Air Toxics Emission Inventory

              for the

       Southeast Chicago Area
          John Summerhays
           Harriet Croke
       Air & Radiation Branch
              Region V
U.S. Environmental Protection Agency
             July 1987

-------
                               Table of Contents
                                                              Page
Introduction                                   .                1
Methods Used in Inventorying Point Sources                     5
Area Source Inventorying Methods                               8
Mobile Source Inventorying Methods                            13
Wastewater Volatilization Inventorying Methods                17
Inventorying Methods for Hazardous Waste Treatment
 Storage and Disposal  Facilities                              18
Summary of Results                                            18
Discussion of Results                                          28
General Observations                                          36
Bibliography                                                  40
Appendix A Sample Questionnaire

-------
                                 List of Tables
Table 1.  Substances included in Inventory                             4
Table 2.  Particulate Matter Species Fraction                          7
Table 3.  Species Fractions Used in Area Source Inventory              10
Table 4.  Emission Factors Used in Area Source Inventory               11
Table 5.  Highway Vehicle Species Fractions                            16
Table 6.  Summary of Emissions" from All Source Types                   19
Table 7.  Area Source Emission Totals                                  20
Table 8.  Highway Vehicle Emissions Totals                             21
Table 9.  Uastewater Treatment Plant Emissions Estimates               22
Table 10. Emissions Estimates for Facilities Receiving
           Questionnaires                                              23
Table 11. Emissions from Coking Operations                             24
Table 12. Point Source Emission Estimates Based on Organic
           Species Fractions                                           25
Table 13. Point Source Emission Estimates Based on Particulate
           Matter Species Fraction                                     27
Table 14. Point Source Emission Estimates Sorted by SCC                29
Table 15. Organic Species Fraction-based Estimates for Facilities
           Receiving Questionnaires                                    33
Table 16. Particulate Matter Species Fraction-based Estimates for
           Facilities Receiving Questionnaire                          34
List of Figures
Figure 1. Southeast Chicago Source Area and Receptor Area
                                                                      Page

-------
                         Air Toxics  Emissions  Inventory
                         for the Southeast  Chicago Area


Increasing national  attention has focussed  on  the health  risks  from "toxic"
(non-criteria) air pollutants that arise in urban areas where a concentrated
level of industrial  activity coexists with  high population  density.  Within
Region V. perhaps the most serious combination of concentrated  industrial
activity with high population density is in Southeast Chicago.   In  particular.
the area is one of the foremast locations for  integrated  steel  production  and
chemical production of the coatings  used in Chicago's substantial manufacturing
economy.  This area also has onp of  the nation's five permitted PCB incinerators
and has a variety of other facilities for treating, storing and disposing  of
hazardous waste.   Therefore, Region V, with assistance  from the Illinois
Environmental Protection Agency (IEPA) and  the Indiana Department of Environ-
mental Management (IDEM), has compiled a comprehensive inventory of "air toxics"
emissions in the Southeast Chicago area.

The inventory described in this report represents the first component of a
three part project.  The next component of this project  is  a modeling analysis,
which will serve to estimate the exposure of Southeast Chicago  residents to the
emitted pollutants.  The third component will  then estimate the health risks
that would be expected from the estimated exposure.

In any inventory of this type, resource considerations require  a choice between
developing a screening inventory covering multiple source types using only
readily available information versus developing a more focussed inventory
investigating only a few source types or pollutants.  This  inventory may he
considered a screening inventory, intended to provide an overview of air
carcinogen emissions in the covered area.

This  inventory has been designed to be comprehensive in  several respects.
First, this study has attempted to include all source types that emit air
toxics.  Second, although the focus of this study is an  exposure in a moderate
sized area (approximately 65 square miles), a  much broader area was inventoried
to include all sources with potentially significant impacts in the selected
receptor area.  Third, this study inventoried  for a comprehensive  list of
potential carcinogens.  Specifically, the inventory included all potential
carcinogens for which a dose-response relationship has been estimated even
including some compounds with minimal evidence of carcinogenicity.   This
inventory also included three compounds which  are not suspected of being
carcinogens but have been measured at relatively high concentrations in the
area.  Unfortunately, apart from these three compounds,  the inclusion of
compounds of the basis of noncarcinogenic health effects was judged to be
beyond to scope of this study, due to the relative sparsity of data on dose-
response  relationship for such health effects, the uncertainties about presence
and magnitude of threshold concentrations, the uncertainties about the relative
importance of short term peak versus  long term average concentrations, and the
analytical difficulties of considering these various parameters.   Also, a fourth
substance, mercury, is also not a suspected carcinogen but was included because
a National Emission Standard for Hazardous Air Pollutants has  been promulgated
for  this  pollutant.

-------
With respect to source types,  this study included all  source types for which air
toxics emissions could be estimated.,  The inventory included point sources,
area sources, and mobile sources,, and further included volatilization from
wastewater treatment plants.,  Emissions estimates for  hazardous waste treatment,
storage and disposal facilities (TSDF's) have been difficult to derive but are
expected to become available in the next few months.  It must be noted that
there may be source categories which are not included  because data for estimating
the emissions are not available,,  On the other hand, it is a reasonable hope
that data are available for the most significant categories.,

With respect to spatial coverage, Figure 1 is a map showing both the target
"receptor area" for the exposure analysis and the broader source area included
in the inventory.  The target  area of the exposure analysis component of this
study is an area approximately 13 kilometers square (8 miles square) for a
total of 167 square kilometers (65 square miles).  The specific boundaries of
this receptor area are: north-87th; south-Sibley Blvd/Pulaski St,; west-Western
Ave,; and east-Indiana/Illinois border.  In order to include all significant
sources, a substantially larger source area was inventoried.  The source area
covers a 46 kilometer (about 29 miles) square area.  Since the -prevailing winds
in the area are from the southwest quadrant, the source area is skewed toward
the south and west of the receptor area.  The specific boundaries of the source
area are, in terms of UTM coordinates from 4584 to 4620 kilometers northing and
from 420 to 466 kilometers easting in zone 16,  This source area extends
approximately 30 kilometers south and west and 16 kilometers north and east
of the center of the receptor  area.  The inventory further includes a few
additional  point sources which are outside this source area but were judged to
be potentially significant sources.

With respect to pollutants covered, the inventory includes 47 compounds for
which quantitative estimates of carcinogenicity (unit  risk factors) have been
made.  The list of compounds is given in Table 1 and includes 22 nonhalogenated
organics compounds, 17 halogenated organic compounds (including 16 chlorinated
and one brominated organics),  and 8 inorganic species  (especially metals).  It
is important to note that the  carcinogenicity of some of these compounds is
quite speculative.  Nevertheless, these compounds were included so that any
error would be on the side of  being comprehensive.  As noted above, the compound
list also included four compounds which are not suspected of being carcinogenic,
but otherwise did not include  any compounds on the basis of noncarcinogenic
health effects.  It might also be noted that several of the compounds on the
inventory are relatively esoteric.  As will be discussed below, the inventory
for Southeast Chicago found emissions for 39 of the 51 compounds.

Some of the pollutants in this study warrant special discussion.  First, form-
aldehyde is not only emitted into the atmosphere but it is also photochemically
formed during atmospheric reactions of other organics.  Current evidence
indicates that substantially more formaldehyde results from photochemical
formation than from direct emissions.  Thus, the ultimate total exposure estimate
is likely to rely more heavily on monitoring data, with the inventorybased
exposure estimate serving only to help characterize the relative importance of
emissions versus photochemical formation.  Somewhat similar considerations
apply for a second pollutant,  carbon tetrachloride.  This compound has an
atmospheric half-life of many  years, and so exposure may be more a function of
historic emissions than of current emissions.  Thus, monitoring data may again

-------
                                             SountAST CHICAGO

                                      SOURCE AREA  &  RECEPTOR AREA
                                                                     ~i
 IWIIUS (|IO»I


HOIINCBNOOK |
       1     I1. •  '' -'T"
       :r.  ill/'  if

-------
Table 1.  Substances included in inventory
         Acrylamide
         Acrylonitrile
         Benzene
         Butadiene
         Coke Oven Emissions
         Diethanolamlne
         Oimethylni trosamine
         Dioctylphthalate
         Ethyl Acrylate
         Fthylene
         Fthylenp Oxide
         Formaldehyde
         Gasoline Vapors
         Isopropylidene Diphenol
         Mel amine
         Nitrobenzene
         Nitrosomorpholine
         Polycyclic Organic Matter
         Propylene Oxide
         Styrene
         Terephthalic Acid
CHLORJ_NAJEP_ YOG (17)

   Allyl Chloride
   Benzyl Chloride
   Carbon Tetrachloride
   Chloroform
   nioxin
   Fpichlorohydrin
   Fthylene Dibromide*
   Ethylene Dichloride
   Methyl Chloride
   Methylene Chloride
   Pentachlorophenol
   Perchloroethylene
   PCB's
   Propylene Dichloride
   Trichloroethylene
   Vinyl Chloride
   Vinylidene Chloride
INORGANIC. (8)

   Arsenic
   Asbestos
   Beryllium
   Cadmium
   Chromium
   Nickel
   Titanium Dioxide
   Radionuclides
                ( 4 )
   Acetone
   Mercury
   Toluene
   Xylene
*Although ethylene dibromide is not, strictly speaking, chlorinated, it is halogenated
   and is included among chlorinated compounds due to chemical similarity.

-------
may be a more reliable basis for assessing exposure.,   A third "pollutant" ,
identified here as polycyclic organic matter, is actually a  class of aromatic
compounds,.  Three of the lightest such compounds, namely napthalene (a two  ring
structure)„ anthracene (a three ring structure), and  phenanthrene (also a three
ring structure), are both among the most highly emitted and  among the least
toxic of this class of compounds.,  In order to focus  on the  more significantly
toxic compounds, this inventory includes not only an  estimate of total polycyclic
organic matter but also an estimate of "heavy" polycyclic organic matter that
excludes these three compounds,  A fourth pollutant,  chromium, also in a sense
represents a class of pollutants.,  In this case the different forms are diffe-
rent valence states, including- Cr*6 (such as found in chromic acid), Cr*3,  and
neutral (metallic) chromium.  It is known that Cr"^ is the most toxic form, but
it is usually not clear what mix of valence states is present in any set of
emissions.  Generally, this inventory conservatively  assumed that all  chromium
in the most toxic valence,

Methods used in Inventorying Point Sources

The first component of the inventory was for discrete industrial facilities
within (or in a few cases, just outside) the source area0  For 29 of the 88
facilities in the source area, questionnaires were sent to the facility by  the
appropriate State agency (Illinois EPA for Illinois sources, Indiana DEW for
Indiana sources).  The selection of facilities was intended  to include the
sources suspected of having the greatest air toxics impact on the receptor
area.  The first step in the selection procedure was  to develop rankings of
criteria pollutant impacts on the receptor area.  Two rankings were developed,
including one for VOC and the other for total suspended particulate matter
(TSP), based on indices defined as the respective pollutant  emissions divided
by distance from source to the center of the receptor area.   For the top 25
sources in each ranking, the second step of the selection procedure was a
subjective screening of sources for the probability of emitting compounds  of
interest.  This second step of the procedure effectively took into account  the
fact that VOC and TSP emissions totals are not necessarily indicative of air
toxics emissions.  This process led to the selection  of 13 facilities.  In
addition, all other chemical manufacturing facilities were included, representing
an additional 16 facilities (above and beyond 2 chemical manufacturing facilities
selected on the basis of the ranking procedure),

A copy of a sample questionnaire is included as Appendix A,   This questionnaire
asked the companies to make their own estimates of emissions of each of the
substances on the inventory list, and asked for a few other  pieces of infor-
mation needed for modeling.  Comments regarding the challenges confronted
during the process of companies answering and U,S, EPA processing these ques-
tionnaires are given in a concluding section of this  report.  It may be noted
here that substantial followup was often necessary to obtain responses, and in
many cases the companies asked U,S,EPA to supply appropriate species fractions
or even to estimate emissions (e,g, from storage tank data).  Also, in U,S,EPA's
review of questionnaire responsesB companies were asked in a few cases to clarify
or to confirm their responses.

For the other 59 facilities in the source area, a method labeled the "species
fraction method" was used.  The first step of this method is to obtain emission
estimates for total volatile organic compounds and total suspended particulate

-------
matter for each operation of each  facility.   The second step of the method is
to obtain species fractions representing the fractions of the total VOC or the
TSP that are emitted as the various individual  species.  For example,  in this
study, the fugitive emissions from purging operations at refineries were esti-
mated to be 2.4% benzene.  These fractions are then multiplied times the VOC
or TSP emissions estimate to obtain estimates of emissions of individual
species.

The two main sources of species fraction information were the Volatile Organic
Compound (VOC) Species Data Manual  and the Receptor Model Source Composition
Library.  The index indicating-which species profiles in the VOC Species Data
Manual (i.e., which sets of species f-ractions) should he applied to which SCC's
was developed by the GCA Corporation for another project (the National Acid
Precipitation Assessment Program)  being undertaken by U.S. EPA.  Both  the
profiles and the index are shown in reference P. 2.  The authors developed the
index indicating which sets of metal species fractions from the Receptor Model
Source Composition Library should  be applied to which SCCs.  This index is shown
as Table 2.  For this study, estimates of total VOC and TSP emissions  were
taken from the National Emissions  Data System (NEDS), which provided emissions
data and identified the Standard Classification Code (SCC) for each operation
at each facility.

One subtle point concerning the use of species fractions with VOC totals con-
cerns compounds that are negligibly photochemically reactive, such as  methane
and methylene chloride.  In some cases, the VOC emissions may be derived in a
way that excludes these compounds.   To the extent this is true, it would be
desirable to apply species-fractions derived on the same basis, i.e.,  species
emissions as fractions of just the reactive VOC.  Unfortunately, it is not
clear what VOC emission estimates  exclude the negligibly reactive species.
Therefore, all species fractions were derived and applied on a fraction-of-total
VOC basis.  In any case, the negligibly reactive portion of VOC tends  to be
small, so that any error here is likely also to he small.

The estimation of emissions by the species fraction method was performed for
all facilities, including facilities that were sent questionnaires.  This
provided the opportunity for quality assurance test of comparing the results
of these two methods.

Limited additional information was obtained from a variety of other sources.
Coke oven emissions were based on  TSP emissions estimates provided as  part of
Indiana and Illinois TSP SIP submittals.  Based on information in reference
P.6, emissions of the pollutant identified as "Coke oven emissions" (including
organics soluble in benzene) were estimated as 1.1 times the TSP emissions for
relevant emission points (coke oven charging and leak emissions).  Estimates of
benzene emissions from coke by-product recovery plants were obtained directly
from reference P.7, (except for Acme, for which an Illinois EPA estimate was
used).  Toluene and xylene emissions estimates were derived from benzene emis-
sions estimates based on information in reference P.7 suggesting that  the coke
oven gases are 60-85% benzene, 6-17% toluene, and 1-7% xylene.

-------
                                       7

Species fraction information for formaldehyde from utility fuel combustion and
from industrial coal combustion was derived from the formaldehyde "locator
document" (Locating and Estimating Air Emissions from Sources of Formaldehyde,
refo P»4e)o  This document provided emissions factors in units of grams per
Joule of heat Input, which was multiplied times the heat content of typical
locally used fuels and compared against the standard VOC emission factor to
derive the formaldehyde species fraction.,

Area Source Inventorying Methods

As with point sources, the most, commonly used method for developing area
source emissions estimates was the species fraction method.  The categories
inventoried by this method, the species fractions, and other relevant
information are shown in Table 3.  In general, the first piece of necessary
information was a county-by-county estimate of emissions from each category
being assessed.,  This information was generally derived by the State from the
1982Qhj_cago ozone State Implement at ion_£lan_l5IP)°  The SIP included VOC
                    based for £Aflmle_jan_ADJ3£flj^             factors
      ne sales or numbprTP tJJBeSL-jaA-^8JbQJlOJpi£JJ&£& TSP pmi SSJJXO— £a£Jbttr»  The
SIP provided county VPOC emissions estimates in kilograms per day  for  1980  and
1987»  Illinois EPA staff converted these estimates to annual emissions  (tons/
year) and interpolated to 1984 emissions for Illinois counties, and  the  authors
performed similar calculations for Lake County, Indiana,,

The second piece of information was then a set of species fra££ions, which
could be multiplied times the county VOC or TSP emi^s 1 ons j£fl_obtflj n  county  totaj
  ecies ejjfTjppSgg^ — mr^nrrs point a thiTd^ step of spatial disaggregati on was
performed „  ThTs step was necessary to support the modeling  analysis of
population exposure, since the exposure in the receptor area to emissions  is  a
function of the spatial distribution of emissions,,  For_ejch source  category, a
          parameter (e.g. _populajJLop ) j-***— ^Lgg£ed whTcTT could be  assumegrto
                                      For each source category, then,  each
2 k i 1 ometer_ar^squa^re was assigned a fraction of the coafrjacrgmijgions  equal
to the grid square "^Traction of the total surrogate parameter  in "tKeTcounty .
For example, if a grid had 1% of Cook County's arterial traffic, the  grid  would
be assigned 1% of Cook County's gas marketing emissions.   Note  that although
most of the surrogate data were 1985 projections made in  1976,  the use of  these
data for performing spatial allocations should not introduce  any significant
errors.

A second method used for inventorying areas sources was an  emission factor
approach.,  The catefories inventoried by this method, the emissions factors,
and other relevant information are shown in Table 4.  This  method was used in
different forms for different categories.  For heating, the available emission
factors for formaldehyde and polycyclic organic matter (POM)  for various types
of fuel use were multiplied times county level estimates  of the quantities of
the respective fuels used.  These emissions .»ort» t.hpn
to estimate emissions in each of the above grids.  For  per  capita  emissions,
(e.g. the emissions fromj/arious consumer products), emissions  in  each  grid
square were estimated T)y"multi plying the per capita emissions  factor times  the
number of people in the grid square.

-------
                  Table 2.  Participate Matter Species Fractions
                           (Taken from Reference  P. 3.  All  data  in percent)
                                      .Sgec_1es_ Fraction*
Source Category        Arsenjic Cadmiuni Chromium Mercury Nickel
Coal combustion
Oil combustion
Steel making:
Sinter plants
.058%   .006%    .054%
.015    .001
 :ay iron foundries


Asphalt roofing

Glass manufacturing

Gypsum manufacturing

Lime and cement
 manufacturing****


Refinery heater
 (gas-fired)

Refinery cat. cracker

Municipal incinerator
                       .012    .006
                       .023    .003

                       .050**  .050**
.020
                               .010

                               .115
                 .089
.550**  .550**  3.000
Open hearth.
basic oxygen furnaces  .050**  .050**   .20
                                         550*
                                  .040%
                                 1.622
                                                  Applicable
                                                     SCCs
                 .036
 .003

5.330
                     1-01-002-01,-02.-03.-23
                     1-02-002-01,-02,-04

                     1-01-004-01,
                     1-02-004-01,-04
                     1-02-005-01,-04
                     3-06-001-03***
                                              3-03-008-01. -02. -08, -09
                                                   -11, -12, -13, -14. -21
                                                       -22, -24, -25, -99
.20
.038
.550**
.218
.550**
.011
3-03-009-04, -10, -11, -12, -13
-14, -15, -31, -32, -33, -99
3-04-007-01, -05, -06. -07
-15, -99
.067 3-04-003-01. -20. -31, -40
-50, -51
3-05-001-01, -02, -03, -04, -05
.004 3-05-015-01. -02, -06, -08. -10
.550** 3-05-015-01, -02, -03, -04
.00012 3-05-006-06
-3-05-016-01. -02, -04, -07, -08
-09, -10. -14, -15, -99
.550**


.43

.014
                                              3-06-001-04


                                              3-06-002-01

                                              5-01-001-02
   * Although reference P.3. shows data for beryllium, beryllium does not show up for any  source
      types in Southeast Chicago.
  ** Where this table (and reference P.3) shows species fraction of .550% and .050%, the original
      study reported values of "trace" (0.1 to 1.0%) and "below detection limits" (<0.1%).
  *** SCC 3-06-001-03 is for oil-fired process heaters at refineries.
  f*** These species fractions are not taken from reference P.3 but are derived from the
       "locator documents" references P.4.f and P.4.g.

-------
                                       9
                                                                               ss*
For dejacgajsing^ county level  emissions estimates were derived from a national
study that estimated tofeal  usage of each of the compounds, dlvidejl-tha-±atal
h  the total number o_f__employees in the Standard Industrial Classification
         levant to the respective compound, and multiplied £h&_ce&»J-feaat
         factor tiJH&s^.^^o niimhQje_JoJE_J&jmaljoj/ees in tJbj&sg-jcJLASLSifi cat ions employed
         mtlgs^Jjg^^            r.hicagcLAce^  County emissions were then
distributed according to total  manufacturing employment, since spatial distri-
bution data were not avaiable for the specific industrial categories.  Similarly
for dry cleaning, national  emissions data were distributed to county level
estimate using employment data for relevant SICs, which were in turn distributed
to grids in the study area using commercial employment data-  For comfort
cooling towers, the draft background, information document for a possible NESHAP
(reference A.13) provided per capita emission factors reflecting the per capita
number of buildings in six building size ranges and emissions factors for each
size range adjusted to reflect the probability of having a comfort cooling
tower and the probability that chromium is used as a corrosion retardanto
These emission factors were provided on a state-by-state basis to reflect the
cooling load in each state,,  Reference A013 provided upper and lower bound esti-
mates reflecting disparate emissions testing results,.  This study used the more
conservative,, upper bound value; the lower bound value is about 28 times lower.
The ratio of Cook County population versus "other" (nonmanufacturing, non-
coiranerical) employment (3.4 people/"other" employee) was used to adjust the per
capita emission factor to a per "other" employee emission factor, which was
used with gridded "other" employment data to estimate grid-by-grid emissions0

An important subcategory within heating is residential wood combustion.  This
category includes woodburning in wood stoves, wood furnaces, fireplace inserts,
and fireplaces.  The principal  data sources for estimating wood usage were a
contractor study for the Department of Energy and the Department of Energy's
Residential Energy Consumption Survey (RECS) data (References A,7 and A,8),
The contractor study estimated state by state wood usage based on th* heating
degree days for the state, the probable proportions of households using wood as
a primary or secondary heat source, and various surveys  indicating the relationship
between these factors and wood usage for each kind of wood user.  This survey
indicated a 1981 statewide Illinois usage of 1830,000 dry tons of wood per year
(about 2,150,000 actual tons/year or 1,510.000 cords/year).  This translates to
about 0,43 dry tons of wood (about 0,50 actual tons or n.35 cords) per year
per Illinois household.  However, the Chicago area is likely to have a lower
percentage of wood burners and a lower rate of wood consumption that the
estimated Illinois average.  Therefore, RECS data on the wood usage in central
cities in the Worth Central region of the United States  versus the regional
average wood usage were used to adjust the Illinois usage estimates.  These
data indicated that in central  cities in the North Central U.S., 0,8 million
cords were used per year, which divided by approximately 5.8 million households
represents 0.14 cords/household.  By comparison, in the  full North Central
region, 10,9 million cords divided by 22075 million households represents 0,48
cords per household.  (For reference, these data suggest that the central city
usage of 0.14 cords/year represents 16% of the households burning 0.9 cords/year,
and the regional usage of 0.48 cords represents 21% of the households burning
2,3 cords/year.)  Thus, the urban usage may be estimated at 0,14/0,48 or 29% of
the regional usage.  On this basis, Chicago area wood usage was estimated to
be 29% of 0.43 dry tons per year, equalling 0.12 dry tons (OdO cords or 0.14
actual tons or 288 pounds) per ypar per household.

-------
            Table 3. Species Fractions used in Area Source Inventory
                     (except as noted, units are % of VOC emissions)
Category

Gasoline
Marketing
P^TJ.utarvts

Benzene
Toluene
Xylene
Ship and        Benzene
Barge Transfer  Toluene
                Xylene
Fractions

  .63%
  .64%
  .18%

  .63%
  .64
  .18%
Spatial dbn
  Parameter

Arterial VMT
                                  Distributed
                                   within port
                                   areas
Reference

   A. 3
                       A.3
Architectural
 Surface
 Coating
Benzene               .18%
Methyl Chloride       .27%
Methylene Chloride   3.37%
Toluene              7.60%
Xylene               1.36%
              Dwelling units
                       A.3
Heating
 (oil-fired)
Chromium
Nickel
 .047%*
 5.36%*
                                                      **
                       P.4
  Heating fractions are fractions of total  suspended particulate emissions.
** Spatial distribution parameter for industrial  distillate oil use was
      manufacturing employment.  Parameter for all commercial/institutional
      and residential  fuel  oil  usage was population.   Emissions for indus-
      trial  residual oil  and coal usage and for all utility fuel usage were
      considered in the point source inventory.

-------
            Table 4.  Emissions  Factors  used  in  Area  Source  Inventory
Source Category
    Pollutant
Heating
   Resid. oil:  Comm/Inst*  Formaldehyde
   Dist. oil:  Industrial
              Comm/Inst
              Residential
   Gas:  Industrial
        Comm/Inst
        Residential
    Emission  Factor
                  .069 ng/J (10.8 mg/gal)
                  .10  ng/J (14.5 mg/gal)
Spatial dbn.   Ref.
   Wood stoves, Fireplaces
   D1st. Oil  (all  users)
   Gas (all users)
       POM
  .038 ng/J  (   41  g/mmcf)
  .095 ng/J  (  103  g/mmcf)
  .43  ng/J  (  464  g/mmcf)

     .078 g/kg wood***
 8.7 pg/J (1.4 mg/gal)***
11.2 pg/J (12  g/mmcf)***
Per Capita Emissions

   Aerosol Cans
   Paint Stripping
   Misc. Products
   Chlorinated
      Drinking Water

Miscellaneous
Methylene Chloride
Methylene Chloride
   Formaldehyde
   Chloroform
      .50 #/capita-yr
      .59
      .048   "
      .029   "
Population
 Mfg. emp.
Population
Population
 mfg. emp.
Population
Population

dwell, units
 population
 population
 population
Hospital  sterilizing    Ethylene oxide
Chrome platers
Chromium
  2.8#/year/bed (> 200 beds)
   .2#/year/bed (< 200 beds)

 82.54/year per known facility
                                                                         ***
                                                                         ***
Comfort cooling towers  Chromium
Degreasing
Dry cleaning
Methylene Chloride
Perchloroethylene
Trichloroethylene

Perchloroethylene
                                                    *****
                                                    *****
                              mfg. emp.
                             comm. emp,
                                           P.4e
P.4
A.10
A.10
P.4e
P.4c
                  2.5#/year/1000 "other"    "other" emp.
                              employees****
             Author's
              survey

               A.14

               A.13
               A.10
               A.11
               A.12

               A.11
*Some of the abbreviations used on this page:  Comm/Inst - commercial  and institutional
    Resid. - Residual:  Dist. - distillate:   ng - nanograms (10~* grams):
    POM - polycyclic organic matter;  mfg.  - manufacturing:  emp. - employment.

**Excludes naphthalene, anthracene, and phenanthrene (see text).  Emission
    factor for woodstoves and fireplaces is  60% of total POM, for distillate oil
    is 90% of total POM. and for gas is 100% of total  POM.

***Distributed at actual location.

****"0ther" employees are nonmanufacturing,  nonretail  employees.

*****Degreasing and dry cleaning emission estimates are derived from national
     emissions totals disaggregate to county totals based on employment within
     specified Standard Industrial Classification (SIC) Codes.  Codes used for
     degreasing were 25, 34, 36, 37, and 39, and for dry cleaning were 7216, 7217
     and 7218.

-------
                                      12
                                                                               ,«»
The emission factor is derived from data given in a draft POM "locator document",
entitled "Locating and Estimating Air Emissions From Sources of Polycyclic
Organic Matter (POM)"0  The emission factor of 78.0 mg POM per kg wood reflects
first an exclusion of three lighter-weight POM (napthalene, anthracene, and
phenanthrene) and then a numerical averaging over all uncontrolled wood stoves
and fireplaces (not including three fireplace measurements which did not reliably
measure gaseous emissions,,)  Note that this emission factor is 60% of the
emission factor that would have been obtained had the three lighter weight
compounds not been excluded*

A third method used might be called the special survey method.  The results of
these surveys are also shown on Table 4.  Line oFinie" special surveys was conducted
for chroma plating,, which emits chromium.,  The chrome plating survey was based
on the "yellow pages" of the telephone directory,,  This survey indicated 17
facilities within the Southeast Chicago source area,,  Information was then
obtained from U.S. EPA's Office of Air Quality Planning and Standards (OAQPS),
indicating that based on the data obtained in development of a potential NESHAP
for chrome plating, one could assume a national average plant emissions of 33
pounds chromium per year.  Unfortunately, the listing of platers in the yellow
pages presumably does not include "captive platers" which perform chrome plating ,_
as one step in manufacturing their own product (e0g«, chrome plating as part of
an auto assembly plant): this listing presumably only includes "job shops" who
specialized in plating work for other customers,,  Data from OAQPS suggest that
"job shops" represent about 40% of total plating, so the emissions estimate
here was adjusted accordingly.  Note that the result should be quite conservative,
both because in reality "job shops" tend to be disproportionately high emitters
and because the survey assumed that all platers performed chrome plating,

A second special survey was for hospital use of ethylpne=o=x=ide for purposes of
,sterilizing.  The first step of this survey was tlTTontact TRe State Department
of Health, which provided an inventory of hospitals in t^e State,,  This inventory
showed four hospitals in this study's "receptor area" and a total of 24 hospitals
in the overall source area.  The second step was to contact the purchasing
departments of the four "receptor area" hospitals to request information on
annual ethylene oxide usage.  The third step was to use this information to
assess an emission factor.  The emission factor was developed on a per bed
basis, since the number of beds is an easily obtained piece" of information
which is considered indicative of the patient load and thus the amount of
sterilization taking place at a hospital.  (The number of beds does not include
"long term care" beds such as found at nursing homes, since such beds are
expected not be associated with significant ethylene oxide use.) The fourth
step was to multiply the per bed emission factor times the number of beds in
each of the hospitals in the source area.

The results of the four hospital survey were as follows: one hospital (427
beds) used 2430 pounds per year: a second hospital (418 beds) used steam, not
ethylene oxide, for sterilization: a third hospital (176 beds) also used steam,
not ethylene oxide: and a fourth hospital (168 beds) used 80 pounds per year
(only a small part of the hospital's sterilization used ethylene oxide).  Four
hospitals is of course a small sample, and so the emissions factors derived are

-------
                                      13
                                                                              x<=*
very approximate,,   Nevetheless,  the survey suggested that small  hospitals are
less likely to use ethyfene oxide sterilizers then large hospitals, so separate
emission factors were derived-   (The outpoint between "large" and "small" was
arbitrarily set at 200 beds.)   Also, obviously not all  hospitals use ethylene
oxide for sterilizing.  Thus,  a large hospital emission factor was estimated
by averaging the per bed usage for hospital  1 (2.8 pounds/bed) and hospital  2
(0), and a separate, small  hospital emission factor reflected an average for
hospital 3 (0) and hospital 4  (.24 pounds/bed).,

The data used in this survey reflected only Illinois hospitals.   Rather than
obtain an inventory of hospitals in Indiana, a per capita approach was used to
estimate hospital  usage of  ethylene oxide in Northwest  Indiana.   The per capita
emissions factor used was 0.010 pounds per year, derived from the Illinois
results indicating 22,172 pounds emissions by hospitals serving 2,160,180 people.

Mobile Source Inventorying  Methods

As with most of the area source categories, highway vehicle emissions were
inventoried using the species  fraction method.  For both Illinois and Indiana.
the analysis differentiated between emission on freeways and arterial emissions
(i«6o, emissions on arterial roadways and local streets).  In Illinois,
total county level VOC emissions were first derived, adjusted, and dvvidjgd
into freeway and arterial emissions.  A variety of species fractions were
deTfve^rancr usea fo~ol5tain  courrcy ^level emissions of individual  compounds.
Finally, the spatial distribution of these emissions data was estimated accord-
ing to qridded IfaTa on freeway Ind arterial =vejilclg_jT'-i Lg£==&ag^alg4=XVMT)°  I"
Indiana, gridded emissions  data were already available, from the 1982=ozone
§XP, so it was only necessary  to make adjustments and apply species'TracTTons
to obtain the desired inventory.  The following discussion documents these
calculations and the basis  of  the underlying data in more detail.

The Illinois county level VOC  emissions data were derived by the Illinois EPA
based on data in the 1982 ozone SIP.  These data were logarithmically inter-
polated between 1980 and 1987  to a 1984 base year and converted by Illinois EPA
from units of kilograms per day to units of tons per year.  The next step of
the analysis used mobile source emissions models to make a .set of adjustments
from conditions inherent in the SIP to conditions appropriate here and
simultaneously to disaggregate the arterial and freeway exhaust emissions and
the evaporative emissions.   The first run, designed to correspond to the SIP
emissions estimate, used MOBILE 2, used a summer temperature of 75°F, the
Federal Test Procedure (FTP) average speed of 19.6 miles per hour  (mph), vehicle
mileage and registration data  given in the SIP, and areawide averages for VMT
mix and hot start/cold start percentages.  Then, two runs of MOBILES were made
to estimate updated, more annual average emissions factors for the Southeast
Chicago area.  Both of these runs used an annual average temperature of 50°F,
and the same mileage and registration data as in the SIP.  The run for arterials
used a speed of 19.6 mph and hot start/cold start percentages given  in the SIP
for the southern Chicago area  arterials, and the run for freeways used a speed
of 45 mph and zero percent  hot starts and cold starts.  The  ratio of these
MOBILES results to the MOBILE2 results was multiplied times  the SIP-derived
county emissions totals to derive adjusted county emissions  totals for arterial
exhaust emissions and freeway  exhaust emissions.

-------
                                      14
                                                                              x«*
A similar procedure was used to estimate adjusted  county  total  evaporative
emissions.  However, one  additional  step in  this procedure was  to assign  all
evaporative emissions to  arterials,  i.e.,,  to reassign  emissions to arterials
that otherwise would have been  assigned  to freeways„   Since most of these
emissions may be assumed  to occur at trip ends, this  reassignment was  intended
to provide a more reliable assignment of these emissions  to where they actually
occur,,  The results of multiplying the MOBILES evaporative emissions estimates
times the respective traffic volumes indicates that  freeway evaporative emissions
in the area are 43% of arterial evaporative  emissions.,  Therefore, the reassign-
ment of evaporative emissions was accomplished by  zeroing out the freeway
evaporative emissions and increasing the arterial  evaporative emissions by  43%,

Two further adjustments were made that are not reflected  in MOBILES-  The first
adjustment only affects evaporative emissions. Based  on  a paper prepared by
UoSo EPA Office of Mobile Source staff,  it appears that whereas MOBILES uses  a
single miles per day figure and a single trip per  day  figure for all vehicle
ages in converting from emissions per day to emissions per mi1e9 the use  of
model year by model year  conversion factors  yields a  higher gram per mile
emission factor.  Based on information derived from the Office of Mobile  Source
paper, all evaporative emissions estimates were increased by 28%,  The second
adjustment only affects exhaust emissions.  Limited work  by John Sigsby et al,
at U,S, EPA's Mobile Source Characterization Branch found that exhaust emission
measurements made with the usual, unheated sampling train measured only 77% as
much VOC as a parallel heated sampling train.  Therefore, all exhaust  emissions
estimates here were divided by  0,77%, i.e.,  increased  by  30%,  It must be noted
that these adjustments do not reflect U,S. EPA policy,  and especially  the
exhaust emissions adjustments is based on a  sparse and possibly unrepresentative
data base.  Nevertheless, given the uncertainties  that exist throughout this
inventory, these adjustments were considered appropriate  for use in this  study.

The results of the above  adjustments may most easily be described in terms of
emissions factors.  The unadjusted, MOBILE? based  emission estimate translates
to ,833 tons per year per 1000  daily vehicle miles traveled (tpy/1000  miles),
(This is approximately 2,35 g/mile).  The various  conversions using MOBILES and
adjustments yielded an arterial exhaust  emission  factor of 1,122 tpy/1000 miles,
an arterial evaporative emission factor  of ,878 tpy/1000  miles, and a  freeway
(exhaust only) emission factor  of ,432 tpy/1000 miles.

The second area of emphasis in  the mobile source  inventory is species  fractions.
The species fractions derived in this study are summarized in Table 4, The
source of the greatest number of species fractions were taken from a paper by
Roy Zweidinger, et al, of U.S.  EPA's Mobile Source Characterization Branch
reporting concentration measurements of  numerous  compounds and of total VOC
near a limited access roadway near Raleigh,  North  Carolina (reference  M,l)0
This paper provided the data for the species fractions for formaldehyde,  ethylene,
toluene, xylene and acetone.  Since the  gas chromatographic method could  not
distinguish butadiene from butane, a butadiene species fraction was also  derived
from this paper using the butane data in conjunction with an assumption (suggested
by John Sigsby, reference M.6)  that 10%  of the material reported to be butane is
in fact butadiene.  Species fractions for benzene  for exhaust emissions and for
evaporative emissions were based on empirical formulae derived by Office  of
Mobile Source staff from  existing testing data.  The input to the calculation

-------
                                      15
                                                                              x-s*
of fractions used here include national  average gasoline benzene content  (1.34%)
and aromatic content (3&5/0  and an approximate North  Central  annual  average
reid vapor pressure (12,6).

Ethylene dlbromide fractions  were calculated from results reported by John
Sigsby et al. in a separate  paper focussing on ethylene  dihromide.  For exhaust
emissions, the average ratio  of EDB emissions to total exhaust hydrocarbon (HC)
emissions is 117,4 ug EDC/g  HC,  However,  because gasoline EDB content is
proportional to lead content  and lead content has declined, this figure must be
adjustedo  The lead content  of Sigsby's  fuel was 1,98  g/gallon.  The actual
1984 lead content data are derived from  a  Motor Vehicle  Manufacturers Association
survey, indicating that the  average lead content of 1984 leaded gasoline was
Io07 g/gallon and indicating  that leaded gasoline was  42.6% of total  sales.
Assuming that emissions are  proportional to gasoline sales (i.e., more pollu-
ting cars also consume more  gasoline), these data indicate a 1984 EDB species
fraction of 2701 ug EDB per  gram of gasoline vehicle exhaust HC emissions.,
The local percentages of gasoline vehicles exhaust to  total vehicles exhaust
for arterials and freeways was then used to derive the species fractions shown
on Table 50

For evaporative emissions, Sigsby et al  reported a best  fit equation that  EDB
evaporative emissions in ug/mile = 42 + (48 x HC emission g/mil).  This
equation reflects FTP operation, so an emission factor for Chicago vehicles
operated as in the FTP was input into this equation.  The result was adjusted
for lead content the same way the exhaust  species fraction was.  The result  was
17.1 ug EDB per gram evaporative HC emissions.  Since  diesel fuel results  in
minimal evaporative emissions, no adjustment from gasoline evaporation to  total
evaporation was necessary,

Sigsby et al. also measured  for but did  not find any ethylene dichloride  in
either their exhaust or their evaporative  emissions.  Therefore, zero species
fractions were used for this  pollutant.

The final step in deriving Illinois emissions estimates  was to determine  the
spatial distribution.  The most recent and applicable  data available for  this
purpose are 1985 VMT projections made in 1977.  However, despite how long  ago
these projections were made,  these data  should be reasonably accurate for  indi-
cating the spatial distribution of emissions.

In Indiana, the 1982 ozone SIP provided  direct estimates of emissions by
traffic zones.  Nevertheless, most of the  same adjustments made in the Illinois
inventory were also necessary in the Indiana inventory.   The SIP provided
summer and and winter emissions estimates  for 1980 and 1987, so SIP data  on
monthly traffic volumes was  used to estimate annual emissions totals.  A  county
total 1984 emissions estimate was derived  by logarithmic interpolation between
1980 and 1987 totals, and the ratio of the 1984 to the 1980 county totals  was
calculated for use in converting 1980 emissions data to  1984 estimates.  For
arterials. Northwest Indiana  vehicle operating characteristics were assumed  to
be similar to those in the southern Chicago area, so the same conversions  and
disaggregations from MOBILE2 total emissions to MOBILES  arterial exhaust  and
evaporative emissions were used.  For freeways, where  the vehicle mix includes

-------
                  Table 5.   Highway vehicle species fractions



                                    Species Fraction (in %)
Pollutant                        Exhaust           Evaporative

Benzene                           3.14%                1.09%

Butadiene                        .0345%                  0

Ethylene dibromide            .OOZ5%/.0018%*           .0017%

Formaldehyde                     .763%                   0

Ethylene                          7.15%                  0

Toluene                           6.59%                 6.3--;

Xylene                            5.84%                 3.8

Acetone                          .072%                   0

Benzo(a)Pyrene                .040%/.061%*               0
* Separate figures are for arterial/freeway exhaust emissions.  The differences
  result from the differences in vehicle mix.

-------
                                       17
                                                                              -<

more trucks, a separate Northwest  Indiana conversion factor was calculated with
the Indiana vehicle mix, translating to an emission factor of .652 tpy/1000
daily miles,,  The same adjustments  as in Illinois were made to reflect possible
undermsasuremant of exhaust emissions and undercalculation of evaporative
emissions.,  Finally, all of the same species fractions were used in Indiana as
were used in Illinois.

Hastewater Volatilization Inventory Methods

National concern has arisen about  volatile organics dissolved in wastewater but
then volatilizing into the atmosphere.   This volatilization can occur during
aeration and during other processing- at sewage treatment plants, during transit
through the sewer system, and during "pretreatment" of company treatment
facilities.  Unfortunately, methods for assessing the extent to which volatili-
zation rather than biodegradation  or transference into sludge and for assessing
losses within the sewers and at industrial pretreatment facilities are highly
uncertain,,  Therefore, no attempt  was made here to quantify the effect of these
factors.
The basis of the emissions estimates derived in thJlSLJAjJidj^Ar^jiiMSMrements i of
wastewater concentrations made at f5k-Sgisi^u^ft_Jx£aJungja^_^lajL£s7  The~Southea=sT=='
Chicago source area includes two major sewage treatment plants, both of which
have substantial industrial  input: the Calumet plant, and the Hest-Southwest
(Stickney) plant.  At both plants, a special set of measurements of a broad set
of organics were taken on seven consecutive days at each of these plants.
Measurements were simultaneously made of the same organics not just in the
incoming wastewater but also in the sludge and the treated effluent.
Unfortunately, the mixing and variable retention times inherent in the treatment
process mean that a given influent measurement cannot be directly compared to any
single effluent measurement.  Nevertheless, calculations were made to assess
whether the quantities in the sludge and the treated effluent for each compound
were. a substantial fraction of the quantities in the inluent.
                *$•

To estimate emissions, the conservative assumption was made that all of the
volatile contaminants coming into these plants actually volatilize.  Thus,
emissions were calculated as simply the product of the wastewater concentration
for each pollutant times the volume of wastewater treated.

This study only includes emissions at the sewage treatment plant.  An employee
of the Metropolitan Sanitary District indicated there is little if any indus-
trial pretreatment, and in any case there was no information available to
estimate any emissions.  A more serious omission is the absence of any estimate
of emissions out sewer system openings.  These emissions may he close to zero
or may be on the same order of magnitude as emissions of material reaching the
sewage treatment plant.  However, no estimate was attempted here, because any
estimate would be highly speculative.

Some wastewater treatment plants use chlorine to disinfect the incoming material.
This leads to some formation of chloroform, which may be presumed ultimately to
volatilize.  However, an employee of the relevant Metropolitan Sanitary District
stated that these plants^ dojnf norfnrm rhiojdLaatJjgg. so none of this source of
chloroform was assumed.  (Note, however, that the area source inventory assumes
that the drinking water is chlorinated and therefore releases chloroform.)

-------
                                       18

Inventory Methods for Hazardous Haste Treatment Storage and Disposal  Facilities^

Emission estimates for this source category are being developed hy a  contractor
and have not yet been completed.  Although the facilities have been identified,
the methods to be used have not yet been selected.   Emission estimates for this
category will be included in the inventory as soon  as they are available.

Summary of Results

A summary of the emissions from all categories of emissions is provided in
Table 6.  This table shows emissions estimates for  each of the 40 compounds
found in this inventory.  One column in this table  shows totals of point source
emissions estimates, one column shows area source totals, one column  shows
highway vehicle totals, one column shows total emissions from the two wastewater
treatment plants. Estimates of emissions from hazardous waste treatment storage
and disposal facilities are not yet available,,  A final column on this table
shows total emissions from all facility types in the Southeast Chicago source
area.

Following Table 6 are several tables showing the emissions estimates  for the
various source types.  A summary of area source emissions is shown in Table 7.
This table shows the estimates of total emissions of each species for each
source category within the full Southeast Chicago source as well as the por-
tions of this total  within each county.  For reference, this table also shows
what population in the Southeast Chicago source area resides within each county,
what the percentages of these populations these are of the respective total
county populations,  and what percentage of the 2,361.474 Southeast Chicago
source area residents live in each county.  These data illustrate that even
though barely a third of Cook County is within the  source area, this  county
still dominates the overall source area.  By comparison, the receptor area has
395,969 people.  Thus, while the receptor area has  about 8% of the area in the
source area, it has  16.5% of the total source area's population.

Table 8 shows similar information for highway vehicle emissions as Table 7
shows for area source emissions.  Table 9 shows emissions from the two waste-
water treatment plants.

Following Table 9 are four tables representing the  point source inventory.
Table 10 shows emissions estimates for those 29 facilities to whom question-
naires were sent.  Table 11 shows coke oven emissions estimates from the four
facilities in the area with operating coke oven batteries.  These emissions
estimates are taken  from a draft background information document for a proposed
coke oven NESHAP, based on questionnaires sent to these firms as part of the
NESHAP development.   Table 12 shows emissions estimates for organic species for
the other 59 facilities present on the NEDS.  Table 13 shows emissions esti-
mates for metals for these other facilities that are estimated to have such
emissions.

One special entry in Table 13, next to the entry for Commonwealth Edison's State
Line plant, is an estimate of polycyclic organic matter from this facility.
Although this table is the most convenient location in this report to place
this emission estimate, this estimate was in fact estimated by special methods.
Specifically, this emission estimate was derived from reference P.5 on fuel use
at this plant multiplied times an emission factor taken from reference P.4.

-------
             Table 6. .Summary of Emissions from all Source Types
Pollutant

Benzene
Methylene chloride
Perchloroethylene
Trichloroethylene
Chloroform

Formaldehyde
Vinyl chloride
Vinylidene chloride
Ethylene
Ethylene dichloride

Ethylene dibromide
Butadiene
Ethylene oxide
Methyl chloride
Benzyl chloride

Styrene
Gasoline vapors
Toluene
Xylene
Acetone

Arsenic
Cadmi urn
Chromium
Mercury
Nickel

"Heavy POM"
Total POM
Coke oven emissions
Diethanolamine
Epichlorohydrin

Propylene oxide
Acrylonitrile
Pentachlorophenol
Acrylamide
Ethyl acrylate

Mel amine
Carbon tetrachloride
Nitrobenzene
Titanium dioxide
PCRs
Beryllium
Point
Area    Mobile
STP
Total
4860.0
7.6
171.0
10.2
.8
29.6
.1

568.8
.2

12.2
16.1
.3
.05
4.2

1972.1
836.3
522.8
21.2
19.8
105.9
3.8
6.5
.01
.01
740.9
.07
.1
.6
10.5
.003
.03
.4
.3
.8
.4
.3
.04
.001
43.9 1113.4 2.0
1490.3 23.6
1126.2 16.4
401.4 5.2
34.2 2.0
, 122.4 229.5

.03
2137.7
2.0
1.0
91.3
12.3
12.0 .2
3.7
6.6
5221.8 20396.5
372.6 2977.6 18.9
70.1 2343.6 43.3
21.4 507.3


1.3

16.3
12.3 12.8
18.6 12.8














6019.3
1521.5
1313.6
416.8
37.0
381.5
.1
.03
2706.5
2.2
1.0
103.5
28.4
12.5
3.8
10.8
25618.3
5341.2
3293.3
1051.5
21.2
19.8
107.2
3.8
22.8
25.3
31.4
740.9
.07
.1
.6
10.5
.003
.03
.4
.3
.8
.4
.3
.04
.001

-------
                     Table 7.   Area Source Emissions Totals
Category
Pollutant
       emissions (tons/year)
Total   Cook   DuPage   Will   Lake. In.
Gasoline
Marketing


Ship 4 Barge
Transfer

Degreasing


Architectural
sfc. coating



Aerosol cans
Paint stripping
Dry cleaning
Misc. products
Drinking water
Heating
(except wood
stoves)


Wood stoves

Chrome plating
Hospital
steri 1 izing
Comfort cooling
benzene
toluene
xylene
gasoline vapors
benzene
toluene
xylene
perchl oroethyl ene
trichloroethylene
methylene chloride
benzene
methyl chloride
methylene chloride
toluene
xylene
methylene chloride
methylene chloride
perchl oroethyl ene
formaldehyde
chloroform
formaldehyde
nickel
chromium
"Heavy POM"
Total POM
"Heavy POM"
Total POM
chromium

ethylene oxide
chromi urn
32.9
33.4
9.4
5221.8
3.0
3.1
.9
242.2
401.4
295.4
8.0
12.0
150.2
338.8
60.6
590.3
454.4
884.0
56.7
34.2
65.7
16.3
.143
3.4
3.4
9.1
15.2
.7

12.3
.46
30.1
30.5
8.6
4771.6
1.4
1.4
.4
194.2
332.4
277.8
6.7
10.1
126.3
284.8
51.0
502.2
364.7
714.7
48.2
29.1
54.3
10.8
.09
2.2
2.2
7.8
13.0
.7

10.4
.39
.2
.2
.05
30.8
0
0
0
1.2
2.3
2.0
.04
.06
.7
1.6
.3
12.5
9.8
21.8
1.2
0.7
1.0
.2
.002"
.04
.04
.1
.2
0

.6
.01
1.4
1.5
.4
230.0
.1
.1
.04
3.1
4.7
3.3
' .2
.3
3.4
7.7
1.4
15.4
9.4
15.1
1.5
.9
1.1
.2
.002
.05
.05
.2
.3
0

0
.01
1.2
1.2
.3
189.4
1.5
1.5
.4
43.7
62.1
12.2
1.1
1.6
19.8
44.7
8.0
60.2
70.5
132.4
5.8
3.5
9.2
5.0
.047
1.1
1.1
1.0
1.7
0

1.3
.05

-------
           Table 8.  Highway Vehicle Emissions  Totals  (in  tons/year)
Pollutant



Benzene



Butadiene



Gasoline vapors



Ethylene



Formaldehyde



Ethylene dibromide



Polycyclic organics



Toluene



Xylene



Acetone
Arterial Exhaust
823.fi
80.0

1871.7

201.2

.7
10.5
1731.4
1520.8
18.7
Freeway Exhaust
llfi.7
11 2
1 A • L.
266 0
(,\J*J * \J
28 3
c. \J • J
.07
2 3
c. • «s
240.2
214.5
?_7
Evaporative
173.1

onooc c
cUjyo. b



.3

1006.1
608.3

Total
1113.4
01 7
71 . O
on o rt £ c
20396. 5
°1 07 7
c 1 J/ . /
OOQ C
£.£.? * 5
1.0
1 O Q
1C . O
2977.6
2343.6
91 A

-------
           Table 9.  Wastewater Treatment Plant Emissions Estimates
Pol Jutant

Benzene
Methylene chloride
Perchloroethylene
Trichloroethylene
Chloroform
Vinylidene chloride
Ethylene dichloride
Methyl chloride
Styrene
Chlorobenzene
Toluene
Xylene
Acetone
 Calumet
  1.7
  3.3*
  1.4
   .8
   .7*
  .03
   .4
   .2
  4.6
   .7
 12.0
  8.7
301.0
West-Southwest

      .3
    20.3*
    15.0*
     4.4*
     1.3*

     1.6

     2.0
     3.0*
     6.9
    34.6
   206.3
*Because contaminant quantities in the treated effluent are a significant
   fraction of the quantities in the raw influent (suggesting limited
   volatilization), these amounts represent the difference between influent
   and effluent quantities.

-------
     Table 10.   Emissions  Estimates  for  Facilities  Receiving Questionnaires
Nalco Chemical
       Methyl  chloride
       Benzyl  chloride
       Diethanolami ne
       Epichlorohydrin
       Formaldehyde
       Pentachlorophenol
                                               PMC
Desoto
       Titanium dioxide
       Acrylamide
       Ethylacrylate
       Mel amine
       Propylene oxide
       Styrene
       Formaldehyde
       Toluene
       Xylene
                            .0016 tpy
                            .048
                            .065
                            .10
                            .278
                            .003
                            .20 tpy
                            .025
                            .375
                            .25
                            .075
                           1.95
                            .080  "
                           2.7
                           2.1
            Formaldehyde
            Toluene
                       .0015 tpy
     Stolt  Terminals
            Perchloroethylene
            Tri chloroethylene
            Methylene Chloride
            Benzene
            Toluene
            Xylene
            Acetone
     Clark
Oil
 Benzene
MORECO (Formerly Motor Oil  Refining)
       Arsenic
       Beryllium
       Cadmi urn
       Chromium
       Nickel
       Benzene
       Carbon Tet.
       Chloroform
       Methylene Chloride
       Nitrobenzene
       PCB
       Trichloroethylene
       Toluene
                            .047  Ib/yr.
                            .0011
                            .18
                            .32
                            .015
                           !.3
                            .77
                            .77
                            .77
                            .41
                            .039
                            .77
     McKesson  Chemical
            Trichloroethylene
            Perchloroethylene
            Toluene
            Xylene
            Acetone

     Union Oi1
            Benzene
     Amoco
            Benzene
            Toluene
     Conoco
            Benzene
                      1.73 tpy
                      6.55
                      5.68
                      1.09
                      3.09
                      1.45
                      1.17
                                15.  tpy
                     2.9 tpy
                     2.0
                     7.2
                     5.7
                     6.8
                       3.59 tpy
                        .39 tpy
                      1.6
                        .09 tpy
                          18.
Koppers
       Styrene
                            .83 tpy
     Marbon Div./Anderson Devel.
            Styrene               .19 tpy
            Butadiene            8.75
                                               Keil
                                               Shell
Ford Motor
       Benzene              .06 tpy
       Methylene Chloride  1.
       Perchloroethylene   1.
       Toluene            70.1
       Xylene            165.8
       Acetone             1.6

Sherwin Williams
       Titanium Dioxide      .06 tpy
       Chromium              .001
       Toluene              2.31
       Xylene               1.347
       Acetone               .07
           Emissions were estimated to be zero for
                                                      F.ttiylene Dichloride   .004 tpy
                                                      Benzene
                                  .88 tpy
                                               Stauffer, IN'
                                                      Perchloroethylene     .010 tpy
                                                      Methylene Chloride    .113
                  Trumbul
                  3 M
                  Si gnode
                  Dynagel
                  Getty
                  Stauffer
                           Asphalt
Invite
Diamond Shamrock
PVS
Unichema (formerly Darling)
Mobil Chemical
Texaco

-------
             Table 11.   Emissions  from coking operations
                     L      (All  data  in  tons per year)
                    Coke  Ovens*
                                   Byproduct  Plants
Interlake
LTV Steel
(Chicago)
Inland Steel
U.S. Steel
"Coke oven emissions"
68.61 tpy
23. 27 .
406.14
242.83
Benzene
474.4 tpy
225.8
1303.4
2666.8
Toluene
75.2 tpy
35.8
206.8
423.0
Xylene
26.2
12.5
71.9
147.1
Total
740.85 tons/year
4670.4 tpy   740.8 tpy
257.7 tpy
*These are total  emissions  of "coke oven emissions"  (benzene soluble  organics)
   that are emitted from charging and l*aks at the coke  ovens.

-------
    •A»L= 1*.    PC:NT SOURCE EMTSSIQNS ESTIMATES BASED
                     :N ORGANIC SPECIES FRACTIONS
1-*:
154?
1T4C
1540

174C
1*40
1?4C
1T4T
1?4C
154C
174C
1*4?
1'4C
1'4-

1740
154?
1-4C
154C
PLTI     OLANT NCME

  i  «H*R:C«N HETBL OECQR
  :  Vtf: L3'8 CONTAINERS
  5  M " 'iiTH PBPER cc
  6  1ILE3 LAEC5ATORIES -
  e
 ir
 11
 i:

 if
 :c
 24
 3*
 42
 r4
 5P
 61
 69

 PC
 el
 87
 PC
 94
 98
IOC
117
                                   ETH  136UT  HCHO
                                                       CcMISSICNS  IN  TONS/TEAR)
                                                      ACE    CTC  PR.O   ACN    MCL
                                                                              EDC  PERC   VCL   XYL  BENZ    TOL   STTR
         '7'r TMiri'AL CAN CCHP
         C:TN PKOPUCTS-ARSO P
         r-a*X TjPNINALS COPP.
         -• U H*CTPIC SPECIAL
         CHTCSO: ^INISHEO MET
         C"TrN CIL t CHEM CC
         :BLUHPT *TE?L CCMPAN
         CH'CAGC HEIGHTS BAR
         "HI-ALL CAP MFG cc PL
         BI'CC ?TL CC - CHICA
         IWfiN GG^OCN CO
         BLLIED TUEE t CONOUI
                  *?*c?s - ELS
                  M-TALS cc
                  . INC. 9IVE
         CB"AVELLr MD30 PRCDU
         •E'RTPCLTTAN
P32C
FTO
      10
                 CHCMTCAL CO
         5»rCRA*T CCPP. OF AM
         U * *T£(FL - 1CUTM WC
         TN-E-LAKI: - CHICAGO
         LTV STEEL CCNPANT IN
         CH'CAGC ?rT«%PINT
         CA-CTLL !HC - CCNHOO
         A*-?:CAN CAN CO
         9LrC G'AVURF INC
         NA'TCNAL CAN COPP
         PHTEM HFG.CS.
         MEfO °ACIC AGING
         LARE-RTVcR TERMINALS
         lOrRC AMERICAN GRAPH
         NA'IINAL CAN CORP -
         UH'CN CHEMICALS 3IV.
             CO - MILL COUNTY
             C PIPELINE co c
.0
11.8
1.8
2«.l
.2
5.8
1.4
.0
.0
.0
.0
9.7
.0
.0
.0
.0
.0
3.8
.0
.0
2.S
.0
.0
.2
79.3
.0
.0
.0
.0
.0
5.8
.0
.0
2.1
.0
.1
.0
.0
2.8
.2
.0
.0
.0
.0
.4
.0
.0
.0
.0
.0
.0
.0
.2
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
, .0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.2
.2
.0
.0
.3
.1
.0
.0
.3
.0
.0
.0
.0
2.3
.0
.0
.1
.6
.0
.9
.1
.1
.0
4.2
6.S
.2
.0
.1
.0
.0
.0
.0
.0
.1
.0
.0
.2
2.6
.0
70.7
27.3
10.3
.0
.0
.0
.2
6.9
27.9
12.6
7.5
.0
10.3
13.5
.0
.0
40.1
124.'
.0
3.6
15.6
.0
.3
10.9
1.4
4.8
.0
.3
21.9
.0
22.8
2.0
1.0
24.4
1.7
.3
11.7
25.4
.0
.0
.0
.0
.c
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.c
.0
.0
.0
.0
.0
.0
.6
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.c
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.2
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.4
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.1
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.1
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.1
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
9. T
.0
.0
.0
.0
5.6
27.9
12.6
6.1
.0
.0
13.5
.0
.0
40.1
10.1
.0
.0
.0
.0
.0
8.9
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
11.7
16.2
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
. .0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.1
.0
.0
17.4
50.0
3.»
.0
11.4
.0
.8
.0
11.2
5.0
.0
..o
6.6
5.4
22.9
.0
16.0
111.3
.0
5.1
28.5
.0
6.2
.0
.0
.0
.0
.0
.0
.0
16.5
.0
5.5
22.4
.0
1.9
4.7
12.. 8
.2
8.2
12.7
.0
.0
.0
17.4
2.1
.0
.0
1.6
.0
.0
1.9
7.0
.0
.0
4.1
.3
.0
.2
.0
.2
.0
.4
1.1
2.6
.0
.9
.0
.0
.0
.0
.0
.0
.0
.0
.0
.3
.0
.0
1.3
.0
2.4
43.0
123.4
4.3
.8
8.2
.0
?.*
5.8
11.2
5.0
6.5
.3
5.2
5.4
16.5
.1
16.0
262.4
.0
12.8
70.5
.2
4.5
9.2
21.5
.5
.0
.0
327.6
.0
1S.1
29.8
4.4
35.4
25.3
1.5
4.7
11.5
.5
5.7
9.2
.0
.0
.0
.0
.0
.0
. 0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
-.0
.a
.4
..0
.0

-------
               PLANT  NAME
                               ETH  13BUT  HCHO
                                              (EMISSIONS IN TONS/TEAR)
                                             ACE   ETC  PR.O    ACN    NCL
                EOC   PCRC    VCL    XrL   BENZ    TOL  STfR
336?
:-»«r
2?*?
2*60
11-
127
::i
rcc
         crcF 650  H
a*" PIP?LINF ST JOHN
NI-SC7 ?rAN H NITCHE
U.'.STfEL C •:AN»
~L~M KN7X COUNOR»  AN
-a-- CM: tir.r HUN re t*1
r •  -"'jc^T nr N510U9
rNLflfJl S'?TL INDIANA
TNLAN1 S'TCL INCIANA
LTV -TTPL crrpANt
US'CN CAi"BIOc C?»P L
U»<*ON TANK CAR CCMPA
.0
.3
.0
.3
108.3
7.8
.0
.0
.0
.0
.0
39.1
4.3
93.9
112.7
50.*
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
1.7
1.2
.0
.0
.0
.0
.3
.7
.0
.4
.0
.2
1.2
.2
.0
1.4
.0
.4
.0
l.B
2.5
.1
.2
.0
.0
.0
.0
.0
3.9
.0
.0
.1
.0
.7
.0
.0
.0
.0
1.4
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.9
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.3
.c
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.6
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.2
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
3.9
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.1
.0
.0
• .a
.0
.0
.0
.0
2.1
.0
.4
1.6
.0
2.3
.0
4.7
.2
.0
.3
.0
4.2
.0
.0
.0
.0
.0
.0
.3
.5
.0
.1
.0
.1
.6
.0
34.6
2.0
47.8
34. 5
2.2
.0
.1
.0
.0
1.3
.2
.7
1.6
.0
l.S
.1
3.4
.6
(.0
.8
2.4
4.5
.9
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.7
.0
.0
.9
.a
.0
        C---   5-?  'LAN'S)
                                      E'H

                                    568.842
                                           HCHO

                                          29.218
                                                 ETO

                                                 1.610
ACN

1.050
EOC

 .176
VCL

 .210
 BENZ

167.085
SMR

 1.190
                                          139UT      ACE      PR.O

                                           3.480   513.189       .490
                                                                    NCL       PERC        XTL       TOL'

                                                                      .315    166.287   402.238  1125.369
                     - ETHTLEN?
           =*0
           ACN
           rrc
           VCL
           9"NZ
                       ETHTLENS  OXIDE
                       ACPTLONITRILE
                       =THTLCNE  DICMLOPIOE
                       VINYL  CHLORIDE
                     - STTREN5
                                         13BUT- 1.3-SUTftOrENE
                                         ACE  - ACET3NE
                                         PR.C - PPOPrLENE OXIDE
                                         MCL  - METHtL CHLORIDE
                                         PERC - PEP.CHLOROETHYLENE
                                         XTL  - XTLENE
                                         TOL  - TOLUENE

-------
     'fiFLc 1?.  STINT
                       S3U»C:  EMISSION
                             ^ MATTER
                                       ESTIMATES  EASED  ON
                                       SPECIES
CN-Y
                 NT NIHr
                                    AS
                                        EMISSIONS  IN
NI
HG
1'4C
1S4C
1T4C
lt4C
1540

154?
1-4"
                 Mtr^^PC -  ~L"
                 6TLS-L:>«5  "L
                 ?,  IMC.  -IVE
       ? S^'LL SIL  CCMFANY  ,
      i! ::~N 9-
      :i :ALUME
1540  to srrs^ai
154?  44 VJLCAN
      f 4 IS
      5? Sr
      ce "C-eLrMEAC  LIME  CC
      -I "i"«LEVEAE  LIME  CC
      "3 it * *TeeL  ~  SCUTW  U ^
      "? H-f-LaKS  -  CHICAGC
      ?: L*V STEEL  CCMCANY  IN
     1G6 Cir!?r?3 INC
     ICE *5?KET7 ENGIMEE»ING
154" 11- LAfE-PTVER  -cPMINALS
154? 1Z1 VA'ICHAL CA»1 CCPP  -
8?2C  21 ??f «D - HILL CCUNTY
236? 11" MI"SCC CEAN H MITCHE
2?*C 120 U. '.STCEL  CCC-P.  C-A~Y

2260 :iO ?:H"CMH=AL~H EDISCN«
2360 ?C5 tLtU KNCX  FCUNOPY AN
216C ::s sa?* CHICAGC  MUNICI?
2~^v ?16 INLAV?  -TE=L  INDIANA
2?.*~ ?1* :NLA'!3  STEEL  INCIANA
236C ?lc
.CGI
.000
.013
.011
.053
.OOC
.582
• C02
.030
.000
.335
2.239
1.962
.929
.54-"
.000
.OOC
.5 1C
.135
1.175
6.S99
.633
.012
.004
.014
.261
1.035
2.784
.OOC
.3CO
.302
.011
.CCS
.COO
.581
.COO
.000
.000
.309
3.2?3
1.962
.929
.547
.000
.OOC
.053
.014
1.175
6.63?
.065
.OC1
• CC&
.083
.261
.954
2. 784
.004
.OCl
.017
.044
.C49
.030
2.453
.014
.009
.02E
1.488
17.646
1S.6C4
5.C70
2.980
.003
.002
.475
.125
6.396
35.972
.589
.069
.018
.026
1.166
5.228
15.059
.081
.016
.012
.000
.036
.000
.115
.263
.000
.COO
2.984
.125
.097
.000
.000
.049
.032
.352
.093
.000
.266
.436
1.265
.OOC
.010
.000
.066
.093
.OOC
.OOC
.00(
.001
.OOC
.00<
.00
,00(
.00
.00(
.001
.001
.001
.00)
.00
.00
.001
.00
.00
.00
.00
.00
.00
.00
3.33
.00
.00
.00
      »§ ^ r*
      L J.C
              'LAN*  NftME
                                         EKISSICNS  IN
                                    AS         en         CR
           H6
                                   21.144     19.612   105.564
                                                                  6.479
                                                                             3.83)
                                    J?  =STIMATPD  TC EMIT .012 TCNS/YEA*  pen
                                             C3  -  CAOMIUK
                                             HG  -  MERCURY

-------
                                       28
                                                                               so*

Discussion of Results  u

The most important question that can be addressed with the results reported in
Tables 6 to 13 is what source types contribute most significantly to emissions
of air toxicants in the Southeast Chicago area.  The first place to look for
this kind of information is Table 6,  This table shows the contribution from
each general kind of source toward the emissions of each pollutant.   Obviously,
the relative contribution from each kind of source varies for different
pollutantSo  For example, benzene, ethylene, and toluene are about equally
emitted by point and mobile sources,,  Methylene chloride, perch!oroethylene,
and trichloroethylene, on the other hand, are emitted predominantly by area
sourceso  Formaldehyde and acetone, much of which come from combustion, are about
equally emitted by area and mobile sources.  The metal species are predominately
emitted by point sources.  Butadiene is estimated to be emitted almost entirely
by point sources.  Sewage treatment plants lead to almost half of the area
chloroform emissions, but emit less than three percent of any of the other
compounds inventoried.

The other tables show more detailed information about particular sources or
source types that contribute to those various pollutant emissions-  For benzene,
the major fraction of benzene emissions from point sources comes from steel
mills (coking operations), and about three quarters of the mobile source benzene
emissions come from exhaust emissions on arterials.  Methylene chloride is
emitted in similar quanties from aerosol can usage and paint stripping.
Perchloroethylene emissions mostly arise from dry cleaning but also arise in
significant quantity from degreasing.  Trichloroethylene emissions are almost
exclusively from degreasing.  Formaldehyde emissions are mostly from combustion
by automobiles and by furnaces used in space heating, though an estimated 10%
arises from volatilization out of assorted consumer products.  Ethylene is
emitted predominantly by steel mills and by automobiles.  About half the
butadiene is emitted by an oil-fired utility power plant and the other half is
emitted by the steel mills.  Steel mills are also the dominant source of
chromium, arsenic, and cadmium.

Another way of analyzing the point source inventory is shown in Table 140  This
table summarizes the point source emissions of each organic pollutant by standard
classification code (SCC).  Note that this table only includes emissions from
facilities to which no questionnaire was sent, since the questionnaire responses
did not identify SCC's for each set of pollutant emissions.

Perhaps the best means of judging the relative importance of emissions of
different compounds from different source categories is to compare the health
risks posed by each set of emissions.  This comparison, however, requires a
dispersion analysis of the exposure to the area's residents from these emissions
and then a risk assessment using dose-response data for each compound to translate
exposure into risk.  A thorough dispersion analysis is planned but has not been
completed, and many of the dose-response relationships are subject to question0
Indeed it must be highlighted that many of the compounds may not in fact be
carcinogens, notwithstanding the availability of an estimate of carcinogenicity
per unit concentration.  Thus, a review of the emissions data and available
dose-response data provide only a qualitative comparison of the relative
importance of different pollutants.

-------
   TAELF :*.  POINT £OURC? EMISSIONS ESTIMATES SORTED B»
                      STANDARD CLASSIFICATION CODE CSCC)
sec
VCC *
                           A. 3RGANICS
ETM.   13 BUT  FORM  ACETONE  ETO
                                                       PRO
ACN   CHCL3   EDC   PBRC  VINTLCL  XTL.  BENZENE  TOL   SffR.
10100231
1C100202
10100203
10I0022?
101C06G1
K20020:
102CC?0*
102COIC1
10200r?l
1G2C0511
io:co;o2
moon*
102007?7
1C5-T106
20:0020;
•>Cl0?p(»
3130CEJ3
30300-5C1
303007M
30700933
•»C??09?4
105C169*
1090019S
•30199<9S
?SOCC*"9
193CC691;
19090702
40250101
4CC00201
4C230S01
40200P01
40200610
40290?C1
4
-------
«/>:'%:*:x
*c:oi?>»<5
ii2*i~?z
4C:M?:?
*:23i:'>5
*?::iT:e
*C23!?'P

4C2frroi
4C:«J-}599
4?:cioci
4-ncicr:
4"Ci:?7
4 Vip-;"
*T"1 101

4 : * - : M i
«*•*./*! * 1 O
. „' 1 1 I ' '
4~**»rflrj
4-torii'.
4TCC131
4:^r(:rci
4r«OC:M
4C6TCU7
*:'."ci"7
4C?-n<;?<)
5c::cn:
U6.070
3.0CC
4'!s.'3co
54.000
17.000
2C1.0?C
37.010
2.000
6?7.000
1.000
rs.coo
l.OOC
ns.oco
4*. CIO
I:!.OTO
72.770
•. • . : o c
l.COC
:i7. .0
. 0.
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
. 0
.0
.0
.0
t .0
1 .0
2 .0
1 .0
9 .0
0 .0
5 .0
3 .0
            vcc :*
            :TM.
                                    FC«M  CCETONF  ETC    PRC    ACN   CHCLS   EDC   PERC  VINUCL  KTL. BENZENE  TOL   srru.
 T'BL
           1-6-7.«4i  "«e.E    3.'   29.2  513.2    1.6
                                                       1.0
                                     .2  166.3     .2  402.2  167.1 1125.4    1.2
nrr:
           F'H
:rC
                - ?THTLENE
                - cC«JHetO =
                - -TMTLENr CKIOE
                - FTHTLENE CICHLOPIDE
                - VINYL CHLOBIOE
13EUT- 1,3-BUTADTENE
ACE  - ACET3NE
PR.C - PBOPflENE OXIDE
HCL  - HETHfL CMLCPICE
PERC - PERCHLPRCETHYtENE
XYL  - XYLESE
TOL  - TCLUSN?

-------
                                                     NICK-L
10I032ri
10103232
1010C2C3
10130222
102302C2
102*0204
1C2334C1
1C2CC404
13230501
?033C?<31
33?00?32
•»f»-« « r e ~ ^
— -*«*»w— Jw
20200?? »
3333C?11
3??3CS12
33?OC«13
3C3CCM4
•s p i •} n » •? i

?C?33522
3C3 ?C?24
293?Cr25
302005"?
f 33CC5C4
?0?OC31C
3C73C511
3330C512
2333C31?
3C233~I4
30?3C31*
33:cc?2i
IQirTZT*
-t W •/ %« T _ to
7 «^ £ «S •»•»
20730991?
•»34nc"'l
234<:0'01
3"4007C6
334CC715
?C43?7'3
-------
                                       32
                                                                               ^
One finding from a review of emissions and dose-response data is ironically and
unfortunately that the compounds that appear most significant are associated
with sme of the greatest uncertainties.   Polycyclic organic matter may be  the
largest contributor to area risk, but is clearly one of the most uncertain
estimates in the inventory. This "pollutant" is actually a class of compounds
with widely varying toxicities,  and measurement of emission factors for this
spectrum of compounds is difficult and has been done only a small number of
times for any given source category.  Emissions of polycyclic organic matter,
which emanate mostly as products of incomplete combustion, are estimated to be
emitted in similar quantities from residential wood combustion (including
wood stoves and fireplaces) and  from highway vehicles,  with a smaller contribu-
tion from other home heating and only a  minor contribution from point sources.

A second compound that may contribute significantly to  the total risk is chromium.
An important uncertainty here relates to the different  valence states of chro-
mium, each of which has different level  of toxicity. Unfortunately, while
for the two major area sources all emissions are the most toxic form (Cr*6),
for the largest source of chromium (steel furnaces) there is little basis  for
estimating the distribution among valence states.  For  this inventory, these
emissions were conservatively estimated  to he all the most toxic form.

Other pollutants that may make significant contributions to the total risk
include coke oven emissions, arsenic, formaldehyde, benzene, and butadiene.
Note that photochemical formation of formaldehyde, which may contribute an
order of magnitude more than direct emissions, was not  included in this study.
There is also significant uncertainty on the most appropriate estimate of  the
human carcinogencity of formaldehyde.

As implied by the above discussion, area and mobile sources may contribute most
of the risk, but point sources also contribute significantly to the total  risk.
Wastewater treatment plants appear to contribute only a very small fraction of
the total risk.  Note that emissions estimates are not  yet available for
hazardous waste treatment storage and disposal facilities, and so no estimate
can yet be made of their contrbution to  the total risk.  Once again, however,
a more quantitative evaluation of the relative risks caused by various source
types and various pollutants must await  the full dispersion and risk assessment
that is planned.

As mentioned previously, for point sources that were sent questionnaires,
emission estimates were also made using  the species fraction approach.  This
may be viewed as providing quality assurance for the methods used here. The
questionnaire results were shown in Table 10.  The results of the species
fraction approach for organics is shown  on Table 15 and for particulate species
is shown in Table 16.  The sum of questionnaire emission estimates for each
compound from the relevant companies is  also shown on Table 15 and 16 to facili-
tate comparison of the results.   Note that only 15 of the facilities to whom
questionnaires show up on Tables 15 and  16, since these tables are based on
updated NEDS data in which many  of the smaller facilities (including several
chemical plants) have been discarded.

-------
                '5
 N'T °LTn
              NAME
t't"
1T41
            A3F"ALT  CC.
            -"ICiL C3.-
 c •>•< rri°tNY - TNOUST"?
i" ' TrnT" rr°p

 IT u •* ~ o " ~ T I ^C~*Mv\lf~ f
 C  . _ - , I L '  . . *< 4 * Ij I,
1"4C
9Tr
• r :*-CW:N-WULTAHS  cc
i: -°"c "nr-p COMPANY c

 1 M--TL CH-r
-------
"BEL: If.   Pac'ICULtT?  MAT-£i? SPECIES FRtCTIZN-BASSO SSTIMATES
                 FTP  FACILITIES DECEIVING S
                  T N&ME

          '?UMS'JLL ACfHtLT  C'
154?  12
1T41?  56
154C 127

8?2C  2C
     T'O'RS CCMPiNY  INC.
     LiPK CILES=^rNISG
     N-CN CIL  C.?.  -  CHIC
     ?rAC1 B^FIN'MG  ANO
     i-c 3IL CC  2831  INC
      "EMISSIONS  IN  73MS
ARSENIC   CADHIU^   CH9CMIUH
                                                            NICK5L    MERCURT
.000
.006
. 001
.005
.000
,C30
• CO*
.125
EMI
i
.001
.ceo
.ccc
.007
.C27
.Cll
.023
SSICN«J IN
CACMIUM
,0f5
.005
.006
.031
1.67Z
.874
.276
.883
TCNS
CH«!CMIUM
3.806
C.323
.000
.004
.114
.568
1.982
2.448
1. 367
14.7S3
NICKEL
21.265
C.015)
.000
.000
.000
.000
.002
.011
.003
.006
NERCUR'
.02'
    CALSC,
                           - .OC11

-------
                                       35
                                                                               x*^
The results in Tables island 16 indicate that total  emissions according to
questionnaire responses in general  were found to he  substantially lower than
total  emissions estimates using the species fraction approach.,   In some cases,
the difference was dramatic:  species fractions were  higher than questionnaire
totals for benzene by a factor of 16, for styrene by a factor of 25,  and for
formaldehyde by a factor of 566,  Also, the species  fraction approach estimated
some emissions of several compounds which were not found in the questionnaires,
including ethylene oxide, propylene oxide, acrylonitrile,  and vinyl  chloride.
In other cases, the species fraction totals were only moderately higher than
questionnaire totals, such as for toluene (a factor  of 6)  and xylene (a factor
of 2,6), or even lower (for cadmium the species fraction total  was 2 times
lower than questionnaire totals),  IR yet other cases, however, some relatively
common chemicals, specifically methylene chloride and trichloroethylene were
not estimated to occur using the species fraction approach but were found in
the questionnaire responses.  Further, it is not surprising that relatively
esoteric compounds such as diethanolamine and epichlorohydrin were found with
questionnaires and were not found with species fraction approach.

In order to make an overall comparison among the multiple compounds, it is
useful to make a qualitative review that considers dose-response relationships
as well as emissions.  This review suggests that overall the species fraction
approach leads to a substantially higher estimate of risk than questionnaire
responses.

The results as just discussed by the questions of why the differences arise and
what kinds of sources have the greatest differences.  In general, differences
between the two sets of results may arise either from differences in total
emissions estimates (i.e., total VOC or TSP) or from differences in species
fractions.  Unfortunately, the questionnaire did not ask for total emissions
estimates, so it is difficult to distinguish these two origins of discrepancies.
Nevertheless, some enlightenment comes from looking in detail at three types of
sources: chemical plants, refineries, and miscellaneous manufacturing operations.

Chemical plants exhibited the most dramatic between  species fraction results
and questionnaire results.  For these facilities, the species fraction approach
appears to rely on a national mix of chemical production,. The result, not
surprisingly, bears little resemblence to the chemicals produced and emitted in
the Southeast Chicago area.  It is these  facilities  which were estimated to but
have been affirmed not to emit ethylene oxide, propylene oxide,  acrylonitrile,
and vinyl chloride.  On the other hand, these facilities did emit non-zero
quantities of butadiene, fliethanolamine,  epichlorohydrin, for which the species
fraction approach estimated little or no  emissions.

The second source type,  refineries, generally requested U,S, EPA assistance
particularly in estimating species fractions.  Thus,  regardless of the actual
accuracy of the species  fraction used, any discrepancy between species fraction
results and "questionnaire results" is caused by a discrepancy in total VOC
estimates.  For example, for the Texaco refinery, the species fraction approach
estimated substantial emissions whereas the questionnaire response indicated
that the facility had largely shut down.   It  is interesting to note further
that the species fraction approach predicted  substantial  refinery emissions  of
chromium, from combustion of chromium-containing oil for heating.  This is
illustrative of the possibility in some cases that the species fraction approach
may find substantial emissions which are  not  suspected by the sources themselves,

-------
                                       36
                                                                              ;*•
The third source type,  miscellaneous  manufacturing operations,  sometimes  shows
surprisingly good correspondence between  species  fraction  results  and  question-
naire results and sometimes  shows firms estimating zero emissions  where species
fractions indicate substantial  emissions.  It  is  possible  in  some  of these
latter cases that the firms  use solvents  for example that  are mixtures which,
unknown to the firms, contain the compounds of interest.

In summary, the species fraction approach often given reasonable screening
estimates of emissions, but  the results sometimes differ substantially from the
emissions estimates made by  the firms themselves.  Particularly for chemical
plants, the use of national  default  species fraction cannot be  expected to pre-
dict what chemicals are used at local  facilities.  For other  facilities,  however,
the species fraction are generally at least comparable to  source estimates, and
in fact the species fraction may in  some  cases be more accurate than the  source's
own estimation of its emissions mix.   In  some  cases, inaccuracies  were a  result
of inaccurate VOC/TSP emissions data  in NEDS.   In general, though, in  the
context that this inventory  is  intended as a screeening inventory, the species
fraction approach appears to give reasonably reliable emissions estimates for
this purpose.

General Observations

Substantial interest has arisen in the urban air toxics problem.  The inventory
described in this report is  part of  a unique project to assess  one urban  air
toxics "hotspot".  It is anticipated  that this inventory will facilitate  the
development of other, similar urban  area  inventories.  Therefore,  it is
appropriate to discuss  the process of inventory development and make other
comments that might he  useful  to other groups  developing similar inventories
for other areas.

Given the goal of compiling  a comprehensive inventory,  this inventory necessarily
relied heavily on information available in the literature.  Even for the  point
source portion of the inventory, resource constaints led to sending direct
questionnaires to only  29 out of 88  facilities.  For the other  59 sources,
emissions estimates for VOC  and TSP were  available as part of NEDS.  By neces-
sity, literature information, particularly information from the VOC Species
Data Manual and the Receptor Model Source Composition Library,  were used  to
assess the toxic compound components  of these  VOC and TSP  emissions estimates.

Fortunately, the above  two references provide  relatively complete information
on point source species fractions.  Unfortunately, species fractions for  area
source categories were  much  more difficult to  obtain.  For the  point source
inventory, the inventory appears to  be relatively complete, and the principal
question is the accuracy of  the available estimates.  For  area  sources, there
is more of a question as to  the completeness of the inventory.   That is,  it is
more evident in the area source inventory than in the point source inventory
that the ability to inventory emissions is limited by the  availability of
information with which  to make estimates.

-------
                                       37
                                                                              x«*
For mobile sources,  substantial  literature  review was  necessary as  part  of this
project, but the result~appears  to be  a fairly complete and  seemingly  relatively
reliable inventory,,   For wastewater volatilization,  the inventory appears
relatively complete  (with the notable  exception that emissions out  of  sewers
are not estimated),  though the assumption of 100% volatilization and the
reliance on a handful  of influent  samples leads to questions about  reliability.
For treatment, storage and disposal facilities, the  estimates here,  despite
being the best available estimates, must be considered highly uncertain, due  to
uncertainties in waste composition, waste quantities,  propensities  to  volatilize,
and the insurmountable difficulties of addressing "special waste" landfills and
unauthorized landfills.

Part of the process  of developing  the  inventory was  developing a plan  for
compiling the inventory.  This was a useful endeavor,  since  it helped  in defining
the source area and  the compounds  to inventory, and  helped  in organizing what
source types warranted investigation.   At the same time, it  must be noted  that
much of the character of the inventory development process was a function  of
what information was available in  the  literature.  Thus, for example,  the
inventory includes methylene chloride  emissions from its use in aerosol  cans
and does not include butadiene from oil-fired home heating,  simply  because      ,,
emissions factor data were available for the one category and not for  the
other.  As a result, a major portion of the work in compiling this  inventory
was performing literature review,  exploratory work which could not  be  antici-
pated in the work plan.

Regarding the literature which was found useful, this  report includes  a  complete
bibliography of literature used  in this study.  The VOC Species Data Manual and
the Receptor Model Source Composition  Library have already  been identified as
key data sources for addressing  point  sources.  Most of the  literature used in
the mobile source inventory was  obtained from U.S. EPA's Mobile Source
Characterization Branch of the Atmospheric  Sciences Research Library in  Research
Triangle Park, North Carolina, based on considerable work they have performed
on the subject.* The literature for area source categories  was more diverse,
including literature focussing on  per  capita emissions (based on national
consumer VOC usage), literature  focussing on particular categories such  as wood
stove emissions, as  well as literature from a few broad-based inventory  efforts.
The wastewater volatilization estimates did not rely on any  literature data.

One other series of  documents which were reviewed was  the documents known  as
"locator documents"  (e.g., "Locating and Estimating Air Emissions from Sources
of Formaldehyde").  However, these documents did not lead to very many of  the
emissions estimates  used in this study.  This is partly because most of  the
facility types described in these  documents are not present  in the Southeast
Chicago source area.  This is also partly because many of the emission factors
provided in these documents rely on information that is not  readily available.
For example, the chromium locator  document  provides an estimate of the quantity
of chromium emissions per quantity of  chrome-containing refractory brick,  but
the available information on the brick production does not  differentiate the
various kinds of brick refractories.  Thus, the locator documents, while
providing some information useful  in this  study, more often  seems oriented
toward focussed investigations of  individual sources or source types than
toward general inventories.

-------
                                       38
                                                                              s*
One surprising aspect of^ the process  of completing this inventory was thp effort
associated with the questionnaires.   Numerous companies misplaced the question-
naire,,  For some companies,  plant  personnel  were unable to process the question-
naire, and we had to send another  copy of the .questionnaire to company head-
quarters.  We attached to each questionnaire the NEDS information on the company,
thinking this would simplify their response, since the companies could just
confirm the accuracy of NEDS.  This was a mistake: companies felt obliged to
have accurate information on every detail on NEDS, and often needed information
to decipher which point of their facility corresponded to which point on NEDS.
Finally, substantial follow-up work was necessary, particularly where responses
appeared incomplete or where companies did not know the composition of the
materials they processed.

This report would not be complete  without discussion of needs for further
investigation0  The goal of this project was to obtain a comprehensive inven-
tory of air toxics emissions, so almost by definition any element of the
inventory could be improved through  further investigation.,  Nevertheless, it
is possible to identify specific elements of the inventory for which further
investigation would particularly improve the overall inventory.  Perhaps the
most important need identified in  this study is the need for improvement of
point source species fractions.  Part of an investigation of this issue would
be a more detailed investigation of the cause of the discrepancy found here
between species fraction-based emission estimates and questionnaire responses,
which would presumably help identified means of improving the reliability of
the species fractions.  Even apart from this investigation, it is clear that
significant improvement to the inventory could be obtained by performing a
broader set of species fraction measurements and doing the collateral work of
assessing which operations (e.g.,  hy  SCC) can be characterized with which sets
of species fractions.  A second important need is the development of more area
source emission fractors (e.g., metals from electroplating and butadiene from
home heating) and improvement of existing area source factors (e.g., benzene
fractions for various portions of  the petroleum/gasoline marketing chain and
the composition of coatings used in house painting and auto refinishing),

A third important need is to improve  emissions estimates for polycyclic organic
matter, as emitted by highway vehicles, by wood stove use,'and by other heating
uses.  Relatively few measurements have been made of the relevant emission
factors.  Also, for wood stoves, there are significant uncertainties in wood
usage in the urban setting of Southeast Chicago area, and there are significant
undertainties about the impact of  differences between actual and laboratory
stove operating characteristics on emissions.  In addition, since this is not a
single compound but rather a complex  mixture of compounds, any assessment of
the biological toxicity of the emissions would benefit from a more thorough
understanding of the composition of the emissions.  Fourth, some studies have
indicated that formaldehyde formation from photochemical reactions can cause
greater exposure than direct formaldehyde emissions and in fact can represent a
major component of ambient toxicity.   No effort was made to assess photochemical
formaldehyde formation in Southeast  Chicago.  A need exists for a simple method.

-------
                                       39
                                                                              x-**
using methods such as correlations with monitored ozone or photochemical  box.
modeling, which could approximate photochemical  formaldehyde formation without
requiring long-term formaldehyde monitoring.   Fifth,  a number of improvements
could be made in the inventory of emissions from treatment, storage,  and  disposal
facilities, both in the estimation of waste quantities of particular  compounds
and in the estimation of the degree of volatilization.  Although the  emissions
from these source types are relatively small,  the public interest in  this
general category translates to an interest in  these improvements.

Other needs for further investigation exist in the other source types.  For
mobile sources, the adjustments, particularly  the adjustment to reflect the
possibility that standard measurement techniques may  only measure 77% of  total
exhaust, warrant further investigation.  For wastewater volatilization, further
investigation could address volatilization from sewers, collection of a more
robust set of influent samples, and refinement of the fraction of influent that
volatilizes.

Another set of further investigations represent a broadening of the emission
investigation.  One such broadening is to include emissions of compounds  with
non-cancer health impacts.  Note that the impacts of  such compounds are often  a
function of short-term average concentrations, and so such an emissions inventory
would probably have to address peak short-term average emission rates.  A second
such broadening would be to assess the impact  of potential emissions  controls,
such as the ozone control program, the total  suspended particulate program, or
such strategies as a focussed permit review program.

This report describes the completion of a project to  estimate emissions of a
broad list of potential air toxicants in the Southeast Chicago area.    At the
same time, this report signals the beginning of a modeling assessment of
population exposure to these pollutants.  It is also  hoped that this  report
will facilitate the beginning of other projects in other urban areas  to assess
emissions of potential air toxicants.

-------
                                   RiM iography


 information on Illinois  and  Indiana  emissions/activity

 T.I. National  Fmissions  r»ata System  (NFDS)  - a computer data hasp maintained by
      M.S.  FPA  containing data on  point  sources (including  npprat.ing character-
      istics and emissions of criteria pollutants as obtained hy State AgenciesU

 1. 2. 1QR?  Ozone STP  Rpvisions hy  minois  and Tndiana for  thp Chicago/Northwest
      Indiana Area  -  a  collective  term for  numprous suhmittals most relevantly
      including a vnr. emissions -inventory.   These suhmittals  were the primary
      hasis for area  and  mobile source county emissions totals and for various
      characteristics affecting mobile sourcp emissions. Three specific references
      included  in the SIP are:

 T.3. P.M.  Sellars, A.M.  Kiddie, I. .A. Raci ,  R..1. Rosy, R.n. Green, Oevelopment of
      the Area  Sourcp Fmissions Tnvpntory for the 1QR? Ozonp State Implementation
      Plan  for  thp  Thicago Metropolitan  Rpgi on of Illinois  and Indiana. RCA Corp.,
                 '
 1. 4. "IQfl? State  Implementation  Plan  Suhmittal  for Northeastern Illinois," Chicago
      Area  Transportation  Study,  ,1unp  ?df 1QR7.   fThis is the highway vphiclp
      emissions  inventory  in  Illinois'  1QR? ozone plan for Chicago.)

 T.S. "Mohile Source  ^missions  Inventory for Lake and Porter Counties, Indiana,"
      Rernardin, l.ochmueller  K  Associates,  September 1QR?.

 T.fi. Rridded Illinois  socio-economic  data  - a computer file containing gridded •
      socioeconomic data  (p.g.,  arterial  and freeway VMT and manufacturing
      employment)  for use  in  spatially distributing area and mohile source emissions.

 1. 7. 1QRO  Census  of  Population  - Illinois  (PC(l)-AlR) and Tndiana (PCM)-Alfi) -
      contains maps of  the urbanized arpa and population data.

 i.R. Cpnsus data  - comput.prizpd  population and  dwelling unit data at the hlock
      group level  were  taken  from Rureau of Census data stored on the Graphical
      Exposure Modeling Systems  (GFMS).

 T.I, Wastewater Concentration  data -  data  were  obtained ^rom the Greater Chicago
      Metropolitan Sanitary ni strict,  in letters fPrank nait.nn to Charles Sutfin)
      dated npcemher  17,  iQRfi,  and April  7, 19R7.

T.in. Hazardous  waste handling  data -  extensive  state data on the quantities of
      each  waste code generated,  treated, stored or disposed hy each facility
      were  obtained  from TFoa and TPIFM.  (This report does not document final
      emission estimates  for  this category, however.^

1. 11.. "Fst.1mat.es of M.S.  Wood Fnergy Consumption from 1.Q7Q to IQfll" (np,F/FTA 0341),
      Fnergy Information  Administration, nepartment of Fnergy - provided state
      by state estimates  of wood  usage per  household. (The authors used these
      data  as cited  in  separate references.

T.I?. "Inventory of Health rare  Facilities  and Need net.ermi nation by Planning
      Arpa", Illinois nepartment  of Public  Health - provided data on local hospitals.

-------
Point_ Source References

P.I. Volatile_0rgarn_c_Cornpou_nH_J_VOC_)_^pecj_es_pata_Mamjal, Office of Air Quality
     Planning" and Standards", USEPA". EPA Report * EPA-4"50/4-80-015. July 1980 -
     contains 174 profiles and 134 organic species, including 19 species from
     this study.

P.?. Compiling Air Toxics Emissions Inventories, Office of Air Duality Planning
     and Standards. EPA Report #450/4-86-OOfi, June ] 98fi - contains an index
     of which species profiles to use for which standard Classification Code
     (SCC), and contains a variety of other useful types of information.

P.3. Receptor Model Source Composition Library. Office of Air Quality Planning
     and Standard's", EPA Report # EPA-450/4-85-00?. November 1984 - covers
     roughly 80 SCC's and 30 metals, including 6 metals in this study (arsenic.
     beryllium, cadmium, chromium, mercury and nickel).

P.4. "Locator documents" e.g.. Locating and Estimating Air-Emissions From Sources
     of forma_l_d_e_hy_de_. EPA Report"* EPA-450/4-84-007e, March 1984 - TTeri¥s of
     documents providing emission factor data and, for source types with a modest
     number of facilities, source location.  Such documents are available for:
       a.  Acrylonitrile         (EPA-450/4-84-007a)
       h.  Carbon Tetrachloride  (    "       -007b)
       c.  Chloroform            (    "       -007c)
       d.  Ethylene Dichloride   (    "       -007d)
       e.  Formaldehyde          (    "       -007e)
       f.  Nickel                (    "    -   -007f)
       g.  Chromium              (    "       -007g)
       h.  Manganese             (    "       -007h)
       i.  Phosgene              (    "       -007i ^
       j.  Epichlorohydrin       (    "       -007j)
       k.  Vinylidene chloride   (    "       -007k)

     At time of report preparation, draft "locator documents" were also avail-
     able for chlorohenzenes. ethylene oxide, polychlorinated Mphenyls, and
     polycyclic organic matter.  For a great majority of these pollutants,
     these reports indicated a probable absence of major point sources.  How-
     ever, the Southeast Chicago area does contain sources for which species
     fractions and emission fractions could be derived based on the locator
     documents for chromium, nickel, formaldehyde, and polycyclic organic
     matter.

P.5. Steam Electric Plant Factors, 198?, National Coal Association - provided
     heat content ~data for »ariou"s fuels, e.g., RTU/1000 gallons oil. (This
     study used Commonwealth Edison fuel data).  These data were used to convert
     emissions factors from units of emissions/heat content (e.g., g/Joule) to
     units of emissions/quantity fuel (e.g., g/1000 gallons oil).

P. fi. C o k e_ Oven Emiss i o rvs_ fr om Wet-Coal Ch a r geoMSy -Product C ok e Oven Ra tteries —
     Ba c k g'r ou n d In forma t i o n~ for P ro p o s ed St a n d a rd s, Draft F.fsTOAOPS. "August 1986.

P. 7. Benzene Emi s s i o n s _frpm Cqk_e Ry-Product Recovery P_1_ant_s _--_ Background
     Information for"P'roposed" Standards. Draft Ffs. OAQPS, TPA Report *4"5'n/3-83-01fia.

-------
 Area Source References

 A.I. W.H.  Lamason,  "Technical  Discussion  of Per Capita  Emission  Estimates  for
      Several Area Sources  of  Volatile  Organic  Compounds",  paper  presented  at
      Air Pollution  Control  Association meeting, June 21-26,  1981 -  this  is  the
      fundamental  source for most  of  the per capita  emission  factors.

 A.2. End Use of Solvents Containing  Volatile Organic Compounds,  The Research
      Corporation of Hew England  (for OAQPS, USEPA),  FPA Report f EPA-450/3-79-032,
      May 1979 - provides additional  information for calculating  per capita  emission
      factors.

 A.3  Improvement of the Emissions Inventory for Reactive Organic Gases and  Oxides of
      Nitrogen In the South  Coast  Air Basin. Systems  Applications, Inc..  and Radian
      Corporation (for Calfornia Air  Resources  Board), May  1985 - provided  species
      profiles for gasoline  marketing,  architectural  surface  coating (see especially
      the main text  chapter  on "species profile development"  and  Appendix E).

 A.4. "Locator documents" -  see reference  data  and description urlder point  source
      references.  Specific  documents from which area source  emissions  factors
      were used included the formaldehyde, chloroform, chromium,  nickel and
      draft polycyclic organic matter locator documents.

 A.5. Hazardous Air  Pollutants: Air Exposure and Preliminary  Risk Appraisal  for
      15 U.S. Counties,  Appendix E. "Area  Source Emission Factor  Documentation."
      Versar, Inc. and American Management Systems,  September 1984 - although
      the documentation  is  scant and  the reliability  of  the data  are uncertain,
      this  report provides  useful  species  profiles  for rtegreasing: other  profiles
      here were superseded  by  other references.

 A.6. Steam Electric Plant  Factors -  see reference data  under point  source
      references.  Again, fuel  heat content were used to convert  from emissions
      per Joule to; emissions/quantity fuel burned.

 A.7. Estimates of U.S.  Wood Energy Consumption, 1949-1981, by Applied  Management
      Sciences for Dept. of  Energy -  supplied Illinois wood usage estimates.

 A.8. Residential  Energy Consumption  Survey: Consumption and  Expenditures April
      1982 through March 1983,  Energy Information Agency (Oept. of Energy)  -
      supplied regional  wood usage information  used  to adjust from statewide to
      urban wood usage,

 A.9. R.L.  Gay, W.T. Greene, J.J.  Smith, "A National  Assessment of Residential
      Wood Combustion Air Pollution Impacts," Nero and Associates, March  15,
      1985 - showed  U.S. Dept.  of  Agriculture wood usage estimates.

A.10. Survey of Methylene Chloride Emission Sources,  USEPA  Report #450/3-85-015,
      June 1985 - References A.10, A.11, and A. 12 document  national  emission
      estimates and  procedures  underlying  county estimates  for degreasing emis-
      sions of these compounds.

-------
A.11. Survey of Perchloroethylene  Emission  Sources,  U.S.  EPA  Report  #450/3-85-017,
      June 1985 - see note under A.10.   Also  supplied  similar information  for dry
      cleaning.

A.12. Survey of Trichloroethylene  Emission  Sources,  U.S.  EPA  Report  #450/3-85-021,
      June 1985 - see note on  A.10.

A.13. Chromium Emissions  from  Comfort Cooling Towers - Background  Information for
      Proposed Standards  (Draft  EIS), Office  of  Air  Quality Planning and Standards,
      March 1987 - provided per  capita  emission  factors used  to  derive  a per
      "other" (nonmanufacturing; nonretail) employee emission factor.
                                     j
A.14. Phone conversation  with  OAQPS  staff (Al  Vervaert),  October 30, 1986  -
      provided data for crude  emissions estimates  for  chrome  plating.

 Mobile Source References

 M.I. R.B. Zweidinger, J.E. Sigsby,  Jr., S.B. Tejada,  F.D.  Stump,  D.L.  Dropkin,
      W.D. Ray, and J. W. Duncan,  "Detailed Hydrocarbon and Aldehyde Mobile Source
      Emissions from Roadway Studies,"  Environmental Sciences Research  Laboratory  -
      provided the majority of exhaust  emission  fractions.

 M.2. J.E. Sigsby, Jr., S.B. Tejada, W.D. Ray, J.M.  Lang, and J.M. Duncan, "Volatile
      Organic Compound Emissions from 46 In-use  Passenger Cars," Environmental
      Sciences Research Laboratory - provided evaporative emission fractions,
      and documented the  possibility of unheated sampling trains under-measuring
      exhaust emissions.

 M.3. J.E. Sigsby, Jr., D.L. Dropkin, R.L.  Bradow, J.M. Lang, "Automotive  Emissions
      of Ethylene Dibromide,"  Environmental Sciences Research Laboratory,  presented
      at Passenger Car Meeting,  Troy, Michigan,  June 7-10,  1982  - provided exhaust
      and evaporative emissions  data for ethylene  dibromide,  and documented absence
      of ethylene dichloride.

 M.4. Conversations with  staff of  Office of Mobile Sources staff - provided data
      for estimating exhaust and evaporative  emissions fractions for benezene.

 M.S. User's Guide to MOBILES  (Mobile Source  Emissions Model) -  MOBILES provided
      data on total highway vehicle emissions factors  and the components of the
      total from each vehicle  type.

 M.6. Personal communication from  John  Sigsby to author  (comments on draft report)
      dated January 30, 1987 --  reports that  butadiene has been  found to be between
      5 and 20% of n-butane.

 Uastewater Volatilization References

      Apart from local wastewater  concentration  data,  no  references were used,
      since simplying assumption of 100% volatilization was used.

-------

    Illinois Environmental Protection Agency    2200 Churchill Road, Springfield, IL 62706
        14. 1
 MoasrcJi Uaatrjr Co.
 140 Kast Ilia Street
 Catcaft. llltatls

 fitttloata:

 Tht Illtatts CM aad USEPA art ••!•§•< ta a ctaatrattft tffort ta caaatla aa
 lavtatary of atr aslftsfoaa of salacttd saastaatat fa /Mr araa. At aatatrKad
 by tat  mtatts Cavtroaaaatal Prefect!** Act aad ta* federal  Cloaa Air Act. M
 arc coat111 of tfels ln»«ot»ry s* Out M CM «SMSS antd^U* kMltft favacts
 o* tat  rvsfaims of tlw Santftiast CMcaf* art*.  F«r Qrpts «f laaixat an
 bdM iaelidcd 1a ttff fnforaatlo* fatterlaf tfftrU  1*tf1w1a»a)
 M*afacturliig/>rDCc«sia9 factlUf**, *arM M«rcu* (stcft AS  §a*»1lM *t*tf
 and dry cIcaMrt). '•oftflt sacrccs* 
-------
   Illinois Environmental Protection Agency    2200 Churchill Road. Springfield. IL 62
Paft.2
III* total  3 1s a •vtstlattalro »»1cft Idtttfflts otfttr iaftrattloa Baadad.  I
atttrtl. this ajtastloMuIrt stats atrtsslots astlnstts far each oparatlot U
>ttr facility for oocft of tftt llstod cttjoutds mlcft yoo tajr aarft. and  sat*
atdltional  itforaatiot htlpful it asttMttf tftt possibility of attBSpftarlc
rtltasts.  Plaast fill att a «Mft1otMlrt tar aacft optratlot ar prodtctlot
11M It yotr facility as idottlflod (tr as should bo idtttlflta) It
Attachment  2.

Atucfttttt  4 rtojuasts additional Information at sncttdary air atJlssloas
ratvltltf frat platt opt rat loot Itmlvtitj •tsttvattr. Hajgld vastas.  aad soli
vast**.
       wu that la tJ» afe«€aca of tftt rtajmt**  1«f«mat1o«. aartwl
your facility  «<11 b* tstiMtcd Mi«d MI tat  fOC  a* T» aan»slo« oo protrld* sftMlo: bo Uft
coafldMtU).  plaaso frorlda Jastlflcation ««d looatlfjr tfta lafotvotlaa) yo»
wlcft kapt cMfldontlal at tfto t1«a jra« lotorit It.  IlllMts EFA will folio*
tfto procoduras 91*011 IB tfto Stato of I1l1a0fs Rftlos tod RafMlat1«m. Tltlo -K
Chapter II:  Part  U1 for ftanillof roojMtU far coafldoatlalitjr. «a* WSCFA wl
                                                           Fallont to ass*
                                                         to tfto ptftllc
•Itftoot furtftor attic*.
                           dllaf
follow tfto procrtoros fllvaa 1o 40 Cfl Soct1o« 2.100 at m.
a cUioi «f co«f1de«tl«Mty •okas tfto lofomatloft avaTTaFlo
Pltast svbsrtt tftt  ranoastad lafomatloa by Santanter 30. 1!
tht USCPA's copy to Ntrrlat Crota (SAft.2ft). USCPA,  230 S. Dtarbort. OUcaoo,
IL 60604.  If ytt  Htuld I1kt additional itfomatian rt«ard1tf tftt
quastlotnairt or tftt projact it atntral. ploast call Joan Sftrnck of tf staff
1217/782-1 CM) or  Jtbt SttBtrhays (312yOt*-4047)  or Or. Htrrlnt Crtte
(312/353-UJ09) tf  tftt USCPA Itflot V off lea.   Ut  «otld also bt happy to shar
vita you. at rtnjttst. a dtscrlptlot tf tftt plat for estimating missions of
otfttr sa«ircts and  sovrca typos 1o tftt arta.  Ut art loocltf forward to yntr
coottratlan It this off art.

Sltctrcly.
            • f  y.  '  r~'
-------
                      i protection Agency    2200 Churchill Road, Springfiel
-------
N E 0 5    POINT

STATEdO: ILLINOIS
COUKTKtUJ): CJ?r CO
iOURCc     LISTING

   A')CRC3o7>:  1FTROPOLITAU CNICAuQ
                    PL AM  ID:
                                                               NJOS-USfR FILE C'fAUt ON  THU«S»M
                                                                                                           Itt.
PIRSONAL counter:'
                                                            POINT  ID: 01
                              co  UO 4 nisr siflffi"    sice     >:   KISSING
                                sccc.i-3r-oo6-.rn: f*Tcoi«  NOILE»     -INDUSTRIAL
                                                               -NATURAL
• GCNMAL I.-jFOtNAllO*
• 0*U OF
! LAil JPOAFL: I'SI
!
• O.W . u JM i p: (•,„! L ; j
!fb>3 OC^^C " r '
• r *• J w . ' * •
{ SOUtCf : INV «(..':>
|
•
| N3BP4L CP.:K*TIOSS
{ •••••••«••••«•••«
• HOU4S/OAV: 1C
• wtEKS/ViAF): Si
f
• t tNMUAL THDOViiMPIJT
J »••••»•••*••••••••«
t D5C-F?.-: .'J V
• HAR-rtV: i) X
1 JUNt-AJ'i: !u X
i ! E P I - "0 4 : 101
i
* £ SPACE HIM: ;
f
! COMPLIANCE INfl

! I.OT s»«.cmtt*
f
J^tMOlU^C *!.*»•"« t
Cd^rLNCE J Alt : /
i COH?LIASCt iTAlui
t UPOATc: / /
1
t C«1£?GCNCT CONT-lJL
f ACTIO'4 PL«k>
| ••••«•••••••«•«*•
(MA • ft^^lil Affk
«u i nciuiK.p
t »L«KT co^it-«r •
i POINT CO*"UNT :
! iCC CO^^tfcT :
! UIX G1I» C30«OlN»ftS ! HAND OLCULAUD POINT MISSIONS
i
*
•
i
i
i

i
i
i
i
i
i
i
•
•
•
i
•
•
!
i
i
i
;
•
;
;
*
I
!

i
i
•
i
i





UTJ» /3Sf: lo ! PAailCULAIf:
! S0«:
HQKilOHIlL: «4/.b <^ ! NOX:
VflTICAL: A»e15.I <* ! WCC :



SHC< htlil-lT; l?5 FI ! ••••••••••••••
SIACK b!A*iLTEit: i.V FT ! PA<;i: filSilON
j»i If ^P^^ATUO.'.: t^O F ! &0«: t, 11 S3 ION
CAi FLOJ >AIC: 11, .5o ACF1 ! N^X: S^iCIAL
PL'J^C HT(NO ST«CKl: FT ! VOC: HUSSION
iAIH iTACK V-NTS »flINJS 1?-D?! CO: E1ISSI3N
COST.10L 0:« 1C -/^'.THCO l9tM IF 1C A T ION
»••»••*•*%•*«••«•»•••»»•••••»•««•««•
P«I<««RI P^^T: CENTP:FUUAL COLLJLf Jl CONTTM: l.«! X 40'JKLV

flJ£L JkSrt CONTfll: X ;IOILE<
VW^ ^v V^V «VVVV •• W V V
0 TQ'tSJVR
0 T?'4i/»!|
? TONS/VR
J fONSMF.
Of f\ MC « V A
1 U ~ > w VI*
ATTON KETN3PS
*•»••*•»«»•»»
FACTOMtcOS SCC FILc)
FACIOttNEOS UC FILE)
STATc EMISSION FtCTOR
FACT04(NfOS SCC FILE)
FACTORED; 3tC FILE)
• CONTROL
! tFFICIENCIES
EFFIClEKCf ! •••«•*»•»•••
• PA?T: T1.2 X
•
i
! SOI: I
;
}
! NOI: X
4
•
! VOC: X
«
«
! .'-CO: X
1C ?«TfS
• «••••»•..
kL OP-4ATIN'i MM:
•
MAIN OSSI.N )ATr: .:
* •
1 Of SIGN CA?ACI Tf :

!
i
;
|

•
•
t
i
•
i
J,
! '
!
!
•
4
•
t
•
i
{
1
!
4
4
4
!
4
!



S2

SO

S?
FJiL MEAT CONT.-4I: 1,001 MILLION i TU/'ILLI ON CWU.Ftei <4U«NCO C'




•' !
J



M.LOUABLE (NISSIONl I0« P«t4i "*.
'*PARTlCULM*f. 	 *™ I
soi- i •!**"* *
ICj(j IftZlM* *
«0t: ( 10MIMI <
t0: tomm »
COf-^UTE* CALCO SCC CniSSlOHS \
A*************************** i
PART: <1 IONS/11 •
t
SOI: «l 10NSIVM •
%
NO*: 2 TONS/IR !
|
VOC: <1 10NS«V« \
4
CO: <1 TONS/VR !
4
•
J
4
•
4
4
4
4
1
4
4
4
4
I


f
MILLION cu»u F?ET "juaiio t
i
BILLION CU8IC FEFT 1U9NEO 4
f
PILLION RTU'H* 4
J-MCNTS: V •
i
i
4

-------
   Illinois Environmental Protection Agency    2200 Churchill Road. Springfield, IL 62706


                                ATTACWCMT 3                       "


A.  Mtottf/ to* c*o«tc«1 listed ea Attocfem A *ricfe 1s oso*  or cr*otod ojr
        orocosft/storcgo/tflspoul ooorotloo*.
            wlssfea strMM (stKk. ^roc«ss. f«fU1v«.  st»r«ft tmk.
         pi MM prorl^t tUt f«11ow1iif UfpfMtlM o« tft* s*t«t tf
    for tftt 19ti c«l«no«r jr««r.
    htlffet  frs« irwno.
      «MUr of «1 sitoa potot
    For oac* cDovfcol s«asu>c« ioootlflotf IB port A «*1cft ts In tirts •
    flvt
                    l  MOM                    Uocontrono* falssio* Ut«
                                                       Tooo/Tr
    Mot oir pollotio* cootrol  «rt In ploco.for Uls eorfsstoo ttr*m

    Control                     Cfiovlcal                   IOBBVOI tfffcloncjr
                               (AtUCtBMt 1)

-------
    Illinois Environmental Protection Agency    2200 Churchill Road, Springfield, IL 62706
                           ATTAOMBft 3
       ttt I** C4lft»4«r /«*r 41* /•• Mft aty tMKtw«il*4 rtlMMS
             Qf 4% Of titt S«tatMCM iatrt1fft« 11 Hit A?
for ttcn SMCH str»a» associate nit* aa •Mch«*jrit4 roloatt 11st for u» total
coltntfar ytar
loapmind                                           fpMJtttr «•*«•« 10 air
           1a attactMrit 1)
                    VS

-------
Illinois Environmental Protection Agency   2200 Churchill Road. Springfield. IL 6270
         mission OF POTHTUUT raxc SIBSTAMCCS

Socoafer? oaltstoaa «rt taaaa air oatfsafaaa rtsaltlag froa taa trtataant
•r ilsaasal of vastaaatar  11«i14 «MtM «r WH4 vaftUs.  F*r t«cJ»
 •r ilsaasal of vastaaatar,
 prvctss Mitt rtnMB «to1cli cMttlw tn» or aon tMlc
            fey f 1111 •§ <• tl» Uftlc toltm  If uapll* 4au an
                   • test ntlMtt f«r tacfc M
      •oat or *1saa*al aattood (if aartloaa of taa stroaa ara
 traataavaiaposod of 41fforaatljr aaacrtaa aataats aaaorataly far oaca taalc
 T«U1 «Dlo« of
 Total v»1«» of )1ojHid wosto
 Tatal voMaw of offlaaat 41scaaraaa to sawr tystaa           aal par da/.


           Saaataaca        Total      % af Sakstaaea   Ptaar Traataaat/
                 itloa    Svaataaca   Wacaaraod to      Bfaaaaal of
            (l

-------