum desiccator, heat to
approximately 120°F using heating lamps. Do not use
vacuum grease on the desiccator. Protect the vacuum pump
from pentane vapors with a carbon trap. The XAD-2 may
be left in the paper thimbles while drying in the
desiccator. Use a filter (i.e., cotton in a flask)
between the carbon trap and the desiccator to prevent
backflow of carbon into the XAD-2 in the event of a
rapid loss of desiccator vacuum.
132
-------
A-3.0 PREPARATION FOR A SAMPLING RUN
A"3.1 Containers, Chemicals, and Laboratory Equipment
Table A-2 lists the samples to be recovered from the SASS train
and the recommended containers used for sample storage and shipping.
In some cases more than one container may be required. All containers
should be cleaned prior to use according to the procedure used for
cleaning the train as described in Section A-2.2.
Laboratory Equipment—
All sample recovery operations, sample weighing, and chemical
cleaning of train components and containers should be performed in a
clean area specially set aside for this work. In the field, this
"clean room" should consist of at least a clean enclosed work bench
or table top and every attempt should be made to observe the following
general recommendations:
1. Avoid drafts and areas with high foot traffic
2. Keep floors swept to minimize air borne dust
3. Use plastic table cloths
4. Inlet filters on air conditioners should be in place
5. Use common sense to avoid contaminating samples with
hair, fingerprints, perspiration, cigarette smoke or
ashes, etc.
6. Use plastic gloves or forceps when handling tared
containers; stainless steel tweezers when handling
filters
In addition to sample containers listed in Table A-2, the
following clean room accessories will be required:
SASS train tool kit
stainless steel tweezers (2)
stainless steel spatulas (2)
disposable plastic gloves
teflon or "guth" unitized wash bottles (3)
teflon policeman (optional)
110°C drying oven
weighing balance 160 gram capacity required
assorted powder and liquid funnels
assorted graduated cylinders, 250 ml to 1000 ml
1/2-gal mixing jugs (3)
133
-------
TABLE A-2. SAMPLE STORAGE/SHIPPING CONTAINERS
Train Component
Sample Type
Container Required*
Probe and nozzle
lOy cyclone
3y cyclone
ly cyclone
Filter holder and
filter
XAD-2 Module:
(1) XAD-2 resin
(2) Condensate
(3) All surfaces
Irapinger #1
Impinger #2
Impinger #3
solid tappings
solvent wash
cup solids
solvent wash
cup solids
solvent wash
cup solids
solvent wash
solid tappings and
filter
solvent wash
solid adsorbent
contents of
condensate cup
solvent wash
contents
rinses
contents
rinses
contents
rinses
Tared 4 02. LPE
500 ml amber glass (16 oz)
Add to probe and nozzle tappings,
Add to probe and nozzle wash.
Tared 4 pz. LPE
500 ml amber glass (16 oz)
Tared 4 oz. LPE
500 ml amber glass (16 oz)
Tared 150 mm glass petri dish
500 ml amber glass (16 oz)
500 ml amber glass (wide
mouth) (16 oz)
1 liter LPE
500 ml amber glass (16 oz)
1 liter LPE, with pressure
relief cap
500 ml amber glass (16 oz)
1 liter LPE
500 ml amber glass (16 oz)
1 liter LPE
500 ml amber glass (16 oz)
*A11 glass containers must have teflon cap liners.
^Linear polyethylene (same as "high density" or "type 3" polyethylene)
Additional sample bottles must be provided for all fuel, process
materials, and ashes to be collected. For train washes and liquids,
particularly the condensate, several bottles may be required.
134
-------
Quality of Chemicals—
An underlying concern in selecting chemicals for impinger solu-
tion and washes is to avoid introducing trace compounds similar to those
being analyzed. Although "blanks" of impinger solutions will be analyzed,
it is preferable to minimize chemical impurities when possible by using
highest quality chemicals rather than adjust the final sample analyses
results. The following chemical grades were used:
Chemical
Impinger Solution:
distilled water
ammonium per (oxydi) sulfate
0. IN silver nitrate
30% hydrogen peroxide
Train Precleaning:
distilled water
isopropyl alcohol
[CH CH(OH)CH3J
methylene chloride (CH
Sample Recovery:
distilled water
methylene chloride
methanol (CH^OH)
isopropyl alcohol
[CH3CH(OH)CH3J
Quality
Commercial
distilled
AR
AR
Commercial
distilled
Spectr AR
Spectr AR
commercial
distilled
Spectr AR
Spectr AR
Spectr AR
If higher grade (lower impurity levels) of chemicals are available they
should be used.
135
-------
A-3.2 Cleaning the SASS Train
Newly purchased or previously unused train components and sample
containers should be washed with tap water and a plastic scouring pad.
All surfaces in the sampling train which come in contact with sample,
as well as all sample containers and impingers, should be prepassivated
by one-hour standing contact with a 50:50% volume solution of pure nitric
acid and distilled water. Remove any remaining traces of acid by rinsing
with tap water, then continue with the solvent cleaning procedure below.
Prior to sampling, all SASS train components and sample con-
tainers are cleaned in two or three successive stages (in the order
listed) using a different solvent in each stage:
All Except Impinger Sample Bottles Impinger Sample Bottles
1. distilled water 1. distilled water
2. isopropyl alcohol 2 isopropyl alcohol
3. methylene chloride (CH Cl )
The distilled water may be dispensed in plastic wash bottles; the iso-
propyl alcohol and CH Cl should be dispensed using teflon or glass wash
bottles. After each part is washed with CH Cl.,, it should be dried in
£ £t
a filtered stream of dry air or nitrogen.
Any solid residues adhering to the internal surfaces should be
removed with tap water and a plastic scouring pad before preceding with
the solvent cleaning procedure.
After cleaning, assemble and cap off the cyclone assembly.
(All caps should be previously cleaned according to the above 3-solvent
procedure.) Cap off other sections of the train including the probe,
XAD-2 module, filter housing, impinger trains, and interconnecting
hoses.
136
-------
A-3.3 Impinger Solutions
Impinger
Reagent
Quantity
Purpose
6M
750 ml
0.2M (NH4) SO 750 ml
i 2 ^ 8
+ 0.02 M AgNO3
Trap reducing gases such as
S02 to prevent depletion of
oxidative capability of trace
element collecting impingers
2 and 3
Collection of volatile trace
elements by oxidative disso-
lution.
0.2M (NH4) S 00 750 ml
Z 2 a
+ 0.02 M AgNO
* Drierite (color 750 g
indicating)
Collection of volatile trace
elements by oxidative disso-
lution.
Prevent moisture from
reaching pumps
Suggested Formulas for Preparing Impinger Solutions —
Impinger #1 (6M H,.O )
To prepare 750 ml of 6M HO dilute 465 ml of standard
-------
Impinger #4 (color indicating Drierite)
Use 750 gm or approximately 750 cc of 8 mesh color indicating
Drierite (CaSO )
When installing the top on the impinger bottle, avoid forcing
Drierite up into the center tube as this results in increased
pressure drop. Lay impinger on side while inserting top.
It may be necessary to replace the Drierite several times
during a SASS run. A marked decrease in Impinger #4 outlet
temperature (moisture absorption by Drierite produces heat)
may signal Drierite depletion if the Drierite color change
is difficult to detect.
The spent Drierite is not kept for analysis and can be dis-
carded or, preferably, rejuvenated for future use by heating
in a drying oven at 220°F to 250°F to blueness.
A-3.4 Filter Preparation
More than one filter will be required when particulate grain
loading is high (i.e., pulverized coal units, cement kilns, etc.).
Using stainless steel tweezers, place each filter in a clean, numbered
150 mm glass petri dish. Bake at 220°F for at least three hours in a
drying oven, then immediately transfer to a desiccator to cool.
Weigh the petri dish (plus filter). Weigh a second time,
preferably several hours later, to confirm the initial weighing. This
is the tare weight used to determine the mass particulate catch on the
filter.
The type of filter used is a Gelman type A/E binderless glass
fiber filter (142 mm diameter), purchased through Scientific Products.
A- 3.5 SASS Train Assembly
Transport each separate train component to the sample port
area with all sealing caps in place. When removing caps for connection
of components, make certain no foreign matter enters the components.
If the ambient dust level is high, the train should be covered with
plastic drop cloths. Before installing the probe nozzle and with the
probe capped, turn on the vacuum pump and leak check the system. Leakage
rate should be held to 0.05 cfm at 20 "Kg pump sruction. Avoid over-
tightening fittings and clamps.
138
-------
A-3.6 SASS Chemical "Blanks"
a. Blanks from impingers #2 and #3 should be prepared in the
field with the same distilled water used in preparing the
impinger solution. To prepare a 1000 ml blank, mix the
following ingredients and dilute to 1000 ml with distilled
water:
1. 45.7 gin crystalline (NH ) SO
2. 200 ml 0.1 N A NO
g 3
b. Blanks of impinger #1 can be prepared in the field
with the same HO and distilled water used for the
impinger solution.
c. Blanks of the wash solutions should be obtained in the
field (i.e., IPA, 50:50 meth. chlor. - methanol, H 0).
d. At least one filter blank should be processed in the
same manner as sample filters; one blank per test
site.
e. At least one blank sample of the XAD-2 resin should be
preserved for each test site.
139
-------
A-4.0 SAMPLING PROCEDURES
The SASS train is basically a high volume Method 5 system modi-
fied to collect trace metal and organic compounds which would normally
pass through the standard Method 5 train. The major design differences
apparent in Figure A-l are the XAD-2 adsorbent module, multiple cyclone
assembly, and new impinger solutions. The SASS train is operated in
much the same fashion as a Method 5 train, but there are a number of
modifications as discussed below.
A-4.1 Sample Flow and Isokinetic Conditions
To preserve the cyclone "cut-off" points, the sampling flow
rate is adjusted to maintain close to 4.0 awcfm (actual wet cubic feet
per minute) at the required 400°F cyclone oven temperature conditions.
Since isokinetic sampling is also still required, both these constraints
are satisfied to as high a degree as possible by selecting the optimum
probe nozzle diameter.
After stack velocities, temperatures, and oxygen levels are
established by the preliminary stack traverse, the nomogram, Section A-5,
may be used to select the proper nozzle diameter and "rough in" the
required sampling rate (but see STEP 5 below). However, if stack con-
ditions are encountered that are not covered by the nomogram, the
following computational procedure may be used for each sampling location.
EQUATIONS:
0.1192 /T /V (1)
s s
281.4 (Vs)(d2)/Ts (2a)
V = ly Ughl/L^Bl.4 (d )] (2b)
Qcy(V860)E1 " (%H20/100)] (3)
140
-------
d - nozzle diameter (inches)
T = stack temperature (°R)
5
V = stack velocity (ft/sec)
s
Q = sample flow rate at cyclones (awcfm)
cy
O = sample flow rate at meter (adcfm)
T = meter temperature (°R)
%H 0 = sample moisture content (% by volume)
These equations are valid only when an oven (cyclone) temperature of
400°F is maintained and when the pressure of the stack and dry test
meter are roughly the same (i.e., +_ 1" Hg) .
STEP 1:
Select the nozzle size closest to the value computed from
Equation (1). Use this value in the following step:
Fractions of inch (nozzle diameter)
00 "3* 00 H CN rH CO X 'ff
X X \XX\\H\
rH rH fl f- rH LT> H CO
I ! i t i i i i i
I I I I T^ 1
n ^ in \D r^ oo
o o o o o o
Decimal Inches
STEP 2:
Compare the cyclone flow rate from Equation (2a) to the desired
rate of 4.0. If the values compare to within +_ 10%, proceed to next
step. Otherwise, calculate a stack' velocity from Equation (2b) using
a value for Q which is within 10% of 4.0 [i.e., use 3.6 or 4.4,
*cy
whichever is closest to the value obtained from Equation (2a)]. This
calculated stack velocity should be within 10% of the actual stack
velocity. If not, stack conditions are very unusual and greater than
10% "tolerances" are necessary (i.e., deviations from isokinetic condi-
tions a/o deviations from 4.0 cfm conditions at the cyclone will be
necessary) .
141
-------
STEP 3:
Calculate the meter flow rate from Equation (3) using the cyclone
flow determined in the previous step.
STEP 4:
Determine the approximate orifice AH corresponding to the flow
rate from the previous step. Use the nomogram plot of AH versus flow
rate determined experimentally for the particular control box and orifice.
This is based on the mid-size orifice of the three in the control box.
STEP 5:
The value of AH determined in the previous step (or from the
nomograph) will be adequate to "rough in" the flow rate when the SASS
train is first turned on. However, as soon as possible/ obtain more
accurate settings using the actual measured meter temperature and the
actual meter flow rate obtained from the meter readout and a stopwatch.
£-4.2 Organic Adsorber Module Operation
When the XAD-2 module is operated "cold" to condense moisture
from the sample, the following procedure may be used to transfer conden-
sate from the condensate cup at the base of the module to the condensate
collection flask. This is necessary to avoid overfilling the condensate
cup which would result in condensate carryover into the impingers.
This transfer procedure can be accomplished without interrupt-
ing the sampling. The procedure should be performed frequently at the
start of a test until the actual condensate collection rate is established.
STEP 1:
Inspect the condensate collection flask and interconnecting tube
to confirm that all fittings are tight.
STEP 2:
Partially close off the large (1/2-inch) ball valve at the inlet
to the XAD-2 module until the vacuum gage on the pump increases by about
2 in. of mercury.
142
-------
STEP 3:
Open the condensate drain valve at the bottom of the module.
Since the collecting flask is initially at a higher pressure than
the inside of the module, air will flow from the flask into the
module (bubbling through the collected condensate) until pressures are
equalized.
STEP 4:
After a few seconds to allow the equilibration of pressures,
open the 1/2-inch ball valve. This raises the pressure in the module
relative to the collection flask, forcing any condensate into the
bottle.
STEP 5:
After all the condensate has been transferred, close the drain
valve.
A-4.3 Drierite
See Section A-2.3 for comments on Drierite depletion and
renewal (Impinger #4).
A-4.4 Filter Changes
When sampling combustion effluents with high particulate loading,
plugging of the filter may occur before adequate sample volume is obtained.
In this event, it will be necessary to shut the train down and install a
new filter.
The rate of filter plugging is evident by the gradual increase
in sample pump vacuum required to maintain sample flow. To minimize
filter changes, the train may be operated with pump vacuums of 15 to
20 "Hg or until desired sample flow cannot be maintained.
143
-------
A- 5.0 TRAIN DISASSEMBLY AND SAMPLE RECOVERY
1. After turning off train and withdrawing probe from stack, open
the cyclone oven to expedite cooling (turn oven cooling fan on)
2. Disconnect probe and cap off both probe ends and inlet to 10y
cyclone.
3. Disconnect the line joining the cyclone oven to the XAD-"2
module at the exit side of the filter and cap off the filter
holder exit and the entrance to the joining line which was
disconnected from the filter holder exit point.
4. Disconnect the line joining the XAD-2 module to the impinger
system at"the point where it exits the XAD-2 module. Cap off
the exit of the XAD-2 module and the ei trance to the joining
line leading to the impinger system.
5. Disconnect the line exiting the Drierite impinger at the
point where it leaves the impinger and cap off the impinger
exit. Discard ice and water from the impinger box to facili-
tate carrying.
6. Carry the probe, cyclone-filter assembly, XAD-2 module (plus
joining line and condensate collection flask) and impinger
train (plus joining line) to the clean room for sample
recovery. Before entering the clean room, clean off all loose
particles from the exterior surfaces of the train components
using compressed air, brushes, etc.
7. Procedure for transferring samples from the various portions
of the SASS train into storage containers is outlined in the
flow diagrams on Figures A-2, A-3, and A-4. Place copies of
these diagrams in an easily visible location in the clean room
for quick reference during the sample recovery and transfer
operations.
144
-------
Figure A-2. SASS train seunple recovery — probe, cyclones, filter, XAD-2 module.
Probe Nozzle
Step Is Hold probe vertically
(nozzle end down) and tap vigor-
ously to clear loose solids from
fittings and drive them into
nozzle.
Step 2i Disconnect nozzle from
probe and tap loose solids into
tared nalgene container.
Step 3: Rinse adhered material
into astoer glass container.
Add to 10 Mm
cyclone solids
Add to probe rinse
Probe
-*»
Rinse into nozzle wash
container-
10 pm Cyclone
_**
Step Is Remove filter housing
from cyclone assembly, cap off
filter housing inlet and 1 p»
cyclone outlet, and aet filter
housing aside.
Step 2: Briefly tap cyclone
assembly to clear solids froos
fittings.
Step 3: Disconnect 10pm cyclone
from cyclone assembly and cap off
10pm cyclone outlet and 3 pm
cyclone inlet. Vigorously tap
10pm cyclone to drive solids into
lower cup.
Step 4: Reconnect cyclone cup
assembly (with vanes) , remove
cyclone top and rinse top into
lower sections of cyclone.
Step 5: Rinse cyclone center I
section into cut. assembly. j
probe/nozzle catch be combined with
the 10 pra cyclone. Rather, Level I
specifies that the 10 pn and 3 pra
cyclone catches should be combined.
Remove cup, lift out
vanes with stainless
steel tweezers and
transfer cup contents
into nozzle tappings
container.
Remove cup assembly, rinse
vanes into cup and transfer
cup contents into nozzle-
fCoat
\ *
[Comb
\ •
s
/ • ' '
>ine)
(Continued)
-------
Figure A-2 (continued). SASS train sample recovery — probe, cyclones, filter, XAD-2 module.
en
Step 1: Briefly tap cyclone
assembly to clear solids from
3pm - 1pm cyclone connecting
fitting.
Step 2: Disconnect 3 pm cyclone
from 1pm and cap off 3pm cyclone
outlet and 1 pm cyclone inlet.
Vigorously tap 3 pm cyclone to
drive solids into lower cap.
Step 3: Reconnect cyclone cap
assembly, remove cyclone top
portion and rinse top portion
of cyclone into lower sections
of cyclone.
Step 4: Rinse cyclone center
section into cup assembly.
Step 1: Vigorously tap cyclone
to drive solids into lower cup.
Step 2: Disconnect upper portions
of cyclone and rinse them and the
cup into amber glass container.
Step 1: Open up filter housing,
remove filter using a stainless
steel tweezers and place filter
(particulate side down) in a
covered tared 150 mm glass petri
dish. Any appreciable solids
adhered onto the filter housing
may be tapped into the petri dish
(i.e. lift edge of the filter,
tap solids into bottom of petri dish
and then cover over with filter),
Remove cup assembly, lift
out vanes with stainless
steel tweezers and trans-
fer contents of cup into a
tared nalgene container.
Remove cup assembly, rinse
vanes into cup and transfer
contents of cup into amber
glass container.
Disconnect cup and transfer
conten'-s Into a tared
nalgane container.
step 2: Rinse both halves of
particulate housing (including
interconnect tubing attached)
into amber glass container.
MOTES:
1. Use 50:50 CH2C12 and CH3OH for
all rinses (use teflon wash
bottles or Guth unitized wash
bottles).
2. Handle all tared containers with
gloves.
3. Transfer of solids may be assisted
by the use of stainless steel
spatulas and powder funnels. Nylon
bristle brushes may also be used
if necessary.
4. All nalgene containers must be
high density polyethylene.
-------
SASS TRAIN SAMPLE RECOVERY — XAD-2 MODULE
STEP NO. 1, XAD-2
AND CONDENSATE REMOVAL
Release clamp joining XAD-2 cartridge
section to the upper gas conditioning
section.
Remove XAD-2 cartridge from cartridge
holder. Remove fine mesh screen from
top of cartridge. Eipty resin into
wide mouth glass amber jar.
Open condensate cup valve, raise con-
densate cup above the condensate
collection bottle and flow condensate
from the condensate cup into the
collection bottle through the Teflon
tube.
Unscrew Teflon tube from collection
h-ittle anc. cap off collection bottle.
ne condensate is sent to the labora-
-cory in this bottle.
Disconnect the Teflon tube at the
condensate cup valve. Rinse Teflon
tube into amber glass bottle.
Install new collection bottle and
connect Teflon tube at the bottle.
Replace screen on canister, reinsert
canister into module. Join module
back together and replace clamp.
STEP NO. 2, XAD-2
MODULE RINSE
Close condensate cup drain valve.
Release upper clamp and lift out inner
well.
Rinse inner well surface into and alone
condenser wall so that rinse runs ^ -™
through the module and into condensate
cup.
When inner well is clean, place to one
side.
Rinse braided entrance tube into moeJ'jle
interior. Rinse down the condenser all
and allow solvent to f lo™ ;«m through th
system and collect in condensate cup.
Release central clamp and separate the
lower sections (XAD-2 and condensate
cup) from the upper section (condenser)
The entire upper section is now clean.
Rinse the now empty XAD-2 canister into
the condensate cup. Remove canister and
place in a clean container. Rinse walls
of XAD-2 section into condensate cup.
Release lower clamp and remove XAD-2
section from condensate cup.
NOTE: USE 50:50 CH.Cl and CH3OH
FOR ALL RINSES.
Figure A-3.
The condensate cup now contains all
rinses from the module. Drain into the
amber glass bottle (via drain valve!
containing the Teflon tube rinse.
Rinse condensate collection flask and
Teflon connecting tubing into the above
amber glass bottle.
Assemble complete module and reconnect
Teflon tube at condensate cup valve.
147
-------
SASS TRAIN SAMPLE RECOVERY - WINGERS
Rinse From
Connecting Line
Leading From XAD-2
Mod to First Impinger
Step No. 1
Impinger No. 1
Impinger
Liquid
Step NO. 2
Impinger No. 2
CD
Step No. 3
*Rinse From
Impinger Bottle
And Tubing
Impinger
Liquid
Impinger No. 3
*
*Rinse From
•Impinger Bottle
And Tubing
Impinger
Liquid
*Rinse From
Impinger Bottle
And Tubing
*NOTE: ALL RINSES ARE
(1) ISOPROPYL ALCOHOL (FIRST)
(2) DISTILLED WATER (SECOND)
IPA AND WATER RINSES SHOULD
BE PLACED IN SEPARATE BOTTLES
Step. No. 4
Impinger No. 4
Drierite
Discard
"Drierite
Figure A-4.
-------
A-6.0 SUPPLEMENTARY REFERENCE MATERIAL
Physical Properties of SASS Chemicals, Figure A-5
Physical Properties of XAD-2, Table A-3
SASS Train Nomogram, Figure A.-6
Miscellaneous Data, Table A-4
149
-------
PHYSICAL CONSTANTS OF ORGANIC COMPOUNDS
No.
Name
Synonyms and Formula
Mol.
*rt.
Color,
crystalline
form,
specific rota lion
and i_. (log it
m.p.
•c
b.p.
•C
Density
"D
Solubility
w
al
eth
ace
bz
other
solvents
Ref.
Methane
Qm252l—.dlcMoro-* |Methylenechloride.CH,a,. | 84.93 |i"« <200
f-95.1 |40'« ||.3266J' |l.4242» | t \ -181 (2.79) -89.5 82.4'" 0.78JSj« 1.3776"
I
oo o= QD s v Bl',1439
Mefhanol>
Carbinol. Methyl alcohol. I 32.04 L'" 183.3 (2.18)1 -93.9
Wood alcohol. CHjOH |
64.96"
IS"
•r-i
0.79l4i° 1.3288" oo oc CD
-------
TABLE A-3. XAD-2 RESIN
XAD-2 is available from:
Fluid Process Department
Rohn and Haas
Philadelphia, Pa.
A contact for questions is:
Mr. Charles Dickert
(215) 592-3000
The material is a styrene/divinylbenzene copolymer and the material is
supplies wet with a salt solution.
Some relevant parameters are:
mesh range:
surface area:
avg. pore dia.:
specific density:
bulk density:
pore volume:
20-50
300-350 n//g
90A
1.02 g/cc
0.4 g/cc
0.85 cc/g
Costs vere $96.50/cu. ft.
Property
Appearance
Solids
Porosity (ml.pore/ml.bead-dry basis)
Surface Area (m.2/g.-dry basis)
Effective Size (mm.)
Harmonic Mean Particle Size (mm.)
Average Pore Diameter (A -dry basis)
True Wet Density indistilled water (g./ml.)
Skeletal Density (g./ml.)
Bulk Density (lbs./ft.3)
(g./cc.)
Amberlite XAD-2
Hard, Spherical
opaque beads.
51 to 55
0.40 to 0.45
330
0.30 to 0.45
0.45 to 0.60
90
1.02
1.07
40 to 44
0.64 to 0.70
•
151
-------
I ' J/8 " no*»l« di»»t«r i
. (lnc>M>)
Figure A-6. SASS Operating nomogram.
-------
Figure A-6. (Continued) SASS operating nomogram.
-------
TABLE A-4. MISCELLANEOUS DATA
Cyclone cup capacities: 3 ym-and 10 ym = 370 cc; 1 ym = 20 cc
XAD-2 canister volume = 402 cc
S-type pitot tube factor = 0.85 +_ 0.2
Screen for XAD-2 canister:
316 stainless steel
80 mesh x 0.055 wire diameter
Purchase from:
Cambridge Wire Cloth Co.
3219 Glendale Blvd
Los Angeles, California
(213) 660-0600
Condensate container volume = 700 cc
XAD-2 module temperature = 68°F (20°C)
154
-------
A-7.0 SAMPLE PREPARATION AND ANALYSIS (Calspan Corp.)
Samples were analyzed by Calspan Corp., Buffalo, NY, by atomic
absorption, gas chromatography and wet chemistry. Spark source mass
spectrographic (SSMS) analyses were performed by Commercial Testing and
Engineering, Golden, CO, as a subcontract to Calspan1s work. Calspan and
CTE analyzed preselected samples that include base samples, blanks, and
duplicates. Additional samples were submitted to Battelle Memorial Institute,
Columbus, OH for analysis of POM by gas chromatograhpy/mass
spectrometry.
A-7.1 Sample Size
The sample size required for analysis is dependent on how much
sample can be obtained from the SASS train. Table A-5 lists the detec-
tion limit and sensitivity for all sample components to be analyzed.
For metal analysis, 200 ml of impinger liquids are necessary. For solid
samples, 4 to 5 grams are necessary. Analysis for chloride, fluoride,
sulfate, and nitrate requires up to 200 ml of liquid sample and 5 grams
of solid sample. PCB and POM analysis requires 10 to 50 grams of solids
and as much liquid as can be obtained (> 500 ml) . Additional sample is
required for SSMS analysis.
The sample amounts given are desired amounts. Analysis can be
achieved on much smaller samples but with a sacrifice in detection
capability for desired components. The detection of individual compo-
nents, however, cannot be greater than the detection limits and sensi-
tivities given in the table. Detection limits may also be higher for certain
types of sample matrix.
A-7.2 Sample Preparation
Analysis of SASS train samples involves pretreatment of the
samples after collection to place them in a form suitable for chemical
analysis. Atomic absorption requires that each sample be predissolved
or be in the liquid phase. The technique for solubilization of the
155
-------
TABLE A-5. DETECTION LIMITS AND SENSITIVITY VALUES
Pollutant
Antimony
Arsenic
Barium
Beryllium
Cadmium
Calcium
Chromium
Cobalt
Copper
Iron
Lead
Manganese
Mercury
Nickel
Selenium
Tellurium
Tin
Titanium
Vanadium
Zinc
Chloride
Fluoride
PCB
POM
Sulfates
Nitrates
Detectipn Limit
Solids
(yg/g) *
10
0.10
1.5
0.25
0.10
0.15
1.0
1.5
0.5
1.0
2.5
0.5
0.01
1.0
0.1
10
5
15
10
0-25
5.0
5
0.1
50
50
5
Liquids
(yg/ml)
0.2
0.002
0.03
0.005
0.002
0.003
0.02
0.03
0.01
0.02
0.05
0.01
0.0002
0.02
0.002
0.2
0.1
0.3
0.2
0.005
1
0.1
0.002
0.1
1
0.1
Sensitivity
Solids
(yg/g)*
25
0.5
20
1.25
1.25
4
5
10
5
5
25
2.5
0.05
7.5
0.1
50
200
100
40
1
250
12.5
0.50
100
250
10
Liquids
(yg/ml)
0.5
0.01
0.4
0.025
0.025
0.08
0.1
0.2
0.1
0.1
0.5
0.05
0.001
0.15
0.002
1
4
2
0.8
0.02
5
0.25
0.01
2
5
0.2
*Values given are for 1 gram of material dissolved in 50 ml of solution.
156
-------
metals is based on methods utilized by'the National Bureau of Standards *
(Ref. A-2) for solubilizing both highly organic materials such as coal
and inorganic materials such as fly ash prior to sample analysis. The
outlined techniques allow for wet chemical ashing of material that
prevents loss of volatile elements like mercury, arsenic, and selenium.
The methods given use concentrated minerals acids, as well as a strong
oxidizing acid, perchloric (HC1C>3) , to decompose organic materials.
One gram of highly organic material (coal, tar residue, fuel
oil, etc.) is transferred to a Teflon beaker. The sample is slowly
digested for several hours in 25 ml of NHO and cooled. A mixture of
5 ml of HF and 10 ml of HC1O4 is added and the digestion is continued
at low heat. Extreme care is necessary, for excessive temperatures
can cause decomposition and explosion. Digestion is continued until
all carbonaceous material has been destroyed. The solution is then
transferred to a 50 ml volumetric flask and diluted to a calibrated
volume.
Samples not as highly organic (fly ash, bottom ash, cement kiln
dust, etc.) are to be accurately weighed to one gram in a Teflon beaker.
A mixture of 5 ml of HNO and 5 ml of HF is added. The beaker is
covered and the sample digested for one hour. After complete cooling,
10 ml of HC1O is added and the digestion is continued until all carbo-
naceous material has been destroyed. The cover is then removed and
the sample evaporated to dryness and baked until the solids turn brown
around the edges. A mixture of 2 ml of HC1 and 35 ml of distilled water
is added and the solution heated slightly until all solids dissolve.
The solution is then transferred to a 50 ml volumetric flask and diluted
to a calibrated volume.
Liquid samples from the SASS train are stabilized with 1 ml
of concentrated nitric acid to every 200 ml of impinger liquid. Whenever
possible liquids are concentrated by boiling to one-half their received
volume to concentrate trace elements.
*EPA Level I specifies the use of Parr bombing to avoid loss of volatiles.
157
-------
Both predissolved and concentrated liquids are analyzed
using atomic absorption spectroscopy using the most sensitive aspiration
techniques available. Analysis for both PCB and POM will involve extrac-
tion and concentration prior to analysis. The PCB and POM are coextracted
by liquid-liquid or liquid-solid extraction.
Solid samples (^ 50 grams) are extracted with pentane using
a Soxhlet extractor. The extract is concentrated using a Kuderna-
Danish evaporator to reduce the extract volume to 10 ml. Aliquots 2 to .
5 yl are injected directly into a gas chromatograph for PCB and
POM analysis after liquid-solid column separation and clean-up.
Both POM and PCB after extraction with pentane are isolated
as a class using adsorption chromatography by a technique called the
Rosen separation (Refs. A-3 and A-4). This technique entails adsorption
of the total sample on a silica gel column. The initial effluent from
the column when washed with pentane will contain an aliphatic hydrocarbon
fraction. The aromatic hydrocarbon fraction is eluted with benzene.
The benzene fraction which contains all POM and PCB is analyzed
using gas chromatography employing FID and EC detectors.
Detection and measurement of POM and PCB are accomplished by
using a gas chromatograph employing a flame ionization detector (FID)
and an electron capture (EC) detector equipped with Ni-63 source.
Confirmation is performed by comparing to POM standards and PCB standards
of known concentration and literature relative retention time data.
A- 7.3 Analysis Procedures
Analysis for chlorine, fluorine, nitrates, and sulfates all
involve wet chemical processing prior to actual measurement. Since all
chlorides, nitrates and most sulfates are water soluble, they can be
extracted from solid samples using a Soxhlet extractor. The extraction
scheme to be used has been effectively used by the Bay Area Air Pollution
Control District, San Francisco (Refs. A-5 and A-6).
158
-------
Fluorides, however, are not sufficiently soluble to allow for
effective aqueous extraction. Solid samples are fused with sodium
hydroxide to convert all fluorides to soluble sodium fluoride. The fused
melt is dissolved in 4 M HC1 and the resulting liquid analyzed as a
soluble fluoride.
Liquid samples analyzed for chlorine, fluorine, nitrates,
and sulfates are analyzed directly by techniques specific for each
anion.
Solubilized chloride is analyzed by adding dilute mercuric
nitrate solution to an acidified sample in the presence of mixed diphenyl-
carbozone-bromophenol blue indicator. The end point of the titration is
the formation of a blue-violet mercury, diphenylcarbozone complex (Ref.
A-7).
An alternative method involves direct measurement of chloride
with a specific ion electrode. Both methods are used and checked
to obtain the best sensitivity on the submitted samples.
Analysis for fluoride in liquid samples or solubilized fusion
products is performed by prior Bellack Distillation to remove
interfering substances. After distillation, the fluoride is deter-
mined potentiometrically using a selective ion fluoride electrode
(Ref. A-8) .
The analysis for nitrate is based upon the reaction of the nitrate
ion with brucine sulfate in a 13N H SO solution at 100 °C. The color of
the resulting complex is measured at 410 nm (Ref. A-8) .
Sulfate analysis is performed by converting sulfate ion to
barium sulfate suspension under controlled conditions. The resulting
turbidity is determined on a spectrophotometer and compared to a curve
prepared from standard sulfate solutions (Ref. A-8) .
159
-------
Metal analyses are performed on liquid and solid samples
after pretreatment and solubilization of materials as outlined earlier.
A Perkin-Elmer Model 460 atomic absorption spectrometer with microcom-
puter electronics is used in conjunction with conventional aspira-
tion and time integration techniques. The Model 460 is a relatively new,
highly sensitive instrument that allows accurate measurement of metal
concentrations. In addition, conventional hollow cathode source - lamps,
electrodeless discharge lamps (EDL), are available for lead, mercury,
arsenic, and selenium. These special lamps are more stable and provide
for more initial energy to allow accurate detection of difficult-to-
analyze elements.
Mercury is analyzed by the cold vapor technique developed
by Hatch and Ott (Ref. A-9). Arsenic and selenium are to be analyzed by
conversion of these elements with hydrogen to arsenic hydride and selenium
hydride vapor. Each of the vapor techniques allows for low-level detec-
tion and quantization for each of these elements.
A listing of the detection limits and sensitivity for each element
in liquid and solid samples is given in Table A-6. In the table, detection
limit is defined as the concentration that produces a signal equivalent to
twice the magnitude of the background. Sensitivity is defined as the con-
centration in micrograms per milliliter of solution to produce a one
percent change in absorption or one percent change in the recording chart
readout.
The detection limits for solid samples are based on a
one gram sample dissolved or extracted into 50 ml volumes of analysis
solution. Each value given is conservative and is based on the possi-
bility of interference between components present. If the sample is
relatively "clean", i.e., no interfering or high background substances,
detection limits may be even lower.
160
-------
Polychlorinated biphenyls (PCB) and polycyclic organic materials
(POM) are analyzed by Calspan using a Hewlett-Packard Model 5700 gas chroma-
tograph equipped with a flame ionization and an electron capture detector.
The electron capture detector contains a radioactive source, Ni-63, and
is highly sensitive to chlorinated and highly conjugated organic compounds.
The flame ionization detector is sensitive to all hydrocarbons. The gas
chromatographic column used in separation of components is four feet
long, packed with a substrate coated with 2.5% by weight of a liquid
crystal.
The analysis column used is the one suggested by Janini (Ref.
A-10) specifically for POM separations. Gas chromatographic column para-
meters are summarized, below:
Column length: 4' x 1/8" OD
Column material: Stainless steel
Stationary phase: 2.5% BMBT*
Support: Chromosorb W HP, 100/120 mesh
Flow: 40 ml/min helium
Temperature: 235°C, isothermal
*N, N-bis [p-methoxybenzylidene]- a, a' - bi-p-toluidine
It should be noted that the gas chromatograph is operated in the
isothermal temperature mode. This is necessary due to the extreme tem-
perature sensitivity of the electron capture detector. Any attempt to
temperature program would result in a gross baseline drift.
Alternate chromatographic methodology and retention time data
has been obtained from an analytical method of Gouw, et al. (Ref. A-ll)
and Lao, et al. (Ref. A-12) . Literature column retention time data is
available for all the desired POM listed in the request for proposal
with the exception of the dibenzo[c,g]carbazole.
161
-------
Four of the eight POM are commercially available and are used in
fixing retention times and in calibrating the instrument response factors for
the various components. The 7,12 dimethylbenz[a]anthracene, benzo[a]pyrene,
dibenz[a,b]anthracene, and 3-methylcolanthrene POM are obtained from the
Eastman Kodak Company in the pure form. The other POM listed are not avail-
able from any commercial source known, so literature relative retention time
data of the other POM is utilized to fulfill analyses requirements.
The quantitization of total POM is made by taking the total
area of all POM and reporting the response area as if it were 9-methylanthra-
cene (C _H , Mol. Wt. 192.26). If PCB is found to be present, the concen-
«L 3 !-•*-
tration is subtracted from the total hydrocarbon response area. The standards
used in measuring PCB response and retention times are known (Aroclor)
standards. The eight individual POM specifically required for identification
are analyzed separately, and reported as such. The eight materials are also
included in the total POM reported values.
A-7.4 Quality Control
Quality control is maintained by two principal modes. Throughout
this study a number of samples are analyzed in duplicate to assure precision
of results. More importantly, however, carefully prepared analytical
standards and blanks are utilized in preparing suitable calibration curves,
thereby assuring accurate measurement of data. To test the accuracy, known
additions are made to samples that can be obtained in large enough quantity
to test for quantitative recoveries.
162
-------
A-8.0 SAMPLE PREPARATION AND ANALYSIS (Battelle)
Selected samples provided by Calspan as directed by KVB were analyzed
for polycyclic organic materials (POM) and polychlorinated biphenyls (PCS)
by Battelle Columbus Laboratories to more positively quantify specific
compounds.
A-8.1 Sample Extraction and Concentration
XAD-2 samples are Soxhlet extracted with pentane for 24 hours.
Liquid samples, except benzene extracts, are liquid-liquid extracted with
methylene chloride and dried over magnesium sulfate. If water is noted in
the pentane extract of the XAD-2 sample, that sample is also dried over
magnesium sulfate. At this point, samples are split in half for PCB and POM
analyses, if both are required. All POM samples are spiked with an internal
standard and then concentrated by use of a rotary evaporator and then a
Kuderna-Danish concentrator.
A-8.2 Sample Separation
All samples for POM analysis are separated by liquid chromatography
on 100-200 mesh Silica gel using a stepwise gradient elution. The POM
fractions are eluted with 20 percent methylene chloride in petroleum ether.
All other fractions are discarded. The samples are concentrated by a Kuderna-
Danish concentrator to a volume of approximately 0.5 ml.
A-8.3 POM Analyses
Analysis for all detectable POM species from anthracene through
coronene is carried out by capillary gas chromatography-mass spectroscopy.
A 30 m SP-2100 glass capillary column is programmed from 150 °C - 270 °C at
2°C/min and held at 270 °C for the remainder of the run. The carrier gas
(helium) flow is 20 cm/sec. All injections are splitless, using the Grob
technique. The Finnigan 3200 mass spectrometer is operated in the chemical
ionization mode using methane as thereagent gas.
Quantitative GC-MS analysis is accomplished by specific ion monitoring.
Total POM and concentrations of the specified POMs are reported. All data is
normalized to represent the entire sample. The detection limit is 0.1
nanogram.
163
-------
Qualitative identification is assured by the elution order and isotope
patterns of compounds of interest. Since specific ion monitoring is used,
almost any potentially interferring species which may be carried through the
LC separation scheme is eliminated duirng mass spectral analysis. Retention
times (relative to internal standards) and elution order of isotopes are
determined by comparison with POM standards.
A-8.4 PCS Analysis
Extracts are placed on previously standardized Florisil to separate
the PCBs from possible contaminates. GC analysis is carried out using a
6 ft x 4 mm ID column packed with 11 percent OV-17 plus QF-1 and an electron
capture detector. Sample peaks at retention times relative to aldrin are
observed for possible interference. Samples are quantitated against
Arochlor 1248. The limit of detection is 150 (total) nanograms. The clean-
up procedure using Florisil assures removal of almost all interferring
chlorinated pesticides.
164
-------
REFERENCES FOR APPENDIX A
A--1 Hamersma, J. W., Reynolds, S. L., and Maddalone, R. F., "IERL-RTP
Procedures Manual: Level I Environmental Assessment," EPA
Report EPA-600/2-76-160a, NTIS Order No. PB 257 850, June 1976.
A-2 Private communication with Theodore C. Rains, U. S. Dept. of
Commerce, National Bureau of Standards, Washington, DC (1974).
A--3 Rosen, A. A., and Middleton, F. M., Anal. Chern^ 27, 790 (1955).
A-4 Moore, G. E., Thomas, R. S., and Monkman, J. L., J. Chromatogr.
26, 456 (1967).
A_5 Levaygi, D. A., et al., J. Air Pollution Association 26 (6),
554 (1976).
A-6 Sandberg, J. S., et al., J. Air Pollution Association 26 (6),
559 (1976).
A-7 ASTM Standards, Part 23, Water & Atmospheric Analysis, p. 273,'
Method 512-67, Referee Method A (1973).
A-8 Methods of Chemical Analysis of Waters and Wastes, US EPA,
EPA-625/6-74-003 (1974).
A-9 Hatch, W. R. , and Ott, W. L., "Determination of Sub-Microgram
Quantities of Mercury in Solution by a Flameless Atomic Absorp-
tion Technique," Atomic Absorption Newsletter 6, 101 (1967).
A-10 Janini, G. M., Hohnston, R., and Zrelinski, W., Anal^ Chem. 47,
(1975).
A-ll Gouw, T. H., Whittemore, I. M., and Jentoft, R. E., "Capillary
Column Separation of Various Poly Cyclic Aromatic Materials,"
Anal. Chem. 42, 1394 (1970) .
A-12 Lao, R. C., Thomas, H., Oja, H., and Dubois, L., "Application
of Gas Chromatograph-Mass Spectrometer Data Processor Combination
to the Analysis of the Polycyclic Hydrocarbon Content of Airborne
Pollutants," Anal. Chem. 45, 908 (1973).
165
-------
APPENDIX B - CONVERSION FACTORS
SI Unita to Metric or English Unita
To Obtain
g/Hcal
106 Btu
MBII/ft2
Mflll/ft3
Btu
10 lb/hr» or HUH
Ib/HDtu
ft
in
ft*
ft3
H- lb
OS
OS Fahrenheit
p»ig
psia
iwg (39.2»F)
c.
10° Btu/hr
GJ/hr
From
ng/J
GJ
GJ'hr"1^"2
GJ-hr"1^'3
gn col
GJ/hr
ng/J
in
cm
m*
n.3
Kg
Celsius
Kelvin
Pa
Pa
Pa
MW
MW
Multiply By
0.004186
On JO
• 3H O
0.08806
0.02684
3.9685 x 10"3
0.948
0.00233
3.281
0.3937
10.764
35.314
2.205
tF - 9/5 (t) 02
r C
t - l.BK - 460
P i " «PM)(l.«OX10-4)-14.7
psig pa .
P^ . - (P . (1.450X10"4)
psia pa)
pi^ "
-------
English and Metric Unita to SI Units
TO Obtain
ng/J
ng/J
GJ.hr"1'*"2
GJ'hr"1.*"3
GJ/hr
cm
n
n
Kg
Cclotus
Kelvin
Pa
Pa
HW
MW
From
Ib/MBtu
g/Mcal
HBII/Ct2
MBH/ft3
10 3 Ib/hr*
or 106 Btu/ht
ft
in
ft2
ft3
lb
Fahrenheit
psig
paia
ivg (39.2*F)
106 Btu/hr
GJ/hr
'Multiply By
430
239
11.356
37.257
1.055
0.3048
2.54
0.0929
0.02832
0.4536
fc - 5/9 (tr-32)
tR - 5/9 (tr-32) 4- 273
V (pPsig*14-7)(6
Ppa-
-------
TECHNICAL REPORT DATA
(Please read Inunctions on the reverse before completing}
. REPORT NO.
EPA-600/7-78-099a
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Emission Reduction on Two Industrial Boilers with
Major Combustion Modifications
5. REPORT DATE
June 1978
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
W.A.Carter, H.J.Buening, and S.C.Hunter
8. PERFORMING ORGANIZATION REPORT NO.
KVB 6004-734
9. PERFORMING ORGANIZATION NAME AND ADDRESS
KVB, Inc.
17332 Irvine Boulevard
Tustin, California 92680
10. PROGRAM ELEMENT NO.
EHE624A
11. CONTRACT/GRANT NO.
68-02-2144
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13 TYPE OF REPORT. AND PERIOD COVERED
Final; 1/76-1/78
14. SPONSORING AGENCY CODE
EPA/600/13
15. SUPPLEMENTARY NOTES IERL-RTP project officer is Robert E. Hall, Mail Drop 65, 919/
541-2477.
16. ABSTRACT The J^Q^QJ^ giV6s results of a study of the effects on pollutant emissions of
extensive combustion modifications on two industrial boilers. Staged combustion,
variable excess air, and variable air preheat were evaluated while firing natural gas
or No. 6 fuel oil in a watertube boiler rated at 16 MW thermal input (55 million Btu/
hr). Reductions in NOx of 31% for natural gas and 42% for No. 6 fuel oil were obtai-
ned when excess air was optimized and staged air was introduced through injection
ports in the furnace side, as far downstream from the burner as practical. Com-
bined lowered air preheat and staged combustion reduced NOx by 70% while firing
natural gas. In a watertube boiler rated at 6. 5 MW thermal input (22 million Btu/hr),
flue gas recirculation, staged combustion, and variable excess air were evaluated
while firing natural gas, No. 2 fuel oil, or No. 6 fuel oil. The maximum NOx reduc-
tion for natural gas was 79% with flue gas recirculation and lowered excess air.
A 77% NOx reduction was obtained for No. 2 fuel oil with combined modifications.
Since NOx reduction for heavy fuel oil has been very difficult to achieve, the most
significant result in the program was a 55% NOx reduction, obtained with a combi-
nation of modifications while firing No. 6 fuel oil. Trace species and organics
emissions were measured while firing No. 6 fuel oil.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Air Pollution
Combustion
Boilers
Burners
Nitrogen Oxides
Smoke
Fossil Fuels
Particle Size
Flue Gases
Circulation
Fuel Oil
Natural Gas
Trace Elements
Organic Compounds
Air Pollution Control
Stationary Sources
Combustion Modification
Industrial Boilers
Particulate; Excess Air
Staged Combustion
Trace Species
13B
2 IB
13A
07B
21D_
14B
06A
07C
13. DISTRIBUTION STATEMEN1
Unlimited
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
17T
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
168
------- |