United States
 Environmental Protection
 Agency
 Industrial Environmental Resea
 Laboratory
 Research Triangle Park NC 27711
 Research and Development
 EPA-600/S7-81-003d  Oct. 1981
 Project  Summary
 Emissions Assessment of
 Conventional  Stationary
 Combustion  Systems:
 Summary  Report
C. C. Shih and A. M. Takata
  Multimedia emissions from  39
source categories of  conventional
stationary combustion systems  are
characterized in this  study. In  the
assessment process, existing emis-
sions  data were first  examined to
determine the adequacy of the data
base.  This was followed by the
conduct of a measurement program to
fill identified data gaps. Emissions
data obtained from the sampling and
analysis program were combined with
existing emissions data to provide
estimates of emission  levels, and to
define the need for additional data.
  The  results of this study indicate
that conventional stationary combus-
tion systems contribute significantly
to the nationwide emissions burden.
Flue gas emissions of NOX. SOz, and
particulate matter from  the 39 source
categories studied account for approx-
imately 86,  66, and  36 percent,
respectively, of the emissions of these
pollutants from all stationary sources.
Additionally, flue gas  emissions of
sulfates and several trace elements
from coal- and oil-fired combustion
sources also require further attention.
POM compounds in flue gas emissions
are mostly naphthalene, phenanthrene,
and pyrene. However, dibenz(a.h)an-
thracene and possibly benzo(a)pyrene,
both active carcinogens, were detected
at a limited number of coal-fired sites.
 Also, dibenz(a,h)anthracene, and
 possibly benzo(a)pyrene and benzo
 (g,h,i)perylene, another active carcin-
 ogen, were detected at one coal- and
 one  wood-fired underfeed stoker
 tested. The possible presence of
 benzo(a)pyrene in significant amounts
 was indicated in the emissions of two
 other wood-fired boilers.
  A second major source of air emis-
 sions in steam electric plants is vapors
 and drifts from  cooling towers. Air
 emissions of chlorine, magnesium,
 phosphorus, and sulfates from me-
 chanical  draft cooling towers were
 found to be comparable to flue gas
 emissions of these pollutants from oil-
 fired utility boilers.
  Wastewater streams are generated
 from several operations in steam
 electric plants, and in industrial and
 commercial/institutional facilities but
 to a  much lesser extent. Overall,
 concentrations of iron, magnesium,
 manganese, nickel, and phosphorus
 are at levels that may be of environ-
 mental concern. Average organic
 levels ranged from 0.01 mg/l for ash
 pond effluents to 6.0 mg/l for boiler
 blowdown.  Also, no POM was de-
tected in wastewater streams.
  Data on coal fly ash and bottom ash
show that from 11 to 16 trace
elements  are present at potentially
harmful  levels. The  only POMs de-

-------
tected, however, were  naphthalene,
alkyl naphthalenes, and other relatively
nontoxic compounds.
  This Project Summary was devel-
oped by EPA's Industrial Environmen-
tal Research Laboratory,  Research
Triangle Park, NC,  to announce key
findings of the research project that is
fully documented in a separate report
of the same title (see Project Report
ordering information at back).


Introduction
  In  response to the need for a compre-
hensive characterization of  pollutants
from stationary conventional combus-
tion  processes,  EPA's Industrial Envi-
ronmental Research Laboratory at
Research Triangle Park  (IERL-RTP) in
North Carolina established the Conven-
tional Combustion Environmental As-
sessment (CCEA)  Program  as the
primary vehicle for filling identified data
gaps. The component project under
which this study was performed  is
known as the Emissions Assessment of
Conventional Combustion Systems
(EACCS) project, whose objectives are
the:
  •  Compilation and evaluation of all
     available emissions data on pollu-
     tants from  selected stationary
     conventional combustion proc-
     esses.
  •  Acquisition of needed new emis-
     sions data from field tests.
  •  Characterization of air emissions,
     wastewater effluents,  and solid
     wastes generated by selected
     stationary conventional combustion
     processes, utilizing combined data
     from existing sources and field
     tests.
  •  Determination of additional  data
     needs, including specific areas of
     data uncertainty.

  The combustion source types assessed
were selected because of their relevance
to emissions and because they are
among the largest, potentially largest,
and most numerous (in use)  of existing
combustion source types. As shown in
Table 1, 39 source types were selected
for study. Selected source types were
classified into five principal categories.
The results of emissionsassessmentfor
these five combustion source categories
are detailed in the following group/cat-
egory reports:
  Volume  I,  Gas- and Oil-Fired Resi-
  dential Heating Sources (EPA-600/7-
  79-029b; NTIS PB  298494).
 Volume II, Internal Combustion Sources
 (EPA-600/7-79-029c; NTIS PB 296390).
 Volume  III,  External  Combustion
 Sources for  Electricity Generation
 (EPA-600/7-81 -003a;  NTIS  PB 81-
 145195).
 Volume IV, Commercial/Institutional
 Combustion  Sources (EPA-600/7-
 81 -003b; NTIS  PB 81-145187).
 Volume V,  Industrial Combustion
 Sources (EPA-600/7-81-003c; NTIS
 PB 81-225559).
 The highlights of these group/cate-
gory reports are documented in this
report summary.

Assessment Methodology
 The assessment method employed in
the project involved a critical examina-
tion of existing emissions data, followed
by  a measurement program to fill data
gaps based on a phased sampling and
analysis strategy. Data acquired as
result of the measurement program,
combination  with  the existing dai
were further evaluated. Data inadequ
cies identified at the completion of tl
project are discussed with respect tot!
need for additional  study.
  Specifically, the phased approach
environmental  assessment providi
comprehensive  emissions informatic
on all process waste streams in a co
effective manner. To achieve this goc
two distinct samplings and analyses ai
employed. Level I utilizes semiquantiti
tive (+  a factor of 3) techniques >
sample collection and  laboratory  ar
field analysis: 1) to provide preliminai
emissions data  for  waste  streams  ar
pollutants not adequately characterizei
2) to  identify potential problem area1.
and 3) to prioritize  waste  streams  an
pollutants in those  streams for furthe
Table 1.     Combustion Systems Considered in the Study
Combustion
Source
Type
External Combustion
Coal
Bituminous
Pulverized dry
Pulverized wet
Cyclone
All stokers
Anthracite
Pulverized dry
All stokers
Lignite
Pulverized dry
Cyclone
All stokers
Petroleum
Residual oil
Tangential firing
All other
Distillate oil
Tangential firing
All other
Gas
Tangential firing
All other
Wood
Stoker
Internal Combustion
Distillate Oil
Gas turbine
Reciprocating engine
Gas
Gas turbine
Reciprocating engine

Electricity
Generation
X
X
X
X
X
X
X
X
X
X
X

X
X

X
X
X
X
user
Industrial
X
X
X


X
X
X
X
X
X
X
X
oecfor
Commercial/
Institutional Residential
X
X
X

X
X X
X X
X
X
X

-------
more  quantitative testing. Using the
information from Level  I,  available
resources can be directed toward Level
II testing which involves specific quanti-
tative  analysis  of components of those
streams that contain  significant pollu-
tant levels. The data developed at Level
II are used to identify control technology
needs  and  to  further define environ-
mental hazards associated with emis-
sions.

Existing Emissions Data  Base
  A major task in the project  was the
identification of gaps and inadequacies
in  the data base. Decisions as to the
adequacy of the data base were made
using criteria developed by considering
both the reliability and variability of the
data.  Estimated environmental risks
associated with the emission of each
pollutant were also considered in
determining the need for, and extent of,
the sampling and analysis program.


Gas-  and Oil-Fired Residential
Heating Sources
  The  sources of emissions  data for
residential gas- and  oil-fired  systems
are  limited to early data used  to
generate EPA emission  factors and
more  recent data developed  by EPA
contractors for criteria pollutants. For
gas-fired systems, the data  base  for
SO2, N0», total hydrocarbons, and CO
emissions is adequate. However,  the
data base for  paniculate and  organic
emissions is inadequate. For  oil-fired
systems, the emissions data  base for
particulate, S02, NOX, HC, and CO is
adequate,  but inadequate for SOs,
particulate sulfate, trace element, and
organic emissions.

Internal Combustion Sources
  The evaluation of emissions data for
electricity generation and industrial
internal combustion sources indicates
that the emissions data base is adequate
for gas-fueled turbines and reciprocating
engines. For distillate-oil-fueled  gas
turbines, the existing data base for NOX,
total  hydrocarbons,  CO, particulate,
S02, and SOs  emissions  is adequate.
However, the data  base  for trace
elements and specific organic emissions
is  inadequate.  For distillate-oil-fueled
reciprocating engines, the data base for
NOX, total hydrocarbons, CO, and S02
emissions is adequate. The data base
for particulates, S03, trace elements,
and  specific organic emissions was
found to be inadequate.
External Combustion Sources
for Electricity Generation
  For flue gas emissions, the status of
the data base can be summarized as
follows:
  • The data base for criteria pollutants
    is generally adequate.
  • For SOs emissions, the data base is
    adequate for bitummous-coal-fired
    boilers, residual-oil-fired boilers,
    and gas-fired boilers, and inade-
    quate for lignite-fired boilers. For
    emissions of primary sulfates, the
    data base is adequate for pulverized
    bituminous dry- and  wet-bottom
    boilers, residual-oil-fired boilers,
    and gas-fired boilers, and inade-
    quate for other combustion source
    categories.
  • For  emissions of particulates by
    size fraction and  trace elements,
    the data base is adequate for gas-
    fired boilers and inadequate for all
    other combustion  source cate-
    gories.
  • For emissions of specific organics
    and polycyclic  organic matter
    (POM), the data base is inadequate
    for all combustion  source cate-
    gories.
  Two other sources of air emissions of
environmental concern are cooling
tower emissions and  emissions from
coal  storage piles. The data bases
characterizing air emissions from these
two sources are inadequate.
  For wastewater effluents from exter-
 nal combustion sources for electricity
 generation,  the data base is adequate
 for wastewater from water treatment
 processes-, and  inadequate for all other
 streams.
  The evaluation of emissions data for
 solid  wastes indicated the inadequacy
 of the organic data base for coal fly ash
 and bottom ash, and the inadequacy of
 the inorganic and organic data bases for
 FGD sludges. On  the  other hand, the
 inorganic data base  for coal ash is
considered to be adequate because of
the adequate  characterization of  the
 inorganic content of coal. Similarly, the
data base for water treatment wastes is
considered to be adequate, because the
waste -constituents are inorganic  and
can be estimated from the raw water
constituents and the treatment method
used.

Commercial/Institutional
Combustion Sources
  Evaluation of emissions data indicates
that the data base for gas- and oil-fired
external combustion sources, although
limited,  is adequate for NOX, total
hydrocarbons, CO,  particulates, and
SOa. However, the data base for specific
organic emissions for these sources is
inadequate, and, for the oil-fired sources,
the  data base  for  SOs and trace
elements is inadequate. Emissions data
from solid-fuel-fired sources are gener-
ally inadequate for all pollutants.
  In the case  of  oil-fired internal
combustion sources, data are  inade-
quate for SOa,  trace element, and
specific organic emissions. Data for
gas-fired reciprocating engines are
adequate; however, one unit was tested
in this program to confirm data adequacy.

Industrial Combustion Sources
  The status  of the  data  base can be
summarized as follows:
  • The data base for criteria pollutants
     is adequate, except for emissions
    from wood-fired combustion
    sources.
  • The  data base for  particulate
    sulfate and sulf uric acid emissions
     is  adequate only for gas-fired
    sources.
  • The data  base for specific organics
    is  inadequate for all  industrial
    source categories.

The Source Measurement
Program
  Because of the deficiencies in the
emissions data base, source tests were
conducted at  a selected number of sites
for each of the five principal combustion
source categories.

Gas- and Oil-Fired Residential
Heating Sources
  Five gas- and five oil-fired residential
sources were  initially selected for
testing.  Upon review of the  results
obtained from the testing of the 10 sites,
1 gas-fired and 2 oil-fired systems were
subsequently tested to study the effect
of cycle  mode on  organic  emissions.
Level II  analyses for SO2, S03, and
particulate sulfate were also conducted
at the two oil-fired sites.

Internal Combustion Sources
  Eleven internal combustion sites
were selected for  testing to better
characterize the emissions associated
with these sources. The sites tested
included  one gas-fueled  gas turbine,
five distillate-oil-fueled gas turbines,
and five distillate-oil-fueled recipro-
cating engines (diesel engines). A gas-

-------
fueled gas turbine site was included to
ensure  that  previously  unidentified
pollutants are  not  being emitted in
environmentally  unacceptable quanti-
ties.
  Test results from the first phase were
evaluated to determine the need for and
type of additional sampling and analysis.
These evaluations led to the recommen-
dation of additional tests to determine
S03 and organic emissions from elec-
tricity-generation distillate-oil recipro-
cating  engines. Level II  tests  were
subsequently conducted at three of the
diesel engine  sites previously tested.
External Combustion Sources
for Electricity Generation
  Forty-six sites were  selected for
sampling and  analysis  of  flue gas
emissions. These 46 sites  include: 3
pulverized dry bottom, 7 pulverized wet
bottom,  6 cyclone, and 3 stoker bitumi-
nous-coal-fired boilers; 3 pulverized dry
bottom, 2 cyclone,  and  2 spreader-
stoker lignite-fired boilers; 4tangentially
fired and 8 wall-fired boilers fueled with
residual oil; and 3 tangentially fired and
5 wall-fired boilers fueled with natural
gas.
  At a selected number of these sites,
wastewater streams and solid wastes
were  also sampled  and  analyzed.
Wastewater  streams sampled  and
analyzed included cooling tower blow-
down,  once-through cooling water,
boiler blowdown, fly ash pond overflow,
bottom ash pond overflow, and combined
ash pond overflow. Intermittent waste-
water streams such as chemical clean-
ing wastes and coal pile runoff were not
sampled. Solid waste streams sampled
and analyzed  included fly ash, bottom
ash, and FGD scrubber sludge.
  In addition  to the modified Level I
tests, comprehensive Level II tests were
also conducted for a bituminous-coal-
fired cyclone  boiler,  two bituminous-
coal-fired pulverized dry bottom boilers,
and an  oil-fired boiler. All these coal-
fired boilers were equipped with flue
gas desulfurization (FGD) systems. The
oil-fired boiler tested used off-stoichio-
metric firing and flue gas recirculation
for NOx control.
   Because  direct   measurements  of
chemical constituents present in cooling
tower  exhausts  have not been  made
(except  for a limited number of  trace
elements), six cooling  towers were
selected for  testing. Cooling tower
streams sampled and analyzed included
air emissions (as evaporation and drift)
and blowdown.
Commercial/Institutional
Combustion Sources
  Twenty-two  external combustion
systems were tested. These included:
five gas-fired, three distillate-oil-fired,
five residual-oil-fired, three anthracite
stokers, three bituminous stokers, two
bituminous pulverized dry  units, and
one wood-fired stoker. Four oil-fired,
one  gas-fired, and one dual-fired
internal combustion reciprocating en-
gines  were also tested.

Industrial Combustion Sources
  Twenty-two  external combustion
systems were tested. These include: 10
gas-fired,  3 distillate oil-fired, and 5
residual-oil-fired boilers; 3 bituminous
pulverized wet bottom and 2 bituminous
pulverized dry bottom units; 3 bitumi-
nous stokers; and 5 wood-fired stokers.

Sampling and Analysis
Methodology

Level I Field Testing
  The  Source Assessment Sampling
System (SASS) train, developed by EPA,
was used to collect both vapor and pani-
culate emissions in quantities sufficient
for the wide range of analyses needed to
adequately characterize emissions from
external combustion  sources. Briefly,
the SASS train consists of a conven-
tional  heated probe, three cyclones, and
a filter in a heated oven which collect
four paniculate size fractions (>10/um,
3-10  //m,  1-3 fjm,  <1  //m); a gas
conditioning system; an XAD-2 polymer
adsorbent trap to collect gaseous
organics and some  inorganics; and
impingers  to collect  the  remaining
gaseous inorganics and trace elements.
The train is run until at least 30 m3 of
gas has been collected.

  In addition to using the SASS train for
stack  gas sampling, other  equipment
was used to collect components that
could  not be analyzed from the train
samples. A gas chromatograph (GC)
with flame ionization detection was
used in the field to analyze hydrocar-
bons in the boiling point range of-160°
to 90°C (reported as d - C6) collected in
gas sampling bags. Additionally, these
samples were analyzed for CO, COz, Oz,
and S02 by GC using a thermal conduc-
tivity detector.
  Water samples were generally taken
by either tap or dipper sampling. Tap
samples, obtained on contained liquids
in motion or  static liquids in tanks or
drums,  was  generally  applicable 1
cooling tower or boiler blowdown. Th
method involved fitting the valve (
stopcock used for sample removal wit
a length of precleaned  Teflon tubin
long enough to reach the bottom of th
container. Dipper sampling, applicabl
to sampling ponds or open discharg
streams, was used to acquire ash pon
discharge samples. The method involve
a dipper with a flared bowl and attache
handle, long enough to reach discharg
areas. After  sample recovery,  wate
analyses using the Hach Kit were per
formed  in the field to  determine pi-
conductivity,  total suspended  solid
(TSS),  hardness,  alkalinity  or acidity
ammonia  nitrogen,  cyanide,  nitrat
nitrogen, phosphate, sulfite, and sulfate
  For solids  sampling, the fractiona
shovel grab samples procedure wa:
used unless the plant had an automatii
sampling system. The concept of frac
tional shoveling involves the acquisitioi
of a time-integrated grab  sampl<
representative of overall process inpu
or output during a given run time period
A standard, square-edged, 12-in,  (30.5
cm) wide shovel was used. For stream;
entering or exiting a process operation
a full cross-stream  cut sample wa:
taken from the belt hourly. Each  hourly
shovel sample was  added  to a pile tc
eventually form a run time penoc
composite. At the conclusion of the run
this  pile was coned and quartered tc
form  a final  representative sample
weighing  from  2.3  to  4.5  kg.  When
plants  were equipped with automatic
samplers  to  remove representative
cross  sections of a stream  while
automatically forming a  homogeneous
composite, these were used in prefer-
ence to the shovel technique.
  In  addition to  the above sampling
methods,  air emissions from cooling
towers were sampled, using a modified
EPA Method  5  train without the filter
assembly.


Modified Level I Laboratory
Analysis
  The basic Level I schematic, outlining
flow of samples and analysis plans for
paniculate and gaseous emissions, is
depicted in Figure 1. The corresponding
schematic for solid, slurry, and liquid
samples is presented in Figure 2.  These
schematics provide a general idea of the
apportionment of samples for analysis.
For example,  Figure 1 shows that the
probe-and-cyclone-rinses combination
will  only  be subjected to inorganic
                                  4

-------
                                      Probe and
                                      Cyclone
                                      Rinses
                                   SASS Train Gas
                                   Conditioner
                                   Condensate
* Weigh
 Individual
 Catches
f// Inorganics
 are Greater than
 10%of Total Catch.
                       3-1
                        1-3/J*
                        Filter
                                     SASS Train
                                     Impingers
                                     Extraction
               Extraction
                            2nc
                                  Organics
                                  Extract
                                                        Inorganics
                  Inorganics
                  As. Sb. Hg
                                 Inorganics
                   Organics
Physical Separation
into LC Fractions,
IR/LRMS
Elements (SSMS) and
Selected Anions
Elements and
Selected Anions
Elements (SSMS) and
Selected Anions

Physical Separation
into LC Fractions. IR/LRMS
                                     Same as Above
                            Chemiluminescence
                             or Method 7
                                                            Elements (SSMS)
                                                            and Selected
                                                            Anions^
                                                            Physical
                                                            Separation
                                                            into LC Fractions,
                                                            'R/LRMS
                         Inorganic
                         (Grab)
                       Organic
                       Material >C6
Organic
Material Ci-C6
One-Site Gas
Chromatography

XAD-2
Absorber,
Module Rinse

On-Site Gas
Chromatography
                    Elements (SSMS) and
                    Selected Anions

                    Aliquot for Gas
                    Chromatographic
                    Analysis
                    Physical Separation
                    into LC Fractions,
                    IR/LRMS
Figure 1.    Basic level 1 sampling flow and analytical plan for particulates and gases.
analysis if the dried sample exceeds 10
percent of the total cyclone-and-filter-
sample weight. Details of the sample
handling, transfer, and analysis proce-
dures are in the IERL-RTP Procedures
Manual: Level I Environmental Assess-
ment. EPA-600/2-76-160a* (NTIS  PB
257850). A brief description of inorganic
and  organic analyses performed  and
deviations from the  basic  Level I
procedure follows.
•Although  superseded by EPA-600/7-78-201
(NTIS PB 293795), the earlier procedures were
used in this study

Inorganic Analyses
  Level I analysis was used for all
inorganic analyses. It was designed to
identify all elemental  species in  the
SASS  train fractions and to provide
semiquantitative data on the elemental
distributions and total emission factors.
The  primary tool for  Level I  inorganic
analysis is Spark  Source Mass Spec-
                 trometry (SSMS).  SSMS data  were
                 supplemented with Atomic Absorption
                 Spectrometry (AAS) data for mercury,
                 arsenic, and antimony, and with specific
                 ion electrode determinations for chlo-
                 rides.
                   The following SASS train fractions
                 were analyzed for elemental composi-
                 tion: 1)the particulate filter, 2) the XAD-
                 2 sorbent, and 3) a composite sample
                 containing  portions of  the  XAD-2
                 module  condensate and HMOs  rinse,
                 and the first impinger solution. Analy-
                 ses of the carbon, hydrogen, nitrogen,
                 oxygen, and trace element contents and
                 heating  values of the fuel were also
                 performed for the  coal- and oil-fired
                 sources.

                 Organic Analyses
                   Level I organic analyses provide data
                 on volatile (boiling point range of 90 to
                 300°C, corresponding to the boiling
                 points of C7 - Ci6 n-alkanesand reported
                                          as C7 - Cie) and non-volatile organic
                                          compounds (boiling point >300°C,
                                          corresponding to the boiling points of
                                          >C16 n-alkanes and reported as >Cie) to
                                          supplement data for gaseous organics
                                          (foiling  point range of  -160 to 90°C,
                                          corresponding to the boiling points of Ci
                                          - C6 n-alkanes and reported as Ci - Ce)
                                          measured in the field. Organics in the
                                          XAD-2 module condensate trap and
                                          XAD-2 resin were recovered by methy-
                                          lene chloride extraction.  SASS train
                                          components including the tubing were
                                          carefully cleaned with methylene
                                          chloride or  methylene chloride/meth-
                                          anol solvent to recover all organics
                                          collected in the  SASS train.
                                            Because all samples are too dilute to
                                          detect organic compounds by the
                                          majority  of instrumental techniques
                                          employed, the first step  in the analysis
                                          was to concentrate the sample fractions
                                          from 1000to10mlinaKuderna-Danish
                                          apparatus in which rinse solvent  is

-------
                                                 Leachable
                                                 Materials
                                                Inorganics
                                                 Organics
                                                Inorganics
                                                 Selected
                                                 Water
                                                 Tests
                                                Organic
                                                Extraction
                                                or Direct
                                                Analysis
                      Selected Anions


                      Physical Separation
                      into LC Fractions.
                      IR/LRMS
                                                Suspended
                                                So/ids
                                                               Elements (SSMSJ and
                                                                                              Physical Separation mtt
                                                                                              LC Fractions, IR/LRMS
                                                                                             Elements (SSMS) and
                                                                                             Selected A n/'ons
                                                     Elements (SSMS) and
                                                     Selected Anions

                                                     Physical Separation into
                                                     LC Fractions, IR/LRMS
                                                                                             Physical Separation into
                                                                                             LC Fractions, IR/LRMS
                                                     Aliquot for Gas
                                                     Chromatographic
                                                     Analysis
  Figure 2.    Basic level 1 sampling flow and analytical scheme for solids, slurries, and liquids.
evaporated while the  organics of
interest are retained.* Kuderna-Danish
concentrates  were then  evaluated by
gas  chromatography  (GC),  infrared
spectrometry (IR), liquid chromatography
(LC) and  gravimetric analysis,  low
resolution mass  spectroscopy (LRMS),
and  sequential gas  chromatography/
mass spectrometry (GC/MS)f.  The
extent of the  organic analysis is deter-
mined by the stack gas concentrations
found for total organics (volatile  and
non-volatile). If the total  organics
indicate a stack gas  concentration
below 500 Aig/m3, a liquid concentra-
tion below 0.1 mg/l, or a solid concen-
tration below 1 mg/kg, further analysis
"Kuderna-Danish is a glass apparatus for evaporat-
 ing bulk amounts of solvents.
fine major modification in the Level I sampling and
 analysis procedure was the addition of GC/MS
 analysis for POM.
is not conducted. If the concentrations
are above these levels, a class fraction-
ation by liquid chromatography  is
conducted followed by GC and IR
analyses. Additionally, if the concentra-
tions in a LC fraction are above these
levels, LRMS is  conducted for that
particular LC fraction.
Level II Sampling and
Analysis
  In addition to the modified  Level I
tests. Level II tests were also conducted
at a number of sites. Level II sampling
and analysis techniques used for these
sites included:
  • Continuous monitoring  of  NO,
     emissions by chemiluminescent
     instrumentation.
  • Continuous monitoring  of  SOa
     emissions  by pulsed flourescent
     analyzer.
 Determination of sulfate emissions
 by the Goksoyr-Ross Controlled
 Condensation System.
 Determination of particle size
 distribution by Polarized Light
 Microscopy (PLM)  and MRI cas-
 cade impactor.
 Determination  of trace element
 concentrations by Atomic Absorp-
 tion  Spectroscopy  (AAS) and In-
 ductively Coupled Plasma Optical
 Emission Spectroscopy (ICP).
 Identification of inorganic com-
 pounds from specific infrared band
 correlations by  Fourier Transform
 IR (FTIR).
 Identification of crystalline material
 in solid samples by X-ray Diffrac-
 tion (XRD).
' Determination of the  surface and
 subsurface sulfur  concentrations

-------
    and oxidation state of bulk samples
    by Electron Spectroscopy for Chem-
    ical Analysis (ESCA).
  • Determination of the surface and
    subsurface composition  of bulk
    samples by Secondary Ion  Mass
    Spectrometry (SIMS).
  • Determination of elemental com-
    position  of single  particles by
    Scanning Electron Microscope
    with Energy Dispersive X-ray
    Fluorescence (SEM-EDX).
  • Identification and quantification of
    non-POM  organic compounds by
    GC/MS.

 Conclusions
  The results of the field measurement
 rogram,  along with supplementary
 alues  from  the  data  base, were
 valuated in terms of data adequacy and
 y using the concept of severity factors.
 wo types of severity factors, ambient
 nd discharge, were used  in data
 valuation. For air  emissions, the
 mbient severity factor, defined as the
 itio of the calculated maximum ground
 ivel  concentration of a pollutant
 pecies to an ambient air quality level or
 lazard  factor,  was  used. The hazard
 ictor for noncriteria  pollutants is a
 educed threshold  limit value (TLV);
 vhile for  criteria pollutants,  it is the
 imbient air quality standard. The TLV is
 educed by a factor of 300 (24/8 x 100)
 o account for length of exposure and an
 idded safety factor  due to the higher
 .usceptibility of the general population
 >f exposure effects. An ambient severity
 actor of greater than 0.05 indicates a
 >otential  problem  requiring  further
 ittention. The "greater than 0.05"
 ;riterion reflects an uncertainty factor
 }f  20  in  the  calculation of  ambient
 severity, because of  potential errors
 ntroduced in  the application of the
 dispersion model, and in Level I sampling
 and analysis. For residential sources, a
 modified ambient severity factor based
 on multiple  sources was used. Maxi-
 mum  ground level concentrations  for
 multiple sources were determined
 using a  dispersion model for an array of
 1000 sources and a grid of houses 80 x
 80 m.
  For  wastewater effluents and solid
 wastes, discharge severities were used
 in data evaluation. Discharge severity is
the ratio of discharge concentration to
the health-based water or solid Dis-
charge Multimedia Environmental Goal
(DMEG). A discharge severity greater
than 1.0 indicates a  potential hazard
requiring further attention. The "greater
than  1.0" criterion,  instead of the
"greater than 0.05"  criterion for
ambient  severity, was  used  because
calculation of discharge  severities was
based on conservative DMEG values.
Also, the uncertainty in  the calculated
values only involved potential sampling
and analysis errors. The error due to the
application of dispersion models was no
longer a component.

Flue Gas Emissions
  The conventional  stationary combus-
tion source categories investigated in
this study contribute significantly to the
nationwide  emissions burden.  As
shown in Table 2, flue gas emissions of
NOX, SOz, paniculate  matter, CO, and
total hydrocarbons from the 39  source
categories studied account for approxi-
mately 86, 66, 36,  10, and 5 percent,
respectively, of the  emissions of these
pollutants from all stationary sources.
  From  an environmental  standpoint,
emissions of NOX, SOz, and particulate
matter are of particular concern.
Ambient severity factors for NOX emis-
sions far  exceed  0.05 for  internal
combustion sources,  utility boilers,
industrial boilers, and coal-fired com-
mercial/institutional boilers. Addition-
ally, ambient severity factors for emis-
sions of SOz and particulate matter far
exceed 0.05 forallbituminous-coal-and
lignite-fired boilers. Emissions of SO2
are also  considered  to be environ-
mentally significant for all residual-oil-
fired boilers.
                 Emissions of total hydrocarbons are
               considered to be a  lesser problem.
               Ambient severity factors for emissions
               of total hydrocarbons exceed 0.05 only
               for large bituminous-coal-, lignite-, and
               residual-oil-fired utility  boilers, indus-
               trial and commercial/institutional coal-
               fired  and wood-fired  stokers, and
               distillate-oil  and gas  reciprocating
               engines.
                  Emissions of CO are not an environ-
               mental concern: their ambient severity
               factors are all well below 0.05.
                 Aside from the criteria pollutants, flue
               gas emissions of SOs (in the form of
               sulfuric acid vapor and aerosol) from
               several combustion  source -categories
               require further attention. These com-
               bustion source categories include: oil-
               fired residential sources,  electricity
               generation and industrial  oil-fueled
               internal combustion sources, bitumi-
               nous-coal- and residual-oil-fired utility
               boilers, bituminous-coal-f ired commer-
               cial/institutional  boilers, and bitumi-
               nous-coal- and residual-oil-fired- in-
               dustrial boilers. Ambient severity
               factors for SOa emissions from these
               sources range from 0.05 to 7.4.  For
               bituminous-coal- and lignite-fired boilers,
               emissions of particulate sulfate are also
               associated with ambient  severity factors
               in  excess  of  0.05  and merit special
               concern.
                 Of the trace elements present in
               bituminous coal, flue gas emissions of
               aluminum, beryllium, calcium, chlorine,
               cobalt, chromium,  fluorine,  iron, lead,
               lithium, nickel, phosphorus, and silicon
Table 2.    Contribution of Conventional Stationary Combustion Systems
           to Nationwide Emissions Burden
    Combustion
    Category
     Emission Contribution as Percentage of All
    	Stationary Sources	

        SOZ   Particulates  Hydrocarbons  CO
 Gas- and Oil-fired
 Residential Heating
 Sources
 Internal Combustion
 Sources
 External Combustion
 Sources for
 Electricity Generation
 Commercial/Institu-
 tional Combustion
 2.5    0.9


18      0.07



50     57
 0.2


 0.1



25
0.2


3.7.



0.6
0.5


2.7



4.3
Sources
Industrial External
Combustion Sources
TOTAL
5.0
10
86
3.0
5.7
66
1.7
9.0
36
0.3
0.4
5
0.5
1.7
10

-------
from  most  coal-fired  boilers  are of
environmental significance. For residual-
oil-fired boilers, flue gas emissions of
beryllium, chlorine, chromium, copper,
lead,  magnesium, nickel, phosphorus,
and vanadium are of principal concern.
Elements with ambient severity factors
in excess of 0.05 also include chromium,
nickel, phosphorus, and vanadium from
distillate-oil-fired industrial boilers,
nickel from distillate-oil-fired commer-
cial/institutional boilers, barium,  cal-
cium, potassium, and phosphorus from
wood-fired boilers, and copper, nickel,
and phosphorus from oil-fueled internal
combustion sources.

  Analysis of organic emissions indi-
cated that the principal constituents of
flue  gas are: saturated straight-chain
and branched hydrocarbons and substi-
tuted benzenes from oil-fueled internal
combustion  sources; glycols, ethers,
ketones, and saturated and aliphatic
hydrocarbons from  utility  boilers; ali-
phatic and  aromatic hydrocarbons,
esters,  ketones, and  carboxylic acids
from commercial/institutional sources;
and esters, ethers, glycols, and aliphatic
and aromatic hydrocarbons from indus-
trial boilers. The most prevalent consti-
tuents are generally  associated with
DMEG values in the 10 to 1000 mg/m3
range. Ambient severity  factors for
these organic  compounds (excluding
POM)  are all  well below 0.05  and
probably not of concern with respect to
human  health.  ROMs  emitted at the
highest concentrations in flue  gas
streams include naphthalene,  phen-
anthrene, pyrene, fluoranthene,  and
chrysene  from  bituminous-coal-fired
sources.  Dibenz(a,h)anthracene  and
possibly benzo(a)pyrene  and benzo
(g,h,i) perylene, all active carcinogens,
were detected  at a  limited number of
sites. POM emissions from wood-fired
boilers were found  to be significantly
higher  than those from  coal-fired
boilers. Dibenz(a,h)anthracene and also
possibly benzo(a)pyrene and benzo(g,h,i)
perylene were detected at some of the
sites tested. The only POMs identified in
flue  gas emissions from  lignite-fired
sources were  biphenyl and trimethyl
propenyl  naphthalene. Carcinogenic
POMs were not detected. For residual-
oil-fired sources, POMs emitted at the
highest concentrations in flue  gas
streams are naphthalene and biphenyl.
Again, carcinogenic  POMs  were not
detected. No POM was detected in flue
gas streams from gas-fired  utility boiler
sites.
Air Emissions from
Cooling Towers
  Two potential environmental problems
associated with the air emissions from
cooling  towers have  been  identified.
First, air emissions of chlorine, magne-
sium, and phosphorus from mechanical
draft cooling towers with high drift rates
are comparable to flue gas emissions of
these elements from  residual-oil-fired
utility boilers and are of environmental
significance. Second, sulfate emissions
from  mechanical draft cooling towers
employing sulfuric acid as an additive,
and with design drift losses in the 0.1 to
0.2 percent range, are of the same
magnitude  as  sulfate emissions from
coal- and oil-fired utility boilers.

Wastewater Discharges
  The major sources of wastewater
discharges from  external  combustion
sources for electricity generation are:
once-through cooling water, blowdown
from  recirculating cooling  systems,
wastes from water treatment processes,
chemical cleaning wastes, and coal pile
runoff.  Discharges from once-through
cooling  system  amount to 7,780,000
I/sec and  account for approximately
99.8  percent of the total wastewater
from conventional utility power  plants.
Of  the  remaining sources, blowdown
from  recirculating cooling  systems is
the largest  contributor to wastewater
discharge.
  From an  environmental  standpoint,
the  pollutants  of  most  concern  in
wastewater effluents from conventional
utility power plants  are  iron, magne-
sium,  manganese,  nickel,  and phos-
phorus. The average organic levels in
the ash pond  effluents sampled were
less than 0.1  mg/l.  Average organic
levels in the cooling tower blowdown
and boiler blowdown sampled were 1 .5
and 6.0 mg/l,  respectively. POMs were
not found above the detection limit of 2
   Based on  discharge severities,  the
 once-through cooling tower  and ash
 pond overflow streams appear to be of
 less environmental significance than
 the  other  wastewater  streams from
 conventional fossil-fueled steam elec-
 tric plants. Total pollutant loading from
 wastewater streams will, however,
 depend on individual discharge flow
 rates.
   Industrial  and commercial/institu-
 tional boilers are smaller contributors to
 wastewater discharges when compared
 with electricity  generation  sources.
Further, characteristics of wastew
discharges from these sources woul
similar to those from electricity gen
tion sources.

Solid Wastes
   Solid waste streams  generatec
conventional utility power plants cor
primarily of coal ash and sludge f
FGD systems. In 1978, total
production was 63.6 Tg and total F
sludge production was 2.1 Tg (ash-fr
Ash production from industrial  .
commercial/institutional  sources  \
proportionally less and FGD slu>
production  from these  sources  \
negligible.
   Leaching of trace elements from c
ash may result in environmental c
tamination. Concentrations of 11 to
trace elements in bituminous coal <
and lignite  ash exceed  their heal
based solid DMEG values. The polluta
of most concern are aluminum, arser
calcium,  chromium, iron, mangane
nickel, potassium, and silicon.

Recommendations
   Because of  inadequacies in the dc
•base that characterizes emissions frc
conventional  stationary  combustii
systems, it is  recommended that ad<
tional studies be conducted to provi<
the following key data needs.

Flue Gas Emissions
  • There is a lack of emissions data f i
     pulverized dry-bottom boilersfirir
     Texas lignite. This is a serious dai
     deficiency because approximate
     16,000 MW of added generatin
     capacity is planned for this sourc
     category in the 1978-1985 perio<
  • Size distribution data for flue ga
     emissions of particulates ar
     inadequate for bituminous-coa
     lignite-,  and  residual-oil-firei
     boilers.
  • The   date bases  for  particulat
     sulfate emissions from bituminous
     coal- and lignite-fired sources an
     inadequate. Also,  SOs emission;
     data for lignite-fired  sources art
     presently unavailable.
  • For bituminous-coal-fired boilers
     equipped with wet  scrubbers oi
     mechanical precipitators, the data
     bases  characterizing flue  gas
     emissions of  trace elements are
     inadequate.  Data for flue  gas
     emissions of trace elements from
     lignite-fired boilers are generally
                                  8

-------
    not available. Analysis of the data
    acquired in this program indicated
    the need for additional characteri-
    zation studies.
 • Although  current data indicated
    that flue gas emissions of specific
    organics  (excluding  POM) are
    probably  not of concern with
    respect to human health, more
    detailed Level II  organic analysis
    would be required to conclusively
    determine the  significance  of
    organic emissions.
 • Emissions of POM  from  bitumi-
    nous coal- and wood-fired sources
    will require  further  characteriza-
    tion, with special emphasis on the
    positive identification and quantifi-
    cation of carcinogenic compounds
    such as dibenz(a,h)anthracene,
    benzo(a)pyrene, and benzo(g,h,i)
    perylene.

fVastewater Discharges
 • The data bases characterizing cooling
    tower blowdown, ash pond over-
    flow, chemical cleaning wastes,
    wet scrubber wastewater, and coal
    pile runoff are inadequate. The
    present study has been  instru-
    mental in applying Level I  techni-
  ques to identification of waste-
  water  constituents which  pose
  potential environmental problems.
  Since  potential  problems were
  detected  by Level I techniques,
  further studies using  Level II
  techniques will  be  required  to
  adequately characterize wastewater
  effluents from utility boilers.
Solid Wastes
  • The data base characterizing flue
    gas emissions of POM from bitu-
    minous-coal-fired sources is ade-
    quate except for dibenz(a,h)an-
    thracene and benzo(a)pyrene.
    Emissions of these specific POMs
    will require further characteriza-
    tion.
C. C. Shih and A. M. Takata are with TRW Environmental Engineering Division,
  One Space Park, Redondo Beach, CA 90278.
Michael C. Osborne is the EPA Project Officer (see below).
The complete report, entitled "Emissions Assessment of Conventional Station-
  ary Combustion Systems: Summary Report, "(Order No PB 82-109 414; Cost:
  $9.50, subject to change) will be available only from.
        National Technical Information Service
        5285 Port Royal Road
        Springfield, VA 22161
        Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
        Industrial Environmental Research Laboratory
        U.S. Environmental Protection Agency
        Research Triangle Park,  NC 27711
                                                                       U. S. GOVERNMENT PRINTING OFFICE: I98I/559-092/3347

-------
                                                    ~ 9

                                                    ss
                                                    o

                                                    in
                                                    
                                                          m > T)
                                                          WO
                                                              O

-------