Environmental Assessment of a Waste-to-Energy
Process: Burlington Electric's Wood and
Oil Co-Fired Boiler
Midwest Research Inst.
Kansas City, MO
Prepared for
Industrial Environmental Research
Lab.-Cincinnati, OH
Aug 80
U.S. DEPARTMENT OF COMMERCE
National Technical Information Service
-------
EPA-600/7-80-148
August 1980
ENVIRONMENTAL ASSESSMENT OF A WASTE-TO-ENERGY PROCESS
Burlington Electric's Wood and Oil Co-Fired Boiler
by
Mark A. Golembiewski
Midwest Research Institute
Kansas City, Missouri 64110
Contract No. 68-02-2166
Project Officer
Harry M. Freeman
Fuels Technology Branch
Industrial Environmental Research Laboratory
Cincinnati, Ohio 45268
INDUSTRIAL ENVIRONMENTAL RESEARCH LABORATORY
"OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
NOTICE
THIS DOCUMENT HAS BEEN REPRODUCED
FROM THE BEST COPY FURNISHED US BY
THE SPONSORING AGENCY. ALTHOUGH IT
IS RECOGNIZED THAT CERTAIN PORTIONS
ARE ILLEGIBLE, IT IS BEING RELEASED
IN THE INTEREST OF MAKING AVAILABLE
AS MUCH INFORMATION AS POSSIBLE.
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/7-80-148
2.
3. REC
4. TITLE AND SUBTITLE
Environmental Assessment of a Waste-to-Energy Process:
Burlington Electricfs Wood and Oil Co-Fired Boiler
5. REPORT DATE
August 1980 Issuing Date.
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Mark A. Golembiewski
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Midwest Research Institute
425 Volker Boulevard
Kansas City, Missouri 64110
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
EPA 68-02-2166
12. SPONSORING AGENCY NAME AND ADDRESS
Industrial Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA/600712
15. SUPPLEMENTARY NOTES
EPA Project Officer: Harry M. Freeman (513/684-4363)
16. ABSTRACT
In July 1978, Midwest Research Institute conducted a series of emission tests at
the Burlington Electric Department's power plant in Burlington, Vermont. The study was
designed to provide multimedia emission data for the purpose of identifying potentially
adverse environmental impacts and to identify pollution control technology needs.
The No. 1 boiler at Burlington Electric, which was tested for this study, is
fueled by a combination of wood chips and No. 2 fuel oil. Approximately 82% of the
heat input (9.3 tons/hr) was provided by the wood fuel and the remaining 18% by the
fuel oil (175 gal/hr). Electrical power generated from this boiler system was about
8 MW. The air pollution control system consists of two mechanical collectors in series.
Four effluent streams were sampled and analyzed for this assessment program:
bottom ash; primary collector ash; secondary collector ash; and stack emissions. Common
to all streams were characterizations for elemental composition and potentially haz-
ardous compounds such as polychlorinated biphenyls and polycyclic aromatic hydrocarbons.
In addition, the boiler exhaust gases were analyzed for particulate, NOX, S02, CO, and
total hydrocarbon concentrations. The Source Assessment Sampling System was also used,
following guidelines established by EPA's Level 1 environmental assessment protocol.
17.
KEY WORDS ANO DOCUMENT ANALYSIS
DESCRIPTORS
b.lOENT!FIERS/OPEN ENDED TERMS
c. COS ATI Field/Group
Air Pollution
Wood Wastes
Environmental Tests
Boilers
Wood-Fired Boiler
SASS Train
SAM-LA Analysis
Multimedia Effluent
Sampling
13. DISTRIBUTION STATEMENT
Release to public.
19. SECURITY CLASS (DiaReportl
Unclassified
21. NO. OP PAGES
179
20. SECURITY CLASS iThis page)
Unclassified
22. PRICE
EPA Form 2270-1 (R«v. 4-77) previous EDITION is OBSOLETE
-------
DISCLAIMER
This report has been reviewed by Che Industrial Environ-
mental Research Laboratory, U.S. Environmental Protection
Agency, and approved for publication. Approval does not
signify that the contents necessarily reflect the views and
policies of the U.S. Environmental Protection Agency, nor
does mention of trade names or commercial products constitute
endorsement or recommendation for use.
-------
FOREWORD
When energy and material resources are extracted, processed, converted,
and used, the related pollutional impacts on our environment and even on our
health often require that new and increasingly more efficient pollution con-
trol methods be used. The Industrial Environmental Research Laboratory-
Cincinnati assists in developing and demonstrating new and improved method-
ologies that will meet these needs both efficiently and economically.
This report documents the results of an environmental assessment of a
power plant operated by Burlington (Vermont) Electric Department which is
fueled with a combination of wood chips and fuel oil. It discusses, in detail,
emissions of criteria and many noncriteria pollutants. It will be of interest
to those considering wood as an alternative fuel. Questions and comments
should be directed to the Fuels Technology Branch, Industrial Environmental
Research Laboratory, Cincinnati, Ohio 45268.
David G. Stephan
Director
Industrial Environmental Research Laboratory
Cincinnati
iii
-------
ABSTRACT
In July 1978, Midwest Research Institute (MRI) conducted a series of
emission tests at the Burlington Electric Department's power plant in
Burlington, Vermont. The study was designed to provide multimedia emission
data for the purpose of identifying potentially adverse environmental impacts
and to identify pollution control technology needs.
The No. 1 boiler at Burlington Electric, which was tested for this study,
was fueled by a combination of wood chips and No. 2 fuel oil. Approximately
82% of the heat input (9.3 tons/hr) was provided by the wood fuel and the
remaining 18% by the fuel oil (175 gallons/hr). Electrical power generated
from this boiler system was about 8 MW. The air pollution control system
consisted of two mechanical collectors in series.
Four effluent streams were sampled and analyzed for this assessment pro-
gram: bottom ash; primary collector ash; secondary collector ash; and stack
emissions. The two fuels used were also analyzed. The analysis included
identification of elemental compositions, and potentially hazardous com-
pounds, such as polychlorinated biphenyls and polycvclic aromatic hydro-
carbons. In addition, the boiler exhaust gases were analyzed for particulate,
CO, NOX, S(>2, and total hydrocarbons. The Source Assessment Sampling System
(SASS) was also used, following guidelines established by EPA's Level 1 en-
vironmental assessment protocol.
Data obtained from the above tests were used to evaluate each effluent
stream on the basis of existing standards and also based on EPA's Source
Analysis Model (SAM-LA). Results of the sampling and analysis efforts indi-
cate a total particulate loading of 0.077 gr/scf at the outlet of the control
device. The control device demonstrated a particulate collection efficiency
of 94%. Average CO concentrations ranged between 180 and 245 ppm; NOX
emissions averaged about 65 ppm; S02 averaged between 120 and 155 ppm, and
total hydrocarbons were of the order of 10 ppm. Based on the SAM-LA method-
ology, the secondary collector ash demonstrated the highest degree of hazard.
iv
-------
CONTENTS
Page
Foreword , m
Abstract iv
Figures . vii
Tables ix
Acknowledgment . xiii
Summary. xiv
1. Introduction ........................ 1
2. Description of the Boiler Facility ...... 2
3. Sampling and Analytical Methodology ............. 6
Wood Chip Fuel . 6
Fuel Oil 11
Bottom Ash ................. 11
Collector Inlet ....... ..... 12
Primary and Secondary Collector Ash .......... 12
Collector Outlet .............. 13
Chemical Speciation .................. 15
4. Presentation and Discussion of Results ........... 16
Boiler Operating Conditions .............. 16
Wood Fuel Analysis ................... 17
Fuel Oil Analysis ................... 20
Bottom Ash ....................... 22
Uncontrolled Air Emissions (collector inlet) ...... 26
Primary and Secondary Collector Ash .......... 32
Controlled Air Emissions (collector outlet) ...... 40
Opacity .« ................ 55
PCS and PAH Compounds ................. 55
SASS - Level 1 Assessment ............... 58
Electron Spectroscopy for Chemical Analysis ...... 67
5. Environmental Assessment of Burlington Data Based on
EPA's SAM-lA 72
Approach ........................ 72
Burlington Data .................... 74
6. Conclusions ......................... 77
Bottom Ash ......*................ 77
Primary Collector Ash ................. 77
Secondary Collector Ash ................ 78
Stack Emissions .................... 78
SAM-lA Effluent Analy_si_s__»_ - 79
v
-------
CONTENTS (continued)
References •••••••••••••••••••••••••••••• 80
Appendices
A* Analytical Methodology ••••••••••••••••••• 81
B* Sampling Locations and Procedures •••••••••••••• 120
C» Comparison of Chemical Analyses of Bottom Ash, Primary Ash,
and Secondary Collector Ash ••••••••••••••••• 135
D. Description of SASS Equipment and Analysis ••••••••• 142
E. Data Tables for SAM-lA . 145
vi
-------
FIGURES
No* Page
1 Layout of the Burlington Electric Plant 3
2 Schematic diagram of the wood feed system (one of four chutes)* 4
3 Test matrix for Burlington Electricfs wood and oil fired
power plant .*.......*............... 7
4 Inlet particle size distribution. - Run 1 ...... 33
5 Inlet particulate size distribution - Run 2..... 34
6 Inlet particle size distribution - Run 3 ........... 35
7 Outlet particle size distribution - Run 1 51
8 Outlet particle size distribution - Run 2 52
9 Outlet particle size distribution - Run 3 ' 53
10 Average particle size distributions (inlet and outlet) .... 54
11 Plot of opacity versus time, 7/24/78 57
12 Plot of opacity versus time, 7/25/78 57
13 Auger survey scan of 1-y, particulate ••...»••••..• 71
A-l Reconstructed ion chromatogram for PAH standard ........ 98
A-2 Chromatogram resolution of phenanthrene, anthracene, and D-10
anthracene ......................... 99
A-3 Chromatographic resolution of 1,2-benzanthracene and chrysene . 100
A-4 Chromatogram of heterocyclic PAH standard ........... 102
A-5 Identification of fluoranthene and pyrene in secondary
collector ash •».».•••..•••••••••••••• 103
A-6 Organic Level 1 analysis flow diagram ••».••••••••• 112
A-7 IR spectrum for LC Fraction 6 of XAD-2 resin ...115 .
A-8 Typical TCO chromatogram of concentrated extract from XAD-2
sample •••••••••••••••••••••«««••• 117
A-9 Direct inlet LRMS of concentrated XAD-2 field sample extract . 118
B-l Layout of Burlington Electric Plant facilities 121
B-2 Example of sampling location for wood feed stream ..••••• 122
B-3 Illustration of the collector hopper arrangement ....... 124
B-4 Schematic diagram of the inlet sampling location ....... 125
B-5 Schematic diagram of the outlet sampling location ....... 126
B-6 Schematic illustration of Method 5 sampling train in sampling
position ...»...«•.•.•...»•••.••••• 128
B-7 Schematic illustration of the MRI dynamic dilution - optical
counter particle sizing system ........••..••• 129
B-8 Photograph of the MRI-developed dynamic dilution system .... 131
vii
-------
FIGURES (continued)
No.
B-9 Photograph of the oarticle counting system in sampling
position .......................... 132
B-10 PCB sampling train •.... 133
B-ll Schematic diagram of the SASS train .............. 134
C-l GC/MS chromatograms for bottom ash, primary collector ash,
and secondary collector ash extracts ............ 138
C-2 Photograph of density gradient experiment ........... 140
D-l Schematic of source assessment sampling system ........ 143
viii
-------
TABLES
No,
1 Burlington test: schedule • ••••••»••»•»•••••. 6
2 Summary of analysis methods •••••••••••••••••• 8
3 Summary of fuel input data •••••••••••••••••• 17
4 Summary of boiler operating parameters •••••••••... 18
5 Analysis of wood fuel ..................... 19
6 Wood fuel composition reported by Burlington Electric ..... 20
7 Elemental analysis of wood by SSMS .............. 21
8 Analysis of No. 2 fuel oil 22
9 Elemental analysis of fuel oil by SSMS 23
10 Elemental analysis of bottom ash by SSMS ........... 24
11 Analysis.of bottom ash for PAH by GC/MS 25
12 Uncontrolled air emissions .................. 26
13 Elemental analysis of inlet Method 5 filter particulate by
SSMS 28
14a Air emission concentrations for selected elements in
uncontrolled particulate . . . . 29
14b Air emission factors for selected elements in uncontrolled
particulate ........ . 29
15 Inlet particle size distribution by number as monitored by the
optical counter ....................... 30
16 Inlet particle counts by size as monitored by the diffusion
battery /condensation nuclei counter ............. 31
17 Inlet particle size distribution resulting from diffusion
battery penetration data .................. 32
18 Elemental analysis of primary collector ash by SSMS concen-
tration .... 37
19 Elemental analysis of secondary collector ash by SSMS 38
20 Comparison of elemental analysis of ash materials ....... 39
21 Analysis of secondary collector ash for PAH by GC/MS ..... 41
22 Boiler exhaust gas composition ................ 4^
23 Summary of particulate test results (metric units) ...... 4f
24 Summary of particulate test results (English units) ...... 45
25 Elemental analysis of outlet Method 5 filter particulate by
SSMS 47
ix
-------
TABLES (continued)
No* Paige
26 Selected elemental concentrations in controlled particulate
emissions •••••.•.......•••.•••••••• 48
27 Outlet particle size distribution by number as monitored by
the optical counter .«.....••........•.«• 48
28 Outlet particle counts by size as monitored by the diffusion
battery/condensation nuclei counter ••••••••••••• 49
29 Outlet particle size distribution resulting from diffusion
battery penetration data .................. 50
30 Summary of BAHCO ash analysis ........•«...•••• 56
31 Simulated particle size distributions developed by UOP . • • • 56
32; Analysis of Florisil train components for PAH by GC/MS . • • • 59
33 SASS particulate analysis (Level 1)..... . . . 60
34 Summary of SASS particulate results .............. 60
35 Elemental analysis of SASS aqueous condensate by SSMS . . . . . 62
36 Elemental analysis of SASS hydrogen peroxide impinger counter
by SSMS 63
37 Level 1 organic analytical results .............. 64
38 IR spectral analysis of concentrated XAD-2 resin extract • . • 66
39 Level 1 organic analysis of liquid chromatographic fractions
of XAD-2 resin sample 66
40 IR spectral analysis of LC Fractions 6 and 7 ......... 67
41 ESCA results for 1-tt. particulate ............... 69
42 ESCA results for bottom ash .................. 70
43 Summary of SAM-lA effluent analysis 74
A-l Proximate and ultimate blind duplicate analyses of wood and
fuel oil 82
A-2 Elemental analysis of NBS SRM 1633 coal fly ash by SSMS .... 90
A-3 SSMS elemental analysis of NBS pine needles SRM 1575 92
A-4 SSMS quality assurance analysis of trace metals in oil . • . . 93
A-5 SSMS analysis of selected elemental fortifications in aqueous
condensate ......................... 94
A-6 Elemental analysis of fortified 30% hydrogen peroxide by SSMS . 95
A-7 GC/MS operating conditions for PAH analysis 97
A-8 Quality assurance data for low level PAH fortification of
selected samples ...................... 104
A-9 GC/MS quality assurance data for selected long retention time
PAH compounds ........................ 105
A-10 Analytical precision of selected PAH compounds in ash extracts. 106
A-ll Recovery of Arochlor 1254 fortifications from selected samples
analyzed by EC/GC 108
A-12 Quality assurance data for GC/MS analysis of PCB fortified
samples »•••••••»•••••••••••••••••• 109
A-l3 Level 1 organic sample summary ................ 113
A-14 LRMS operating conditions
-------
TABLES (continued)
No. Page
C-l Quality assurance data for low level PAH fortification of
selected samples •••••••..•••••«••...•. 135
C-2 GC/MS quality assurance data for selected long retention time
PAH compounds ........................ 137
0-1 Analysis matrix for SASS train components ••••••••••• 144
xi
-------
METRIC CONVERSION FACTORS USED IN THIS REPORT
To convert from
English
Btu
Btu/lb
ft
ft2
ft3/min
gal/min
gr/scf
Ib
lb/ft3
lb/106 Btu
psi
ton
To
Metric
kJ
kJ/kg
m
m2
m3/min
liter/min
g/dscm
kg
kg/m3
kg/MJ
kPa
Mg
Multiply by
1.055
2.326
3.048 E-01
9.290 E-02
2.832 E-02
3.785
2.288
4.536 E-01
1.603 E-01
4.300 E-01
6.895
9.072 E-01
xii
-------
ACKNOWLEDGMENT
This report was prepared as partial fulfillment of Environmental Protec-
tion Agency Contract No. 68-02-2166. The overall program, "Environmental As-
sessment of Waste-to-Energy Processes," is managed by Mr. M. P. Schrag and the
project leader is Dr. K« P. Ananth. The principal author of this report was
Mr* Mark Golembiewski• The author is grateful for the assistance of Or* Glenn
Trischan, who supervised the analytical work, authored the appendix on analy-
sis methodology, and provided a great deal of interpretive input. Mr. Emile
Baladi is also recognized for his supervision of the field sampling activi-
ties.
Midwest Research Institute would like to express its sincere appreciation
for the generous assistance and cooperation provided by Mr. Thomas Carr, Super-
intendent of the Burlington plant, and his operating staff throughout this test
program.
xiii
-------
SUMMARY
The sampling study described in this report was conducted at the
Burlington Electric Department plant in Burlington, Vermont, during July 1978.
The objective was to provide information which would help define the hazard
potential of each effluent stream and identify control technology needs* The
test program and results are summarized below*
The No. 1 boiler at Burlington Electric has been modified to burn a com-
bination of wood chips and No* 2 fuel oil and produces sufficient steam to
power an electrical generator rated at 10 MW. Wood is the primary fuel and oil
is used as a supplementary fuel to maintain steam production near maximum ca-
pacity. Air emissions are controlled by two mechanical collectors in series*
Bottom ash and fly ash collected by the particulate control system are pneu-
matically conveyed to a storage silo before being transported to a landfill.
Fuel feed and effluent streams sampled for this environmental assessment
program included:
• Wood chip feed;
• Fuel oil feed;
• Bottom ash;
• Primary collector ash;
• Secondary collector ash; and
• Stack emissions (collector outlet).
No wastewater streams were associated with the Burlington boiler.
Three complete sets of samples from each of these streams were taken over
a 2-day period. Sample analyses included conventional pollutants plus elemental
composition and selected hazardous materials (polychlorinated biphenyls - PCS,
and polycyclic aromatic hydrocarbons - PAH). In addition, a modified Level 1
assessment was carried out using the Source Assessment Sampling System (SASS).
xiv
-------
The Level 1 analysis protocol included determinations for particulates, organic
constituents and vaporous metals. Major findings of the study, plus brief de-
scriptions of the test methods, are summarized below in the order of the input/
output streams listed above*
SYSTEM OPERATING PARAMETERS
Wood chips (roughly 3-cm square) are pneumatically injected into the
boiler and onto a traveling grate at a rate of about 9 tons/hr. The supplement
tary fuel oil (approximately 175 gal/hr) is fired from the sides, several feet
above the grate* The wood/oil feed ratio during the Midwest Research Institute
(MRl) tests was approximately 80% wood - 20% oil, on a heat input basis* A
slightly higher ratio was used on the second test day. Total heat input to the
boiler averaged about 140 x 106 kJ/hr (135 x 106 Btu/hr).
WOOD FEED
Grab samples of wood chips were taken from the feeder system during each
of the three test runs* Sample analyses included moisture content determina-
tion, proximate and ultimate analyses, and assessment of heating value* Ele-
mental composition of the wood was also determined* Results of the proximate
analysis indicated an average composition of 4.37o ash, 70.0% volatile matter,
and 25*7% fixed carbon (dry basis)* The average sulfur content was 0.35% and
the average heat of combustion (as received) was 13,650 kJ/kg (5,870 Btu/lb)*
On a dry basis, the heating value was 22,040 kJ/kg (9,480 Btu/lb).
OIL FEED
One grab sample of No. 2 fuel oil was collected for each day of testing.
Heat of combustion and elemental composition were determined* The samples also
underwent ultimate analysis* Sulfur content of the oil was 0.35%, while its
heat of combustion averaged 45,400 kJ/kg (19,500 Btu/lb or 138,300 Btu/gal).
BOTTOM ASH
Three composited samples of the bottom ash were obtained for subsequent
analysis* Determinations were made for elemental composition and PCB and PAH
material content. As expected, most elements were more concentrated in the bot-
tom ash relative to the fuel inputs. Those elements exhibiting the largest in-
creases in concentration included Ba, Zr, Sr, and Li*
No PCB materials could be detected in the bottom ash samples above the
0.05 ng/g detection limit. One PAH compound, phenanthrene, was identified at
an average concentration of 0.89 p,g/g.
xv
-------
The amount of bottom ash generated during the MRI test program was esti-
mated to be about 91 kg/hr (200 lb/hr). On a dry weight basis, slightly less
than 2% of the wood fuel input was discharged as bottom ash*
UNCONTROLLED AIR EMISSIONS
Particulate concentrations in the boiler exhaust gases before the control
device were determined by completing three Environmental Protection Agency
(EPA) Method 5 runs* The resulting average concentration was 2.96 g/dscm (1.30
gr/dscf)• On the basis of heat input, uncontrolled particulate emissions aver-
aged 1.47 g/MJ (3.43 lb/106 Btu). Oxygen and carbon dioxide contents were moni-
tored during the particulate runs and averaged 12.3% and 8.2%, respectively.
Filter samples from the Method 5 particulate tests were analyzed for ele-
mental composition. The elements emitted at concentrations greater than 10 ^g/
dscm were Pb, Ba, Sr, As, Ga, Zn, Cu, Fe, Mn, Ti, and P»
An optical/diffusional particle counting system was used to measure the
particle size distribution of the uncontrolled emissions. Three runs using this
system were made. Particles in the range of 0.005 to 0.10 ^m were counted by a
diffusion battery/condensation nuclei counter arrangement, while those in the
0.3 to 2.6 urn range were counted by an optical counter. Because the dilution
system consistently became plugged with larger particles during operation, no
particle counts could be obtained in the size region above 2.6 |J,m. Therefore,
the mean particle size could not be determined. Within the size range of parti-
cles counted (0.005 to 2.6), the majority of the particles appeared to be be-
tween 0.3 and 0.5 (j,m in diameter.
PRIMARY AND SECONDARY COLLECTOR ASH
Grab samples of collected fly ash were taken from the hoppers beneath the
primary and secondary mechanical collectors. The samples were analyzed for ele-
mental composition and for concentrations of PCB and PAH compounds. In general,
the elements detected at the highest concentrations in the primary and second-
ary collector ashes were the same elements that were most prominent in the bot-
tom ash. Furthermore, many elements were more concentrated in the secondary ash
relative to the primary ash. Several elements showed a trend of increasing con-
centration from bottom ash to primary collector ash to secondary collector ash.
These included Hg, Br, Se, As, Cl, F, and B.
Analysis for PCB materials did not yield any positive responses above the
.0,05 ng/g detection limit in either of the ash samples. No PAH compounds were
identified in the primary ash sample extracts. However, several compounds were
confirmed in the secondary ash samples, including acenapthylene, phenanthrene,
fluoranthene, and pyrene. One sample contained 10 p,g/g of phenanthrene, which
was the highest PAH concentration observed.
xvi
-------
CONTROLLED AIR EMISSIONS
Using continuous gas analyzers, concentrations of 02, NOX, S02» CO, and
total hydrocarbons (THC) were measured in the boiler exhaust stream* NOX and
SC>2 concentrations averaged 66 and 138 ppm, respectively* CO readings averaged
213 ppm* THC concentrations averaging only 9 ppm were observed*
Three Method 5 particulate runs were made simultaneously with the sam-
pling runs at the collector inlet* The average particulate concentration was
0,18 g/dscm (0,08 gr/dscf). The particulate emission rate averaged 0.09 g/MJ
(0.17 lb/106 Btu), on the basis of heat input. The average efficiency of the
two-stage mechanical collection system, as determined from the simultaneous
inlet/outlet tests, was 94.27. for total particulate and 95.1% for filterable
particulate only. Visual plume opacity readings taken during testing indicated
an average opacity of about 20%.
Elemental analysis of the Method 5 particulate filters indicated moder-
ately high elemental concentrations. Pb, Ba, Sr, Zn, and Ti were present at
the highest concentrations, approaching 100 |j,g/dscm, while Hg, Sb, Zr, Br, Se,
As, Ga, Cu, Ni, and V were in the range of 1 to 75 (j,g/dscm. The remaining ele-
ments had concentrations which were less than 1 (j,g/dscm.
Particle size data were obtained by the same method used at the collector
inlet (optical/diffusional particle counter). As with the inlet measurements,
data for particles > 2.6 (im in diameter could not be obtained. The number of
small particles (< 2.6 (im) appeared to increase in the controlled gas stream
relative to the uncontrolled emissions. The reasons for this increase are not
clear.
Plume opacity data, obtained using EPA Method 9, averaged about 20% on
both test days. Samples for analysis of PCB and PAH materials were collected
in a special sampling train utilizing impingers and a Florisil adsorbent trap.
Two sampling runs were made with this collection system and the samples were
split for the two, separate analyses. PCB analysis did not produce any re-
sponses greater than the 1 fig/sample detection limit of the GC/MS analytical
technique. Similarly, no PAH compounds were identified at levels which per-
mitted structural confirmation.
Organic analysis of the SASS components, in accordance with Level 1 guide-
lines, revealed low levels of organic constituents. Characterization of the
organic emissions was difficult, although they appeared to be composed mainly
of carbonyl-containing groups.
xvii ,
-------
SOURCE ASSESSMENT MODEL (SAM-lA)
The EPA's SAM-lA methodology was applied to the four effluent streams as
a means of interpreting the emission measurement results* The SAM-lA analysis
indicated that the secondary collector ash contained the highest degree of
hazard^ although all three ash streams were similar in the magnitude of their
hazard values* These values were primarily the result of just a few elements
which had low Minimum Acute Toxicity Effluent (MATE) values assigned to them.
Considerations of the physical nature of the ash discharges and the actual dis-
posal conditions could substantially reduce the true hazard potential of these
streams from that estimated by the SAM-lA methodology. Stack emissions showed
a relatively low degree of hazard*
The primary collector ash stream had the highest Toxic Unit Discharge Rate
(TUDR) which would seem to indicate that this effluent should receive the first
priority for control measures* However, for the reasons discussed above, further
work is needed to provide a better estimation of the environmental control
needs*
xviii
-------
SECTION 1
INTRODUCTION
The environmental assessment of Burlington Electric's wood and oil fired
boiler described in this report Is one of a series of similar field studies
performed by MRI under contract to EPA's Industrial Environmental Research Lab-
oratory, Fuels Technology Branch, in Cincinnati. Other waste-to-energy systems
studied to date have included a wood and coal fired boiler, an air classifier
at a refuse processing plant, a refuse-fed pyrolysis reactor, and a waterwall
refuse incinerator. (1-4)
Because of renewed interest in the use of wood and wood waste as a pri-
mary boiler fuel in certain regions of the country, the Burlington Electric
plant was selected for inclusion in the EPA/MRI environmental assessment pro-
gram* The Burlington facility is the only one in the United States that is
presently firing oil with wood waste to generate electric power.
The following sections of this report contain a description of the test
facility, an overview of the sampling and analytical procedures used, and a
discussion of the results of the test program. An assessment of these results
by applying EPA's SAM-lA and a summary of the study's major findings are also
included.
-------
SECTION 2
DESCRIPTION OF THE BOILER FACILITY
The No, 1 unit at Burlington Electric's J. Edward Moran Generating Sta-
tion was originally a coal fired boiler which has been modified to accommodate
a wood fuel (chips) with supplementary oil injection. Steam production is
rated at 45,360 kg/hr (100,000 Ib/hr), which powers a 10 MW turbine generator.
Tandem mechanical collectors are used for emission control* Figure 1 shows a
layout of the plant and storage yard*
Wood chips from the storage pile are pushed by bulldozer into a ground-
level feed conveyor hopper* Because of an adjacent coal storage pile, small
amounts of coal fines are invariably mixed with the wood chips* Through a ser-
ies of conveyors, the chips are brought to a storage bin near the No. 1 boiler.
Four parallel screw conveyors continuously remove wood chips from the bottom
of the storage bin and transfer them to four gravity-fed chutes. Compressed
air is used at the base of the chutes to inject the chips into the boiler and
to prevent clogging in the chutes* A schematic of the feed system is shown in
Figure 2.
The wood chips fall onto a horizontal, traveling grate which is supplied
with underfire air* Because of the high moisture content of the chips, the
boiler cannot provide the desired steam output based on wood alone* Therefore,
supplementary fuel oil (No* 2), along with overfire air, is introduced above
the grate bed from both sides of the firebox, thus insuring adequate steam
production. Residual ash is discharged at the end of the grate into a hopper
and then is pneumatically transferred to an outdoor storage silo.
During the MRI tests, the fuel blend ratio, on a Btu input basis, was ap-
proximately 82% wood chips and 18% fuel oil. The corresponding feed rates were
roughly 8.2 Mg/hr (9 tons/hr) and 660 liters/hr (175 gal/hr), respectively.
Under these firing conditions, the No* 1 boiler produced about 40,800 kg/hr
of steam (90% of rated capacity)*
-------
Feed Hopper
to Conveyor
System
Power House
3 Boilers + 3 Generators
Figure 1. Layouc of the Burlington Electric Plant,
-------
From Wood Chips
Storage Bins
Sampling
Location
Boiler
Compressed Air
Figure 2, Schematic diagram of the wood feed system
(one of four chutes).
-------
The boiler was supplied by the Wickes Boiler Company and has a furnace
volume of 170 m^» A waterwall surface area of 216 m^ precedes the boiler tubes
and an economizer section follows* The design steam output specifications are
6,200 kPa at 438°C (900 psig at 820°F).
Flue gases, after leaving the economizer section, are ducted to an emis-
sion control system which consists of two, high efficiency mechanical collec-
tors in series* The units were supplied by Union Oil Products (UOP)/Air Cor-
rection Division* Each collector contains 260, 15*2 cm diameter tubes in a 13
by 20 array* For a flue gas flow rate of 1,700 acmm at 166°C (60,000 acfm at
330°F), the collectors were designed for an overall pressure drop of 16*5 cm
H20 and a collection efficiency of 97.757.*
After exiting the collectors, the flue gases enter an 10 fan and are dis-
charged through the No* 1 unit's exhaust stack* Exhaust flow rates are of the
order of 1,275 dscmm (45,000 dscfm)*
Fly ash collected by the control system is pneumatically conveyed to a
storage silo and later transported by truck to a landfill*
-------
SECTION 3
SAMPLING AND ANALYTICAL METHODOLOGY
The sampling program conducted at the Burlington Electric facility was
designed to provide environmental assessment data by characterizing all efflu-
ent streams from the boiler system* The two fuels used were also sampled and
analyzed to allow comparison with the effluent streams. Figure 3 presents the
test matrix followed for the Burlington program.
The sampling chronology is shown in Table 1.
TABLE 1. BURLINGTON TEST SCHEDULE
Test No.
1
Date
7/24/78
Test period
11:08-17:32
Comments
Included both PCB/PAH
2 and 3 7/25/78 08:20-19:57 Included SASS run!/
a/ In addition, sampling was done for total particulate, particle
sizing, and gases.
A summary of the laboratory analyses performed on each sample, and the metho-
dologies used, is shown in Table 2. Detailed descriptions of the analytical
procedures are contained in Appendix A* The sampling and analysis schemes used
for each of the effluent streams are discussed next.
WOOD CHIP FUEL
For each of the three tests, a sample of the wood chips was obtained from
the feed augers just prior to the wood supply chutes. A long-handled scoop was
used to extract a sample from each of the four augers.
-------
Wood Feed
Collector Intel
Collector Outlet
Sompllngi Take Three 5-Liter Samples
Each Day. Mix and Extract
a l-Liler Composite.
Analysis) Determine H2O Content
HHV, Proximate/Ultimate
Elemental Analysis by
55M5
Oil Feed
Sampling) Take One Or Two 0.5-
Uter Samples
Analysis: HHV. Proximate/Ultimate
Elemental Analysis by
SSMS|
Bottom Ash
Sampling! Take Three 1 kg Samples •
Each Day. Mix and Extract
1 kg Composite
Analyslsi • Elemental Analysis by
. SSMS;
'PCB/PAH
Sampling and Analysis!
a. Method 5 Participate - I Per Day
Elemental Analyst* by SSMS*
b. Orsol (O2 &COj)
c. Particle Sizing- 1 Per Day
Bottom Ash
AA
YY
Primary Secondary
Collector Collector
Ash Ash
Sampling and Analysis)
a. Method 5 Particulote - 1 Run Per Day
Elemental Analysis by SSMS* :
b. Orsat (Oj &COj)
c. Particle Sizing - I Run Per Day
d. Opacity (Method 9) - Two I Hr Tests/Day
e. Continuous Analyien (O2, NOX, SO2, HC, CO)
f. PCB/PAH with Florlsll Train - 2 Runs
a. SASS - I Run
Analyie Per Level I Requirements
Primary & Secondary Collector A»h
Sampling! Take A I kg Grab Sample Each Hour •
Mix & Extract I kg Composite
Analysis) SSMS
PCB/PAH
•Atomic Absorption Analysis May Be Conducted
Based on Results of SSMS Analysis
Figure 3. Test matrix for Burlington Electric's wood and oil fired power plant.
-------
TABLE 2. SUMMARY OF ANALYSIS METHODS
Sample
Parameter
Analytical method
Fuel oil
00
Wood
Proximate
Moisture
Ash
Volatile matter
Fixed carbon
Sulfur
Heat of combustion
Ultimate
Hydrogen
Carbon
Nitrogen
Oxygen
Elemental composition
Proximate
Moisture
Ash
Volatile matter
Fixed carbon
Sulfur
Heat of combustion
Ultimate
Hydrogen
Carbon
Nitrogen
Oxygen
Elemental composition
Modified ASTM Method D-3172-73
Modified ASTM Method D-3176-74
Spark source mass spectrometry and atomic
absorption spectrometry (Hg)
Modified ASTM Method D-3172-73
Modified ASTM Method D-3176-74
Spark source mass spectrometry and atomic
absorption spectrometry (Hg)
(continued)
-------
TABLE 2. (continued)
Sample
Parameter
Analytical method
Bottom ash
Collector inlet
Primary collector
ash
Secondary collector
ash
Elemental composition
Polynuclear aromatic hydrocarbons
Polychlorinated biphenyls
Chemical speciation
Particulate mass
Moisture content
02 + C02
Particle sizing
Elemental composition
Elemental composition
Polynuclear aromatic hydrocarbons
Polychlorinated biphenyls
Elemental composition
Polynuclear aromatic hydrocarbons
Polychlorinated biphenyls
Spark source mass spectrometry
Gas chromatography/mass spectrometry
Gas chromatography/mass spectrometry and
electron capture gas chromatography
Electron spectroscopy for chemical analysis
EPA Method 5
EPA Method 5
Orsat/Fyrite
Particle counting system (optical)
Spark source mass spectrometry and atomic
absorption spectrometry (Hg)
Spark source mass spectrometry and atomic
absorption spectrometry (Hg)
Gas chromatography/mass spectrometry
Gas chromatography/mass spectrometry and
electron capture gas chromatography
Spark source mass spectrometry and atomic
absorption spectrometry (Hg)
Gas chromatography/mass spectrometry
Gas chromatography/mass spectrometry and
electron capture gas chromatography
(continued)
-------
TABLE 2. (continued)
Sample
Parameter
Analytical method
Collector outlet
Particulate mass
Moisture content
02 + C02
Particle sizing
Opacity
Gases
S02
NOX
CO
°2
HC
Level 1
Particulate sizing
Organic
Elemental composition
Chemical speciation
EPA Method 5
EPA Method 5
Orsat with 02 also determined by continuous
analysis
Particle counting system (optical)
EPA Method 9
Continuous monitors
SASS - Level 1 gravimetric
SASS - Level 1 gravimetric, gas chromatog-
raphy, infrared, liquid chromatography,
and mass spectrometry
Spark source mass spectrometry and atomic
absorption spectrometry (Hg)
Electron spectroscopy for chemical analysis
and auger electron spectroscopy
-------
This was done three times during each test and the collected chips were placed
in a plastic bag* At the end of each test, the chips were thoroughly mixed and
a 1-liter sample was withdrawn and placed in a glass jar*
In the laboratory, the wood samples were first dried at a temperature
slightly above ambient to remove the "free" moisture* The samples were then
ground and submitted for proximate and ultimate analysis* The fuel's heating
value was also determined* Another portion of each sample was analyzed by
Spark Source Mass Spectrometry (SSMS) to determine elemental composition*
The quantity of wood chips fed to the boiler each hour was determined
using a counting mechanism connected to the shafts of the augers* The number
of revolutions were totaled each hour by plant personnel and translated into
pounds of wood chips using previously determined calibration data.
FUEL OIL
Fuel oil samples were drawn from a tap in the oil feed line near the point
of injection into the boiler* A 1-liter sample was collected on each of the 2
days of testing*
As with the wood chips, the fuel oil was subjected to ultimate analysis
and determination of heat of combustion. In addition, an elemental analysis
was performed using SSMS and the samples were further analyzed for FOB mate-
rials*
The fuel flow rate to the boiler was determined using differential volume
readings from a cumulative flow meter*
BOTTOM ASH
During each test period, three 1-liter samples of bottom ash were col-
lected at regular intervals from the ash hopper and placed in a sample bucket.
After the third sample was taken, the composited ash was thoroughly mixed and
a 1-liter sample was extracted for analysis*
After size reduction, laboratory analysis of the bottom ash consisted of
elemental characterization by SSMS and analysis for Hg by atomic absorption
(AA). The bottom ash sample was also assayed for PAH and PCB materials*
A precise determination of the bottom ash generation rate was not possi-
ble because the ash accumulates in a hopper beneath the discharge end of the
traveling grate and is removed only periodically through a pneumatic system*
Also, the depth of ash in the hopper could not be determined, which precluded
a calculation of the material volume based on the hopper dimensions.
11
-------
Therefore, on the 2nd day of testing, the ash hopper was allowed to fill
over a defined period of time. After the boiler was shut down, the hopper con-
tents were transferred manually into 55 gal» drums» One of the drums was sub-
sequently weighed and the result used as an average weight for all the drums»
Thus, the bottom ash generation rate was estimated by calculating the total
weight of material collected during that particular time period.
COLLECTOR INLET
Emission measurements of the, flue gases leaving the boiler were made at
the inlet to the mechanical collectors. The determinations consisted of partic-
ulate mass concentration, particle size distribution, and gas composition
and C^). Each of these measurements is briefly described next.
Particulate Mass, Concentration
Three EPA Method 5 sampling runs were completed for the program. One run
was made on the first day and two on the 2nd day of testing. Particulate con-
centrations were determined from both the front and back halves of the sampling
train. After the weights of the collected particulate samples were determined,
the filter samples were analyzed for elemental composition by SSMS»
A Fyrite gas analyzer and an orsat gas analyzer were used to measure the
percentages of Q£ and C0£ in the inlet gas stream. Triplicate readings were
obtained during each particulate sampling run and the average value used in
subsequent calculations.
Particle Size Distribution
A fine particle counting systsn, designed by MM, was used to size the
uncontrolled particulate emissions. Larger particles were counted by an optical
counter, while the number of smaller particles was determined by a diffusion
battery/condensation nuclei counter arrangement (see Appendix B for a descrip-
tion of this system). Because these devices can only accommodate a relatively
small number of particulates per unit volume of sample, a dynamic dilution sys-
tem was used to extract the sample from the inlet duct.
Particle counts were read directly from the counting instruments and were
recorded for subsequent data reduction and analysis.
PRIMARY AMD SECONDARY COLLECTOR ASH
Because the primary and secondary collector ash hoppers are emptied pneu-
matically, samples of the collected ashes could not be taken until the system
was shut down at the end of the test day. UOP/Air Correction Division person-
nel, who were present at the site to monitor the collector operation and to
12
-------
make some of their own measurements, used a vacuum system to empty the collec-
tor hoppers into 208 liter (55 gal.) drums* One-liter samples of both the pri-
mary and secondary ash were obtained from the drums at the end of both test
days* The ash samples were analyzed for elemental composition by SSMS and AA,
and by GC/MS techniques for PGB and PAH materials.
The quantities of fly ash captured by the tandem mechanical collectors
per unit time were estimated from measurements made by UOP personnel (who emp-
tied the ash hoppers and weighed the collected material) and from calculations
using the emission rates (kg/hr) measured by MRI at the inlet and outlet loca-
tions*
COLLECTOR OUTLET
The most extensive characterization of emissions was carried out on the
postcollector flue gas exhaust* Each of the measured gas parameters is dis-
cussed below*
Particulate Mass Concentration
Simultaneous with the inlet sampling, three EPA Method 5 particulate sam-
ples were collected at the outlet of the mechanical collectors* The samples
were subsequently weighed and the filters analyzed for elemental composition
by SSMS and by AA for mercury*
Particle Size Distribution
The same particle counting system used at the collector inlet location
was also used to determine the outlet size distribution* Three sampling runs
were made at times adjacent to those of the inlet runs*
Orsat Analysis
Three grab samples of the flue gas were taken during each of the Method 5
particulate runs and were passed through an Orsat gas analyzer to measure the
percentages of 02 and C02 in the gas stream* These values were later used to
calculate the molecular weight of the flue gases and the percentage of excess
air used in the combustion process*
Opacity
Stack plume opacity was determined using EPA Method 9. Visual readings
were taken for several hours each day to determine the average stack plume
opacity* These periods of observation generally overlapped the times of the
particulate tests*
13
-------
Gas Composition
Continuous gas analyzers were used to monitor real-time concentrations
of 02» CO, NO » SC-2» and THC. A Teflon sample line transported the flue gas
sample from the outlet duct to a field trailer which housed the monitoring
equipment* There a manifold diverted a portion of the sample to each gas ana-
lyzer. Instrument responses were registered on strip chart recorders and tran-
scribed to log sheets at 15-«iin intervals.
PGB and PAH
To collect PCB and PAH materials from the gas stream in both the particu-
late and gaseous phases, a special impinger/adsorption sampling train was used
(see Appendix B for additional information)* The measurement system consisted
of a series of Greenburg-Smith impingers (containing water) with a solid ad-
sorbent trap (Florisil adsorbent) placed between the third and fourth imping-
ers* Samples were collected from 45 points in the duct cross-section, at a
constant flow rate of approximately 1*7 liters/min (0.06 cfm). Two runs were
obtained using this method, both on the first day of testing.
For laboratory analysis, the samples were extracted with an organic sol-
vent and divided equally. GC/MS techniques were then used to analyze one sam-
ple portion for PCB compounds and the other for PAH materials.
SASS
The SASS is a device developed by the EPA for environmental assessment
work (Level 1 environmental assessments)* A brief explanation of the train and
a schematic diagram of the system are Included in Appendix D* The SASS train
is designed to collect particulates in four size ranges, and to collect gase-
ous emissions simultaneously by adsorption on XAD-2 resin (for organic analy-
sis) and by absorption in appropriate solutions (for vaporous trace metals).
One sampling run using the SASS, over nearly a 5-hr period, was completed
on the final day of testing. Consistent with Level 1 protocol, the sample was
taken isokinetically from a single point in the gas stream.
Analysis of the SASS components, as prescribed by the EPA's Level 1 envi-
ronmental assessment procedure, is quite complex and includes analyses utiliz-
ing SSMS, AA, solvent extraction, infrared spectrophotometry (IR), low resolu-
tion mass spectrometry (LRMS), and liquid chromatography (LC). A detailed
description of the analytical methodology employed for the SASS in this pro-
gram is included in Appendix A*
14
-------
CHEMICAL SPECIATION
In an effort to ascertain the chemical species of the most prevalent trace
elements, selected particulate samples (based on SSMS results) were analyzed
using electron spectroscopy for chemical analysis (ESCA). This technique, which
is designed to determine the chemical composition of the surface layers of par-
ticles, is further discussed in the next section and in Appendix A.
15
-------
SECTION 4
PRESENTATION AND DISCUSSION OF RESULTS
Boiler operating parameters and sampling/analytical results are discussed
in this section* As presented in the previous section, each fuel input or ef-
fluent stream is examined separately, in general order of material flow through
the system. The boiler operating conditions are presented first to provide a
background for the discussions of the process streams*
BOILER OPERATING CONDITIONS
Operation of the No. 1 boiler during the MRI test program was monitored
by Burlington Electric personnel. The boiler operator established and main-
tained firing conditions which produced the least visible stack plume. This
was consistent with the plant's normal operating mode.
As described earlier in Section 2, the boiler fuel consisted of a combi-
nation of wood chips and No. 2 fuel oil. Wood is considered the primary fuel,
but because of its inherently high moisture content (37 to 41%), the oil is
added to ensure sufficient steam production for the electric turbine genera-
tor.
Fuel usage data for the 2 days of testing are summarized in Table 3. In-
put rates for the wood and oil fuels were obtained from the daily records main-
tained by Burlington Electric personnel. The values shown are averages of hourly
fuel usage data.
The higher heating values shown in the table are the result of sample
analyis by MRI. It should be pointed out that the heating values measured for
the wood chips are probably slightly greater than comparative data compiled
by other researchers for a similar type of wood. (5) This is most likely due
to contamination from coal fines. In the plant's storage yard, the wood chip
pile is adjacent to a coal storage area and the wood fuel invariably becomes
"contaminated" with coal fines through wind erosion and the operation of the
bulldozer which pushes the chips over to the feed conveyor pit. The heating
values reported for this study, nevertheless, are representative of the actual
fuel firing conditions at the plant during the time of the MRI test program.
The values in Table 3 are not truly indicative, however, of the heat content
16
-------
TABLE 3. SUMMARY OF FUEL INPUT DATA
Wood chips No. 2 fuel oil
7/24/78 7/25/78 7/24/78 7/25/78
Fuel feed rate
Mg/hr (tons/hr) 7.92 (8.73) 8.92 (9.83)
Liter/hr (gal/hr) 700 (185) 644 (170)
Higher heating value
Btu/lbS/ 5,990 5,810 19,540 19,510
kJ/kg 13,720 13,510 45,450 45,390
Heat input rate
106 Btu/hr 104.6 114.2 25.6 23.5
106 kJ/hr 110.4 120.5 27.0 24.8
Percent heat input
contribution 80.3 82.9 19.7 17.1
a/ Heating values shown for the wood chip fuel are on an "as fired" basis.
Also, these values may reflect the contribution of minute amounts of
coal fines mixed in with the wood as a result of storage conditions.
of "clean" wood chips. By comparison, Burlington Electric has reported a Btu
content of 5,240 Btu/lb for a sample of "clean" wood chips that was taken from
the delivery truck near the time of the MRI test program.
Table 4 presents the average daily operating parameters for the No. 1
boiler. As shown by the data, boiler operation remained relatively steady, with
only a small increase in the heat input rate (and consequently an increase in
power output) on the 2nd day.
WOOD FUEL ANALYSIS
Results of the proximate and ultimate analyses of the wood fuel samples
are shown in Table 5. The wood fuel's heating value was also determined and
is presented on the basis of "as received," dry, and ash-free conditions. The
total moisture values shown were obtained by combining the results of two mois-
ture determinations. First, the samples were dried to a constant weight at
slightly greater than ambient temperatures, yielding a "free" moisture value.
After drying, the wood chips underwent size reduction for the subsequent proxi-
mate/ultimate analyses. The moisture values resulting from the proximate analy-
sis are included in the total moisture data in Table 5.
17
-------
TABLE 4. SUMMARY OF BOILER OPERATING PARAMETERS
Test day
7/24/78
7/25/78
Steam flow
kg/hr
Ib/hr
42,200
93,000
40,400
89,000
Steam pressure
kPa
psig
5,520
800
5,585
810
Steam temperature
°G
°F
427
800
440
823
Furnace draft
cm H20
in. H20
Economizer outlet draft
cm H20
in.
Boiler heat input
106 J/hr
106 Btu/hr
-W.23
-W.09
-10.2
-4.0
137.4
130.2
•W.91
•W.36
-9.7
-3.8
145.3
137.7
Generator power output
MW
7.9
8.0
18
-------
TABLE 5. ANALYSIS OF WOOD FUEL
Run No.
Date
Total moisture (%)
Proximate analysis (%)— '
Ash
Volatile matter
Fixed carbon
Total
Ultimate analysis (%)•=
Ash
Carbon
Hydrogen
Oxygen
Sulfur
Nitrogen
' Total
Heat of combustion (kJ/kg)
As received
Dry basis
Dry, ash-free basis
1
7/24/78
37.01
5.59
66.68
27.73
100.00
5.59
53.31
5.76
34.60
0.44
0.27
99.97
13,930
21,840
23,130
2
7/25/78
39.02
3.80
70.12
26.08
100.00
3.80
54.11
5.78
35.74
0.35
0.26
100.04
13,670
22,080
22,950
3
7/25/78
41.23
3.49
73.35
23.16
100.00
3.49
51.74
5.96
38.38
0.25
0.12
99.94
13,360
22,210
23,010
Average
39.09
4.29
70.05
25.66
100.00
4.29
53.05
5.83
36.24
0.35
0.22
99.98
13,650
22,040
23,030
a/ Dry basis.
19
-------
As mentioned previously, coal particles had unavoidably been mixed in
with the wood chips through storage and handling, so that the wood fuel char-
acteristics presented in Table 5 reflect this condition. Percentages of ash
and sulfur, in particular, are higher than typical published values for wood.
Furthermore, Burlington Electric Department has reported results from analysis
of a wood sample taken directly from a delivery van in September 1978. This
sample was representative of clean, fresh wood chips. The resulting composi-
tion is shown in Table 6.
TABLE 6. WOOD FUEL COMPOSITION REPORTED BY BURLINGTON ELECTRIC
As received (%) Dry (%)
Total moisture 38.29
Volatile matter 48.33 78.32
Fixed carbon 12.14 19.67
Ash 1.24 2.01
Sulfur 0.04 0.06
Heating value (Btu/lb) 5,240 8,491
In contrast to the ash content measured by MRI (4.29%), the Burlington .
Electric data show a value of 2.01% (on a dry basis). Likewise, the 0.06% sul-
fur reported in Burlington Electric's sample, was much lower than the 0.35%
average resulting from the MRI analyses. Periodic wood sample analysis has
never shown sulfur contents in excess of 0.1% according to plant personnel.
These differences are too large to be attributed to the natural variations in
wood composition alone, and therefore coal infiltration is probably the major
cause for the discrepancy.
Elemental composition of the wood fuel was also determined, using SSMS,
and the results are displayed in Table 7. With the exception of many conmon
elements (i.e., Fe, Mn, Ti, Ca, K, S, P, Si, and Mg which were detected at lev-
els > 100 |j,g/g), the data show very low concentrations (< 1 |ig/g) of most ele-
ments. Several notable exceptions include Ba, Sr, As, Zn, Cr, Cl, and Fl, whose
concentrations were in the range of 25 to 100 ng/g.
FUEL OIL ANALYSIS
Results of the ultimate analysis of the two No. 2 fuel oil samples are
shown in Table 8. The amount of ash in the oil was negligible (< 0.001%). Sul-
fur content averaged 0.35%. The heat of combustion of the fuel oil was deter-
mined and averaged 45,400 kJ/kg (19,520 Btu/lb or 138,100 Btu/gal) for the two
samples.
20
-------
TABLE 7.1 ELEMENTAL ANALYSIS OF WOOD BY SSMS
Concentration (PK/K)
Klumenc
llrunluui
Thorium
Bismuth
U-aJ
•Ilia II lum
Mercury
Cold
Platinum
Irldlum
Osm 1 urn
Rhenium
Tungsten
Tantalum
llaf itlum
l-utetlum
Ytterbium
Thulium
Erbium
lloliuium
Dysprosium
Ter lilum
Cudo 1 liilum
Europium
Samarium
Neodymlum
Praseodymium
Cerium
Lanthanum
Barium
Cutj lum
Iodine
Tel lurluu
Antimony
Tin
Indium
Cadmium
Silver
Palladium
Rhodium
Run \i'
0.
1
0.
4
0.
4
01
4
NK
-
-
-
-
-
0.
<
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
2
0.
4
4
68
0.
0.
.
5
0.
2
0.06
3
04
2
02
1
2
3
1
4
09
8
8
3
2
6
STl)
0.
0.
-
-
1
2
Kim I
Duplicate*
< 0.2
< 0.2
.
2
- '
NK
-
-
-
-
-
-
-
-
-
-
-
- -
-
-
-
-
0.03
O.I
0.1
0.05
0.4
0.6
25
0.3
O.I
-
0.4
Q.tt
STl)
-
0.1
-
-
Run I-1
0.2
0.4
O.t
4!l/
0.07
0 . 20£/
-
-
-
-
-
O.I
-
0.2
0.02
0.1
0.02
0.07
O.I
0.2
0.09
0.2
O.I
0.4
0.5
0.3
1
1
82i/
0.07
0.07
.
0.7V
2J)/
STI>
0.3
0.08
-
-
Range
< 0.2 -
< 0.2 -
< 0.02
2 - 4
< 0.02
-
< 0.02
< 0.02
< 0.02
< 0.02
< 0.02
< 0.02
< 0.06
< 0.02
< 0.02
< 0.02
< 0.02
< 0.02
< 0.02
< 0.02
< 0.02
< 0.02
0.03 -
0.4
1
- O.I
- 0.4
- 0.2
- 0.3
- 0.04
- 0.2
- O.I
- 0.2
- 0.3
- 0.1
- 0.4
O.I
0.1 - 0.8
0.1-2
0.05 -
0.4 - 1
0.6 - 1
25 - 1)2
0.07 -
0.07 -
< 0.02
0.7 - 5
0.6 - 2
-
< 0.02
0.08 -
< 0.02
< 0.112
0.8
0.3
0.2
- 0.3
0.2
K lenient
Ruthenium
Molybdenum
Niobium
Zirconium
Yttrium
Strontium
Rubidium
Bromine
Selenium
Arsenic
German lum
Gallium
Zinc
Copper
Nickel
Coba 1 t
Iron
Manganese
Chromium
Vanadium
Titanium
Scund lino
Calcium
Potassium
Chlorine
Sulfur
Phosphorus
Silicon
Aluminum
Magnesium
Sodium
Fluorine
Oxygen
Nitrogen
Carbon
Boron
Beryllium
Lithium
Run \S.<
.
0.7
2
7
2
57
6
0.6
0.9
7
0.8
5
43
14
29
49
MC
MC
12
16
MC
0.9
HC
HC
22
MC
HC
MC
> 16
MC
> 40
•-- 74
im
NK
NK
|
0.2
3
Concentration (MR/R)
Kun 1
Duplicate!/
.
0.3
0.3
3
0.6
15
16
0.8
0.3
48
0.2
15
35
25
10
0.8
MC
HC
21
2
MC
0.3
MC
MC
29
HC
MC
HC
> 23
MC
> 60
= 80
NR
Nil
NR
17
0.01
0.8
Run 1S.I
.
I
1
10
3
62
6
2
0.9
5
0.6
5
56
7
18
1
HC
HC
HC
19
MC
1
MC
HC
27
MC
MC
HC
> 19
MC
> 50
MC
NK
NR
NR
0.8
o.oa
I
Range
< 0.02
0.3 - 1
0.3 - 2
3-10
0.6 - 3
15-62
6-16
0.6 - 2
0.3 - 0.9
5-48
0.2 - 0.8
5 - 15
35 - 56
7 - 25
10 - 29
0.8 - 49
> 100
> 100
12 - > 100
2-19
> 100
0.3 - 1
> 100
> 100
22 - 29
> 100
> 100
> 100
> 16
> 100
> 40
74 - > 100
-
-
-
0.8 - 17
0.01 - 0.2
0.8 - 3
Note; NR « not reported.
STI> * Internal standard.
HC 3 major component, > 100
a/ Kl^tuuntu not UutcctuJ < 0.02 Pg/g.
li/ llccerogunuuut*.
c/ Flametess atomic absorption.
-------
TABLE 8. ANALYSIS OF NO. 2 FUEL OIL
Run No.
Date
1
7/24/78
2 and 3
7/25/78
Average
Ultimate analysis (%)
Ash
Carbon
Hydrogen
Oxygen
Sulfur
Nitrogen
Total
Heat of combustion (kJ/kg)
< 0.001
84.94
12.93
1.79
0.33
0.01
100.00
45,450
< 0.001
81.84
11.38
6.41
0.36
0.01
100.00
45,360
< 0.001
83.39
12.16
4.10
0.35
0.01
100.01
45,400
As with the wood fuel, the fuel oil was analyzed for elemental composi-
tion. All resulting concentrations, which are shown in Table 9, appear very
low. Many elements could not be detected at a lower limit of 0.06 |lg/g. The
only elements observed at concentrations greater than 10 ^g/g were Fe, Ca, S,
Si, and F. The highest single concentration reported was for iron at 25 |Ig/g.
BOTTOM ASH
The bottom ash generation rate, resulting from combustion of wood chips,
could only be estimated on the 2nd day of testing. As described in the previous
section, the total quantity of ash produced during the 12 hr of boiler opera-
tion was measured by weighing 55 gal* drums which contained the removed mate-
rial* By this method, a total ash weight of about 1,090 kg was obtained, which
corresponds to an ash generation rate of approximately 90.7 kg/hr. This amount
is just slightly over 1% of the wood chips feed rate (8.9 Mg/hr) on a wet
weight basis, or about 1.7% on a dry weight basis*
Elemental analysis of the bottom ash using SSMS was conducted and the data
are presented in Table 10* Almost without exception, the elemental concentra-
tions were much greater in the ash relative to the wood and oil fuels. The major
elements (Fe, Mn, Ti, Ca, K, S, P, Si, Al, Mg, and Na) were present at levels
greater than 1,000 (ig/g* Other elements (Ba, Zr, Si, and Li, in particular)
showed substantial increases in concentrations when compared to their levels
in the fuels* However, several of the more volatile elements, such as Sb, As,
and Se, exhibited very little or no increase in concentration, as expected.
22
-------
TABLE 9. \ELEMENTAL ANALYSIS OF FUEL OIL BY SSMS
rO
LO
Element Run 1—'
Uranium
Thorium
Illsmuth
Uad <0.2
Thallium
Mercury NR
(told
Platinum
Irldiim
Osmium
Rhenium
Tungsten
Tantalum
Hafnium
Ultctlum
Ytterbium
Thu 1 1 um
Erbium
Ho 1m lion
Dysprosium
Tc rb 1 um
Gadolinium
Europium
Samarium
Neodymlum
Praseodymium
Ce r 1 um
lanthanum
Barium £ 0. 1
Cesium
Iodine
Tellurium
Antimony < 0.04
Tin
Indium SID
Cadmium
Silver < 0.04
Palladium
Rhodium
Concentration (HK/c)
Run 1
Duplicate^' Run 2- Range
.
-
-
0.4 < 0.3 < 0.2 - 0.4
.
NR 0.281'
-
.
.
-
-
-
.
_•
.
-
-
-
.
.
-
.
-
-
-
.
-
-
0.6 <0.2 < 0.1 - 0.6
.
. .
.
<0.09 - < 0.09
.
STD STI)
.
, 0.09 - < 0.04 - 0.09
-
-
Concentration (MK/x)
Element
Ruthenium
Molybdenum
Niobium
7.1rconlum
Yttrium
Strontium
Rubidium
Bromine
Sc lenlum
Arsenic
Germanium
Gallium
Zinc
Copper
. Nickel
Cobalt
Iron
Manganese
Chromium
Vanadium
Titanium
Scandium
Calcium
Potassium
Chlorine
Sulfur
Phosphorus
Silicon
Aluminum
Magnesium
Sodium
Fluorine
Oxygen
Nitrogen
Carbon
Boron
Beryllium
l.lthlun
Run IS/
0.1
-
-
.
< 0.04
-
< 0.04
-
< 0.04
-
< 0.04
0.1
0.3
0.2
< 0.02
2
0.02
< 0.1
0.02
0.2
< 0.007
12
2
0.3
3
2
7
0.5
2
2
— 7
NR
NR
NR
< 0.09
-
0.01
Run 1
Duplicate^/
< 0.05
-
-
_
0.04
-
<0.1
-
0.1
-
< 0.07
0.2
0.3
2
0.02
25
0.3
0.8
0.3
< 0.6
<0.03
13
1
0.2
11
0.6
12
0.3
5
3
~ 2
NR
NR
NR
< 0.1
-
0.01
Run 2$J
< 0.06
-
-
.
< 0.06
-
<0.06
-
<0.06
-
< 0.06
0.4
0.4
2e/
0.01
3
0.07
0.2
0.1
< 0.6
<0.08
17
2
5^'
4
3
16
7£/
3
3
— 16
NR
NR
NR
<0.2
-
0.02
Range
< 0.1
-
-
.
< 0.06
-
< 0.1
-
< 0.04 - 0.1
-
<0.07
0.1 - 0.4
0.3 - 0.4
0.2 - 2
< 0.02
2-25
0.02 - 0.3
< 0.1 - 0.8
0.02 - 0.3
<0.6
<0.08
12 - 17
1 - 2
0.2 - 5
3-11
0.6 - 3
7-16
0.3 - 7
2-5
2 - 3
2-16
-
-
-
<0.2
-
0.01 - 0.02
Note: NR - not reported.
STO • Internal standard.
s./ All elements not detected < 0.04 Mg/g.
t>/ All cl orient s not delected < 0.05 Hg/g.
£/ All elements not detected < 0.06 I'g/g.
AJ Flameless atonic absorption.
e/ Heterogeneous signal.
-------
TABLE 10.. ELEMENTAL ANALYSIS OF BOTTOM ASH BY SSMS
Concentration (ctt/tO
Element
Uranium
Thorium
Bismuth
Lead
TlulJlJuiii
Mercury
Cold
I'lal Ilium
1 r 1 d 1 uu
Osmium
Khenltim
Tungsten
TantaJ uai
llafn lum
l.utet lum
Ytterbium
Thul luui
Kr blum
llolmlum
Dysprosium
Terb lum
(ladol Inlum
Luroplum
Samar turn
Neodyral um
Praseodymium
Cer lum
Lanthanum
Bar lum
Ce£ 1 uia
lod Ine
T*;l lur Juiii
Ant luiony
Tin
Indium
Cadmium
Silver
Palladium
Khodlinu
Run 1 Oi
9
16
-
8
-
NK
-
-
-
-
-
2
-
4
0.4
2
0.3
1
2
4
1
3
1
10
28
13
88
87
MC
6
-
-
4
2
STI)
0.6
0.3
-
-
HC = uia.)or cooiponellt , >
STI) • 1
Dashes
nternal standard
Run 1
ii|>) Icate
6
11
-
4
< 0.7
NK
-
-
-
-
-
3
-
3
0.3
2
0.4
2
2
3
0.8
2
1
11
12
6
68
27
IIC
3
-
-
2
2
STI)
< 0.4
< 0.2
-
-
1,000 HB/U.
Kun 2
4
8
.
4
.
0.092/
-
-
-
-
-
< 1
-
2
-
-
0.2
0.9
1
2
1
3
1
9
20
9
58
58
MC
7
-
-
1
2
STI)
< 0.4
0.4
-
-
Range
4-9
8-16
< 0.2
4-8
< 0.7
0.09
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 1 - 3
< 0.2
2-4
< 0.2 - 0.4
< 0.2 - 2
0.2 - 0.4
0.9 - 2
1 - 2
2-4
0.8 - 1
2-3
1
9-11
12-28
6-13
58 - 88
27 - 87
\> 1,000
3 - 7
< 0.2
< 0.2
1 - 4
2
-
£ 0.4 - 0.6
< 0.2 - 0.4
< 0.2
< 0.2
h/ lie terogeneouti.
Element
Ruthenium
Mol ybdenuiit
Niobium
Z 1 rconlum
Yttrium
St ront lum
Rubidium
Bromine
Sit 1 en 1 urn
Aruenic
Germanium
Call lum
7. lac
Copper
Nickel
Cobalt
1 ron
Manganese
Chromium
Vanad lum
Tf tan lum
Scand lura
Ca Ic turn
Potassium
Chlorine
Sulfur
Phosphorus
Silicon
Alurolura
Magnesium
Sod him
l-'luor ine
Oxygen
Nitrogen
Carbon
Boron
Beryl 1 lura
Lithium
'
Run 1
14
65
HC-'
140
MC
110
1
2
16
4
50
55
100
100
58
MC
MC
480
360
MC
JIO^'
MC
MC
21
MC
MC
HC
MC
MC
MC
^89
NK
NK
NK
71
8
530
Concentration (PK/R)
Run 1
Duplicate
7
30
340
69
940
90
2
3
11
6
35
17
88
48
27
HC
530
240
160
MC
20
MC
MC
33
MC
MC
MC
MC
HC
HC
=^ 180
NH
NR
NH
62
3
920
Kun 2
.
6
46
190
73
800
87
2
1
19
2
59
150
67
93
17
HC
MC
210
230
MC
13
HC
HC
50
HC
HC
MC
HC
HC
MC
= 200
NR
NR
NR
84
9
320
Range
< 0.2
6-14
30 - 65
l90-> 1,000
69 - HO
SOU<- 1,000
87 - 110
1 - 2
1 - 3
11-19
2-6
35 - 59
17 - 150
67 - 100
48 - 100
17-58
> 1,000
> 1,000
210 - 480
160 - 360
> 1,000
13 - 110
> 1,000
> 1,000
21 - 50
> 1,000
> 1,000
> 1,000
> 1,000
> 1,000
> i.ooo
89 - 200
-
-
-
62 - 84
3-9
320 - 920
Indicate not: defected (< 0.2 p.g/g)
-------
Bottom ash samples were also analyzed for PCB using electron capture gas
chromatography (EC/GC) and gas chromatography-mass spectrometry (GC/MS). No
PCB signals were obtained above the detection limit of 0.05 lig/g.
Analysis of the ash samples for PAH materials resulted in some positive
confirmations, as shown in Table 11• However, only phenanthrene could be veri-
fied as being present at levels exceeding the limit of detection. The signals
observed for fluoranthrene, pyrene, 1,2-benzanthracene, and benzo[ajpyrene
could not be positively confirmed due to the absence of observable confirma-
tory ion fragments.
.TABLE 11. ANALYSIS OF BOTTOM ASH FOR PAH BY GC/MS
Compound
Acenaphthylene
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Chrysene
1, 2-Benzanthracene
Benzo [ajpyr ene
Perylene
Indeno[l,2,3-c,d]pyrene
1,2,5,6 -Dibenzanthracene
1 , 12-Benzopery lene
Detection
limitS/
-------
UNCONTROLLED AIR EMISSIONS (COLLECTOR INLET)
Air emissions from the boiler were sampled and analyzed, at the collector
inlet, to determine pollutant constituents* Measurements included particulate
mass concentration, elemental composition of the mass samples, gas composition
(C02 and 02) and particle size distribution. Each of these aspects is discussed
below*
Particulate Mass Concentration
*
i
(
The concentration of particulate in the uncontrolled gas stream was mea-
sured by EPA Method 5* Three samples were collected and analyzed to yield the
data shown in Table 12. The average particulate concentration was 2.96 g/dscm
(1.30 gr/dscf)• On a heat input basis, the particular emission rate averaged
1.47 g/MJ (3.43 lb/106 Btu).
TABLE 12. UNCONTROLLED AIR EMISSIONS
Constituent
Particulate^/
g/dscm
gr/dscf
kg/hr
Ib/hr
g/MJ heat input
lb/106 Btu
1
3.32
1.45
241
531
1.75
4.08
Run No.
2
2.62
1.15
183
403
1.26
2.93
3
2.94
1.29
205
451
1.41
3.28
Average
2.96
1.30
210
462
1.47
3.43
Percent C02
Percent 02
Elemental composition
Particle size distribution
7.7 11.0
12.5 10.5
(see Table 13)
(see Tables 15-17)
6.0
14.0
8.2
12.3
a/ Results shown are for front half of the Method 5 train only.
26
-------
The emission values shown in Table 12 represent particulate collected by
the front half of the EPA Method 5 train since it is this portion that is nor-
mally reported and compared to emission regulations. An additional 0.09 g/dscm
(average) was measured in the back half of the sampling train* These condensi-
ble emissions represent only about 2.5% of the total particulate sample. A
complete sumnary of the particulate emission results is presented later in
this report with the discussion of outlet emissions.
Gas Composition
Oxygen and carbon dioxide concentrations in the boiler exhaust gases aver-
aged 12.3 and 8.2%, respectively. Oxygen contents ranged from 10.5 to 14.0%,
while the G02 content varied between 6.0 and 11.0%. Boiler excess air values,
calculated from these readings, ranged from 103 to 197% and indicated some
fluctuation in boiler operation. The lowest and highest excess air values were
observed during the second and third particulate runs, respectively, which were
both conducted on the 2nd day of the testing. For the first particulate sam-
pling run on the first day, the calculated excess air usage was 146%.
Particulate Elemental Composition
Portions of the Method 5 particulate filter samples were analyzed by SSMS
to determine the elemental composition of the uncontrolled particulate emis-
sions* Table 13 presents the resulting data.
Concentrations of many of the most common elements (Ca, K, Cl, S, Si, Al,
Mg, Na, F, and B) could not be quantified because background levels of these
elements from the glass fiber filter matrix exceeded the SSMS analytical range.
Of the remaining elements, the highest concentrations (> 100 pg/g) were
observed for Pb, Ba, Sr, As, Ga, Zn, Cu, Fe, Mn, Ti, and P. The emission rates
for these elements are shown in Table 14a (air concentrations) and Table 14b
(emission factors).
Particle Size Distribution
The size distribution of uncontrolled particulate emissions was measured
using an optical/diffusional particle counting system constructed by MRI. This
sampling method was selected in place of the more traditional cascade impactor
technique because it was anticipated that a majority of the emitted particu-
late would be in the submicron size range. The optical/diffusional system is
capable of providing data in size regions well below the lower limits of cur-
rently used impactors, while still operating comparably in the upper ranges.
By diffusion methods, particles can be sized down to 0.002 urn, whereas cascade
impactors can only yield data to about 0.1 ^m.
27
-------
TABLE 13. ELEMENTAL ANALYSIS OF INLET METHOD 5 FILTER PARTICULATE BY SSMS
NJ
oo
Concentration (lift/ft)
Element
Uranium
Thorium
Bismuth
biad
I1.nl Mum
Mercury
Gold
Platinum
Irldluis
Osmtuo
Rhenium
Tungsten
Tantalum
llnfnlum
Uitetlius
Ytterbium
Thul |IM>
Erbium
Holmlum
Dysprosium
Terbium
Gadol Inlum
Europium
Samarium
Ncodymlum
Run |i'
Run 1
Duplicate^'
Run 2£/ Run \i/
Range element
Run 1*'
Concentration (UK/H)
Run 1
Duplicate^'
Run 2£'
Run 3S/
1 - < 0.03 - 1 Ruthenium - ...
_
<0.03£'
200
5
11. Oi/
.
3001'
80
50
300I'
1
B
B
g
B
>300I'
B
a
B
B
B
NR
NR
NR
B
< 0.032'
2
3£'
2
20
5
300
40
40
20
500
10
50
> sool'
200
40
<0.05i'
>5001'
>200
< 0.051'
40
> 5001/
2
B
B
B
B
> sooi'
B
B
B
B
B
NR
NR
HR
B
0.092'
50
10
8
40
20
300
40
80
40
>400I/
30
300
>400l/
100
40
< 0.042'
>4001/
> 200
<0.04l/
30
>40ol/
4
«ool/
NR
NR
NR
B
0.6«/
20
20
7
70
30
300
10
70
40
100
30
50
>70fli/
200
70
< 0.041/
>700i/
400
< 0.07S'
50
700l/
700«'
NR
NR
NR
B
< 0.072'
2
Range
< 0.07
3-20
2 - B
20 - 70
5 - 30
80 - 300
10 - 40
40-80
20 - 40
100 - > 400
10 - 30
50 - 300
>300
80 - 200
40 - 70
< 0.01
> 300
>200
< 0.07
10 - 50
> 300
1 - 7
-
-
-
-
> 300
-
-
-
-
-
-
-
-
.
<0.03 - 0.6
2-50
< 0.07
not reported.
• Internal standard.
r 8
b/ Ikitect
Ion llolt • 0.05 vg/g. f/
£/ IIQlectlun llBll — w.v— fa'S' B.r
d/ ivteetlon limit - 0.07 ug/g. .
a
> 20X of
Major component.
-------
TABLE 14a. AIR EMISSION CONCENTRATIONS FOR SELECTED ELEMENTS
IN UNCONTROLLED PARTICULATE
Concentration (ue/dscm)
Element
Lead
Barium
Strontium
Arsenic
Gallium
Zinc
Copper
Iron
Manganese
Titanium
Phosphorus
Mercury
Run 1
38
57
"15
38
10
2 57
15
2 57
38
* 57
* 57
5.9
Run 2
34
69
51
2 69
51
2 69
17
* 69
2 34
* 69
2 69
17
Run 3
11
22
34
11
6
2 79
22
2 79
45
* 79
* 79
6.2
Range
11-38
22-69
15-51
11-2 69
6-51
2 57
15-22
2 57
2 34
2 57
* 57
6-17
TABLE
14b. AIR EMISSION
FACTORS
FOR SELECTED ELEMENTS
IN UNCONTROLLED P ARTICULATE
Element
Lead
Barium
Strontium
Arsenic
Gallium
Zinc
Copper
Iron
Manganese
Titanium
Phosphorus
Mercury
Run 1
20
30
7.9
20
5.3
2 30
7.9
2 30
20
a 30
2 30
3.1
Emi s s i on
Run 2
16
33
25
2 33
25
2 33
8.2
2 33
* 16
* 33
i 33
8.2
factor (mg/MJ)
Run 3
5.3
11
16
5.3
2.9
* 38
11
a 38
22
2 38
i 38
3.0
Average
14
25
16
2 19
11
2 34
9
2 34
19
2 34
2 34
4.8
29
-------
The original intention of the MRI sampling team was to use the particle
counting system in its unrestricted mode to obtain data over the full particle
size spectrum* However, during the first sampling run it was discovered that
the larger particles in the air stream were causing plugging in the dynamic
dilution/conditioning system. To alleviate this problem a cyclone was added
to the front of the system (see Appendix B). Since the cyclone has a particle
diameter cutoff of 2.65 y,m, those particles larger than 2«65 pan were captured
and prevented from entering the particle counting system. Thus, this portion
of the particle size spectrum is not included in the reported data*
The data obtained from the optical/diffusional system are presented in
Tables 15 through 17* Particles in the range of 0«3 to 2.6 |im were counted by
the optical system which segregated the particle counts into six discrete
ranges, as shown in Table 15* These data indicate a progressive increase in
the number of particles observed as the particle size decreases*
The diffusion battery/condensation nuclei counter data shown in Tables
16 and 17 represent counts of particles in the range of about 0*005 to 0.10
p,m* Values in Table 16 are the measurement output of the nuclei counter as the
sample stream was passed through the cumulative stages of the diffusion bat-
tery* Fractional penetration data were calculated from the corresponding par-
ticle concentrations* Using this information and a graphical stripping tech-
nique, the size distribution data presented in Table 17 was developed* The size
ranges listed vary for each run because of limitations imposed by the graphical
data reduction method* An average of 47% of the particles measured by the con-
densation nuclei counter were larger than 0*040 \m, while another 40% were be-
tween 0,025 and 0.040 \m in diameter* Only 1 to 2% of the measured particles
were less than 0.02 (im.
TABLE 15. INLET PARTICLE SIZE DISTRIBUTION BY NUMBER AS
MONITORED BY THE OPTICAL COUNTER
Particulate concentration
CIO6 particles/cu m)SJ
Size range (nm)
Run 1
Run 2
Run 3
Average
Channel 6 (1.5-2.6)
Channel 5 (1.2-1.5)
Channel 4 (0.9-1.2)
Channel 3 (0,7-0.9)
Channel 2 (0.5-0.7)
Channel 1 (0.3-0.5)
59.8
98.0
200
354
746
2,250
10.8
22.2
50.8
105
250
943
13.1
25.9
58.5
115
239
913
27.9
48.7
103
191
419
1,370
a/ At standard conditions (29.92 in. Hg and 68°F).
30
-------
(-0
TABLE 16. INLET PARTICLE COUNTS BY SIZE AS MONITORED BY THE DIFFUSION
BATTERY/CONDENSATION NUCLEI COUNTER
Diffusion
battery
port No*
0
1
2
3
4
5
6
7
8
9
10
Particulate concentration
(106 particles/cu m)2/
Run 1
1,795
1,795
1,795
1,346
898
448
216
107
45
20
13
Run 2
1,520
1,094
1,016
760
577
182
106
27
15
12
7
Run 3
1,368
1,216
1,216
760
380
152
76
37
15
10
6
Average
1,561
1,368
1,342
955
452
261
133
57
25
14
7
Fractional
Run 1
1.000
1.000
1.000
0.750
0.500
0.249
0.120
0.060
0.025
0.011
0.007
Run 2
1.000
0.720
0.668
0.500
0.380
0.120
0.070
0.018
0.010
0.008
0.005
penetration
Run 3
1.000
0.889
0.889
0.556
0.278
0.111
0.056
0.027
0.011
0.007
0.004
Average
1.000
0.876
0.860
0.612
0.290
0.167
0.085
0.037
0.016
0.009
0.006
a/ At standard conditions (29.92 in. Hg and 68 F).
-------
TABLE 17. INLET PARTICLE SIZE DISTRIBUTION RESULTING FROM
DIFFUSION BATTERY PENETRATION DATA
Size range
Run No* <|im)
1 > 0.045
0.030-0.045
0.025-0.030
< 0.025
2 > 0.055
0.025-0.055
0.020-0.025
0.010-0.020
< 0.010
3 > 0.035
0.020-0.035
< 0.020
Average > 0.040
0.025-0.040
0.020-0.025
< 0.020
No. of particles!/
(106 particles/cu m)
718
790
269
18
155
784
456
58
67
711
595
62
734
624
184
19
Percent of particles
in stated size
range
40.0
44.0
15.0
1-P.
10.2
51.6
30.0
3.8
4.4
52.0
43.5
4.5
47.0
40.0
11.8
1.2
a/ At standard conditions.
Graphical representations of the optical and diffusional sizing results
are presented in Figures 4 through 6. A definite data gap is evident between
the lower limit of the optical range and the upper limit of the diffusional
range. No extrapolation of the curves was made because of the difficulty in
accurately predicting the changing curve slope in the area of the disfunction•
PRIMARY AND SECONDARY COLLECTOR ASH
Samples of collected fly ash were taken from both the primary and second-
ary collector hoppers each day. Analyses were then conducted to determine ele-
mental composition and concentrations of PCB and PAH compounds.
32
-------
I010
u
a
108
3
U
L DIFFUSION AL •»! L OPTICAL
1Q7, 1 1 1
0.01 0.1 1.0 '0-0
PARTICLE DIAMETER,^.m
Figure 4. Inlet particle size distribution - Run 1
(optical and diffusional).
33
-------
10
10
U DIFFUSIONAL J
• OPTICAL
o
a
.03
5
5
107
0.01
O.I 1.0
PARTICLE DIAMETER,/*™
10.0
Figure 5. Inlet particulate size distribution - Run 2
(optical and diffusional).
34 :
-------
10'°
U DIFFUSIONAL—«4 U OPTICAL
s
-------
During the planning for this test program, it was anticipated that the
quantities of fly ash captured by the tandem collectors could be measured on
a unit time basis* But because of the pneumatic ash removal system, a direct
measurement of fly ash quantities was not possible* As an alternative measure-
ment method, the removal system was blocked off and the hoppers were allowed
to fill over the course of the test period. At the end of the test day when
the boiler was shut down, the depth of material in the hoppers was to be mea-
sured and the quantities of fly ash calculated* However, at the conclusion of
the first test day, the ash levels in the primary hoppers were well above the
access doors which precluded access to the interiors of the hoppers for mea-
surement of ash depth*
Personnel from UOP/Air Correction Division, the suppliers of the mechani-
cal collectors, were at the test site on the 2nd day and provided assistance
in this effort* When testing was completed and the boiler shut down, the UOP
crew evacuated the secondary collector hoppers via a vacuum system and quanti-
tated the material using 19 liter drums* The primary collector hoppers could
not be emptied by this method because of the large amount of ash involved*
The ash capture rate of the secondary collector was determined to be 4*9
kg/hr by UOP* The primary catch can be estimated using results of the MRI par-
ticulate tests* The difference between the inlet and outlet emission rates was
183*9 kg/hr, which should correspond to the total amount of fly ash collected
by both multiclones* Therefore, the primary collector capture rate should have
been approximately 179 kg/hr* If the same primary/secondary capture ratio is
assumed for the first test day, the ash collection rates would be 222 kg/hr
for the primary collector and 6.3 kg/hr for the secondary collector* The rea-
son for the greater amounts of collector ash on the first day could not be
readily explained*
SSMS analysis results showing elemental composition of the primary and
secondary collector ashes are presented in Tables 18 and 19, respectively.
In general, there were distinct increases of elemental concentrations in the
secondary ash relative to the primary ash. The elements detected at the high-
est concentrations were basically the same elements that were most prominent
in the bottom ash* Those identified as major components (> 1,000 ng/g) included
Fe, Ti, Ca, K, S, P, Si, Al, Mg, and Na. These elements were present as major
components in both primary and secondary ash. In the secondary ash, additional
major elements were Ba, Sr, Mi, and F. Other elements of interest which were
observed in the secondary ash at moderately high concentrations were As, Ni,
Cr, V, and Cl* Levels of lead, mercury, antimony, and beryllium were low (< 10
Hg/g) in both ashes*
Table 20 presents a comparison of elemental concentrations in the bottom
ash, primary collector ash, and secondary collector ash* This comparison is
noteworthy because it is apparent that many elements are most concentrated in
the secondary collector ash, which has the smallest mean particle size of the
three ash types*
36
-------
TABLE 1ft. ELEMENTAL ANALYSIS OF PRIMARY COLLECTOR ASH BY SSMS CONCENTRATION (^g/g)
1
Element
Uranium
Thorium
Ulmuuth
Lead
Thallium
Mercury
Cold
t'lutlmiiu
Irldlunt
Omulinu
Hlienlum
Tungbtcn
Tantalum
llal'nf um
l.utei luiu
Ytterbium
Thulium
L'rbiuw
lloliulum
Dysprosium
'1'erbliim
(Jailol Inium
Kurjpium
Sumar 1 um
Neodymlum
1'raseudyiiilum
Cerium
Lanthanum
Barium
Ces iiim
Iodine
Te 1 1 u r 1 um
Ant itiiony
Tin
1 nd 1 uin
CadmiuiD
Sliver
I'al ladlu.u
Khud 1 um
Kim 1
6
10
< 0.02
9
< 0.02
NK
< 0.02
< 0.02
< 0.02
< 0.02
< 0.02
1
< 0.02
1
0.3
2
0.2
1
2
3
1
3
1
I)
12
b
31
25
MC
1
2
< 0.02
±1
2
STU
0.7
0.2
< 0.02
< 0.02
Concent rat ji
Kim 1
Dupl Kale
3
7
< 0.02
8
< 0.02
NK
< 0.02
< 0.02
< 0.02
< O.U2
< 0.02
< 0.02
< 0.02
1
< 0.02
< 0.02
0.2
0.7
)
2
0.6
2
0.8
f,
8
4
25
22
510
1
2
< 0.02
2
1
STI)
< 11.3
o.:>
< 0.02
< 0.02
HI (I'K/K)
Kuilt, 2
und I
< 2
2
< 0.2
4
< 0.2
0.66!!/
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< O.2
< 0.2
0.2
O.b
0.5
3
9
3
31
22
450
0.8
1
< 0.2
1
0.6
STI)
0.4
< 0.2
< U.2
< 0.2
Concent rut Inn (MM/K)
Kan^e
S 2 - 6
2 - 10
< 0.2
4-9
< 0.2
-
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2
< 0.2 - 1
< 0.2
<0.2 - 1
< 0.02 - 0.3
< 0.02 - 2
< 0.2 - 0.2
< 0.2 - I
< 0.2 - 2
< 0.2 - 3
0.2 - 1
0.6 - 3
0.5 - 1
3-11
8-12
3-6
25 - 31
22 - 25
450-> 1.000
0.8 - 1
1 - 2
< 0.2
1 - 2
0.6 - 2
-
« 0.3 - 0.7
< 0.2 - 0.2
< 0.2
< 0.2
K lenient
Kuthenliim
Mol ybdenum
Niobium
Zirconium
Yttrium
Strontium
Kuhldlum
Bromine
Selenium
Arsenic
(icrmanlutii
Call 1 um
£inc
Copper
Nickel
Colin 1 1
Iron
Manganese
Chromium
Vannd lum
Tl tan lum
Scandium
Ca Ic lum
Hot ass lum
Chlor ine
Sulfur
Hlin.snliiirui)
Silicon
Aluminum
Ma^neulum
Sod lum
Fluor Ine
Oxygen
Nitrogen
Carbon
Uoron
Beryll lum
Lithium
Knn 1
< 0.02
6
12
120
31
380
28
17
7
99k/
5
2b
39
64
25
8
MC
350
120
140
HC
14
HC
MC
600^
MC
MC
MC
MC
MC
MC
= 840
NK
NK
NK
100
4
74
Knn 1
Dupl Icate
< 0.02
5
9
61
18
460
41
19
10
60
5
18
62
64
22
7
HC
720
110
130
HC
12
MC
MC
96
HC
MC
MC
HC
MC
MC
= 330
NK
NK
NK
40
3
30
Kuns 2
and 3
< 0.2
592/
11
55
28
460
15
24
10
58
5
28
110
63
35
11
MC
270
110
160
HC
12
MC
MC
120
MC
HC
HC
HC
MC
HC
= 250
NK
NK
NK
16
2
20
Kuilye
< 0.2
5-59
9-12
55 - 120
18 - 31
380 - 460
15 - 41
17-24
7 - 10
58 - 99
5
18 - 28
39 - 110
63 - 64
22 - 35
7-11
> 1,000
270 - 720
110 - 120
130 - 160
> 1,000
12 - 14
> 1,000
> 1,000
96 - 600
> 1,000
> 1,000
> 1.000
> 1,000
> 1,000
> 1,000
250 - 840
-
-
-
16 - 100
2-4
20 - 74
Note; NK a not tc|>orLud
MC « muior component, > !,0()U
STI) * liUotiutl bi;mdurd
b^ atomic uhiiorpt ion
-------
TABLE 19. > ELEMENTAL ANALYSIS OF SECONDARY COLLECTOR ASH BY SSMS
oo
Concent rat Ion (HH/K)
t'jemeiit
Uranium
Thorium
bismuth
I-ead
•Ilia 11 lum
Mercury
Cold
Platinum
Irldlum
Ouiuium
Rhenium
Tungsten
Tantalum
lla Irnlum
Lutet lum
Ytterbium
Thulium
Erbium
llolmlum
Dysprosium
Tei-blum
Gndol In lum
Europium
Sauiarlfuu
Neodymiuiu
Klin 1
17
25
< 0.6
5
< 0.6
NR
< 0.6
< 0.6
< 0.6
< 0.6
< 0.6
5
< 0.6
< 0.6
0.7
4
1
3
3
4
2
6
3
15
33
Praseodymium 15
Ce r I uiu
Lanthanum
Barium
Cesium
Iodine
Tc 1 1 ur lum
Antimony
Tin
Indium
Cadmium
Silver
Palladium
Rhodium
Note: NR
MC
STI)
170
61
MCU/
3
9
< 1
14
5
STD
2
< l
< 0.6
< 0.6
x. not reported.
» major component ,
Run 1
Duplicate
11
23
< 0.6
4
< 0.6
NK
< 0.6
< 0.6
< 0.6
< 0.6
< 0.6
3
< 0.6
7
< 0.6
3
0.5
2
3
4
1
4
2
14
19
9
99
94
MC
3
3
< 0.6
11
5
STD
1
< 0.6
< 0.6
< 0.6
> i.ooo im/8.
- Internal standard.
Runs 2
and 3
7
15
< 0.4
3
< 0.4
5.87i/
< 0.4
< 0.4
< 0.4
< 0.4
< 0.4
3
< 0.4
3
0.5
3
0.4
2
2
4
0.9
2
1
9
22
16
110
100
MC
7
0.9
< O.tt
16
4
STD
1
0.5
< 0.4
< O./i
Range
7 - 17
15 - 25
< 0.6
3-5
< 0.6
-
< 0.6
< 0.6
< 0.6
< 0.6
< 0.6
3-5
< 0.6
< 0.6 - 7
< 0.6 - 0.7
J - 4
0.4 - 1
? - 3
2-3
4
0.9 - 2
2-6
1 - 3
9 - 15
19 - 33
9 - 16
99-170
61 - 100
> 1,000
3 - 7
0.9 - 9
< 1
11-16
4-5
-
1 - 2
< 1
< 0.6
< 0.6
1 1 b 1
Element
Ruthenium
Molybdenum
Niobium
Zirconium
Yttrium
Strontium
Rubidium
Bromine
Selenium
Arsenic
Germanium
Gallium
Zinc
Copper
Nickel
Cobalt
Iron
Manganese
Chromium
Vanadium
Titanium
Scandium
Calcium
Potass iuoi
Chlorine
Sulfur
Phosphorus
Silicon
Aluminum
Magnesium
Sodium
Fluorine
Oxygon
Nitrogen
Ca rbon
Boron
Beryllium
Lithium
Run 1
< 0.6
33
38
190
96
940
42
28
72
380
16
97
48
99
240
56
MC
MC
93
440
MC
43
MC
MC
330
MC
MC
MC
MC
HC
MC
MC
NR
NR
NR
270
5
100
Concentration (IIK/K)
Run 1
Duplicate
< 0.6
34
78
190
89
MC
59
26
83
630
31
220
89
230
280
69
MC
MC
620
MC
MC
79
MC
MC
380
MC
MC
MC
MC
MC
MC
MC
NR
NR
NR
140
8
42
Runs 2
and 3
< 0.4
51
52
130
150
MC
98
31
140
940
41
150
130
150
370
46
MC
MC
410
4 SO
MC
53
MC
MC
340
MC
MC
MC
MC
MC
MC
MC
NR
NR
NR
190
5
28
Kangti
< 0.6
33 - 51
38 - 78
130 - 190
89 - 150
> 940
42 - 98
26 - 31
72 - 140
380 - 940
16 - 41
97 - 220
48 - 130
99 - 230
240 - 370
46 - 69
> 1.000
> 1.000
93 - 620
440-> 1,000
> 1,000
43 - 79
> 1,000
> 1.000
330 - 380
> 1.000
> 1,000
> 1,000
> 1,000
> 1,000
> 1.000
> 1.000
> 1,000
-
-
140 - 270
5 - 8
28 - 100
b/ Heterogeneous .
-------
TABLE 20.' COMPARISON OF ELEMENTAL ANALYSES OF ASH MATERIALS
LO
vO
Clement
Uranium
Thori urn
Bismuth
Lead
Thull lum
Mercury
(Jo Id
I'lutinum
Irldluio
OsilllUIH
Rhenium
Tungsten
Tuntaluin
Hafnium
l.utct lum
Ytterbium
Thulium
Erbium
llolmluui
Dysprosium
Terbium
Cudol Inl urn
Kurop 1 uui
Siiinur ium
Neodynil um
1' raseo J yin 1 um
Cerium
Lanthanum
Uur luu)
Cesium
Iodine
Tellurium
Ant liuuny
Tin
I ml luu
Cmllullim
Silver
Palladium
llhodliia
Average3'
Bottom
Ash
7.5
13.5
-
6
_
(0.09)-'
-
-
-'
-
-
2.5
-
3.5
0.35
2
0.35
1.5
2
3.5 "
0.9
2.5
1
10.5
20
9.5
78
57
MC
4.5
-
-
3
2 .
STD
-
-
-
-
Avera oS'
I'rlmary
Col lector Ash
4.5
8.5
-
8.5
_
(0.66)^'
-
-
-
-
-
-
-
1
-
-
0.2
0.85
1.5
2.5
0.8
2.5
0.9
8.5
10
5
28
24
MC
1
2
-
2
1.5
STD
-
0.2
-
-
Aveiagui'
Si.'condury
Collector Ash
14
24
-
4.5
_
(5.87)^
-
-
-
-
-
4
-
-
_
3.5
0.75
2.5
3
4
1.5
5
2.5
14.5
2h
12
135
78
MC
3
6
0
12
5
STD
1.5
-
-
-
Element
Kuthenluu
Molybdenum
N loh lum
Zirconium
Yttrium
St ront lum
Rubidium
Bromine
Sel en tun
Arsenic
Cermanlum
Call lum
Zinc
Copper
Nickel
Cobu 1 t
Iron
Manganese
Chromium
Vanad 1 urn
Titanium
Scand I urn
Calc lum
I'otasslum
Chlorine
Sulfur
1'hosphorus
Silicon
A lum lum
Magnesium
Sodium
Fluorine
Oxygen
Nl trogen
Carbon
Boron
Bery 11 ium
Lithium
Average^
Bottom
Ash
10
48
MC
104
HC
100
1.5
2.5
14
5
42
36
94
74
42
MC
HC
360
260
HC
65
MC
HC
27
MC
HC
HC
HC
HC
HC
134
NO
NK
NR
66
6
720
Average—
Primary
Collector Ash
.
6
10
90
24
420
34
18
8.5
80
5
22
50
64
24
8
HC
540
120
140
MC
13
MC
MC
350
MC
HC
HC
MC
MC
MC
580
NR
NK
NR
70
4
52
Averag<£/
Secondary
Collector Ash
34
58
190
92
HC
50
27
78
510
24
160
68
160
260
62
HC
HC
360
MC
MC
61
MC
MC
360
MC
MC
MC
MC
HC
MC
MC
NK
NR
NK
205
6
71
Nute: NK - not reported.
MC - oiajur component, > 1,000
STD " Internal standard.
n_/ Baued on an average of two values.
b/ Baaed on one value.
-------
Another interesting trend is exhibited by Hg, Br, Se, As, Cl, F, and B.
Each of these elements shows progressively increasing concentrations from bot-
tom ash to primary collector ash to secondary collector ash. Since these ele-
ments are relatively volatile, their presence in the primary and secondary
collector ash suggests that a condensation mechanism may be responsible for
the observed increases in concentrations with decreasing particle size.
Analysis of the collector ash samples for PCB materials did not yield any
confirmed responses. The primary ash samples were eliminated from further con-
sideration after the initial EC/GC screening. Secondary ash samples were also
analyzed using GC/MS, but no positive signals were observed at the 0.05 p,g/g
detection limit.
The primary and secondary collector ashes were also analyzed for PAH com-
pounds. No PAH materials were detected in any of the primary ash extracts. How-
ever, positive confirmation of several compounds resulted from analysis of the
secondary collector ash. These data are shown in Table 21. The compounds iden-
tified were acenaphthylene, phenanthrene, fluoranthene, and pyrene. The highest
sample concentration observed (10 (ig/g) was that of phenanthrene in the second-
ary ash sample collected on the 2nd day of testing.
CONTROLLED AIR EMISSIONS (COLLECTOR OUTLET)
The air stream emitted from the mechanical collectors was extensively
characterized due to its importance as a pollutant stream. Flue gas parameters
assessed for the test program included:
• Gas composition (0-, CO-, NO , SO-, CO, and HC);
• Particulate mass concentration;
• Particulate elemental analysis;
• Particle size distribution;
• Opacity;
• PCB and PAH compounds; and
• EPA Level 1 assessment using SASS.
Results of these analyses are discussed individually below.
40
-------
TABLE 21. ANALYSIS OF SECONDARY COLLECTOR ASH FOR PAH BY GC/MS
Compound
Acenaphthylene
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Chrysene
1 , 2-Benzanthracene
Benzo [ajpyrene
Perylene
Indeno [l,2,3-c,d ]py r ene
1,2,5 , 6-Dibenzanthracene
1,12-Benzoperylene
Detection
Iimit2/ Run lk/
(iig/g) (n-g/g)
0.56£/ 2.2
0.54£/
0.24C/ 6.0
0.15S/
0.172/ 1.6
0.16S/ 0.9
0.50^7
0.472/
6.7£./
0.561/
1 .5l/
2.2l/
1.4l/
Run 1
duplicate
2.5
_
3.5
—
0.62
0.32
_
^
.
-
-
-
-
Run 2k/
_
10
_
3.5
1.0
_
< 0.47S/
.
-
-
-
-
a/ Assinning a 20 g sample weight and a 1,000 counter minimum signal.
b/. Values corrected to D-10 anthracene internal standard.
c/ Based on response factors determined using 1 u,l of a 1 ng/ul standard.
d/ Based on response factors determined using 1 |il of a 5 ng/y.1 standard.
e/ Compound identity could not be verified by GC/MS.
Gas Composition
During the course of air emissions testing, the exhaust gases were con-
tinuously monitored for levels of 02, NOX, S02» CO, and THC. Concentrations of
C02 and 02 were also periodically measured using Orsat and Fyrite gas analysis.
The continuous monitors were operated during the entire period of testing each
day. Results are displayed in Table 22.
Oxygen and carbon dioxide levels remained relatively constant throughout
the test program. Concentrations of 02 in the exhaust stream, as determined by
the continuous monitor, averaged 13.0 and 12.1% on the first and 2nd days, re-
spectively. The Fyrite analysis yielded slightly different values of 12.7 and
41
-------
TABLE 22.! BOILER EXHAUST GAS COMPOSITION*/
Date
7/24
7/25
Time
11:45-17:00
09:00-16:00
02
Avg.
13.0
12.1
tt)
Range
12.6-13.3
11.4-12.6
C02
Avg.
7.9
7.3
M±f
Range
7.9
6.8-7.9
SO2
Avg.
155
120
(ppra)£/
Range
110-230
90-160
CO
Avg.
180
245
(ppm)
Range
135-225
125-400
THC (ppra)l/
Avg. Range
7 2-13
10 4-21
NOX (ppm)
Avg. Range
68 60-77
64 57-71
a/ Gas samples were drawn from the duct downstream from the mechanical collectors and were analyzed by continuous
monitoring equipment. Averages were computed from data points transcribed from the strip chart records at 15
mln Intervals.
b/ Determined with a Fyrlte analyzer.
c/ H20 Interference contributed about 20 ppm to all S02 readings. Values shown have been corrected for this dis-
crepancy.
d/ Total hydrocarbon results were reported as propane.
-------
12.97o, by comparison. Fyrlte analyzers, however, are usually only accurate
to within about 170. CC>2 concentrations averaged 7.9% on the first test day and
7.37o on the second*
Excess air in the gas stream downstream of the collector was calculated
using the Fyrite data and showed very little variation (+ 270) during the study.
The average value for measurements which correspond to the three particulate
tests was 1557. excess air. Since the excess air values determined upstream of
the mechanical collectors varied between 103 and 197%, these observed readings
may have been the result of incomplete mixing of the flue gases*
Concentrations of CO appeared to be a little high, averaging ISO ppm the
first day and 245 ppm on the second. Readings as high as 400 ppm were observed
during the last day of testing. It is possible that incomplete combustion of
some of the wood chips could have led to elevated CO levels, although it would
seem that the overfire air introduced with the fuel oil above the combustion
grate should have aided in completing the combustion of any suspended wood par-
ticles. Nonetheless, some charred, wood-like particles were observed in the
primary ash samples.
S02 concentrations were fairly low, and most likely would have been even
lower if the wood chips were not "contaminated" with coal dust particles as
described earlier. Theoretically, if it is assumed that all of the sulfur in
wood (0.17o) is released as S02, the S02 concentration should have been only
about 38 to 46 ppm if wood were used as the sole fuel. Similarly, theoretical
maximum emissions of S02 from combustion of the fuel oil alone during testing
would have been about 10 ppm.
Emissions of nitrogen oxides and THCs were both very low. NO levels
varied from 57 to 77 ppm over the two test days, while THC concentrations were
in the range of 2 to 21 ppm.
Particulate Mass Concentration
Results of the EPA Method 5 particulate concentration determinations are
presented in Tables 23 and 24 in both metric and English units. The uncontrolled
emission data are included for comparative purposes.
The total particulate concentration averaged 0.177 g/dson (0.077 gr/dscf),
of which 0.143 g/dson was filterable material. On a heat input basis, the total
emission rate averaged 0.094 g/MJ (0.166 Ib/lO^ Btu).
On August 12, 1978, the Vermont Agency of Environmental Conservation adopted
new air pollution control regulations for wood fired boilers (where wood fuel
contributes 5070 or more of the total Btu input) operating in the state. For Unit
No* 1 at Burlington Electric, which commenced operation with wood fuel prior to
December 5, 1977, the particulate emission limitation is 0.45 grains per dry
43
-------
TABLE 23. / SUMMARY OF PARTICULATE TEST RESULTS (METRIC UNITS)
Flue cag parameters
Test No. Date Location
1 7/24/78 Inlet
Outlet
2 7/25/78 Inlet
Outlet
3 7/25/78 Inlet
Outlet
Average Inlet
Outlet
Moisture
content
Ci)
10.0
9.3
11.9
10.4
12.0
12.1
11.3
10.6
Avg.
temp.
C'C)
154
146
159
158
167
154
160
153
Flow
rate
(dscnnt)
1,210
1,301
1,162
1,246
1,160
1,253
1,177
1,267
Part iculate
concent ral Ion
(j>/dficra)
* o2
12.5
12.7
10.5
12.8
14.0
12.9
12.3
12.8
* C02
7.7
7.9
ll.O
7.4
6.0
7.1
8.2
7.5
Front
ha If i<
3.32
0.163
2.62
0.173
2.94
0.093
2.96
0.143
Total£/
3.36
0.192
2.77
0.204
2.97
0.134
3.03
0.177
Collector
Partlculate emission rate efficiency
(kK/l>r>
Front
half
241
13
183
13
205
7
210
11
Total
244
15
193
15
207
10
215
13
(a/KJ) «)
Front
half
1.75
0.095
1.26
0.089
1.41
0.048
1.47
0.077
Front
Total half
1.78
0.109 95.1
1.33
0.103 93.4
1.43
0.069 96.8
1.51
0.094 95.1
Total
94.3
92.6
95.5
94.2
g/ "Front half* refers Co filterable partlculatc only (l»e*f filter cntch plus rinses of probe and connections)*
b/ "Total" values Include contents of iropinger solutions (condenslblcs)*
-------
TABLE 24.; SUMMARY OF PARTICULATE TEST RESULTS (ENGLISH UNITS)
Flue eas par.Tmoters
Te.it No. Date Location
I 7/24/78 Inlot
Outlet
2 7/25/78 Inlet
Outlet
3 7/25/78 Inlet
Outlet
.!
: Average Inlet
|i Outlet
Mo is cure
content
(X)
10.0
9.3
11.9
10.4
12.0
12.1
11.3
10.6
Avg.
temp.
C'F)
310
294
318
317
313
309
320
307
Flow
rate
(dscfin)
42,700
46,000
41,000
44,000
41,000
44,200
41,600
44,700
%02
12.5
12.7
10.5
12.8
14.0
12.9
12.3
12.8
7. C02
7.7
7.9
1 1.0
7.4
6.0
7.1
8.2
7.5
Pnrt iculal.e
concent rat Ion
-------
standard cubic foot (gr/DSCF) of exhaust gas corrected to 12% C(>2» The average
emission concentration measured by MRI was 0,10 gr/DSCF» Therefore, the boiler
easily complies with the new state regulations.
Measured efficiencies of the mechanical collection system were very good,
averaging 95.1% of the filterable particulate. Values ranged from 93.4 to
96.8%. When condensible particulate was included, the average efficiency dropped
slightly to 94.2%.
Particulate Elemental Composition
After gravimetric analysis of the Method 5 filter particulate, elemental
composition of the samples was determined using SSMS. Table 25 presents the
resulting data.
As was the case with SSMS analysis of the inlet particulate filter samples,
several elemental concentrations were obscured by background levels in the fil-
ter substrate. The concentrations of those elements which could be quantified
were generally higher than corresponding levels in the other particulate sam-
ples (bottom ash, collector ash, or inlet particulate). Elements with concen-
trations greater than 100 ng/g, as well as some elements whose hazard potential
is of interest, are shown in Table 26. The data in the table are presented on
the basis of an air emission concentration (|o,g/dscm).
A review of the data in Table 26 indicates moderate air concentrations for
most of the elements listed. Those approaching 100 ^g/dscm are lead, barium,
strontium, zinc, and titanium.
Particle Size Distribution
The particle counting system employed at the mechanical collector outlet
was identical to that used at the inlet. The mode of operation was also identi-
cal.
Table 27 summarizes the data output of the optical counter portion of the
sizing system, while data from the diffusional portion is shown in Tables 28
and 29. Similar to the inlet data, no values are available for the size region
above 2.6 y,m.
Figures 7 through 10 are graphical representations of the outlet particle
size data. The curves shown in these figures parallel those obtained for the
uncontrolled particulate stream. However, the absolute number of particles in
each size range is greater in the outlet stream compared to the inlet. This is
most easily seen in Figure 10, in which the average cumulative size distribu-
tions for both streams have been plotted. The discontinuity between the diffus-
sional and optical portions of the particle size curve is due to the nonover-
lapping nature of the two particle counting systems.
46
-------
TABLE 25.i ELEMENTAL ANALYSIS OF OUTLET METHOD 5 FILTER PARTICULATE BY SSMS
Concentration (|iu/u)
Element
Uranium
Thorium
Bismuth
Lead
Thallium
Mercury
Cold
Plat Inuin
Iridlum
Osmium
Rhenium
Tungsten
Tantalum
Hafnium
Lutetluui
Vt terbium
Thulium
Erbium
Itolmium
Uyuproslum
Terbium
Gadolinium
Europium
Samarium
Neodyuilum
Praseodymium
Cerium
Lanthanum
BHi-luia
Cesium
Iodine
Tellurium
Antiauny
Tin
Indium
Cailmlum
Sllvur
Pal luditim
Rhodium
Run 1—
.
.
-
200
5
59.8I/
-
-
.
.
.
-
.
-
.
-
.
_
.
-
-
.
-
-
5
2
30
10
500
2
10
.
6
10
STO
2
4
-
"*
Run 1
duplicate6-/
0.4
.
0.6
300
10
46 .&!/
-
-
-
.
-
-
-
-
_
.
.
_
-
-
-
-
-
-
0.5
0.6
6
10
100
0.6
10
-
6
6
SIB
2
2
-
™
Run 2^/
_
.
i 0.05JJ/
400
20
20l/
-
.
-
-
-
-
-
-
_
-
-
.
-
-
-
-
-
-
4
-
30
40
> 500
2
16
-
11
11
STI>
£ 0.05J>/
30
-
™
Run V—' Kang6
.
.
!£/
> 900 200-0 900
70 5-70
64 .Ol/ 20-64
< 0.3£j-K/
.
-
.
.
-
.
-
.
.
-
.
-
-
-
-
-
-
a 0.5-8
6 0.6-6
40 6-40
ou 10-60
> 900 100-0 900
4 0.6-4
40 10-40
-
50 6-50
50 6-50
STD
40
9 2-30
-
—
Element
Ruthenium
Molybdenum
Niobium
Zirconium
Yttrium
St ront lum
Rubidium
Oiotnlne
Selenium
Arsenic
Germanium
Gallium
Zinc
Copper
Nickel
Cobalt
Iron
Manganese
Chromium
Vanadium
Titanium
Scandium
Calcium
Potassium
Chlorine
Sulfur
Phosphorus
Silicon
Aluminum
Magnesium
Sodium
Fluorine
Oxygen
Nitrogen
Carbon
Boron
Beryl Hum
Lithium
Hydrogen
Run 1-'
m
10
4
30
10
200
40
50
60
400
20
200
> 600
300
200
< 0.06S/
> 600
> 400
< 0.06S/
60
> 600
5
B
B
B
B
> 600
B
B
B
B
B
NR
NR
NK
B
2
40
NR
Concentration (ua/ft)
Run 1
duplicate6-'
_
10
2
10
4
60
30
60
20
200
20
20
> 600
100
20
4
> 600
60
20
10
100
5
B
B
B
B
> 600
B
B
B
B
B
NR
NR
NR
B
0.6
10
NR
Run 2£/
_
16
5
20
10
300
50
100
50
300
30
100
> 500
100
2 of.'
< 0.052/
> 500
> 200
< 0.05S/
40
> 500
0.7£/
B
B
B
B
> 500
B
B
B
B
B
NK
NR
NR
B
< 0.05H/
10
NK
Run 3^
_
60
20
200
30
> 900
90
300
300
200
200
400
> 900
800
300
< 0.03*/
> 900
> 900
< O.j£/
200
> 900
20
B
B
B
B
> 900
B
B
B
B
B
NR
NR
NR
B
t,
80
NR
Range
10-60
2-20
10-200
4-30
60-> 900
20-90
50-300
20-300
200-400
20-200
20-400
100-800
20-300
60 -> 900
10-200
100-> 900
0.7-20
10-80
B =* background levels exceed analytical range.
NK " no! reported.
STD — Internal standard.
n/ Detection limit = 0.06 ng/g.
b_/ Detection limit = 11.06 |ig/g.
c/ Detection limit = (1.05 ug/g.
d/ Detection limit = 0.3 ug/g.
e/ Corrected for background levels > 201 of sample content.
£/ Plomelesd atomic absorption analyuls.
g^/ Heterogeneous signal.
-------
TABLE_2_6_.__ SELECTED ELEMENTAL CONCENTRATIONS IN CONTROLLED
PARTICULATE EMISSIONS
Concentration (ue/dscm)
Element
Run 1
Run 2
Run 3
Range
Lead
Mercury
Barium
Antimony
Cadmium
Zirconium
Strontium
Bromine
Selenium
Arsenic
Gallium
Zinc
Copper
Nickel
Chromium
Vanadium
Titanium
Beryllium
38
11.4
95
1.1
0.38
5.7
38
9.5
21
76
38
> 114
57
38
< 0.01
11
> 114
0.38
69
3.4
> 86
1.9
< 0.009
3.4
51
17
8.6
51
17
> 86
17
3.4
< 0.009
6.9
> 86
> 0.009
> 101
7.2
> 101
5.6
4.5
22
> 101
56
56
22
45
> 101
90
34
< 0.03
22
> 101
0.45
38-> 101
3.4-11.4
> 86
1.1-5.6
< 0.009-4.5
3.4-22
38-> 101
9.5-56
8.6-56
22-76
17-45
> 86
17-90
3.4-38
< 0.03
6.9-22
> 86
< 0.009-0.45
TABLE 27. OUTLET PARTICLE SIZE DISTRIBUTION BY NUMBER AS MONITORED
BY THE OPTICAL COUNTER
Particulate concentration (10^ particles/cu m)—'
Size range (^m) Run 1 Run 2 Run 3 Average
Channel 6 (1.5-2.6)
Channel 5 (1.2-1.5)
Channel 4 (0.9-1.2)
Channel 3 (0.7-0.9)
Channel 2 (0.5-0.7)
Channel 1 (0.3-0.5)
292
344
563
877
1,530
3,500
17.9
46.2
448
318
867
3,250
25.1
59.0
153
347
888
3,080
112
150
388
514
1,100
3,280
a/ At standard conditions (29.92 in. Hg and 68°F).
48
-------
TABLE 28. OUTLET PARTICLE COUNTS BY SIZE AS MONITORED BY THE DIFFUSION
BATTERY/CONDENSATION NUCLEI COUNTER
Diffusion
battery
port No*
0
1
2
3
4
5
6
7
8
9
10
Particulate concentration
(106 particles/cu m)S/
Run 1
1,133
866
550
472
283
220
79
32
23
13
12
Run 2
791
791
791
462
329
198
132
66
45
12
10
Run 3
791
725
435
382
369
132
93
59
26
15
10
Average
905
794
592
439
327
183
101
52
31
13
11
Fractional
Run 1
1.000
0.764
0.485
0.417
0.250
0.194
0.070
0.028
0.020
0.011
0.010
Run 2
1.000
1.000
1.000
0.584
0.416
0.250
0.167
0.083
0.057
0.015
0.013
penetration
Run 3
1.000
0.917
0.550
0.483
0.466
0.167
0.118
0.075
0.033
0.019
0.013
Average
1 .000
0.877
0.654
0.485
0.361
0.202
0.112
0.057
0.034
0.014
0.012
a/ At standard conditions (29.92 in. Hg and 68°F).
-------
TABLE 29. OUTLET PARTICLE SIZE DISTRIBUTION RESULTING FROM
DIFFUSION BATTERY PENETRATION DATA
Size range
Run No* (p>m)
1 > 0.200
0.030-0.200
0.025-0.030
0.020-0.025
0.005-0.020
< 0.005
2 > 0.055
0.035-0.055
0.025-0.035
0.020-0.025
< 0.020
3 > 0.110
0.035-0.110
0.030-0.035
0.025-0.030
0.005-0.025
< 0.005
Average > 0.065
0.030-0.065
0.025-0.030
0.005-0.025
< 0.005
No. of particles^/
(106 particles/cu m)
26.1
521
176
86.1
313
11.3
348
293
103
32.4
15.1
52.2
372
117
50.6
187
12.6
215
489
90.5
94.1
17.2
Percent of particles
in stated size
range
2.3
46.0
15.5
7.6
27.6
1.0
44.0
37.0
13.0
4.1
1.9
6.6
47.0
14.8
6.4
23.6
1.6
23.7
54.0
10.0
10.4
1.9
50
-------
10
10
o
10'
u
c
Z
o
V
O
Z
108
-DIFFUSIONAL-
• OPTICAL -
3
5
U
10?
0,01
0.1
1.0
PART1CU DIAMETER,
10.0
Figure 7. Outlet particle size distribution - Run 1.
51
-------
y
Q
O
107
OIFFUStONAL
0.01
0.1 1.0
PARTICLE DIAMETER,iun
10.0
Figure 8. Outlet particle size distribution - Run 2,
52
-------
10'°
o
£ 10'
trt
Q
uj
U
a
z
O
5! 108
2
u
107
-OIFFUS1ONAL-
0.01
0.1
1.0
PARTICLE DIAMETER,>im
Figure 9. Outlet particle size distribution - Run 3
10.0
53
-------
4
O
10*
Ul
•p-
z
A
i/}
^
O
O
108 —
13
5
o
• INLET
o OUTLET
10?
I
- DIFFUSION AL-
OPTICAL
0.001
0.0!
1.0
O.I
PARTICLE DIAMETER,/im
Figure 10. j Average particle size distributions (inlet and outlet),
10.0
-------
The cause of the increased numbers of small particles in the gas stream
leaving the mechanical collectors is not clear. One hypothesis is that some of
the particulate entering the multiclones are fractionated as a result of impac-
tion within the collectors. If this phenomenon was indeed occurring, one would
expect a more steeply sloped curve for the outlet distribution in the optical
range including an intersection point with the inlet curve somewhere between
2 and 10 p,m. Therefore, additional phenomena may be affecting the controlled
emission stream particle_size distribution. _
It is not possible to make a definitive statement concerning mean particle
size in the controlled emissions from the available data. However, some infor-
mation about the size distribution of the collected fly ash was provided by UOP
and may give a rough indication of the mean size.
Table 30 presents results of an analysis of the primary and secondary
ashes which was done by UOP using a BAHCO centrifugal analyzer.
Working from the ash size analysis data and the design efficiency curves
for the mechanical collectors, UOP used a computer program to simulate the par-
ticle size distributions in the air streams entering and exiting the control
system. These derived distributions are shown in Table 31.
The UOP data show that, theoretically, 98.2% of the controlled particu-
late emissions (by weight) should have been comprised of particles smaller than
4 y,m. Although this simulated distribution may be somewhat unreliable due to
the inaccuracies of the BAHCO analysis and the use of design efficiency curves,
it still indicates that the great majority of the particles are in the range
measured by MRI's optical counter. It seems likely that the mean particle di-
ameter is less than 1 )j,m.
OPACITY
Results of the EPA Method 9 visual plume opacity readings are shown in
Figures 11 and 12. Plume opacities averaged very nearly 207, on each test day.
Maximum opacities observed were 27.9% on Day 1 and 31.47, on Day 2. The read-
ings were taken for a total 131 min on the first test day and 245 min on the
second.
PCB AND PAH COMPOUNDS
A special sampling train, using a Florisil adsorbent trap, was used to
collect two samples of both particulate and vaporous emissions. These were ana-
lyzed for PCB and PAH compounds using GC/MS techniques.
55
-------
TABLE 30. SUMMARY OF BAHCO ASH ANALYSIS
Size range
> 60
40-60
30-40
20-30
15-20
10-15
7.5-10
5.0-7.5
3.5-5.0
2.5-3.5
1.5-2.5
< 1.5
Primary ash
(% by wt.)
40
11
7.7
8.8
4.5
7.5
5.5
4.0
2.8
2.0
4.3
1.9
Secondary ash
(% by wt.)
0.9
0.5
0.5
1.4
1.7
4.3
6.2
13.5
19.0
24.2
21.2
6.6
TABLE 31. SIMULATED PARTICLE SIZE DISTRIBUTIONS
DEVELOPED BY UOP
Size range Collector inlet Collector outlet
(% by wt.) (% by wt.)
> 60
40-60
30-40
20-30
15-20
10-15
8-10
6-8
4-6
2-4
< 2
35.48
9.75
6.82
7.79
3.93
6.69
4.01
3.32
3.73
7.22
11.26
0.02
0.01
-
-
-
0.01
0.08
0.27
1.39
12.17
86.05
56
-------
30
S 20
v
o ;
8.: 10
O: ~
No Observation
11:30 12:00 12:30 13:00 13:30 ;14:pO :U:30 15:00
Time Hours
Figure 11.; Plot of opacity versus time, 7/24/78
30
§ 20
u
g. 10
O
No Observation
8:00 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00
Time, Hours
Figure 12. Plot of opacity versus time, 7/25/78<
57
-------
Contents of the two impingers which preceeded the Florisil trap were ana-
lyzed, along with the Florisil adsorbent, for PCB materials. Preliminary screen-
ing using EC/GC indicated that most of the samples did not contain sufficient
quantities of PCB materials to permit mass spectral verification. Using Arochlor
1254 and 1260 standards, a detection limit corresponding to a total of 1 g,g of.
PCB per sample was obtained.
Only the Florisil trap from Run 1 and the second impinger solution from
Run 2 were submitted for further analysis by GC/MS because of the complexity
of their EC/GC chromatograms. A selective ion monitoring (SIM) technique, using
an Arochlor 1254 standard, was employed. All sample responses were found to be
less than 1 ^g total PCB content for both samples.
Results from analysis of the Florisil train components for PAH compounds
are shown in Table 32. In all the samples analyzed, no PAH materials were iden-
tified at levels which permitted structural confirmation by GC/MS. The possible
compounds observed are therefore listed as "less than" the detection limit.
SASS - LEVEL 1 ASSESSMENT
In addition to the air emissions testing previously described, one sample
was collected using the EPA-developed SASS in conjunction with EPA's Level 1
environmental assessment procedure. The SASS run was made near the end of the
2nd day of testing. Brief explanations of the operation of the train and the
required analytical procedures are included in Appendices A and C of this re-
port. Results are discussed in the subsections below.
Particulate Concentration
The high-volume SASS collects a particulate sample in a manner similar
to EPA Method 5, except that the SASS probe is positioned at a single sampling
point in the air stream for the entire period of sample collection. Therefore,
the SASS sample can only serve as an approximation of the true particulate
emission concentration.
Weights obtained for the various portions of the SASS sample are shown
in Table 33. Sixty-five percent of the sampled particulate was collected by
the 1 um cyclone and the final filter, indicating a predominance of very fine
particles in the stack emissions. Table 34 presents a summary of the particu-
late concentrations and emission rates corresponding to each portion of the
SASS sample. Both metric and English units are included.
The total particulate concentration measured by the SASS was 0.0906 g/dscm
(0.0396 gr/dscf). This compares very well to the concentration of 0.093 g/dscm
(0.041 gr/dscf) which resulted from the third Method 5 sampling run (completed
just prior to the SASS run). Particulate emission rates measured by the SASS,
in terms of weight per unit time and weight per unit of heat input, are also
shown in the table.
58
-------
TABLE 32. ANALYSIS OF FLORISIL TRAIN COMPONENTS FOR PAH BY GO/MS
Ln
vO
Gas volume (cu m)
Compound
Acenaphthylene
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Chrysene
1 , 2-Benzanthracene
BenzoCaJpyrene
Perylene
Indeno[l,2,3-c,d]pyrene
1,2,5, 6-Dibenzanthracene
1,12-Benzoperylene
Detection
limitS/ Run 1 Run 2k/
0.060 0.061 0.059
ug/cu m
(jtg/cu m Florisil Impinger 1 Impinger 2 Florisil Impinger 1 Impinger 2
37
35
16 < 16l/ < 16l/
10 < 10
-------
TABLE 33. SASS PARTICULATE ANALYSIS (LEVEL 1)
Constituent
Front end 10 ^m 3 pi
rinse Cyclone Cyclone
Cyclone Filter
Weight (g)
CH2C12 extract
0.2588
0.1331
0.4441
0.7351
0.8080
Combined
TCO (mg)
GRAV (mg)
IR (qualitative
identification)
LC (F fractions)
0.017
4.3
a/
b/
0.28
2.2
a/
b/
a/ Samples did not meet the Level 1 criteria for these analyses to be
performed.
b/ IR absorptions of sufficient intensity to allow interpretation were not
observed*
TABLE 34. SUMMARY OF SASS PARTICULATE RESULTS
Particulate
concentration
Particulate emission rate
Sample fraction
Front end rinse
10 |j,m cyclone
3 \an. cyclone
1 nm cyclone
Filter
Total
g/dscm
0.0099
0.0051
0.0169
0.0280
0.0308
0.0906
gr/dscf
0.0043
0.0022
0.0074
0.0122
0.0134
0.0396
kg/hr
0.75
0.38
1.29
2.12
2.33
6.89
Ib/hr
1.65
0.84
2.83
4.67
5.13
15.18
g/MJ
0.005
0.003
0.009
0.015
0.016
0.047
lb/106 Btu
0.012
0.006
0.021
0.034
0.037
0.110
60
-------
Inorganic Analysis
Under Level 1 protocol, each portion of the SASS sample is to be analyzed
for elemental composition using SSMS. The SASS particulate fractions were not
submitted for this analysis, however, since the Method 5 particulate samples
(filters) were used instead to provide more representative information regard-
ing elemental compositions*
SSMS analysis was only performed on the aqueous condensate from the con-
denser module and the first impinger contents from the final portion of the
SASS. Results of the condensate analysis are shown in Table 35 in terms of an
air concentration (p,g/m^)« Sample volume through the SASS was approximately
925 dscm.
Many elements were not detected at the 0.8 p,g lower limit (0.009
Only a few elements were observed at air concentrations greater than 1
These included Zn, Ni, Fe, Mn, Cr, Ca, K, Cl, S, P, Si, Mg, Na, and F. Iron
and chromium were present at the highest concentrations (> 43 and 22 |j,g/m3,
respectively).
Table 36 summarizes results of the SSMS analysis of the SASS hydrogen
peroxide impinger contents. Many elements were not detected or had air concen-
trations less than 1 iig/m? • Pb, Mo, Mn, Ca, P, Si, and Na showed concentrations
between 1 and 10 jj,g/m^, while Zn, Cu, Ni, Fe, Cr, and S exceeded the upper range
of the SSMS screening technique (> 13 ng/m3). The data in Tables 35 and 36 in-
dicate the relative magnitude of trace element emissions from the Burlington
boiler in the vaporous phase.
It is important to note that concentrations of certain trace metals, par-
ticularly Fe, Cr, and Ni, in the aqueous condensate and impinger solution are
probably greatly elevated due to corrosion of the stainless steel internal
components of the SASS. Corrosion products have been verified in past MRI stud-
ies in which the SASS was used, (3,4) and the appearance of the sample solu-
tions in this study indicates that some corrosion also occurred during the SASS
run at Burlington Electric.
Organic Analysis
Additional analysis of the SASS particulate, as prescribed by the Level 1
procedures, included a methylene chloride extraction for subsequent qualitative
and quantitative determination of organic compounds. The particulate extracts
were analyzed for total chromatographable organics (TCO) and gravimetric resi-
due (GRAV), with the resulting data shown in Table 33. In all concentrated ex-
tracts, numerous additional peaks were observed in the TCO analysis which were
beyond a C-17 hydrocarbon retention time. Because the Level 1 protocol does
not include peaks beyond C-17 for TCO calculation, these data are not reflected
in the tabulated results (Table 33). Attempts to qualitatively assess the nature
61
-------
TABLE 35. ELEMENTAL ANALYSIS OF SASS AQUEOUS CONDENSATE
BY SSMS
Element
Uranium
Thorium
Bismuth
Lead
Thai liven
Mercury
Gold
Platinum
Iridium
Osmium
Rhenium
Tungsten
Tantalum
Hafnium
Lutecium
Ytterbium
Thulium
Erbium
Holmium
Dysprosium
Terbium
Gadolinium
Europium
Samarium
Neodymium
Praseodymium
Cerium
Lanthanum
Barium
Cesium
Iodine
Tellurium
Antimony
Tin
Indium
Cadmium
Silver
Palladium
Rhodium
Concentration
(|ig/cu m)2/
ND
NO
ND
< 0.03
.
0.009k/
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.043
ND
0.022
ND
ND
ND
STD
ND
ND
ND
ND
Element
Ruthenium
Molybdenum
Niobium
Zirconium
Yttrium
Strontium
Rubidium
Bromine
Selenium
Arsenic
Germanium
Gallium
Zinc
Copper
Nickel
Cobalt
Iron
Manganese
Chromium
Vanadium
Titanium
Scandium
Calcium
Potassium
Chlorine
Sulfur
Phosphorus
Silicon
Aluminum
Magnesium
Sodium
Fluorine
Oxygen
Nitrogen
Carbon
Boron
Beryllium
Lithium
Hydrogen
Concentration
kg/cu m)i/
ND
0.43
ND
< 0.011
ND
0.043
ND
0.22
< 0.011
0.086
ND
ND
1.1
0.32
8. 6
0.22
> 43
1.1
22
0.043
0.22
< 0.002
11
l.l
5.4
11
3.2
l.l
0.43
l.l
2.2
2.2
MR
NR
NR
0.22
ND
0.002
NR
Note: STD = internal standard; NR = not reported; ND = not detected
(< 0.009 ug/cu m).
a/ Detection limit = 8 ug» based on a collected volume of 400 ml.
b/ Flamelesa atomic absorption analysis.
62
-------
TABLE 36.. ELEMENTAL ANALYSIS OF SASS HYDROGEN PEROXIDE
IMPINGER COUNTER BY SSMS
Element
Uranium
Thorium
Bi south
Lead
Thallium
Mercury
Cold
Platinum
Irtdium
Osmium
Rhenium
Tungsten
Tantalum
Hafnium
Lutecium
Ytterbium
Thulium
Erbium
Holmium
Dysprosium
Terbium
Gadolinium
Europium
Samarium
Neodymium
Praseodymium
Cerium
Lanthanum
Barium
Cesium
Iodine
Tellurium
Antimony
Tin
Indium
Cadsium
Silver
Pal ladiua
Rhodium
Concentration
(yig/eu at)
m
ND
ND
1.6
ND
3§/
SD
ND
ND
SD
SD
ND
SD
N0
ND
ND
ND
ND
.ND
ND
ND
ND
ND
SD
SD
ND
ND
ND
o.ooa
ND
ND
ND
0.003
0.16
ND
0.013
0.029
ND
ND
Element
Ruthenium
Molybdenum
Niobium
Zirconium
Yttrium
Strontium
Rubidium
Bromine
Selenium
Arsenic
Germanium
Gallium
Zinc
Copper
Nickel
Cobalt
Iron
Manganese
Chromium
Vanadium
Titanium
Scandium
Calcium
Potassium
Chlorine
Sulfur
Phosphorus
Silicon
Aluminum
Magnesium
Sodium
Fluorine
Oxygen
Carbon
Boron
Beryllium
Lithium
Hydrogen
Concentration
(p,g/cu m)
ND
1. 1
0.029
0.003
ND
0.046
ND
0.029
0.12
0.12
B
0,11
> 16
> 16
> 16
0.66
> 16
4.3
> 16
0.065
0.29
SD
a. 2
0.32
0.043
> 13
3.0
9.1
0.12
0.26
2.8
B
MS
MR
0.10
ND
0.009
NR
Note: B =" background levels equal or exceed sample levels; STD = Internal
standard} MR =* not reported; ND =" not detected (< 0.003 ng/cu m).
a/ Flameless atomic absorption analysis.
63
-------
of the TCO/GRAV components through infrared (IR) analysis were not successful
since IR absorptions of intensity sufficient to allow interpretation were not
observed.
Further organic analysis of the particulate extracts via LC fractionation
was not carried out since the combined TCO and GRAY values were below the 15
mg minimum requirement which is stipulated in the Level 1 procedures*
The organic module which follows the particulate filter in the SASS serves
to trap gaseous organic constituents in the sample stream using XAD-2 resin
adsorbent. After sample collection, both the resin itself and the rinse of the
internal surfaces of the organic module are collected for analysis. In addi-
tion, the aqueous condensate from the organic module is also collected and ana-
lyzed (the sample air stream is quickly cooled just prior to the XAD-2 resin
cartridge to allow optimum adsorption). Results of the TCO/GRAV analyses for
these sample fractions are shown in Table 37. Direct interpretation of these
data is difficult. However, one method of assessment is the use of EPA1s SAM-
1A, which is presented in Section 5 of this report.
TABLE 37. LEVEL 1 ORGANIC ANALYTICAL RESULTS
Sample fraction
XAD-2 resin
Organic module rinse
Aqueous condensate
TCO (mg)^!^/
5.9£/
0.087
0.29
GRAV (mg)k/
34d/
2.0
£/
a/ Sample values corrected for reagent blanks.
b/ Based on analysis of concentrated sample extracts.
c/ Corrected for XAD-2 resin background contributions.
d_/ Determined from the unconcentrated sample. No background level
was observed in the unconcentrated blank.
e/ Not detected.
64
-------
IR spectra obtained from analysis of the GRAV residues produced inter-
pretable data only for the concentrated XAD-2 resin extract. These data are
shown in Table 38 and indicate absorbance peaks primarily characteristic of
carbonyl-containing hydrocarbon materials.
Only the organic content of the XAD-2 sample was above the minimum re-
quirement for further analysis (15 mg) and therefore the extract was fraction-
ated by the LC method. Resulting TOO and GRAV values are shown in Table 39.
These analyses revealed that the bulk of the organic material was concentrated
in the more polar LC6 and LC7 fractions. The TOO analysis also indicated the
presence of organic material in fraction LC3.
The LC fractions 6 and 7 were the only ones to produce interpretable spec-
tra from IR analysis. The observable absorbance peaks and their functional
group assignments are presented in Table 40. The most prominent feature con-
sists of a carbonyl band in the 1700 cm region. Additional peaks of lesser
intensity corresponding to aliphatic, aromatic, and possibly hydroxyl materials
were observed. Distinctions between classes of acids, aldehydes, esters, and
ketones were not possible due to the limited data; however, the presence of
acids or esters appears most likely based on the LC fractionation scheme. Acids
and esters have been reported to elute in LC fractions 6 or 7, in contrast to
aldehydes and ketones which reportedly elute in LC fraction 4. Although a
strong absorption suggesting 0-H stretching (3400 to 3550 cm~l) was observed
in the unfractionated XAD-2 extract, which may be indicative of a carboxylic
acid or an alcohol, the absence of a similar absorbance in the IR data for the
LC fractions would suggest that the prevalent functional group is an ester.
The large absorbance observed in the unfractionated sample may be due to the
presence of water in either the sample extract or the salt plates used for
the IR analysis. Since the XAD-2 resin was found to be wet after sampling, the
presence of water at its solubility limit in the initial sample extracts would
not be unlikely. The IR spectra were found to be essentially the same for both
the concentrated XAD-2 extract and the LC fractions. Of the two LC fractions,
LC6 was found to produce the more intense spectra. Interpretation of IR spec-
tra of other concentrated sample extracts was not generally possible due to
the low concentrations of organic materials present and the resultant absence
of observable absorptions above background levels. IR spectra for the remain-
ing LC fractions were not taken due to the absence of measurable organic mate-
rial as identified in the gravimetric analyses of those fractions.
The final phase of the organic analysis of the XAD-2 resin extract con-
sisted of direct inlet LRMS. Thermograms and ion range plots were constructed
to identify scanning regions of interest. Then spectra from each selected scan
range were obtained. A complex mixture of ions was observed for each range of
interest, which was further complicated by high background levels of the less
volatile components. Because of these problems, detailed interpretation of the
LRMS data was not performed.
65
-------
TABLE 38. IR SPECTRAL ANALYSIS OF CONCENTRATED
XAD-2 RESIN EXTRACT
Frequency
V (cm'1)
1460
1600-1650
2850
2920
2960
3400-3550
Intensity^/
M
S
M
S
W
S
Functional group
assignment
C-H bending,
0-H bending
G=O stretching
C-H stretching
C-H stretching
C-H stretching
0-H stretching
(broad )
a/ S = strong, < 857. T; M = medium, 85 to 95% T; W = weak, > 95% T.
TABLE 39. LEVEL 1 ORGANIC ANALYSIS OF LIQUID CHROMATOGRAPHIC
FRACTIONS OF XAD-2 RESIN SAMPLE
LC Fraction
1
2
3
4
5
6
7
TC03/ (mg)
0.10
0.0055
1.1
0.010
-
0.80
0.44
Gravimetric
-b/
-
-
-
-
6.8
6.4
(mg)
a/ Sample values corrected for reagent blank contributions.
b_/ Blanks denote insufficient material to be detected.
66
-------
TABLE. 40. IR SPECTRAL-ANALYSIS OF LC FRACTIONS 6 AND 7
Frequency
LC Fraction I/ (cm"1)
6 1130
1290
1460
1710-1740
2870
2940
2970
7 1070-1140
1230-1300
1390-1420
1550-1600
1690-1720
2850
2920
2950
Intensity*/
W
M
W
S
M
S
W
W
M
W
W
S
M
S
W
Functional group
assignment
Unas signed
C-0 stretching
C-H bending
0=^) stretching
C-H stretching
C-H stretching
C-H stretching
Unassigned
C-0 stretching
Unassigned
Unassigned
0=^) stretching
C-H stretching
C-H stretching
C-H stretching
a/ S = strong, < 85% T; M = medium, 85 to 95% T; W = weak, > 95% T.
ELECTRON SPECTROSCOPY FOR CHEMICAL ANALYSIS
Elemental analysis of particulate samples from the Burlington Electric
test program was conducted to obtain information on potential hazards. However,
elemental analysis conducted using SSMS only indicates the total elemental com-
position and does not differentiate between surface and internal concentrations.
Moreover, SSMS does not offer any potential information about the specific chem-
ical species contained in the particulate material, which is very important for
establishing the hazard potential. Therefore, particulate samples from the
Burlington process were analyzed by ESCA to identify any hazardous elements
present at surface concentrations greater than 0.5%. Selected elements detected
were then speciated to determine the oxidation state and the chemical form of
the element as it exists on the surface of the solid. Auger analysis was also
performed to achieve improved sensitivity for certain elements.
67
-------
Samples of particulate from each cyclone and-the filter from the SASS
train, primary and secondary collector ash, and bottom ash were submitted for
analysis. The 1 (jjm cyclone catch from the SASS train was analyzed first to es-
timate the amount of information that could be found from stack particulates.
The 1 jum particulate had the greatest available surface area of the SASS par-
ticulate samples and therefore offered the greatest potential for detection
and speciation using ESGA. However, few elements of interest were detectable
in the 1-p.m particulate fraction. Therefore, the rest of the SASS train samples
and the collector ash samples were not analyzed. Bottom ash elemental composi-
tion, however, was expected to be different from the stack particulates and
was therefore analyzed.
ESCA and Auger analysis were performed by Physical Electronics Industry
using a Model 550 series instrument built by the same company. An ESGA survey
scan was made of each sample and peaks were identified by element and electron
transition type. General composition elements such as carbon, oxygen, calcium,
sodium, potassium, aluminum, and silicon were noted but were not of interest
with respect to characterization of hazardous elements present in the samples.
No transition metals (except iron) or heavy metals were identified in the sur-
vey scan. Sulfur and fluorine were detected in the SASS 1 jum particulate survey
scan and selected for high resolution analysis to determine their chemical
types. The SSMS results for collector ash and bottom ash were used to select
additional elements with concentrations greater than 250 ppm for the 1-y.m par-
ticulate and bottom ash samples, respectively. The elements were searched in
the samples by repetitive energy scans to enhance the signal. The energy region
of the most intense emission (+ 10 ev) was repeatedly scanned and the signal
averaged to increase the detection limit. Although 250 ppm is well below the
~0.57o detection limit for ESCA, if any of these selected elements were en-
riched on the particulate surface, the surface concentration could be 10 to 50
times higher than the bulk concentration and therefore should be potentially
detectable by ESCA.
Table 41 summarizes the results from the survey scan (atom percent) and
high resolution scans for the 1-^m particulate. Carbon Is binding energy is
used as a reference to compensate for sample charging which shifts binding
energy. In addition to the general composition elements, fluorine, sulfur and
titanium were detected. The fluorine species is rather surprising because fluo-
ride salts were expected. However, the observed fluorine binding energy pre-
cludes the presence of fluoride salts and suggests the presence of carbon-
fluorine bonding. The presence of carbon-fluorine bonds suggests the presence
of reactive fluorine species in the stack which react with carbon on the par-
ticulate surface. Table 42 summarizes the results from the survey scan (atom
percent) and high resolution scans for the bottom ash.
68
-------
TABLE 41. ESCA RESULTS FOR l-|i PARTICULATE
Element
Oxygen
Carbon
Silicon
Aluminum
Potassium
Sulfur
Fluoride
Titanium
Arsenic
Manganese
Vanadium
Chromium
Nickel
Barium
Atom
(%)
41
37
7
5
4
4
1
Trace
ND
ND
ND
ND
ND
ND
Corrected
binding
energy
(ev)
532
285
100
118
297
17 Ok/
687k/
458k/
46
642
517
575
855
781
Orbit
Is
Is
2p
2s
2p
2p
Is
2p
Is
2p
2p
2p
2p
3d
Chemical
species Comments
a/
a/
aJ
a/
a/
804
C-F Organically bound
fluorine
Ti02
c/
c/
c/
c/
c/
c/
a/ Species for major composition elements not determined.
b_/ Data from high resolution analysis.
c/ Element could not be detected.
Because few elements of interest were detected by ESCA, the l-/im particu-
late was analyzed by Auger techniques. Auger is more sensitive than ESCA for
certain elements and the detection of selected elements was attempted. Figure
13 is the survey scan of the 1 ^ particulate sample. Magnesium was the only
additional element detected. Fluoride was not observed in the Auger analysis
when compared to the ESCA results. This is probably due to the decreased sen-
sitivity of Auger analysis to fluoride species.
69
-------
TABLE 42. ESGA RESULTS FOR BOTTOM ASH
Element
Carbon
Oxygen
Silicon
Aluminum
Calcium
Iron
Titanium
Manganese
Chromium
Zirconium
Strontium
Vanadium
Atom
(7.)
45
34
12
9
Trace
Trace
Trace
Trace
b/
b/
b/
b/
Corrected
binding
energy
(ev)
285
532
100
118
350
710
458£/
642£/
575
182
134
517
Orbit
Is
Is
2p
2s
2p
2p
2p
2p
2p
3d
3d
2p
Chemical
species
a/
a/
a/
a/
a/
a/
Mn02
b/
b/
b/
b/
Comments
Other possible compounds
MnO and MnX2
a/ Species for major composition elements not determined.
b/ Element could not be detected*
c/ Data from high resolution analysis.
70
-------
REFERENCES
1. Riegel, S. A., and K, P. Ananth. Stationary Source Testing at Power Plant
of University of Missouri at Rolla. EPA Contract No. 68-02-2166. MRI Draft
Final Report. July 1977.
2. Golembiewski, M. A., and K. P. Ananth. Evaluation of Fabric Filter Per-
formance at Browning Ferris Industries/Raytheon Service Company Resource
Recovery Plant - Houston, Texas. EPA Contract No. 68-02-2166. MRI Draft
Report. September 1977.
3. Gorman, P., M. Marcus, K. Ananth, and M. Golembiewski. Environmental As-
sessment of Waste-to-Energy Process: Union Carbide Purox®System. EPA
Contract No. 68-02-2166. MRI Revised Final Report. April 1979.
4. Golembiewski, M. K. Ananth, G. Trischan, and E. Baladi. Environmental As-
sessment of a Waste-to-Energy Process: Braintree Municipal Incinerator.
EPA Contract No. 68-02-2166. MRI Revised Final Report. April 1979.
5. Sanborn, C. R. Evaluation of Wood-Fired Boilers and Wide-Bodied Cyclones
in the State of Vermont. Prepared for U.S. EPA Region I and Vermont Agency
of Environmental Conservation. March 1, 1979.
80
-------
APPENDIX A
ANALYTICAL METHODOLOGY
PROXIMATE ANALYSIS
Samples
One wood and one fuel oil sample for each of two test days were submitted
for proximate analysis. A blind duplicate of each sample type was also analyzed.
Sample Preparation
Three 5-liter samples of wood were taken for each test day and mixed to
produce a 1 liter composite sample. The samples were weighed as received and
dried at 5°G above ambient temperature until a constant weight was obtained.
The total sample was then ground until it passed a No. 60 sieve. A 1-liter sam-
ple of No* 2 fuel oil was taken for each test day and analyzed directly.
Sample Analysis
The ASTM procedure for coal was used for the analysis of both wood and
fuel oil. The method, ASTM Method D-3172, includes: moisture determination ac-
cording to Method D-3173; ash according to Method D-3174; volatile matter ac-
cording to Method D-3175; and fixed carbon by difference. Heat of combustion
was also determined using a calorimeter according to ASTM Method D-271. Addi-
tionally, oil density was determined by pycnometer. All of these analyses were
performed by Industrial Testing Laboratory, Kansas City, Missouri.
Quality Assurance
Analytical precision was monitored by submitting blind duplicate samples
of ground wood and fuel oil for one test run of the analytical subcontractor.
The data for duplicate analyses for both proximate and ultimate analyses
are presented in Table A-l» All results are within + 5% of the averaged values
for duplicate tests with the exception of sulfur content. The high sulfur vari-
ability may be due to sample inhomogeneity and/or the relatively low sulfur
content of the materials which were analyzed. Additionally, the wood samples
81
-------
TABLE A-l. PROXIMATE AND ULTIMATE BLIND DUPLICATE ANALYSES OF WOOD AND FUEL OIL
Analysis
As received:
Moisture (%)
Ash (%)
Volatile matter (%)
Fixed carbon (%)
Sulfur (%)
Heat of combustion (Btu/lb)
Density (Ib/gal)
Dry basis:
Ash (%)
Volatile matter (%)
Fixed carbon (%)
Sulfur (%)
Heat of combustion
Hydrogen (%)
Carbon (%)
Nitrogen (%)
Oxygen (%)
Run
2A
2.53
3.70
68.35
25.42
0.34
9,253
-
3.80
70.12
26.08
0.35
9,492
5.78
54.11
0.26
35.74
Wood
Run
2B
2.74
4.12
68.72
24.42
0.72
9,266
-
4.24
70.65
25.11
0.74
9,527
5.81
54.38
0.29
34.59
Relative
deviation
(+%)
4
5
< 1
2
36
< 1
-
5
< 1
2
36
< 1
< 1
< 1
5
2
Fuel oil
Run Run
2A 2B
— — .
< 0.001 < 0.001
-
-
0.23 0.36
19,525 19,501
7,056 7,054
_ —
_
-
_
- -
11.98 11.38
78.45 81.84
0.00 0.01
9.34 6.63
Relative
deviation
(± %)
—
-
-
-
22
< 1
< 1
-
-
-
-
—
3
2
50
17
-------
were observed to contain coal powder which was introduced by the boiler feed
system and could significantly influence the low level analyses, especially
if the coal were not homogeneously dispersed. Losses of sulfur-containing fuel
oil components during sample handling may also be a source of error in the
analysis of fuel oil for sulfur*
Problems
The main problems encountered in this analysis were those of obtaining a
representative sample of wood and in measurement of the lower concentration
components such as sulfur*
ULTIMATE ANALYSIS
Samples
Ultimate analysis was performed on one wood and one fuel oil sample for
each of two test days; a blind duplicate of each sample type for one test day
was also analyzed.
Sample Preparation
The sample preparation for ultimate analysis was identical to that used
for proximate analysis*
Sample Analysis
The ASTM method for coal was used for the analysis of wood and fuel oil.
The method, ASTM Method D-3176, includes: determination of carbon and hydrogen
according to Method D-3178; sulfur determination according to Method D-3177;
nitrogen determination according to Method D-3179; ash is determined using the
same methods as in proximate analysis; and oxygen is determined by difference.
Quality Assurance
Analytical precision was monitored by submitting blind duplicates of wood
and fuel oil for one test run* The quality assurance data are included with
the proximate testing quality assurance data listed in Table A-l. All results
are within + 5% of the averaged values for duplicate tests with the exception
of analysis for nitrogen and oxygen in fuel oil. Variability in the nitrogen
and oxygen analyses may be due to either loss of volatile nitrogen- or oxygen-
containing compounds or air entrainment in the samples as analyzed. Since oxy-
gen is determined by difference and is at a relatively low level, variations
in other analytical results also will influence the quantity reported and ap-
pears to be the most probable source of variation. The extremely low quantity
of nitrogen reported approaches detection limit levels, resulting in large per-
cent deviations for extremely small amounts of material.
83
-------
Problems
The main problems encountered in these analyses involved representative
sampling and lack of analytical reproducibility at low concentrations.
PARTICULATE WEIGHT
Samples
Particulate weights were determined for the probe rinse, cyclone, filter,
ether/chloroform extract residue, and water residue from the collector inlet
Method 5 trains; the probe rinse, filter, ether/chloroform extract residue, and
water residue from the collector outlet Method 5 trains; and the front half
rinse, cyclones, and filter from the SASS train* Laboratory and field blanks
were also included in the analyses. These blanks include: filters, solvents
used for rinsing and sample extraction, and water used to fill the impingers.
Sample Preparation
Gelman Type A/E (fiberglass) filters for the Method 5 and SASS trains were
dessicated prior to subsequent taring in a controlled humidity environment.
Reweighings under comparable conditions were performed to determine the weight
of particulate on the filters.
Acetone was used to rinse the probes of the Method 5 trains. The rinse
from the front half of the filter holder was included in the probe rinse. These
rinses were evaporated as separate samples and included as part of the front
half weight. For the SASS train, methylene chloride was used to rinse the probe,
cyclones, cyclone connecting lines, and front half of the filter holder. The
rinses were combined and weighed as a single sample according to Level 1 pro-
tocol.— Cyclone catches were dessicated and weighed in a controlled humidity
environment.
Liquid samples were transferred to previously dessicated and tared beak-
ers and evaporated to dryness. Liquid samples were analyzed for organic mate-
rial by extraction with three 25-ml portions of chloroform followed by extrac-
tion with three 25-ml portions of ethyl ether.!/ The chloroform and ether
extracts were combined in previously dessicated and tared beakers. The extracts
were evaporated to dryness under a stream of prepurified nitrogen. After extrac-
tion the water samples were poured into similarly treated beakers and evaporated
to dryness on a steam bath.
Sample Analysis
Particulates, and dried samples, in beakers, were dessicated overnight,
equilibrated at controlled temperature and humidity for 3 hr, and weighed to
the nearest 0.1 mg on an analytical balance. Filters were dessicated to a con-
stant weight.
84
-------
Quality Assurance
Field blanks of filters, methylene chloride, acetone, and water were pre-
pared with the samples. Laboratory blanks consisting of chloroform/ether and
control beakers were also prepared and weighed at the same time as the samples.
The tare weights and final weights of the filters were made under controlled
humidity and temperature conditions* The balance was calibrated daily using a
set of Class S reference weights. Multiple weighings of dried sample materials
to constant weight were performed to assure accurate weight determinations.
Problems
No specific problems were encountered in these analyses.
References
1. Federal Register. Vol. 36, No. 159. Tuesday, August 17, 1971.
2. IERL-RTP. Procedures manual: Level 1 environmental assessment. EPA-60012-
76-160a, June 1976.
ELEMENTAL COMPOSITION
Samples
All input and outlet streams were analyzed by SSMS for elemental compo-
sition. Samples of wood, fuel oil, bottom ash, primary collector ash, second-
ary collector ash, primary collector inlet particulate and secondary collector
outlet particulate were analyzed for each test day. Outlet samples were col-
lected using the SASS train on one test day. From these samples portions of
aqueous condensate and hydrogen peroxide impinger solution were selected for
elemental analysis.
Quality assurance samples for precision determinations included analysis
of blind duplicate samples of refuse, bottom ash, primary collector ash, sec-
ondary collector ash, primary collector inlet particulate, and secondary outlet
particulate. Standard reference materials (SRM), when available, were analyzed
as unknown samples to determine analytical accuracy. SRMs which were analyzed
included NBS pine needles (SRM 1575) and NBS fly ash (SRM 1633). A mixed metal
standard in a hydrocarbon matrix was prepared using Conostan® organometallic
standards.* Samples of aqueous condensate and hydrogen peroxide were fortified
with selected metals to determine the analytical accuracy for these sample types
because suitable reference materials were not available. Blanks of Gelman A/E
* Conostan Division, Continental Oil Company, Ponca City, Oklahoma,
85
-------
fiberglass filter and hydrogen peroxide solution were also analyzed. Mercury
analyses, which are not accurately analyzed by SSMS techniques, were performed
using vapor atomic absorption methods*
All elemental composition analyses were subcontracted to Commercial Test-
ing and Engineering Company (GTE), Instrumental Analysis Division, Golden,
Colorado*
Sample Preparation
The solid samples were prepared by mixing a 0.1-g sample with 0.1 g of
ultrapure graphite in an agate mortar, adding the internal element standard
along with a few drops of redistilled alcohol, and evaporating the alcohol and
water from the mix with an infrared lamp. At this point the temperature reached
can be as high as 180°C and may, depending upon compound form, contribute to
loss of more volatile elements from the resulting SSMS sample pins.
Solid samples on fiberglass filters were prepared by first sectioning the
sample plus filter and placing a section in a Pyrex centrifuge tube. The section
was then covered with redistilled ethanol and the tube placed in an ultrasonic
bath to separate as much sample as possible from the filter. The ethanol was
then evaporated off and the tube and contents dried for 2 hr at 100°C. The tube
and contents were then weighed accurately. The filter was then removed and the
sample reslurried with ethanol and centrifuged to settle the sample while leav-
ing any filter fibers suspended. The ethanol plus any filter fibers were re-
moved by suction and the tube plus sample was again dried for 2 hr at 100°C.
After cooling in a dessicator the tube and contents were weighed accurately to
determine the weight of the filter. The sample was reslurried with ethanol and
washed into an agate or quartz crucible and at this point was treated as a nor-
mal solid sample. The tube was then dried a third time for 2 hr at 100°C and
weighed accurately to determine the tare weight.
Due to the small weight obtained, all glass filter samples were mixed in
a 1:1 ratio with graphite and the internal element standard was ratioed to the
individual sample weight.
The four wood samples were initially ashed to a constant weight using a
LFE-302 low temperature RF plasma asher. The resulting ash was dissolved in a
1:1 ratio of ultrapure nitric acid and water. The entire solution was taken
for analysis and the sample preparation carried out as for liquid samples.
The liquid samples were prepared by mixing a 20-ml aliquot of sample with
0.2 g of graphite in an agate mortar, adding the internal element standard
along with a few drops of redistilled alcohol and evaporating the alcohol and
water from the mix with an infrared lamp.
86
-------
The resulting mix was compacted into polyethylene slugs, and the slug in-
serted into a metal die* Hydraulic pressure was applied to the die and slug,
forming a graphite electrode, which was then mounted in the spark source for
analysis*
The fuel oil samples were weighed accurately into quartz crucibles and
mixed with the appropriate amount of graphite and the internal element stan-
dard* The mix was then placed in a cold muffle furnace and slowly brought to
350°C* The samples were maintained at this carbonizing temperature for 1 hr in
order to eliminate hydrocarbon interference. The samples were then treated as
solid samples*
All sample preparation was performed by GTE*
Samples for mercury analysis were specially prepared by GTE. The solid
samples were combusted in a quartz combustion tube in the presence of a mercury-
free oxygen flow* The oxygen carried the mercury vapor over two gold traps ar-
ranged in series* The gold traps were heated successively and the mercury vapor
passed through a quartz cell set in the path of a mercury hollow cathode lamp.
The fuel oils were treated in the same manner as the solid samples*
The liquid samples were analyzed for mercury by placing an aliquot of the
sample in a reaction vessel and adding a known volume of sodium borohydride to
reduce the mercury. The mercury vapor was then carried by mercury-free oxygen
to the gold traps and ultimately to the quartz cell.
The samples on filter media were analyzed for mercury by heating a section
of known weight in the combustion tube and treating them as solid samples. By
knowing the weight of the unused portion, the weight in micrograms was then
calculated*
Sample Analysis
All SSMS analyses were performed on an AEI MS-7 mass spectrometer equipped
with a photographic plate detector.
The prepared sample pins were loaded into the source of the mass spec-
trometer* A source vacuum pressure of at least 1 x 10~? mm Hg is maintained
during sparking. A series of decreasing exposures, i.e., 100, 60, 30, 10, 6,
3, 1, 0.6, 0.3, 0.1, 0.06, 0.03, 0.01, 0.006, and 0.001 nC, totaling 210 nano-
coulombs is deposited on a photographic plate. This produces a permanent photo-
plate spectrum for interpretation.
After exposure, the photographic plate was removed from the instrument
and developed using controlled developing times.
87
-------
Photoplate interpretation is carried out using the "just disappearing
line" method. This allows simultaneous detection of all ion spectra and spec-
tra can be recorded at the highest available resolving power of the instrument.
Reproducibility of the plate response can be as good as + 3 to + 5%.
The use of the photoplate allows the recording of spectra without prior
knowledge of the sample, i.e., in some cases unexpected interferences in the
spectra would lead to erroneous results using electrical detection methods.
Mercury was analyzed by double-gold amalgamation-flameless AA.
Quality Assurance
Precision of SSMS photoplate results, normally interpreted to be spectra
read from a photoplate several times by the same analyst, can be as good as
+20%. However, a precision value that is normally of interest is the total
precision of the method on trace analysis from sample to sample. The sample-
to-sample value observed in the Instrumental Analysis Division of GTE experi-
ence in a coal matrix is not better than + 100% and is probably in the range
of + 100 to 300% at a parts per million concentration utilizing the visual
"just disappearing" line technique.
Accuracy of the SSMS is, of course, related direcly to trace element stan-
dards in a given matrix and the ability to repeat analysis with precision in
relation to a known standard.
It is generally agreed that the spark source value can be within a factor
of three to five times the absolute elemental concentration at the parts per
million level. Major component (MC) in a solid is an elemental concentration
greater than 1,000 ng/g unless otherwise stated. Major component in a liquid
is normally an elemental concentration greater than 10 |ig/ml unless otherwise
stated. When an element is reported as "heterogeneous," it means that while
sparking, an enrichment or elemental inclusion was passed through and what is
normally a linear function of exposures on the photographic plate is now very
nonlinear. The concentration reported on a heterogeneous element is therefore
probably a maximum value.
A variety of samples were analyzed to monitor analytical accuracy and
precision. NBS reference fly ash (SRM 1633) and pine needles (SRM 1575) were
employed as primary monitors of analytical accuracy. Possible sample matrix
interferences in aqueous condensate and hydrogen peroxide solution were moni-
tored through analysis of samples fortified with known quantities of metals
of interest. Analytical accuracy of fuel oil analyses were monitored using a
mixture of Conostan® organometallic standards. To monitor analytical preci-
sion, blind duplicate samples of fuel oil, wood, bottom ash, primary and sec-
ondary collector ash, and inlet and outlet Method 5 filter catch particulates
88
-------
for one run were submitted for analysis. The results of blind duplicate analy-
ses are reported in the respective analysis sections* All quality assurance
samples were submitted without origin identity.
Results for reference materials are summarized in Tables A-2 and A-3 for
NBS fly ash and pine needles* Metal concentrations observed in NBS fly ash were
compared to certified and uncertified values reported by NBS and additional
values reported by Ondov et al», in Anal. Chem.. 47:1102 (1975). Generally good
agreement (50 to 150%) between observed and reported values was found. Of the
exceptions, Ta « 11%), Yt (43%), Cs (47%), I (31%), Br (25%), and Cr (206%),
only chromium content is certified by NBS, which suggests that the remaining
elements may be inhomogeneously dispersed in the reference materials, and
thereby may produce variable results. Elements having the lowest recoveries
(I and Br) may be lost as volatile species in the spectrometer vacuum system.
Losses of Ta, Yt, and Cs are unexplained; however, the observed losses do not
appear serious* The analysis of SRM fly ash was considered to reflect accuracy
for all ash and particulate samples.
Selected elements in NBS pine needles (SRM 1975) were analyzed by SSMS as
one measure of quality assurance for the wood elemental analyses. The results
are summarized in Table A-3. In many cases, the certified values were found to
exceed the SSMS analytical range, as found for Fe, Mn, Ca, K, P, and Al. There-
fore, these results were not amenable to recovery calculations. An additional
group of elemental concentrations are reported by NBS but not certified. These
elements are Tl, Ce, La, Sb, Cd, Br, Ni, and Co. The recoveries of noncertified
materials generally fell within the expected accuracy range of SSMS (+ 100 to
+ 300%) with the single exception of bromine. Of the certified elements, results
were generally acceptable within the limits of SSMS.
The organometallic oil standard was prepared from Conostan organometallic
materials. Agreement with actual values was generally acceptable (30 to 300%)
as summarized in Table A-4. Several outliers which were noted include: B (< 3%);
Cd « 2%); Ca (1,000%); Fe (350%); Pb (< 7%); Mg (420%); and V (350%). Inter-
actions between Ca, Mg, and V during AA analyses performed on similar materials
by Conostan® were noted; however, the specific source of enhancement, inter-
ference, or contamination has not been identified. The low recovery of boron
may be due to hydrolysis and subsequent losses during handling. Losses of cad-
mium and lead may be due to loss of volatile organocadmium or lead materials
during sample preparation. The causes of high iron and magnesium recoveries
are unknown.
As summarized in Table A-5, the recovery of selected metal fortifications
of aqueous condensate are generally good (50 to 150%) with the exception of
Ni and Be. Spectral interferences precluded accurate analysis of nickel. The
loss of beryllium is currently unexplained, but may be due to loss of volatile
compounds.
89
-------
TABLE A-2. ELEMENTAL ANALYSIS OF NBS SRM 1633 COAL FLY ASH BY SSMS2/
Element
Uranium
Thorium
Bismuth
Lead
Thallium
Mercury
Cold
Platinum
Iridium
Osmium
Rhenium
Tungsten
Tantalum
Hafnium
Lutecium
Ytterbium
Thulium
Erbium
Holmium
Dysprosium
Terbium
Gadolinium
Europium
Samarium
Neodymium
Praseodymium
Cerium
Lanthanum
Barium
Cesium
Iodine
Tellurium
Antimony
Tin
Indium
Cadmium
Silver
Palladium
Rhodium
Ruthenium
Molybdenum
Niobium
Observed
concentrations/
Wg)
10
25
0.6
54
2
SR
-
-
-
-
-
3
-
5
0.7
3
0.4
3
4
6
1
3
2
15
26
12
76
76
MC
4
0.9
<. 0.4
5
2
STD
1
<_ 0.4
-
-
-
26
38
Actual
concent rat ionk'
fcg/g)
11.6 i 0.2£/
-
-
70 i 4£/
-
-
-
-
-
-
-
4.6 -I- 1.6
1.8 i 0.3
7.9 i 0.4
1.0 + 0.1
7 + 3
-
-
-
-
1.9 + 0.3
-
2.5 i 0.4
12.4 > 0.9
-
-
146 -I- 15
82 + 2
2,700 + 200
8.6 + 1.1
2.9 + 1.2
-
6.9 + 0.6
-
_
1.45 + 0.06£/
-
-
-
.
-
-
Recovery
(7.)
89
-
-
77
-
-
-
-
-
-
-
65
< 11
63
70
43
-
-
-
-
53
-
80
121
-
-
52
93
-
47 .
31
-
72
• -
_
69
.
-
-
-
-
-
(continued)
90
-------
TABLE A-2, (continued)
Element
Zirconium
Yttrium
Strontium
Rubidium
Bromine
Selenium
Arsenic
Germanium
Gallium
Zinc
Copper
Nickel
Cobalt
Iron
Manganese
Chromium
Vanadium
Titanium
Scandium
Calcium
Potassium
Chlorine
Sulfur
Phosphorus
Silicon
Aluminum
Magnesium
Sodium
Fluorine
Oxygen
Nitrogen
Carbon
Boron
Beryllium
Lithium
Hydrogen
Observed
concentration^/
(lig/g)
190
64
MC
63
3
14
49
15
65
260
99
97
22
MC
500
270
220
MC
17
MC
MC
41
MC
MC
MC
MC
MC
MC
•^ 460
NR
NR
NR
740
INT
_> 820
NR
Actual
concentration^/
1,000 ug/s; and INT = interference.
aj All values in micrograms per gram except when noted*
b/ Values reported by J. M. Ondov et al., Anal. Chem.. 47(7):1102 (1975).
c/ Values certified by NBS.
91
-------
TABLE A-3. SSMS ELEMENTAL ANALYSIS OF NBS
PINE NEEDLES SRM 1575
Element
Observed
concentration
NBS
certified
concentration
Recovery
Uranium
Thorium
Lead
Thallium
Mercury
Cerium
Lanthanum
Antimony
Cadmium
Strontium
Rubidium
Bromine
Arsenic
Copper
Nickel
Cobalt
Iron
Manganese
Chromium
Calcium
Potassium
Phosphorus
Aluminum
< 0.01
< 0.01
3
< 0.01
0.12
0.2
0.1
0.1
< 0.1
2
5
0.1
?!/
4
1
0.09
> 100
> 100
2
MC
MC
MC
> 14
0.020 ± 0.004
0.037 ± 0.003
10.8 ± 0.5
0.05^/
0.15 ± 0.05
0.41/
0.21/
O.Zi/
< 0.51/
4.8 ± 0.2
11.7 + 0.1
9a/
0.21 ± 0.04
3.0 + 0.3
3.5JL/
O.Li/
200 ± 10
675 + 15
2.6 ± 0.2
0.41% + 0.02%
0.37% ± 0.02%
0.12% ± 0.02%
545 + 30
< 50
< 27
28
< 2C£/
80
sok/
50^'
50^'
42
43
330
133
29—'
90^'
-
77
-
-
-
a^l Noncertified values reported for informational
purposes only.
])/ Recovery data based on noncertified values.
£/ Analysis by atomic absorption methods.
d/ The observed value reflects its maximum amount of
material present due to a heterogeneous signal.
92
-------
TABLE A-4. SSMS QUALITY ASSURANCE ANALYSIS OF TRACE METALS IN OIL
Element
Observed
concentration
(ug/g)
Reported^/
concentration
(ug/g)
Recovery
Aluminum
Antimony
Arsenic
Barium
Boron
Cadmium
Calcium
Chromium
Copper
Iron
Lead
Magnesium
Manganese
Mercury
Molybdenum
Nickel
Phosphorus
Selenium
Silver
Silicon
Sodium
Tin
Titanium
Vanadium
Zinc
0.5
0.5
0.5
2
£ 0.05
< 0.03
15
3
0.7
5
< 0.1
6
1
2.42k/
1
1
1
0.3
0.9
MC£/
0.2
0.6
3
5
0.2
1.44
1.50
1.61
1.44
1.44
1.44
1.44
1.44
1.44
1.44
1.44
1.44
1.44
1.72
1.44
1.44
1.44
1.51
1.44
1.44
1.44
1.44
1.44
1.44
1.44
35
33
31
140
< 3
< 2
1,000
210
49
350
< 7
420
69
140
69
69
69
20
63
> 100
14
42
210
350
14
a/ Based on dilutions of stock concentrations of Conostan®organometallic
standards, supplied by Conostan Division, Continental Oil Company,
Ponea City, Oklahoma.
b_/ Flameless atomic absorption analysis.
c/ MC = major component.
93
-------
TABLE A-5. SSMS ANALYSIS OF SELECTED ELEMENTAL FORTIFICATIONS IN
AQUEOUS CONDENSATE
Element
Lead
Mercury
Antimony
Cadmium
Selenium
Arsenic
Zinc
Copper
Nickel
Manganese
Chromium
Vanadium
Titanium
Beryllium
Total
observed
content
(us)
200
115
150
125
250
225
225
200
INT
> 250
> 250
> 250
> 250
< 0.05
Background
condensate
content
(us)
< 2
0.5
< 0.5
< 0.5
£ 0.75
5
75
20
500
75
1,200
2.5
15
< 0.5
Observed
fortification
(us)
200
114.5
150
125
250
220
150
180
-
> 175
> 250
> 248
> 235
-
Actual
fortification
200
200
200
200
200
200
200
200
400
200
1,000
200
200
200
Recovery
(%)
100
58
75
63
125
110
75
90
-
>. 90
UNK
>125
>118
0
Note: INT = interference.
UNK = unknown.
Conclusive data for Mn, Cr, V, and Ti was not available because the quantita-
tive analytical range of the SSMS is exceeded for these elements; however, min-
imum recoveries were calculated for Mn, V, and Ti.
The recovery of selected metal fortifications in 30% hydrogen peroxide
solution also was generally good (50 to 150%) as tabulated in Table A-6. Low
recoveries of Sb (40%), Cd (20%), and Pb (40%) were observed which has been
attributed to the relatively high volatility of these elements and their com-
pounds. Losses most likely occur during sample preparation or during analysis
in the spectrometer vacuum system. Two other potentially volatile elements, As
and Se, were recovered at levels of 100 and 140%, respectively, indicating no
losses during sample preparation or analysis. Unidentified interferences pre-
cluded the analysis of beryllium in the fortified peroxide solution.
94
-------
TABLE A-6. ELEMENTAL ANALYSIS OF FORTIFIED 30% HYDROGEN PEROXIDE
BY SSMS
Observed Reported
concentration concentration Recovery
Element
Antimony
Arsenic
Beryllium
Cadmium
Chromium
Copper
Iron
Lead
Manganese
Nickel
Selenium
Titanium
Vanadium
Zinc
2
5
INT
1
3
6
- 12
2
3
6
7
7
3
9
5
5
5
5
5
10
10
5
5
10
5
5
5
10
40
100
-
20
60
60
120
40
60
60
140
140
60
90
Note: INT = interference.
Problems
General good agreement between observed and reported values for reference
materials and fortified samples were observed. The most prevalent problem ob-
served was the loss of the more volatile elements such as cadmium or lead which
may be expected* Other problems associated with a specific element or sample
matrix are discussed in the quality assurance section.
Reference
Conostan Division, Continental Oil Company, Ponca City, Oklahoma.
POLYNUCLEAR AROMATIC HYDROCARBON ANALYSIS
Samples
Samples of bottom ash, primary collector ash, secondary collector ash,
and aliquots of Florisil train extracts were analyzed for selected hydrocarbon
95
-------
and heterocyclic PAH materials* A duplicate of each ash sample type, a forti-
fied sample of each sample type, and reagent blanks were analyzed for quality
assurance.
Sample Preparation
Ash samples were prepared specifically for PAH analysis as described in
the following section* Florisil train components were extracted for PCB anal-
ysis and were subsequently aliquoted for PAH analysis* The preparation of
Florisil train samples is only briefly described in this section. Additional
detail on preparation of Florisil train samples is presented in the PCB analy-
sis section. All samples were fortified to a concentration of 40 to 80 ng/(j,l
of D-10 anthracene as an internal standard prior to GC/MS analysis*
Approximately 20 g of each ash sample were mixed with preextracted anhy-
drous sodium sulfate to form a free-flowing mixture prior to extraction with
150 ml of methylene chloride in Soxhlet extractors for 8 hr. The Soxhlet ex-
tractors were wrapped with aluminum foil to protect the samples from light.
The sample extracts were subsequently concentrated to a volume of 5 ml with
aluminum foil-wrapped Kuderna-Danish evaporators. All samples were stored in
the dark at 4°C prior to analysis.
Samples from the Florisil train components were aliquoted prior to the
sulfuric acid treatment, which is prescribed for PCB analysis, and analyzed
directly.
Sample Analysis
Preliminary sample analysis was attempted using a capillary GC equipped
with a flame ionization detector (FID); however, the complex chromatograms ob-
served precluded successful interpretation and necessitated use of GC/MS analy-
sis. Samples were analyzed by GC/MS using a 10 m, SE 30 SCOT capillary column.
Identification and quantitative determination of specific PAH compounds was
made based on retention times and extracted ion plots characteristic of the
compounds of interest.
The chromatographic conditions used for the capillary column analysis of
samples for PAH are given in Table A-7* A reconstructed ion chromatogram (RIG)
of the 14-component standard is presented in Figure A-l* As illustrated, two
clustered peaks are produced by more than one compound. With selective ion mon-
itoring, it was found that chromatographic resolution was achieved for phenan-
threne and anthracene at scans 561 and 574, respectively, as shown in Figure
A-2. Although anthracene and D-10 anthracene were not clearly resolved chro-
matographically, mass resolution was accomplished for ions at m/e 178 and 188,
respectively, as also illustrated in Figure A-2. Similarly, the cluster at scan
1173 produced by 1,2-benzanthracene and chrysene was resolved as illustrated
in Figure A-3.
96
-------
TABLE A-7. GC/MS OPERATING CONDITIONS FOR PAH ANALYSIS
Column: 10 m SE 30 SCOT capillary
Column temperature: Initial hold 100°C for 4 min, program rate
8°C/min to 260°C
Instrument: Finnigan 4000
Injection mode: Splitless
Column pressure: 2 psi
Scan rate: 1*2 sec/mass decade
Scan range: 400-400 amv
lonization potential: 70 ev
Emission current: 0.2 ma
Multiplier voltage:
97
-------
lOO.O-i
VD
oo
75.0-
50.0-
25.0-
D-10 Anthracene
Anthracene
Acenaphthylene
Phenanthrene
\
Fluorene
200
4:00
400
8:00
600
12:00
Fluoranthene
Pyrene
1,2 Benzanthracene
Chrysene
BenzaQa] Pyrene
800
16:00
1000
20:00
Indeno (1,2.3-c. d) Pyrene
1.2.5,6 Dibenzanlhracene
Benzoperylene
1400
28:00
1600
32:00
SCAN
TIME
Figure A-l. Reconstructed ion chromatogram for PAH standard.
-------
18. On
ny/e » J76
Intensity =9408
100.0-i
ny/e = 178
Intensity "52.224
14.4-,
m/e =229
Intensity « 7520
100.0-1
m/e »188
Intensity =45,632
450
9:00
-i 1
500
10:00
i Anthracene
Phenanthrene
D-10 Anthracene
550
11:00
600
12:00
650 SCAN
13:00 TIME
Figure A-2. Chromatographic resolution of phenanthrene, anthracene,
and D-10 anthracene.
-------
28.7-1
o
o
100. On
19.8n
m/e = 226
Intensity =9152
m/e = 228
Intensity =31,904
Chrysene
1,2 Benzanthracene
m/e = 229
Intensity = 6312
1100
22:00
1)50
23:00
1200
24:00
1250
25:00
1300 SCAN
26:00 TIME
Figure A-3, Chromatographic resolution of 1,2-benzanthracene and chrysene.
-------
A relatively simple heterocyclic PAH standard was prepared from carbazole,
dibenzothiophene, and three benzoquinoline isomers, 3,4-, 5,6-, and 7,8-benzo-
quinoline. The RIG from a 20 ng/ml standard solution analyzed using the same
chromatographic conditions as used for the hydrocarbon PAH materials is illu-
strated in Figure A-4. Chromatographic separation was achieved for each com-
pound type; however, the benzoquinoline isomers were neither chromatographi-
cally nor mass resolved* Sample data was screened using selective ion plots
for the three most intense ions of each heterocycle standard; however, these
heterocyclic compounds were not observed in any sample.
The RIG for a secondary collector ash extract is shown in Figure A-5 and
is generally indicative of the chromatographic complexity of secondary collec-
tor ash and bottom ash extracts. Relatively simple RICs were observed for pri-
mary collector ash and Florisil train extracts. In all cases, extracted ion
plots were made using the three ions of highest intensity characteristic for
each of the PAH compounds* Identifications were based on the presence of the
three ions maximized at the correct retention time and with the correct ratio
of intensities. Figure A-5 illustrates the positive identification of fluor-
anthene and pyrene in a secondary collector ash extract. Confirmation was ob-
tained from the full mass spectral data of the highest intensity chromatographic
scan corresponding to a suspected PAH* Mass spectral data were visually inter-
preted and submitted to computerized searching of a 25,000 compound library to
identify compounds producing similar mass spectra. Quantitative analysis was
achieved by converting digital peak intensity values to concentration values
using instrumental response factors based on analyses of standards.
Quality Assurance
One sample of each ash type was prepared in duplicate for determination of
analytical precision. Fortified samples of each ash type, and fortified blanks
of methylene chloride, Florisil, and impinger solution were prepared and car-
ried through the complete extraction and concentration steps to monitor prepar-
ative efficiency. Blanks of methylene chloride, Florisil, and impinger solution
were analyzed to identify background reagent PAH levels. All samples were pro-
tected from heat and light. An internal standard of D-10 anthracene was added
to compensate for retention time variations and response variability. Prior to
analysis, the MS was calibrated with perfluorokerosene. Standards at three lev-
els were analyzed intermittently throughout the sample analyses'.
Recoveries for fortified samples, as shown in Tables A-8 and A-9, were
found to vary from one PAH to another and also among different sample types.
Difficulty in the recovery of many PAH materials from both the primary and sec-
ondary collector ash appears to be a result of the inability of methylene chlo-
ride to extract PAH from collector ashes in a Soxhlet extractor. This conclusion
is based primarily on low extraction recoveries of long retention time PAH com-
pounds fortified at both low and high (10 times low) fortification levels.
101
-------
lOO.O-i
o
10
75.0-
S 50.0
25.0-
0-10 Anthracene
500
10:00
Dibenzotliiophene
550
11:00
3,4-Benzoquinoline
5,6-Benzoquinoline
7,8-Benzoquinoline
Carfauzole
600
12:00
650
13:00
700 SCAN
14:00 TIME
Figure A-4. Chromatogram of heterocyclic PAH standard.
-------
u.3n
m/e = 100
Intensity = 698
O
03
1
17.1-
100. 0-
Ve = 101
Intensity = 1062
A Fluoranthene
m/e = 202 ||
Intensity = 6200 | I
i Pyrene
800
6:00
850
17;00
900
18:00
950
19:00
1000 SCAN
20:00 TIME
Figure A-5. Identification of fluoranthene and pyrene in secondary collector ash.
-------
TABLE A-8. QUALITY ASSURANCE DATA FOR LOW LEVEL PAH FORTIFICATION
OF SELECTED SAMPLES
Total
Fortified
fortification Observed
quantity quantity
Compound
Acenaphthy lene
Kluoreiie
Hhenanlhrene
Anthracene
Klnoranthene
I'yrene
Cbrysene
1 ,2-Benznnthracene
benzol a Ipyrene
I'ei'ylene
indenol 1,2, 3-c,d Ipyrene
1,2,5 ,6-l)ib
-------
TABLE A-9. GC/MS QUALITY ASSURANCE DATA FOR SELECTED LONG RETENTION
TIME PAH COMPOUNDS
Total
Primary collector
ash
Secondary
collector ash
Compound
fortification Observed Observed
quantity quantity Recovery quantity Recovery
(*>
Chrysene
1 , 2-Benzanthracene
Benzo [ajpy rene
Perylene
Indeno[l,2,3-c,d]pyrene
1,2,5, 6-Dibenzanthracene
1,12-Benzoperylene
1,162
1,202
1,328
826
1,320
763
1,256
92.2
205
58.0
44.8
4.6
0
18.1
8
17
4
5
0.4
0
1
71.8
181
46.4
34.9
32.5
0
21.8
6
15
3
4
2
0
2
Analytical precision for selected PAH materials in ash extracts is sum-
marized in Table A-10. A relative deviation of 17% for two analyses of phenan-
threne was the only precision measurement possible for bottom ash. Relative
deviations ranging from 6 to 48% were observed for PAH compounds detected in
secondary collector ash. The most likely source of large analytical variabil-
ity is in the relatively low concentrations actually analyzed and large back-
calculation multipliers which tend to enhance small variations. Precision cal-
culations were not made for analyses of primary collector ash and Florisil
train component extracts, since the quantity of PAH materials were at or below
the analytical detection limits.
Problems
The most serious problem encountered was the inability to quantitatively
recover selected PAH compounds from primary and secondary collector ashes. The
problem observed appears to result from the inability to extract the PAH com-
pounds efficiently from the collector ash matrix. This conclusion is based on
the good recoveries of fortifications from the bottom ash materials. Addi-
tionally, the recoveries of PAH compounds from primary collector ash have been
generally lower than the recoveries of the same materials from secondary collec-
tor ash. Similar observations during the PCB analysis reinforce the conclusion
that primary and secondary collector ashes are significantly different in their
ability to absorb or adsorb organic material and most likely are different in
their chemical composition. Since many PAH sample concentrations were found
at levels at or below the analytical detection limits, precision calculations
could not be made in most cases.
105
-------
80.00 280.00 480.00 680.00
880.00 1080.00 1280.00
Electron Energy, eV
1480.00 1680.00 1880.00 2080.00
Figure 13.i Auger survey scan of 1-y, particulate.
-------
observed in the TOO analyses* With the exception of the XAD-2 LC Fractions 6
and 7 and concentrated XAD-2 extract, sample extracts were not found to pro-
duce IR spectra of sufficient intensity to allow interpretation.
*
The TCO chromatogram for concentrated XAD-2 resin extract prior to LC
fractionation is shown in Figure A-8, which depicts a complex mixture contain-
ing in excess of 50 components, many beyond the specified C-17 cutoff for TCO
analysis* Additional low concentration materials may be unobserved in the con-
tinuum of overlapping peaks* After LC fractionation, the bulk of the collected
material was observed in the polar sixth and seventh fractions*
The direct inlet LRMS analysis of XAD-2 sample and blank extracts was per-
formed* Thermograms and ion range plots were constructed to identify scanning
regions of interest* Spectra from each selected scan range were compared. As
illustrated in Figure A-9, a complex mixture of ions was observed for each
range of interest* Detectable ions were observed through the mass range of each
scan* Comparison of sample and blank data indicated serious background problems
from the less volatile components in both extracts* Due to the mixture complex-
ity, spectral complexity, and high background levels, detailed interpretation
of this data was not attempted*
Quality Assurance
Quality control was carried out in accordance with those procedures speci-
fied in the draft revision for Level 1 organic analysis. Samples for quality
assurance include those for calibration procedures and analysis of blanks. Spe-
cific procedures used are described below*
TCO analyses were performed using commercially prepared standard mixtures
containing Cg-C^Q compounds and C^g-C^g compounds for retention time and detec-
tor response standards* Standards were chromatographed at the beginning and
end of an analysis day to monitor column performance or detector response* The
IR spectrophotometer was calibrated using polystyrene during each day of use.
Calibration of the analytical balance was made using a set of NBS Class S qual-
ity weights before proceeding with any GRAY determinations*
Burdick and Jackson distilled in glass solvents (methanol, methylene chlo-
ride) were used to preextract the XAD-2 resin* All methylene chloride used for
extraction purposes was prepared from the same manufacturer's lot* A blank of
XAD-2 resin was prepared from the same manufacturer's lot and preextraction
batch as the XAD-2 resin used in the actual SASS sampling run* This blank was
taken to the field and exposed to the field lab conditions as was the actual
sample* This XAD-2 blank subsequently was carried through all Level 1 analysis
procedures used for the XAD-2 sample used in the SASS run, including LC frac-
tionation and analysis* A methylene chloride blank volume greater or equal to
that used for particulate and/or aqueous extractions was prepared from the sol-
vent lot used for all analyses* The methylene chloride blank was carried through
116
-------
fr
,2
i
Retention Time
Figure A-8. Typical TCO chroraatogram of concentrated extract
from XAD-2 sample.
-------
00
Figure A-9. Direct inlet LRMS of concentrated XAD-2 field sample extract.
-------
all Level 1 analysis procedures including TCO, GRAV, and IR but not LC frac-
tionation*
Problems
Generally insufficient material was present In GRAV residues to produce IR
spectra of sufficient intensity for interpretation. This may be due to either
the small absolute quantity of available material, the large number of compo-
nents present, and/or some combination of these factors*
TCO chromatograms indicated the presence of a complex mixture of materi-
als* It was found that a large number of the observed materials were excluded
from the TCO analysis* This was a result of the limitation of the TCO analysis
to those materials eluting earlier than C-17 carbon atom standard for the GC
conditions used in the Level 1 protocol*
The analysis of both the XAD-2 resin field sample and field blank indi-
cated a significant background level in both TCO and GRAV samples* Pretest TCO
analysis of the XAD-2 resin had indicated the resin to be acceptably cleaned
(according to Level 1 requirements)* It would appear that background levels
were encountered under either field lab or sample preparation conditions* Cor-
rections for background levels were made to the TCO and GRAV values for XAD-2
resin when appropriate* Interpretation of direct inlet LRMS data was not at-
tempted due to the large number of components observed in the TCO analysis,
the spectral complexity, and the high background contribution observed in the
XAD-2 blank.
The batch inlet LRMS analysis of the XAD-2 field sample was not performed*
The exclusion of this analysis is based on the relative abundance of TCO mate-
rials (£0.05% w/v) and the complexity of the mixture present as evidenced in
the TCO analysis (> 50 components). Although the absolute quantity of the sam-
ple TCO material (5.6 mg) exceeds the Level 1 batch inlet LRMS requirements,
previous experience has proven the inability to introduce sufficient sample
material into the MS without solvent swamping problems*i/
Reference
1* Golembiewski, M., K* Ananth, G* Trischan, and E. Baladi. Environmental
Assessment of Waste-to-Energy Process: Braintree Municipal Incinerator.
Final report.
119
-------
APPENDIX B
SAMPLING LOCATIONS AND PROCEDURES
LOCATION OF SAMPLING POINTS
For this assessment program, samples of the following feed and effluent
streams were taken:
• Wood chip feed;
• Fuel oil feed;
• Bottom ash discharge;
• Primary and secondary collector ash discharge;
• Boiler exhaust gases prior to the mechanical collectors; and
• Stack emissions after the collectors*
Specific sampling locations for each of these material streams are described
below. Figure B-l shows an overall view of the test site, including locations
of the MRI field trailers and opacity observer locations*
Wood Fuel
Wood samples were taken from each of the four augers which feed Boiler
No. 1 and composited at the end of each particulate run* Figure B-2 shows a
typical feeder arrangement and the location from which the samples were ob-
tained*
Oil Feed
Samples of the No* 2 fuel oil feed were drawn from an in-line valve which
was located several feet upstream from the oil burners.
120
-------
N
Power House
3 Boilers +3 Generators
Observer /•
Location
A.M.
Figure B-l. Layout of Burlington Electric Plant facilities.
121
-------
Sampling
Location
From Wood Chips
Storage Bins
Boiler
Wood Chips
Compressed Air
Figure B-2« Example of sampling location for wood
feed stream.
122
-------
Bottom Ash
Boiler bottom ash is discharged gravimetrically into two storage bins di-
rectly beneath the end of the stoker grate* Samples were collected via clean-
out doors located at the base of each bin* Normally, these hoppers are emptied
pneumatically and the ash is conveyed to a storage silo* But during the MRI
tests, the pneumatic system was not operated to allow for the ash sampling*
Primary and Secondary Collector Ash
A schematic illustration of the primary and secondary collectors (and hop-
pers) is shown in Figure B-3* Separate samples were collected from each of the
four hoppers (two per collector) at the conclusion of the test day, during
physical removal of the collected ash* The pneumatic removal system was not
in operation during the MRI test program.
Boiler Air Emissions Prior to Collector
At this location, samples were taken for determination of total particu-
late concentration (EPA Method 5), gas analysis (Orsat) and particle size dis-
tribution* Figure B-4 is a schematic diagram showing the duct dimensions and
locations of the individual sampling points. The nine sampling ports were 1.5
equivalent duct diameters downstream and 1*0 duct diameters upstream from the
nearest flow disturbance, and were located in a plane perpendicular to the
air flow. The number of traverse points (54) was determined according to EPA
Method 1*
The particle size samples were drawn through a separate port at a point
located 18 in. inside the duct (see Figure B-4).
Stack Emissions (After the Collector)
An illustration of the sampling location arrangement at the outlet of the
collectors is shown in Figure B-5* The nine sampling ports were located 3.8
equivalent duct diameters downstream and 2.3 equivalent duct diameters upstream
from the nearest flow disturbance. The 45 traverse points used in the Method 5
particulate sampling were determined according to EPA Method 1.
Particle size and SASS samples were taken at average velocity points
through additional ports as shown in Figure B-5. The sampling point for the
continuous analyzers was located deep inside the duct, about 4 ft upstream of
the sampling ports. This minimized the possibility of air in-leakage and dilu-
tion of the sample stream going to the continuous monitors.
123
-------
From Boiler To Stack
"T T
Secondary
Collector
Pneumatic Conveyor
*Samples were drawn during the ash
removal through the hatch.
Figure B-3» Illustration of the collector
hopper arrangement.
124
-------
1.
' '
43"
T
18"
•• 1
— »-v^^— >_*• —
108"
444
4-44-
4- + 4-
+ • 4 4-
+ 4-4
444-
njuuLJir
a ) Top View
^~~-^ — »• • — ,
iJ2" , 6"'
+• 4- 4-
-h 4- -h
4-44-
4-44-
4- 4- 4-
nmj2^^
TlO°
^-^s * *-^
3.6"
From Boiler
Bend
Particle Sizing Sampling Ports, 4" I.D.
466 6~6°6
^Sampling Ports 3-1/2" I.D.
Grating
6
66
Bend
To ^Collector
90" = 1.5Dia.
62" = 1.0 Dia.
b ) Front View
9 Particulate Sizing Sampling Point
4 Particulate & Orsat Sampling Points
Figure B-4. Schematic diagram of the inlet
sampling location.
125
-------
108"
Grating
•v_
a) Top View
Bend
t
To Stock
4" I.D. Particulate
Sizing Sampling Ports
\56
666 666 6 6 5.
^3- 1/2" I.D. /
Sampling Ports /
Grating t
ii
Bend
t
— :
123" =2.3 Oia.
To Continuous
Analyzers
208"=3.8Dia.
From Collector
b) Front View
• Particulate Sizing Sampling Point
O SASS Train Sampling Point
A Continuous Analyzer Sampling Point
+• Particulate, Orsat 4 PCB Sampling Points
Figure B-5» Schematic diagram of the outlet sampling location!
126
-------
SAMPLING AND GRAVIMETRIC ANALYTICAL PROCEDURES
Several sampling and analytical procedures were followed during the con-
duct of this study. All of these procedures are either approved, referenced,
or recommended by EPA. Published EPA reference methods were followed whenever
they were applicable.
Particulate Mass Concentration
EPA Methods 1 through 5 of the Federal Register (42:160, August 18, 1977;
and 36:159, August 17, 1971) were followed in the sampling and analysis of par-
ticulate matter. Figure B-6 is a schematic illustration of the sampling train
used.
Initial velocities, temperatures, and moisture contents of the flue gas
were estimated from a previous compliance test report. The data from the ini-
tial estimate were used to preset the sampling equipment for the start of iso-
kinetic sampling.
Particulate sampling and analysis were accomplished according to EPA
Method 5 of the above mentioned Federal Register. Fifty-four traverse points
were utilized to sample for particulate from the collector inlet duct. The sam-
pling time at this inlet was 2 min/traverse point for a total of 108 min for
each test. The collector outlet was divided into 45 traverse points for partic-
ulate sampling. The sampling time at the outlet was 2 min/traverse point for
a total of 90 min for each test.
Particle Sizing
Particle size measurements were accomplished by utilizing the real-time
particulate sizing system described below. MRI-established sampling procedures
and the manufacturers' manuals were followed in conducting the tests.
Two optical particle counters were used to count the number of particu-
late in the flue gas: the Climet Model 0208A optical counter, manufactured by
Climet Instrument Company; and the GE Model 112 L428 Gl condensation nuclei
counter, manufactured by General Electric Company. Since condensation nuclei
counters count the total number of particles present in a stream, a diffusion
battery was used to classify those particles into different sizes. The TSI
Model 3040 diffusion battery, manufactured by Thermo-System, Inc., was used
to classify the submicron particle sizes through the diffusion principle.
Optical counters are capable of measuring only low particle concentrations
(ambient level); therefore, a dynamic dilution system, designed and fabricated
by MRI, was incorporated in front of the counters to sample, dilute, and con-
dition the sample continuously. Figure B-7 is a schematic illustration of the
particle counter's sampling system used for this project. A photographic view
127
-------
Thermometer"
NJ
00
Thermocouple
Reverie-Type'1
Pilot Tube
Healed
Compartment
•» Console
q/ Impingers 1,3 and 4 are of the Modified Greenburg-Smith Type
Impinger 2 is of the Greenburg-Smith Design
Impinger I and 2 Contain 100ml Water
Impingcr 3 Empty
lniping«r 4 Contains 200-300 Grams Silica Gel
Figure B-6. Schematic illustration of Method 5 sampling train in sampling position,
-------
Stack Wall
Optical
Counter
To CN Counter
1-Cyclone (l.OOacfm, Participate Cutoff Diameter 2.
2, 3-Flowrate Meters
4-Dried & Filtered Pressurized Air Supply
5-Dilutor
6-Neutralizer
7-Filter
8-Condenser
Figure B-7. Schematic illustration of the MRI dynamic dilution
optical counter particle sizing system*
129
-------
of the dilution system is given in Figure B-8. Figure B-9 is a photographic
view of the particle counting system in sampling position.
Continuous Monitoring for Gases
The continuous monitoring system for gases consisted of the following gas
analyzers:
1» IBC 02 analyzer (polarographic cell);
2* Bendix NOX analyzer (chemiluminescent);
3. Beckman S02 analyzer (NDIR);
4. Beckman HG analyzer (NDIR); and
5. Beckman CO analyzer (NDIR)*
The continuous monitoring system drew its sample from a common manifold*
The sampling interface between the sampling point and the manifold, which is
about 30 ft in length, consisted of a 1/4-in* heated Teflon tubing*
Analyzer manuals were followed in operating this continuous sampling sys-
tem* Each monitor was connected to a strip chart recorder*
PGB/PAH Sampling System
Figure B-10 is a schematic diagram of the PCB/PAH train used for this proj-
ect* The procedures published in the EPA environmental monitoring series (EPA-
600/4-77-048, November 1977) were followed in sampling and analyzing for PCB
pollutants* PAH analysis was conducted using MRI-established procedures*
Source Assessment Sampling System (SASS)
Operating procedures for the SASS train as published by EPA "Level 1 En-
vironmental Assessment" (EPA-600/2-76-160a, June 1976 and subsequent modifica-
tions) were followed for these tests* This system is shown schematically in
Figure B-ll in sampling position*
130
-------
Figure B-8. Photograph of the MRI-developed dynamic dilution system*
-------
u>
ro
Figure B-9» Photograph of the particle counting system In sampling position.
-------
Probe ff*t**
Reverse-Type
Pi tot Tube
Manometer
Control Box
I
Figure B-10. PCB sampling train.
-------
Stack Temperature T.C.
Filler
Holder
3-Way Solenoid Valve
— To Ice Both
From Ice Bath
Impingers Check
Valve
Probe Temperature T.C.
Clquld Foliage —
Gal Passage-
Goi Cooler
» Cooling Fluid
Reiervolr
W_ Immersion
Heater
XAD-2 Cartridge—-
Condeniate
Reiervolr
__ Temperature
Controller
Fine Adjustment
By Paw Volve
Vacuum
Goge
Coane
Adjustment
Air Tiglil Valve
Vacuum
Pump
Orifice &P
Mognehelic Gage
_g/lmplnger No.) contains 750ml of 6MHyOj
Impingen No.2 ond 3 each contains 750ml
of 0.2M (NH4)2S2O8 *0.2MAgNO3
Impinger No.4 contains 750 grams Orierite
Figure B-ll. Schematic diagram of the SASS train.
-------
APPENDIX C
COMPARISON OF CHEMICAL ANALYSES OF BOTTOM ASH,
PRIMARY ASH, AND SECONDARY COLLECTOR ASH
Chemical analyses of bottom ash, primary collector ash, and secondary col-
lector ash have yielded data which may indicate significant differences between
the various ash types. Prior to chemical analyses, bottom ash was ground to
produce a powder; primary and secondary collector ashes were analyzed as re-
ceived without grinding. The primary collector ash, which included fragments
of charred wood, was found to be noticeably coarser than the secondary collec-
tor ash*
Analysis of PAH fortifications from the various ash matrices were highly
variable as sunmarized in Table C-l» Recovery of PAH fortifications from the
primary collector ash for a 5 ng/g fortification was generally less than 20%
with the exceptions of acenaphthylene (35%) and fluorene (55%). Recoveries of
a similar fortification from secondary collector ash were somewhat better for
the earlier eluting PAH compounds fluoranthene (5 to 12%) and chrysene (0 to
10%). Detectable quantities of the later eluting PAH materials (benzo[ajpyrene,
perylene, indeno[l,2,3-c,d]pyrene, 1,2,5,6-dibenzanthracene, and 1,12-benzo-
perylene) were not observed for the 5 ng/g fortification in either primary or
secondary collector ash. As also shown in the table, the possibility of losses
due to heating, handling, and/or simple surface interactions appears unlikely
based on the recoveries of the same PAH materials from heated solvent and from
ground bottom ash fortified at the 5 ng/g level and subsequently Soxhlet ex-
tracted. Analyses of 50 ng/g fortifications of the later eluting PAHs in pri-
mary and secondary ashes are summarized in Table C-2. The recovery of later
eluting PAH compounds generally was low for both primary and secondary ash
extracts.
The peculiar difference of primary collector ash from either bottom ash
or secondary collector ash is illustrated in Figure C-l (a, b, and c), which
depicts the GC/MS reconstructed ion chromatograms (RIG) for the three ash types
from a single test run. Figure C-la illustrates the RIG of numerous compounds
in a methylene chloride extract of bottom ash. In contrast, the RIG of primary
collector ash which has been extracted and analyzed under the same conditions
produces only two responses in addition to that of the internal standard, as
shown in Figure C-lb.
135
-------
TABLE C-l. QUALITY ASSURANCE DATA FOR LOW LEVEL PAH FORTIFICATION
OF SELECTED SAMPLES
Total
fortified solvent
fortification Observed
quantity quantity
Compound
Aceiiaphthylene
Kluorenc
I'lionaiulirene
Anthracene
Kluoianthene
Pyrene
Chrysenc
1 ,2-Uenzunthrut:ene
Henzo(a]pyrene
Perylcne
Indeiio[ 1 , 2,3-c,d]pyrone
] ,2 ,5,6-l)llienzanthracene
1 ,12-Uenzoperylene
(MK
187
185
162
131
146
123
95
133
135
73
129
112
79
)
.8
.3
.1
.1
.7
.1
.5
.5
.8
.7
.3
.0
.1.
(Mg)
116.8
134.1
129.0
106.4
142.6
118.6
106.0
133.7
146.0
86.5
183.6
164.6
109.2
Recovery
Boctotr ash
Observed
quantity^'
(%) (MS)
62
72
80
81
97
96
111
100
108
117
142
147
138
94.
103.
HI.
57.
87.
75.
77.
39.
53.
35.
61.
41.
44.
8
9
2
4
4
3
0
7
0
3
0
6
0
Recovery
00
51
56
69
44
60
61
81
30
39
48
47
37
56
Primary collector ash
Observed
quuntlty^'
(MK)
65.8
101
24.5
9.67
7.95
6.11
0
1.45
0
0
0
0
0
Recovery
m
35
55
15
7
5
5
0
1
0
0
0
0
0
Secondary collector ash
Observed
quant lty£/
(MR)
84.9
142
104k/
99.0
18.2k/
42.6b/
9.6
10.3k/
0
0
0
0
0
Recovery
(%)
45
77
64k/
75
12k/
35k/
10
ek/
0
0
0
0
0
u/ Values corrected to D-10 ,mtlir;iecne Internal standard.
l_ Values corrected for sniuplo contrihutIon.
-------
TABLE C-2. GC/MS QUALITY ASSURANCE DATA FOR SELECTED LONG
RETENTION TIME PAH COMPOUNDS
Primary collector Secondary
Total ash collector ash
fortification Observed Observed
quantity quantity Recovery quantity Recovery
Compound (^g) (ng) (%) (pig) (%)
Chrysene
1 , 2-Benzanthracene
Benzo [ajpyrene
Perylene
Indeno[l,2,3-c,d]pyrene
1,2,5, 6-Dibenzanthracene
1 , 12-Benzoperylene
1,162
1,202
1,328
826
1,320
763
1,256
92.2
205
58.0
44.8
4.6
0
18.1
8
17
4
5
0.4
0
1
71.8
181
46.4
34.9
32.5
0
21.8
6
15...,.,,
"3
4
2
0
2
137
-------
11
IV
*« •« 'MB
ill
tJD 1100
Figure C-l. GC/MS chromatograms for bottom ash, primary collector
ash, and secondary collector ash extracts.
138
-------
Finally, Figure C-lc presents the RIG of secondary collector ash extract which
contains numerous components and appears very comparable to the bottom ash ex-
tract. From these observations, it is evident that either organic material is
absent from the primary collector ash or the organic material is not efficiently
extracted* The observation of strongly colored extracts observed from secondary
and bottom ash in contrast to the pale yellow to colorless extracts observed
from primary ash further reinforces the conclusions* Additional aliquots of
primary collector ash were extracted with similar visual and GC/MS results.
A similarly unexpected phenomenon was observed during the analyses of ex-
tracts from ash samples which had been fortified with Arochlor 1254 PCS mate-
rial* As discussed in the quality assurance section, recovery of Arochlor 1254
fortifications from both bottom ash and secondary collector ash were on the
order of 100%* In the case of fortified primary collector ash, certain PCB com-
pounds were found to be selectively absent from the characteristic Arochlor
1254 patterns* The selectively absent compounds were not found to be related
to a particular number of chlorine atoms per molecule; losses of di-, tri-,
tetra-, and pentachlorinated biphenyls were found* The presence of a strongly
absorptive/adsorptive material in the primary collector ash is strongly sug-
gested by these results*
Comparison of averaged SSMS data for the series of duplicate ash analy-
ses provides some additional information* Although many analyses were within
an order of magnitude, primary collector ash generally was found to contain
low or intermediate component concentrations for many elements* Secondary col-
lector ash was generally found to contain intermediate to high component con-
centrations compared to primary collector ash* Bottom ash was most frequently
found to contain elemental concentrations between the levels observed for pri-
mary and secondary ashes* It was also noted that several of the more volatile
elements, e.g., Hg, As, Se, Br, and F, appeared to be successively enriched
from bottom ash through primary collector ash and secondary collector ash* It
cannot be determined from the SSMS data whether the primary collector ash con-
tains significantly different proportions of major constituents, such as car-
bon or silicon, when compared to bottom ash or secondary collector ash because
in all cases these elements are identified as major components in the SSMS
results.
To investigate the presence of major differences in the particulate which
could be expected between carbonaceous and silicate materials, a crude density
gradient experiment was performed on the three ash types. A density range of
1*6 to 2*8 g/ml was found using carbon tetrachloride and tetrabromoethane in
varying proportions* The differences in the effective densities of ground bot-
tom ash, primary collector ash, and secondary collector ash are illustrated
in Figure C-2, The bulk of the ground bottom ash is observed in the higher
density regions above a density of 2*4 g/ml* This would be expected of a sili-
cate material* The primary collector ash is found to contain particles having
139
-------
1.6
•
1.8
2.0
2.2
2.4
2.6
2.8
Ground Primary Secondary
Bottom Collector Collector
Ash Ash Ash
Figure C-2. Photo of density gradient experiment,
140
-------
widely variable effective densities* However, the bulk of the primary collec-
tor ash is less dense than either the bottom ash or secondary collector ashes•
The relatively uniform intermediate density of secondary collector ash is also
illustrated. During the course of this experiment the presence of relatively
large particulates resembling charred wood were also noted in the primary col-
lector ash* It appears that the most significant sources of difference between
the solid materials tested which may affect sample extraction and analysis are
major component composition and associated physical properties, as evidenced
by the observed differences in effective particle density.
141
-------
SECTION 5
ENVIRONMENTAL ASSESSMENT OF BURLINGTON DATA BASED ON EPA'S SAM-lA
APPROACH
The environmental assessment data obtained from MRI's study of the
Burlington power plant provides a good deal of quantitative and semiquantita-
tive information about pollutants in the air, water, and solid waste streams
resulting from thermal processing of refuse. To ascertain the significance of
these data and thereby determine the need for control of any particular pollut-
ants that may be of concern, a comparative set of emission criteria is needed.
Although some emission level goals have been established in the form of EPA's
New Source Performance Standards and Effluent Guidelines and Standards, these
regulations are source-specific and extrapolation to other types of processes
may not be valid. There are also many pollutant species, such as many forms of
hydrocarbons, certain pesticides, trace metals, etc., for which emission limi-
tations have never been promulgated.
Therefore, a standardized data evaluation methodology recently developed
by EPA's Industrial Environmental Research Laboratory - Research Triangle Park/
Energy Assessment and Control Division (IERL-RTP/EACD) was employed to inter-
pret the Burlington test results. The IERL-RTP/EACD evaluation scheme consists
of using a data calculation procedure known as SAM-lA to compare effluent
stream contaminant levels to Multimedia Environmental Goals (MEGs), which are
also being developed by IERL-RTP. MEGs are guideline limitations for contami-
nants that are judged to be: (a) appropriate for preventing certain negative
effects in the surrounding populations or ecosystems (ambient air, water, or
land); or (b) representative of the control limits achievable through applica-
tion of the best available technology. These emission level goals are approxi-
mated using: (a) ambient concentration goals which are based on hazards posed
to public health and welfare as a result of long-term or continuous exposure
to emissions; (b) natural background levels which can be used as ultimate goals
for the elimination of discharges; and (c) hazards to human health or to ecol-
ogy from short-term exposure to emissions. Those MEGs derived from short-term
exposure data are known as MATE values and are intended to serve both as rela-
tive hazard indicators and very approximate emission guidelines. Pollutants
72
-------
TABLE E-3. (continued)
CONTINUATION SMftr FOR ITEM NO 4. FORM IA02. ICVCl 1
Page
3/3
snimrf/rnNiHni nniifiN Burlington EFFUIEN1 SIRFAU nn 302
A
VIMPIt (KACItOM
UNI IS
67 /Ta
68/Cr
69/Ho
70/W
71 /Mil
72/Fe
73/Ru
74/Co
75/Rh
76/Nl
77/Pt
78/Cu
79/Ag
flO/Au
8l/Zn
82/Cd
83/Hg
84/La
B
r*ACIIUN
CONT.tN
IBAIION
tig/8
<0.2
110
23
<0.2
447
>1,000
<0.2
8.7
<0.2
27
<0.2
64
<0.2
<0.2
70
<0.47
0.66
23
C
KAini
MM[
CONCCN
IflAIION
H8/8
1.SE2
0.50
1.SE2
3.0E1
0.50
3.0
N
1.5
0.03
0.45
O.OG
l.OEl
0.50
N
5.0E1
0.10
0.02
3.4E3
0
(oxonoit
MA 1C
CONUN
1RAIUN
|lg/8
N
0.50
1.4E1
. N
0.20
0.50
N
0.50
N
0.02
N
0.10
0.10
N
0.20
0.002
0.50
N
E
offiutt or
nntm
IMtAllll)
(8/O
—
O.001
220
0.15
<0.007
894
>333
-
S.8
<6.7
60
<3.3
6.4
<0.4
-
1.4
<4.7
3.3
0.007
r
OHOIfMt
POSIIION IN
IITAllll MAII
IAU1[
.
G
MOTCC or
HA2ARO
(rcntoaoAii
2,000
-
17
-
1,350
-
640
<2
-
350
<235
1.3
-
H
ORUN41.
POSIIION IN
ICOt MAir
IAOII
1
%/,r
IK Aim
MAIC
OCtCMO
—
/
/
/
/
/
/
/
/
/
/
J
»
IC.OL
MM[
iicrinin
—
/
/
/
/
/
/
/
/
/
v/
/
K
L
IOXIC UHtl CHSCHARGC RAII
(IHAIIH
BAVOI
(E > i we ii
g/sec
<0.06
1.23B4
8.4
<0.39
4 .99E4
>1.86E4
-
3.24E2
O.74E2
3.35E3
<1.84E2
3.57E2
<22.3
-
78
262
1 .84E2
0.39
(CcaoncAi
BA$(OI
in > IIHC ji
g/scc
-
1.23E4
89
-
I.25E5
M.12E5
-
9.49E2
-
7.53E4
-
3.57E4
<1.12E2
-
1.95E4
1.31E4
72.5
-
-------
LO
TABLE E-4.
SAM/IA WORKSHEET FOR LEVEL 1
SECONDARY COLLECTOR ASH WORKSHEET
Form IA02 Level 1
i SOURCE/CONTROL OPTION
BurllngLoii Electric Wood and Oil Fired Boiler
2 crn.ucNi SIIICAM
303 Secondary Collector Ash
cnoc • NAMC
Pjge 1 ' 3
3 EfFLUENI STREAM FLOW RATE
Q. 1.55 g/sec
(gis • m'/jec — liquid « l/iec — tolid *>sle ° g/»«)
4 COMPIEIE THE rniLOWING TABLE FOR THE tfflUENl STREAM OF LINE 2 (USE BACK OF FORM FOR SCRATCH WORK)
A
lAMPU IftACIION
1INI1S
SSHS
27/U
28/Na
29/K
30/Rb
31/Ce
32/Be
33 /MR
34/Ca
35/Sr
36/Ba
37/n
38/A1
39/fia
LI
tRACIIOM
CONCIN
IRAIhlN
tig/g
57
>1,000
>1,000
66
4.3
6
>1,000
>l,000
>1,000
>1,000
200
> 1.000
156
C
lit AllH
MAII
CIKTfN
IRAIIUN
H8/8
0.70
1.6E3
N
3.6E3
2.5E3
0.06
1.8E2
4.8E2
9.2E1
1.0E1
9.3E1
1.6E2
1.5E2
u
[COtdOtTAI
uAir
CONCfN
inAIKIN
l'8/8
0.75
N
N
N
N
0.11
1.7E2
3.2E1
N
5.0
5.0E1
2.0
N
E
rannic m
HA/ARO
(IKAIIIII
EFFLUENT STREAM DEGREE (
HEALTH MATE HASFD (I COL
ECOIOUICAL MATE HASED (I
(INTER HERE AND AT LINE H.
F H A: Ann
rife 6'961E3
C(IL 0) 5bl.072E6
IORM IAOI)
6 NIIMREf
COMPAR
HEAIIH 6.
ECOIOGIC'
1 OF ENTRIES
ED 10 MATES
a 6h._
H
nxoiNAi
itismnti IN
I cot UAII
IAIIII
-
1
N/ll
III A| III
MAII
mcumo
-
/
v/
/
/
/
/
/
/
\
v/«
ICOl
UAII
UCIIIIIO
-
v/
y
/
/
/
v/
K
IOKIC UMtl 0
IK Al III
BASCDI
If . IINI Jl
g/sec
1.26E2
0.98
_
0.03
0.003
1.55E2
8.7
3.3
17.1
1.55E2
3.4
9.8
1.6
L
v:>
-------
TABLE E-4. (continued)
CONllNUAflON SHEET FOR ITEM NO «. TORM IA02. LEVEL I
Tags
2 1 3
Ln
qnnnrr/r.nNiHOi OPIION Burlington EfflUl.NI S1REAM NO _1Q3 .
A
SAUrtf (BACIION
IINI1S
40/In
41/Tl
42/C
43/S1
44/Ge
45/Sn
46/Pb
48/P
49/As
50/Sb
51/Bl
53/S
54/Se
55/Te
56/F
57/Cl
58/Br
59/1
60/Sc
61/Y
62/Tl
63/Zr
64/llf
65/V
66/Nb
0
r BAT not)
CONCtN
IHAIIUN
Hg/g
Internal St
< 0.6
Not Repo
> 1,000
29
5
4
>1,000
650
14
<0.6
>l,000
98
<0.8
>1,000
350
28
4.3
58
112
>1,000
170
<3.5
>630
56
C
m AIIII
MAU
CONCIN
IDAIION
Ug/8
andnrd
3.0
Led
3.0E2
1.7E1
N
0.50
3.0E1
0.50
1.5E1
1.2E1
N
0.10
3.0
7.5E1
2.6E3
N
N
1.6E3
3.0E1
1.8E2
1.5E1
1.5
5.0
6.5E2
O
icmoiiicAi
MAIC
CONCIN
IHAIKJN
|ig/8
N
N
N
N
0.10
0.001
0.10
0.40
N
N
0.05
N
N
0.01
N
N
N
N
N
N
N
0.30
N
E
WGRtC Or
HAfAPn
OICAllll)
(H/CI
—
0.02
3.3
1.7
-
8
33
1.3E3
0.93
0.05
-
980
0.27
13
0.14
-
-
0.04
3.7
33
11
2.3
126
0.09
F
ODOrNAl
POSH IflN IN
IITAIIHUAII
MBit
—
G
otnmi or
HAIADII
uroioGiCAi)
m'Di
—
.
_
• -
-
40
1.0E6
6.5E3
35
-
.
1,960
-
-
3.5E4
-
-
.
.
-
-
-
2.1E3
-
H
OAUlNAI
POSItlOH IN
tax UAH
lAOIf
1
v/u
tIC AIIII
MAIt
OCIincn
—
/
•
/
/
/
/
/
/
/
/
/
_/_
J
-Ja
(CO).
MAIC
occcniD
—
/
/
/
/
/
/
/
K
IOXIC UNII 0
(IrtAUM
BAS(U)
it . imi Jl
g/sec
0.03
5.1
2.6
-
1.2
51.2
2.02E3
1.4
0.08
.
1.52E3
0.42
20.2
0.22
-
-
0.06
5.7
51.2
17.1
3.6
195
0.14
L
SCIIARfJ DAK
CICOtO-JCAl
DAUOI
10 i nut i>
g/sec
_
_
.
.
62
1.55E6
1.0E4
54.3
.
-
3 .04E3
-
-
5.43E4
-
-
.
.
-
-
-
-
-
-------
TABLE E-4. (continued)
CONTINUATION SIIEEI FOR ITEM NO 4. FORM IA02. IEVEI 1
Page
3/ 3
Ui
Ui
tniiurF/rnMifini DPIIIIN Iliirl l.np.i:oii ._ IFFlUtNl SIRf.AM NO 303...
A
SAUTII rnAciHW
IINIIS
67/Tn
68/Cr
69/Ho
70/W
71 /Mil
72/Fc
73/Ru
74/Co
75/Rh
76/N1
77/Pt
78/Cu
79/Ag
80/Au
81/Zn
82/Cd
83/llR
84/La
PAII
\cenaphthyl
'liennnthrei
'luoranlhei
'yrenc
B
rHACIIUN
COMCfN
IRAtlUN
(18/8
<0.6
374
39
3.7
> 1,000
> 1,000
<0.6
57
<0.6
297
<0.6
160
<0.6
<0.6
89
1.3
5.9
85
cne2.3
e 6.5
c 1.9
0.74
C
IITAUM
MA1E
COMCtN
iRAnm
1^8/8
1 .5E2
0.50
1.5E2
3.0F.1
0.50
3.0
N
1.5
0.03
0.45
0.06
l.OEl
0.50
N
5.0E1
0.10
0.02
3.4E3
N
4.8E1
2.8E3
6.9E3
n
rcoior.icAi
MAIl
cow.rN
inAfuw
HB/8
N
0.50
1.4E1
N
0.20
0.50
N
0.50
N
0.02
N
0.10
0.10
N
0.20
0.002
0.50
N
N
N
N
N
E
M'GKtf 01
mum
(Ml All Ml
IB/O
—
0.004
748
0.26
0.12
2.0E3
333
-
38
20
660
10
16
1.2
.
1.8
13
295
0.03
-
0.14
0.0007
0.001
F
OMNHAl
POSIIIQN IN
IHAIIM MAIt
IABII
—
G
txr.mi n
HAlAftt)
(fcmocicAii
(0/01
—
-
748
2.8
-
5.0E3
2.0E3
-
114
-
1 .49E4
-
1.6E3
6
.
445
650
12
-
-
•
-
-
II
ORDINAL
POSITION m
ICIX MA 1C
IABII
1
%/il
HI AIIX
MAU
Kcicnco
—
/'
/
/
/
v/
/
/
/
/
/
/
/
J
S/ll
cent
MAI[
ixcftncn
—
/
/
V/
/
/
/
/
/
/
/
/
K
L
lOXC UNII DISCHARGE RAII
OUAIIH
OASfOI
(C I I "4[ 3)
g/sec
0.006
1.16E3
0.40
0.19
3.1E3
5.16E2
-
5.9
31
1.02E3
15.5
24.8
1.9
-
2.8
20.2
4.57E2
0.05
-
u.n
0.001
0.002
(ICO OGKAl
OAttOI
in • UIK i)
g/sec
-
1.16E3
4.3
-
7.75E3
3.1E3
-
1.77E2
-
2.31E4
-
2.48E3
9.3
-
6.90E2
1.01E3
18.6
-
-
™
-
-
-------
TABLE E-5. STACK EMISSIONS WORKSHEET
SAM/IA WORKSHEET FOR LEVEL 1
Form IA02 Level 1
1 SOURCE /CONTROL OPIION Page , , 6
Burlington Electric Wood and Oil Fired Boiler
2. EFFLUENT SIREAM 3 EFFLUENT STREAM FLOW RATE
I/M „ . „ . . „ 21.1 cu in/sec
IOI Sf--"-|r Fmissl""s Q »
Cone * NAMI (gas « ni'/ser - Itqujd > l/jec — lolu] wasle * (/tec)
4 COMPLETE THE FOLLOWING TABLE FOR THE EFFLUENT SIREAM OF LINE 2 (USE GACK Of FORM FOR SCRATCH WORK)
A
SA1IHI IRACIION
IINIIS
CO
S02
NOX (N02)
SSMS of 1
27/L1
28/Na
29/K
30/Rb
31/Ce
32/Be
33/Mg
34/Ca
35/Sr
G
fllArlfON
COOIS
Ug/cu m
2.43E5
3.59E5
1 .2/.E5
eLhod 5 Pai
5.1
a/
aJ
8.0
0.32
<0.28
a/
a/
>63
C
in AIIH
MAII
ClINUN
1RA1ION
ug/cu m
tt.0f.lt
1 .3E4
9.0E3
ticulate
2.2E1
5.3E4
N
1.2E5
8.2E^>
2.0
6.0E3
1 .6F.4
3.1E3
D
[CfMOTilT.AI
MAII
r.OMCIM
IDAIIUN
UR/CII m
1.2E5
N
N
N
N
N
N
N
N
N
N
N
E
otiwtf nt
HA/ANA
lU'CI
-
6.1
28
14
0.23
-
-
6.7E5
3.9E6
<0.14
-
.
>0.02
F
IHKIINAI
IDSIIKIN IN
HCAIIM MAII
1*111 (
-
G
on, mi (»
IfA/ANU
ilcmnr.n AH
dcni
-
2.0
.
-
-
-
-
-
-
-
-
.
-
II
ONIXNAI
I*OSI1I(JN IN
ltd MAII
IAOII
-
1
V/.I
MtAllll
MAII
I>CIIMU
-
\/
y
v/
J
V/ll
fCOl
MAIt
IICICPCD
-
/
K
KUC Ulfll 0
BAUD)
1C • IINI Jl
cu ra/sec
129
591
295
A .9
-
-
1.4E3
8.2E5
O.O
-
-
>0.42
I
v:nAnc( RAIC
iir.otiir.irAi
RA^LOI
|G > IMf 1|
cu ro/sec
42
_
-
-
-
-
-
-
-
-
-
-
a MORI STACI it Htliifo use A COHIINIIAIIIW silici
S EFFLUENT SIREAM DEGREE U
HEALTH MAIE BASED (I COI
tCOIOT.ICAl MAIE HASFD (I
(ENTER HERE AND AT LINE 8.
F HAZARD
OS. 178'6
COI 0) 5l> _.
FORM IA01)
15.9
6 NUMPEf
COMPAH
HEALTH 6j
ECOLOCICA
OF ENTRIES 7 TC
ED TO MATES |fl
EC
1 6b . (E
IXIC UNIT DISCHARGE SUM
:ALTH MATE RASED d COL. K)
OLOniCAI. MATE FIASfD C COL
NTER HERE AND AT LINE 8. FOJ
1f 3.763E3
U 7h 33A.3
)M IAOI)
-------
TABLE E-5. (continued)
CONTINUATION SHEET rOH ITEM NO «. fORM IA02. LEVEL 1
Page 2 / 6
Ln
sniinrr/rnNiHni OPIION Burlington ffFlUfNI SIREAM NO JO1
A
SAMTlC IRAOIION
UNI IS
36/Bn
37/U
38/Al
39/Ga
41/Tl
43/S1
44/Ce
'.5/Sn
46/Pb
48/P
49/As
50/Sb
51/Di
53/S
54/Se
5S/Te
56/F
57/C1
58/Dr
59/1
60/Sc "
6i/v
62/Ti
63/Zr
6'i/lie
B
IRACIKIN
CONCIN
1RAIION
Hg/cu in
>95
a/
a/
33
<0.03
«/
8.7
2.6
>69
>104
50
2.9
<0.06
a/
29
<0.03
a/
a/
28
2.5
1.1
1.9
>100
8.3
<0.03
C
iir.At.iH
MAIC
CONCtN
IBAIITOI
Ug/cu ra
5.0E2
3.1F.3
5.2E3
5.0E3
1 .OE2
1.0E4
5.6E2
N
1.5E2
1.0E2
2.0
5.0E2
A1.E2
N
2.0E2
1.0E2
2.5E3
N
N
N
5.3EA
1.0E3
6.0E3
5.0E3
5.0E2
0
icmnnirAi
UAIt
COW IN
IRAIMN
|ig/CU HI
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
I
Menu or
IIAIAUn
(IIUIIM)
(n/ci
—
>0.19
-
-
6.6E3
O.OEA
-
0.02
-
>O.A6
>1.0
25
5.8E3
<1 .5E/t
-
0.15
<3.0E4
-
-
-
-
2.1E5
I.9E3
>0.02
1 .7E3
<6.0E5
r
OITDINAl
lOSIIIOM IN
IUAIIII UAI[
IAOK
—
G
ofcnri 01
IHIU>U
(CCOlOClCAll
(0/0)
—
-
-
-
-
• -
.
.
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
H
ORIMNAl
position m
(COt MAIC
lABIt
1
v/lf
IKAIIH
UAK
[>CC[MO
—
/
}
\/ll
crm
UAIt
f ici coin
—
K
L
IOKIC uwr OISCIMOGC R«rc
((OllH
OASCD)
(I • UNI Jl
cu in/ sec
:>4.0
-
-
0.1/t
<6.3E3
-
O.A2
-
>9.7
>21
528
0.12
<3.2E3
-
3.2
<6.3E3
-
-
-
-
4.4E4
O.OA
>0.42
0.04
C.3E3
ucoi OOCAI
OAHOI
1C < UNC 11
cu ra/sec
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
~
-------
TABLE E-5. (continued)
CONTINUATION SHEET fOR ITEM NO 4. rORM IA02. LEVEL 1
Page
3/ 6
Ol
00
KDiini-F/rnMiBDi nriinw Burlington EniUFN! SI RE AM NO _1Q1_. ....
A
SAMFIt (MArilOH
UNI IS
65/V
66/Nb
67 /Ta
68/Cr
69/Mo
70/W
71 /Mi,
72/Fe
73/Ru
74/Co
75/Rli
76/Ni
77/Pt
78/Cu
79/AR
80/Au
81/Zn
82/Cd
83/llg
84/I.a
SASS XA
LCI
LC2
LC3
B
rRACIION
CONCIN
1RAIION
Hg/cu m
13
l.l
<0.03
<0.016
3.3
<0.03
56
>104
<0.03
<0.2
<0.03
25
<0.03
55
1.8
<0.03
>100
7.3
7.7
- 4.4
-2 Resin Ei
0.11
0.006
1.2
C
tVALIH
MAIf
COMCtN
1 DAI KIN
|ig/cu m
5.0E2
2.4E4
5.0E3
1.0
5.0E3
I.OE3
5 .OE3
1.0E3
N
5.0E1
1.0
1.5E1
2.0
2.0E2
l.OEl
N
4.0E3
l.OEl
5.0E1
1.1E5
tract
100
110
0.02
0
icoiw.ir.Ai.
MA IE
CWM N
INAIKttl
)ig/cu m
1.0
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
l.OEl
N
N
N
N
E
OtGKU Of
IcAMrni
OltAI III)
0.10
-
<4 .OE3
<0.03
1.7
<0.02
0.28
0.18
-
>0.03
0.73
0.15
4.0E5
0.001
5.5E5
60
F
(lOlllNAl
TOSIIIOM IN
HfAIIHUAlt
itnit
—
G
Menu (*
HAiABO
(CcniociCAii
(O/DI
—
13
"
-
-
• -
-
-
-
-
.
.
.
-
-
-
.
-
-
0.77
-
-
-
-
H
OAUNAl
rosniON IN
tax UAIC
lABll
—
1
Ja
IUAIIII
MAIC
Encitnro
—
/
/
i
-Jo
ICO.
MAII
KCtinio
—
/
K
L
milC UNII OlSCMAROf RAIC
IIKAllll
BAWD)
(t • UM1 J)
cu ra/eec
0.63
9.7E4
<1.3E4
<0.42
0.014
<6.3E4
0.21
>2.1
-
<0.08
<0.63
36
<0.42
5.9
3.8
.
>0.63
1.5
3.2
8.4B4
0.02
0.001
1.27E3
{(COICGKM
OASCOI
(C t UNt ))
cu m/sec
274
-
-
-
-
-
-
-
-
.
.
.
-
-
-
-
-
-
16
-
-
-
-
-------
TABLE E-5. (continued)
Ln
vO
CONTINUATION SHEET FOR ITEM NO 4. FORM IA02. LEVEL I
Page 4 / 6
•yiunrf/rnNinni OPTION Burlington ... EMi.UENI STREAM NO I0_l
A
SAUPIE mACIIOt
UNI IS
LC4
LC5
LC6
LC7
SSMS of
27/U
28/Na
29/K
30/Rb
31/Ce
32/Ue
33/Mg
34/Ca
35/Sr
36/Ba
37/B
38/Al
39/Ga
41/T1 -
43/Sl
44/Ce
45/Sn
46/Pb
47/P
B
IRACIION
COW 1 N
IBM MM
M>g/cu m
0.011
_
8.2
7.4
Aqueous Coi
0.011
5.2
1.4
<0.009
<0.009
<0.009
1.4
19
0.09
0.05
0.32
0.54
0.11
<0.009
11
<0.009
0.16
1.6
6.2
C
IIIA11H
MAIl
COMtCH
1HAIKIN
LLg/CU ID
990
1.3E3
170
100
densaCe ;
2.2E1
5.3E4
N
1.2E5
8.2E4
2.0
6.0E3
1.6E4
3.1E3
5.0E2
3.1E3
5.2E3
5.0E3
1.0E2
1.0E4
5.6E2
N
1.5E2
I.OE2
n
ICIHOfiir.AI
MAIE
CONCCN
IRAIION
HS/cu m
N
N
N
N
nd Pe^roxi
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
I
otnntl or
HAMHO
(IITAUM)
IH'C)
—
1.1E5
„
0.05
0.07
le Imping
5.0E4
9.8E5 .
-
<7.5E8
-------
TABLE E-5. (continued)
CONTINUATION SHEET FOR ITEM NO 4. fOnM IA02. LEVEL I
Page
a. I
o i
S/limrf/rONIMOl OPIUM H,irlinglnn . .. ffflUlNF SIRCAM NO Ifll —
A
SAUPII IRACIION
tINIIS
/.9/As
50/Sb
51/Bi
53/S
54/Se
55/Te
56/F
57/Cl
58/Br
59/1
60/Sc
61/Y
62/Ti
63/Zr
64/llf
65/V
66/Nb
67 /Ta
68/Cr
69/Mo
70/W
71/Mn
72/Fe
73/Ru
74/Co
B
UUCIION
COMCIN
IRAIIOfl
Hg/cu m
0.21
0.003
<0.009
>24
0.13
<0.009
2.2
5.4
0.25
0.02
<0.002
<0.009
0.51
<0.014
<0.009
0.11
0.029
<0.009
>38
1.5
<0.009
5.4
>59
<0.009
0.88
C
lltAIIM
MAir
an: in
INAHCIM
|Lg/cu m
2.0
5.0E2
4.1E2
N
2.0E2
1.0E2
2.5E3
N
N
N
5.3E4
1.0E3
6.0E3
5.0E3
5.0E2
5.0E2
2.4E4
5.0E3
1.0
5.0E3
1.0E3
5.0E3
1.0E3
N
5.0E1
0
icmoccAi
MA 1C
CO"C[N
IRAtlUN
|ig/cu m
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
1.0
N
N
N
N
N
N
N
N
N
E
nionit of
IIA/MIU
IIHAllll)
(II'C)
—
0.11
6.0E6
<2.2E5
-
6.5E4
<9.0E5
8.BE4
-
-
-
<3 .8E8
<9.0E6
8.5E5
<2 .8E6
<1.8E5
2.2E4
1.2E6
<1 .8E6
>38
3.0E4
<9.0E6
1.1E3
>0.06
-
0.018
f
OnVINAl
nmiinN m
lit Allll MAlt
IABIE
—
G
Mnnci or
HA/ARO
rccm.oGiCAi >
(n/oi
—
-
-
-
-
• -
-
-
-
-
-
.
-
-
-
-
0.11
-
_
-
-
-
-
-
-
-
M
ORWNAl
rosi'ion IN
ICOt UAIt
IABII
1
V/l»
IMAIIH
UA1C
cucirnco
—
/
J
v/ir
tea
MAlt
oriroro
—
«
K
lOIK UNII CM
(I9AUH
BAUD)
(I . lINf 3)
cu m/sec
2.3
1.3E4
802
6.3E3
<1.9E4
0.023
>1.3
-
0.38
L
SCMABOf RAIC
(ICO! OOCAl
OAUO)
1C • I INC 1)
cu m/sec
-
-
-
-
-
-
-
-
-
-
.
-
-
-
-
2.3
-
-
-
-
-
-
-
-
-
-------
TABLE E-5. (continued)
CONTINUATION SHEET TOR ITEM NO 4. IORM IAOZ. LEVEL I
Page 6/6
SOtjncf/rnKiinni nr-linN Burlington EfflUENr SIIICAM NO 101
A
S>MFU IttACHOH
UNI IS
75/Rl!
76/Ni
77/Pt
78/Cu
79/Ag
80/Au
81/Zn
82/Cd
83/llg
84/La
-
0
(RACIION
COHC.tH
IRAIOTN
|ig/cu m
<0.009
>25
<0.009
>16
0.029
<0.009
>16
0.013
0.009
<0.009
C
Hi; Alllt
UAII
CONtfN
IRAIintl
|ig/cu ra
1.0
1.5E1
2.0
2.0E2
1.0E1
N
4.0E3
1.0E1
5.0E1
LIES
0
ICOIOUICM
U«IC
co»r(N
IRA (KIN
|ig/cu m
N
N
N
N
N
N
N
N
l.OEl
N
E
Menu or
HA<«ru
0tt»LIII>
(U/CI
—
<9.0E3
>1.7
<^i.5E3
>0.08
2.9E3
.
>4.0E3
1.3E3
1.8E4
<8.2E8
F
(MDINAl.
n)SIIION IN
IICAIIIIUAII
IABII
—
G
mnRcc w
HA/ADO
(tCOIOGIC«ll
(6/n>
—
-
.
_
_
• -
.
.
.
9.0E4
-
M
ORDINAL
POSIIIOH IN
(COL UAK
IAOLC
'
1
>/lf
lit Aim
UAIi
KCKMO
J
•Jtt
t cm.
MAI[
txciroto
—
/
K
L
IO>K UWI OISCIIAROC RAIC
(IIOIIH
RASTO)
([ • i mi n
cu in/sec
<0.19
>35.9
<0.095
>1.7
0.061
.
0.084
0.027
0.031
1.7E6
-------
emitted at levels lower than these guidelines are not expected to evoke'sig-
nificant harmful or irreversible responses from exposures of less than 8 hr
duration*
For each of the three environmental media (air, water, and land), there
are two sets of MATE values* The first, designated as health-based MATEs, is
based on evidence of acute and chronic effects on humans and also on studies
of animal toxicity. The ecology-based MATEs are derived from studies of pol-
lutant effects on plant and aquatic life*
MATE values, therefore, can be a very useful tool by revealing potential
environmental hazards in source assessment data. The SAM technique does not
work in all cases, however, since there are many contaminants for which MATEs
have not yet been established*
The SAM is basically a screening technique designed to provide a quanti-
tative means of assessing the pollution potential of a process or source. The
SAM format focuses on each separate effluent stream, whether it evolves from
the process itself or associated pollution control equipment. The information
resulting from the SAM procedure can provide the guidance required to progress
from a Level 1 to a Level 2 source analysis program, where specific chemical
compounds are targeted for study.
In the most basic version of SAM (designated as SAM-lA), which was used
to evaluate the data in this report, the emitted contaminants are assumed to
be noninteractive with the ambient environment (i.e., effluent components are
released without transformation and are not involved in synergistic activities)!
Also, no assumption is made about the dispersion modes of individual pollutants
to their receptor sites, except that such dispersion would have at least the
equivalent effect of the safety factors which are normally applied when con-
verting short-term exposure data to estimates of longer term chronic ambient
exposure levels*
The SAM-lA approach compares each sample fraction or specific pollutant
concentration in a given effluent stream to a corresponding MATE value. The
ratio of these quantities is known as the degree of hazard (H) for that par-
ticular pollutant. Any contaminant whose H value is greater than unity is
flagged as a potential problem pollutant. The product of a contaminant's H
value and the effluent stream flow rate establishes that contaminant's Toxic
Unit Discharge Rate (TUDR). The H values and TUDRs are then summed for each
effluent stream. This derived information can subsequently be used to rank
effluent streams on the basis of the magnitude of their TUDRs, establish pol-
lutant priorities and identify problem pollutants, and determine the need for
control/disposal technology development.
73
-------
BURLINGTON DATA
The Burlington Electric power boiler system has four major effluent
streams that were monitored for this study. They included: (a) bottom ash dis-
charged from the traveling stoker grate; (b) collected fly ash from the primary
mechanical collector; (c) collected fly ash from the secondary mechanical col-
lector; and (d) exhaust gas emissions. The constituents of each effluent stream,
as measured by MRI and discussed in the previous section, were compared to their
corresponding MATE values via the SAM-lA procedure. Subsequently, the H and TUDR
values were calculated for each stream as shown in Table 43. The data which were
used for these compilations appear in the tables in Appendix D.
TABLE 43. SUMMARY OF SAM-lA EFFLUENT ANALYSIS
Health-based Ecological-based
Degree of hazard
Bottom ash 4,400 1,020,000
Primary collector ash 2,100 1,020,000
Secondary collector ash 7,000 1,070,000
Stack emissions 180 16
Toxic Unit Discharge Rate
Bottom ash (g/sec) 110,000 25,600,000
Primary collector ash (g/sec) 115,000 56,900,000
Secondary collector ash (g/sec) 11,000 1,700,000
Stack emissions (cu m/sec) 3,800 330
Results of the SAM-lA analysis show the secondary collector ash to have
the highest health-based and ecological-based degrees of hazard. However, the
degrees of hazard (health) for the other two ash streams (bottom ash and pri-
mary collector ash) are of the same order of magnitude as the value for the
secondary ash, and the degrees of hazard (ecological) are also very similar
for all three ash streams. The boiler air emissions showed low degrees of haz-
ard in comparison to the other effluents.
74
-------
Because the primary collector ash has the highest material discharge rate
of the three ash effluent streams, it yielded the largest TUDR value. TUDRs
calculated for the bottom ash stream were just slightly less than the primary
ash values*
When assessing the data in Table 43, it must be noted that the summations
shown include some "less than" and "greater than" values which add a factor of
inaccuracy to the cumulative totals. For lack of a better strategy, absolute
values were used in computing the H and TUDR values presented in the table (see
the SAM-LA data sheets in Appendix D)«
The high degrees of hazard (health based) associated with each of the ash
streams are primarily the result of high sample concentrations/low MATE concen-
trations for a few trace metals. H values for the metals Li, Be, Cr, Mn, Fe,
and Ni in the bottom ash, for example, contributed 4,090 to the total stream
degree of hazard of 4,400. Likewise, the very large ecological-based degree of
hazard for the bottom ash stream (1,020,000) is predominantly the result of
the high H value for the element phosphorus (1,000,000). An additional 15,300
of the 1,020,000 total is due to the elements Li, Ba, Al, V, Cr, Mn, Fe, Ni,
Cu, and Cd. Similar patterns are evident from analysis of the data generated
for the two collector ash streams. Generally, the same elements were responsi-
ble for the high degrees of hazard in both cases.
The results of the SAM-lA analysis described above may not reflect the
true hazard potential of the ash streams because of the physical nature of the
ash. For example, the bottom ash is discharged as agglomerated chunks of vary-
ing size. The MATE concentrations used in the SAM-lA procedure are based on
the potentially hazardous effects of each element or compound. No provision ap-
pears in the methodology to relate the amount of a particular element measured
in the bottom ash to what may actually be released into the environment. It is
possible that much of the element may never be leached from the pieces of ash
material. Therefore, the degree of hazard, as determined by SAM-lA, could be
very overestimated.
Since TUDRs are the product of H values and the effluent stream discharge
rate, it follows that the primary collector ash would have the highest TUDR
values. Health- and ecological-based TUDRs for the bottom ash stream were only
slightly lower, while secondary collector ash TUDRs were at least an order of
magnitude less than the primary collector ash values. A combination of low de-
grees of hazard and a moderate effluent flow rate yielded TUDR values for the
stack emissions which were lowest among the four effluent streams.
The health-based degree of hazard determined for the stack emissions (180)
is mainly the result of high H values for CO, S02» NOX, and several trace met-
als, namely As, Ni, and Cr, plus one group of organic components (the LC3 frac-
tion from the SASS organic sample). It is possible, however, that the high Ni
and Cr concentrations could have resulted from corrosion of some stainless
75
-------
steel components within the SASS and do not accurately reflect the levels of
these metals in the boiler exhaust gases* Also, the designated MATE concentra-
tion for SASS organic fraction LC3 (0.02 ^g/m^) is a "worst case" assumption
based on the hazard potential of benzo[ajpyrene, since the exact composition
of LC3 is unknown* PAH analysis of the SASS organic sample did not indicate
the presence of benzo[ajpyrene. Therefore, it is certain that the H (health)
value for the LC3 fraction of the sample is suspect. This case points out a
deficiency in the EPA SAM-lA methodology.
The ecological-based degree of hazard is very low for the stack emissions
because only a few ecological-based MATE concentrations have been established
for airborne emissions.
The degree of hazard values obtained for S02 and NOX also point out the
conservative nature of some MATE concentrations. When viewed on a parts per
million basis, these gaseous constituents appear to be of little concern be-
cause of their relatively low levels. The SAM-lA analysis, however, shows them
as warranting concern since their degrees of hazard are greater than one.
Although many of the constituents analyzed in each of the effluent streams
had H values greater than unity, many others had very low degrees of hazard.
Therefore, the SAM-lA analysis methodology at least provides a means for iden-
tifying those pollutants which may or may not be of environmental concern. It
allows an initial analysis of environmental assessment data that would other-
wise not be available.
76
-------
SECTION 6
CONCLUSIONS
Summarized below are the major findings of the Burlington Electric wood
and oil fired boiler environmental assessment. They are presented according to
each of the effluent streams and the SAM-lA results*
BOTTOM ASH
• Elemental analysis of the bottom ash residue yielded a number of ele-
ments at concentrations greater than 1,000 |j,g/g. These included Ba,
Fe, Mn, Ti, Ca, K, S, P, Si, Al, Mg, and Na. Ba, Zr, Sr, and Li ex-
hibited the largest increases relative to their concentrations in the
wood and oil fuels*
• PCB analysis did not reveal the presence of any PCB materials in the
bottom ash at a detection level of 0.05
• Phenanthrene was the only PAH compound which could be confirmed as a
constituent of the bottom ash. The highest level of phenanthrene de-
tected was 1.7 p-g/g.
PRIMARY COLLECTOR ASH
• Major elemental components (> 1,000 p,g/g) of the fly ash captured by
the primary mechanical collector were Fe, Ti, Ca, K, S, P, Si, Al, Mg,
and Na.
• No PCB materials were detected in the primary collector ash.
• No measurable levels of PAH compounds were found in any of the primary
ash samples.
77
-------
SECONDARY COLLECTOR ASH
• The same elements found in the bottom ash and primary collector ash
at concentrations greater than 1,000 (j,g/g were also present at these
levels in the secondary collector ash, with the addition of Sr and F.
Elements in the range of 100 to 1,000 p,g/g included Ge, Zr, As, Ga,
Cu, Ni, Cr, V, Cl, and B.
• A trend was observed for the elements Hg, Br, Se, As, Cl, F, and B.
Each showed progressively greater concentrations from the bottom ash
through to the secondary collector ash, which suggests a fine particle/
condensation phenomena.
• No positive signals for PCB materials were observed at the 0.05 p,g/g
detection level*
• Several PAH compounds, however, were identified in the secondary col-
lector ash, namely acenaphthylene, phenanthrene, fluoranthrene, and
pyrene. Concentrations ranged from 0.3 to 10 p,g/g.
STACK EMISSIONS
• Concentrations of the gaseous criteria pollutants, S02, NOX, and CO
were low. Total hydrocarbon emissions were minimal.
• Intermittent elevated CO readings (up to 400 ppm) can probably be re-
duced by more consistent combustion conditions.
• The S02 levels observed would have been lower had the wood fuel not
contained a small amount of coal fines as a result of adjacent storage
piles.
• Total particulate emissions averaged only 0.18 g/dscm (0.08 gr/dscf).
On a heat input basis, the emissions averaged 0.09 g/MJ (0.17 lb/10^
Btu).
• The dual mechanical collection system had an average particulate re-
moval efficiency of 95%.
• PCB materials could not be detected in the stack gases.
• Likewise, the presence of any PAH compounds in the air effluent could
not be confirmed.
78
-------
SAM-1A EFFLUENT ANALYSIS
• Assessment of the four effluent streams using the SAM-lA methodology
showed that the secondary collector ash contained the highest degree
of hazard, while the primary collector ash, because of its greater
stream discharge rate, had the largest TUDR. The stack emissions had
both the lowest stream degree of hazard and the lowest TUDR*
• High TUDRs for all three of the ash streams were primarily due to low
MATE concentrations and relatively high measured concentrations of
phosphorus and several trace metals* The calculated degree of hazard
and TUDR values for these streams may give an unrealistically high ap-
praisal of their actual hazard potential, since the physical form of
the effluents and their disposal method is not fully considered under
the SAM-lA approach. For solid waste streams, other factors, such as
teachability, should be considered to determine which pollutants have
the highest potential of being released to the environment.
' 79
-------
TABLE A-10. ANALYTICAL PRECISION OF SELECTED PAH COMPOUNDS
IN ASH EXTRACTS
Run 1 Relative
Run 1 duplicate Average deviation
Compound (U8/§) (y.g/g) ((J.g/g) (± %)
Bottom ash
Phenanthrene 0.56 0.40 0.48 17
Secondary collector ash
Acenaphthylene
Phenanthrene
Fluoranthene
Pyrene
2.2
6.0
1.6
0.90
2.5
3.5
0.62
0.32
2.35
4.75
1.11
0.61
6
26
44
48
POLYCHLORINATED BIPHENYL ANALYSIS
Samples
Bottom ash, primary collector ash, and secondary collector ash samples
were specifically prepared and analyzed for PCB material. Impinger contents
and Florisil from Florisil sampling trains for vaporous PCB materials were also
prepared and analyzed. Samples of wood and fuel oil were not analyzed for PCB
materials due to low levels observed in the effluent materials. One sample of
each ash type was analyzed in duplicate.
Sample Preparation
Approximately 20 g of each ash sample was mixed with preextracted anhy-
drous sodium sulfate to form a free-flowing mixture prior to extraction with
150 ml of hexane in Soxhlet extractors for 8 hr. The hexane extracts were con-
centrated to approximately 5 ml in Kuderna-Danish evaporators, -transferred,
and the volumes adjusted to 1 ml. Florisil train extracts were aliquoted into
two equal volume quantities for PCB and PAH analysis. All ash extracts and one
aliquot of each Florisil train extract were cleaned by shaking with 5 ml of
concentrated sulfuric acid.
106
-------
Sample Analysis
All extracts were screened on a Varian 1400 GC using a scandium tritide
electron capture detector. The glass column (1.8 m x 2 mm ID), packed with 37,
SP-2401 on 100/120 mesh Supelcoport, was operated isothermally at 190°C with a
nitrogen flow of 30 ml/min« The extracts were screened against approximately
1 ng/jj.1 standards of Arochlor 1254 and 1260. Samples producing no responses
corresponding to any single component in either Arochlor standard at approxi-
mately 1 ng/jj.1 concentrations were eliminated from additional analyses. Samples
which produced responses which corresponded to Arochlor components or which
produced significantly higher responses than the 1 ng/^,1 Arochlor standard were
submitted for subsequent GC/MS analysis.
GC/MS was used to confirm the presence of PCB residues in portions of se-
lected extracts. The GC/MS system consisted of a Varian MAT 311-A MS inter-
faced with a Varian 3700 GC via a two-stage Watson-Bieman separator. The MS,
controlled by a Varian 620/i MS data system, was sequentially focused on m/e
222, 224, 256, 258, 290, 292, 324, and 326 for di-, tri-, tetra-, and penta-
chlorobiphenyls. The GC column, as described for screening, was held isother-
mally at 180°C and was eluted with helium at a flow rate of 30 ml/min. PCB ma-
terials were identified by coincident peaks in the two plots for a particular
chlorobiphenyl with the characteristic relative intensities. The sensitivity
of the system was established by assaying an Arochlor 1254 standard.
Quality Assurance
Quantitative analysis of PCB was made using pattern recognition and re-
sponse measurement as outlined in "Methods for Determining the PCB Emissions
from Incineration and Capacitor and Transformer Filling Plants." The results
for EC/GC analysis of selected fortified samples are presented in Table A-ll.
Good recoveries (96 to 130%) of Arochlor 1254 from heated solvent, Florisil,
and bottom ash were found. Recoveries of Arochlor 1254 components from forti-
fied primary collector ash were found to be highly variable, ranging from 9 to
91%* This observation suggests the selective sorption of specific PCB compounds
and is indicative of a significantly different sample matrix than previously
encountered* The results of GC/MS analysis of Florisil train impinger water,
primary collector ash, and secondary collector ash fortified with Arochlor 1254
are presented in Table A-12* Good recoveries ( 10070) of the Arochlor fortifi-
cation were observed for secondary collector ash; however, recoveries of both
low and high level fortifications from the primary collector ash were found to
be highly variable and similar to those observed in the low level EC/GC forti-
fication* Since compound stability and preparative losses do not appear to be
serious at low levels ( 0.8 ^g of Arochlor 1254), losses at high fortification
levels strongly implicates selective compound sorption* Since all sample re-
sponses were at or below the instrumental detection limits, precision calcula-
tions were not possible*
107
-------
TABLE A-11. RECOVERY OF AROCHLOR 1254 FORTIFICATIONS FROM SELECTED
SAMPLES ANALYZED BY EC/GC2/
Component
number
1
2
3
4
5
6
7
8
Average
recovery
Fortified
reagent
103
91
97
95
94
97
98
96
96 + 3
Florisil (%)
108
112
100
117
139
152
154
155
130 + 23
Bottom ash (%)
98
80
67
76
118
122
124
122
101 + 24
Primary
collector ash
0
91
80
29
0
58
9
30.
-b/
a/ Determined for a fortification of 0.8 )j,g of Arochlor 1250 in all cases.
b/ An average recovery of 37% was found for a recovery range of 0 to 9170.
108
-------
TABLE A-12. QUALITY ASSURANCE DATA FOR GG/MS ANALYSIS OF PGB FORTIFIED SAMPLES
o
vo
Component
Dlchlorobiphenyl
Peak 1
2
3
4
Trlchloroblphenyls
1
2
3
4
5
6
Tel: racliloroblplicnyls
1
2
3
4
5
6
7
Pentachloroblplicnyls
1
2
3
4
5
6
Fortified
aolvcnt.i'
100
100
100
100
89
97
95
96
96
111
96
94
120
101
130
96
96
118
103
94
123
107
113
Fortified
implnger
INT
INT
75
100
95
103
105
96
144
156
180
135
184
165
174
153
193
221
194
141
524
302
300
Primary
5 ng/gi'
39
119
63
0
61
47
32
6
0
0
20
0
0
7
65
42
24
118
56
50
16
0
0
collector
ash
50 |ig/g£/
76
55
0
0
72
66
58
20
0
0
76
65
18
0
91
65
41
82
69
56
18
0
0
Secondary
a
5 |ig/g
-------
The recovery of an Arochlor 1254 fortification after extraction from a
blank Florisil train impinger water and rinse was found to be approximately
100% or greater as summarized in Table A-12. The elevated recoveries have been
attributed to background contamination of this sample by Arochlor 1254* Flori-
sil and reagent blanks were analyzed in all phases of analysis to identify
potential contamination sources*
Problems
Extracts of bottom ash and secondary collector ash were found to produce
complex chromatograins by EC/GC which necessitated GC/MS analysis for identifi-
cation and verification. The variable, selective recoveries of PCS fortifica-
tions, and the relatively simple chromatograms from primary collector ash are
indicative of a significantly different sample matrix which suggests the pres-
ence of a strong sorbent material.
A single sample of impinger water from the second run of the Florisil
train produced patterns corresponding to Arochlor 1254. Considering the lack
of detectable PCB in the remaining train components, location of the Impinger
in the train, and the direct correspondence of sample and standard responses,
the sample responses were discounted as low level contamination, most likely
originating in the concentration step.
Reference
Haile, C. L., and E. Baladi. Methods for Determining the Total Polychlorinated
Biphenyl Emissions from Incineration and Capacitor - and Transformer-Filling
Plants. EPA-600/4-77-048, November 1977.
LEVEL 1 ANALYSIS METHODS
Samples
Level 1 environmental assessment methods for particulate-containing gases
were included in the test program. Samples of the outlet gas stream were col-
lected using the SASS train during the second test day. Organic analyses were
performed on all SASS train samples including particulates, XAD-2 resin, aque-
ous condensate, train rinses, and reagent blanks. Inorganic analyses were per-
formed only on the aqueous condensate and the hydrogen peroxide impinger so-
lutions. These analyses were performed according to the method contained in
"IERL-RTP Procedures Manual, Level 1 Environmental Assessment" (EPA-600/2-76-
160), and included any revisions available at the time of sample analysis.
Only selected inorganic gas analyses were performed on site, and no organic
GC analyses for C-l through C-6 compounds were performed on site. Continuous
on-site analyses for the following gases were performed: 02» NOX, S02, hydro-
carbons, C02, and CO.
110
-------
The organic Level 1 analysis protocol is unique to Level 1 and is not
covered in other sections of this appendix* Therefore, only organic Level 1
analyses will be discussed in this section.
Sample Preparation
Sample preparation methods used were those defined in EPA-600/2-76-160a,
a draft revision of Chapter 8 received November 22, 1977, and subsequent ap-
plicable revisions, which were received May 31, 1978. Samples were extracted
with methylene chloride, aliquoted, and concentrated to a volume of 10 ml using
Kuderna-Danish evaporators. A sample aliquot which contained an amount of TOO
material between 10 and 100 mg was either evaporated to dryness or solvent-
exchanged depending on its TOO level* LC fractionation on silica gel was per-
formed on this sample aliquot and seven fractions collected for subsequent
analysis* The above sequence of sample preparation steps, and decision making
criteria, are schematically shown in Figure A-6. A summary of samples prepared
and the analyses subsequently performed on them are presented in Table A-13*
Particulate samples were combined for extraction of organic material* The
combined particulate samples were subsequently extracted with methylene chlo-
ride* The results of particulate analyses are reported as "total particulate"
values due to the combination of particulates*
XAD-2 resin was prepared before use according to the procedure in the
draft revision of Chapter 8 of the Level 1 protocol. The XAD-2 samples were
Soxhlet-extracted using only methylene chloride, after testing, as specified
in the draft revisions* Aqueous condensate was adjusted to pH 2 and 12, re-
spectively, .and extracted with methylene chloride at each pH* The extracts
were then combined before analysis*
Since the particulate weight contained in the front half rinse accounted
for greater than 10% of the total particulate collected after weighing, the
rinse particulates were Soxhlet-extracted with methylene chloride for organic
analysis*
Sample Analysis
Organic Level 1 methods include gravimetric measurement of low volatility
components (GRAV), IR analysis, determination of TCO, and mass spectral (direct
inlet and batch inlet) procedures (LRMS). Analytical methods for each procedure
are discussed below.
GRAV determinations were performed on all unconcentrated samples and LC
fractions by evaporation of an aliquot (5 ml) of methylene chloride extract or
LC solvent fraction. GRAV determinations of concentrated samples were made by
evaporation of 1-ml aliquots.
Ill
-------
•lm.li LC fixil
Figure A-6. Organic Level 1 analysis flow diagram.
-------
TABLE A-13. LEVEL 1 ORGANIC SAMPLE SUMMARY
Preliminary analysis
Sample prepared TCO GRAV IR
Participate
Front rinse
XAD-2 field sample
Aqueous condensatc
Organic module
rinse
Mctliylene chloride
> blank
H« ' XAD-2 blank
£ Kilter blank
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Co
TCO
X
X
X
X
X
X
X
X
icentrate analysis
GRAV
X
X
X
X
X
X
X
X
IR LRHS^' t
X
X
X X
X
X
X
X X
X
Analysis Solvent
cnnlnatcd exchange
X
X
X
X
X
X
X
X
Analysis of all
LC fXj fractions Analysis
tract lonat ion TCO GRAV IR terminated
...
_ . _
X X X ^ X
. . -
-
_
X X X - X
-
ill Direct Inlet LRHS only.
_b/ IR spectra were taken only of samples producing measurable gravimetric weight, LCC and I.CF.
-------
IR spectra were taken from the GRAY residues* The samples were prepared
by dissolving the residue in a minimum amount of methyl chloride and trans-
ferring the solution to KBr plates. The methylene chloride was then evaporated
and the IR spectra taken. GRAY samples resulting in no measurable weight gain
were not submitted for IR analysis. A Beckman IR-12 spectrophotometer was used
to obtain the IR spectra.
TCO analyses were performed using a Varian 2400 GC with flame ionization
detector. A glass column (1.83 m x 2 mm ID), packed with 1.5% OV-101 on Gas
Chrom Q, 100/120 mesh, was used. The temperature conditions were: 4 min initial
hold at 50°C; temperature programmed from 50 to 250°C at 20°C/min; inlet tem-
perature, 200°C; and detector temperature, 275°C. Chromatograms were recorded
on an HP Model 3380A recording integrator. Sample volumes of 5 |j,l were used
for TCO analyses.
Direct inlet LRMS of the XAD-2 field sample and XAD-2 field blank was per-
formed. MS analyses were performed using a Varian MAT CH-4 with a Varian 620/i
data system. Thermograms from 50 to 460°G at approximately 50°C/min were per-
formed on the direct inlet samples. The MS operating conditions are listed in
Table A-14.
TABLE A-14. LRMS OPERATING CONDITIONS
Ionizing voltage 70 eV
Ionizing current 40 y,a
Multiplier voltage 2.3 kV
Mass range 0-500 m/e
Scan rate 5 sec/mass
decode
IR spectra obtained from concentrated XAD-2 resin and XAD-2 LC Fractions
6 and 7 consisted primarily of absorbance peaks characteristic of carbonyl-
containing hydrocarbon materials. The IR spectrum observed for the XAD-2 LC
Fraction 6 is illustrated in Figure A-7. The major absorption peaks are gen-
erally characteristic of aliphatic or aromatic carbonyl-containing compounds.
The low prevalence, or absence, of materials containing hydroxyl and/or car-
boxylic acid functional groups is suggested by the absence of intense broad
band absorption in the region of 3000 to 3400 cm"*- for LC Fractions 6 and 7.
These functional group assignments were typical of the IR data for samples
producing interpretable IR spectra. Compound identification beyond the func-
tional group assignments was not attempted due to the large number of peaks
114
-------
ijl
J L.
1100 MOO nra ?«oo 1400 mo noo ira
tno ino
WAVINUMCII Ctf '
1400 l»0 1400
IIOO MOO 1000 «00 400 700
Figure A-7. IR spectrum for LC Fraction 6 of XAD-2 resin.
-------
APPENDIX D
DESCRIPTION OF SASS EQUIPMENT AND ANALYSIS
The SASS is a method prescribed by EPA for environmental assessment work*
It is somewhat similar to, but more complex than, the EPA Method 5 train* Basic
components of the SASS train are shown in Figure D-l* It is different from the
Method 5 train in that the filter is preceded by three cyclones and the filter
is followed by an XAD-2 cartridge for collection of organics* Also, the imping-
ers contained special solutions for collection of vaporous metals (e.g., Hg).
Analysis of the SASS train components, as prescribed by EPA's Level 1 en-
vironmental assessment procedure, is quite complex* A tabular presentation of
the required analysis is shown in Table D-l* This matrix is in accordance with
a Level 1 assessment, with the exception of morphology and bioassay* However,
for the purposes of this test program, MRI has added certain other analyses as
indicated in Table D-l.
142
-------
I.C.
u>
Hilt.
lloldo
l'..4.0 lui^Mti.iluia l.C.
ssSs
(f~^-j xx's^^iTSSSv^!
V_»- — ._^_ — _-_- _j_; • ••
tfu», !yuc_l
»«.. lypc
Pilal
Cycloiwi
O.CM
i.e. xx
ftp
U<|uU PtUM<
Out Puiji^
Go* Cualaf - *
S-—,n
-l^v
XAD-J C(^tuitr
Canlfollttr
.l cunlalm 7SOml of
.2 and 3 «acit cuntuiia */5
l 0.2M (NIU);.ljOfl t 0.2M««NO]
Figure D-l. Schematic of source assessment sampling system.
-------
TABLE D-l. ANALYSIS MATRIX FOR SASS TRAIN COMPONENTS
Particulate
Probe and XAD-2
cyclone rinse 10 \ua 3 \t,m 1 pirn Filter resin
EPA Level 1 protocol
SSMS
Hg by atomic
absorption
CH2C12 extraction Combined
Gravimetric X XX
IR analysis X X
LRMS analysis X X
LC (7 fractions) X X
Additional analyses selected by MRI
PCS material X X
PAH compounds X X
Aqueous Imp Inge r
condensate solutions
X X2/
X X
X
X
X
a/ Hydrogen peroxide impinger only.
-------
APPENDIX E
DATA TABLES FOR SAM-lA
TABLE E-l. SAM-lA SUMMARY SHEET
farm 1*01
i. SOURCE AND APPLICABLE CONTROL OPTIONS
BURLINGTON ELECTRIC DEPARTMENT WOOD AND OIL FIRED BOILER
2. PROCESS THROUGHPUT OR CAPACITY
J.Q
(100,000 Ib sceam par hour)
3. USE THIS SPACE TO SKETCH A BLOCK DIAGRAM OF THE SOURCE AND CONTROL ITEMS SHOWING ALL EFFLUENT
STREAMS. INDICATE EACH STREAM WITH A CIRCLED NUMBER USING 101-199 FOR GASEOUS STREAMS.
201-299 FOR LIQUID STREAMS, AND 301-399 FOR SOUO WASTE STREAMS A
PRIMARY SECONDARY ft *™£IQM
11 \
FUEL
OIL -
WOOD-
CHIPS
BOILER
MECHANICAL
COLLECTORS
/• v
^,
' — V
\}
L
- — ^
BOTTOM ASH
COLLECTOR ASHES
4. LIST AND DESCRIBE GASEOUS EFFLUENT STREAMS USING RELEVANT NUMBERS FROM STEP 3.
ns (after mechanical collectors)
102
103
104
105
106
107
5. LIST AND DESCRIBE LIQUID EFFLUENT STREAMS USING RELEVANT NUMBERS "ROM STEP 3.
201
202
203
204
205
206
6. UST
301
302
303
304
30!
306
AND DESCRIBE SOLID WASTE EFRUENT STREAMS USING RELEVANT NUMBERS FROM 3TE? 3.
Bottom ash
Primary collector ash
Secondary collector ash
7. IF YOU ARE PERFORMING A LEV& 1 ASSESSMENT. COMPLETE THE IA02-LEVEL 1 FORM .-CR EACH Er-'.UENT
STREAM LISTED ABOVE. IF YOU ARE PERFORMING A LEVEL 2 ASSESSMENT COMPLETE THE IA02-L£VEL 2 fQS
FOR EACH AFFLUENT STREAM LISTED ABOVE.
145
-------
TABLE E-l. (continued)
S LIST SUMS
FROM LINE 7. FORMS IA02. IN TABLE BELOW
DEGREE OF HAZARD AND TOXIC UNIT DISCHARGE RATES BY EFFLUENT STREAM
GASEOUS
STftCAM
coot
101
A
CtGHU Of
x€Ai.rn
-
1.3E2
tax
3ASCO
-
1.6E1
i
3
C
TOXIC UNIT
OOOtUKSt RATH
nuirw
BASIC
(mv
3.8 E3
0
CCOL
i»«0
inA- 3.3^E2
inp . ft r-m n inn-
inr 2.4LE5 r roi r» !O(~ 3.47E7
UENT STREAMS
I
na
uc_
0
3
'.2. LIST POLLUTANT SPECIES KNOWN OR SUSPECTED TO 3E cMimo .-OR WHICH A .MATE is NOT AVAILABLE.
146
-------
SAM/IA WORKSHEET FOR LEVEL 1
TABLE E-2. BOTTOM ASH WORKSHEET
Form IA02 Level 1
1 SOURCE /CON IROL OPTION
Burlington Elecirlc Wood and Oil Klred Boiler
2 ft'FLUEN! STREAM
301 bo c com Ash
CUOf I NAUt
Page 1 / 3
3 EFRUENI STREAM FLOW RATE
Q » 25.2 |>/sec
(tMS ' in'/tcc — liquid • I/tec — tolul «»sle * (/tec)
4 COMPLETE THE FOLLOWING TABLE FOIt THE EFFLUENT SI HE AM OF LINE 2 (USE BACK Of FORM FOIt SCRATCH WORK)
A
SAMTli IHACIUN
UNI1S
SSHS
27/L1
28/Na
29/K
30/Rb
31/Ce
32/Bts
33 /Mg
34/Ca
35/Sr
36/Bu
37/B
38/Al
39/Uu
B
I'UNU N
KAIUN
Hg/g
590
>1.000
> 1,000
96
5
7
>1.000
>1,000
>800
Vl.OOO
72
>1,000
48
C
III Al III
MAII
IHAIlUN
Hg/g
0.70
1.63E3
N
3.6E3
2.5E3
0.06
1.8E2
4.8E2
9.2E1
1.0E1
9.3EI
1.6E2
1.5E2
1)
ICUKXilCAl
MAII
CONCIN
IRAIIUN
HB/S
0.75
N
N
N
N
0.11
1.7E2
e.2El
N
5.0
5.0E1
2.0
N
E
Oil Jill (X
IIA/AHO
IHCAI IHI
lll'Cl
-
843
>0.6
-
0.03
0.002
117
>5.6
>2.1
>8.7
>10
0.8
>6.3
0.3
F
OHIHNAI
III Al III UAIt
lAUlf
-
G
HA/AUU
llCOlOCCAL}
IH'Dl
-
787
-
-
-
-
64
>5.9
>31
-
>500
1.4
>500
-
H
UHLMNAI
ICO. UAU
lAUlf
-
1
nCAlIN
UAIt
flCIIMO
-
y
/
y
y
/
/
y
i
v'u
rcoi
UAU
OCflMO
-
/
\/
*/
l/
\/
/
/
/
K
IO>lC IIMII U
IKIAIIII
BAUD)
(f • UN[ J)
g/sec
2.12t!4
>1.51E1
-
0.76
0.05
2.95E3
>1.41E2
>5.29E1
>2.19E2
>2.52E2
2.02E1
V1.59E2
7.6
L
y.lAW.I UAU'
1C > IIN1 Jl
g/sec
1 .98E4
.
-
-
-
1.61E3
>1.49E2
>7.81E2
-
>1.26E4
3.53E1
>1.26E4
-
II MIWI &TACC IS NIIIUU. MSt A COIIIINIIAIIIMJ Slllll
'j tmuENI SIHLAM IU GKt t 11
IIEALIII MA If UAUO a *
COI G) bb J
I GIIM IAUI)
.016E6
6 NUMUEt
COMPAR
IUALIH 64
LCOLOGlCX
OF ENDUES 7 K
ED TO MATES H
Et
tl 6I> (E
tXlC UNIT DISCHARGE SUM
EALTH MATE BASED It COL K)
OLOT.ICAL MATE BASTb (i. COL
NIER HERE AND AT LINE 8. FO
1.097E5
1} 7h 2.559E7
RM IAOI)
-------
TABLE E-2. (continued)
oo
CONTINUATION SHEET fOR ITEM NO. 4. fORM IA02. LEVEL 1 Pjgc 2 ' 3
SOURCE /CON 1 001 OP1ION Jlurltngtmi CffUJENI STREAM NO . 30_1
A
SAMTU IB AC ltd
IINIIS
40/In
41/T1
«/C
43/Si
WGe
45/Sn
46/Pb
48/Pb
49/As
50/Sb
51/Bi
53/S
54/Se
55/Te
56/F
57/C1
58/Br
59/1
60/Sc
61/Y
62/T1
63/Zr
64/Hf
65/V
| 66/Nb
U
IDACIION
CONCIM
IBAIOM
UK/K
Internal
<0.7
•>l,OQO
It
2
5
>l,000
15
2
<0.2
>l,000
2
<0,2
156
35
2
<0.2
48
94
>1,000
265
3
250
47
C
HI ALII*
MAU
court n
IAAIIOH
itK/e
Standarc
3.0
Led
3.0E2
1.7E1
N
0.50
3.0E1
0.50
1.5E1
1.2E1
N
0.10
3.0
7.5E1
2.6E3
N
N
1.6E3
3.0E1
1.8E2
1.5EI
1.5
5.0
6.5E2
0
ECOlOCICAl
MAIC
tOHC[N
TRAIIUM
I1E/R
N
N
N
N
0.10
0.001
0.10
0.40
N
N
0.05
N
N
N
N
N
N
N
N
N
N
0.30
N
E
IHCHEI or
MA/ADI)
(IKAIlin
(H/CI
—
-
<0.2
-
>3.3
0.2
-
10
>33
30
0.1
<0.02
-
20
<0.07
2.1
0.01
-
-
0.03
3.1
>5.6
18
2
50
0.07
r
UKOWAl
mSIIION IN
IICAUX MAII
IAIIII
—
G
Monti or
MA/ARU
(tCOlOGICAII
ID/0)
-
-
-
-
•-
-
50
>1.0E6
15
5
-
-
40
-
-
-
-
-
-
-
-
-
-
833
-
H
ORDINAL
POSMION IN
icoi UAir
1*011
—
1
%/•
iirAUM
UAII
tucttotu
—
/
/
/
/
/
/
/
/
/
/
/
J
%/ll
[COl
UAIC
KCIIOCO
—
/
/
/
/
/
•
K
L
tOIIC Uinl DISCHARGt RAIt
lllfAtIM
BAsro*
II • UNI »
g/sec
-
15.0
-
>8.32E1
5.0
-
2.52E2
>fl.32E2
7.56E2
2.5
<0.5
-
5.04E2
<1.8
5.29E1
0.25
-
-
0.76
7.81EF
>1.41E2
4.54Er
50
1.26E3
1.8
UCOtOr.ir.Ai
OASID)
1C > l»« 1)
g/sec
-
-
-
-
-
-
1.26E3
>2.52E7
3.78E2
1.26E2
-
-
1 .OR3
-
-
-
-
-
-
-
-
•
-
2.10E4
-
-------
TABLE E-2. (continued)
CONTINUATION SHEET FOR ITEM NO 4. FORM IA02. LEVEL I
Page 3/3
<;oiinrF/roNiROi OPTION Burlington EFFIUENI STRF.AM NO JQJ
A
wurti FRACIION
IINIIS
67/Ta
68/Cr
69/Ho
70/W
71/Hn
72/Fe
73/Ru
74/Co
75/Rh
76/N1
77/Pt
78/Cu
79/Ag
80/Au
8l/Zn
82/Cd
83/llg
84/La
B
FflACIIOM
CONTIN
IHAIION
llg/S
<0.2
310
9
2
>l,000
> 1,000
<0.2
34
<0.2
80
<0.2
85
0.3
<0.2
74
<0.4
0.09
57
C
IKAllll
UA1C
CONCIN
IRAIinN
|lg/g
1.5E2
0.50
I .5F.2
3.0EI
0.50
3.0
N
1.5
0.03
0.45
0.06
1.0121
0.50
N
5.0E1
0.10
0.02
3.4E3
I)
[COIOGICAI
MA II
CONCtN
in/unw
|ig/g
N
0.50
1.4E1
N
0.20
0.50
N
0.50
N
0.02
N
0.10
0.10
N
0.20
0.002
0.50
N
E
otoiict or
MARAUD
(1ICA11HI
(0/C)
—
<0.001
620
0.06
0.07
>2,000
>333
-
23
<6.7
178
<3.3
8.5
0.6
-
1.5
<4
4.5
0.002
F
(XIOIHAl
IWIIION IN
IKAllll MAK
IAOII
—
G
wr.Htf or
HAIARO
((CUIOGICAI |
(B/0>
-
620
0.6
-
>5,000
>2,000
-
68
-
4,000
-
850
3
-
37
200
0.2
-
H
ORIXNAt
POiMION IN
IC(X MAK
1*611
1
Ju
IKAIIK
MAII
KClFOtO
I/
/
/
/
/
/
/
/
v/
/
/
J
S/»
cent
UAH
tiacnio
—
/
/
/
/
/
v/
v/
/
/
K
I
ICItC UWI OlSCIlAROt DAK
IHU11H
BASED)
5.04E4
>8.39E3
-
5.80E2
<1 .69E2
4./.9E3
<8.32E1
2.14E2
15
-
3.78E1
<1.01E2
1.13E2
0.05
UCOIOGICAl
o«s(ui
in • UNI it
g/sec
-
1.56E4
15
-
>1.26E5
>5.04E4
-
1.71R3
-
1.01E5
-
2.14E4
7.56E1
-
9.32E2
5.04E3
5.0
-
-------
TABLE E-3. PRIMARY COLLECTOR ASH WORKSHEET
SAM/IA WORKSHEET FOR LEVEL 1
Form IA02 Level 1
1 SOURCE /CONTROL OfTION
Burlington ElecCrlc Wood and Oil Fired Boiler
2 EFFLUENT STHEAM
302 Primary Collector Ash
COUt • NAUI
fit' 1 / 3
3 EFFLUENT SIKEAM FLOW RATE
Q „ 55.8 g/sec
(gj» • m'/sec — liquid • l/lec — wlitl w*tle " g/>ec)
4 COMPIE1E HIE FOLLOWING IAUI.E FOR THE EFFLUENT STREAM Of LINE 2 (USE BACK OF FORM FOR SCRATCH WORK)
A
&AWII IHAUIUN
UNI IS
SSMS
27/L1
28/Nu
29/K
30/Rb
31/Ce
32/Be
33/Mg
34/Ca
35/Sr
36/Bu
37/B
38/At
39/Ga
0
IHACIII1N
cawriN
IMAllUN
ng/g
41
>1,000
> 1,000
28
1
3
> 1,000
>1,000
433
>653
52
vt.OOO
24
C
IUAI IM
UAII
IHAIUIN
Hg/g
0.70
I.6E3
N
3.6E3
2.5E3
0.06
1 .8K2
4.8E2
9.2E1
1.0E1
9.3E1
1.6E2
1.5E2
0
UAII
i:ONLfN
IHAIlUN
l'g/8
0.75
N
N
N
N
0.11
1.7E2
3.2E1
N
5.0
5.01U
2.0
N
E
01C.IKI in
IIA;«HO
IHlAllll)
tb'Cl
-
59
>0.6
-
0.008
0.0004
50
>5.6
>2.1
4.7
>65
0.56
>6.3
0.16
F
UNIHNAl
HJilllUX IN
IUAI III UAII
IAIUI
--
G
LXI.kll IK
IIA/Alll)
-
55
-
-
-
-
27
>5.9
>31
-
>131
1.0
>500
-
H
CMJ4NAI
K1SIIIUN IN
(rot UAK
1AIUI
-
1
MAU
uuiotu
-
/
y
/
y
J
\/
\/
1
J«
ecu
MAIi
HCllCHO
-
J
/
/
/
/
/
K
L
IO>lC Utlll DI^CllAHCf DAK
OUAllH
luscoi
ic • mil ii
g/sec
3.29E3
>3.3
-
0.45
0.02
2.79E3
3.12E2
1.17E2
2.62E2
3.67£3
31.2
3.52K2
8.9
KCOiOOCAi
1C > ilN( 1)
g/sec
3.07E3
-
-
-
-
1.51E3
3.29E2
1.73E3
-
7.31E3
55.8
2.79E4
-
U UOHI SI-ACl li till IX U Ubt A CUtlllNUAIum SlUll
b iniUEtll SIHEAM OfGHIt (J
HEALTH MATE OASfD (> COI
ECOlOTilCAL MA1E LlASEO (1
(INIEH HIKE AND Al LINE b.
1 HA2AKD
E, t,a_.2_.°65
COL. G)hb 1.019E6
KlhM IAOI)
6 NUMUEK OF EN1HIES ' TOXIC UNIT DISCHARGE SUM
COMPARED TO MATES
HEAL III 6d _ ECOLOGICAt MATE FJASEO (i COL
LCOLOCICAI 6b._. _ (ENUH HERE AND Al LINE 8. FOI
1.154E5
1)Jh 5.689E7
)M IAOI)
-------
TABLE E-3. (continued)
CONTINUATION SHU! I OR ITfM NO. 4. FORM IA02. IEVU 1
Page
2/ 3
sniiiirr/rnNinni oriinN Burlington CFHUINT SCREAM NO Jfil
A
SAuru rRAciioN
UNIIS
40/ln
41/T1
42/C
43/Si
tttt/Ge
45/Sn
46/Pb
48/P
49/As
50/Sb
51/Bi
53/S
54/Se
55 /Te
56/F
57/Cl
58/flr
59/1
60/Sc
61/Y
62 /Ti
63/Zr
64/HE
65/V
66/Nb
B
IB AC IIOH
COMCI N
IkAImN
»»g/g
Internal
<0.2
Not Repoi
>1,000
5
1.2
7
>1,000
72
2
<0.2
>1,000
9
<0.2
473
108
20
2
12
26
" >1,000
79
<0.7
143
11
C
m Aim
UAH
CONCCN
1RAIKIN
V-e/R
Standard
3.0
ted
3.0E2
1.7R1
N
0.50
3.0E1
0.50
1.5E1
1.2E1
N
0.10
3.0
7.5EI
2.6E3
N
N
1.6E3
3.0E1
1.8E2
1.5E1
1.5
5.0
6.5E2
0
JCOl.OGtf.Al
MAII
CONCCN
IRAIK1N
ll8/8
N
N
N
N
0.10
0.001
0.10
0.40
N
N
0.05
N
N
0.01
N
N
N
N
N
N
N
0.30
N
C
WGflEt Of
HA/ANU
(IIIAIIII)
(0/CI
—
<0.07
>3.3
0.29
-
14
>33
144
0.13
<0.02
-
90
<0.07
6.3
0.04
-
-
0.008
0.87
>5.6
5.3
<0.47
29
0.02
r
omuNAi
rnsiiiON IH
HCAIIIIUAII
IA8LI
—
G
mam or
HAZARD
([cniociCAii
IB/O)
—
-
-
•-
-
70
>1.0E6
72
5
-
-
180
-
-
1.1E4
-
-
-
-
-
-
-
477
-
H
OROINAl
rosiriONiN
ICCR MAIf
lAStt
1
\/lf
i if Aim
MAIC
ocaofo
—
/
/
/
/
/
/
/
/
/
J
V'.r
ecu
MAII
cucrinfo
—
/
/
/
/
/
/
K
I
IO«IC UNII OISCIIARGC RAIC
(IK»UH
BASfOI
1C > IINI 1)
g/sec
<3.9
>l.84E2
16.2
-
7.81E2
>1.84E3
8 .04F.3
7.3
<1.1
-
5.02E3
<3.9
3.52E2
2.2
-
-
0.45
48.5
>3.13E2
2.96E2
<26.2
1.62E3
1.1
nccM(yjc«i
OAStD)
1C • IINI ])
g/ sec
-
-
-
-
3.91E3
>5.58E7
4.02E3
2.79E2
-
-
1 .OE'i
- •
-
6.14E5
-
-
-
-
-
-
2.66E4
-
------- |