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_
-------
  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
 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
-

-------