-------
-------
EPA-600/7-79-029b
May 1979
Emissions Assessment of Conventional
Stationary Combustion Systems;
Volume I. Gas- and Oil-fired
Residential Heating Sources
by
N. F. Surprenant, R. R. Hall, K. T. McGregor,
and A. S. Werner (GCA Technology Division)
TRW, Inc.
One Space Park
Redondo Beach, California 90278
Contract No. 68-02-2197
Program Element No. EHE624A
EPA Project Officer: Ronald A. Venezia
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
ABSTRACT
Emissions from gas- and oil-fired residential heating sources were
assessed through a critical examination of existing emissions data, followed
by the conduct of a phased measurement program to fill gaps in the emissions
data base. Initially, five gas-fired and five oil-fired residential sources
were selected for testing. Mass emission rates of criteria pollutants, trace
elements, and organics, including polycyclic organic matter (POM), were deter-
mined. Subsequent evaluation of the test program led to a decision to conduct
additional tests at one gas-fired and two oil-fired sites. The principal ob-
jective of this second test program was to determine the effect of burner
on/off cycle on emissions. Particulate sulfate, S(>2, and 863 emission data
were also obtained at the oil-fired sites.
The results of the emissions assessment indicate that residential sources
are of potential significance based on multiple source severity factors calcu-
lated for an array of houses burning gas or oil. Pollutants.for which multi-
ple source £eyerity_factors.excejed 0.05, the level which may be potentially
significant, are NOX from gas-fired sources and 803, NOX, and Ni from oil-fired
sources. Measured criteria pollutant emission factors were generally compara-
ble to EPA emission factors given in AP-42 with the exception of total hydro-
carbon emissions from oil-fired sources which were three times greater. How-
ever, POM compounds known to be carcinogenic were not found above the detection
limit of 0.3 yg/m3. In contrast to previous studies, variations in the burner
operating cycle had no measurable effect on emissions. Failure to detect an
effect may be due to the accuracy limitation (± a factor of three) inherent in
Level I measurements.
iii
-------
CONTENTS
Abstract iii
Figures vii
•Tables viii
Acknowledgment xii
1. Summary and Conclusions 1
1.1 Residential Source Description 2
1.2 The Existing Residential Source Emissions Data Base . 3
1.3 The Residential Source Measurement Program 3
1.3.1 Field Testing 4
1.3.2 Laboratory Analysis 4
1.3.3 Results 5
1.4 Conclusions 7
2. Introduction . 10
3. Source Description 14
3.1 Size of Industry and Geographic Distribution 14
3.2 Population Characteristics 18
3.3 Characteristics of Combustion Equipment 19
4. Emissions 24
4.1 Evaluation of Existing Emissions Data . 24
4.1.1 Criteria for Evaluating the Adequacy of
Emissions Data ; . . . 24
4.1.2 Sources of Existing Emissions Data 26
4.1.3 Existing Emissions Data: Gas-Fired Sources . . 26
4.1.4 Existing Emissions Data: Oil-Fired Sources . . 29
4.1.5 Status of Existing Emissions Data Base for
Gas- and Oil-Fired Residential Sources ... 31
-------
CONTENTS (continued)
4.2 Emissions Data Acquisition 31
A.2.1 Selection of Test Facilities 31
4.2.2 Field Testing Procedures 36
4.2.3 Laboratory Analysis Procedures 42
4.2.4 Test Results 56
4.3. Analysis of Existing Data and Test Results 68
4.3.1 Emissions of Criteria Pollutants and S02, 803,
and Particulate Sulfate 68
4.3.2 Emissions of Trace Elements 74
4.3.3 Emissions of POM 77
4.3.4 Summary of Status of Emissions Data Base ... 79
5. Total Emissions 80
5.1 Current and Future Fuel Consumption 80
5.2 Current and Future Nationwide Emissions 85
6. References 89
Appendices
A. Conversion Factors and Metric Prefixes 92
B. Fuel Consumption By Residential Space Heating Sources ... 94
C. Criteria for Evaluating the Adequacy of Existing Emissions
Data for Conventional Stationary Combustion Sources . . . 104
D. Summary of Existing Emissions Data 119
E. Data Reduction Procedure 133
F. Gas- and Oil-Fired Residential Source Laboratory Data . . . 137
vi
-------
FIGURES
Number
1
2
3
4
5
6
7
8
9
C-l
Geographical distribution of residential fuel consumption
for space heating, 1975
Basic Level I sampling and analytical scheme for
SASS schematic ....
Sampling locations for oil-fired residential sources .....
Level I inorganic analysis methodology for residential
sources
Level I organic analysis methodology for residential
sources .....
Sample identification and coding for residential sources . . .
Step 1 screening mechanism for emissions data
Page
11
17
35
38
40
41
43
48
66
106
vii
-------
TABLES
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Summary of Results of Emissions Assessment for Residential
Sources
Estimated Contribution of Stationary Combustion Sources to
Anthropogenic Air Pollutants, 1973
U.S. Residential Space Heating Fuel Use By Region, 1975 ....
Population Distribution of Residential Units Now in
Service, 1974 ...
Size Distribution of Residential Oil-Fired Systems, 1972 . . .
Average Age of Residential Oil Burners, 1972
Design and Operating Parameters for Gas- and Oil-Fired Forced
Mean Severity Factors for Gas- and Oil-Fired Residential.
Sources
a
Summary of Existing Emissions Data for Gas-Fired Residential
Sources
POM Emissions from Gas-Fired Residential Sources .......
Summary of Existing Emissions Data for Oil-Fired Residential
Sources
POM Emissions from Oil-Fired Residential Sources
Trace Metal Emissions from an Oil-Fired Burner . .
Summary of Evaluation of Existing Emissions Data
Characteristics of Residential Combustion Units Tested ....
Mass to Charge Values Monitored
Page
6
10
16
18
19
20
20
22
26
27
28
29
30
32
33
37
53
viii
-------
TABLES (continued)
Number
18 Minimum List of POM's Monitored ................ 54
19 Field Data: Gas-Fired Residential Combustion Sources ..... 57
20 Field Data:. Oil-Fired Residential Combustion Sources ..... 58
21 Summary of AAS Results for Hg, As, and Sb ........... 61
22 Total Organic Emissions from Gas- and Oil-Fired Residential
Combustion Sources ..................... 62
23 Organic Analysis Results of LC Fractionation of SASS Samples
from Gas- and Oil-Fired Residential Sources ......... 65
24 Classes of Compounds Identified in IR Spectra of XAD-2 Resin
LC Fractions from Oil-Fired Residential Sources ..... . . 67
25 POM Emissions from Oil-Fired Residential Sources ....... 69
26 Measured Emission Factors for Particulate, NOx, and
Total Organics ................ ....... 70
27 Comparison of Criteria Pollutant Emission Factors for
Gas- and Oil-Fired Residential Combustion Sources ...... 71
28 Criteria Pollutant and SOa Severity Factors for
.Residential Sources ..................... 75
29 Trace Element Emission Factors for Oil-Fired Residential
Sources ........................... 76
30 Trace Element Severity Factors for Oil-Fired Residential
Sources . .......................... 77
31 POM Emissions and Severity Factors for Oil-Fired Residential
Sources ....... .................... 78
32 Consumption of Energy in the Residential and Commercial
Sector, 1974, and Projections to 1985 ............ 81
33 Estimates of Residential Space Heating Fuel Consumption
to 1985 Based on Population Growth ...... . ...... 84
34 Estimates of Residential Space Heating Fuel Consumption
to 1985 Based on Fuel Type ................. 85
ix
-------
TABLES (continued)
Number
35 Relationship of Gas- and Oil-Fired Residential Heating
Source Emissions to Total Estimated Emissions from
Combustion Sources 86
36 National Trace Element Emissions from Oil-Fired Residential
Sources, 1978 87
37 National POM Emissions from Oil-Fired Residential
Sources, 1978 88
B-l Residential Space Heating Fuel Use By State, 1975 97
B-2 U.S. Residential Space Heating Fuel Use By Region, 1975 .... 99
C-l Maximum Ratio of Extreme Ranking Observations 117
D-l Emission Rates of CO, HC, NOx, S02, Particulates, and
Aldehydes from Gas-Fired Residential Sources ........ 120
D-2 Emission Rates of CO, HC, NOx, S02, and Particulates, and
Bacharach Smoke Numbers from Oil-Fired Residential Sources . 128
E-l Fuel Composition and Combustion Characteristics 135
F-l SSMS Analytical Data: Site 100 138
F-2 SSMS Analytical Data: Site 101- 140
F-3 SSMS Analytical Data: Site 102 142
F-4 SSMS Analytical Data: Site 103 144
F-5 SSMS Analytical Data: Site 104 146
F-6 SSMS Analytical Data: Site 300 .. 148
F-7 SSMS Analytical Data: Site 301 150
F-8 SSMS Analytical Data: Site 302 152
F-9 SSMS Analytical Data: Site 303 .......... 154
F-10 SSMS Analytical Data: Site 304 156
F-ll SSMS Analytical Data: Sites 326 and 327 158
x
-------
TABLES (continued)
Number Page
F-12 Distribution of Volatile (C8-C12) and Nonvolatile (* C16)
Organics in SASS Train Samples: Gas-Fired Sources ..... 160
F-13 Distribution of Volatile (Ce-C^) and Nonvolatile (> C16)
Organics By Gas-Fired Site ................ . 161
F-14 Distribution of Volatile (Cy-Cie) and Nonvolatile (> C16)
Organics in SASS Train Samples: Oil-Fired Sources ..... 162
F-15 Distribution of Volatile (Cy-Cie) and Nonvolatile (> C16)
Organics By Oil-Fired Site ..... . ........... 163
xi
-------
ACKNOWLEDGMENT
The authors gratefully acknowledge the technical guidance of the follow-
ing individuals who contributed to the review and revision of this report:
Ronald A. Venezia, Project Officer; Edward G. Bobalek; Robert E. Hall;
William B. Kuykendal; Warren Peters; Wade H. Ponder; Kenneth E. Rowe; and
W. Gene Tucker, of the Industrial Environmental Research Laboratory (IERL) ,
Research Triangle Park, North Carolina; Don Gilmore, Environmental Monitoring
and Support Laboratory, Las Vegas, Nevada; Victor F. Jelen, IERL, Cincinnati,
Ohio; and Robert G. Hooper of the Electric Power Research Institute. The
authors further acknowledge the contribution of the GCA and TRW field and
laboratory personnel who were responsible for sample collection and analysis.
A special acknowledgment is extended to Sandra M. Sandberg who played a major
role in the technical preparation and editing of this document.
xii
-------
1. SUMMARY AND CONCLUSIONS
Emissions from gas- and oil-fired residential combustion sources for
space heating have been assessed through a review and evaluation, of the exist-
ing emissions data base in conjunction with a phased sampling and analysis
measurement program designed to fill data gaps. Similar assessments of emis-
sipns from electricity generation and industrial internal combustion sources,
electricity generation external combustion sources, industrial external com-
bustion sources and commercial/institutional external combustion sources are
being conducted and results will be presented in future reports.
This phased approach to an emissions assessment is designed to provide
comprehensive information in a cost-effective manner through two distinct
sampling and analysis levels. Level I procedures^ utilize semiquantitative
(± a factor of three) techniques of sample collection and laboratory and
field analyses: (1) to provide preliminary emissions data for waste streams
and pollutants not adequately characterized, (2) to identify potential pfob-
lem areas, and (3) to prioritize waste streams and pollutants in those
streams for further, more quantitative testing. Using the information from
Level I, available resources can be directed toward Level II testing which
involves specific, quantitative analysis of components of those streams
which do contain significant pollutant levels. The data developed at
Level II .are used to identify control technology needs and to further define
the environmental hazards associated with emissions. A third phase,
Level III, which is outside the scope of this program, employs continuous
or periodic monitoring of specific pollutants identified at Level II so that
the emission rates of these critical components can be determined exactly as
a function of time and operating condition.
-------
1.1 RESIDENTIAL SOURCE DESCRIPTION
Residential space heating sources are defined as combustion units with
fuel input capacities below 422 x 106 joules per hour (0.4 x io6 Btu/hr)* in
accordance with recent U.S. Environmental Protection Agency (EPA) sponsored
studies. Residential combustion systems consume about 15 percent of the fuel
used by conventional stationary combustion systems. The residential sector
accounts for about 6.8 x 10*8 joules of the 1978 estimated fuel consumption
total of 45 x IO18 joules. This source uses primarily gas (58 percent) and
oil (38 percent). It is estimated that in 1974 there were about 34,000,000
gas-fired, 13,000,000 oil-fired, 740,000 coal-fired, and 660,000 wood-fired
residential space heating systems in the United States.
Heating systems for residential sources are concentrated in areas of
high population density such as the northeast, midwest, and parts of Cali-
fornia. Oil consumption is most heavily concentrated in the northeast with
the states of Pennsylvania, New York, New Jersey, Massachusetts, and Con-
necticut consuming 53 percent of the U.S. total. Only very small amounts of
oil are burned in the west and south. Residential gas consumption for space
heating is more widely distributed than oil, but is still most heavily con-
centrated in the upper midwest and northeast. States that account for more
than 5 percent of the U.S. total residential gas consumption include Illinois
(8.9 percent), New York (8.3 percent), Ohio (8.1 percent), California
(7.8 percent), Michigan (7.6 percent), and Pennsylvania (6.0 percent).
Residential gas- and oil-fired space heating equipment is subject to a
number of design variations related to burners, combustion chambers, excess
air, heating medium, etc. Residential systems generally operate only in an
on/off mode with no variation in fuel input rate in contrast to load modula-
tion encountered with larger commercial, industrial, and electric utility
systems.
Gas-fired systems are inherently less complex and easier to maintain
than oil-fired units because the fuel is cleaner and atomization is not
*
1 Btu = 1,055 joules (J). Although it is EPA policy to use the metric
system, this publication uses certain nonmetric units for convenience.
A conversion table is presented in Appendix A.
-------
required. Residential gas burners use natural aspiration and are very simi-
lar in design, whereas several burner designs are used for oil atomization.
However, high pressure (-100 psig) atomization burners account for about
90 percent of the total.2 Low pressure and rotary burners are being phased
out because of their complexity. Although air pollution control equipment is
not available, for residential combustion systems, emission reduction measures
are being evaluated. The EPA is active in the development and evaluation of
residential gas- and oil-fired burners and furnaces.
1.2 THE EXISTING RESIDENTIAL SOURCE EMISSIONS DATA BASE
A major task in this program has been the identification of gaps and
inadequacies in the existing emissions data base for residential combustion
sources. Decisions as to the adequacy of the data base were made using cri-
teria developed by considering both the reliability and variability of the
data. Environmental risks associated with the emission of each pollutant
were also considered in the determination of the need for, and extent of,
the phased sampling and analysis program.
The sources of emissions data for residential gas- and oil-fired systems
are limited at the present time to early data used to generate EPA emission
factors and more recent data developed by EPA contractors for criteria pollu-
tants. For gas-fired systems, the existing data base for S02, NO , HC, and
X
CO emissions is.adequate. However, the existing data base for particulate
and organic emissions is inadequate. For oil-fired systems, the existing
emissions data base for particulate, S02> NO , HC, and CO is adequate, but
X
inadequate for 803, particulate sulfate, trace element, and organic emissions.
1.3 THE RESIDENTIAL SOURCE MEASUREMENT PROGRAM
To remedy deficiencies in the existing emissions data base, five gas-
fired and five oil-fired residential sources were initially selected for
testing. The choice of specific sites within the two source categories was
based on the representativeness of the sites with respect to such important
system characteristics as burner type and age, firing rate, and heating
medium (hot air, hot water, and steam). Upon review of the results obtained
from the testing of the 10.sites, one gas-fired and two oil-fired systems
were subsequently tested to study the effect of cycle mode on organic
-------
emissions. Level II analyses for SC>2> $03, and particulate sulfate were also
conducted at the two oil-fired sites.
1.3.1 Field Testing
The Source Assessment Sampling System (SASS) train was used to collect
both gas phase and particulate emissions in quantities sufficient for the
wide range of analyses needed to adequately characterize residential source
emissions. In addition to using the SASS train for stack gas sampling, other
equipment was employed to collect those components not analyzable from the
train samples. A gas chromatograph (GC) with flame ionization detection was
used in the- field to analyze low boiling hydrocarbons (boiling point < 100 C).
Additionally, CO, 02, and C02 were field analyzed by GC using a thermal con-
ductivity detector. Detection tubes were used for CO in the second series of
tests to increase measurement sensitivity because the detection limits attain-
able for CO by the field GC were above CO levels normally measured in the
stacks of residential heating systems. Analyses for NO were carried out at
. X
the gas-fired sites electrochemically using a Theta detector. NO emissions
X
were not analyzed at the oil-fired sites. Goksoyr-Ross sampling and analysis
for S02, 803, and particulate sulfate were also conducted at the two addi-
tional oil-fired sites tested later in the program.
1.3.2 Laboratory Analysis
The basic Level I sampling and analytical scheme for particulate and
gaseous emissions is depicted in Figure 3 (Section 4). The analytical scheme
was modified, however, for this emissions assessment program. The major modi-
fication of the Level I sampling and analysis procedure was that gas chroma-
tography/mass spectroscopy (GC/MS) analyses for POM were performed on the
samples collected in this program. Level II analyses for S02, 803, and par-
ticulate sulfate were also conducted for the two oil-fired sites tested in
the second stage of the program.
1.3.2.1 Inorganic Analyses1—
The Level I analysis scheme was used for all inorganic analyses. This
scheme was designed to identify elemental species in the SASS train fractions
and to provide semiquantitative data on elemental distributions and total
emission factors. The primary tool for Level I analysis is the Spark Source
-------
Mass Spectrograph (SSMS). The SSMS was chosen for Level I for its capability
of detecting 70 elements simultaneously with sub-ppm sensitivity. The SSMS
data were supplemented with Atomic Absorption Spectrometry (AAS) data for Hg,
As, and Sb, and with Level II Goksbyr-Ross determinations of 862, 863, and
particulate sulfate emissions from two oil-fired sources.
1.3.2.2 Organic Analyses—
Level I organic analysis is a technique designed to identify organic
compounds in greater detail than a total hydrocarbon equivalent analysis.
It provides data on gaseous organics with boiling points less than 100 C,
volatile organics boiling between 100 and 300 C, and nonvolatile organics
with boiling points greater than 300°C.
Because all samples were too dilute to detect organic compounds by the
majority of instrumental techniques employed, the first step in the. analysis
was to concentrate the SASS train samples as much as 100-fold. The concen-
trated samples were then evaluated by gas chromatography (GC), gravimetric
analysis, infrared spectrometry (IR), and Level II sequential GC/MS. The
extent of the organic analysis was determined by the stack gas concentrations
found for total organics (volatile and nonvolatile). If the total organics
per sample fraction indicated a stack gas concentration below 500 yg/m3, then
no further analysis was conducted. If the concentration was above 500 yg/m3,
then a class fractionation by liquid chromatography (LC) was conducted fol-
lowed by GC, gravimetric, and IR analyses of each fraction.
1.3.3 Results
The results of the field measurement program along with supplementary
values for certain pollutants obtained from the existing data base are
summarized in Table 1. Severity factors, defined as the ratio of the maximum
ground level concentrations of species emitted from the source to an ambient
air quality level or hazard factor, are also listed in Table 1. The hazard
t'
factor for noncriteria pollutants is a reduced threshold limit value (TLV),
while for criteria pollutants it is the ambient air quality standard. The
t
TLV is reduced by a factor of 300 (24/8 x 100) to account for length of
exposure and an. added safety factor due to the higher susceptibility of the
general population to exposure effects.
-------
TABLE 1. SUMMARY OF RESULTS OF EMISSIONS ASSESSMENT FOR RESIDENTIAL SOURCES
Gas-fired sources Oil-fired sources
Emissions
Pollutant ,.
from
AP-423
(ng/J)
Particulate 2-6
S02* 0.26
S03 ND
NOX 33.
CO 8.4
Organics 3 . 3
(total)
Trace
elements
Lead
Cadmium
Copper
Nickel
Chromium
Multiple Emissions
Emissions Severity source from Emissions
(ng/J) factor severity AP-423 (ng/J)
factor (ng/J)
1.0 1.7 x IQ-5 4.3 x 10-4 7.7 3.1
ND 3.2 x 10~6 8.0 x 10"5 106. ND
ND ND ND 1.5 5.9
33. 2.8 x 10~3 7.0 x ID"2 55. ND
ND 1.6 x 10~6 4.0 x 10~5 15. ND
2.6 1.0 x IQ-1* 2.5 x 10~3 3. 9.2
7.5 x ID'2
2.2 x ID"2
0.25
- - 0.49
- 5.5 x ID"2
Severity
factor
Multiple
source
severity
factor
7.7
1.9
1.6
6.2
4.2
5.3
1.0
1.0
2.4
10.
1.0
x 10~5
x 10" 3
x 10"2
x 10" 3
x ID"6
x lO-4
x 10- 3
x 10- 3
x 10" 3
x 10- 3
x 1Q- 3
1.9
4.8
4.0
1.6
1.1
1.3
17
23
12
250
25
x 10~3
x 10~ 2
x 10- l
x 10~ *
x lO"1*
x ID" 2
x 10- 3
x ID" 3
x 10- 3
x 10- 3
x 10- 3
Based on fuel sulfur content of 2000 grains/106 ft3 for gas and 0.25 weight percent for oil.
Multiple source severity factors for all elements dashed (-) or not listed were less than 0.01.
bound values of emissions were used to calculate severity for oil-fired sources.
ND - Not Determined
Upper
-------
Severity factors calculated from program emission data or, in the case
of pollutants not measured in the program, from EPA (AP-42)3 emission factors
are shown for a single source and for multiple sources, Maximum ground level
concentrations for multiple sources were determined using a dispersion model
for an array of 1000 sources. The model assumes a Class C stability (slightly
unstable) and'a windspeed of 4,5 m/sec (10 mph), Using a grid of houses
80 x 80 m and the average stack parameters found in this study, the maximum
ground level concentrations determined by the model were about 25 times
greater than those from a single source. As shown in Table 1, multiple source
severity factors for several pollutants (NOX for gas-fired sources and NOx,
863, and Ni for oil-fired sources) exceed 0.05, a value which indicates that
emissions are potentially significant.
Data for POM obtained by GC/MS are not reported in the table. POM was
not found in the emissions from gas-fired residential sources; the concentra-
tions of POM measured for oil-fired sources were at least two orders of mag-
nitude below levels that are considered hazardous. Compounds considered par-
ticularly hazardous, such as benzo(a)pyrene and benzo(a)anthracene, were not
detected.
In contrast with earlier studies, a change in the on/off cyclic mode of
burner operation from a 50 minute on/10 minute off cycle to a 10 minute on/
20.minute off cycle did not result in increased HC (or POM) emissions.
1.4 CONCLUSIONS
Several conclusions, as listed below, can be drawn from this emissions
assessment of gas- and oil-fired residential sources.
• Multiple source severity factors determined for an array
of 1000 residential sources exceed 0.05 for a number of
pollutants. These pollutants, of potential environmental
significance, are NOX from gas-fired sources and NOX, 803,
and Ni from oil-fired sources.
• The average emission factors for criteria pollutants
measured in this program, despite large source-to-source
variations, are in agreement with EPA emission factors
(AP-42) within the Level I accuracy of a factor of three.
-------
A singular exception is the HC emission factor which is
3.1 times greater than the EPA emission factor. The
multiple source severity factor for HC emissions from
oil-fired sources is 1.3 x 10~2. Emissions of criteria
pollutants are adequately characterized.
Particulate, S02, and NOX emissions from residential
sources account for about 0.4, 1.2, and 2.5 percent,
respectively, of emissions from all stationary combus-
tion sources based on a previous estimate of total
nationwide emissions.14 Residential CO emissions account
for about 7 percent of the total CO emissions from sta-
tionary sources, with gas- and oil-fired sources con-
tributing equally. HC emissions from residential
sources account for about 10 percent of the total HC
emissions from stationary combustion sources, with
oil-fired sources contributing 62 percent of the gas-
and oil-fired residential total.
803 emissions from oil-fired residential sources
represent a potential hazard. The 803 emission factor
measured in this program'is three times greater than
the EPA emission factor.3 Further work is needed to
determine 803 emission factors.
Trace element emissions from gas-fired residential
sources are insignificant. The only element emitted
from oil-fired sources of potential environmental
significance is Ni. The multiple source severity
factor for Ni is 0.25.
POM emissions, as measured in this program, are not
environmentally significant. No POM emissions were
detected from gas-fired sources. POM multiple source
severity factors from oil-fired units are generally
two to five orders of magnitude below levels considered
hazardous. Although the data obtained in this study
8
-------
represent a valuable contribution to a sparse existing
data base, further work is desirable because of the
high level of potential hazard associated with POM.
• Within the accuracy limitations of Level I (± a factor
of three), a change in burner cycle mode from 50 minutes
on/10 minutes off to 10 minutes on/20 minutes off had no
effect on HC and POM emissions. The effect of cycle on
emissions noted by other investigators is undoubtedly a
real effect and merits further study to determine its
magnitude and significance.
On the basis of the above, further work is recommended to determine 803
concentrations and the effect of cycle variations on emissions from oil-fired
sources. Additional work to determine emission factors for POM should be
undertaken to build a larger data base. Modeling studies are also recom-
mended to determine severities from a multiple array of oil-fired sources,
using EPA emission factors and the trace element content of oil to provide
emission rate data for the model. The emission data base for gas-fired
residential sources is adequate and no further work is needed.
-------
2. INTRODUCTION
Conventional stationary combustion systems are major sources of pollu-
tion of air, water, and land. A preliminary assessment1* of the relative
significance of stationary combustion systems as sources of pollution esti-
mated that these combustion sources contribute a major portion of the total
manmade emissions of nitrogen oxides, .sulfur oxides, and particulates
(Table 2). This preliminary assessment also identified as generally inade-
quate the emissions data for a number of potentially harmful pollutants,
including trace elements, 803, and particulate sulfate.
TABLE 2. ESTIMATED CONTRIBUTION OF STATIONARY COMBUSTION
SOURCES TO ANTHROPOGENIC AIR POLLUTANTS, 19734
., . . Percent of
n 11 4. ^ Emissions n ,
Pollutant ,-,n^ i \ total manmade
(10° tonnes/yr) . .
3 emissions
Particulates
SO
X
NO
X
HC
CO
6,420
20,050
10,000
320
980
20
71
44
1
1
The overall objective of the current program is to provide a comprehen-
sive assessment of emissions from selected conventional stationary combustion
systems. The assessment process is based on a critical examination of exist-
ing data, followed by a phased measurement approach to resolve data gaps
(Figure 1). In the first phase, modified Level I sampling and analysis pro-
cedures are used to provide results accurate to a factor of three so that
preliminary assessments can be made and problem areas identified. Evaluation
of results from the first phase will determine pollutants requiring a more
10
-------
PHASE 1
PHASE 2
PHASE 3
NECESSARY
DATA
Existing
Adequate
Data Base
Nonproblem
Substances
Problem
Substances
Missing or
Inadequate
Data Base
LEVEL I
ANALYSIS
Nonproblem
Substances
Problem
Substances
LEVEL II
ANALYSIS
LEVEL III
MONITORING
AND CONTROL
DEVELOPMENT
EVALUATION
K
Not within the scope of the present program.
Figure 1. Program plan.
-------
detailed and accurate Level II sampling and analysis program. The character-
ization of combustion source emissions from this program will allow EPA to
determine the environmental acceptability of combustion waste streams and
pollutant levels and the need for control of those pollutants which are
environmentally unacceptable.
A third phase, Level III, which is outside the scope of this program,
employs continuous or periodic monitoring of specific pollutants identified
at Level II so that the emission rates of^ these critical components can be
determined accurately as a function of time and operating condition.
The combustion source types to be assessed in this conventional combus-
tion emissions assessment program have been selected because they are among
the largest, potentially largest, or most numerous stationary combustion
source types. A total of 50 source types have been selected for study and
classified under the following principal categories:
1. Electricity generation - External combustion
2. Industrial - External combustion
3. Electricity generation and industrial - Internal combustion
4. Commercial/Institutional - Space heating
5. Residential - Space heating.
These five principal categories have been further divided into subcategories
based on fuel type, furnace design, and firing method because their emission
characteristics are dependent upon these parameters.
This program report is the first in a series of five reports, and is
concerned with the emissions assessment of gas- and oil-fired residential
combustion sources used for space heating (category 5, above). Residential
coal and wood combustion sources are being studied under a separate EPA
contract.
The approach utilized in the emissions assessment of residential combus-
tion sources is similar to that utilized for the assessment of other combus-
tion source types. First, available information concerning the process and
population characteristics of residential combustion sources and their emis-
sions was reviewed to determine the adequacy of the available data base.
12
-------
Modified Level I sampling and analysis was then conducted at selected repre-
sentative sites (initially five gas-fired and five oil-fired residential
sites) to remedy inadequacies in the existing data base. The results of the
modified Level I sampling and analysis program were then evaluated to deter-
mine the need for Level II or additional Level I sampling and analysis and to
identify the potentially hazardous substances emitted from these combustion
sources. As a result of this evaluation, a decision was made to acquire SC>2,
S0$, and particulate sulfate emission data and to study the effect of burner
cycle mode on emissions. This additional testing and analysis have been
conducted and results are included in this report. Lastly, emissions data
obtained from the sampling and analysis programs were combined with existing
emissions data to provide estimates of current and future nationwide emis-
sions of pollutants from gas- and oil-fired residential combustion sources.
13
-------
3. SOURCE DESCRIPTION
Residential combustion systems for space heating are classified as units
with fuel input capacities below 422 x 1Q6 J/hr (0.4 x 106 Btu/hr) in accor-
dance with recent U.S. Environmental Protection Agency sponsored studies.
The focus of this study is gas- and oil-fired residential heating systems.
Gas and oil are the most important residential space heating fuels, repre-
senting 5.8 percent and 38 percent, respectively, of 1978 fuel consumption..
Although these systems are not considered to be a major source of nationwide
pollutant emissions,1* their environmental impact is enhanced by high seasonal
fuel consumption, winter meteorological conditions, proximity of the popula-
tion to the emissions sources, and the release of emissions at close-to-
ground level. To provide a better understanding of the emission problem
associated .with gas- and oil-fired home heating sources, and to aid in the
selection of representative test units, brief descriptions of industry size
and geographic distribution, population characteristics, and the design and
operating characteristics of combustion equipment are discussed in this
section.
3.1 SIZE OF INDUSTRY AND GEOGRAPHIC DISTRIBUTION
Residential combustion systems consume about 6,750 x 1015 J/yr of fuel
for space heating based on 1978 estimates presented in this report. Other
conventional stationary combustion systems burn a total of 38,000 x 1015 J/yr.
Residential systems use primarily gas (58 percent) and oil (38 percent) in
contrast to the electric utility sector which burns primarily coal (55 percent)
Pollutant mass emissions estimates developed in a previous study1* indi-
cated "that gas- and oil-fired residential combustion sources account for
roughly 2 percent of total particulates, 5 percent of SO , 3 percent of NO ,
X . X
13 percent of HC, and 17 percent of CO emissions from the stationary combus-
tion sources considered in this study.
14
-------
Heating systems for residential sources are concentrated in areas of
high population density. Regional residential space heating fuel consumption
data for 1975 are presented in Table 3 and Figure 2. The distribution of
fuel usage by region is probably very similar for the years 1975 and 1978.
Data inputs used to develop Table 3 and Figure 2 include the number of dwell-
ing units using each fuel and the average heating degree days in each state,
typical heat requirements for an average dwelling unit, and Bureau of Mines
fuel sales data. A complete discussion of the methodology used to develop
Table 3 and Figure 2 can be found in Appendix B. An update of national fuel
consumption values to 1978 has been made and is discussed in Section 5.
Table 3 and Figure 2 show that about 60 percent of residential fuel oil
is burned in the northeast region. This region also contains 37 percent of
the U.S. population.5 Residential gas consumption (natural and LPG) for
space heating is more widely distributed than oil, but is still most heavily
concentrated in the upper midwest and the northeast. States that consume
more, than 5 percent of the U.S. total include Illinois (8.9 percent), New
York (8,3 percent), Ohio (8.1 percent), California (7.8 percent), Michigan
(7.6 percent), and Pennsylvania (6.0 percent).
In 1974 there were 70,831,000 occupied single and multiple dwelling
units in the United'States.6 The majority of these dwelling units, 60,500,000,
were heated by gas or oil. Single unit structures accounted for 50,000,000
units; multiple unit dwellings numbered another 4,000,000. Further analyses
indicate that there were about 34,000,000 gas-fired and 13,000,000 oil-fired
residential .space heating systems in service in 1974. Similar analyses indi-
cate that in 1974 about 740,000 coal-fired and 660,000 wood-fired residential
units were in operation.
Although prediction of fuel use trends is subject to many uncertainties,
consumption of gas and oil for residential space heating is expected to de-
crease 2.9 percent and increase 3.5 percent, respectively, from 1978 to 1985
(see Section 5). The prediction of future trends in the pattern of fuel use
for residential space heating is difficult due to uncertainties in inter-
national oil production, distribution, and pricing; however, it would appear
inevitable that reliance on oil must eventually diminish. Gas heating
15
-------
TABLE 3. U.S. RESIDENTIAL SPACE HEATING FUEL USE BY REGION, 1975, 1015 J/yr*
Region
United States
Northeast
New England
Middle Atlantic
North Central
East North Central
West North Central
South
South Atlantic
East South Central
West South Central
West
Mountain
Pacific
Gas
Natural
3,794
801
119
681
1,686
1,208
478
716
237
135
342
597
231
366
LPG*
316
23
7
16
150
85
63
105
35
31
40
38
22
15
Oil5
2,550
1,515
515
999
593
403
191
310
290
19
1
133
27
105
Coal
Anthracite
64.3
50.8
1.5
49.4
9.6
9.3
0.3
3.5
3.5
0.0
0.0
0.0
0.0
0.0
**
Bituminous
54.9
0.2
0.2
0.0
23.0
19.4
3.6
26.4
9.2
17.5
0.1
5.1
3.9
1.2
Wood§§
49.9
4.6
2.6
2.1
7.9
3.0
5.0
28.6
13.5
10.2
4.9
8.8
2.7
5.9
All
fuels
6,833
2,395
647
1,748
2,468
1,727
741
1,189
589
212
388
781
287
494
*Totals may not agree because of rounding.
f38 x 106 J/m3 (1,022 Btu/ft3).
+25.1 x 109 J/A (90,000 Btu/gal).
§39 x 109 J/£ (140,000 Btu/gal).
#30 x 106 J/kg (26 x 106 Btu/ton).
**27.9 x 106 J/kg (24 x 106 Btu/ton).
§§13.9 x 106 J/kg (12 x 106 Btu/ton).
-------
WEST NORTH CENTRAL
6AS 541
OIL 191
OTHER 8
EAST NORTH CENTRAL
GAS 1293
OIL 403
OTHER 32 MIDDLE ATLANTIC
NEW ENGLAND
GAS 127
OIL 515
OTHER 4
PACIFIC
GAS 381
OIL 106
OTHER 7
GAS 253
OIL 27
OTHER 6
Figure 2. Geographical distribution of residential fuel consumption
for space heating, 1975, 1015 J/yr.
-------
appeared to be enjoying increased popularity, but again its future has been
clouded by the natural gas shortage - a shortage which may be alleviated by
importing LNG. Electric heating is also gaining in popularity, particularly
in those regions where the winters are less severe. This trend shifts emis-
sion problems from the home to the utility. Solar heat for residences is
technically, if not economically, feasible at the present time, but probably
will not perturb fuel use patterns significantly in the near future. The
most crucial unanswered question at the moment is whether or not oil and gas
shortages and/or price increases will initiate a significant trend back to
coal and/or wood for residential heating. Their increased use as residential
fuels will result in increased emissions and require further study.
3.2 POPULATION CHARACTERISTICS
The population distribution of gas- and oil-fired residential sources,
based on the heating medium used, is shown in Table 4. In the size range
less than 0.16 x 109 J/hr, gas-fired heating systems are 76 to 94 percent
warm air furnaces while in the range above 0.16 x 1Q9 J/hr they are 86 per-
cent steam or hot water.7 A similar, but less pronounced, pattern exists for
oil-fired units. New steam or hot water systems are more expensive than warm
air systems and are usually installed only in sizes above 0.14 x 109 J/hr.8
TABLE 4. POPULATION DISTRIBUTION OF RESIDENTIAL UNITS
NOW IN SERVICE, 1974,7 PERCENT
Type of unit
Gas burners
Steam or hot water
Warm air
Oil burners
Steam or hot water
Warm air
0 to 0
6.1
93.9
28.3
71.7
Rated capacity (109
.1 0.1 to 0.16
24.0
76.0
60.5
39.5
J/hr)
0.16 to 0.42
85.6
14.4
46.8
53.2
18
-------
The.size distribution of residential oil-fired sources, based on 1972
data,2 is presented in Table 5. Some changes from 1972 can be anticipated
due to the relative growth of multiunit dwellings. About 69 percent of the
units have fuel firing capacities below 0.2 x 109 J/hr. Heating units in the
New England and middle Atlantic areas tend to be larger than in other areas.
TABLE 5. SIZE DISTRIBUTION OF RESIDENTIAL OIL-FIRED SYSTEMS,
1972,2.PERCENT IN SIZE RANGE BY NUMBER
Size (109 J/hr)
Region
New England
Mid-Atlantic
South Atlantic
Midwest
West
All sections
< 0.16
19.6
24.3
51.8
44.6
75.4
34.6
0.16 to
0.2
43.9
39.7
29.8
27.1
14.2
34.7
0.2 to
0.24
12.8
17.8
9.3
12.8
5.9
13.9
0.24 to
0.3
8.9
8.6
5.7
8.8
3.3
8.0
0.3 to
0.44
7.0
5.7
1.6
4.4
0.9
4.9
> 0.44
7.8
3.9
. 1.8
2.3
0.3
3.9
Information is also available concerning the population and age of oil
burner designs, and this is shown in Tables 6 and 7. High pressure burners
predominate. Trend'information can be inferred from Table 7, indicating a
growth in nonconventional high pressure burners and a decrease in the use of
low pressure, rotary, and vaporizing units. Although data are not available,
members of fuel oil related institutions indicate that the efficient flame
retention burner is now used in the majority of new installations.
Details on the sizes and ages of gas-fired systems are unavailable. It
can be assumed that the size distribution will be very similar to oil-fired
units and that gas-fired systems may tend to be newer.
3.3 CHARACTERISTICS OF COMBUSTION EQUIPMENT
Residential gas- and oil-fired space heating units are subject to a
number of design and operating variations. These variations are related to
burner and combustion chamber design, excess air, heating medium, etc. Resi-
dential systems operate only in an on/off mode with constant fuel firing rate
19
-------
TABLE 6. POPULATION OF RESIDENTIAL OIL BURNER DESIGNS, 19722
Percent of
Region
High
Conventional
pressure
Shell
head
Less than
New England
Mid-Atlantic
South Atlantic
Midwest
West
All sections
New England
Mid-Atlantic
South Atlantic
Midwest
West
All sections
All sections
44
66
74
63
80
63
61
69
89
73
80
71
All
68
.9
.3
.8
.0
.3
.1
0.
.4
.0
.4
.3
.3
.2
7.2
8.4
1.9
9.4
0.6
7.2
15 to 0
7.9
5.2
2.3
13.1
0.6
6.9
oil burners
.3
7.0
Flame
retention
0.15 x
14
3
13
13
5
9
.44 x
13
8
7
6
2
8
up to
8
total
Low
pressure
Rotary
Vaporizing
109 J/hr
.7
.8
.0
.1
.2
.1
26
10
0
5
8
11
.4
.6
.8
.6
.1
.5
5
8
0
8
0
6
.8
.5
.1
.6
.5
.6
1.
2.
9.
0.
5.
2.
0
4
4
3
3
5
109 J/hr
.6
.2
.3
.2
.9
.3
0.44 x
.6
9
10
0
6
13
8
109
9
.0
.7
.8
.0
.2
.6
J/hr
.6
7
6
0
1
1
4
5
.6
.8
.2
.3
.8
.7
.4
0.
0.
-
0.
1.
0.
1.
5
1
1
2
3
1
TABLE
7 . AVERAGE
AGE OF
RESIDENTIAL
OIL
BURNERS ,
19722
Region
New England
Mid-Atlantic
South Atlantic
Midwest
West
All sections
High
Conventional
13
12
10
14
12
12
.1
.4
.1
.1
.7
.6
pressure
Shell
head
9.5
8.9
7.0
8.8
7.0
8.7
Flame
retention
4
4
3
3
3
3
Years
.1
.2
.0
.7
.7
.9
Low
pressure
14
16
19
17
16
16
.4
.2
.3
.8
.6
.5
Rotary
16
17
17
17
.8
.7
-.
.8
-
.5
Vaporizing
15.
13.
17.
16.
15.
6
7
0
7
9
20
-------
during the "on" cycle in contrast to load modulation encountered with larger
commercial, industrial, and utility systems.
The typical oil-fired heating system consists of a burner mounted in a
refractory-lined combustion chamber. Flue gas leaves the combustion chamber
through a water- or air-cooled heat exchanger. The burner system includes a
combustion air blower, fuel pump, spark ignition system, and a fuel nozzle.
Fuel, flow is determined by the fuel nozzle orifice size while combustion air-
flow is determined by the blower characteristics and a damper in the flue gas
duct.7 Steam, hot water, or air from the exchanger is usually forced through
the heating system by a pump or fan, although gravity or natural draft circu-
lation is used in some instances.
Several burner designs are used for oil atomization2 as previously shown
in Table 6. High pressure atomization burners are the most common, repre-
senting 84 percent of total units in 1972 and over 90 percent in 1978. These
units operate by forcing oil at a pressure of 100 psig through a small ori-
fice or orifices in the nozzle. The conventional high pressure burner is the
most common, but newer designs such as the shell head and retention head are
increasing in popularity. These newer burners are designed for,improved com-
bustion efficiency and generally result in lower particulate, HC, and CO
emissions. SO emissions are unchanged, but NO emissions may increase due
X X
to a more intense, compact flame.
Gas burners are simple and relatively maintenance free compared to oil
burners. Most residential gas burners are very similar. They use natural •
aspiration and consist of three to four Venturis with distribution pipes con-
sisting' of rows of small orifices. Primary air is aspirated and mixed with
the gas as it passes through the venturi. Secondary air enters the furnace
around the burners. Flue gases pass through a heat exchanger and a stack.
On gas-fired systems, the stack always contains a draft diverter that pro-
vides dilution air and prevents downdrafts that could blow out the pilot
light.
Typical design and operating characteristics of gas- and oil-fired resi-
dential warm air furnaces are presented in Table 8. The large variation in
excess air levels presented in Table 8 for the gas-fired systems is due to
21
-------
to
N>
TABLE 8. DESIGN AND OPERATING PARAMETERS FOR GAS- AND OIL-FIRED FORCED AIR RESIDENTIAL
SPACE HEATING SYSTEMS,8 HEAT INPUT 1.055 x 108 J/hr
Parameter
Excess air
Flue exit diameter
Heat exchange area
Overall heat transfer coefficient
Exit flue gas temperature
Combustion chamber pressure
Temperature rise on air side
Overall steady state efficiency
Units
Percent
Centimeters
Square meters
Btu/hr/ft2/°F
°C
Newton/meter2
°C
Percent
Gas-fired sources
20
8.9
0.3
2
232
+
20
70
- 500
-12.7
- 3.2
- 3
- 316
50
- 24
- 80
Oil-fired
10 -
12.7 -
1.9 -
2 -
260 -
12 -
24 -
70 -
sources
100
17.8
2.8
3
482
50
27
80
-------
several factors. Because natural gas burners are naturally aspirated, pres-
sures are near atmospheric. Therefore, it is possible to use the natural
draft created by the stack to draw the secondary air into the furnace. The
amount of secondary air will vary with meteorological conditions.
Air pollution control equipment is not installed on residential heating
systems. Excess air, residence time, flame retention devices, and mainte-
nance are major factors in the control of air pollutants and performance of
residential units.^ Proper attention to the above factors can reduce emis-
sions of CO, HC, and particulates. Emissions of NO generally tend to remain
X
unchanged when adjustments are made to improve system efficiency and reduce.
emissions of other pollutants.2'7
One of the major factors that reportedly affects particulate, HC, and
CO emissions from oil-fired residential systems is cycling. Emissions of
particulate, HC, and CO peak during ignition and after burner shutdown.2'7
However, results obtained during this study and discussed in Section 4 did
not confirm these observations for particulates and HC. Emissions of HC and
particulates per unit of fuel were essentially unchanged when both gas- and
oil-fired units were operated in a 10 minute on/20 minute off cycle as com-
pared to a 50 minute on/10 minute off cycle.
23
-------
4. EMISSIONS
Air emissions from the flue gas stack are the only significant emissions
from gas- and oil-fired residential combustion sources used for space heating.
Fugitive emissions from these combustion sources are negligible because the
liquid fuels used have low volatility leading to minimum evaporative losses,
and the gaseous fuels are for the most part received continuously from a
pipe, which must be tightly sealed for safety reasons, rather than via a fuel
storage tank and fuel pump.
Although the two residential combustion source categories considered in
this study are not major contributors to nationwide pollutant emission totals,
they are of potential concern because the sources are numerous and have a
close source-receptor relationship. Both source categories ranked high when
their total impact on the environment was estimated based on the air impact
factor developed by Monsanto Research Corporation (MRC). This air impact
factor provides a mathematical ranking of impacts using a model which takes
into account the number of sources, the associated population density, ground
level concentrations of the pollutants and their environmental hazard poten-
tial. The mathematical expression derived by MRC is shown in Appendix C,
Attachment B.
4.1 EVALUATION OF EXISTING EMISSIONS DATA
4.1.1 Criteria for Evaluating the Adequacy of Emissions Data
A major task in this program has been the identification of gaps and
inadequacies in the existing emissions data base for combustion sources.
The results of this effort determine the extent of the sampling and analysis
program required to complete an adequate emissions assessment for each of
the combustion source types.
The criteria for assessing the adequacy of emissions data are developed
by considering both the reliability of the data and the variability of the
24
-------
data. A detailed presentation of the procedures used to identify and evaluate
emissions data is given in Appendix C. Briefly, the general approach for iso-
lated sources is to use a three-step process. In the first step, the avail-
able data are screened for adequate definition of process and fuel parameters
that may affect emissions as well as for validity arid accuracy of sampling
and analysis methods. .In the second step of the data evaluation process,
emission data deemed acceptable in Step 1 are subjected to further engineering
and statistical analysis to determine the internal consistency of the test
results and the variability in emission factors. The third step in the pro-
cess uses a method developed which is based on both the potential environ-
mental risks associated with the emission of each pollutant and the quality
or variability of the data. The potential environmental risks associated with
pollutant emissions are determined by the use of a source severity factor
which is defined as the ratio of the calculated maximum ground level concen-
tration of the pollutant species for a typical source to the level at which
a potential environmental hazard exists. If the variability of emission
factor data is greater than 70 percent, then the need for further measurement
will be based on calculated severity factors for each pollutant.. The data
will be considered adequate if the upper bound of the source severity factor
is less than 0.05 even if the variability is greater than 70 percent.
Severity factors calculated by MRC from nationwide emission estimates1*
for the residential gas- and oil-fired systems considered in this report are
all less than 0.05 as shown in Table 9.1(^ Normally, as just noted, no further
testing would be required. However, in the case of source types such as those
considered in this report, which consist of a large number of small sources,
additional criteria based on MRC calculated air impact factors (see Appendix C)
are used to determine the need for further testing. Application of the crite-
ria described in Appendix C indicated that further testing was required. The
environmental significance of emissions from residential sources was evaluated
following the measurement program by use of multiple source severity factors
using a dispersion model to determine the ambient concentrations from an array
of typical residential sources.
25
-------
TABLE 9. MEAN SEVERITY FACTORS FOR GAS- AND OIL-FIRED
RESIDENTIAL SOURCES10
Pollutant
Particulates
SO
X
NO
X
HC
CO
Benzo(a)pyrene
Gas-fired sources
< 0.0001
< 0.0001
0.0010
0.0001
< 0.0001
0.0001
Oil-fired sources
0.0001
0.0026
0.0016
0.0001
< 0.0001
0.0001
4.1.2 Sources of Existing Emissions Data
Emissions data for residential gas- and oil-fired heating units have
been obtained by the-EPA,9 Battelle,2.11>12 TRC,13.14 KVB,15 Walden Re-
search,16^7 and the Public Health Service.18 Most of the data are for the
five criteria pollutants (particulates, S02, NO , HC, and CO), although some
X
measurements of aldehydes, trace metals, and POM emission rates were made.
POM measurements were made for both gas- and oil-fired systems, trace metals
were measured only for oil-fired units, and aldehydes were measured only for
gas-fired systems. Bacharach smoke numbers were also taken for most of the
oil-fired systems which were tested. Additional data are now being gathered
by other investigators working under EPA sponsorship.
The tests made by TRC, KVB, Walden Research, and the Public Health
Service were field tests on burners in the "as found" condition. The EPA
data were obtained in laboratory tests of burners.which were well tuned.
Battelle performed three studies (1971,n 1973,2 and 197412), two of which
(1971 and 1973) involved field testing. In these studies, emission rates
were measured before and after tuning. The other study (1974) was done in
the laboratory.
4.1.3 Existing Emissions Data; Gas-Fired Sources
The data used to evaluate the status of the existing data base are pre-
sented in Appendix D for gas- and oil-fired residential sources. A summary
of'the data is presented below along with the conclusions drawn from evalua-
tion of the adequacy of the data.
26
-------
Emission rates of particulates, S02, NO , HC, CO, and aldehydes from
X
gas-fired sources were measured by TRC,13*14 Battelle (1971),n EPA,9 and
KVB.15 One hundred units were tested by TRC, two units by Battelle, four .
units by EPA, and thirty-one units by KVB. Data from all studies were com-
bined and mean emission rates, the standard errors and variability of the
means, and the severity factors calculated. This information along with
EPA emission factors3 is presented in Table 10. There are significant dif-
ferences in the results of the various studies used to compile Table 10,
particularly in the case of particulates, HC, and CO emissions. Emissions
of S0£ and NO were in reasonable agreement with EPA emission factors.
X
TABLE 10. SUMMARY OF EXISTING EMISSIONS DATA FOR GAS-FIRED
RESIDENTIAL SOURCES
Pollutant
Particulates^
SO 2
NO
X
HC
CO
Number
of
units
101
9
133
98
105
Mean
(ng/J)
0.41
0.21
56.6
39.4
38.7
Standard
error
(ng/J)
0.10
0.15
4.1
5.8
8.3
TTPA
Variability Severity emission
(%) factor .factor3
(ng/J)
49
164
15
29
43
8
3
5
2
8
x 10~6 2-6
x 10~6 0.26*
x 10~3 33
x 10~3 3.3
x 10-6' 8.4
. Based on a firing rate of 99 x 106 J/hr.
Filterable particulate.
Based on a fuel sulfur content of 4,600 g/106 Nm3.
Most of the available particulate emissions data is for filterable par-
ticulate. The mean emission factor of 0.41 ng/J is based almost entirely on
data obtained by TRC, using nonstandard sampling methods, and is a factor of
5 to 10 lower than the Battelle data and the EPA emission factor. TRC fur-
ther reported that filterable particulates made up only about 2 percent of
total particulate emissions. This is in contrast to the result obtained by
Battelle in a single test. Battelle reported that about one-third of the
total particulate was filterable. Further testing should be conducted to
27
-------
resolve the above differences. In the case of HC and CO emissions, TRC values
exceeded those measured by EPA and Battelle by a factor of 10 and 4, respec-
tively. The discrepancy cannot be attributed to the test methods used, as was
true of the particulate emission values, and further testing should be con-
ducted, particularly in the case of HC because of its.higher severity factor.
The Public Health Service measured the emissions of POM from three resi-
dential gas-fired heating units, as shown in Table 11. Samples were obtained
by passing the flue gas through two water impingers at 6 C, a series of freeze-
out traps at -17 C, and a high-efficiency filter. The samples were then ex-
tracted with benzene and separated by chromatography. Concentrations of
several specific compounds were measured by ultraviolet-visible spectroscopy.
The mean emission factor for total benzene-soluble organics was 2,150 pg/J.
Measures of variability such as standard errors are not shown in the table.
Since there is only one measurement for three source types, no meaningful
measure of variability is possible.
TABLE 11. POM EMISSIONS FROM GAS-FIRED RESIDENTIAL SOURCES,18 pg/J
Heating unit
Pollutant
Double shell boiler,
0.19 GJ/hr input
Hot air furnace,
0.22 GJ/hr input
Wall space heater,
0.03 GJ/hr input
Total benzene-
soluble organics
Benzo (a)pyrene
Pyrene
Benzo (e)pyrene
Perylene
Benzo (ghi)pyrene
Anthanthrene
Coronene
Anthracene
Phenanthrene
Fluoranthene
900
< 0.02
0.16
ND
ND
ND
ND
ND
ND
ND
0.30
620
< 0.02
0.11
0,02
ND
ND
ND
ND
ND
0.07
0.10
4,920
0.26
15.2
1.40
ND
2.20
0.07
0.79
ND
ND
7.60
ND - Not Detected.
28
-------
Significant quantities of trace elements are not present in natural gas,
and measurements are not justified except for purposes of confirmation.
4.1.4 Existing Emissions Data: Oil-Fired Sources
The results of tests in which emission rates of total particulates, SC>2,
NO , HC, and CO were measured for oil-fired units are presented in Appendix D.
X .
Information about the condition of the units tested, such as age, efficiency,
excess air, and stack gas compositions, is also presented. The tests were
carried out by Battelle and EPA. Data from Walden Research is also shown,
although no criteria pollutant data were obtained in their study of oil-fired
source operations.
Data from the EPA and Battelle studies were combined. The mean emission
factors calculated from these tests are shown in Table 12. Standard errors
and variabilities of the mean, severity factors, and EPA emission factors are
also shown in Table 12.
TABLE 12. SUMMARY OF EXISTING EMISSIONS DATA FOR OIL-FIRED
RESIDENTIAL SOURCES
Pollutant
Particulates1"
S02
NO
x
HC
CO
Number
of
units
35
20
58
59
61
Mean
(ng/J)
24.9
93.0
43.8
10.2
47.4
Standard
error
(ng/J)
5.2
7.1
3.0
6.1
16.0
Variability
42
15
13
121
67
*
Severity
factor
1.4 x 10~3
3.8 x 10~3
1.1 x 10~2
1.4 x 10~3
3.1 x 10~5
EPA
emission
factor3
(ng/J)
7.7*
106§
55
3
15
t
§
Based on a firing rate of 298 x 106 J/hr.
Total particulate.
'Filterable particulate.
AP-42 gives the S02 emission factor as 142 S lb/1000 gal, where S is the
fuel sulfur content. In tests of 20 residential fuel oils, Battelle
(1971) found the average sulfur content to be 0.24 percent. This content
was used to calculate the emission factor.
29
-------
As shown in Table 12, the average emission factors for S02 and NO are
, X
in reasonable agreement with EPA emission factors. In the case of the par-
ticulate emission data, which was obtained by Battelle, the average emission
factor is three times greater than the EPA emission factor of 7.7 ng/J. How-
ever, filterable particulate data obtained from the same units averaged about
40 percent of the total particulate. Battelle's emission factor estimate for
filterable particulate is, therefore, about 8.8 ng/J. The measured HC emis-
sion factor shown in Table 12 is three times greater than the EPA emission
factor. The average value, however, is highly influenced by the inclusion of
data from two units tested by Battelle. If these two data points are not in-
cluded in the summary, the mean emission factor drops to about 2 ng/J. Simi-
larly in the case of CO emissions, the exclusion of four high emission values
from the data summary reduces the mean emission factor to 17 ng/J, a value
close to the EPA emission factor of 15 ng/J.
The Public Health Service measured the emissions of POM from three resi-
dential oil-fired heating units, as shown in Table 13.
TABLE 13. POM EMISSIONS FROM OIL-FIRED RESIDENTIAL SOURCES,18 pg/J
Heating unit
Pollutant
Cast iron boiler,
0.25 GJ/hr input
Hot air furnace,
0.15 GJ/hr input
Hot air furnace,
0.09 GJ/hr input
Total benzene-
soluble organics
Benzo(a)pyrene
Pyrene
Benzo(e)pyrene
Perylene
Benzo (ghi)pyrene
Anthanthrene
Coronene
Anthracene
Phenanthrene
Fluoranthene
7,700
< 0.04
1.70
ND
ND
ND
ND
ND
ND
8.40
4.70
3,400
< 0.06
0.01
ND
ND
ND
ND
ND
ND
ND
0.07
3,300
< 0.11
1.10
ND
ND
ND
ND
ND
ND
ND
14.20
ND - Not Detected.
30
-------
For the measurements shown in Table 13, the Public Health Service ob-
tained samples by passing the flue gas through two water impingers at 0 C,
a series of freeze-out traps at -17 C, and a high-efficiency filter. The
samples were then extracted with benzene and separated by chromatography.
Concentrations of several specific compounds were measured by ultraviolet-
visible spectroscopy. The mean emission factor for total benzene-soluble
organics was 4,800 pg/J. Variability data have not been calculated because
of the limited number of tests.
In their 1971 study, Battelle measured the emission rates of some trace
metals in particulate matter for one burner in the "as found" condition
(Unit 26, Appendix D). Particulates were collected using a revised version
(no cyclone was used) of EPA Test Method 5. The probe wash, impinger wash,
and filter catch were analyzed for trace metals using optical emission spec-
trometry. Emission rates were determined for all of the metals detected:
iron, boron, silicon, magnesium, manganese, lead, nickel, aluminum, copper,
calcium, chromium, barium, bismuth, cobalt, potassium, tin, vanadium, silver,
sodium, zinc, and titanium. The emission rates are presented in Table 14.
Since only one burner was tested, it is impossible to calculate variability
of the data.
4.1.5 Status of Existing Emissions Data Base for
Gas- and Oil-Fired Residential Sources
In summary, the evaluation of the adequacy of existing emissions data
on the basis of variability and severity factor criteria for gas- and oil-
fired residential heating units has led to the findings shown in Table 15.
4.2 EMISSIONS DATA ACQUISITION
4.2.1 Selection of Test Facilities
Because residential combustion sources are associated with areas of high
population density, operate at relatively low efficiency, and emit pollutants
at essentially ground level, emissions from these combustion sources have been
regarded with some concern. The concern is largely due to the inadequacy of
the previously described existing emissions data base. The combination of the
potential for significant air pollution impact and data inadequacy led to a
decision to test five gas-fired and five oil-fired residential sources. This
31
-------
TABLE 14. TRACE METAL EMISSIONS FROM AN OIL-FIRED BURNER11
Element
Fe
B
Si
Mg
Mn
Pb
Ni
Al
Cu
Ca
Cr
Ba
Bi
Co
K
Sn
V
Ag
Na
Zn
Ti
Emission factor (pg/J)
Probe
0.
0.
1.
0.
< 0.
< 0.
< 0.
0,
0.
0.
< 0.
0.
wash
49
10
97
68
10
68
10
68
30
98
10
10
ND
< 0.
< 0.
0.
< 0.
< 0.
< 0.
< 0.
0.
10
98
10
10
10
98
98
20
Filter catch
5
.53
< 0.98
0
0.49
< 0
7
0
14
< 0
1
< 0
< 0
< 29
.40
.37
.49
.74
.98
.47
ND
.98
.40
ND
.5
ND
ND
ND
ND
ND
ND
Impinger wash
9
1
24
9
0
1
0
14
2
29
0
1
.89
.97
.57
.83
.30
.97
.68
.74
.98
.80
.98
.97
2
0
9
1
3
1
2
ND
2
9
2
< 0
0
67
2
0
.98
.83
.98
.10
.10
.56
.98
.80
3
10
0
67
3
Total
15
.0
26
11
.3
.2
.2
30
.0
32
.0
.0
< 0
.0
.0
3
< 0
.1
.5
.0
1
.9
- 3.0
.5
.0
- 0.8
- 9.8
- 1.3
.2
- 4.0
.3
- 1.1
- 3.0
.40
- 3.1
- 40.0
.1
.2
- 0.11
- 68.5
- 4.0
.0
Severity
factor
0.0018
10-"
0.018
10-"
10-"
0.0075
10-"
10-"
10-"
10~3
0.0024
10~3
-
.0.003
0.003
10-"
io-5
ID"3
0.006
10-"
10- 5
ND - Not Detected.
32
-------
TABLE 15. SUMMARY OF EVALUATION OF EXISTING EMISSIONS DATA
10
Co
Pollutant
Gas-fired sources
Particulates
S02
NO
x
HC
CO
POM
Trace elements
Oil-fired sources
Particulates
S02
NO
x
HC
CO
POM
Trace elements
f Variability (%) Severity factor Comments concerning
, - ^ (criterion: < 70) (criterion: < 0.05) data requirements
0.4
0.2
56.6
39.4
38.7
2.2
24.9
93.0
43.8
10.2
47.4
4.8
0-30 pg/J
49
164
15
29
43
•
42
15
13
121
67
'
—
8 x 10~6 Further tests required due to
data inconsistency
3 x 10~6 No further tests required
5 x lO"3 No further tests required
2 x 10~3 Further tests required due to
data inconsistency
8 x 1Q~6 No further tests required
- Further tests required
No significant quantities in
gaseous fuel, test for
confirmation only
1.4 x 10~3 No further tests required
3.8 x ID"3 NO further S02 data required,
but 803 data are lacking
1.1 x 10~2 No further tests required
1.4 x 10~3 No further tests required
3.1 x 10~5 No further tests required
Further tests required
0.018 - 10~5 Further tests required for
confirmation
-------
is the maximum number of test sites being studied by modified Level I sampling
and analysis for any source category in the total program. The basic Level I
sampling and analysis scheme is shown in Figure 3.
Following a review of the initial data obtained from testing of the 10
selected residential sites, it was decided that further testing was warranted
to resolve questions concerning certain aspects of the test program. A major
concern was the selection of the 50 minute on/10 minute off cycle used at all
10 residential test sites, as opposed to a 10 minute on/20 minute off cycle
used in other studies9'11 of residential emissions. These studies have re-
ported that hydrocarbon emissions peak during startup and shutdown of the
burner. The fivefold increase in the number of cycles required for the 10/20
cycle should result, therefore, in an increase in emissions. Further, it was
postulated that, because of lower furnace and flame temperatures, the 10/20
cycle could lead to increased organic emissions and considerably greater POM
emissions. The effect of cycle mode on organic emissions was subsequently
measured at one gas-fired and two oil-fired sites. Additional testing in-
volved Level II analysis for S02, 803, and particulate sulfate emissions from
the oil-fired sites, and parameter checks of all test units to determine if
their operations were within the normal range of residential furnace operation.
All of the oil-fired units were serviced during the months of September and
October prior to both of the two test periods. One unit, gas-fired site 103,
which had emitted high levels of C}-C3 organics during the first series of
tests, was found to be defective, and these C^-C3 values are riot included in
estimations of organic emission rates and severity factors presented in this
report.
The choice of specific sites is based on the representativeness of the
sites as measured against important characteristics of systems within each
source category. Typical values for important characteristics; e.g., burner
type and age, firing rate, and types of systems, were obtained from the data
base (see Section 3). The gas-fired units selected for testing were warm air
furnaces which comprise over 90 percent of units in the size range tested
(< 1.0 x 108 J/hr). Both warm air and forced hot water oil-fired systems
were tested. All of the oil-fired systems used conventional high pressure
burners, the type of burner used in 68 percent of oil-fired home heating
34
-------
Ul
PARTICIPATE
MATTER
S02/S03
SOURCE -»l
OPACITY
(STACKS)
pAC
PROBE AND
RINjC-i rYTnnrr PHYSICAL SEPARATION
1 » rioeANTr<; INTO FRACTIONS:
UKhANlLb LR/IR/MS
SASS TRAIN GAS
f CONDITIONER FI FMFNT<; K iunDraNir<: ELEMENTS (SSMS) AND
> 10w — * EXTRACTION » INORGANICS SELECTED ANIONS
1 - 3p* — p .
SAME AS ABOVE
FILTER " *
NOX CHEMILUMINESCENCE
(GRAB) CHROMATOGRAPHY > INORGANI
ORGANIC IS^nLro . rv-rn.rTTnM , ORGANIC
MATCDTAI r ADSORBER, ' — * EXTRACTION * r r
MATERIAL > C6 HOOULE RINS£ C7-C16
ORGANIC ON-SITE GAS ^ ORGANIC
MATERIAL Cj-C6 CHROMATOGRAPHY . * > C16
, TNORMNirs ELEMENTS (SSMS) AND
INORGANIC^ SELECTED ANIONS**
r— : 1 PHYSICAL SEPARATION
1 — » ORGANICS INTO FRACTIONS:
1 ' LR/IR/MS
P, ELEMENTS (SSMS) AND
Li SELECTED ANIONS
. ALIQUOT FOR GAS
CHROMATOGRAPHIC
ANALYSIS
PHYSICAL SEPARATION
s INTO FRACTIONS:
LR/IR/MS
Weigh Individual catches.
*
If inorganics are > 105 of total catch.
Figure 3. Basic Level I sampling and analytical scheme for participates and gases.
-------
systems. As will be discussed in the results section, all of the units tested
exhibited acceptable combustion characteristics, as determined by the param-
eter checks, with the exception of sites 103, 129, and 326. High CO values
were measured at these sites.
The manufacturer, rated capacity, and age of the units selected for test-
ing are shown in Table 16. While it is common practice in the case of oil-
fired sources to replace burner nozzles at frequent intervals, often on a
yearly basis, burner age is still important to provide some indication of the
age of moving parts and parts exposed to high temperatures.
A.2.2 Field Testing Procedures
Field testing procedures were based on Level I environmental assessment
methods. The SASS train was used to collect particulate, organic, and trace
metal samples at the exit of the stack. This sampling train (Figure 4) is a
high volume (approximately 0.14 m3 per minute) system designed to extract
particulates and gases from the stack, separate particulates into four size
fractions, trap organics in an adsorbent, and collect volatile trace metals
in liquid solutions. A high volume system is required to collect adequate
quantities of trace materials for subsequent laboratory analyses. The train
is constructed such that all sample contacting surfaces are of type 316 stain-
less steel, Teflon, or glass.
The residential combustion tests were carried out without the cyclones
in the SASS train due to the low concentrations of particulates and their
characteristic small particle diameters. The particulates were collected on
Spectrograde® glass fiber filters in the heated oven. The sample stream was
then cooled and the organic material collected by adsorption on the XAD-2
resin (a styrene, divinylbenzene copolymer). The gas then passed through an
impinger containing a hydrogen peroxide solution to collect oxidizable con-
stituents. A second impinger with a solution of ammonium peroxydisulfate and
silver nitrate and a third impinger containing ammonium perosydisulfate solu-
tion were used to collect any volatile trace elements not collected in up-
stream SASS sections. A fourth impinger containing silica gel was used to
remove the remaining moisture from the sample stream.
36
-------
TABLE 16. CHARACTERISTICS OF RESIDENTIAL COMBUSTION UNITS TESTED
Combustion
source Site no.
type
Gas-fired 100/127
101/128
102/125
103/126
104
129/130/131
Oil-fired 300/300-2
301/301-2
302/302-2
303/303-2
304
326-1/326-2
327-1/327-2
Manufacturer
Gaffers & Sattler, Model 580
Gaffers & Sattler, Model 580
Day and Night, Model C 80 U,
Series 510
Sears Homart, Model 735
Sears Homart, Model 735
Rheem, Model 3204-80
American Standard,
Model MF-A
American Standard,
Model W0391
General Electric No. 5.KI,
ME-1 Burner
U Labs No. L8626 LD
Embassy Steel Products
Arcoliner No. W.O. 351,
Series 3AS3
Armstrong, Model L61-95 A527
Burner type
Conventional
Conventional
Conventional
Convent ional
Conventional
Conventional
Conventional
high pressure
Conventional
high pressure
Conventional
high pressure
Conventional
high pressure
Conventional
high pressure
Conventional
high pressure
Conventional
high pressure
Heating
medium
Forced air
Forced air
Forced air
Forced air
Forced air
Forced air
Forced
hot water
Forced
hot water
Forced air
Forced air
Forced air
Forced air
Forced air
Rated .
capacity
(Btu/hr)
80,000
80,000
80,000
100,000 '
75,000
80,000
182,000
175,000
78,000
112,000
87,000
141,000
120,000
Age
(yr)
14
8
12
20
6
15
27
25
21
5
16
16
6
1 Btu/hr is equivalent to 1,055 J/hr.
-------
STACK T.C.
OJ
co
CYCLONES
FUJER | GAS COOLER
I "
GAS
TEMPERATURE
*—' CONVECTION A
_ _S&__OVEN_ L-L-J
DRY GAS METER ORIFICE METER
CENTRALIZED TEMPERATURE
AND PRESSURE READOUT
CONTROL MODULE
OVEN
T.C.
XAD-2
CARTRIDGE
CONDENSATE
COLLECTOR
IMP/COOLER
TRACE ELEMENT
COLLECTOR
IMPINGER
T.C.
10CFM VACUUM PUMPS
Figure 4. SASS schematic.
-------
Samples of the flue gas were obtained using a 1-meter probe with a 1.6-
centimeter nozzle at a single traverse point approximating the average flow
rate of the flue gas, as determined by a multipoint traverse. Sample time
for the SASS train was from 4 to 6 hours as required to obtain a total sample
volume of 30 cubic meters or greater. Additional sampling time was required
to obtain 90 cubic meters of sample volume in the case of the five oil-fired
sites initially tested. The longer sampling time was required to provide
samples to EPA for additional analyses. Sampling locations for an oil-fired
residential unit are shown in Figure 5.
Samples of the flue gas were collected for on-site analyses of flue gas
constituents using a stainless steel probe, condenser, diaphragm pump and gas
sampling bags. 'The collected samples were injected into a gas chromatograph
through a heated gas sampling valve for analysis. Low molecular weight hydro-
carbons with boiling points below 100 C were measured in the field using a
flame ionization detector. Other flue gas constituents, C02, 02, and CO were
measured using a thermal conductivity detector. Standard mixes of the gases
were used for calibration.
In the second series of tests conducted in 1978, measurement of combus-
tion parameters was conducted at both the stack and furnace locations shown in
Figure 5 in order to improve characterization of furnace operating conditions.
Combustion parameter measurements of C02, 02, CO, and smoke spot numbers were
conducted at both the previously tested sites and the new sites selected for
the study of burner duty cycle. CO detection tubes were used in these tests
because GC detection limits were well above the normal CO concentration range
of the combustion gases. Smoke spot numbers were determined at the furnace
exit location using a Bacharach smoke spot tester.
The two oil-fired residential heating units tested in 1978 (sites 326 and
327) were sampled at the stack exit for S02, 803, and particulate sulfate using
a Level II controlled condensation system (CCS), which is a Goksoyr-Ross pro-
cedure modified by TRW. The CCS sampling train, shown in Figure 6, consists
of a quartz nozzle; a heated quartz probe liner; a heated quartz filter holder
containing a quartz filter; a Pyrex modified Graham condenser; three impingers,
the first containing 3 percent hydrogen peroxide, the second empty, and the
third containing silica gel; a dry gas meter; and a leakless pump.
39
-------
SAMPLING PORTS
• SASS TRAIN
• BAG SAMPLES
• GOKSOYR ROSS TRAIN
STACK EXTENSION
•CHIMNEY WALL
-STACK
BAROMETRIC DAMPER
SAMPLING PORT FOR PARAMETER CHECK
FURNACE
Figure 5. Sampling locations for oil-fired residential sources.
40
-------
ADAPTER FOR CONNECTNG HOSE
TCWELL
ASBESTOS CLOTH
INSULATION
GLASS-COL
HEATING
MANTLE
STACK
RUBBER VACUUM
HOSE
DRY TEST
METER
PUMP
THREE WAY
VALVE
SILICA GEL
EMPTY
REQRCULATOR
THERMOMETER
Figure 6. Schematic of controlled condensation system.
-------
For sampling, the nozzle is oriented into the gas stream and sample gas
is withdrawn at a constant rate (0.3 liter/minute). The gas temperature is
kept above the dewpoint of sulfuric acid through the probe and filter holder
to ensure that only particulate sulfate is filtered. The condenser, however,
is maintained at a temperature below the acid dewpoint by a heated circulating
water bath which causes the 863 in the sample gas to condense out as sulfuric
acid. Sulfur dioxide remains in the gas phase. Subsequently, the S02 is
withdrawn from the sample gas in the solution of the first impinger where it
is oxidized to sulfate. Sample recovery consists of combining the filter and
the particulate recovered from the probe for measurement of particulate sul-
fate, flushing the condenser coil several times with distilled deionized water
for recovery of §03, and packaging the contents of the first impinger for SC>2
analysis.
Sample recovery, including precleaning and handling of sample containers,
of SASS and Goksb'yr-Ross train components was carried out according to Level I
procedures and specifications.1
4.2.3 Laboratory Analysis Procedures
The laboratory analysis procedures used to characterize the residential
sources sampled during this portion of the program are summarized below. In
general, the analytical scheme of lERL's Level I Procedures Manual1 was fol-
lowed. More details are given in "Emissions Assessment of Conventional Sta-
tionary Combustion Systems: Methods and Procedures Manual for Sampling and
Analysis," EPA-600/7-79-029a, January 1979.
4.2.3.1 Inorganic Analysis—
The inorganic analytical scheme consisted of a SSMS elemental survey
analysis for the determination of some 70 elements and Atomic Absorption
Spectrometric (AAS) analysis for mercury, arsenic, and antimony. The ana-
lytical scheme is depicted schematically in Figure 7.
As shown in Figure 7, the flue gas sampled by the SASS train produced
four laboratory samples for inorganic analysis. The particulate filter (PF),
the XAD-2 resin (XR), and a composite sample (CH) containing portions of the
XAD-2 module condensate, the HNOjj rinse of the module, and the first impinger
42
-------
CO
>
i V
PARTICULATE
FILTER
\
SASS TRAIN
SAMPLES
V
' \ ' \ f \f
XAO-2
RESIN
AQUA RE6IA
EXTRACTION
>
i
f
PARR BOMB
COMBUSTION
\
^ > '
CONDENSATE H „ | 1 1
AND NITRIC * f ' 1 1
ACID RINSE IMPIN6ER j j IMPIN6ERS j
\> \t
J^
... , V ! ,. .._ ( AAS }
(SSMS j (AAS )
AAS
SSMS
• LEVEL B ANALYSIS
Figure 7. Level. I inorganic analysis methodology for residential sources.
-------
were analyzed by AAS and SSMS. AAS analysis was also conducted on the second
and third impinger composite sample (CI). In addition, for the oil-fired
sources, the fuel feed was analyzed for trace metal composition.
Figure 7 also indicates the sample preparation steps which were used to
prepare the samples for laboratory analysis. For the primarily organic mate-
rials, XAD-2 resin, and fuel oil, a Parr oxygen bomb combustion of the sample
was employed in order to eliminate the organic matrix. The particulate fil-
ters, were extracted with aqua regia to facilitate analysis. No preparative
steps were necessary for the impinger composite samples.
The complete analytical scheme was implemented for the five gas-fired and
five oil-fired sources initially tested. A review of the resulting data indi-
cated that no significant trace metal emissions are produced by the gas-fired
sources. . The data analysis also showed that the trace metal emissions from
oil-fired sources can be adequately characterized by analyzing the fuel feed.
Therefore, in the later series of tests of residential sources, inorganic
characterization of the oil-fired sites was restricted to a fuel feed analysis.
No inorganic analyses were conducted for the gas-fired sources sampled in 1978.
The two residential oil-fired sites sampled to evaluate cyclic effects
were also tested for SC>2, SC>3, and particulate sulfate using the Goksb'yr-Ross
sampling train. This Level II procedure was instituted because of inadequa-
cies in the data base with regard to 803 and particulate sulfate emissions.
Brief descriptions of the analytical techniques used for the inorganic
characterizations are provided below.
• SSMS — SSMS was used in the laboratory to perform a
semiquantitative elemental survey analysis on all types
of Leyel I samples. The analysis was performed using a
JEOL Analytical Instruments, Inc., Model JMS-01BM-2 Mass
Spectrograph. The JMS-01BM-2 is a high resolution, double-
focusing mass spectrometer with Mattauch-Herzog ion optics
and ion sensitive photoplate detection. The instrument is
specially designed to carry out high sensitivity trace
element analysis of metals, powders, or semiconductor type
materials using an RF spark ion source. Elemental analysis
by SSMS involves the incorporation of a sample aliquot into
44
-------
two conducting electrodes which are decomposed and subse-
quently analyzed by a mass determination using a double-
focusing mass spectrometer. Decomposition of the sample
electrodes is accomplished by the application of a radio
frequency (~1 MHz) potential of about 4 kV. This induces
an electrical discharge in the form of a spark plasma.
Because of the high energy associated with the discharge,
the spark plasma created is composed primarily of elemental
species. The positively charged ions contained in the plasma
are accelerated and formed into an ion beam by a high poten-
tial electric field (-30 kV). The beam is then energy-
focused and momentum-dispersed to produce a mass spectrum
which is recorded by an ion sensitive photoplate.
SSMS can be used to detect elemental species contained in
the sample electrodes at levels down to 10~9 grams. Al-
though, the sensitivity varies somewhat, depending on the
element of interest and the sample type, practically all
elements in the periodic table can be detected. Using
photoplate detection, all elements having masses in the
range 6 to 240 can be detected simultaneously. Concentra-
tion data are derived from the intensities (optical density)
of the mass spectral lines. There are several methods for
determining concentration data from photoplate spectral line
densities. The methods vary widely in terms of their com-
plexity and corresponding precision and accuracy of the
results. The photoplate interpretation procedures followed
for this program and for Level I survey work in general are
designed to yield concentration data accurate to within a
factor of two for 70 elements.
Mercury - Cold Vapor — The cold vapor mercury analysis is
based on the reduction of mercury species in acid solution
with stannous chloride and the subsequent sparging of ele-
mental mercury, with nitrogen, through a quartz cell where
its absorption at 253.7 nm is monitored.
Arsenic - Hydride Evolution — This procedure entails the
reduction and conversion of arsenic to its hydride in acid
solution with either stannous chloride and metallic zinc
or sodium borohydride. The volatile hydride is swept from
the reaction vessel, in a stream of argon, into an argon- '
hydrogen flame in an AAS. There, the hydride is decomposed
and the arsenic concentration is monitored at its resonance
wavelength 193.7 nm. Excess hydrogen peroxide and nitric
acid present in certain Level I samples interfere with the
analysis and must be removed prior to the addition of either
the zinc slurry .or sodium borohydride used to generate the
arsenic hydride.
45
-------
• Antimony - Hydride Evolution — Antimony-containing
compounds are decomposed by adding sulfuric and nitric
acids and evaporating the sample to fumes of sulfur
trioxide. The antimony liberated is subsequently re-
acted with potassium iodide and stannous chloride, and
finally with sodium borohydride to form stibine. The
stibine is removed from solution by aeration and swept
by a flow of nitrogen into a hydrogen diffusion flame
in an AAS. The gas sample absorption is measured at
217.6 nm. Since the stibine is freed from the original
sample matrix, interferences in the flame are minimized.
• Goksb'yr-Ross Analysis for SO?, SOa, and Particulate
Sulfate — Four separate analyses are done for the
Goksb'yr-Ross samples. The particulate matter, which
includes filter and probe rinse particulates, is ana-
lyzed for sulfate. The filter is allowed to desiccate
overnight and is weighed. The probe rinse (acetone) is
evaporated to dryness in air and the particulate weight
recorded. This is then combined with the filter and
extracted with hot water. The hot water extract is
analyzed and reported as water soluble sulfate. Any
solids remaining from the hot water extraction are
extracted in hot hydrochloric acid; the extract is
analyzed and reported as water insoluble sulfate.
The coil rinse (water) is analyzed by an acid-base
titration against 0.02 N sodium hydroxide which has
been standardized against primary standard potassium
acid phthalate. The results are reported as mg 803.
The peroxide impinger is analyzed by titration with
standard barium perchlorate. Prior to the titration,
sodium carbonate >is added to bring the pH into the
range of 8.9, and the sample is boiled to remove per-
oxide. The results are reported as mg SC>2.
4.2.3.2 Organic Analysis—
Level I organic analysis is designed to provide a semiquantitative (± 3)
determination of the classes and concentrations of organic substances con-
tained in waste streams emitted by stationary energy and industrial processes,
In general, three .categories of organic compounds are defined according to
their boiling points (BP) by Level I: gaseous, volatile, and nonvolatile.
Gaseous organics boiling below 100°C, which are measured in the field, have
been discussed earlier. Volatile organics are defined as those which boil
between 100° and 300°C; nonvolatile organics boil above 300 C. Organic
46
-------
analyses were performed on all SASS train components except the impingers.
All stainless steel components were rinsed with methylene chloride or a 50/50
(v/v) mixture of methylene chloride/methanol to recover organics. Organics
in the condensate, particulate filter, and XAD-2 resin were recovered by
methylene chloride extraction.
Sample collection and laboratory analyses were performed during two
different time periods (summer 1977 and spring-summer 1978). In the interim,
procedural changes were made in the Level I sampling and analytical methods.
These changes are described, as appropriate, in the following discussion of
specific procedures and analyses.
The laboratory analysis scheme and decision criteria for residential
sources are depicted in Figure 8. All organic liquids and solvent extracts
were first concentrated to 10 m£ in a Kuderna-Danish (K-D) evaporator, with
the concentrated samples then analyzed in two stages. The first stage of
the analysis consisted of four different methods. A sample aliquot was
evaporated to.dryness and weighed. The residue was then taken up in methy-
lene chloride, transferred to salt plates, the methylene chloride evaporated,
and its IR spectrum scanned between 2.5 and 15 microns by a grating IR spec-
trophotometer. The output of these steps was, respectively, a measure of the
amount of nonvolatile organic matter (> 0^5) present in each sample and an
indication of the functional groups present.
Another sample aliquot was injected into a gas chromatograph (GC). The
instrument was calibrated so that the organic compounds boiling between 100
and 300 C (i.e., 67 to Gig, the total chromatographable organics or TCO) were
quantified relative to n-decane. If the TCO was greater than 75 yg/m3, the
quantity of organics boiling within specific ranges was also determined. For
samples from the initial five gas-fired and five oil-fired sources (sites 100
through 104 and 300 through 304), compounds boiling in the range 110 to 220°C
(C8 to C12) were determined. In the later tests (sites 129, 326, and 327),
procedure modifications were made in order to quantify organics over the full
100 to 300°C BP range.
47
-------
r~
ORGANIC
ANALVSIS
1 LIQUID SAMPLES 1
i
SOLID SAMPLES |
1 1
1
SASS TRAIN
SOLVENT RINSES
1 ml ALIQUO
FOR GC-TCO
AND GC/MS
METHOD 1
TOTAL ORGAN I CS
> 500 uj/m]
TCO > 10% TOTAL
I
ALIQUOT FOR LC
9 ma TO 100 ng
UP TO 8 ml
SOLVENT
EXCHANGE
AND TRANSFER
TO COLUMN
I 1
CONDENSATE
1
PARTICULATE XAD-2
FILTER RESIN
1 1
CHiCl.
EXTRACTION
i
J TCO INPUT (
ufl/ms
.XQUAN
/ nr T
"\ ORGA
^•^ug
METHOD 2
TOTAL 0
> SOD
TCO « 1
ALIQUOT
9 ma TO
UP TO
GRAV INPUT AND IR
riTVX.
1TAI ^v
1ICS. yS
$/
RGANICS • TOTAL ORGAN I CS
uQ/m9 < 500 ug/m3
l\_^
FOR LC (STOP )
100 mg \_J
6ml
•
EVAPORATE
AND TRANSFER
TO COLUMN
TCO * GRAV * IR
GRAV + IR
FRACTION
> SOO ug/m> OR OF
SPECIAL INTEREST
SPECIAL FRACTIONS
LEVEL II ANALVSIS
Figure 8. Level I organic analysis methodology for residential sources,
48
-------
As shown in Figure 8, the samples were not analyzed further if the total
quantity of volatile and nonvolatile organic emissions (the sum of the TCO
emissions and the gravimetric components) was less than 500 yg/m3.
In the second stage of analysis, those samples with organic emissions
greater than 500 yg/m3 were fractionated by LC. (The LC method used provides
some separation of components according to polarity.) The fractions were
analyzed by the gravimetric and IR methods described previously. If the un-
fractionated sample contained more than 10 percent volatile organic material,
fractions 1 through 7 were also analyzed for TCO components. Fractions which
contained more than 15 mg of material or which were of special interest were
analyzed by low resolution mass spectroscopy (LRMS). LRMS is an instrumental
technique which may provide molecular weights and compound identification on
a "most probable" basis for samples of low complexity. In Level I analysis,
it is used to.supplement the compound classification derived from IR spectra.
Finally, aliquots from each sample concentrate were analyzed by GC/MS for POM.
Brief descriptions of the analytical techniques used in conducting the
Level I organic analysis and the GC/MS analysis for POM are presented below.
• Extraction of Aqueous Samples — These liquid/liquid
extractions were performed with standard separatory
funnels. Whenever necessary, the pH of the sample was
adjusted to neutral with either a saturated solution
of sodium bicarbonate or ammonium chloride. The sample
was extracted three times with a volume of high-purity
methylene chloride equal to approximately 5 percent of
the sample volume. The volume of the resulting extract
was measured and concentrated.
• Extraction of Solid Samples — The particulate filters
and XAD-2 resin samples from the SASS train were ex-
tracted in appropriately sized Soxhlet extractors.
Each sample was placed in a glass thimble and extracted
for 24 hours with Distilled-in-Glass® or Nanograde®
purity methylene chloride. The resulting extracts were
measured and concentrated.
• K-D Concentration — The solvent extracts of solid and
liquid samples and the solvent rinses of sampling hard-
ware were concentrated in K-D evaporators. Heat pro-
vided by a steam bath was sufficient to volatilize the
solvents with minimal loss of other organic components.
49
-------
All samples were concentrated to a volume between
5 and 10 mil, allowed to cool, transferred to a
volumetric flask, and diluted to a final volume
of 10 m£ with methylene chloride.
Gravimetric Determination — The weight of nonvola-
tile organic species was determined on the concen-
trates obtained from the K-D concentration of
solvent extract and rinse samples. The samples
were transferred to tared aluminum weighing dishes,
evaporated at ambient temperature, and stored in a
desiccator to constant weight. Weights of organic
residues as small as 0.1 mg were measured.
IR Analysis — IR analysis was used to determine the
functional groups in an organic sample or LC fraction
of a .partitioned sample. The interpreted spectra pro-
vide information on functionality (e.g., carbonyl,
aromatic hydrocarbon, alcohol, amine, aliphatic hydro-
carbon, halogenated organic, etc.). Compound identi-
fication is possible only when that compound is known
to be present as a dominant constituent in the sample.
The minimum sample amount required for this analysis
is 0:5 mg. A compound must be present in the sample
at 5 to 10 percent (w/w) at least for the character-
istic functional groups of a compound to appear suf-
ficiently strong for interpretation. Organic solvents,
water, and some inorganic materials cause interferences.
Water, in particular, decreases the resolution and
sensitivity of the analysis.
The initial organic sample concentrate or LC fraction,
after evaporation, was either (1) taken up in a small
amount of carbon tetrachloride or methylene chloride
and transferred to a NaCl window, or (2) mixed with
powdered KBr, ground to a fine consistency, and then
pressed into a pellet. A grating IR spectrophotometer
was used to scan the sample in the IR region from 2.5
to 15 microns.
TCP Analysis — GC was used to determine the quantity
of low boiling hydrocarbons (BP between 100 and 300°C)
in the K-D concentrates of all solvent rinses and or-
ganic extracts and in LC fractions 1 through 7 (when
the volatile organics were greater than 10 percent of
the total organics in the unfractionated sample).
Whenever the TCO concentration exceeded 75 yg/rn3, fur-
ther GC analysis was conducted to determine the amount
of individual species.
50
-------
The extent of compound identification was limited to
the representation of materials as normal alkanes
based upon comparison of boiling points. The analy-
sis is semiquantitative because only one hydrocarbon,
n-decane, is used for calibration. The differences
in instrument response, or sensitivity, to other
alkanes are well within the desired accuracy limits
for Level I analysis and are not taken into consider-
ation in data interpretation.
LC Separation — This procedure was designed to
separate samples into eight reasonably distinct
classes of compounds and was applied to all organic
samples which contained a minimum of 500 yg/m3 of
combined volatile (TCO) and nonvolatile (gravimetric)
organics. A sample weighing from 9 to 100 mg was
placed on a silica gel liquid chromatographic column,
and a series of eight eluents of sequentially increas-
ing polarity was employed to separate the sample into
eight fractions for further analyses. As the use of
HC1 in the final eluent results in partial degradation
of the column material, data were derived from only
the first seven fractions.
Two distinct methods were used to prepare samples for
LC fractionation and subsequent analysis. The selec-
tion of "Method 1" or "Method 2" (Figure 8) was based
on the results of gravimetric and TCO determinations
on the concentrated organic sample. Method 1 was used
whenever the volatile organic content determined by the
TCO analysis was in excess of 10 percent of the total.
Method 2 was used whenever the TCO was low - less than
10 percent of the total.
In Method 1, the low boiling components must be pre-
served for LC separation and subsequent analysis.
This requires a solvent exchange step to transfer the
sample from methylene chloride to the nonpolar solvent
hexane before placement on the column. In Method 2,
where there are few volatile components, a simple,
direct solvent evaporation step is sufficient to pre-
pare the sample for fractionation. Gravimetric and
IR analyses were performed on the first seven frac-
tions of all LC separations. In addition, whenever
Method 1 was used, a TCO analysis was also performed
on each of the seven fractions.
LRMS — This procedure is a survey analysis used to
determine compound types in an organic sample or in
an LC fraction of a sample. The analyst is specif-
ically searching for hazardous compounds or compounds
51
-------
which may be generally considered toxic; e.g., aromatic
hydrocarbons and chlorinated organics. Analysis using
different sample ionizing parameters results in molecular
weight data which, combined with IR and sample source
data, can provide specific compound identifications on
a "most probable" basis.
The mass spectrometer (MS) used in this procedure has
sufficient sensitivity such that 1 nanogram or less
presented to the ionizing chamber results in a full
spectrum with a signal ratio of 10:1. A dynamic range
of 250,000 is achievable. The detection limit for a
specific compound related to the size of an air sample
or liquid sample varies widely depending on the types
and quantities of the species in the mixture. This is
because of interfering effects in the spectrum caused
by multiple compounds. The impact of this interference
is reduced by lowering the ionization voltage to pro-
duce spectra containing relatively more intense molecu-
lar ions.
Solid samples are placed in a sample cup or capillary
for introduction via the direct insertion probe. More
volatile samples are weighed into a cuvette for intro-
duction through a batch or liquid inlet system. The
probe or cuvette is temperature programmed from ambient
temperature to 300°C. Periodic MS scans are taken with
a 70 eV ionizing voltage as the sample is volatilized
during the program. A lower ionizing voltage range
(10 to 15 eV) can be used at the discretion of the
operator if the 70 eV data are complex. Spectra are
interpreted using reference compound spectral libraries,
IR data, and other chemical information available on
the sample. The results of LRMS analysis give quali-
tative information on compound types, homologous series
and, in some cases, identification of specific compounds.
This information is then used to assess the hazardous
nature of the sample.
GC/MS Analysis for POM — This is a combined GC/MS
method for qualitative and quantitative POM determina-
tions. Microliter quantities of concentrated sample
extracts are used for this analysis. This technique
is classified as a Level II procedure.
Microliter sized samples are injected onto a GC column
and are separated by the differences in the retention
characteristics between the sample components and the
column material. As the components elute from the
column, they are transported via an instrument inter-
face to the MS, which is being operated in a Total
Ion Monitoring (TIM) mode.
52
-------
In the MS, the various compounds are ionized and all
ion fragments in the mass range of 40 to 400 amu are
monitored. The resulting mass spectra are stored by
the computerized data system. All compounds eluting
from the GC in detectable quantities could be identi-
fied, including aromatic compounds containing hetero-
atoms, depending upon the desired scope of the analysis.
The computer was used to search the stored spectra for
the specific mass fragments shown in Table 17.
TABLE 17. MASS TO CHARGE VALUES
MONITORED, m/e
128 180 242
154 184 252
162* 192 256
166 202 278
178 216 300
179 228 302
*
Internal Standard -
Chloronaphthalene.
The spectra of POM's are quite distinctive because they
yield very strong molecular ions with little fragmenta-
tion. Using molecular ions to find POM's in a mixture
involves reconstructing the GC trace from the stored data
using only a single mass to charge (m/e) value. Any in-
flection in this mass chromatogram indicates the possi-
bility of a POM of that molecular weight. The spectrum
is then displayed and the operator judges if the spectrum
is consistent with a POM. The GC retention time as well
as the spectrum is used to make this identification, al-
though it is often difficult to confirm which isomer is
causing a peak without standards for the specific material.
Using this technique, a large number of POM's can be
screened in a short period of time and good identification
of POM type is possible. More time is'required for exact
identification. Table 18 lists POM's which are sought in
all samples; any POM with a molecular weight on this list
will be determined. If other POM's with different molecu-
lar weights are desired, all that is needed for their iden-
tification is the molecular weight and a relative retention
time or a standard. The compounds listed in Table 18 rep-
resent essentially all POM compounds within the molecular
weight range of 128 to 302. Many of these compounds have
been identified from previous studies of combustion sources.
53
-------
TABLE 18. MINIMUM LIST OF POM's MONITORED
Compound name Molecular weight
Naphthalene 128
Biphenyl 154
Benzindene 166
Fluorene 166
Phenanthrene 178
Anthracene 178
Benzoquinoline 179
Acridine 179
9,10-dihydro-phenanthrene 180
9,10-dihydro-anthracene 180
1-Methyl-fluorene 180
2-Methyl-fluorene 180
9-Methyl-fluorene 180
2-Methyl-phenanthrene 192
3-Methyl-phenanthrene 192
2-Methyl-anthracene 192
Fluoranthene 202
Pyrene 202
Benzo(a)fluorene (1,2-benzofluorene) 216
Benzo(b)fluorene (2,3-benzofluorene) 216
Benzo(c)fluorene (3,4-benzofluorene) 216
2-Methyl-fluoranthene 216
1-Methyl-pyrene 216
3-Methyl-pyrene 216
4-Methyl-pyrene 216
(continued)
54
-------
TABLE 18 (continued).
Compound name Molecular weight
Benzo(c)phenanthrene 228
Benzo(ghi)fluoranthene 228
Benzo(a)anthracene 228
Chrysene (Benzo(a)phenanthrene) 228
Triphenylene (9,10-Benzophenanthrene) 228
4-Methyl-benzo(a)anthracene 242
1-Methyl-chrysene 242
6-Methyl-chrysene 242
Benzo(b)fluoranthene 252
Benzo(f)fluoranthene . 252
Benzo(k)fluoranthene 252
Benzo(a)pyrene 252
Benzo(e)pyrene 252
Perylene 252
7,12-Dimethy1-benzo(a)anthracene 256
9,10-D.imethyl-benzo (a) anthracene 256
Benzo(c)tetraphene 256
1,2,3,4-Dibenzanthracene 278
2,3,6,7-Dibenzanthracene 278
Benzo(b)chrysene 278
Picene (3,4-Benzochrysene) 278
Coronene 300
Benzo(ghi)perylene 302
1,2,3,4-Dibenzpyrene 302
1,2,4,5-Dibenzpyrene 302
55
-------
During the search of the data for POM compounds, non-POM
compounds may interfere, especially if they coelute with
a POM. Computer data interaction techniques, such as ion
mapping, kept these interferences to a minimum. If a POM
was confirmed, the peak was quantified using an internal
standardization method.
The GC/MS sensitivity varies with several parameters,
including the type of compound, internal instrument
cleanliness, resolution of closely eluting peaks, etc.
Under "everyday" operating conditions, 20 nanograms (ng)
eluting in a peak about 5 seconds wide yields an MS sig-
nal with a usable signal-to-noise ratio. Typically, this
represents at least 100 yg of any single POM compound in
a concentrated extract of a sample.
A. 2. 4 Test Results
A. 2. 4.1 Field Measurements and Emissions of Criteria Pollutants
and S02, SOa, and Particulate Sulfate —
Field data for all units tested are shown in Tables 19 and 20. Measure-
ments of gaseous (boiling below 100°C) hydrocarbon, also made in the field,
are reported later with the emission data for volatile (100 to 300?C BP range)
and nonvolatile (BP greater than 300 C) organics measured in the laboratory.
Combustion parameter data are shown for two locations: at the stack exit
where emissions were sampled, and at the location near the exit from the fur-
nace. The gas composition data at the furnace exit location were obtained
during the second test period to assess the operating condition of the units
tested. Excess air levels at the furnace exit, as determined from C02 and ©2
measurements, ranged from 50 to 175 percent for the gas-fired units and from
50 to 150 percent for the oil-fired units. Normal excess air values are re-
portedly 20 to 500 percent and 10 to 100 percent for gas- and oil-fired sys-
tems, respectively. 8 Two of the gas-fired units, sites 127 and 129, show
exceptionally high CO emission levels; and one oil-fired unit, site 326, also
shows much higher than normal CO concentrations. Generally, CO levels of the
magnitude measured would indicate highly inefficient combustion and lead to
heavy soot formation and hydrocarbon emission levels significantly higher than
those measured at these sites. Thus, the CO measurements at these sites appear
invalid.
56
-------
TABLE 19. FIELD DATA: GAS-FIRED RESIDENTIAL COMBUSTION SOURCES
Ul
Unit
No.
1
1
2
2
3
3
4
4
5
6
6
6
Test
No.
100
127f
101
128f
102
1251"
103
126f
104
129f
130
131
Cycle,
on/off
50/10
-
50/10
-
50/10
-
50/10
-
50/10
-
50/10
10/20
Date
i
6/10/77
3/13/78
5/24/77
3/13/78
5/26/77
3/10/78
6/08/77
3/17/78
6/14/77
3/13/78
3/14/78
3/15/78
(106 J/hr
85
85
85
85
85
85
106
106
79
85
85
85
CO 2
6.4
9.3
1.4
7.9
3.0
4.4
1.7
5.7
1.1
5.4
4.1
3.5
02
16.7
9.7
12.9
11.3
19.5
14.3
19.1
11.6
16.8
10.5
15.8
14.3
CO
(ppmv)*
ND
1220
ND
0
ND
16
ND
0
ND
920
ND
ND
Particulate
emissions
(mg/m3)
0.46
ND
0.40
ND
0.49
ND
0.64
ND
0.62
ND
ND
ND
(ng/J)
0.55
ND
0.26
ND
1.7
ND
1.7
ND
0.76
ND
ND
ND
NOX
emissions
(ppmv)
53
ND
11
ND
119
ND
66
ND
21
ND
ND
ND
(ng/J)
31
ND
6
ND
66
ND
33
ND
11
ND
ND
ND
Bacharach
smoke
number
_
0-1
_
0-1
—
-
-
0-1
-
1
-.
-
ppmv values at 3 percent 02.
Parameter check at exit from heat exchanger; other tests at stack exit.
ND - Not Determined.
-------
TABLE 20. FIELD DATA: OIL-FIRED RESIDENTIAL COMBUSTION SOURCES
Ui
oo
Unit
No.
1
1
2
2
3
3
4
4
5
6
6
7
7
Test
No.
300
300-21"
301
301-21"
302
302-21"
303
303-2f
304
326-1
326-2
327-1
327-2
Cycle,
on/off
50/10
-
50/10
-
50/10
-
50/10
-
50/10
50/10
10/20
10/20
50/10
Date
5/18/77
4/24/78
5/24/77
4/19/78
5/27/77
4/17/78
6/02/77
4/18/78
6/29/77
5/23/78
5/24/78
5/31/78
6/02/78
UO6 J/hr
192
192
185
185
83
83
118
118
92
149
149
127
127
CO 2
3.8
6.5
3.7
7.2
1.2
7.3
2.9
10.4
2.6
3.1
6.6
8.3
1.5
°.2
17.2
14.3
17.4
11.4
19.6
14.6
17.3
7.5
17.5
14.5
13.9
12.8
15.0
C0 *
(ppmv)
ND
40
ND
< 18
ND
8
ND
< 2
ND
620
565
< 2
< 3
Particulate „ , ,
Bacharach
emissions • ,
smoke
(mg/m3)
2.3
ND
1.2
ND
2.2
ND
2.5
ND
1.9
1.7
1.6
1.9
1.5
(ng/J)
3.1
ND
1.7
ND
8.3
ND
3.5
ND
2.8
1.3
1.3
1.2
1.3
number
_
1
_
1
_
1
_
1
-
2
2
1
1
ppmv values at 3 percent 02.
Parameter check at exit from heat exchanger; other tests at stack exit.
ND - Not Determined.
-------
As discussed previously in section 4.1, the existing data base was found
to be adequate for 802, NOX, and CO emissions for both gas- and oil-fired
residential heating sources. Additional S02, NOX, and CO measurements were,
therefore, unnecessary for the construction of the emissions data base.
Nevertheless, NOX emissions were determined using a theta sensor for the five
gas-fired units initially tested.
Particulate emissions from oil-fired units exceed those from gas-fired
.units. Further, cycle mode did not affect particulate emissions from the two
oil-fired units tested. The average NO emission factor of 33 ng/J for the
X
.five gas-fired systems is identical to the EPA emission factor. S02 emis-
sions, although not measured, would be equivalent to 106 ng/J for the oil-
fired sites, based on an average fuel sulfur content of 0.24 .percent. The
emission factor for S02 from the gas-fired sources would be 0.26 ng/J, based
on a sulfur content of 4600 g/106 Nm3.
Sulfur component analyses of oil-fired sites 326 and 327, obtained using
the Goksb'yr-Ross sampling train and Level II analysis, determined that an
average of 95 percent of fuel sulfur was emitted as S02, 4.4 percent as 50%,
and 0.55 percent as particulate sulfate. 803 conversions measured were 6.5
and 4.5 percent for the two tests at site 326 and 2.1 and 4.4 percent for the
two tests at site 327. The 803 emissions are greater than those normally en-
countered' in larger combustion systems (1 to 3 percent), but similar results
have been noted in other studies.19*2" Experimental error due to overtitra-
tion was proposed as the reason for the apparently high 803 concentrations
observed for low sulfur fuel oils.19 The analytical procedure used in this
study differed from that used by KVB19 and does not appear to be the cause of
the higher than expected 803 emissions. However, the results appear to be
within the normal range of values found by Goksoyr-Ross analysis of 803 emis-
sions from combustion sources.2" Further study will be needed to determine
if the high S03 emissions are a result of normal analytical data scatter or
are a real effect resulting from high excess air levels.
The data reduction procedures for converting emission concentrations
(ppmv or mg/m3) to emission factors (ng/J) are based on calculation of the
combustion of fuel with air, as described in detail in Appendix E.
59
-------
4.2.4.2 Inorganic Analysis Results—
Trace element data, obtained by SSMS analysis, are contained in Appen-
dix F in 11 tables. Tables F-l to F-5 contain the results from the gas-fired
sites; Tables F-6 to F-ll provide results from the oil-fired sites. These
tabulated results are presented for up to 65 elements for each section of the
SASS train analyzed and are summed to provide a total value and to calculate
emissions. However, in the case of the oil-fired sites, the fuel was also
analyzed. Based on the assumption that the total elemental content of the
fuel is emitted with the flue gas, mass emissions for these sites were deter-
mined from the fuel analytical results.
Trace element emissions from the gas-fired sites were low. Only about
25 elements were positively detected. With the exception of two obvious out-
liers; i.e., Si in site 100 and Cu in site 102, all identified trace element
emission levels were less than 0.025 mg/m .
Trace element emissions from the oil-fired sites were greater than those
from the gas-fired sites. Several elements (Al, Ca, Mg, Ni, K, and Na) were
emitted in quantities ranging from 0.1 to 0.45 mg/m^, based on the fuel con-
tent analysis by SSMS. Emissions were calculated using the methods shown in
Appendix E.
A summary of the data for specific inorganic analyses for mercury,
arsenic, and antimony, as determined by AAS, is given in Table 21. Arsenic
and antimony emissions were also calculated from the results of SSMS analysis.
Arsenic emission values determined by SSMS were roughly a factor of five
higher than those obtained by AAS for the oil-fired sites. Antimony values
were in reasonable agreement; however, mass emissions of all three elements
as measured by AAS are quite low.
4.2.4.3 Organic Analysis Results—
- Total Organic Matter—A summary of the organic analyses is presented in
Table 22. As noted previously, the GC determination of hydrocarbons boiling
in the same range as methane (Ci) through n-hexane (Cg) was done in the field
while the analysis of hydrocarbons boiling in the range of n-heptane (Cy)
60
-------
TABLE 21. SUMMARY OF AAS RESULTS FOR Hg, As, AND Sb
Combustion
source
type
Gas-fired
Average
Oil-fired
Average
Site No.
100
101
102
103
104
300
301
302
303
304
Mass
Hg
0.00003
0.00025
0.003
0.004
0.001
0.0076
0.0007
< 0.0008
< 0.0008
< 0.0008
< 0.0005
< 0.00072
emissions (mg/m3)
As
< 0.0002 <
0.0004 <
< 0.0002
< 0.0002 <
< 0.0002 <
< 0.00025 <
< 0.00025 <
< 0.00025
< 0.0001 <
< 0.00025 <
< 0.00025
< 0.00022 <
Sb
0.0007
0.0009
0.004
0.0007
0.0007
0.0014
0.0015
0.0017
0.0016
0.002
0.0033
0.002
61
-------
TABLE 22. TOTAL ORGANIC EMISSIONS FROM GAS- AND OIL-FIRED RESIDENTIAL COMBUSTION SOURCES
Gas-fired sources
Volatile Organlcs
Analyzed in Field
(ug/mj)
Cl
C2
C3
C^
. cs
C6
Volatile Organlcs
Analyzed la Laboratory
(ug/m3)
C7 (BP 90-110°C)
Ce (BP 110-140°C)
C9 (BP 140-160°C)
CIQ (BP 160-180°C)
Cn (BP 180-200°C)
C12 (BP 200-220°C)
Cia (BP 220-240°C)
C;1( (BP 240-260°C)
C15 (BP 260-280°C)
Ci6 (BP 280-300°C)
C7-Ci6 (BP 90-300°C)
NOD volatile Orgaoics
Analyzed in Laboratory
(ug/m*)
> C16 (BP > 300°C)
Total Organics (mg/m3)
100
LD
LD
LD
LD
LD
LD
ND
< 10
< 10
1490
560
60
ND
ND
ND
ND
ND
400
2.5
101
LD
LD
LD
LD
LD
LD
ND
< 10
< 10
< 10
< 10
< 10
ND
ND
ND
ND
ND
1240
1.2
102
LD
LD
LD
LD
LD
LD
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
780
0.8
103*
39,400
37,800
5,500
LD
LD
LD
ND
< 10
< 10
770
600
< 10
ND
ND
ND
ND
ND
480
1.9
104
LD
LD
LD
LD
LD
LD
ND
< 10
< 10
1960
2680
370
ND
ND
ND
ND
ND
920
5.9
130
LD
LD
LD
LD
LD
LD
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
117
340
0.5
131f
LD
LD
LD
LD
LD
LD
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
140
304
0.4
300
9600
1700
LD
LD
LD
LD
ND
6
5
46
25
1
ND
ND
ND
ND
170
650
12.1
301
9720
1900
4400
LD
LD
LD
ND
1
3
14
7
33
ND
ND
ND
ND
180
1800
11.0
302
2000
400
LD'
LD
LD
LD
ND
8
9
26
44
45
ND
ND
ND
ND
560
420
2.8
Oil-fired
303
1800
LD
LD
LD
LD
LD
ND
1
15
43
47
65
ND
ND
ND
ND
560
1300
3.7
304
2800
140
LD
LD
LD
LD
ND
5
7
43
49
81
ND
ND
ND
ND
320
1210
9.3
sources
326-1
LD
LD
LD
LD
LD
LD
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
260
1900
2.2
326-2'
LD
LD
LD
LD
LD
LD
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
610
1200
1.8
327-r
LD
LD
LD
LD
LD
LD
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
385
370
0.7
327-2
LD
LD
LD
LD
LD
LD
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
235
620
0.9
values not included in total.
'10/20 cycle; all other tests 50/10 cycle.
LD - Lower than detection linit: 100 to 1000 'ug/m3 for C)-C6; 1 ug/m3 for C7-C16.
ND - Not determined.
-------
through n-hexadecane (Cig) was performed in the laboratory. The sensitivity
of the GC in the laboratory is about a thousand times greater than the sensi-
tivity of the instrument taken into the field. The corresponding limit of
detection for the field GC is about 100 to 1000 yg hydrocarbon/m3, whereas
the laboratory GC can detect 1 yg/m3. This is the principal reason for the
limited data in Table 22 for Cj through Cg organics measured in the field.
The nonvolatile organic content of the samples was also determined in the
laboratory, using the procedure for gravimetric analysis.
There is a large variation in the C\ to Cg concentrations found in the
oil-fired systems. Sites 300 to 304 all show the presence of C\ to C$ organ-
ics. Only one gas-fired site (103) contained Cj to C$ organics in excess of
the GC detection limit. An examination of this site revealed a defective
unit, and the large values reported in Table 22 for GI to C$ from site 103
are a result of this defect. The emission levels detected should not be con-
sidered representative of gas-fired sources and are not included in subse-
quently determined emission rates.
The concentration of materials boiling between 110 and 220°C (C& to Ci2)
varied from < 10 to 5000 yg/m3 in the samples from the gas-fired sources and
from 60 to 185 yg/m3 in the samples from the oil-fired sources. The nonvola-
tile organics (> Cjg) from the gas-fired sites showed less variability than
the volatile fraction, ranging from 300 to 1240 yg/m3. The variability of
nonvolatile emissions from the oil-fired sites was within reasonable limits;
370 to 1900 yg/m3.
The organic content of SASS train samples from the gas-fired sites aver-
aged 1880 yg/m , while the average from the oil-fired sites was 1430 yg/m3.
Total organics, however, were greater on the average for the oil-fired sources
(4.9 mg/m3) than those measured for the gas-fired sources (1.9 mg/m3), due to
higher gas phase organic emissions from the oil-fired sources.
Nonvolatile organics from both gas- and oil-fired units were found on the
walls of the XAD-2 resin module as well as in the resin, in contrast to the
For the initial five gas-fired and five oil-fired residential sources, the
laboratory GC procedure measured only CQ, C9, CIQ, Cn, and C12 individually,
and total Cy through CIG-
63
-------
volatile components which were collected almost entirely by the resin. Fur-
ther details on the distribution of organics collected by the SASS train are
given in Appendix F, Tables F-12 through F-15.
Total organic emissions as measured did not show a dependence upon cycle
mode. This result is in contrast to findings of other investigations and is
discussed in the section on Analysis of Test Results.
Organic Component Analysis—Further characterizations were conducted on
SASS samples containing high levels of organic material. Because most of the
organics were.found in the XAD-2 resin sample and in the condensate and module
rinse combination sample, these samples were subjected, as required by the
analysis strategy, to LC fractionation with subsequent volatile and nonvola-
tile mass determinations and IR analyses of the LC fractions.
LC fractionation—For gas-fired sources, SASS components which con-
tained > 240 yg/m3 of nonvolatile organics were analyzed by LC fractionation.
Table 23 shows the results of gravimetric analysis of the resulting LC frac-
tions. All XAD-2 resin samples from the oil-fired sources were fractionated,
and the results of both TCO and gravimetric analyses of the fractions are also
shown in Table 23. (An explanation of the sample identification codes is
given in Figure 9.) There is no definite trend as to the quantity of organics
contained in the LC fractions. Aliphatic and aromatic hydrocarbons are eluted
in fraction 1; while fractions 5, 6, and 7 contain such polar species as
esters and other carboxylic acid derivatives, aldehydes and ketones, phenolics
and amines.
IR analysis—IR analyses were not conducted on the gas-fired SASS train
samples because of phthalate ester contamination from plastic components used
throughout the laboratory. Corrective measures have since been successful in
eliminating this source of interference. While no data are available, a gen-
eral interpretation of the LC fraction gravimetric data (Table 23) can be
made by comparison with the theoretical types of compounds which are usually
present in LC fractions.
An interpretation of IR spectra obtained from the LC fractions of the
XAD-2 resin extracts from the oil-fired sites is shown in Table 24. The
classes of compounds identified apply only to the nonvolatile (> Cjg) portion
64
-------
TABLE 23. ORGANIC ANALYSIS RESULTS OF LC FRACTIONATION OF SASS SAMPLES
FROM GAS- AND OIL-FIRED RESIDENTIAL SOURCES, yg/m3
Ul
Source
category
Gas-fired
Oil-fired
Site
No.
101
102
103
300
301
302
303
304
Sample
code
XR
CDMR
XR
CDMR
CDMR
XR
XR
XR
XR
XR
Analysis
Grav
Grav
Grav
Grav
Grav
TCO
Grav
Total
TCO
Grav
Total
TCO
Grav
Total
TCO
Grav
Total
TCO
Grav
Total
LC-1
150
<. 0.8
110
< 0.5
< 0.8
43
88
131
68
30
98
91
1
92
< 0.5
266
266
250
2250
2500
LC-2
66
< 0.8
< 0.4
< 0.5
100
36
< 0.5
36
16
23
39
62
2
64
10
40
50
27
140
167
LC-3
23
12
38
< 0.5
61
17
< 0.5
17
15
35
50
48
6 -
54
84
288
372
23
280
303
LC-4
18
430
23
280
520
15
24
39
20
15
35
84
4
88
135
152
287
19
400
419
LC-5
2
110
76
100
< 0.8
17
107
124
36
23
59
179
10
189
155
152
307
5
280
285
LC-6
87
120
110
72
< 0.8
45
91
136
14
402
416
101
17
118
144
342
486
4
730
734
LC-7
44
140
< 0.4
16
150
< 0.2
157
157
9
65
74
< 0.6
250
250
' 29
19
48
400
400
Total
391
815
350
474
843
173
467
640
165
593
758
565
290
• 855
557
1059
1616
328
4480
4808
-------
SITE I DENT
IFICATION
SAMPLE TYPE
Consecutively numbered Codes and corresponding
by sampling team: sample types are as
- —
AAA-iA-AA-AA ^—
SAMPLE PREPARATION INORGANIC ANALYSIS
1
ORGANIC ANALYSIS
Codes and corresponding Codes and corresponding Codes and corresponding
preoaration steos are procedures are as procedures are as
follows:
100-199 - TRW West Coast FF -
200-299 - TRW East Coast PR -
300-399 - GCA
XR -
MR -
m -
CO -
HM -
HI -
CH -
AI -
PF -
1C -
FC -
3C -
IOC -
CC -
Fuel feed (oil)
Solvent probe/
cyclone rinse
XAD-2 resin
Solvent XAD-2
module rinse
XR extract plus MR
Condensate from
XAD-2 module
HN03 XAH-2
module rinse
H202 inpinger
CD plus HM olus HI
APS impingers
Particulate
filter(s)
l-3u cyclone
PF plus 1C
3-10u cyclone
> lOu cyclone
3C plus IOC
0
LE
SE
KD
A
B
PB
HW
AR
as follows: follows:
- No preparation SS - SSHS
- Liquid/liquid AAS - Hg, As, Sb
extraction JQ - SO =
- Soxhlet extraction NQ ^Q -
- K-D concentration ~p _ ^- p-
- Acidified aliquot
- Basified aliquot
- Parr bomb combustion
- Hot water extraction
- Aqua regia extraction
follows:
First Level Second Level
GC - C7-Ci6 GC Resulting LC fractions
GI Grav IR are numbered in
" ' order 1-7 for:
®* " GC/MS GC - C7-C16 GC
LC - LC separation 6I _ Grav-> IR
MS - LRHS
Figure 9. Sample identification and coding for residential sources.
-------
TABLE 24. CLASSES OF COMPOUNDS IDENTIFIED IN IR SPECTRA OF XAD-2 RESIN LC FRACTIONS
FROM OIL-FIRED RESIDENTIAL SOURCES
Site
300
-
301
303
304
327-1
327-2
LC-1
Alkanes
Aliphatics
Aliphatics
Aliphatics
and
aromatics
Aliphatics
and aryl
ketones
Aliphatics
LC-2
_
Aliphatics
and halogen
substituted
aliphatics
Substituted
aliphatics
Aliphatics
and
aromatics
Aliphatics
Aliphatics
LC-3
_
Substituted
aliphatics
Substituted
aliphatics,
ethers, esters
and ketones
Substituted
aliphatics,
ethers, esters
and ketones
Aliphatics
Aliphatics
LC-4
Substituted
aliphatics;
R-CN
Substituted
aliphatics
Substituted
aliphatics,
ethers, esters
and ketones
Substituted
aliphatics,
ethers, esters
and ketones
Esters and
ketones
Unsaturated
or aryl
esters and
ethers
LC-5
Substituted
aliphatics
and ethers
Substituted
aliphatics
Ethers,
esters and
ketones
Carbonyl
compounds
and
aromatics
Esters and
ketones
Unsaturated
or aryl
esters and
ethers
LC-6
Substituted
aliphatics
and alcohols
Aliphatics
and/or aromatic
esters, ketones
and phenols
Ethers, esters
and ketones
Carbonyl
compounds
and
aromatics
Esters, ketones;
possible ethers,
carboxylic
acids, alcohols
and phenols
Esters, ketones,
amides , amines ,
alcohols and
phenols
LC-7
Substituted
aliphatics,
carboxylic
acids
Aliphatic
acids
Aliphatic
acids
Substituted
aromatics,
aliphatics,
carboxylic
acids
Carboxylic
acids,
alcohols
and phenols
Unsaturated
or aryl
esters
-------
of the LC fractions. The IR spectra were obtained on a spectrophotometer
which, although in conformance with Level I requirements, was insufficiently
sensitive to allow detailed evaluation of the classes of compounds present.
GC/MS Analysis for POM—For the gas-fired sources, only the XAD-2 resin
extracts contained amounts of POM's which exceeded the detection limits of
the analysis (0.3 yg/m3). Naphthalene was found in resin samples from sites
100 and 102 and its source is believed to be the result of leaching of mate-
rial from the XAD-2 resin.
The results of the GC/MS analyses for POM's from the oil-fired sources
are given in Table 25. Compounds not listed or for which no values are
listed were below the detection limit. As expected, most of the detectable
POM's were found in the XAD-2 resin samples, and the compounds found were
relatively low molecular weight species. The compounds detected were .not
.particularly hazardous. The most hazardous POM constituents; e.g., benzo(a)-
pyrene and benzo(a)anthracene, were not detected. Cycle mode did not have a
measurable effect on POM emissions.
4.3 ANALYSIS OF EXISTING DATA AND TEST RESULTS
4.3.1 Emissions of Criteria Pollutants and SO?, 50$, and
Particulate Sulfate
The particulate, NO , and total organic emission factor data resulting
A
from this emissions assessment program for gas- and oil-fired residential
combustion sources are presented in Table 26. As shown in this table, data
variability is large for both the gas- and oil-fired systems for all pollu-
tants, reflecting the inherent variability in emissions from residential
sources and the semiquantitative nature of Level I analysis.
A comparison of the new data with existing data and EPA emission factors3
is presented in Table 27. For the gas-fired units, the measured particulate
emission is over twice as large as that given in the literature but only one-
half that of the emission factor obtained by Battelle.2 The Battelle data
were used by EPA to modify its emission factor values for many of the criteria
pollutants emitted by gas-fired residential systems. The measured emission
factor for NOV is identical to EPA's value. The organic emission factor
X
68
-------
TABLE 25. POM EMISSIONS FROM OIL-FIRED RESIDENTIAL SOURCES,* pg/m3
Compound
Acenaphthene
Acetonaphthone
Anthracene
Azulene or naphthalene
Benzo(c) cinnoline
Biphenyl
Butyl phenanthrene
Dimethyl naphthalene
Dimethyl phenanthrene
Ethyl naphthalene
Fluorenone
Methyl anthracene
Methyl dlbenzo thiophene
Methyl naphthalene
Methyl phenanthrene
Octyl phenanthrene
Phenanthrene
Phenanthrene qulnone
Site 301 Site 302 Site 303 Site 304 Site Site
Module XAD-2 Module XAD-2 XAD-2 Conden- XAD-2 ;
rinse resin rinse resin resin sate resin t t
9.3
2.3 0.5
1.5
20 64
1.2
23 1.0 1.1
20
6.0 2.5 0.2
0.4
23 6.5
4.2s 4.7 0.7
2.4
0.1
7.6 15.4 1.1 0.6
0.2
3.7
0.5 4.7 0.1 3.0 1.7
4.2s 1.2 0.1
Compounds not listed were below the detection limit of -0.3 yg/m3. No POM compounds were detected
in other SASS fractions.
10/20 cycle; all other tests 50/10 cycle.
^Combined sample of XAD-2 resin and particulate filter extracts and .solvent module and probe rinses.
Fluorenone or phenanthrene qulnone.
-------
TABLE 26. MEASURED EMISSION FACTORS FOR PARTICULATE, NOX,
AND TOTAL ORGANICS
Combustion
source
type
Gas-fired sources
Oil-fired sources
Site No.
100
101
102
103
104
130
Mean 3c
s(x)
ts(x)/3c
300
301
302
303
304
326-1 •
327-2
Mean 3c
s(x)
ts(x)/x
Particulate
0.55
0.26
1.7
1.7
0.76
ND
0.99
0.30
0.83
3.1
1.7
8.3
3.5
2.8
1.3 .
1.3
3.1
0.92
0.72
Pollutant
NOX
28
6
66
53
12
ND
33
11.58
0.97
ND
ND
ND
ND
ND
ND
ND
—
-
-
(ng/J)
Total organics
3.0
1.5
0.9
2.3
7.1
0.5
2.55
0.98
0.99
16.2
15.4
11.7
5.1
13.4
1.7
0.7
9.2
2.47
0.66
ND - Not Determined.
Note: "x - Mean, ng/J
s(x) - Standard error, ng/J
ts(x)/x - Variability, percent/100
70
-------
TABLE 27. COMPARISON OF CRITERIA POLLUTANT EMISSION FACTORS FOR
GAS- AND OIL-FIRED RESIDENTIAL COMBUSTION SOURCES
Combustion
source
type
Gas-fired
sources
Oil-fired
sources
Data source
Test program
Existing data
Existing data2
EPA AP-423
Test program
Existing data
EPA AP-423
Pollutant (ng/J)
Particulate
1.0
0.4
2.1
2-6
3.1
8
7.7
NOX
33
57
36
33
ND
44
55
S02
ND
0.21
-
0.26
ND
93
106
CO
ND
38
8
8.4
ND
47f
15
HC
2.6
39
2.0
3.3
9.2
10f
3
The existing data base is discussed in Section 4.1.
Includes tests with excessively high HC and CO values; if these data are
not included, HC and CO values are reduced to 2 and 17, respectively.
ND - Not Determined.
71
-------
obtained in this program is only slightly less than the published EPA value
but well below the average value obtained from the existing literature.
For the oil-fired sources, the measured emission factor for filterable
particulates is less than half of the emission factor obtained by Battelle2
and adopted by EPA.3 The total organic emission factor is comparable to the
existing data base but three times greater than the published EPA emission
factor. It should be noted that the existing data base value is highly
biased by the presence of two extremely high readings (see Appendix D). If
these outliers are omitted from the data base, as was done by EPA in arriving
at its emission factor, the existing data emission factor drops from 10 to
about 2 ng/J. The higher value obtained in this test program probably is a
result of experimental procedures. The total hydrocarbon emission data de-
termined by previous investigations were obtained using gas chromatography
with flame ionization detection, and some of the heavier hydrocarbons may
have condensed in the sampling lines and, therefore, were not measured.
Data for the 10/20 duty cycle tests are not included in Tables 26 and 27.
Particulate emissions from the oil-fired sites, as measured for the 10/20 duty
cycle, were identical to the 50/10 duty cycle emissions. Organic and POM emis-
sions from both the gas- and oil-fired sites were also essentially unaffected
by duty cycle. This is in contrast to an earlier study2 which indicates that
organic emissions decrease when the percent "on" time is increased. For ex-
ample, a decrease in emissions of roughly 20 percent was found when the per-
cent "on" time was doubled for a 15-minute total on/off duty cycle. The
difference is attributed to combustion temperature considerations and is de-
pendent in part upon the temperature response time of the combustion chamber.
The failure to detect any change in organic emissions in this study is proba-
bly due to the inherent accuracy limitations of the Level I measurements and
analyses.
Emission factors for CO and 862 are available in the existing data base.
S02 emission factors published by EPA and shown in Table 27 are based on a
fuel sulfur content of 4,600 g/106 m3 for gas and 0.24 percent for oil. CO
emission factors in the existing literature are much higher than the EPA
values and reflect, in part, the contribution of outliers in the existing
72
-------
data base. If these outliers are discarded, the emission factor drops to
about 17 ng/J. CO emission factors for approximately 6 percent of gas-fired
units13,11* an(j IQ percent of oil-fired units2* ^J12 ^n t^e existing data base
were in excess of 100 ng/J in the as-found condition. The high CO emissions
are due to poorly functioning units. These same units contributed to the
high hydrocarbon emission factor noted in the existing data base.
Data on the composition of SOX from combustion sources indicate that
90 to 100 percent of the emitted SOX is 802- The remaining fraction of SOX
emission is 803 and its derivatives. The main 803 derivative is sulfuric
acid; metallic sulfates appear to be directly emitted only in trace quantities.
The Goks8yr-Ross train, described previously, was used at sites 326 and 327 to
measure 802, 803, and particulate sulfate. The percent conversion of sulfur
to particulate sulfate averaged 0.55 percent, based on the results of two
tests at each site. The average measured conversion of sulfur to 803 of 4.4
percent was higher than anticipated, since values of 1 to 3 percent have been
reported generally for larger combustion systems. However, as noted previ-
ously, high conversions have been obtained in previous studies.19'20 In cer-
tain cases, these high conversions have been attributed to errors in the
chemical analyses of the combustion products of low sulfur fuels. However,
the resolution of these observations will require further study to determine
if the high 803 emissions from residential sources are real or are the result
of experimental data scatter.
The significance of the emissions of criteria pollutants and 803 from
gas- and oil-fired residential combustion sources can be assessed using the
source severity factor. The source severity factor has been discussed briefly
in section 4.1, and detailed methods for the calculation of single source
severity factors are described in Appendix C. Basically, the source severity
factor is defined as the ratio of the calculated maximum ground level concen-
tration of .the pollutant species to the level at which a potential environ-
mental hazard exists. Source severity factors below 0.05 are deemed insig-
nificant. In the case of residential sources, the multiple source severity
factor is used.to indicate the potential environmental significance of
emissions.
73
-------
Multiple source severity factors were determined by using a standard
dispersion technique and a residential area of 1,000 residential units.
The model assumes a class C stability (slightly unstable) and windspeed of
4.5 m/sec (10 mph). Using a grid of houses 80 x 80 m and the average stack
parameters found in this study, the dilution was equivalent to a factor of
3,200. The multiple source severity factors calculated from the ambient
concentrations determined by the model are 25 times greater than single
source severity factors.
As shown in Table 28, multiple source severity factors exceed 0.05 for
several pollutants: NOX from gas-fired sources and 863 and NOX from oil-fired
sources. A potential hazard is associated with these pollutants, given the
multiple array of sources and the meteorological condition used in this study
for the modeling of ambient concentrations.
4.3.2 Emissions of Trace Elements
Existing trace element data for gas- and oil-fired residential combustion
sources are inadequate. During this program, trace element emissions were
measured by SSMS for several elements and by AAS for Hg, As, and Sb. The
trace element content of the oil was also determined and potential emissions
calculated assuming complete release to the atmosphere. In almost all cases,
the measured stack emissions were lower by a factor of roughly two than those
calculated from the fuel analysis (see Appendix F).
Trace element concentrations as determined by SSMS for the gas-fired
sources were lower than those determined for the oil-fired sites. .Emissions
from gas-fired sources either were not detectable or were lower than blank
values for over 90 percent of the elements. Trace element emissions from
gas-fired sources are not an environmental hazard.
Trace element emission factors for the oil-fired sources are presented
in Table 29. Elements shown are those that are present in appreciable quan-
tities plus some elements which have low TLV values. The upper bound emis-
sions were calculated from analysis of the oil, thus representing the worst
case condition. Data variability, also shown in Table 29, is greater than
0.7 for over 50 percent of the elements.
74
-------
TABLE 28. CRITERIA POLLUTANT AND SO 3 SEVERITY FACTORS FOR RESIDENTIAL SOURCES
Gas-fired sources
Pollutant
Particulate
SO 2
SO 3
NOX
CO
Organics
(total)
Emissions*
(ng/J)
1.0
0.26
ND
33
8.4
2.6
Severity
factor
1.7 x 10- 5
3.2 x 10~6
ND
2.8 x ID-3
1.6 x 10~6
1.0 x 10-4
Multiple
source
severity
factor
4.3 x 1Q-1*
8.0 x 10~5
ND
7.0 x 1Q-2
4.0 x 10" 5
2.5 x 10-3
Oil-fired sources
Emissions
(ng/J)
3.1
106
5.9
55
15
9.2
Severity
factor
7.7 x 10~5
1.9 x ID-3
1.6 x ID'2
6.2 x 10" 3
4.2 x 10-6
5.3 x IQ-4
Multiple
source
severity
factor
1.9 x 10-3
4.8 x 1Q-2
4.0 x 10-1
1.6 x 1Q-1
1.1 x 10~4
1.3 x ID-2
S02 and CO emissions from gas-fired sources and S02, NOX, and CO emissions from oil-fired sources
were not measured. EPA emission factors3 were used to calculate source .severities for these
pollutants.
ND - Not Determined.
-------
TABLE 29. TRACE ELEMENT EMISSION FACTORS FOR OIL-FIRED
RESIDENTIAL SOURCES, ng/J
Element
Pb
Ba
Sb
Cd
As
Zn
Cu
Ni
Fe
Cr
V '
Ca
K
Al
Mg
. Hg*
X
0.042
0.016
0.0057
0.011
0.0015
0.11
0.16
0.29
0.14
0.029
0.0029
0.45
0.23
0.25
0.21
0.0012
ts(x)/x
0.81
0.61
1.32
0.97
1.17
0.41
0.59
0.65
5.62
0.95
1.21
0.56
0.96
0.69,
1.1
-
xu
0.075
0.026
0.0013
0.022
0.003
0.16
0.25
0.49
0.92
0.055
0.006
0.71
0.45
0.42
0.44
-
*
Determined by AAS; all others by SSMS.
Note: "x - Mean
ts(x)/x - Variability, percent/100
^u ~ Upper bound of mean
76
-------
Multiple and single source severity factors for trace elements are shown
in Table 30 for the oil-fired sites. Only Ni with a multiple source severity
factor of 0.25 exceeds the value of 0.05, thus representing a potential en-
vironmental hazard.
TABLE 30. TRACE ELEMENT SEVERITY FACTORS FOR
OIL-FIRED RESIDENTIAL SOURCES
Element
Pb
Ba
Sb
Cd
As
Zn
Cu
Ni
Fe
Cr
V
Ca
K
Al
Mg
Hg*
TLV
(yg/m3)
150
500
500
50
500
5,000
1,000
100
5,000
100
500
5,000
2,000
10,000
10,000
50
Maximum
severity
factor
1.
0.
0.
1.
0.
0.
0.
10
0.
1.
0.
0.
o.
0.
0.
0.
0
11
05
0
005
06
5
7
0
03
3
47
09
09
05
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
10-
10"
10-
10-
10"
10-
10"
10~
10~
10"
10-
3
3
3
3
3
3
3
3
3
3
3
10" 3
10-
10-
10-
10-
3
3
3
3
Multiple
source
severity
factor
17
2.
1.
23
0.
2.
12
250
10
25
0.
• 7.
12
2.
2.
1.
8
3
1
0
8
4
3
3
3
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
10~
10-
10-
10-
10"
10-
10"
10-
10-
10-
10-
3
3
3
3
3
3
3
3
3
3
3
10-3
10~ 3
10-
10-
10"
3
3
3
Determined by AAS; all others by SSMS.
4.3.3 Emissions of. POM
Emissions of POM from gas-fired sites, with the exception of naphthalene,
could not be differentiated from the blank values and are, therefore, consid-
ered insignificant. The naphthalene concentration observed was 101* orders of
magnitude less than the MATE value.
77
-------
The concentrations of POM's found in the flue gas from oil-fired resi-
dential sources were two to three orders of magnitude below published MATE
.values, as shown in Table 31. Multiple source severity factors, based on
maximum measured concentrations, were small (10~3 to 10"7). POM emissions
are not significant from the gas- and oil-fired residential combustion
sources tested in this program.
TABLE 31. POM EMISSIONS AND SEVERITY FACTORS
FOR OIL-FIRED RESIDENTIAL SOURCES
Compound
Acenaphthene
Acetonaphthone
Anthracene
Azulene or naphthalene
Benzo(c)cinnoline
Biphenyl
Butyl phenanthrene
Dimethyl naphthalene
Dimethyl phenanthrene
Ethyl naphthalene
Fluorenone
Methyl anthracene
Methyl dibenzo thiophene
Methyl naphthalene
Methyl phenanthrene
Octyl phenanthrene
Phenanthrene
Phenanthrene quinone
Maximum measured
stack concentration
(yg/m3)
9.3
2.3
1.5
20
1.2
23
20
6.0 .
0.4
23
8.9
2.4
0,1
15.4
0.2
3.7
5.2
5.4
MATE
value
(yg/m3)
480
225,000
1,000
225,000
225,000
480
200,000
225,000
30,000
1,600
Multiple
source
severity
factor
3.0 x io~"
6.0 x io~5
2 x 10"3
1.5 x 10"5
6.5 x ID'5
4.5 x 1Q-"
4 x ID"8
4.5 x iQ-5
7 x io~7
3.0 x 10""
See Table 25 to identify sites of maximum emissions.
78
-------
4.3.4 Summary of Status of Emissions Data Base
Based on the analysis of the program test results and the existing data
base, the status of the emissions data base for gas- and oil-fired residential
sources can be summarized as follows:
• Emissions of criteria pollutants are adequately characterized.
There is no need for Level II or additional Level I tests.
• For oil-fired sources, the high 803 levels measured at the two
sites tested indicate a potential environmental risk, although
the high emission levels measured may be due to the analytical
technique employed. Further work to resolve this uncertainty
is indicated.
• Trace element data for the gas-fired sources are adequate.
• Trace element emission data are also adequate for oil-fired
sources; only Ni is emitted in amounts which could be poten-
tially significant. Because additional data for Ni and other
trace elements may be obtained through an analysis of fuel
samples, there is no need for further Level II or additional
Level I tests.
• POM emissions, as measured from the sources tested in this
program, do not constitute an environmental problem. How-
ever, . because of the potential hazard associated with these
compounds, additional testing appears warranted.
• No effect of cycle on/off mode (50 minutes on/10 minutes off
versus 10 minutes on/20 minutes off) on emissions was found,
as measured by Level I analysis. However, this conclusion is
based on a very limited number of tests and is contrary to
the results of earlier studies. Further work to determine
the effect of cycle mode on emissions appears necessary.
This work should also investigate in more detail than was
possible in this program the effect of burner and furnace
design parameters on emissions.
79
-------
5. TOTAL EMISSIONS
Based on the results of this sampling and analysis program and the exist-
ing emissions data base, estimates of current national emissions and projected
1985 national emissions from gas- and oil-fired residential combustion sources
for space heating have been made using recent and projected fuel consumption
rates.
5.1 CURRENT AND FUTURE FUEL CONSUMPTION
During the year 1978, residential heating systems consumed 2550 * 1015
Joules (2400 x 1012 Btu) of oil and 3900 x 1Q15 Joules (3700 x 1012 Btu) of
gas. These fuel consumption figures were derived using the methods discussed
in Appendix B.
Consumption of gas and oil for residential space heating is predicted to
decrease 2.9 percent and increase 3.5 percent, respectively, from 1978 to
1985. Emissions will increase almost proportionately, although the introduc-
tion of new burner and furnace designs and other factors will lead to some
reduction of the projected values for most pollutants. NOX emissions may
increase slightly due to increased use of flame retention burners.
Prediction of fuel use trends from the present time to 1985 is subject
to many uncertainties. Population growth, technology changes, economic
growth (Gross National Product (GNP) and all sectors comprising the GNP),
fuel unavailability, governmental regulations, and imported oil prices and
availability which may be affected by political factors are a few of the
parameters that need to be considered in developing a complete fuel use pro-
jection model. The Federal Energy Administration (FEA) completed a projection
of fuel use trends in 1976 that included most of the factors mentioned above
as well as additional factors. Any complex projection is obviously beyond
the scope of this project. Predictions of future energy trends are subject
80
-------
to radical changes from year to year. Therefore, estimates discussed in this
report are based on two recent studies21'22 and our interpretation of these
studies and related data.
The U.S. Department of the Interior, Bureau of Mines (BOM), has pub-
lished estimates of energy use trends to the years 1980, 1985, and 2000.22
Their projections are "based essentially on the evaluation of Bureau of Mines
fuels data" and the assumption that "existing patterns of resource utiliza-
tion will continue."
BOM and FEA fuel projections for the combined residential/commercial
sector for the year 1985 are presented in Table 32.
TABLE 32. CONSUMPTION OF ENERGY IN THE RESIDENTIAL AND COMMERCIAL
SECTOR, 1974, AND PROJECTIONS TO 1985, 1015 Joules
Energy
source
Coal
Oil
Gas
Electricity
Total
1974
307
6,740
7,510
3,890
18,447
BOM22
1985
106
8,440
9,190
8,240
.25,976
Percent
change
from 1974
- 66
+ 25
+ 22
+ 117
+ 41*
1974
330
6,390
7,930
3,570
18,220
FEA21
1985
120
8,680
6,790
6,810
22,400
Percent
change
from 1974
- 63
+ 36
- 14
+ 90
+ 23f
Fuel consumption would increase 22 percent.
Fuel consumption would increase 6.5 percent.
The method used by the FEA in developing the data in Table 32 is briefly
described below:
"The Project Independence Evaluation System (PIES) is a model of
the technologies, leadtimes, costs and geographical locations
which affect energy commodities from the point of discovery,
through production, transportation, conversion to more useful
forms, and ultimately consumption by all sectors of the economy.
Consumption (final demand) for a particular fuel depends on
81
-------
prices for that fuel, the prices of substitute fuels, the general
level of economic activity, and the ability of consumers and capi-
tal stocks to adjust to these factors. For each year of analysis,
FEA forecasts the demand for refined petroleum products, natural
gas, electricity, and coal. These fuel demands are made for each
Census region and for each end-use consuming sector — residential
and commercial, industrial, and transportation. These demand
forecasts are based on estimated prices and vary as prices
change.
"Energy supply is estimated separately for oil, natural gas, and
coal. For each fuel, many different regions are separately eval-
uated to assess the differences between OCS and Alaskan oil or
Appalachian and Western coal. For each region and fuel, reserve
estimates are combined with the technologies and costs of finding
and producing these fuels to estimate the cost of increasing
supply. Major improvements have been made in the oil and gas
models to estimate drilling patterns, link finding rates and
enhanced recovery directly to revised reserve estimates, and
account for changes in the depletion allowance. The coal supply
estimates distinguish between various sulfur and Btu contents.
"The PIES then attempts to match these energy demands as a
function of fuel, sector, and price with the available supply
in the regions which can supply these needs at the lowest price
to find a balance or equilibrium. If supply is not available
to satisfy the specific demands in an area, the prices are
allowed to vary until supply and demand are brought into
balance."
The FEA conducted the above analyses for three imported oil prices ($8,
$13, and $16 per barrel) and four alternative energy strategies (business as
usual, accelerated supply, accelerated conservation, and a combination of
accelerated supply and conservation).
The FEA and BOM estimates differ in two important areas. First, with
regard to gas, FEA predicts a 14 percent decrease in residential/commercial
82
-------
consumption while BOM predicts a 22 percent increase. For total U.S. con-
sumption of natural gas, the FEA and BOM again differ but in the opposite
direction from the residential sector; FEA predicts a 10 percent increase
and BOM predicts a 10 percent decrease.
The second major difference is in electricity consumption where FEA pre-
dicts a 90 percent residential/commercial increase and BOM predicts a 117 per-
cent increase. The BOM also predicts higher overall growth in electricity
production of 112 percent compared to the FEA estimate of 77 percent. The
BOM estimate is in line with the historical growth of 7 percent per year, but
most experts believe that a lower growth rate near the 5 percent figure cor-
responding to the FEA estimate will prevail.
If the conflicts between the FEA and BOM estimates can be resolved, then
the next problem is to determine how residential fuel use for space heating
is related to the projections. The FEA projections include commercial fuel
use and residential fuel for both space heating and air conditioning (heat
pumps are used in the south and may increase in use).
An alternative approach is to consider changes in population and their
probable impact on residential space heating requirements. One would expect
that in a given area the increase in fuel for space heating would be directly
proportional to population growth. Energy conservation measures and better
uses in new houses could lead to a slower growth in residential space heating
than the growth in population. Changes in population to 1985 are much easier
to predict than changes in fuel use. The data in Table 33 show.regional
growth patterns and estimated 1985 fuel consumption based on these patterns.
The predicted increase in fuel is less than the change in population because
of the high growth rates in southern areas. An even smaller increase may
occur because of energy conservation, improved heating efficiency and insula-
tion, and a larger increase in electric heat. Residential fuel use in 1977
was down 2.5 percent from that used in 1976 according to recent BOM data.
This decline was attributed largely to energy conservation measures although
increased use of electricity (~ 1 percent) for residential heating also
contributed.
83
-------
TABLE 33. ESTIMATES OF RESIDENTIAL SPACE HEATING FUEL CONSUMPTION TO 1985
BASED ON POPULATION GROWTH
Region
New England
Middle Atlantic
East North Central
West North Central
South Atlantic
East South Central
West South Central
Mountain
Pacific
Total
1975
fuel consumption
(10 15 Joules)
647
1,748
1,727
741
589
212
388
287
494
6,833
Population
change
(percent)
+ 4.2
0.0
+ 1.9
+ 2.0
+ 16.1
+ 10.1
+ 17.1
+ 23.8
+ 15.8
+ 8.8
1985
fuel consumption
(1015 Joules)
674
1,748
1,760
755
682
234
454
355
572
7,234*
*
Fuel consumption would increase 6 percent from 1975 to 1985.
Our best estimate is that fuel usage for residential space heating will
increase 4 percent from 1978 to 1985. Coal use decreased almost 50 percent
over the 3-year period 1974 to 1976. However, an upswing in coal consumption
for residential sources is anticipated through 1985, based on increased sales
of coal-burning residential furnaces and space heaters.23 A similar increase
in wood consumption, over the estimated 1978 usage21* of 105 x 1015 Joules, is
anticipated. Considering all the data discussed in this section, we have
projected a decrease in gas consumption of 2.9 percent by 1985. To achieve
a total fuel usage increase of 4 percent, oil consumption will increase
3.5 percent. Projections for residential heating are summarized in Table 34.
Fuel usage for 1978 was estimated from references in Appendix B.
84
-------
TABLE 34. ESTIMATES OF RESIDENTIAL SPACE HEATING
FUEL CONSUMPTION TO 1985 BASED ON FUEL
TYPE
Fuel
Coal
Oil
Gas
Wood
Total
Fuel
(101
1978
211
2,532
3,903
105
6,751
consumption
5 Joules)
1985
400
2,620
3,790
210
7,020
Percent
change
+ 90
+ 3.5
- 2.9
+ 100
+ 4.0
5.2 CURRENT AND FUTURE NATIONWIDE EMISSIONS
Total 1978 national emissions from gas- and oil-fired residential com-
bustion sources for space heating were determined based on combined test
program and existing data emission factors and the estimated 1978 fuel con-
sumption rates discussed previously. Nationwide emission totals for the
criteria pollutants are presented in Table 35. Particulate, SOX, and NOX
emissions from residential sources are relatively small, accounting for
about 0.4, 1.2, and 2.5 percent, respectively, of emissions from all sta-
tionary combustion sources based on total estimates given in. reference 4.
Carbon monoxide emissions account for about 7 percent of emissions from
stationary combustion sources, but less than 0.1 percent of total manmade
carbon monoxide emissions. Motor vehicles are the major source of carbon
monoxide emissions. Hydrocarbon emissions from the residential sources are
roughly 10 percent of total emissions from stationary combustion sources.
However, hydrocarbon emissions from the gas- and oil-fired residential com-
bustion sources represent only about 0.1 percent of hydrocarbon emissions
from all manmade sources of hydrocarbon emissions.
85
-------
TABLE 35. RELATIONSHIP OF GAS- AND OIL-FIRED RESIDENTIAL HEATING
SOURCE EMISSIONS TO TOTAL ESTIMATED EMISSIONS FROM
COMBUSTION SOURCES
Pollutant
Participates
sox
NOX.
CO
HC
Total POM
Trace elements
Pb
Cd
As .
Ni
Cr
Emissions
Gas-fired
8,200
1,000
128,800
32,800
12,800
Neg
11
23
6
ND
< 1
(tonnes/yr)
Oil-fired
20,100
268,400
139,300
38,000
23,300
~ 500
106
28
4
734
73
Percent of national
emissions
Gas-fired Oil-fired
0.1 0
< 0.01 1
1.2 1
3.2 3
3.7 6
Neg < 0
-
-
-
-
-
.3
.2
.3
.4
.5
.01
-
-
-
-
-
Based on reference 4 data.
national emissions.
ND - Not Detected.
Trace element data are insufficient to calculate
Current trace element emissions from oil-fired residential source heating
systems are presented in Table 36. Emissions from gas-fired sources are con-
sidered to be negligible in comparison and are not reported in the table.
Elements emitted in the largest amounts, with the exception of Ni, are rela-
tively harmless; e.g., Ca, Al, and Mg. The trace element emissions listed in
Table 36 represent about 22 percent of the total particulate emissions from
oil-fired residential sources. The high percentage contribution of trace ele-
ments to total particulate is due in part to the assumption that all elements
in the fuel are emitted with the flue gases.
86
-------
TABLE 36. NATIONAL TRACE ELEMENT EMISSIONS FROM
OIL-FIRED RESIDENTIAL SOURCES, 1978
„.. . Emissions
Element , . N
(tonnes/yr)
Pb 106
Ba 40
Sb 2
Cd 28
As 4
Zn 290
Cu 410
Ni 734
Fe 360
Cr 73
Ca 1150
Al 710
Mg 540
Emissions of POM from oil-fired residential sources are summarized in
Table 37. POM emissions from the gas-fired residential sources were not
detectable, with the exception of small quantities of naphthalene and its
derivatives. The emission quantities have been calculated using the maximum
values found in the test program.
87
-------
TABLE 37. NATIONAL POM EMISSIONS FROM OIL-FIRED
RESIDENTIAL SOURCES, 1978
_ , Emissions
Compound , , ,
(tonnes/yr)
Acenaphthene 31
Acetonaphthone 10
Anthracene 7
Azulene or naphthalene 68
Benzo(c)cinnoline 6
Biphenyl 67
Butyl phenanthrene 68
Dimethyl naphthalene 21
Dimethyl phenanthrene 1
Ethyl naphthalene 77
Fluorenone 29
Methyl anthracene 9
Methyl dibenzo thiophene 1
Methyl naphthalene 51
Methyl phenanthrene 1
Octyl phenanthrene 11
Phenanthrene 19
Phenanthrene quinone 19
88
-------
6. REFERENCES
1. Hamersma, J. W., S. L. Reynolds, and R. F. Maddalone. IERL-RTP Procedures
Manual: Level 1 Environmental Assessment. TRW Systems Group, Prepared
for the U.S. Environmental Protection Agency. EPA-600/2-76-160a. NTIS
No. PB 257 850. June 1976.
2. Barrett, R. E., S. E. Miller, and D, W. Locklin. Field Investigation of
Emissions from Combustion Equipment for Space Heating. Battelle-Columbus
Laboratories. Prepared for the U.S. Environmental Protection Agency.
EPA-R2-73-084a. June 1973.
3. Compilation of Air Pollutant Emission Factors. Third edition, including
Supplements 1-7. U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina. Publication No. AP-42. NTIS No. PB 275 525.
October 1977.
4. Surprenant, N., R. Hall, S. Slater, T. Susa, M. Sussman, and C. Young.
Preliminary Emissions Assessment of Conventional Stationary Combustion
Systems, Volume II - Final Report. GCA/Technology Division. Prepared
for the U.S. Environmental Protection Agency. EPA-600/2-76-046b. NTIS
No. PB 252 175. March 1976.
5. Statistical Abstract of the United States, 1976. U.S. Department of
Commerce, Bureau of Census. 1976.
6. Current Housing Reports, Series H-150-74. Annual Housing Survey: 1974,
Part A, General Housing Characteristics for the United States and Regions
(advance report). U.S. Department of Commerce, Bureau of Census. 1976.
7. Offen, G. R., J. P. Kesselring, K. Lee, G. Poe, and K. J. Wolfe. Control
of Particulate Matter from Oil Burners and Boilers. Aerotherm Division,
Acurex Corporation. Prepared for the U.S. Environmental Protection
Agency. EPA-450/3-76-005. NTIS No. PB 258 495. April 1976.
8. Brown, R. A., C. B. Moyer, and R. J. Schreiber. Feasibility of a Heat
and Emission Loss Prevention System for Area Source Furnaces. Aerotherm
Division, Acurex Corporation. Prepared for the U.S. Environmental Pro-
tection Agency. EPA-600/2-76-097. NTIS No. PB 253 945, April 1976.
Hall, R. E., T. H. Wasser, and E. E. Berkau. A Study of Air Pollutant
Emissions from Residential Heating Systems. Control Systems Laboratory,
Research Triangle Park, North Carolina. EPA-650/2-74-003. . January 1974.
89
-------
10. Eimutis, E. C. Mean Severities of Air Emissions From Combustion Sources,
Monsanto Research Corporation. Interim report to the U.S. Environmental
Protection Agency. December 1, 1976,
11. Levy, A., S. E. Miller, R, E. Barrett, E. J. Schulz, R. H. Melvin,
W. H. Axtman, and D. W. Locklin. A Field Investigation of Emissions
from Fuel Oil Combustion for Space Heating. Battelle-Columbus Labora-
tories. Prepared for the American Petroleum Institute. API Publication
A099. November 1971.
12. Barrett, R. E., D. W. Locklin, and S. E. Miller. Investigation of Par-
ticulate Emissions From Oil-Fired Residential Heating Units. Battelle-
Columbus Laboratories. Prepared for the U.S. Environmental Protection
Agency. EPA-650/2-74-026. March 1974.
f \
I 13. Kalika, P. W., G. T. Brookman, and J. E. Yocom. Final Report to the
*~^^ . American Gas Association for a Study on Measuring the Environmental
Impact of Domestic Gas-Fired Heating Systems. The Research Company of
New England. June 1974.
14* Kalika, P. W., G. T. Brookman, and J. E. Yocom. Second Quarterly Report
y to the American Gas Association for a Study on Measuring the Environ-
mental Impact of Domestic Gas-Fired Heating Systems. The Research Com-
pany of New England. July 13, 1973.
15.. Bartz, D. R. et al. Control of Oxides of Nitrogen from Stationary
Sources in the South Coast Air Basin of California. KVB Engineering,
Incorporated. Prepared for the California State Air Resources Board.
ARB-R-2-1471-74-31. September 1974.
16. Katzman,' L., and W. Weitzman. Study to Evaluate the Effects of Reducing
the Firing Rates on Residential Oil Burner Installations. Walden Re-
search Division of Abcor. Prepared for the U.S. Department of Commerce,
National Bureau of Standards. Contract No. 6-35738.
17. Katzman, L., and R. D'Agostino. A Study to Evaluate the Effect of Per-
forming Various Energy Saving Procedures on Residential Oil Burner In-
stallations in the New England Area and to Gather Information on the
Steady State and Dynamic Performance of These Installations. Walden
Research Division.of Abcor. Prepared for the U.S. Department of Com-
merce, National Bureau of Standards. Contract No. 5-35781.
/ ^
y 18. Hangebrauck, R. P., D. J. von Lehmden, and J. E. Meeker. Sources of
^~~^ Polynuclear Hydrocarbons in the Atmosphere. Control Technology Research
and Development Center for Air Pollution Control. Prepared for the U.S.
Department of Health, Education and Welfare, Public Health Service.
Environmental Health Series 999-AP-33. NTIS No. PB 260 073. 1967.
90
-------
19. Cato, G. A., H. J. Buenlng, C. C. DeVivo, B. G. Morton, and J. M. Robin-
son. Field Testing: Application of Combustion Modification to Control
Pollutant Emissions from Industrial Boilers, Phase I. KVB Engineering,
Incorporated. Prepared for the U.S. Environmental Protection Agency.
EPA-650/2-74-078a. NTIS No. PB 238 920. October 1974.
20. Homolya, J. B., and J. L. Cheney. An Assessment of Sulfuric Acid and
Sulfate Emissions From the Combustion of Fossil Fuels. Workshop Pro-
ceeding on Primary Sulfate Emissions from Combustion Sources, Volume 2:
Characterization. Prepared for the U.S. Environmental Protection Agency.
EPA-600/9-78-020b. August 1978.
21. National Energy Outlook. Federal Energy Administration, Washington, D.C.
NTIS No. PB 252 224. February 1976.
22. Dupree, W. G., and J. S. Corsentino. United States Energy Through the
Year 2000 (Revised). U.S. Department of the Interior, Bureau of Mines.
NTIS No. PB 250 600. December 1975.
23. DeAngelis, D. G., and R. B. Reznik. Source Assessment: Residential
Combustion of Coal. Preliminary Report. Monsanto Research Corporation.
Prepared for the U.S. Environmental Protection Agency. Contract No.
68-02-1874. 1978.
24. ^Personal Communication. W. Lyman Garrett, United States Forest Service,
Burlington, Vermont.
91
-------
APPENDIX A
CONVERSION FACTORS AND METRIC PREFIXES
92
-------
CONVERSION FACTORS
To convert from
To
Degree Celsius ( C)
Gram (g)
Kilogram (kg)
Metric ton (tonne)
Gram/second (g/s)
Centimeter (cm)
Meter (m)
Meter (m)
Cubic meter (m3)
British thermal unit (Btu)
Joule (J)
Degree Fahrenheit ( F)
Pound-mass
Pound-mass
Pound-mass
Pound/hour (Ib/hr)
Inch (in.)
Inch (in.)
Foot (ft)
Cubic foot (ft3)
Joule (J)
British thermal unit (Btu)
Multiply by
1.8°C + 32
2.205 x 10~3
2.205
2..205 x IO3
7.94
0.394
. 39.4
3.281
35.3
1055
9.48 x ifl-1*
METRIC PREFIXES
Prefix
Symbol
Giga
Mega
Kilo
Centi
Milli
Micro
Nano
Pico
G
M
k
c
m
y
n
P
Multiplication factor
109
106
103
10-2
io-3
ID'6
io-9
io-12
93
-------
APPENDIX B
FUEL CONSUMPTION BY RESIDENTIAL SPACE HEATING SOURCES
94
-------
FUEL CONSUMPTION BY RESIDENTIAL SPACE HEATING SOURCES
There is no generally accepted method for determining residential fuel
use for space heating and similarly no agreement on the total amount of fuel
used.
One approach is to determine the fuel required (Btu) for each dwelling
JU
unit (d.u.) for each heating degree-day (d.d.). Climatological data are
available from the U.S. Department of Commerce showing degree-days data annu-
ally as well as long-term averages for sites throughout the U.S.1 Data are
also available on the number of dwelling units in each state using each fuel.2
These latter data are compiled by the Bureau of Census during deciannual sur-
veys. Regional housing data are published annually.3
The amount of fuel required for each dwelling unit degree-day depends on
housing characteristics, the amount of area heated, and the efficiency of the
heating system. This factor can be determined through field surveys or engi-
neering analyses. A heated area of 1600 ft2 is generally accepted as typical.
There is no debate about the steady state efficiency of residential heating
systems (70 to 80 percent), but estimates of the overall efficiency that con-
sider the effects of cycling and cold air infiltration range from 40 to 80
percent.** Available estimates for heating requirements (gas-fired systems)
are 17,000 Btu/d.u.-d.d. from the U.S. Environmental Protection Agency,
25,000 Btu/d.u.-d.d. from Walden Research, 26,000 Btu/d.u.-d.d. from Hittman
Associates, and 32,000 to 34,000 Btu/d.u.-d.d. from the U.S. Federal Energy
Administration.6
*
A heating degree-day is a measure of the heating requirement. For a single
day it is the difference between 65°F and the mean temperature. If the mean
temperature is above 65°F, the heating degree-day total is zero.
17,000 Btu/d.u.-d.d. is equivalent to 32.2 10 Joules/d.u.-d.d. with degree-
days given in degrees Celsius. EPA has also published a figure for oil of
25,000 Btu/d.u.-d.d. in Reference 5.
95
-------
A second approach is to use fuel consumption data published annually in
Mineral industry Surveys by the U.S. Bureau of Mines.7"11 However, the pub-
lished residential gas consumption data include fuel used for space heating,
cooling, water heating, and clothes drying. The problem is to separate space
heating from other uses. On a national basis, 70 percent of residential gas
consumption can be assumed to be used for space heating based on a study by
Stanford Research Institute.12 However, on a state-by-state basis this
assumption is not valid; i.e., Florida probably uses a smaller percentage for
heating. Data for other residential fuels are grouped with the commercial
category, or grouped in other categories that make the determination of resi-
dential space heating fuel consumption difficult and uncertain.
A combination of the methods discussed above was used to develop the
state-by-state residential space heating fuel consumption data presented in
Table B-l.. Regional data are presented in Table B-2. These data were devel-
oped from 1974 and 1975 fuel use data, 1970 and 1974 housing characteristics
(the number of units that use each fuel in each state), and long-term average
degree-day data. The most current estimate of the number of dwelling units
in each state using each fuel for space heating was determined by updating
the 1970 data to 1974 according to the regional growth rates in the period
1970 to 1974. These estimates can be considered as 1975 data but they would
also apply to other years depending on the changes in the number of units and
the heating requirements for that year.
National fuel use data for 1975 presented in Tables B-l and B-2 have been
updated, using 1977 data from the Bureau of Mines Industry Surveys, trade jour-
nals,13 and various government agency energy data reports. *' National
totals for 1976 and 1977 were obtained from these reports and extrapolated to
1978. The updated 1978 national fuel consumption totals are presented in
Section 5 of this report. The methodology used for derivation of the fuel
consumption data is discussed below.
U.S. consumption of natural gas for residential space heating was esti-
mated to be 70 percent12 of the total residential consumption. To determine
the total U.S. dwelling unit-degree days, the number of housing units using
gas heat was multiplied by the degree-days in each state, and these products
96
-------
TABLE B-l. RESIDENTIAL SPACE HEATING FUEL USE BY STATE, 1975,20
1012 Btu/yr
State
New England
Maine
Vermont
New Hampshire
Massachusetts
Connecticut
Rhode Island
Middle Atlantic
New York
New Jersey
Pennsylvania
East North Central
Ohio
Indiana
Illinois
Michigan
Wisconsin
West North Central
Minnesota
Iowa
Missouri
North Dakota
South Dakota
Nebraska
Kansas
South Atlantic
Delaware
Maryland
District of Columbia
Virginia
West Virginia
North Carolina
South Carolina
Georgia
Florida
Gas
Natural
1
1
4
71
24
11
314
102
230
309
115
330
285
107
124
83
108
12
14
47
65
5
57
13
36
37
16
11
42
6
LPG1"
0.7
1.2
0.7
2.5
1.6
0.4
8.9
1.9
3.8
8.3
11.3
17.4
10.1
13.2
13.5
15.5
25.0
3.9
5.5
8.6
9.2
8.4
1.8
0.4
1.8
1.0
5.0
3.7
12.3
6.2
Oil*
63.9
24.0
42.0
216.1
109.9
32.4
548.7
167.6
231.1
63.6
57.3
68.6
98.7
93,4
84.6
39.8
14.1
19.0
14.7
7.0
1.3
14.2
66.8
7.8
67.2
5.7
76.4
21.0
4.0
11.5
Anthracite
0.144
0.231
0.112
0.682
0.179
0.035
6.893
2.150
37.803
2.634
1.971
0.887
3.138
0.170
0.278
0.000
0.000
0.000
0.000
0.000
O.OOQ
0.147
1.351
0.107
0.224
1.474
0.000
0.000
0.000
0.000
Coal
5 Bituminous^
0.144
0.020
0.041
0.015
0.007
0.000
0.000
0.000
0.000
2.539
0.766
12.643
0.671
1.746
0.998
0.771
0.491
0.740
0.196
0.134
0.067
0.000
0.338
0.388
0.359
3.729
2.235
0.851
0.835
0.005
Wood**
1.309
0.383
0.414
0.204
0.131
0.035
0.936
0.124
0.899
0.454
0.553
0.269
0.657
0.869
-
1.004
0.165
2.928
0.025
0.210
0.119
0.259
0.068
0.552
0.005
3.160
0.383
3.506
2.065
2.758
0.285
(continued)
97
-------
TABLE B-l (continued).
State
East South Central
Kentucky
Tennessee
Alabama
Mississippi
West South Central
Arkansas
Louisiana
Oklahoma
Texas
Mountain
Montana
Idaho
Wyoming
Colorado
New Mexico
Arizona
Utah
Nevada
Pacific
Washington
Oregon
California
Alaska
Gas
Natural* LPG1"
51
30
30
16
49
59
52
164
28
10
15
86
22
19
34
6
35
17
290
6
7.3
4.6
9.7
7.8
7.9
3.1
9.1
18.3
. 3.3
1.6
2.3
7.5
2.9
1.1
1.4
1.3
2.5
4.4
14.2
0.2
Oil* -
I
9.7
6.6
1.2
0.2
0.3
0.2
0.2
0.7
5.3
12.9
0.6
1.8
0.9
0.2
2.2
2.3
62.2
28.8
2.3
10.3
Coal
Vnthracite^
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
Bituminous'
8.436
6.171
1.847
0.163
0.100
0.001
0.022
0.011
0.321
1.206
0.255
0.838
0.039
0.006
1.045
0.026
0.665
0.109
0.023
0.290
- Wood
1.925
3.429
2.360
2.031
2.425
0.380
0.813
0.944
0.461
0.503
0.085
0.150
0.920
0.355
0.095
0.073
1.568
2.250
1.402
0.423
1,022 Btu/ft3.
90,000 Btu/gal.
^140,000 Btu/gal.
26 x 106 Btu/ton.
24 x IQ6 Btu/ton.
12 x 106 Btu/ton.
Note: To convert from 1012 Btu to 1015 Joules, multiply by 1.055.
98
-------
TABLE B-2. U.S. RESIDENTIAL SPACE HEATING FUEL USE BY REGION, 1975, 1015 J/yr"
vo
Region
United States
Northeast
New England
Middle Atlantic
North Central
East North Central
West North Central
South
South Atlantic
East South Central
West South Central
West
Mountain
Pacific
Gas
Natural"*"
3,794
801
119
681
1,686
1,208
478
716
237
135
342
597
231
366
LPG^
316
23
7
16
150
85
63
105
35
31
40
38
22
15
Oil5
2,550
1,515
515
999
593
403
191
310
290
19
1
133
27
105
Coal
Anthracite*
64.3
50.8
1.5
49.4
9.6
9.3
0.3
3.5
3.5
0.0
0.0
0.0
0.0
0.0
Bituminous
54.9
0.2
0.2
0.0
23.0
19.4
3.6
26.4
9.2
17.5
0.1
5.1
3.9
1.2
Wood§§
49.9
4.6
2.6
2.1
7.9
3.0
5.0
28.6
13.5
10.2
4.9
8.8
2.7
5.9
All
fuels
6,833
2,395
647
1,748
2,468
1,727
741
1,189
589
. 212
388
781
287
494
3?
Totals may not agree because of rounding.
f38 x 106 J/m3 (1,022 Btu/ft3).
*25.1 x 109 J/fc (90,000 Btu/gal).
§39 x 109 J/8, (140,000 Btu/gal).
*30 x 106 J/kg (26 x 106 Btu/ton).
JLJL
27.9 x IQ6 J/kg (24 x 106 Btu/ton)
§§13.9 x 106 J/kg (12 x 106 Btu/ton)
-------
were summed, resulting in a U.S. total of 15.8 x 1010 dwelling unit-degree
days. Comparing the estimated U.S. natural gas consumption of 3600 x 1012 Btu
(3800 x 1015 J) in 1975 to the total dwelling unit-degree days indicates a
heating requirement of 22,780 Btu/d.u.-d.d. which is consistent with estimates
previously discussed. This heating requirement was multiplied by the number
of dwelling units using gas heat and the average degree-days in each state to
obtain state-by-state consumption.
Consumption of LPG in the U.S. is reported only as a residential/commer-
cial total. The residential fraction was assumed to be the same as the resi-
dential fraction for natural gas. Also, 70 percent of the calculated resi-
dential consumption of LPG was assumed to be used for space heating, in line
with published estimates for natural gas.12 On this basis, U.S. residential
space heating consumption of LPG was estimated to be 300 x 1012 Btu
(316 x 1015 J). Calculation of the total U.S. LPG dwelling unit-degree days
indicated a heating requirement of 19,600 Btu/d.u.-d.d. This heating require-
ment was used to determine LPG use for space heating by state from the number
of dwelling units using LPG and the heating degree-days in each state.
Residential consumption of fuel oil for space heating was estimated from
BOM and DOE energy data reports which provide annual fuel oil sales data for
heating purposes. Approximately 76 percent of the fuel oil sold for heating
is consumed by the residential sector.16 This percentage was used to generate
the 1975 residential fuel oil consumption total of 2400 x 1012 Btu
(2550 x 1015 J). State-by-state-consumption was estimated using the method-
ology discussed above for natural gas.
Residential consumption of coal (anthracite and bituminous) on a state-
by-state basis was determined from the number of dwelling units using coal,
the average degree-days, and a reported consumption of 0.0012 ton/d.u.-d.d.5
which is equivalent to about 30,000 Btu/d.u.-d.d. for an average house. Dif-
ferences in housing characteristics on a state-by-state basis were also con-
sidered by multiplying the heating requirements by the ratio of the average
rooms per house in the state to the national average of five rooms per house.
In accordance with procedures outlined in reference 5, the calculated resi-
dential coal consumption was compared to anthracite deliveries in the selected
100
-------
state (reference 7). If the calculated consumption was less than the delivered
anthracite, the difference was assumed to be bituminous. Total coal consump-
tion for residential space heating in 1978 was estimated as 200 x 1012 Btu
(211 x 1015 J). This figure is almost twice that reported for 1975 and is
based primarily on sales data for solid burning residential furnaces and space
heaters. Lignite consumption by residential sources is not significant,20
(= 2 x 1012 Btu in 1978).
Minor quantities of wood are also used in rural areas for residential
space heating. Wood consumption on a state-by-state basis was calculated from
the number of dwelling units using wood for space heating,2'3 the average heat-
ing degree-days, and the reported heating requirement of 0.0017 ton/d.u.-d.d.
(20,000 Btu/d.u.-d.d.). Estimated total U.S. consumption was 100 x 1012 Btu
(105 x 1015 J) in 1978, up from an estimated 47 x 1012 Btu (49.9 x 1015 J) in
1975.
101
-------
REFERENCES
1. Local Climatological Data, Annual Summary With Comparative Data. U.S.
Department of Commerce, Environmental Sciences Service Administration.
Washington, D.C.
2. 1970 Census of Housing. Detailed Housing Characteristics. HC-B Series.
U.S. Department of Commerce, Bureau of the Census. 1972
3. Current Housing Reports, Series H-150-75 and H-150-76, Annual Housing
Survey: 1975 and 1976, Part A, General Housing Characteristics for the
United States and Regions (advance report). U.S. Department of Commerce,
Bureau of Census. 1976, 1977.
4. Brown, R. A., C. B. Moyer, and R. J. Schreiber. Feasibility of a Heat
and Emission Loss Prevention System for Area Source Furnaces. Aerotherm
Division, Acurex Corporation. Prepared for the U.S. Environmental Pro-
tection Agency. EPA-600/2-76-097. NTIS No. PB 253 945. April 1976.
5. Guide for Compiling a Comprehensive Emission Inventory, Second Edition.
U.S. Environmental Protection Agency. APTD 1135. December 1974.
6. Benesh, F. Growth Effects of Major Land Use Projects: Volume II - Com-
pilation of Land Use Based Emission Factors. Walden Research Division of
Abcor. Prepared for the U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina. EPA-450/3-76-012b. NTIS No. PB 255 302.
September 1976.
7. Mineral Industry Surveys: Coal - Pennsylvania Anthracite in 1976. U.S.
Department of the Interior, Bureau of Mines.
8. Mineral Industry Surveys: Bituminous Coal and Lignite Distribution, 1976.
U.S. Department of the Interior, Bureau of Mines.
9. Mineral Industry Surveys: Natural Gas Production and Consumption, 1976.
U.S. Department of the Interior, Bureau of Mines.
10. Mineral Industry Surveys: Liquefied Petroleum Gas Sales, Annual Summary
for 1975, 1976, and 1977. U.S. Department of the Interior, Bureau of
Mines.
11. Mineral Industry Surveys: Sales of Fuel Oil and Kerosene in 1975. U.S.
Department of the Interior, Bureau of Mines.
102
-------
12. Patterns of Energy Consumption in the United States. Stanford Research
Institute, Office of Science and Technology. 1972.
13. Gas Facts. Annual Statistical Record for 1973 and 1977. Department of
Statistics, American Gas Association.
14. Energy Data Reports: Fuel Oil Sales. Annual Summary for 1976 and 1977.
U.S. Department of Energy, Energy Information Administration.
15. Energy Data Reports: Natural Gas Production and Consumption. Annual
Summary for 1976 and 1977. U.S. Department of Energy, Energy Informa-
tion Administration.
16. National Emission Data Systems: Fuel Use Report, 1974. U.S. Environ-
mental Protection Agency. EPA-450/2-77-031. NTIS No. PB 284 658.
April 1978.
17. Electric Power Statistics for 1974, 1975, and 1976. Federal Power
Commission. Washington, D.C.
18. Quarterly Report: Third Quarter 1978. U.S. Department of Energy,
Energy Information Administration. October 1978.
19. Energy Monthly Review: January 1979. U.S. Department of Energy, Energy
Information Administration. DOE/EIA/0035/K79).
20. Surprenant, N., R. Hall, S. Slater, T. Susa, M. Sussman, and C. Young.
Preliminary Emissions Assessment of Conventional Stationary Combustion
Systems, Volume II - Final Report. GCA/Technology Division. Prepared
for the U.S. Environmental Protection Agency. EPA-600/2-76-046b. NTIS
No. PB 252 175. March 1976.
103
-------
APPENDIX C
CRITERIA FOR EVALUATING THE ADEQUACY
OF EXISTING EMISSIONS DATA FOR
CONVENTIONAL STATIONARY COMBUSTION SOURCES
104
-------
CRITERIA FOR EVALUATING THE ADEQUACY
OF EXISTING EMISSIONS DATA FOR
CONVENTIONAL STATIONARY COMBUSTION SOURCES
A major task in the present overall emissions assessment program was the
identification of gaps and inadequacies in the existing emissions data base
for conventional stationary combustion systems. The output from this effort
was used in the planning and performance of a combined field and laboratory
program as required to complete an adequate emissions assessment for each of
the combustion source types.
The criteria for assessing the adequacy of emissions data were developed
by considering both the reliability of the data and the variability of the
data. The general approach was to utilize a three-step process as described
below. This approach was applicable to the evaluation of the existing emis-
sions data as well as emissions data collected during the course of this pro-
gram. The following approach was used as appropriate in this assessment of
gas- and oil-fired residential heating sources.
STEP 1
In the first step of the evaluation process, the emissions data were
screened for adequate definition of process and fuel parameters that might
affect emissions, as well as for validity and accuracy of sampling and analy-
sis methods. The screening mechanism was devised to reject emissions data
that would be of little or no use. Acceptance of emissions data in this
screening step only indicated the possibility for further analysis, and in no
way suggested that these data were valid or reliable. As such, the data
screening criteria were often expressed in terms of minimum requirements.
These screening criteria are depicted in Figure C-l and discussed in detail
below.
The first criterion applied was that only source test data would be
accepted. A significant portion of the data base, and especially those
105
-------
P
EMISSIONS DATA
ARE DATA ACQUIRED BY
SOURCE TESTING?
NO
YES
IS THERE SUFFICIENT INFORMATION
TO DESIGNATE THE COMBUSTION SOURCE
ACCORDING TO GCA CLASSIFICATION CODE?
NO
YES
IS THERE INFORMATION
ON FUEL CONSUMPTION RATE?
FOR NOX EMISSIONS DATA,
IS THERE INFORMATION
ON OPERATING LOAD?
NO
YES
FOR PARTICULATE EMISSIONS DATA
FROM COAL BURNING UTILITY BOILERS,
IS THERE INFORMATION
ON PARTICULATE CONTROL
DEVICE PERFORMANCE?
NO
YES
FOR TRACE ELEMENT EMISSIONS DATA
FROM COAL AND OIL COMBUSTION,
ARE THERE CORRESPONDING DATA
ON TRACE ELEMENT CONTENT OF THE FUEL?
FOR SOX EMISSIONS DATA
FROM COAL AND OIL COMBUSTION,
ARE THERE CORRESPONDING DATA
ON SULFUR CONTENT OF THE FUEL?
NO
YES
IS THERE INFORMATION
ON THE SAMPLING AND ANALYSIS
METHODS EMPLOYED?
YES
YES
CAN SAMPLING AND ANALYSIS
METHODS EMPLOYED
PROVIDE EMISSION ESTIMATES
WITH AN ACCURACY BETTER
THAN A FACTOR OF 3?
YES
NO
INCLUDE EMISSIONS DATA IN USABLE DATA BASE FOR FURTHER ANALYSIS
PROCEED TO STEP 2
Figure C-l. Step 1 screening mechanism for emissions data.
106
-------
contained in the National Emissions Data System (NEDS), was developed by the
us.e of standard emission factors and not derived from actual test data. The
inclusion of these estimated emissions data in the data base would have led to
the obviously biased conclusion that the actual emissions were the same as
those predicted by the standard emission factors.
The second criterion to be applied was an adequate description of the
source. In order to further analyze the emissions data, sufficient informa-
tion must be available to designate the combustion source according to the
appropriate GCA classification code. As a minimum, the information provided
should include the following: the function of the combustion source (elec-
tricity generation, industrial, commercial/institutional, or residential); the
type of combustion (external or internal); the type of fuel used (coal, oil,
gas, or refuse); and, in the case of coal combustion, the type of furnace (pul-
verized dry bottom, pulverized wet bottom, cyclone, or stoker). For emissions
data that were judged to be valuable^ and otherwise acceptable, efforts were
made to acquire the needed source description information directly from the
investigator or the plant operator.
The third criterion for acceptance of emissions data for further analysis
was an adequate definition of the combustion system operating mode. For exam-
ple, operating load has a large effect on NOX emissions from combustion sys-
tems. It was, therefore, important to have an adequate definition of the test
conditions that might affect emissions. As a minimum, information on the fuel
consumption rate was required for the emissions data to be accepted. The fuel
consumption rate was necessary for the calculation of emission factors. For
NOX emissions data, field and test results that did not include information on
operating load were considered unacceptable because they could not be used to
estimate emissions from a typical combustion system, nor could they be used to
estimate emissions at any specific load. For other types of emission data,
the operating load information was considered as a useful parameter for data
correlation but not an absolute requirement for data acceptance.
Emission factors obtained in most instances from "Compilation of Air Pollu-
tant Emission Factors," U.S. EPA Publication AP-42.
In this context, emissions data for trace elements, POM, PCB, and organics
were considered to be more valuable because of the paucity of data.
107
-------
The fourth criterion for acceptance of emissions data for further analy-
sis was an adequate definition of the pollution control device performance.
Control device performance affects not only total emissions but also influ-
ences, for example, the particle size distribution and composition of flue
gas emissions. The application of design efficiencies must be approached
with caution in estimating uncontrolled emissions. If a design efficiency of
99 percent is used, and if the control device operating efficiency is only
90 percent, the calculated uncontrolled emissions would be 10 times larger
than the actual case. Since most coal burning utility boilers are equipped
with particulate control devices, particulate emissions data from the coal
burning utility sector were not considered acceptable unless accompanied by
the particulate control device performance data. The application of particu-
late control devices is less common in the industrial, commercial/institu-
tional, and residential sectors; it is also much less common in the oil burn-
ing utility sector and is nonexistent in the gas burning utility sector. For
these combustion source types, emissions data were accepted as uncontrolled
emissions data unless there was information implying the contrary. As noted
in the foregoing discussions, acceptance of emissions data at this screening
step did not suggest that the data were necessarily valid or reliable. In
the second step of the data evaluation process, methods for rejecting outlying
data points are defined. Controlled emissions data that have been mistakenly
assumed to be uncontrolled emissions data due to lack of information were
identified as outlying data points and were rejected in this second step.
The fifth criterion employed in judging the usefulness of the emissions
data was the availability of fuel analysis data. This was especially true
for emissions of trace elements and SOX. The trace element content of coal
can vary by one to two orders of magnitude, and emissions are closely related
to the trace element content of the coal. No trace elements are present in
appreciable amounts in gaseous fuels; however, Ni, V, and Na are present in
appreciable amounts in some fuel oil. In order to estimate trace element
emission levels from all sources within a given category, the fraction of
each trace element exiting the system in each effluent stream must be esti-
mated. Thus, trace element emissions data from coal and oil combustion that
were not accompanied by analysis data on the trace element content of the
108
-------
fuel were not accepted. Similarly, SOX emissions are directly related to the
sulfur content of the fuel. SOX emissions data from coal and oil combustion
that did not also include information on the sulfur content of the fuel were
therefore not accepted.
The last criterion that was applied was an evaluation of the accuracy of
the sampling and analysis methods employed. In order to determine emissions
from a given site to within a factor of three, both the sampling and analysis
procedures employed must be capable of providing an accuracy which is better
than a factor of three. The list of methods available for the sampling and
analysis of general stream types and chemical classes and species is very ex-
tensive and is described in detail in two TRW reports.^»^ In general, most of
the sampling and analysis procedures recommended in these two references are
adaptations of standard EPA, ASTM, and API methods, and have an accuracy and/or
precision of ± 10 to 20 percent or better. Emissions data obtained by these
recommended methods or techniques were considered acceptable. Emissions data
obtained by methods or techniques not listed in these two references were sub-
jected to careful review and rejected if it was determined that the sampling
or analysis method employed would not be able to provide emission estimates
within an accuracy factor of three or better. Special emphasis was placed on
the review of sampling and analysis methods used for obtaining PCB, POM, par-
ticulate sulfate, and trace element emissions data. In cases where information
on the sampling and analysis methods employed was unavailable, the date of test-
ing was used as the criterion for inclusion or rejection of the emissions data
in the usable data base. Emissions data obtained before 1972 were generally
considered as unacceptable due to the probable use of unreliable sampling or
analysis procedures. The 1972 cut-off date was selected on the basis that EPA
Method 5, which has been more or less recognized nationally as the standard
method for sampling particulates, was introduced in late 1971. Furthermore,
most of the more sophisticated sampling and analysis techniques for obtaining
emissions data, and especially those for measuring pollutants for which data
are lacking (such as trace elements and particulate sulfate), were not intro-
duced or used prior to 1972.
109
-------
STEP 2
In the second step of the data evaluation process, emissions data which
had been identified in the screening step as usable were subjected to further
engineering and statistical analysis to determine the internal consistency of
the test results and the variability in emission factors.
Emissions data included in the usable data base were first categorized
according to the five-column GCA combustion system classification code and
the unit operation from which the pollutants were emitted. For NOX, the emis-
sions data were further categorized according to the method of NOX control:
no control, staged firing, low excess air, reduced load, or flue gas recircu-
lation. Emission factors for individual sites, normally expressed in the form
of Ib/MM Btu or Ib/ton, were then claculated for each pollutant /unit operating
pair. In the case of trace element stack emissions from coal and oil combus-
tion, these emission factors were calculated in the form of the fraction of
each trace element emitted to the atmosphere.
The emission factors calculated for each pollutant/unit operation pair
were evaluated in terms of consistency of test results among sites. All the
data points that were outside the upper and lower limits of reasonable data
were subjected to detailed scrutiny and discarded unless there was additional
information to reclassify the data into the correct category. The decision
whether an outlier was a reasonable result or whether it could be discarded
as being an improbable member of the group was based on the method of Dixon.
The method of Dixon is a statistical technique applicable to the rejection of
a single outlying point from a small group of data, and is described in detail
in Attachment A.
The variability of the emission factors was then calculated. The varia-
bility is defined as
where x is the estimated mean value of the emission factor,
s(x) is the estimated standard deviation of the mean, and
t is a multiple of the estimated standard deviation of
the mean value s(x).
110
-------
The value of t depends on the degree of freedom and the confidence level of
the interval containing the true mean and is given in standard statistics
texts. For the present program, the t values at 95 percent confidence level
were used in calculating the variability of emission factors.
The main thrusts in this second step are (1) to determine the emission
factors for each pollutant/unit operation pair and for each combustion source
category; (2) to discard outlying data points using the method of Dixon; and
(3) to calculate the percent variability of the emission factors. The values
calculated in this step are used in Step 3.
STEP 3
The final step in the data evaluation process involves a method developed
by the Monsanto Research Corporation (MRC) for the evaluation of data adequacy.
This quantitative method indicates where additional emissions data are needed.
The method is based on both the potential environmental risks associated with
the emission of each pollutant and the quality of the existing emissions data.
The potential environmental risks associated with pollutant emissions are
determined by the use of source severity factors, S, For emissions to the
atmosphere, the source severity, S, is defined as the ratio of the calculated
maximum ground level concentration of the pollutant species to the level at
which a potential environmental hazard exists. The simple Gaussian Plume
equation for ground level receptors at the plume centerline is the dispersion
model used for determining the ground level concentration. The potential en-
vironmental hazard level is taken to be the Threshold Limit Value (TLV) divided
by 300 for noncriteria pollutants and the ambient air quality standard for the
criteria pollutants,
The mean source severity, S, for noncriteria pollutants is calculated as
follows: .
o _ 5.5 Q
~ (TLV)h*
where Q = emission rate, g/s
TLV = threshold limit value, g/m3
h = stack height, m.
Ill
-------
For the five criteria pollutants, the equations for calculating mean
source severity, S, are as follows:
Pollutant Severity equation
Particulate S = 70 Qh~2 (3)
SOX S = 50 Qh~2 (4)
NOX S = 315 Oh"2-1 (5)
Hydrocarbons S = 162.5 Qh~2 (6)
CO S = 0.78 QtT2 . (7)
The emission rate is calculated by the following equation:
TP
Q = ~ (EF)(GPP)(YPS) (8)
where TC = total fuel consumption, tons/yr
TNP = total number of plants (or sites)
EF = emission factor, Ib/ton
GPP =. 453.6 g/lb
YPS = 3.1688 x 10~8 yr/s.
For discharges to the water, the source severity factor, S, is calcu-
lated as follows:
s = ^^
where V = discharge flow rate, m3/s
C = discharge concentration, g/m3
S = leachable solid waste generation, g/s
d
f = fraction of the solid waste to water
f_ = fraction of the material in the solid waste
V,, = river flow rate, m3/s
K
D = drinking water standard, g/m3
The mean source severity factor, S, for each pollutant/unit operation
pair was used in the evaluation of data adequacy. The method for evaluating
data adequacy is outlined below.
112
-------
Case 1: When Emissions Data Are Available and Usable
1. Determine the mean emission factor x and the variability
of the emission factor ts(x)/(x) for each pollutant/unit
operation pair. (This is done in Step 2 of the data
evaluation process.)
2. Determine the mean severity factor S for each pollutant/
unit operation pair by using the mean emission factor x,
3. If the variability in emission factor < 70 percent,
there is no need for additional data.
4. If the variability in emission factor > 70 percent and
S > 0.05, the current data base is judged to be inade-
quate and there is need for additional data.
5. If the variability in emission factor > 70 percent and
S _< 0.05, determine the severity factor Su by using the
emission factor xu: .
xu = x + ts(x)
Su is the upper bound for the severity factor S. The
current data base is judged to be adequate if Su _<. 0.05
and inadequate if Su > 0.05.
Case 2: When Emissions Data Are Not Available
1. Determine, if possible, from fuel analysis, mass balance,
and physico-chemical considerations the upper bound xu of
the emission factor x. For trace element stack emissions,
for example, xu can be determined by assuming that all the
trace elements present in the fuel are emitted through the
. stack.
2. Determine the upper bound Su of the severity factor S for
each pollutant/unit operation pair by using the emission
factor xu.
3. The current data base is judged to be adequate if Su <_ 0,05,
and inadequate if Su > 0.05.
As discussed in a recent Monsanto report,3 an allowable uncertainty in
emission factor of ± 70 percent (factor of three) would lead to an uncertainty
of less than 10 in Sca^c, which has been defined as the acceptable uncertainty
factor for S.
113
-------
The above decision criteria are based primarily on source severity factor
considerations. A discrepancy arises, however, when comparing a source type
having a few large plants with a source type having many small plants. In the
latter case, each small plant may be emitting pollutants with S < 0.05, but
the total impact on the environment may be significant because of the large
number of plants involved. To overcome this problem, other criteria based on
the air impact factors developed by MRC1* (and defined in Attachment B) are
added. These criteria have been applied to the residential source program:
1. If S < 0.05 for all pollutants emitted by the source type,
but the environmental impact of these emissions is such
that the ratio of the air impact factor for the source
type to the largest air impact factor for conventional
combustion sources exceeds 1 percent, determine those
pollutant/unit operation pairs that contribute more than
10 percent, to the value of the air impact factor for the
source type.
2. If the variability in emission factor for any pollutant/
unit operation pair determined is > 70 percent, the cur-
rent data base is judged to be inadequate and there is
need for additional data.
As a result of the application of the above data evaluation criteria,
pollutant/unit operation pairs that had been inadequately characterized were
identified to permit the planning of field tests for acquisition of additional
emissions data.
114
-------
REFERENCES
1. Hamersma, J. W., S. L. Reynolds, and R. F. Maddalone. IERL-RTP Procedures
Manual: Level 1 Environmental Assessment. TRW Systems Group. Prepared
for the U.S. Environmental Protection Agency. EPA-600/2-76-160a. NTIS
No. PB 257 850. June 1976.
2. Maddalone, R. F., and S. C. Quinlivan. Technical Manual for Inorganic
Sampling and Analysis. TRW Document 29416-6038-RU-OO. Review copy of
report prepared by TRW Systems and Energy for the U.S. Environmental Pro-
tection Agency, April 1976.
3. Source Assessment: Analysis of Error. Draft copy of report prepared by
Monsanto Research Corporation for the U.S. Environmental Protection
Agency. November 1976,
4. Eimutis, E. C., C. M. Moscowitz, J. L. Delaney, R. P. Quill, and
D. L. Zanders. Air, Water, and Solid Residue Prioritization Models for
Conventional Combustion Sources. Prepared for the U.S. Environmental
Protection Agency. EPA-600/2-76-176, p. 54, July 1976.
115
-------
ATTACHMENT A
METHOD OF DIXON FOR DISCARDING OUTLYING DATA*
The method of Dixon provides a test for extreme values using range. If
the observations in the sample are ranked, the individual values can be iden-
tified X}, X2, X3, . , ., xn_i, xn. It is immaterial whether the ranking
proceeds from high values to low or from low values to high, The Dixon
extreme-value test gives the maximum ratio of differences between extreme-
ranking observations to be expected at various probability levels and for
different sample sizes. Table C-l gives the test ratios and maximum expected
values. For samples less than about eight observations, the ratio of the
difference between the extreme and the next-to-extreme value to the total
range is compared with the tabulated values for the same sample size. If the
observed ratio exceeds the tabulated maximum expected ratio, the extreme value
may be rejected with the risk of error set by the probability level. For
samples between about 9 and 14, test the ratio of the difference between the
first and third ranking observations to the difference between the first and
next to last. For samples of 15 or more, use the ratio of the difference
between the first and third ranking observations to the difference between
the first and the second-from-last observation.
In the evaluation of the emissions data, the 0.05 probability level will
be used as the basis for discarding outlying data.
*
Volk, W. Applied Statistics for Engineers. 2nd Edition. McGraw-Hill, Inc.,
New York, 1969. pp. 387-388.
116
-------
TABLE C-l. MAXIMUM RATIO OF EXTREME RANKING OBSERVATIONS
Recommended Rank Sample
for difference size
sample size ratio (n)
x2 - xx
n - n i
n < o j
xn - xl 4
5
6
7
X3 - xl
8£ n - 1 1 1
*• n > ij o
Xn-l - x.l 9
10
11
12
• . 13
14
x3 - X!
1-1 - i r i r
n > J.J j-j
xn-2 - xl 16
17
18
19
20
Maximum ratio
Probability level
0.10 0.05 0.01
0.886
0.679
0.557
0.482
0.434
0.650
0.594
0.551
0.517
0.490
0.467
0.448
0.472
0.454
0.438
0.424
0.412
0.401
0.941
0.765
0.642
0.560
0.507
0.710
0.657
0.612
0.576
0.546
0.521
0.501
0.525
0.507
0.490
0.475
0.462
0.450
0.988
0.889
0.780
0.698
0.637
0.829
0.776
0.726
0.679
0.642
0.615
0.593
0.616
0.595
0.577
0.561
0.547
0.535
117
-------
ATTACHMENT B
AIR IMPACT FACTOR
The mathematical model proposed by MRC to rank the impacts of the com-
bustion sources is based on the impact factor defined below:
K
where I. = impact factor, persons /km2
K = number of sources emitting materials associated with
source type x
P. = population density in the region associated with the
jt» source, persons/km2
N = number of materials emitted by each source
X = calculated maximum ground level concentration of the
3 itn material emitted by the jtl1 source, g/m3
F. = environmental hazard potential factor of the
.
material, g/m3
X1 = ambient concentration of the i^ material in the
^ region associated with the jtn source, g/m3
S = corresponding standard for the ifc material
(used only for criteria emissions, otherwise
set X1. ./S equal to 1)
118
-------
APPENDIX D
SUMMARY OF EXISTING EMISSIONS DATA
119
-------
TABLE D-l. EMISSION RATES OF CO, HC, NOX, S02, PARTICULATES, AND ALDEHYDES
FROM GAS-FIRED RESIDENTIAL SOURCES
NJ
O
Test
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Text
ref-
er-
ence
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
Furnace
type*
1
1
1
1
1
1
1
1
4
2
2
2
2
2
2
2
2
3
3
2
Firing
rate
(106 J/hr)
100
100
100
100
100
100
. 100
100
162
84
84
84
84
84
84
84.
84
118
118
84
Emissions (ng/J)
CO
38
24
50
17
15
11
19
40
26
65
57
10
28
40
21
14
20
18
18
44
Total
HC
54
27
70
8
7
-
2
19
55
49
38
10
25
84
12
17
14
38
37
46
(as N02) S°2
0
59
17
48
33
110
16
35
129
82
0
16
104
45
4
72
0
66
32
34
Particulates
Filterable Total
0.67
0.74
0.27 55
0.02
0.23
0.13
0.08
1.66
0.46
0.21
0.23
0.08
0.13
0.78
0.08
0.21
0.13
0.19
0.23
0.08
Aldehydes
3
8
4
0
19
12
10
4
7
9
4
1
7
3
1
1
.7
2
1
5
(continued)
-------
TABLE D-l (continued).
Test
No.
21
22
23
24
25
26
27
28
29f
30
31
32
33
34
35
36
37
38
39
40
Text
ref-
er-
ence
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
13
Furnace
type*
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1
1
Firing
rate
(106 J/hr)
84
84
84
84
84
84
84
84
84
84
84
84
84
84
84
84
84
84
100
100
Emissions (ng/J)
CO
60
86
62
20
24
43
21
17
75
70
22
37
25
34
15
24
13
29
31
19
Total
HC
-
-
77 .
16
34
61
44
27
-
71
8
79
18
20
28
10
28
19
-
18
(as N02) S°2
8
17
104
12
92
65
62
33
90
40
96
119
59
13
34
65
41
70
74
108
Particulates
Filterable Total
0.50
0.13
0.36 48
0.34
0.48
0.27
0.17
0.04 19
0.61
0.69
1.47
0.36
1.37 11
0.27
0.23
0.29
0.08
0.19 - 0
0.84 14
0.38
Aldehydes
4
-
6
8
15
6
8
8
8
0
0
0
0
10
12
3
1
7
0
10
(continued)
-------
TABLE D-l (continued).
NJ
Test
No.
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Text
ref-.
er-
ence
13
13
13
13
13
13
13
13
13
13
14
14
14
14
14
14
14
14
14
14
Furnace
type*
6
7
2
2
5
2
7
5
2
2
8
8
8
8
8
8
10
8
10
10
Firing
rate
(106 J/hr)
84
113
84
84
79
84
113
79
84
84
140
140
140
140
140
140
140
140
140
140
Emissions (ng/J)
I CO
43
46
17
40
55
103
23
53
17
63
6
10
9
27
19
29
11
14
10
16
Total
HC
27
38
30
-
80
64
26
52
35
-
10
13
2
37
10
103
11
26
8
30
Particulates
(as N02) S°2
41
84
39
10
21
53
46
63
57
34
15
89
40
66
75
43
40
0
0
71
Filterable
0.27
0.93
0.17
0.55
0.34
1.05
0.13
0.32
0.29
0.21
Or86
0.27
0.02
0.00
0.08
0.02
0.04
0.02
0.00
0.00
r\_LU.cuy u.co
Total
8
17
13
8
9 17
8
11
15
11
1
87
184
4
18
5
5
5
3
5
4
(continued)
-------
TABLE D-l (continued).
Test
No.
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
Text
ref-
er-
ence
14
14
14
14
14
14
14
14
14
14
14
14
14
14
14
14
14
14
14
14
Furnace
type*
8
8
8
8
8
8
8
8
8
8
8
8
8
10
9
8
8
8
8
10
Firing
rate
(106 J/hr]
140
140
140
140
140
140
140
140
140
140
140
140
140
140
38
140
140
140
140
140
Emissions (ng/J)
1 CO
15
20
5
12
14
13
31
15
13
451
38
15
11
8
10
20
15
13
15
21
Total
HC
31
24
17
5
102
4
4
22
14
435
38
13
25
12
23
19
36
36
136
42
(as N02) S°2
18
22 . -
25
40
315
50
48
44
0
20
37
32
79
40
48
118
90
126
230
39
Particulate
Filterable Total
0.02
0.02
0.02
0.17
0.11
0.13
0.17
0.02
0.00
0.15
0.02
0.06
0.02
0.06
0.25
0.34
0.02
0.02
0.32
0.25
Aldehydes
1
12
1
1
0
5
4
3
4
3
5
6
5
2
5
3
3
5
5
6
(continued)
-------
TABLE D-l (continued).
Test
No.
81
82
83
84
85
86
87
88
89
90
91
92*
93
94
95*
96
97§
98
99
100
Text
ref-
er-
ence
14
14
14
14
14
14
14
14
14
14
14
14
14
14
14
14
14
14
14
14
. Furnace
type*
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
Firing
rate
(106 J/hr)
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
Emissions (ng/J)
CO
52
11
12
11
40
18
26
31
241
22
28
8
11
11
14
733
12
16
26
19
Total
HC
13
30
141
12
36
15
42
48
29
147
51
42
6
11
30
315
14
28
27
15
NOX
(as NQ2)
149
83
0
63
26
59
39
144
10
39
63
98
22
81
63
57
16
0
220
47
Particulates
Tt
Filterable Total
0.13
0.11
0.97
0.76
1.77
0 . 04
0.06
0.21
0.46
0.48
0.00
0.11
0.00 0.11
0.04 0.38
0.61
1.00 0.40
1.39 0.08
0.00 0.13
0.00 0.17
0.00 9.19
Aldehydes
4
5
15
11
16
34
28
5
4
60
54
153
5
21
53
11
5
11
29
5
(continued)
-------
TABLE D-l (continued).
to
Ul
Test
No.
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
Text
ref- Furnace
er- type*
ence
9 11
9 12
9 13
11 14
11 15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
riring
rate
(106 J/hr) CO T°£al
106 5 0
106 24 0
106 8 2
158 10 1
63
35 -
31 - -
30 - -
41 - -
45 - -
75 -
51 - -
116
40 -
133 -
59 - -
133
44 -
8 -
129
Emissions (ng/J)
Particulates
x "0
(as N02) Fiiterable Total
13 -
18 -
17
35 0.00 2.12 6.5
37 0.00
57
44
44 . -
53
57
31
49
40
40
40 -
44
40
53
18
57 - - -
Aldehydes
-
-
-
-
-
-
-
-
-
-
-
-
-
-
.
-
-
'
.
-
(continued)
-------
TABLE D-l (continued).
Test
No.
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
Text
ref- Furnace
er- type*
ence
15
15
15
15
15
15
is
15
15
15
15
15
15
15
15
15
Firing
rate
(106 J/hr)
44
32
49
135
114
89
44
160
38
38
145
119
45
39
89
228
Emissions (ng/J)
_• Particulates
rr> Total N°x -n
HC (as N02) 2
40
31 - - -
31 - -
44 - -
44 _
57
49 __
35
66
35 - -
- - 35 - -
40
49 _ _
31
- - 31
80
Aldehydes
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
(continued)
-------
TABLE D-l (continued).
*
Furnace types: 1 - Thatcher (Comet) Forced Warm Air: 95,000 Btu input; 76,000 Btu output.
2 - Crane Sunnyland Forced Warm Air: 80,000 Btu. input; 64,000 Btu output.
3 - Crane Sunnyday Hot Water Boiler: 112,000 Btu input; 89,600 Btu output.
4 - Crane Hot Water Boiler: 154,000 Btu input; 123,200 Btu output.
5 - Crane Hot Water Boiler: 75,000 Btu input; 60,000 Btu output.
6 - Crane Hot Water Boiler: 80,000 Btu input; 64,000 Btu output.
7 - Trimline Forced Warm Air: 107,000 Btu input; 85,000 Btu output.
8 - New Yorker Gas-Fired Boiler Model RG 133: 133,000 Btu input; 107,000 Btu
output; 96,500 Btu net output.
9 - Rheem Hot Air System with 36,000 Btu/hr input hot water heater.
10 - Furnace 8 with a 36,000 Btu/hr output hot water heater.
11 - Williamson Furnace: 20 percent excess air.
12 - Bryant Boiler: 40 percent excess air.
13 - Bryant Furnace: 60 percent excess air.
14 - Forced Air Sectional Furnace: 137,000 Btu/hr firing rate.
15 - Forced Air Drum Furnace, Multiport Burner: 150,000 Btu/hr.
Maladjusted furnace.
'Questionable aldehyde sample.
§
Questionable S02 sample.
-------
TABLE D-2. EMISSION RATES OF CO, HC, NOx, S02, AND PARTICULATES, AND BACHARACH SMOKE NUMBERS
FROM OIL-FIRED RESIDENTIAL SOURCES
NJ
co
Test
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Text
ref-
er-
ence
2
2
2
2
2
2
2
2
2
2
2
2
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
Burner
type*
FR
C
FR
S
FR
FR
C
C
C
C
FR
C
C
C
C
C
C
FR
FR
FR
C
C
C
C
C
C
C
FR
Firing
rate
(106 J/hr)
177
154
199
263
186
133
111
162
202
145
126
89
162
109
443
180
160
193
310
216
443 .
414
66
142
177
220
517
236
Burner
age
1
4
1
2
5
2
18
18
20
5
3
11
12
13
9
30
15
1
12
1
1
15
1
3
4
15
1
1
Sys-
tem
age
21
4
1
9
5
8
18
18
20
5
20
11
12
13
9
30
15
1
12
1
1
30
1
3
15
20
1
8
Stack gas
r f f-f '
cien- Temper-
cy • .fature
(°C)
243
341
254
271
310
282
188
299
277
285
371 .
227
288
288
360
354
246
210
210
285
343
260
371
310
293
243
360
271
Per-
cent
C02
9.2
7.1
8.4
7.1
9.1
6.6
9.4
7.5
7.9
7.0
6.1
9.4
7.2
6.7
9.0
7.0
4.5
8.3
10.0
9.4
12.5
7.5
9.2
7.4
7.6
6.6
9.5
6.7
Per-
cent
02
9.1
12.0
9.0
11.1
8.1
11.9
7.1
10.3
10.5
11.4
12.6
7.4
10.0
10.8
9.5
10.6
14.5
8.8
6.9
7.6
3.5
10.8
8.2
10.3
7.0
11.2
7.5
11.3
Per-
cent
excess
air
68
120
75
109
64
126
55 •
95
94
114
143
55
98
112
73
105
216
75
49
59
22
101
. 64
97
74
119
57 •
118
Emissions
CO
2
125
5
3
6
94
11
7
21
186
893
9
0.5
8
4
8
231
0
0
0.6
0
0
0
21
0
2
3
4
HC
2
5
2
2
-
5
1
3
1
5
-
1
1
2
1
0
146
1
1
1
1
2
1
3
2
3
3
3
NOX
68
83
47
64
68
86
62
74
64
76
103
46
64
69
51
57
44
53
46
45
37
50
44
68
45
54
68
61
(ng/J) '
S02
-
-
-
-
-
-
-
-
-
-
-
-
114
73
62
126
-
97
123
143
146
112
109
70
138
97
62
65
Total
partic-
ulates
9
9
11
18
17
25
35
17
17
26
71
19
18
14
17
13
96
11
7
9
12
14
37
17
19
23' .
30
13
Bacharach
smoke
number
0.3
0.4
1.0
0.5
1.0
1.3
9.0
0.2
0
0.2
oily
2.7
3.0
3.0
1.0
7.0
-
1.0
3.0
4.0
7.0
2.0
7.0
3.0
6.0
8.0
6.0
2.0
(continued)
-------
TABLE I>-2 (continued) .
ro
vo
Test
No.
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
Text
ref-
er-
ence
11
11
11
11
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9 .
9
9
. 9
9
9
9
Burner
type*
S
C
C
C
C
CI
CI
CI
CI
CI
FR
FR
FR
FR
FR
FR
FR
FR
FR
C
C
C
C
C
C
C
C
C
Firing „ Sys-
6 Burner 3
rate tern
(106 J/hr) ag age
282 3 15
177 3 3
310 20 20
86 20 20
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
111
Effi-
cien-
cy
-
-
-
-
77
71
75
72
76
83
80
81
80
73
80
70
75
83
78
. -
-
74
-
-91
91
82
-
-
Stack gas
Temper- Per-
ature cent
(°C) C02
288 9.5
293 8.4
338 7.6.
138 3.8
9.9
8.2
10.8
9.1
9.4
12.6
10.9
11.6
10.7
9.3
13.5
7.9
10.7
12.2
10.2
10.3
9.9
8.0
9.3
12.6
13.7
12.6
4.4
4.4
Per-
cent
02
8.2
9.4
10.2
17.0
7.2
9.5
5.4
8.5
7.5
3.7
5.3
4.6
5.7
7.9
2.7
9.3
5.8
3.7.
6.2
6.1
7.2
9.3
7.9
3.7
2.6
3.7
1.0
1.0
Per-
cent
excess
air
61
78
94
346
53
80
40
66
60
20
38
31
34
63
16
86
42
18
48
47
53
86
63
20
11
20
6
6
Emissions (ng/J)
CO
35
26
39
311
12
14
14
14
7
12
12
12
7
14
6
23
12
5
9
91
7
65
13
12
125
58
13
13
HC
4
2
5
338
1
1
1
1
1
1
1
1
0
1
2
2
1
0
0
4
3
3
2
5
0
<4
5
1
Total
N0x s°2 partic-
ulates
55 76 11
56 53 45
38 71 14
15 78 65
26 -
30 -
39 -
29 -
39 -
29 -
18
33 -
41 -
27 -
23 -
34 -
27 -
27 -
34 -
21 -
30 - -
29 -
28 -
41 -
3 -
11 -
9 -
10 -
Bacharach
smoke
number
6.0
4.0
4.0
-
2.9
1.3
3
2
2
1.2
2
2.5
2.6
1.1
3.5
2.2
1.7
1.4
2.1
-
1.8
1.45
-
2.5
-
1.8'5
0.2
-
(continued)
-------
TABLE D-2 (continued).
U)
o
Test
No.
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
Text
ref-
er-
ence
9
9
12
12
12
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
Burner
type*
C
C
FR
FR
FR
FR
C
FR
C
C
FR
FR
C
FR
C
C
FR
S
. S
'FR
FR
c
FR
c
c
c
c
c
Firing
rate
(106 J/hr)
111
111
148
143
148
244
199
244
199
148
177
222
148
185
199
222
199
222
177
148
185
185
258
222
185
199
148
177
Burner
age
-
-
-
-
-
15
25
6
-
18
3
3
-
4
15
10
3 •
20
21
3
2
26
5
-
20
-
26
20
Sys-
tem
age
-
-
-
-
-
40
48
13
-
18
20
20
-
4
20
-
3
21
21
3
26
26
25
-
45
-
26
-
Effi-
cien-
cy
84
-
-
-
-
56
47
69
74
70
75
72
71
72
71
72
72
73
78
73
75
75
75
73
71
76
65
64
Stack gas _ Emissions (ng/J)
Temper-
ature
(°C)
.
-
-
-
-
477
396
308
309
297
177
243
310
352
271
329
332
196
235
339
243
231
371
366
361
235
320
310
Per- Per- ° Total
cent cent ex"ss CO HC NOx S02 partic-
C02 02 ulates
1.8 3.7 25 47 6 28 -
8.9 8.5 68 10 11
-.- 5 1-- 5
12 1 - - 4
--- 61--3
7.0 _____
4.5 _____
7.0 _____
9.0 _____
7.0 - - _____
5.0 ______
6.0 - - - -
7.5 - - _____
12.0 - _____
6.5 _____
8.5 - - _____
8.8 - - _____
5.0 _____
8.5 - - -
10.5- - - ___
5.0 - - - -
7.0 - - . -
11.5 - _____
10.0 - - - - - . -
9.0 - - _____
7.5 ' - - - _ _ _
6.0 - - - - _ _
5.5 - - ____•-
Bacharach
smoke
number
-
-
-
-
-
.7.0
7.0
2.0
0
3.0
0
0
3
1
4
1
0
9
0
3
4
1
1
1
9
0
0
1
(continued)
-------
TABLE D-2 (continued).
Test
No.
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
Text
ref-
er-
ence
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
17
17
17
17
17
17
17
17
17
17
17
17
17
Burner
type*
FR
S
C
C
C
C
C
C
FR
FR
C
C
C
C
FR
FR
C
C
C
FR
C
R
C
C
FR
C
S
S
Firing
rate
(106 J/hr)
177
222
185
148
185
295
185
148
199
185
185
332
126
665
148
199
148
126
148
295
222
244
185
11.1
244
332
443
177
Burner
age
3
15
35
25
20
30
20
30
4
4
-
5
40
-
-
-
-
-
-
-
-
-
-
•
-
-
-
-
Sys-
tern
age
20
15
35
-
35
30
20
35
-
-
' -
40
40
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Effi-
cien-
cy
75
72
58
76
82
77
67
67
74
81
71
61
69
73
-
79
-
-
-
70
63
-
72
-
74
80
79
78
Stack gas „ Emissions (ng/J)
Temper-
ature
(°C)
293
432
271
263
177
316
318
268
359
291
275
359
338
299
-
233
-
-
-
282
285
-
178
-
383
143
256
260
Per- Per- Cent Total
cent cent ex^ess CO HC NO^ S02 partic-
C02 02 alr ulates
9.0 - - -----
11.5 - - _ _ _ -
4.0 - - -----
8.5 - - ------
8.5 - - _____
11.0 - - _____
6.5 - - -___-
5.5 - - _____
10.5 - - - -
13.0 - - _____
6.5 - - _____
6.0 _____
7.5 _____
8.0 _____
__'_ _____
9.5 - - _____
___ _____
___ _____
___ _____
6.5 - - _ _ _ _
5.0 - - _____
_'._ _ ___
7.0 - - _____
___ _____
11.0 - _____
6.0 _____
10.0 - - _____
9.5 - - -----
Racharach
smoke
number
0
9
1
6
2
1
5
'6
0
5
8
0
0
0
-
0
-
-
-
0
2
-
0
-
0
3
2
5
(continued)
-------
TABLE D-2 (continued).
to
NJ
Text
Test ref- Burner
No. er- type*
ence
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
*
Burner
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
types:
FR
C
C
C
FR
FR
C
C
S
C
C
C
S
FR
C
S
FR
C
FR
C
C
S
FR
C
C
: FR
C
S
CI
R
Firing
5 Burner
rate
(106 J/hr) age
244
244
295
177
148
148 . -
199
185
185
244
199
295
222
148
148
222
369
185
244
199
199
199
126
177
199
- Flame Retention
- Conventional
- Shell
- Conventional with £
- Rotary.
Sys- Effi-
tem den-
age cy
-
-
72
74
80
62
72
74
-
75
60
71
73
76
79
78
75
73
67
78
77
-
-
-
i combustion
Stack gas _ Emissions (ng/J)
Temper-
ature
(°C)
-
-
271
346
206
427
326
239
-
286
318
337
352
256
234
284
234
296
403
243
287
-
-
-
improving
Per- Per- C6nt Total
cent cent ex^ess CO HC HO*. S02 partic-
C02 02 3lr ulates
--- _____
--- _____
7.0 - - _____
11.0 - - _____
8.5 - - ______
7.5 - - _____
9.0 _____
6.5 - - -----
'--- _____
8.5 - - -----
5.0 - - -
8.0 - - _____
9.5 - - _____
8.0 - - -
9.0 - - _____
10.5- - - --_
7.0 - - - _ _ _
8.0 - - _-___.
8.5 - - - - - -
8.5 - - -----
10.0 - - - -
___ _____
_ _____
- - - - -
device
Bacharach
smoke
number
:
-
0
0
0
3
0
0
-
0
1
0
0
0
0
2
1
0
2
1
0
-
-
-
-------
APPENDIX E
DATA REDUCTION PROCEDURE
133
-------
DATA REDUCTION PROCEDURE
Stack emissions data reported from field measurements or laboratory
analyses are often expressed in terms of volume concentration (ppmv) or mass
concentration (mg/m3, ug/m3). To convert these emissions data to the emis-
sion factor form, the following data reduction procedure1 is used.
The number of gm moles of flue gas per gm of fuel can be computed using
the fuel composition analysis and effluent 02 concentration:
A.762 (nc + ng) + 0.9405 (nR) - 1,881 (nQ) F
n =
1 - 4.762 (02/100) 1 - 4.762 (02/100)
where -n = .gm moles of dry effluent/gm of fuel under actual operating
conditions
n. = gm moles of element j in fuel per gm of fuel
02 = volumetric 02 concentration in percent
F = gm moles of dry effluent/gm of fuel at stoichiometric
conditions.
The average values of F for natural gas and various liquid fuels are
given in Table E-l. The value of F for coal must be computed on an individual
basis because of the variation in the elemental composition of different coals,
For emission species measured on a volumetric concentration basis (ppmv),
the emission factor expressed as kg/GJ can be computed using the following
equation:
( Volumetric \
\n • f
(Emission) ,t /pn )Concentration/
Factor } (kS/GJ)
/ Fuel \ ,. T/]• . .. 1 - 4.762 (02/100)
\ TI j TT i l (kJ/kg fuel) ^
(Heating Value) 6
134
-------
where s = subject emission species
M = molecular weight of species s.
S
For emission species measured on a mass concentration basis (mg/m3 or
yg/m3) at 20°C, the emission factor expressed as kg/GJ can be computed using
the following equation:
( Mass )
(Emission) ( /QJ) =
1 Factor }
(Concentration)
(yg/m3) x F x 0.02404
( Fuel
(Heating Value
} (kJ/kg fuel) X - 4'762 (°2/100)
The higher heating values of natural gas and various liquid fuels are also
given in Table E-l.
TABLE E-l. FUEL COMPOSITION AND COMBUSTION CHARACTERISTICS*
Fuel
Natural gas
No. 2
Distillate oil
Kerosene
Residual oil
Elemental composition
nc
0.06221
0.06994
0.06994
0.7160
ns
0
0.00006
0
0.00031
nH
0.23116
0.13889
0.15873
0.10913
no
0.00040
0.001125
0
0.00125
F
Factor
0.51215
0.45983
0.48234
0.44037
Higher
heating
value
(kJ/kg)
53,310
45,040
47,710
43,760
The elemental composition and higher heating value data are obtained from
Reference 2.
It should be noted that the data reduction procedure described here sig-
nificantly minimizes errors introduced in data reduction by eliminating terms
which are subject to large measurement errors, such as stack velocity and tem-
perature measurements. The only stack parameter, needed in data reduction is
the volumetric Q£ concentration, which usually can be determined with great
accuracy by gas chromatography.
135
-------
EXAMPLE CALCULATION
The NOX emission from a gas-fueled gas turbine is reported to be 200 ppmv
at an 02 effluent concentration of 15 percent. The emission factor for NOX
(as N02) in kg/GJ is calculated as follows:
/„.,., - v - 200 x 0.51215 x 46.0 1
Emission Factor \ = x = 0>309
Uor NOX (as N02)f 53310 - 1 - 4.762 x 0.15
REFERENCES
1. Coppersmith, R. M., R. F. Jastrzebski, D. V. Giovanni, and S. Hersh.
Con Edison's Gas Turbine Test Program: A Comprehensive Evaluation of
Stationary Gas Turbine Emission Levels. Paper presented at the 67th
Annual Meeting of the Air Pollution Control Association, Denver,
Colorado, June 9-13, 1974.
2. Steam/Its Generation and Use. Revised 38th Edition. The Babcock and
Wilcox Company, New York, New York. 1975.
136
-------
APPENDIX F
GAS- AND OIL-FIRED RESIDENTIAL SOURCE
LABORATORY DATA
137
-------
TABLE F-l. SSMS ANALYTICAL DATA: SITE 100
CO
uEWENl
U
TM
61
PB
TL
AU
IR
OS
RE
w
HF
LU
Y6
TM
ER
MO
OY
IB
GO
EU
SM
NO
HH
Ct
LA
BA
CS
1
Tt
SB
SM
CO
PD
Rh
r FILTER
CATCH
(*6)
< 0.0001
< 0.0001
< 0.0001
< 0.016
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0003
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0007
< 0.0001
< 0.0001
< 0.0001
< C.0001
< 0.0001
0.0004
< 0.0001
< 0.0001
< 0.0001
C.0056
< 0.0001
< 0.0001
< 0.0001
0. 001 0
0.0015
< O.OU13
< 0.0001
< ^.0001
XAD
RE3IN
IMG)
< 0.030
< 0.044
< 0.014
< 0,42
< 0.017
< 0.023
< 0.033
< 0.038
< 0.024
< 0.32
< 0.040
< 0.0094
< 0.014
< 0,0084
< 0.025
< 0.0081
< 0.014
< 0.0038
< . 0.014
< 0. 0064
< 0.024
< 0.031
< 0.0054
< 0.031
0.0059
< 0.34
< 0.0013
0.022
< 0.018
0.039
< 1.0
< 0.15
< 0.018
< 0.0061
COMPOSITE
SAMPLE
<«6)
< 0.0035
< 0.0051
< 0.0016
< 0.14
< 0.0020
< 0.0026
< 0.0040
< 0.0044
< 0.0028
< 0.0037
< 0.0046
< 0.0011
< 0.0016
< 0.0010
< 0.0028
< 0.0009
< 0.0016
< 0.0004
< 0.0016
< 0.0007
< 0.0028
< 0.0035
< 0.0006
< 0.0008
< 0.0009
0.045
< 0 . 0 n 0 3
«. 0.0024
< u.0020
< 0,0029
< 0.19
< u.016
< C.0020
< 0.0007
TOTAL
SASS
CMG)
< 0,034
< 0.049
< 0.015
< 0,56
< 0.019
< 0.025
< 0.039
< 0.043
< 0.627
< 0.33
< 0.045
< 0.011
< 0.015
< 0.0094
< 0.028
< 0.0097
< 0.016
< 0.0043
< 0.016
< 0.0071
< 0.027
0.0004 10 0,034
< 0.0060
0,0001 TO 0.032
0.0059
0.050 TO 0.34
< 0.0016
0.022
< 0.020
0,040
0.0015 TO 1.2
< 0.17
< 0.020
< 0 . 0 u 6 9
EMISSION
FOUND
(MG/OSCM)
< 0.0010
< 0.0015
< 0.0005
< 0.018
< 0.0006
< 0.0006
< 0.0012
< 0.0013
< 0.0006
< 0.0100
< 0.0014
< 0.0003
< 0.0005
< 0.0003
< 0.0006
< 0.0003
< 0.0005
< 0.0001
< 0.0005
< 0.0002
< 0.0008
0.0001 TO 0.0010
< 0.0002
0.0001 TO 0.0010
0.0002
0.0015 TO 0.0100
< 0.0001
0.0007
< 0.0006
0.0012
0.0001 TO 0.037
< 0.0052
< 0.0006
< 0.0002
-------
TABLE F-l (continued).
u>
.CKEN1
*U
f*O
NB
/«
Y
SR
RB
BR
SE
AS
GE
GA
ZN
cu
NI
CO
Ft
MH
CR
V
Tl
CA
R
CL
S
H
SI
AL
»G
NA
f
b
HE
r FILTER
CATCH
(MG)
< 0.0001
0.0002
< 0.0001
0.0009
< 0.0001
< 0.0004
0.0003
0.0025
0.0007
0.0004
< 0.0001
< 0.0001
< 0.32
< 0.15
< 0.41
0.0019
< 0.077
0.0031
0.0014
0.0002
0.0075
0.069
< 0.12
0.20
< 0.37
0.014
< 1.3
MC
< 0.098
MC
0.0059
0.032
< 0.0001
XAD
RESIN
(MG)
< 0.023
< 0.090
0.0026
0.074
0.087
< 0.16
< 0.0086
< 0.096
< 0.079
< 0.047
< 0.0084
< 0.0034
< 1.2
< 26.
< 3;7
< 0.035
< 5.8
0.076
< 1.5
< 0.022
< 0.053
< 5.3
< 1.9
< 7.7
< 9.3
< 1.9
230.
< 1.6
< 1.2
< 11.
< 0.052
0.99
< 0.0017
COMPOSITE
SAMPLE
(MG)
< 0.0626
< 0.051
< 0.0003
0.0040
< 0.0047
< 0.0088
< O.OOU4
< 0.035
< 0.0014
< 0.0054
< 0.0010
< 0.0004
« 0.67
< 0.069
< 0.36
< 0.0061
< 0.67
< 0.052
< 0.35
< 0.0035
< 0.041
< 0.62
< 0.35
< 6.2
< 16.
< 0.16
< 12.
< 0.47
< 0.13
< 0.39
< 0.027
< 0.0049
< 0.0004
TOTAL
SASS
(MG)
< 0.025
0.0002 TO
0.0028
0.079
0.087
< 0.17
0.0003 TO
0.0025 TO
0.0007 TO
0.0004 TO
< 0.009U
< 0.0001 TO
< 2.2
< 26.
< 4.5
0.0019 TO
< 6.5
0.0/9
0.0014 TO
0.0002 TO
0.0075 TO
0.069 TO
< 2.3
0.14
0*013
0.13
o.oao
0.053
0.0038
0.042
1.6
0.025
0.093
6.0
0.20 TU 14.
< 26.
0.014 TO
230.
< 1 ."
0.0059 [0
1.0
< 0.0001 TO
2.1
0.078
0.0022
EMISSION
FOUND
(MG/DSCM)
< 0.0006
< 0.0001 TO 0.0043
< 0.0001
0.0024
0.0027
< 0.0051
< 0.0001 TO 0.0004
< 0.0001 TO 0.0041
< 0.0001 TO 0.0025
< 0.0001 TO 0.0016
< 0.0003
< 0.0001 TO 0.0001
< 0.068
< 0.79
< 0.14
< 0.0001 TO 0.0013
< 0.20
0.0024
< 0.0001 TO 0.055
< 0.0001 TO 0.0008
0.0002 TO 0.0029
0.0021 TO 0.18
< 0.072
0.0060 TO 0.43
< 0.79
0.0004 TO 0.064
7.1
< 0.043
0.0002 TO 0.0024
0.031
< 0.0001 TO 0.0001
INDICATES A MAJO» COMPHNE*JT OF THE
INDICATES THAT THE IOIAL AMD EMISSION v«LUts *EPe >.oi
CALCULATED 0«lN& TU Tht ^HtSENCf Of AN. N (. CONCENTRATION.
-------
TABLE F-2. SSMS ANALYTICAL DATA: SITE 101
EMENT
U
TM
HI
PB
TL
AU
IH
OS
Rfc
M
HF
LU
YB
TM
ER
HU
OY
TU
GO
tu
SM
NO
PK
Ct
LA
bA
CS
I
Tt
SB
SN
CD
PU
HM
XAD
RESIN
(MG)
< 0.014
< 0,020
< 0.0061
< 0.40
< 0,0076
< 0,0100
< 0.016
< 0.017
< 0.011
< 0.49
< 0.018
< 0.0042
< 0.0061
< 0.0038
< 0,011
< 0.0036
< 0.0063
< 0,0017
< 0.0063
< 0.0028
« 0.011
< 0.014
< 0.0024
< 0.023
< 0.0035
< 0.34
0.0020
< 0.0094
< 0.0079
0.019
1.4
< 0.0096
< 0.0079
< 0.0027
COMPOSITE
SAMPLE
(MG)
< 0.0018
< 0,0026
< 0,0008
< 0.35
< 0.0010
< 0.0014
< 0,0021
< 0.0023
< 0,0014
0,0067
< 0.0024
< 0.0006
< 0.0008
< 0.0005
< 0.0015
< 0.0005
< 0.0008
< 0.0604
< 0.0008
< 0.0004
< 0.0015
< 0 . 0 0 1 ft
< 0.0004
< 0.0004
< 0.0005
< 0.045
< 0.0004
< 0.0013
< 0.0011
< u.073
< 0.44
0.44
< 0.0010
< 0.0004
TOTAL
SASS
(MG)
< 0.015
< 0.022
< 0.0069
< 0.75
< 0.0087
< 0.012
< 0.018
< 0,019
< 0.012
0.0067 TU 0.49
< 0.020
< 0.0048
< 0.0070
< 0.004.3
< 0.013
< 0.0041
< 0.0072
< 0.0022
< 0.0071
< 0.0033
< 0.012
< 0.016
< C . 0') 2 9
< O.u23
< 0.0040
< 0.39
0.01)20
< 0.011
< 0.0090
0 . 0 1 V 'TO 0.073
i ^*
0.44
< 0.0089
< U.OU32
EMISSION
FOUND
.(MG/DSCM)
< 0,0005
< 0,0007
< 0.0002
< 0,025
< 0.0003
< 0.0004
< 0.0006
< 0,0006
< 0.0004
0.0002 TU 0.017
< 0.0007
< 0.0002
< 0.0002
< 0.0001
< 0.0004
< 0.0001
< 0.0002
< 0.0001
< 0.0002
< 0.0001
< 0.0004
< 0.0005
< 0,0001
< 0.0006
< 0,0001
< ' 0.013
< 0.0001
< 0.0004
< 0.0003
O.OOOo TU 0.0025
0.047
0.015
< 0.0003
< 0.0001
-------
TABLE F-2 (continued)
EMEN1
RU
MO
N8
ZR
Y
SH
RB
BR
SE
AS
GE
GA
ZN
cu
Nl
CO
FE
MN
CR
V
CA
K
CL
S
P
SI
AL
MG
NA
Y
0
BE
LI
XAO
RESIN
(MG)
< 0.0100
< 0.20
< 0.0012
0,0046
< 0.0016
< 0.071
< 0.012
< 0.042
< 0.023
< 0.015
< 0.0038
< 0.0015
< 0.37
< 16.
< 0.69
< 0.0053
< 1.9
< 0.045
< 0.28
< 0.0045
< 5.3
< 0.84
< 3.5
< 6.2
< 1.3
< 0.7
< 1.8
< 0.80
< 5.1
< 0.069
0.063
< 0.0005
< 0.0026
COMPOSITE
SAMPLE
(MG)
< 0.0013
0,053
0.0008
0.0062
< 0.0094
< o.ose
< 0,0051
< 0,059
< 0,0007
0.086
< 0.0005
< O.U004
< 11.
< 3.5
< 6.7
0,49
< 15.
0,45
< 2.4
< 0.013
< 4. a
7.2
4.9
• *
< 0,80
< 28.
< 5.2
< 1.1
< 2.0
46.
< 0.17
< 0.0002
< 0.0035
TOTAL
SASS
(MG)
< 0.011
0.053 TO 0.20
0.0008
• 0.011
< 0.0023
< 0.100
< 9.017
< 0.100
< 0.024
0.086
< 0.0043
< 0,0020
< 11 .
< 20.
< 7.4
0.49
< 17.
0.45
< 2.7
< 0,017
< 9.7
7.2
3.9
* *
< 2.1
< 35.
< 7.0
< 1.9
< 7.1
ub.
O.&oi T(J 0.17
< 0 . 0 0 0 b
< 0 , 0 0 o (.•
EMISSION
FOUND
(MG/DSCM)
< 0.0004
0.0018 TO 0.0066
< 0.0001
0.0004
< 0.0001
< 0.0034
< 0.0006
< 0.0034
< 0,0008
0.0029
< 0.0001
< 0.0001
< 0.37
< 0.66
< 0.25
0.017
< 0.56
0.015
< 0.091
< 0.0006
< 0.33
0.24
0.13
110.
< 0.071
< 1.2
< 0.24
< 0.062
< 0.24
1 .6
0.0021 TO 0.0057
< 0.0001
< 0.0002
*• INOICAItS A VALUE E
-------
TABLE F-3. SSMS ANALYTICAL DATA: SITE 102
S3
EMENT
U
TH
BI
P8 '
TL
AU
IR
OS
RE
M
HF
LU
Yb
TM
ER
HO
OY
TB
GU
tu
SM
MJ
PR
CE
LA
BA
CS
I
TE
Sb
SN
CD
PU
KM
XAD
RESIN
C"G>
< 0.014
< 0.020
< 0.0061
< 0.54
< 0.0076
< 0.0100
< 0.016
< 0.017
< 0.011
< 0.014
< 0.018
< 0.0042
< O.OOol
< 0.0038
< 0.011
< 0.0036
< 0.0063
< 0.0017
< 0.0063
< 0.0026
< O.ul 1
0.05V
C . 0 1 6
0.15
0.076
0.2o
< 0.0007
< 0 . 0 0 9 u
< 0.00/9
< 0 . 0 1 o
< 0. lo
0.100
< 0.0079
« 0.0027
COMPOSITE
SAMPLE
(WG)
< 0.0074
< 0.01 i
< 0.0033
< 0.14
< 0.0041
< .. 0.0.055
< 0.0085
< 0.0092
< 0,0058
< 0.0078
< 0,0097
< 0.0023
< 0.0033
< 0.0020
< o.oo&o
< 0.0020
< 0.0044
< 0.0099
< 0.0034
< 0.0015
< O.OL'59
< 0.007U
< 0.0013
. < 0.0018
< 0.0019
0.040
0.0001.
< 0.0051
< O.OOU3
0.029
< 0.12
< 0.12
< 0.0043
< 0 . 0 0 1 5
TOTAL
SASS
0.021
.0,030
0.009u
o!oi2
0.016
0.024
0.026
0.017
0.022
0.028
0,0065
0.0095
0.0058
0.017
0.0056
0.0097
0.0026
0.0097
0,
0,
0 ,
0! 7
055
0.018
0.15
u . 0 / o
0.30
0.0001 TO
< 0 . 0 1 u
< 0.012
0.029
< 0.28
0.100
< 0.012'.
< 0 . 0 o u 2
0 . 0 0 C 7
EMISSION
FOUND
(MG/OSC*)
< 0,0006
< 0,0009
< 0,0003
< 0,021
< 0.0004
< 0,0005
< 0.0007
< 0.0008
< 0.0005
< 0.0007
< 0.0006
< 0.0002
< 0.0003
< 0,0002
< 0.0005
< 0.0002
< 0.0003
< 0.0001
< 0.0003
< 0.0001
< 0.0005
0.0017
0.0006
0.0047
0.0023
0.0091
0.0001 TO 0.0001
<' 0.0004
< 0.0004
0.0009
< 0.0085
0.0031
< 0.0004
< 0.0001
-------
TABLE F-3 (continued)
.EMEN
RU
MO
NB
ZR
Y
SR
R6
BK
SE
AS
6E
GA
in
cu
NJ
CO
Ft
MN
CR
V
TI
CA
K
CL
S
P
SI
AL
MG
i.A
F
B
at
LI
T XAD
RESIN
-------
TABLE F-4. SSMS ANALYTICAL DATA: SITE 103
.EMENl
U
IH
BI
PB
TL
AU
IR
OS
RE
w
HF
LU
YB
TM
tR
MO
DY
Tb
GO
EU
SM
NO
Prt
Ct
LA
HA
cs
I
TE
ab
SN
CO
PC
Hh
' XAO
RESIN
(MG)
< 0.022
< 0.032
< 0.0100
< 0.65
< 0.012
< 0.017
< 0.026
< 0.02B
< 0.018
< 0.02a
< 0.029
< 0.0069
< 0.0100
< 0.0061
< 0.018
< 0.0059
< 0.0100
< 0.0028
< 0.0100
< 0.00<46
< 0.018
< 0.022
< o.oooo
< 0.0080
< 0.0056
< 0.25
0.0007
< 0.015 .
< 0.013
< 0.025
< 0 ,5a
< 0 . 036
< 0.013
< 0.00<45
COMPOSITE
SAMPLE
(MG)
< 0.0033
< 0,00"8
< 0.0015
< 0.097
< .0.0019
< 0.0025
< 0.003B
< 0.0042
< 0.0026
0.013
< 0 . 0 0 U a
< 0.0010
< 0.0015
< 0.0009
< 0.0027
< 0.0009
< 0.0015
< o.oooa
< 0.0015
< 0.0007
< 0.0027
< 0.0033
< 0.0006
0.0017
< 0.0009
< 0.037
0.0007
< 0.0023
< O.U019
< O.OOiS
< 0.18
< 0.080
< 0.0019
< 0.0007
TOTAL
SASS
(MGJ
< 0.026
< 0.037
< 0.011
< 0.75
< 0.0 la.
< 0.019
< 0.029
< 0.032
< 0.020
0.013
< 0.034
< 0.0079
< 0.012
< 0.0071
< 0,021
< 0.0068
< 0.012
< 0.0032
< 0.012
< 0.0053
< 0.021 .
< 0.026
< 0.00a5
0.0017 TO 0.00*0
< 0.0066
< 0.29
o.ooiy
< 0.01*
< 0.0)5
< 0.029
< 0.72
< 0.12
< 0.015
< 0.0052
EMISSION
FOUND
(HG/DSCK)
< 0.0008
< 0.0011
< 0.0003
< 0.023
< O.OOOa
< 0.0006
< 0.0009
< 0.0010
< 0.0006
o.oooa
< 0.0010
< 0.0002
< O.OOOi
< 0.0002
< 0.0006
< 0.0002
< o.oooa
< 0.0001
< o.oooa
< 0.0002
< 0.0006
< 0.0008
< 0.0001
0.0001 TO 0.0002
< 0.0002
< 0.0087
< 0.0001
< 0.0005
< 0 . 0 0 0 a
< 0.0009
< 0.022
< 0.0035
< o.oooa
< 0.0002
-------
TABLE F-4 .(continued)
.EMEN1
kU
HO
N8
IK
Y
SK
R6
BR
SE
AS
GE
UA
ZN
CU
NI
CO
FE
MM
CH
V
Tl
CA
K
CL
S
P
SI
At
MG
NA
^
ti
6E
LI
XAD
RESIN
(HG)
< 0.017
< 0.048
< 0.0019
0.0036
< 0.0030
< 0.078
< 0.0090
< 0.098
< 0,0088
< 0.015
< 0.0062
< 0.0025
< 0,91
< 2.7
< 5.3
< 0.019
< 2.1
< 0.22
< 0.66
0.012
< 0.086
< 8.0
< 3.2
< 40,
< 68.
< 3.0
< 12.
< 1.3
< 1.3
< 5.1
< 0.11
.0.032
< 0.0040
< 0.013
COMPOSITE
SAMPLE
CMC).
< 0.0025
0.019
0.0003
< 0.0011
< 0.0004
0.015
< 0.0028
< 0.024
< 0.0013
< 0.016
< 0.0009
< 0.0004
< 0.91
< 0.13
< 0.80
< 0.013
< 2.7
0,088
< 1.1
< 0.0049
< 0.013
< 1.9
< 1 .0
2.8
< 11.
< 0. 15
< 2.5
< 0.32
< 0,62 •
< 3.7
< 0.017
< 0,0065
< 0.0001
< 0.0013
10TAL
SASS
(WG)
« 0,019
0,019 TO 0.048
0.0003 TO 0.0019
0.0036
< 0.0030
0.015 TO 0.078
< 0.012
< 0.12
< 0.0100
< 0.031
< 0,0071
< 0,0029
< 1.8
< 2.8
< 6.1
< 0.031
< 4.8
0.088 TO 0.22
< 1.8
0.012
< 0.099
< 9.9
< 4.2
2.6 T(j «iO.
< 79.
< 3.1
<. 14.
< 1 ,t>
< 1.9
< a.e
< 0.13
0.032
EMISSION
FOUND
(MG/OSCM)
0.015
0.
0.
0.
0.
0.
< 0.0006
0006 TO 0
0001 TO 0
0.0001
< 0.0001
0005 TO 0
< 0.0004
< 0.0037
< 0.0003
< 0.0009
< 0.0002
< 0.0001
< 0.055
< 0.086
< 0.19
< 0.0009
< 0.15
0027 TO 0
< 0.054
0.0004
< 0.0030
< 0.30
< 0.13
086 TO 1
< 2.4
< 0 , 094
< 0.44
< 0.049
< 0.058
< 0.27
< 0.0040
0.0010
< 0.0001
< 0.0004
.0015
.0001
.0024
.0067
.2
-------
TABLE F-5. SSMS ANALYTICAL DATA: SITE 104
.EMENl
U
TH
BI
Pfa
TL
AU
IR
US
Rt
rt
K.f
LU
YB
IM
EH
MO
ov
re
GO
EU
SM
NO
HR
tt
LA
oA
CS
I
_It
StJ
SN
CO
fO
KM
FILTER
CATCH
(MG)
< 0.0001
< 0.0001
< 0.0002
< 0.0071
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.000
< 0.000
< 0.000
< 0.000
< 0.000
< 0.0001
< 0.0001
< 0,0001
< 0.0001
< o.oooi
< 0.0001
< 0,0001
< 0.0001
< 0.0001
< 0.0041
< 0.0001
< 0.0001
< 0.0001
< 0.0002
< O.OOU9
< 0,0009
< 0.0001
< O.U001
XAD
RESIN
(MG)
< O.OM
< 0.020
< 0.0061
< 0.27
< 0.0076
< 0.0100
< 0.016
< 0.017
< 0.011
< 0.014
< 0.018
< 0.0002
< O.OOol
< 0.0038
< 0.011
< 0.0036
< 0.0063
< 0.0017
< 0.0063
< 0.0028
< 0.011
< 0.014
< 0.0024
< 0.0033
< 0. 01)35
< 0.34
< 0.0007
< 0 . 00 9 4
< 0.0079
< 0.052
< 0 . « 5
< 0.031
< 0.0079
< 0.0027
COMPOSITE
SAMPLE
(MG)
< 0.0073
< 0.011
< 0.0033
0.075
< O.OOU1
< 0.0055
< o.oo64
< 0.0091
< 0.0058
< 0.36
< 0.0097
< 0.0023
< 0.0033
< 0.0020
< 0.0059
< 0.0019
< 0,0034
< 0.0009
< 0.0034
< 0.0015
< 0.0059
< 0.0073
< 0.0013
0.0061
< 0.0019
0.080
0.0015
< 0 . 0 U 5 0
< 0.0042
< 0.021
< 12.
< 0.17
< 0.0042
< 0.0015
TOTAL
SASS
(HG)
* 0.021
< 0.030
< 0.0096
0.075 TO 0.27
< 0.012
< 0.016
< 0.024
< 0.026
< 0.017
< 0.37
< 0.028
< 0.0065
< 0.0094
< 0.0058
< 0.017
< 0.0056
< 0.0097
< 0.0026
< 0.0096
< 0.0044
< 0.017
< 0.021
< 0.0037
0.0061
< 0.0055 •
0.080 TL1 0.34
.0.0015
< 0.015
< 0,01?
< • 0.053
< 12.
< 0.21
< O.C12
< 0.0042
EMISSION
FOUND
(MG/DSCM)
< 0.0006
< 0.0009
< 0.0003
0.0023 TO 0.0065
< 0.0004
< 0.0005
< 0.0007
« 0.0008
< 0.0005
< 0.012
< 0.0009
< 0.0002
< 0.0003
< 0.0002
< 0.0005
< 0.0002
< -0.0003
< 0.0001
< 0.0003
< 0.0001
< 0.0005
< 0.0006
< 0.0001
0.0002
< 0.0002
0.0025 TO 0.011
< 0.0001
< 0.0004
< 0.0004
< 0.00 lb
< 0.38
< 0.0064
< 0.0004
< 0.0001
-------
TABLE F-5 (continued).
EMEN1
KU
MO
N8
ZR
r
SH
RB
BR
SE
AS
GE
GA
ZN
cu
M
CU
FE
MN
CR
V
TI
LA
K
CL
S
p
SI
AL
MG
NA
f-
6
8E
LI
r FILTER
CATCH
(MG)
< 0.0001
< 0.0001
< 0.0001
0.0008
< 0.0001
< 0.0006
< 0.0001
< 0.0008
< 0.0009
< 0.0002
< 0.0001
< 0.0001
0.80
< 0.19
< 0.41
< 0.0001
< 0.022
< 0.0012
O.OOOfl
< 0.0001
< 0.0042
0.037
< C.078
< 0.029
< 0.53
< 0.0099
< 1.3
MC
< 0.066
MC
< 0.0038
0.047
< 0.0001
< 0.0001
XAO
RESIN
(MG)
< 0.0100
< 0.020
< 0.0012
< 0.0044
< 0.00.18
« O.OT1.
< 0.0024
< 0.02:9
< 0.0054
< 0.015
< 0.0038
< 0.0015
< 0.39
< 1.6
< 0.49
< 0.0053
< 1.3
< 0.064
< 0.14
< 0.0065
< 0.016
< 3.4
< 1.4
< 7.3
< 14.
< 1.8
< 3.3
< 1.3
< 0.74
< S. 1
< 0.100
0.013
< 0.0002
< o.on
COMPOSITE
SAMPLE
(MG)
< 0.005«
0.036
< 0.0006
< 0.0024
< 0.0097
0.0051
< 0.0021
< 0.016
< 0.0029
< 0.017
< 0.0020
< 0.0008
< 0.63
< 0.30
< 0.82
< 0.013
< 6.V
< 0.72
< (J.65
< 0.0073
< 0.0004
< 1.3
< 1.1
5.1
< 37.
< ".7
< /.6
< 0.31
< 0.40
< K.2
0.015
< 0.006f.
< 0.0003
< 0.0014
0
0
0
0
0
TOTAL
SASS
(MG)
.< 0.016
0.036
< 0.0018
.0008 T(J 0.0068
.0001 TO 0.012
.0051 TO .0.072
< 0.0045
< 0.046
< 0.0092
< 0.03?
< 0.0056
< 0.0024
0.80
< 2.1
< 1.7
< 0.018
< 8.2
< 0.79
.0008 TO 0.79
< 0.014
< 0.020
.037 TO 4.7
< 2.5
5.1
< 51.
< 12.
< 12.
EMISSION
FOUND
(MG/OSCM)
< 0.0005
0.0011
< 0.0001
< 0.0001 TO 0
< 0.0001 TO 0
0.0002 TO 0
< 0.0001
< 0.0014
< 0.0003
< 0.0010
< 0.0002
< 0.0001
0.025
< 0.066
< 0.053
< 0.0006
< 0.26
< 0.024
< 0.0001 TO 0
< 0.0004
< 0.0006
0.001? TO 0
< 0.078
0.16
< 1.0
< 0.36
< 0.38
.0002
.0004
.0022
.024
.15
< 1.2
0.015 T(l 0.11
0.060
< 0.0005
< 0.0001 TO 0.013
< 0.037
0.0005- TU 0.0033
0.0019
< 0.0001
< 0.0001 TO 0.0004
INDICATES A
CO"PUME:-:T
- INDICATES TnAT TH£ TQT4L At.O EMISSION VALUES rtfcPt r.'-'T
CALCULATED OrtlNG TO Tnfc PKEStNCE Of- 4*. MC CONiCt •'« T rr A T I Of. .
-------
TABLE F-6. SSMS ANALYTICAL DATA: SITE 300
00
.EMENl
U
Td
HI
PB .
TL
AU
IK
OS
RE
w
MF
LU
re
TM
EH
HO
UY
TB
GO
tu
SM
>^D
PH
C£
LA
BA
IS
I
TE
Sb
SN
CD
HO
Hh
r FILTER
CATCH
(MG)
< 0.0001
< 0.0002
0.0004
1.5
< o.ooei
< 0.0001
< o.oeoi
< 0.0002
< 0.0001
0.0002
< 0.0002
< 0.0001
< 0.0001
< o.ooei
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0 ,0001
< 0.0001
< o.oOOl
< 0.0005
< e.o«e
< 0.0001
< 0 . 0 0 0 1
< 0.0001
< 0.0015
< 0.022
u.067
< 0.0001
< O.oOOl
XAO
HESIN
(MG)
< 0.0036
< 0,0052
< o.on
< 0.003U
< 0.0020
< 0.0027
< 0.0042
< 0.00«5
< 0.0029
< O.OOJ8
< O.C008
< 0.0011
< 0.0016
< o.ooto
< 0.0029
< 0.0010
< 0.0017
< 0.0004
< 0.0017
< 0.0007
< 0.0029
< 0.0036
< 0 . 0 0 0 h
< O.OObl
< 0.0020
0.078
0.016
< O.'.i(i?5
< 0 . a 0 2 1
0.0070
o.um
< 0.0026
< O.U021
< 0.0007
COMPOSITE
SAMPLE
(MG)
0.015
< 0.0062
< 0.0019
0.077
< 0.0024
< 0.0032
< 0»0050
< 0.0054
< 0.0034
0.023
< 0.0057
< 0.0013
< 0.0019
< 0.0012
< 0.0035
< 0.0011
< 0.002U
< 0.0005
< 0.0020
< 0.0009
< 0.0 ') 35
< O.OOi-3
< 0.0008
0.0016
0.0039
0.76
< 0.0013
< 0.0030
< O.O.V5
0 . '.' 0 / 1
< 0.070
0.012
< 0.0025
< U . 0 0 o 9
TOTAL
SASS
(MG)
0.015
< 0.012
0.0004 TO 0.013
1.6
< 0.0045
< 0,0060
< 0.0093
< 0.0100
< 0.0060
0.023
< 0.011
« 0.0025
< 0.0036
< 0.0022
< 0.0065
< 0.0021
< 0.0037
< 0.0010
< 0.0037
< 0.00-17
< 0.0065
< o.ooei
< 0. 00 1 u
0.0016 ID O.OObl
0.0039
0.83
0.016
< 0.0056
< 0.0047
0.01 u
0.011 ro c.092
0.079
< 0.0047
< 0 . 0 0 ! o
EMISSION
FOUND
0.0002
< 0.0001
0.0001 TO 0.0001
0.017
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
0.0002
< 0.0001
< 0.0001
< 0.0001
< 0*0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001 TO
0.0001
< 0.0001
0.0091
0.0002
< 0.0001
< 0.0001
0 . 0 0 0 1
0,0004 TU 0.0010
0.0009
< o.oooi
< o.oooi
FUEL
(PPM)
0.062
0.090
0.059
2.0
0,035
0,046
0.072
0.078
O.OU9
0.066
0,082
0.019
0.028
0.017
0.050
0,017
0.029
0.0078
0.029
0.013
0.050
0.11
0.024
0.23
0.17
1.1
0.0026
0,043
0.036
0.036
0.24
0.22
0.036
0.013
EMISSION
CALCULATED
(MG/DSCM)
0.0010
0.0015
0.0010
0.033
0.0006
0.0008
0.0012
0.0013
0.0008
0.0011
0.0013
0.0003
0.0005
0.0003
0.0008
0.0003
0.0005
0.0001
0.0005
0.0002
0.0008
o.ooie
0.0004
0.0038
0.0028
0.018
0.0001
0.0007
0.0006
0.0006
0.0039
0.0037
0.0006
0.0002
-------
TABLE F-6 (continued).
VD
.fcMENl
RO
MO
NB
/R
Y
SR
Rtt
8R
SE
A3
(it
tiA
ZN
CU
NI
CO
FE
MN
CH
V
II
CA
K
CL
S
P
SI
AL
MG
NA
F
a
bt
LI
r FILTER
CATCH
(MG)
< 0.0001
0.0037
< 0.0001
0.036
0.0001
< 0.0047
0.0006
< 0.0029
< 0.0072
< 0.0044
< 0.0001
< e.oooi
< 0.55
0.17
0.66
< 0.0052
< 0.53
< 0.044
< 0.020
< 0.0013
0.041
< 0.75
< 0.60
< 0.15
12.
< 0.085
< 7.2
< 0.36
< 0.23
< 0.50
< 0.0032
< 0.026
< 0.0001
< 0.0004
XAD
RESIN
0.0027
0.0098
0.0609
0.0062
0.0004
e.017
0.023
0.026
0.014
0.018
0.0010
0.0020
0;31
0.31
0.60
0.0029
0.40
0.100
0.068
0.0089
0.025
1.4
0.22
0.096
2.6
0.16
2.0
0.<43
1 .0
1.3
0.14
0.064
0.0.006
0.0030
COMPOSITE
SAMPLE
< 0.0032
0.0100
0.0039
0.026
< 0.0006
Oj0059
0.0018
0.030
< 0.023
0.0046
< 0.0812
< 0.021
1 .0
< 0.51
< 1 .0
0 . 006h
2.3
0.064
< 0.057
0.0031
0.12
2.3
1.3
U.31
1.5
0.16
9.5
- 2.0
1 .?
1 .7
1 .1
1 .4
0.0086
0.029
0
0
0
0
TOTAL
SASS
(MG)
< 0.0060
0.024
0.0048
o.oba
0.0005
0.083
0.025
0.030
.01« TO 0.030
0.022
< 0.0022
.0020 TO 0.021
1.0
.17 TO 0.81
.66 TO 1.6
0.00
-------
TABLE F-7. SSMS ANALYTICAL DATA: SITE 301
.EMEN1
U
TH
BI
PB
TL
AU
IR
OS
HE
tut
HF
LU
YB
TM
ER
HO
OY
TB
GO
EU
SM
ND
PH
CE
LA
8A
cs
1
TE
SB
SN
CO
PD
KH
r FILTER
CATCH
(M6)
< 0.8017
< 6.0025
< 0.0008
< 1.1
< 0.0010
< 0.0013
< 0.0020
< 0.0022
< 0.0014
< 0.0018
< 0.0023
< 0.0005
< 0.0008
< 0.0005
< 0.0014
« 0.0005
< 0.0008
« 0.0002
< 0.0006
< 0.0004
< 0,0014
< 0.0017
* 0.0003
0.0034
< 0.0009
< 0.43
< 0.0003
< 0.0012
< 0.0010
< 0.0049
< 0.0010
0.0096
< 0.0010
< 0.0003
XAD
RESIN
(MG)
< 0.0046
< 0.0067
< 0.0021
< 0.30
< 0.0026
< 0.0034
< 0.0053
< u.0058
< 0.0030
< 0.0049
< 0,0061
< 0,0014
< 0,0021
< 0.0013
< 0.0037
< 0.0012
< 0.0021
< 0.9006
< 0.002.1
< 0.0010
< 0.0037
0.0078
0.0029
< 0.023
< 0.0051
< 0.11
< 0.0002
< 0.0032
< 0.0027
< 0.0056
0.020
0/028
< 0.0027
< 0.0009
COMPOSITE
SAMPLE
(MG)
< 0,021
< 0.030
< 0.0093
C.41
0.012
0.016
0.024
0.026
< O.Olo
< 0 ,2
-------
TABLE F-7 (continued).
.EMCN1
RU
MO
NB
2R
r
SR
RB
BR
SE
AS
G£
GA
ZN
CU
HI
CO
Ft
MN
CH
V
TI
CA
K
CL
S
P
Si
AL
Mt
IVA
r FILTER
CATCH
(MS)
< 0.0013
< 0.0051
< 0.0001
« 0.012
< 0.0002
0.0025
0.0011
< 0,0076
0.0071
0.0075
< 0.0005
< 0.0021
1.3
0.38
0.91
< 0.-0020
< 2.3
< 0.057
< 0.026
< 0.0026
0.06U
< 1.4
< 1.1
0.31
20.
< 0.23
< 6.5
0.62
< 0.62
MC
XAO
RESIN
<*G)
< 0;803«
Oi«095
0;<614
< 0.922
< o,es»6
0.015
0 i 0 9 1 <
< 0.046
< 9| 9818
< 0 i 0 1 1
< 0.0013
< 0.0005
0.94
< 0.89
< 1.1
0.0051
< 1.8
< 0.068
< 0.15
< 0.033
< 0.038
1.4
< 1.4
0.41
54.
< 0.61
< 3.4
< 2.0
< 1.8. '
<
3
0
83
4
11
0.82
0.32
O.I
.020
.32
.0052
10 0.034
.0036
.13
TO 0.59
.074
TO 0.019
,03b
.0075
.0050
.2
10 3.4
TO 3.6
.031
,8
*19
.0
TO
.19
.8
.1
.72
.3
•
TO
TU
I u
0.035
2.0
2.4
0. 1«
EMISSION
FOUND
(MC/OSCM)
< 0.0002
0.0036
< 0.0001
o.oooi TO e.oooa
< 0.0001
0,0015
0.0091 TO 0.0067
< 0.0008
0.0001 TO 0.0002
0.0004
< 0.0001
< 0.0001
0.025
0.0043 TO 0.039
0.0160 TO 0.041
0.0004
0.032
0.0022
0.01 1
0.0001 TO 0.0004
0.0021
0.077
0.035
0.0062
0.94
0.049
0.13
0.0093 TO 0.022
0.0036 TO 0.028
-
0.0001 TO 0.0016
FUEL
(PPM)
< 0.01J
0.082
0.0100
0.826
0.0033
0.13
0.9100
0.05S
< 0.0062
0.012
< 0.0044
< 0.0020
2.2
5.7
13.
0.19
3.8
0.29
0.51
< 0.852
0.19
< 8.9
4.1
< 2.8
300.
< 1.0
< 12.
5.1
1.5
< 18.
< 0.027
EMISSION
CALCULATED
(MC/OSCM)
< 0.0002
0.0013
0.0002
0.0004
< 0.0001
0.0020
0.0002
o.oooe
< 0.0001
0.0002
< 0.0001
< 0.0001
0.035
0,088
0.20
0.0029
0.058
0.0045
0.0079
< 0.0008
0.0029
< 0.14
0.064
< 0,044
4.6
< 0.016
< 0.18
0.079
0.023
< 0.27
< 0.0004
C INDICATES A MAJUR COWPUNE.'j! OF T fit- 'SAw^cfc .
INFLATES T^AI THE TOTAL «'»0 £WISSI'.',. VALUES AtHfc •'•0!
CALCULATED U«IMG TO Tuf PRESENCE OF AN MC CU\C t'
-------
TABLE F-8. SSMS ANALYTICAL DATA: SITE 302
.EMCN1
U
TM
BI
PB
TL
AU
IB
OS
RE
w
HF
LU
YB
TM
ER
HO
DY
TB
GD
tu
SM
ND
PR
CE
LA
BA
CS
I
TE
SB
SN
CD
PO
RM
F FILTER
CATCH
(MS)
< 0.0009
< 0.0013
< 0.0004
< 0.36
< 0.0005
« 0.0007
< 0.0010
< 0.0011
« 0.0007
< 0.0009
< 0.0012
< 0.0003
< 0.0004
< 0.0002
< 0.0007
< 0.0002
< 0.0004
< 0.0001
< 0,0004
< 0.0002
< 0.0007
< 0.0009
< 0.0002
< 0.0021
< 0.0002
< 0.022
< 0.0001
< 0.0006
< 0.0005
< 0.0010
0.014
0.0059
< 0.0005
< 0.0002
XAD
RESIN
(MO
< 0,0095
< 0.014
< 0.0043
0.31
< 0.0053
< 0,0071
< 0.011
< 0.012
< 0.0075
< 0.0100
< 0.013
< 0.0029
< 0.0043
< 0.0026
< 0.0077
< 0.0025
< 0.0044
< 0.0012
< 0.0044
< 0.0020
< 0.0076
< 0,0095
< 0.0017
< 0.0023
< 0.0025
< 0.24
< 0.0005
< 0.0065
< 0.0055
0.0039
< 0.073
0.71
< 0.0055
< 0.0019
TOTAL
SASS
< 0.0100
< 0.015
< 0,0047
0.31
* 0,0058
< 0.0077
< 0.012
< 0,013
< 0.0082
< 0,011
< 0.014
< 0.0032
< 0.0047
< 0.0029
< 0,0084
< 0,0028
< 0.0048
< 0.0013
< 0.0048
< 0.0022
< 0.0083
< 0.0100
< O.OOlb
< 0,0044
< 0,0027
< 0.26
< C.0006
< O.U072
< 0.0060
0.0039
O.Ola TO 0,073
0.72
< 0.0060
< 0.0021
EMISSION
FOUND
(MG/DSCM)
< 0.0001
< 0.0001
< 0.0001
0.0030
< 0.0001
< 0.0001
< 0,0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0,0001
< 0,0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0025
< o.oooi
< 0.0001
< 0.0001
< 0.0001
0,0001 TO 0.0007
0.0069
< .0,0001
< 0.0001
FUEL
(PPM)
« 0.025
< 0.036
< 0.011
< 0.49
< 0.014
< 0.019
< 0.029
< 0.031
< 0.020
< 0.027
< 0.033
< 0.0078
< 0,011
< 0.0069
< 0.020
< 0,0067
< 0.012
< 0.0032
< 0.012
< 0,0052
< 0,020
< 0.025
< 0.0045
0.030
0.0100
0.67
< 0.0020
< 0.017
< 0.015
< 0.029
0. 15
0.36
< 0.014
< 0.0051
EMISSION
CALCULATED
(MG/D3CM)
0.0001
0.0002
0.0001
0.0029
0.0001
0.0001
0.0002
0.0002
0.0001
0.0002
0.0002
0.0001
0.0001
0.0001
0.0001
0.0001
0.0001
0.0001
0.0001
0.0001
0.0001
0.0001
0.0001
0.0002
0.0001
0.0040
0.0001
0.0001
O.U001
0.0002
0.00.09
0.0023
0.0001
0.0001
-------
TABLE F-8 (continued).
EMEN1
RU
MO
N6
ZR
Y
SR
RB
BR
3E
AS
GE
GA
ZN
CU
HI
CO
FE
MN
CR
V
Tl
CA
K
CL
S
P
SI
AL
MG
NA
F
a
BE
LI
r FILTER
CATCH
(«G)
< 0.0007
0.0035
< 0.0001
< 0.020
< 0.0001
0.0053
0.0013
< 0.0029
< 0.0011
< 0.0029
< 0.0002
< 0.0001
< 1.2
0.74
0.19
< 0.0035
0.98
0.024
0.059
< 0.0030
0.044
< 1.0
< 1.3
< 0.48
31.
< 0.24
3.7
< 0.85
< 0.52
MC
< 0.015
< 0.17
< 0,0001
< 0.0012
XAO
«fc.31N
(M&)
< 0.0070
0.023
< 0,0008
< 0.0031
< 0.0013
< 0,16
< 0.0081
< 0.098
< 0.0037
< 0.063
< 0.0026
< 0.0011
2.2
0.61
0.72
< 0.022
< 1.8
< 0,14
< 0.13
< 0.031
0.055
25.
< 2.9
1.2
28.
0.96
< 9,8
*C
1.4
< 11.
< 0.048
< 0.018
< 0.0005
< C.0056
TOTAL
SASS
(MG)
< 0
0
< 0
< 0
.0077
.026
.0009
.023
< «»0014
0
0
0
0
i
.0053
.0013
< 0
< 0
< 0
< 0
< 0
^
\
0
< 0
0
.024
.059
< 0
0
25
< 4
1
59
0
.7
1
< 0
< 0
< 0
< 0
TO 0,16
TO 0.0061
.100
.0048
.066
.0029
.0012
.2
.4
.91
.026
.98
TO 0.14
TO 0.13
.034
.099
.
.1
.2
•
.96
TO 9.8
-
.4
-
,063
.19
.0005
.0068
EMISSION
FOUND
FUEL
(PPM)
(MG/OSCM)
< 0
0
< 0
< 0
< 0
< 0,0001
< 0.0001
< 0
< 0
< 0
< 0
< 0
0
0
0
< 0
0
0.0002
0.0006
< 0
0
0
< 0
0
0
0
0.035
0
< 0
< 0
< 0
< 0
.0001
.0002
.0001
.0002
*0001
TO 0,0015
TO 0,0001
.0010
.0001
.0006
.0001
.0001
.022
.013
.0087
.0002
.0094
TO 0.0013
TO 0.0012
.0003
.0009
.24
.040
.012
.57
.0092
TO 0.094
-
.014
-
.0006
.0019
.0001
.0001
< b,
0,
< 0,
0,
< 0,
0.
0,
0,
< 0.
0,
< 0.
< 0,
1,
1,
15,
0.
< 10,
0,
< 1 ,
< 0,
0,
< 4,
< 3,
< 20,
29,
< 2,
< 5,
< 2,
1,
< 9,
< 0,
< 0,
< 0,
0,
,019
,018
,0022
,028
,0033
,040
,0036
,090
,021
,041
,0070
,0028
,3
,2
1
,049
I
,16
, 1
,083
,069
,4
,6
i
»
,4
,3
.4
.5
.3
.043
.023
.0004
.013
EMISSION
CALCULATED
(MG/OSCM)
0.0001
0.0001
0.0001
0.0002
0.0001
0.0003
0.0001
0.0005
0.0001
0.0002
0.0001
0.0001
0.0081
0.0071
0.088
0.0003
0.060
0.0010
0.0064
0.0005
0.0004
0.026
0.022
0.12
0.18
0.014
0.032
0.014
0.0091
0.056
0.0003
0.0001
0.0001
0.0001
MC INDICATES t- MAJOR
Of THfc SAMPLE.
• INDICATES THAT rut TUTAL A\C EMISSION VALUES .-.tst M.-T
CALCULATED CAlNU TO THE PftESE'-.-CE OF Ar. "-C CDNCE'vTRAI-K.'N.
-------
TABLE F-9. SSMS ANALYTICAL DATA: SITE 303
tLtMfcNT FILTER
C 4 1 C H
I-UEL
(PPM)
EMISSION
CALCULATED
(MG/DSCM)
u
Th
bl
Pb
TL
AU
IH
US
Ht
rt
Hh
LU
YB
TM
EH
HU
UY
IB
GD.
tu
SM
NO
HK
CE
LA
bA
CS
I
IE
SB
SM
10
PO
KH
< 0.0006
< 0.0009
0.0009
1.5
< O.OOOi
< O.OOOU
< 0.0007
< O.OOOH
< O.OOOb
< 0.0006
< O.OOOb
< 0.0002
< U.0004
< 0.0002
< o.ooos
< 0.0002
< U.OUOi
< 0.0001
< 0.000?
< 0.0001
< 0.0005
0.0007
0.0001
< 0.001S
< 0.0011
< 0.045
< 0.0002
< O.OOOM
< O.OOOi
< 0.0069
0.012
0.017
< O.OOOJ
< 0.0001
O.OS1
O.Olb
0. / 1
0.020
0.027
o.oui
0.044
0.026
O.OiB
0.047
0.011
0.016
0.009B
0.029
0.0094
O.Olb
0.0045
O.Olb
0.0074
0.029
0.12
0.0066
0.0.40
0.020
0.44
0.0020
0.024
0.021
0.020
0.020
0.089
0.021
(J.0072
0.0006
o.oooe
0.0002
0.011
0.0003
0.0004
0.0006
0.0007
0.0004
0.0006
0.0007
0.0002
0.0002
0.0002
0,0905
0.0001
0.0003
0.0001
0.0003
0.0001
0.0004
0.0019
0.0091
0.0005
0.0003
0.0070
0.0001
0.0004
0.0003
0.0003
0.0003
0.0014
0.0003
0.0001
-------
TABLE F-9 (continued)
Ui
Ul
,EM£M
HU
MO
1MB
^H
Y
SK
HB
bR
3E
AS
Gk
GA
iH
CU
M
CU
FE
MN
CM
V
II
CA
K
CL
S
P
SI
AL
MG
NA
F
H
bt
LI
1 f I L T F. «
CATCH
(M&)
< 0.0004
0.0033
< 0.0002
0.021.
0.0006
0.013
0.0018
< 0.013
< o.oioo
0.0025
< 0.0002
< 0.0001
5.4
1.7
0.95
< 0.00/1
< 2.4
0. 12
0.092
0.0020
0.13
< 1.5
< 1.3
2.2
31.
0.51
< 6.3
< 1.7
< 1.1
< 2.3
< 0.0100
< 0.39
< u.oooi
< 0.0036
FUEL
(PPM)
< 0.026
0.052
0.0032
O.OStt
< 0.0047
0.062
0.050
O.li
< 0.19
0.041
< 0.0098
< o.oo«o
1.3
2.1
8.1
< 0.062
< 10.
0.27
0.55
< 0.053
0.69
< 6.2
< 3.6
< 9.0
170.
< 2.2
< 57.
< 10.
i.l
13.
0.027
0.049
0.0043
0.014
EMISSION
CALCULATED
(MG/DSCM)
0.0004
0.0006
0.0001
0.0009
0.0001
0.0010
0.0006
0.0020
0.0030
0.0007
0.0002
0.0001
0.020
0.034
0.13
0.0010
0.16
0.0043
0.0067
0.0006
0.011
0.099
0.057
2^7
0.036
0.90
0.16
0.049
0.21
0.0004
0.0006
0.0001
0.0002
A MAJUK
ur
SAMPLE.
S TM4I |Ht fOFAL 4NU f-MISSlON VALUES KERt NOT
CALCULATtO UwINl. TO THE HHESEME HF AN MC CONCENTRATION.
-------
TABLE F-10. SSMS ANALYTICAL DATA: SITE 304
Ul
.EMEf
U
TH
Bl
P6
TL
AU
IR
OS
RE
w
HF
LU
YB
TM
EK
HO
OY
T8
GO
EU
SW
NO
PR
CE
LA
6A
CS
I
Tfc
SB
SM
CD
PO
KM
«T FILTER
CATCH
(MO
< 0.0010
< 0.0014
< 0.0004
< 0.26
< 0.0005
< 0.0007
< 0.0011
< 0.0012
< 0.0008
< 0.0010
< 0.0013
< 0.0003
< 0.0004
< 0.0003
< 0.0008
< 0.0002
< 0.0004
< 0.0001
< 0.0004
< 0.0002
< 0.0006
< 0.0010
< 0.0002
< 0.0016
< 0.0005
< 0.051
0.0012
< 0.0007
< 0.0006
< 0.0011
< 0.032
< 0.0007
< 0.0005
< 0.0002
XAD
RESIN
(MG)
< 0.0019
< 0.0028
< 0.0009
< 0.038
« 0,0011
< 0.0015
< 0.0022
< 0.0020
< 0.0015
< 0.0021
< 0.0026
< 0,0006
< 0.0009
< 0.0005
< 0.0016
< 0.0005
< 0.0009
< 0.0003
< 0.0009
< O.OOOu
< 0.0016
< 0.0020
< 0.0011
< 0,0098
0.0018
< 0.069
< 0.0012
< 0.0013
< 0.0011
< 0.011
< 0.032
< 0.029
< 0.0011
< O.OOOU
TOTAL
SASS
(MG)
0.0029
0.0042
0.0013
0.32
0,0016
0.0022
0.0033
0.0036
0.0023
0.0031
0.0038
0.0009
0,0013
0.0008
0,0023
0,0008
0,0013
0.0004
0.0013
0. OOOb
0.0023
0.0029
0,0013
0.011
0.0018
0. 12
0.0012
0.0020
0.0017
0.012
u . 0 6 3
0.030
0.0017
0,0006
EMISSION
FOUND
(MG/D3C*)
o.ooei
o.oooi
o.oooi
0.0059
0,0001
0.0001
0.0001
0.0001
0.0001
0.0001
0.0001
0.0091
0.0001
0.0001
0.0001
0.0001
0.0001
0.0001
0.0001
0.0001
0.0001
0.0001
0.0001
0.0002
0.0001
0.0022
0.0001
0.0001
0.0001
0.0002
0.0012
0.0005
0.0001
0.0001
FUEL
CPPM)
0.025
0.036
0.023
1.7
0.014
0.019
0.031
0.020
0.026
0.033
0.0077
0.011
0.0069
0.020
0.0066
0.012
0.0031
0.011
0.0052
0.020
0.042
0.015
0.21
0.068
1.4
0.0020
0.017
0.01«
O.b8
0.014
0.89
0.014
0.0050
EMISSION
CALCULATED
(MG/OSCM)
0.0004
0.0005
0.0003
0.026
0.0002
0.0003
0.0004
0.0005
0.0003
0.0004
0,0005
0.0001
0.0002
0.0001
0.0003
0.0001
0.0002
0.0001
0.0002
0.0001
0.0003
0.0006
0.0002
0.0032
0.0010
0.021
0.0001
0.0003
0.0002
0.0100
0.0002
0.013
0.0002
0.0001
-------
TABLE F-10 (continued).
EMENI
RU
MO
NB
2«
Y
SR
MB
BR
SE
AS
GE
GA
I*
CU
NI
CO
ft.
MN
c«
V
CA
H,
CL
S
P
SI
AL
HU
NA
a
Bt .
Ll
r FILTER
CATCH
(MG)
< 0.0007 <
0.0072 <
< 0.0001 <
0.014
< o.ooei <
0.0033 <
0.0019 <
< 0.015 <
< 0.0004 <
0.0031 <
< 0.0003 " <
< 0.0001 <
0.93 -<
0.79 <
0.64 <
0.0077 <
6.2 <
0,049 <
0.022 <
0.0038 <
< 0.72 <
< 0.9b <
0.21 <
11. <
< 0 . 1 9 <
22. <
0.74 <
< ,37 <
< 1 .1 <
0.51 <
< O.OU02 <
< 0.0038 <
XAO
RE'SIN
(MG)
0.0014
0.026
0,0035
0.014
: 0.0003
0.032
0.0025
0.042
0.0034
0.0003
0.0005
0.0069
0.37
0.49
0.70
0.0075
0.37
0.029
0.058
0.0019
16.
: 0.42
1.5
25.
•0.53
6.7
0.55
2.5
2.2
u , ••' 1 3
0.0001
0.0026
TOTAL
SASS
(MG)
< 0.0022
0.0072 TO 0,828
< 0.0036
0.028
< 0.0004
0.0033 TO 0.032
0,0019
< 0.057
< 0.0037
0.0031
< 0.0008
< 0.0070
0.91
0. 79
0.64
0.0077
8.2
0.049
TO 0.058
0.0038
17.
1.4
TU
10
< 0.72
22.
0.74
< 2.9
< J.2
O.bl
0.0001 TU (
< 0.0064
0.022
0.21
11.
1.5
25.
0
0
0
0
0
0
EMISSION
FOUND
(MG/DSCM)
< 0.0001
.0001 TO 0*0005
< 0.0001
0.0005
< 0.0001
.0001 TU 0.6006
< 0.0001
< 0.0011
< 0.0001
< 0,0001
< 0.0001
< 0.0001
0.017 .
0.01S
0.012
0.0001
0.15
0.0009
.0004 TO 0.0011
< 0.0001
< -0.32
< 0.026
.0039 TO 0.029.
.20 TO 0.47
< 0.013
0.41
0.014
< 0.054
< 0. ObO
0. 0094
.0001 TU 0.0001
< 0.0001
FUEL
(PPM)
< 0.018
0.27
0.023
0.028
< 0.0033
0.14
0.016
0.055
0.066
0.13
< 0.0069
< 0.0026
4.0
9.7
8.7
0.22
71.
1.1
0.95
0.16
< 19.
< 1 1 .
< 4.4
290.
< 2.4
< 26.
33.
2.1
< 28.
< 0.072
0.0039
0.013
EMISSION
CALCULATED
(MG/DSCM)
<
<
<
<
<
<
<
<
<
<
<
<
0.0003
0.0000
0.0003
0.0004
0.0001
0.0022
0.0002
0.0006
0.0010
0.0020
0.0001
0.0001
0.061
0.15
0.13
0.0033
1.1
0.017
0.014
0.0023
0.28
0.17
0.067
4.3
0.036
0.39
0.50
0.032
0.42
0.0011
0.0001
0.0002
-------
TABLE F-ll. SSMS ANALYTICAL DATA: SITES 326 AND 327
Silt 526-1: SITE 32o-2:
SITE 327-1:
ECEMEMT
FUEL
(PPM)
EMISSION
CALCULATED
(MG/DSCM)
FUEL
(PPM)
EMISSION
CALCULATED
(MG/DSCM)
FUEL
(PPM)
EMISSION
CALCULATED
(MG/DSCM)
00
u
TH
ai
PB
TL
AU
IR
OS
RE
M
NF
LU
YB
TM
ER
HO
OY
TB
GD
Eu
SM
NO
PR
CE
LA
8A
CS
I
IE
Sb
SN
CO
PO
HM
« 0.14
< 0.21
< 0.064
4.4
< 0.080
< 0.11
< 0.16
< 0.18
< 0.11
< 0.43
< 0.19
< 0.044
< 0.065
< 0.039
< 0.12
< 0.036
< 0.066
< 0.016
< 0.066
< 0.030
< 0.12
< 0.14
< 0.025
< 0.066
< 0.037
0. 14
< 0.0060
< 0.099
< 0.083
< 0.058
< 0.16
< 0.100
< 0.083
< 0.029
0.0054
0.0076
0.0024
0.17
0.0030
0.0040
0.0062
0.0067
0,0043
0.016
0.0071
0.0017
0.0024
0.0015
0.0044
0.0014
0.0025
0.0007
0.0025
0.0011
0.0043
0.005«
0.0009
0.0025
0.0014
0.0054
0.0002
0.0037
0.0031
0.0022
0.0058
0.0036
0.0031
0.0011
0.14
0.20
0.063
2.2
0.079
0.11
0.16
0.18
0.11
1.2
0.19
0.044
0.063
0.039
0.11
0.037
0.065
0.018
0.065
0.029
0.11
0.14
0.025
0,034
0.037
0.29
0.0059
0,097
0.062
0.057
0.0«0
0.100
0.081
0.02ft
0.0043
0.0062
0.0019
0.067
0.0024
0.0032
0.0049
0.0054
0.0034
0.036
0.0057
0.0013
0.0019
0.0012
0.0035
0.0011
0.0020
0.0005
0.0020
0.0009
0.0035
0.0043
0.0006
0.0010
0.0011
0.0089
0.0002
0.0030
0.0025
0.0017
0.0025
0.0031
0.0025
0.0009
0.38
0.55
1.4
0.99
0.21
0.26
0.44
0.46
0.30
1.7
0.50
0,12
0.17
0.11
0.31
0.100
0.18
0.048
0.16
0.080
0.31
0.38
0.068
0.091
0.099
17.
O.Olh
0.
-------
SITE 326-lt
TABLE F-ll (continued).
SITE
SJ1E I27-U
Ul
LtMEN
KU
MO
N8
2ft
y
SH
KB
BR
SE
AS
&t
UA
2N
CU
NI
CO
f-E
MN
CH
V
II
CA
n
S
P
SI
AL
i* i,
NA
tf
bt
LI
T FUEL
(PPM)
« 0.11
< 0.26
« 0.035
< 0.046
< 0.019
0.096
0.012
< 0.21
< 0.47
< 0.069
< 0.040
< 0.016
6.4
19.
2.6
0.052
< 25.
< 0.86
0.84
< 0.23
< 0.53
< 14.
6.8
220.
4.0
< 10.
7.5
0.83
51.
< 0.11
0.072
0.040
EMISSION
CALCULATED
(MG/DSCM)
0.0040
0.0100
0.0013
0.0017
0.0007
0.0037
0.0004
0.0077
0.018
0.0033
0.0015
0.0006
0.24
0.73
0.100
0,0020
0.95
0.032
0.031
0.0085
0.020
0.54
0.25
ft.4
0.15
0.36
0.28
0.031
1.9
0.0041
0.0027
0.0015
FUEL
(PPM)
0.100
1.1
0.034
0.046
0.019
0.065
0.0080
0.39
0.055
0.043
0.039
0.016
4.3
4.4
3.1
0.091
18,
0.84
1.6
0.44
0.36
5.1
5.b
270.
1.5
< 5.0
2.8
< 2.4
< 0. IS
0.071
< 0.025
EMISSION
CALCULATED
(MG/DSCM)
0.0032
0.014
0.0010
0.0014
0.0006
0.0020
0.0002
0.012
0.0017
0.0013
0.0012
0.0005
0.13
0.14
0.096
0.0026
0.54
0.026
0.050
0.014
0.011
0.16
0.17
8.1
0.047
0.15
0.066
0.075
0.43
0.0046
0.0022
0.0008
FUEL
(PPM)
«
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
0
0
0
0
0
0
0
0
0
0
0
0
1
14
6
0
0
0
1
0
19
19
560
2
37
6
10
0
0
0
.28
.52
.033
.12
.051
.26
.043
.55
.30
.12
.11
.24
.4
,5
.1
.17
!96
.85
.7
.35
*
•
•
.3
*
.5
•
!o9V
.018
.052
EMISSION
CALCULATED
(MG/DSCM)
0.012
0.022
0.0014
0.0053
0.0022
0.011
0.0018
0.023
0.013
0.0050
0.0045
0.0100
0.059
0. 19
0.26
0.0074
1.0
0.042
0.036
0.072
0.015
0.82
0.82
24.
0.096
1.6
0.28
0.45
5.9
0.0008
0.0022
-------
TABLE F-12. DISTRIBUTION OF VOLATILE (C8-C12) AND
NONVOLATILE (> C16) ORGANICS IN SASS
TRAIN SAMPLES: GAS-FIRED SOURCES
Sample type
Solvent probe
rinse (PR-0)
Solvent XAD-2
module rinse
(MR-0)
XAD-2 resin -
solvent extract
(XR-SE)
Site
100
101
102
103
104
100
101
102
103
104
100
101
102
103
104
Volatile
(mg/m3)
< 0.01
< 0.01
-
0.36
< 0.01
< 0.01
< 0.01
< 0.01
< 0.01
2.11
< 0.01
1.09
5.01
Nonvolatile
(mg/m3)
0
0.14
. 0.031
0.03
0
0.18
0.81
0.47
0.27
0.86
0.22
0.29
0.28
0.18
0.06
160
-------
TABLE F-13. DISTRIBUTION OF VOLATILE (C8-C12) AND NONVOLATILE (> C16)
ORGANICS BY GAS-FIRED SITE
Site
100
101
102
103
104
Sample
type
PR-0
MR-0
XR-SE
PR-0
MR-0
XR-SE
PR-0
MR-0
XR-SE
PR-0
MR-0
XR-SE
PR-0
MR-0
XR-SE
Volatile
(mg/m3)
< 0.01
< 0.01
2.11
< 0.01
< 0.01
< 0.01
_
-
-
0.36
< 0.01
1.09
< 0.01
< 0.01
5.01
Nonvolatile
(mg/m3)
0.0
0.18
0.22
0.14
0.81
0.24
0.03
0.47
0.28
0.03
0.27
0.18
0.0
0.86
0.06
Total organics
(mg/m3)
< 0.01
0.18
2.33
0.14
0.81
0.24
> 0.03
> 0.47
> 0.28
0.39
0.27
1.27
< 0.01
0.86
5.07
161
-------
TABLE F-14. DISTRIBUTION OF VOLATILE (C7-Ci6) AND NONVOLATILE (> C16)
ORGANICS IN SASS TRAIN SAMPLES: OIL-FIRED SOURCES*
Sample type
Solvent probe
rinse (PR-0)
Solvent XAD-2
module rinse
(MR-0)
i
Condensate -
solvent extract
(CD-LE)
XAD-2 resin -
solvent extract
(XR-SE)
Site
300
301
302
303
304
300
301
302
303
304
300
301
302
303
304
300
301
302
303
304
Volatile
Total yg mg/m3
< 31+
< 217
< 140
< 116
< 285
< 31
644 0.0073
< 140
< 116
< 285
< 300
< 34
1,880 0.021
1,840 0.026
< 341
15,500 0.17
15,400 0.17
56,600 0.54
42,900 0.53
24,600 0.32
Nonvolatile
Total yg mg/m3
< 1,000
< 5,000
< 13,000
< 6,000
114,000 1.49
< 1,000
20,200 0.23
< 13,000
< 6,000
14,900 0.19
15,600 0.17
87,800 1.00
< 2,000
< 6,000
92,700 1.21
43,000 0.47
52,600 0.60
30,200 0.29
96,600 1.20
237,000 3.09
Volume
of air
sampled
(m5)
91.9
88.1
104
80.6
76.7
91.9
88.1
104
80.6
76.7
91.9
88.1
104
80.6
76.7
91.9
88.1
104
80.6
76.7
t
All values have been blank-corrected.
All numbers reported as "less than" (<) are sample values found to be
less than the blank value; number reported is value of blank.
162
-------
TABLE F-15. DISTRIBUTION OF VOLATILE (C7-C16) AND NONVOLATILE (> C16)
ORGANICS BY OIL-FIRED SITE*
Volume
. of air Sample
sampled type
300 91.1 PR-0
MR-0
CD-LE
XR-SE
XRB-SE
Totals
301 88.. 1 PR-0
MR-0
CD-LE
XR-SE
XRB-SE
Totals
302 104 PR-0
MR-0
CD-LE
XR-SE
XRB-SE
Totals
303 80.6 PR-0
MR-0
CD-LE
XR-SE
XRB-SE
Totals
304 76.7 PR-0
MR-0
CD-LE
XR-SE
XRB-SE
Totals
Volatile
Total yg
<
<
15
(3
15
<
<
15
(1
16
<
<
1
56
(1
58
<
<
1
42
(9
44
<
<
<
24
(6
24
31+
31
300
,500
,370)
,500
217
644
34
,400
,670)
,044
140
140
,880
,600
,670)
,480
116
116
,840
,900
,230)
,740
285
285
341
,600
,930)
,600
mg/m3
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
17
17
01
17
18
02
54
56
03
53
56
32
32
Nonvolatile
Total yg
< 1
< 1
15
43
(4
58
< 5
20
87
52
(22
160
< 13
< 13
< 2
30
(22
30
< 6
< 6
< 6
96
(6
96
114
14
92
237
(64
458
,000
,000
,600
,000
,000)
,600
,000
,200
,800
,600
,400)
,600
,000
,000
,000
,200
,400)
,200
,000
,000
,000
,600
,930)
,600
,000
,900
,700
,000
,000)
,600
mg/m3
0
0
0
0
0
0
1
0
0
1
1
1
0
1
3
5
.170
.469
.639
.229
.997
.597
.823
.290
.290
.20
.20
.49
.194
.21
.09
,984
Total organics
Total yg
15
58
74
20
87
68
176
1
86
88
1
139
141
114
14
92
261
483
,600
,500
,100
,844
,800
,000
,644
,880
,800
,680
,840
,500
,340
,000
,900
,700
,600
,200
mg/m3
0.17
0.64
0.81
0.24
1.0.0
0.77
2.01
0.02
0.83
0.85
0.03
1.73
1.76
1.49
0.19
1.21
3.41
6.30
t
All values have been blank-corrected.
All numbers reported as "less than" (<) are sample values found to be
less than the blank value; number reported is value of blank.
163
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/7-79-029b
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE Emissions Assessment of Conventional
Stationary Combustion Systems; Volume I. Gas- and
Oil-fired Residential Heating Sources
5. REPORT DATE
May 1979
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
N.F.Surprenant, R.R.Hall, K.T. McGregor,
A. S.. 'Werner (GCA/Technology Division)
and
8. PERFORMING ORGANIZATION REPORT NO.
GCA-TR-77-30-G(l)
9. PERFORMING ORGANIZATION NAME AND ADDRESS
RW,.Inc.
One Space Park
Redondo Beach, California 90278
10. PROGRAM ELEMENT NO.
EHE624A
11. CONTRACT/GRANT NO.
68=02=2197
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final; 9/76 - 3/79
14. SPONSORING AGENCY CODE
is. SUPPLEMENTARY NOTES ffiRL-RTP project officer is Ronald A. Venezia, MD-62, 919/541-
2547.
16. ABSTRACT The ygpo^ gi^gs results of an assessment of emissions from gas- and oil-
fired residential heating sources, through a critical examination of existing emis~
sions data, followed by a phased measurement program to fill gaps in the emissions
Initially, five gas-fired and five oil-fired residential sources were tested,
Mass emission rates of criteria pollutants, trace elements, and organics (including
•were determined. Subsequent test program evaluation led to a decision to con-
duct additional tests at one gas-fired and two oil-fired sites, to determine the effect
of the burner on/off cycle on emissions. Particulate, SO4, SO2, and SOS emission
data were also obtained at the oil-fired sites. Assessment results indicate that resi-
smfcial sources are of potential significance based on multiple source severity fac-
tors calculated for an array of homes burning gas or oil. Pollutants for which mul=
pie source severity factors exceed 0.05 (the level which may be potentially signifi-
cant) are: NOx from gas-fired sources, and SOS, NOx, and Ni from oil-fired sour-
as. Measured criteria pollutant emission factors were generally comparable to EPA
emission factors (in AP-42), except for total HC emissions from oil-fired sources
which were 3 times greater. However, POM compounds known to be carcinogenic
were not found above the detection limit of 0. 3 micrograms/cubic meter.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Pollution
Assessments
Emission
Combustion
Natural Gas
Dust
Sulfur Oxides
Organic Compounds
Nitrogen Oxides
Hydrocarbons
Pollution Control
Stationary Sources
Residential Furnaces
Particulate
T3B
14B
13H,13A
21B
21D
TIG"
07B
07C
8. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
176
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
164
-------
------- |