October 1974
          PREFERRED
STANDARDS  PATH  REPORT
      FOR POUYCYLIC
     ORGANIC  MATTER
  U.S. ENVIRONMENTAL PROTECTION AGENCY
 Office of Air Quality Planning and Standards
    Strategies and Air .Standards Division
       Durham, North Carolina

-------
          PREFERRED
STANDARDS  PATH REPORT
      FOR POLYCYLIC
     ORGANIC  MATTER
       U.S. ENVIRONMENTAL PROTECTION AGENCY
     OFFICE OF AIR QUALITY PLANNING AND STANDARDS
       STRATEGIES AND AIR STANDARDS DIVISION
          DURHAM, NORTH CAROLINA

             October 1974

-------
                              CONTENTS
                                                                   Page

LIST OF FIGURES	    v

LIST OF TABLES	    vi

EXECUTIVE SUMMARY 	    vii

1. INTRODUCTION 	    1

   1.1  Background	    2
   1.2  Scope of Problem	    4

2. RATIONALE AND RECOMMENDATIONS FOR EPA STRATEGY FOR BaP .  .  .  .    7 '•'

   2.1  Criteria Under the Clean Air Amendments of 1970	    7

        2.1.1  National Emission Standards for Hazardous
               Air Pollutants	    9
        2.1.2  National Ambient Air Quality Standards 	   10
        2.1.3  Standards of Performance for New Stationary
               Sources	   11
        2.1.4  Emission Standards for Moving Sources  	   12

               a.  Motor Vehicle Emission Standards 	   13
               b.  Regulation of Fuels	13
               c.  Aircraft Emission Standards  	   14

        2.1.5  Combination of Options	   15
        2.1.6  Total Ban	16

   2.2  The Preferred Standards Path	   17

        2.2.1  NESHAP	17
        2.2.2  NAAQS	18
        2.2.3  ESFMS	19
        2.2.4  NSPS	19

               2.2.4.1  Alternatives to Control Under Section 111   21

        2.2.5  Alternative Control Strategy 	   21
        2.2.6  Advisability of Federal Control for BaP	24

   2.3  Recommendations	'	25

3. SOURCES OF BaP	27

   3.1  Stationary	27

        3.1.1  Heat and Power Generation	28
        3.1.2  Refuse Burning	29
        3.1.3  Industrial Activity  	   31

   3.2  Mobile	33
                                 iii

-------
                                                                   Page
        3.2.1  Gasoline-powered Vehicles 	   33
        3.2.2  Diesel-fuel-powered Vehicles  	   34
   3.3  Natural	   36

4. CONTROL TECHNOLOGY  	   37
   4.1  Stationary	   37
   4.2  Mobile	   41
   4.3  Natural	   43

5. AMBIENT CONCENTRATIONS  	   45
   5.1  Measurement Technique  	   45
   5.2  Ambient BaP Concentrations 	   47
   5.3  Trends Analysis for BaP	   49
   5.4  National Urban and Nonurban Trends for BSO	   51

6. EFFECTS OF BaP	   55

   6.1  Epidemiological Studies  	   55
        6.1.1  Classification of Studies 	   56
        6.1.2  Consideration of Other Variables  	   57
        6.1.3  Conclusions from Epidemiological Studies  ....   58
   6.2  Other Effects	   59

7. PROGRAMS TO STUDY BaP (AND OTHER POM)	   61

   7.1  Current Research Projects  	   62
        7.1.1  Analysis Methodology  . . .	   62
        7.1.2  Effects Investigation 	   62
        7.1.3  Stationary Sources  	   64
        7.1.4  Ambient Analysis  	   64
   7.2  Recommended Research Projects  	   64

8. REFERENCES	   67

APPENDICES	   71
   Appendix A - Overview of a Preferred Standards Path	   A-l
   Appendix B - Lung Cancer Mortality in Selected SMSAs  ....   B-l
   Appendix C - National Air Surveillance Network Ambient
                Air Measurements for Benzo(a)pyrene  	   C-l
   Appendix D - Description of By-product Coke Production  ...   D-l
   Appendix E - Costs of Control and Growth Patterns for
                Coke Ovens	   E-l
   Appendix F - Emission Estimates for Coke Ovens	   F-l
                                  iv

-------
                          LIST OF fIGURES

Figure                                                            Page

  1    Typical Structural Formula of POM,  Benzo(a)pyrene  ....    3

  2    Typical Structural Formula of an Aza-arene,
       Dibe.nzo(c,g)carbazole	    4

  3    Annual POM Concentrations  (10 Compounds)  Versus Annual
       BaP Concentrations in Birmingham, Ala.,  1964 - 1965  ...    6

  4    Preferred Standards Path:  Guide for Determination
       of Regulatory Action 	    8

  5    Trends in Annual BaP Concentrations in Cities with and
       without Coke Ovens	48

  6    Trends in BSO and in BSO Percentage of TSP at 32 Urban
       and 19 Nonurban Stations	52

 A-l   Preferred Standards Path:  Guide for Determination
       of Regulatory Action 	   A-4

 D-l   Koppers-Becker Underjet Low-differential  Combination
       Coke Oven with Waste-gas Recirculation	D-2

-------
                           LIST OF TABLES
Table                                                            Page

   I   Major Sources of Estimated BaP Emissions in the United
       States by Descending Quantity of Contribution (1971-73) .   20

  II   Estimated Benzo(a)pyrene Emissions from Heat and Power
       Generation Sources in the U.S. (1972)	29

 III   Estimated Benzo(a)pyrene Emissions from Refuse Burning
       in the United States (1968)	30

  IV   Summary of Estimated Industrial Benzo(a)pyrene
       Emissions in the United States (1972) 	   32

   V   Estimated Vehicular Benzo(a)pyrene Emissions
       in the United States (1970)	. .   35

  VI   Estimated Benzo(a)pyrene Emissions from Catalytic
       Cracking Sources in the United States (1968)	39

 VII   Contributions to National Totals of Benzo(a)pyrene
       by Source and State (1972)	42

VIII   Automotive Benzo(a)pyrene Emission Factors  (1972) ....   A3

  IX   Three Year Summary of TSP, BSD, and BaP for 121 NASN
       Sites in the United States -  1968, 1969, and 1970 ....   47

   X   Applicable Current POM Research Projects	63

 B-l   White Male Deaths and Death Rates per 100,000 Population
       per Year by SMSA for Malignant Neoplasm of Trachea,
       Bronchus, and Lung (1959-1961)	B-2

 C-l   National Air Surveillance Network Ambient Air
       Measurements for Benzo(a)pyrene:  121 Selected Sites
       (1968-1970)	C-2

 C-2   Listing of 40 NASN Sites Selected for 1971-72
       BaP Analysis	C-7

 C-3   Annual BaP Averages for Selected Cities (1966-1972)  .  . .  C-7

 E-l   Capital Costs (in $1,000)	E-3

 E-2   Annual Costs (in $1,000)	E-4

 E-3   Summary of Projected New Oven Construction  (U. S.).  .  . .  E-6

                                   vi

-------
                        EXECUTIVE SUMMARY




    Polycyclic Organic Matter (POM) is  an aromatic  hydrocarbon  group




that includes certain compounds which are proven carcinogens  at ele-




vated levels in laboratory animals.  Some of  these  compounds  have been




linked with the occurrence of cancer in humans,  but none have been




directly related to cancer from exposures to  ambient air.   Nevertheless,




it seemed prudent to consider POM a potential air pollution problem and




to evaluate the need for additional formal regulatory action.  The




objective of this report is to assess the POM problem,  determine the




requirement for additional regulatory action  under  the  Clean  Air Act,




and identify the most appropriate regulatory  tool if such  action




is necessary.





      POM  includes  thousands of compounds which vary widely in physical




and  chemical characteristics and in  their capacity to act as  carcinogens.




Data  on total  POM  are extremely limited.  Most source emission estimates




are  for a  single POM compound, benzo(a)pyrene (BaP), and not  for total




POM.   BaP  is a proven carcinogen commonly used as a surrogate for total




POM.   No  analytical techniques exist for ambient air measurements of




total POM.   The majority  of these  measurements are for BaP only.




Limited dose-response data that have been assembled also are  for BaP




and  not total  POM.




      Sources of BaP normally are associated with the incomplete combustion




of  organic compounds, especially coal.   The nationwide emissions of BaP




.are  estimated  at about  900 tons per  year based on  1971-1973 data.



                                   vii

-------
Major source categories and their estimated contribution to this total

are listed below.


                                                  BaP Emissions  % of
          Source Type                              (tons/year)   Total

                                                   (1971-1973)

1.   Coal Refuse fires                                 310       34.7

2.   Residential furnaces, coal (hand-stoked)          300       33.6

3.   Coke production                                   170*      19.0

4.   Vehicle disposal (open burning)                    25        2.8

5.   Wood burning (fireplaces, etc.)                    25        2.8

6.   Mobile sources, gasoline                           11        1.2

7.   Forest and Agricultural burning                    11        1.2

8.   Tire degradation                                   11        1.2

9.   Open burning (domestic—municipal)                 10        1.1

10   Intermediate coal furnaces                          7        0.8

11.  Petroleum refineries                                7        0.8

12.  Enclosed incineration (apartment—municipal)        3        0.3

13.  Other                                             	4_       0.5

                                        TOTALS         894      100.0
*A range of emissions has been given for this value.  Appendix F

contains the details.
                                 viii

-------
     As illustrated by this list, stationary sources account for about


97% of the nationwide emissions of BaP.  The inefficient combustion of


coal is the largest single source of BaP.  Thus the three largest


emission sources are coal refuse fires, residential coal furnaces, and


coke ovens.  Together, they comprise more than 85% of national emissions


of BaP.


     Annual concentrations of BaP range from 0.07 to abour 17 nanograms

                                    3
per cubic meter of air sampled (ng/m ).  Urban levels are higher than non-


urban levels by as much as 10 times under some conditions.  Annual


ambient air concentrations of BaP have declined significantly over the


past few years.  In order to assess these declines, BaP trends data


were compiled for selected cities from 1966 through 1972.  Cities


containing large coke oven facilities were singled out as a separate


category since coke ovens are a major source of POM and were considered


for special regulatory action.  In addition, trends in the benzene


soluble organic fraction  (BSO) of total suspended particulates were


analyzed for urban and non-urban sites since data for this fraction


are more complete than data for BaP.  Both BaP and POM are found in


the benzene soluble portion of total particulates.


     Trends data for BaP  (shown in the following figure) indicate a


significant decline in ambient concentrations.  From 1966 to 1972,


concentrations have decreased about 55% in coke oven cities and 77%


in non-coke oven cities.  For all cities sampled, the urban composite


average BSO concentrations (shown in the second figure) have decreased
                                  ix

-------
      5
I    3
o
o
O-
ra
CD
            1966
1967
1970
1971
                          1968          1969


                                  TIME, year


Trends in annual BaP concentrations in cities with and without coke ovens.
1972
     12
     10
cc
o


o
o

o
KQ
CO
                                Till
                                                           32 URBAN STATIONS
  (ONLY 18 NONURBAN STATIONS

  INCLUDED IN THIS COMPOSITE

  AVERAGE)
                                                              19 NONURBAN STATIONS
      0

     1960     1961      1962     1963     1964     1965      1966     1967     1968     1969      1970


                                              TIME, year



         Trends in  BSD and  in BSO percentage of TSP at 32 urban and 19 nonurban stations.

-------
              3                    3
from 10.6 ug/m  in 1960 to 4.8 ug/m  in 1970, or a 55% decrease.



These decreasing trends are significant and can be attributed to  the



success of existing control programs as well as the move awav from



the use of coal in small, inefficient furnaces.



     Coke ovens are a special problem.  Based on available data,



they are suspected of being a major source of BaP and are difficult



to control.  Demonstration projects sponsored by the Environmental



Protection Agency and industry groups indicate that particulate



emissions from coke ovens can be controlled by a combination of



process alteration and equipment design modifications.  Preliminary



results show that application of good particulate emissions control



will also reduce BaP emissions by 85-90%.  Federal emission standards



for particulates being developed for new coke ovens, and enforcement



actions underway against existing ovens, ensure adequate regulation



of this remaining significant source of BaP.



Existing Regulatory Control Program



     Coal refuse fires, residential coal-fired furnaces, and coke ovens



(the three major source categories of POM emissions) are being con-



trolled, or are scheduled for control, through the following actions:



     1.   Coal refuse fires are regulated in the four States which



include the majority of coal refuse fires.  The nine other States



with coal refuse fires can control emissions from this source by open



burning regulations or other regulations.  In addition, the Bureau of



Mines, Department of Interior, has proposed Federal regulatory action



for this source.  They presently require control of spontaneously



started fires under existing Federal regulations.




                                   xi

-------
     2.   Emissions from coal-fired residential sources decreased 28%




between 1968 and 1972 and this trend is expected to continue for the




following reasons:




          a)   Socio-economic conditions have established a trend




               away from coal-burning in the home because of its dirtiness




               and inconvenience.




          b)   All states except one are imposing particulate emission




               limits on coal-fired residential furnaces.  In many




               cases, these limits, by requiring particulate control




               systems, discourage the continued use of coal for res-




               idential heating.




          c)   Residential coal-fired furnaces are banned directly in




               some urban areas such as Chicago, St. Louis, and Milwaukee.




               Others have indirect bans through stringent sulfur regu-




               lations which preclude the use of most, if not all,




               sources of coal.




     3.   New coke ovens will be controlled bv national particulate




emission standards being developed for this source.  Control equipment




has already been installed on several existing coke ovens as a result




of local regulations and Federal enforcement actions.  EPA has fourteen




enforcement actions against coke ovens for particulate controls.  Pre-




liminary data from EPA demonstration projects indicate that about 90%




control can be achieved for particulate emissions during charging




operations, which historically has been responsible for most emissions




from this source.  These same data show an equivalent degree of BaP




emission control can be expected.





                                  xii

-------
     Most of the remaining sources for BaP are small,  numerous and widely


dispersed.   Several of these sources are not possible  to control effectively


(e.g., home fireplaces, forest and agricultural burning).   Others


(petroleum refineries, vehicle disposal, and open burning)  are being


regulated through New Source Performance Standards and State Implementation


Plans.  Although some concern has been expressed regarding  increased


use of coal for electrical power generation under our  current national


energy policy, such increased use is not anticipated to cause increased


emissions of POM because power plants utilize very efficient combustion


systems and emit essentially no POM.


Conclusions and Recommendations


     The health effects of ambient air concentrations  of POM or BaP


are not well documented and information is not available for selecting


harmful air concentrations.  Although the presence or  magnitude of the


threat from POM cannot be determined at this time, selected polvnuclear


organic materials have been shown to be carcinogenic in some situations.


Therefore, it seems prudent to minimize emissions of POM and to use


all practical means to reduce levels of POM in the air.

                                         . i
     Ambient concentrations of BaP and benzene soluble organics have


declined significantly over the period for which trends data are


available.  From 1966 to 1972, BaP concentrations have decreased 55%


in coke oven cities and 77% in non-coke oven cities.  This  decline in


ambient levels is attributed to current particulate control programs


and the phase-out of coal-fired home furnaces.  Although these actions


have not been oriented specifically to reducing BaP emissions, they do


have a direct effect on those sources responsible for most  BaP emissions



                                  xiii

-------
and are an effective national control strategy.   An analysis of additional




regulatory steps does not suggest other effective measures.




     Control regulations specifically for BaP or POM are not warranted




nor practical at this time.  Ambient monitoring for BaP should be under-




taken in selected cities to ensure the continued adequacy of the control




program.  Also, special emphasis should be given to the prompt enforce-




ment of existing regulations for incinerators, open burning, coal




combustion, burning refuse piles, and coking operations.
                                  xiv

-------
                 PREFERRED
    STANDARDS  PATH  REPORT
             FOR  POLYCYLIC
          ORGANIC  MATTER
                   1. INTRODUCTION



   Polycyclic organic matter (POM) and benzo(a)pyrene (BaP) were two

of several pollutant groups for which control strategies were to be

considered as listed under National Objectives in the Administrator's

"Guidelines for Development of Fiscal Years 1973-1978 Program Plans."

Since BaP is a specific POM compound and is generally used as a

surrogate for POM the two tasks were combined.

   The procedure used in preparing this report and arriving at

recommendations included:

   (1) studying control options available under the 1970 Clean Air

      Act, an internal EPA procedure known as a preferred standards

      path analysis.

   (2) reviewing and evaluating comprehensive studies available on

      POM, including:


                          1

-------
                                                     3*
         (a)  "Particulate Polycyclic Organic matter"   by the National



              Academy o,f Sciences, 1972.



         (b)  "Draft NERC/RTP Position Paper on Particulate Polycyclic



              Organic Material (PPOM)"   by PPOM Task Force Panel, U.S.



              EPA, NERC/RTP, 1972.



         (c)  "Preliminary Air Pollution Survey of Organic



              Carcinogens"   by Litton Systems, Inc., for HEW, 1969.


                                                                     8
         (d)  "Sources of Polynuclear Hydrocarbons in the Atmosphere"



              by R.P. Hagebrauck, e_t ^1., AP-33, 1967.



    (3)  evaluating available ambient air data.



         After thorough evaluation of this information, the following



         abridgement is presented.






1.1  Background






    Polycyclic organic matter (POM) is organic matter with a



multiple ring (two or more) structure, as shown in Figure 1.  A



typical structure of five rings is shown in Figure 1 with



the dashed circle representing the carbon-carbon (C-C) bond and the



numbers around the outside representing points where, in this case,



hydrogen atoms are attached.  Other atoms (or groups of atoms) may be



substituted for carbon at the numbered points.  When such a



substitution occurs, polycyclic azaheterocyclic compounds (aza-arenes)



are produced.  Other names which are given to POM include (1)




    *  Numbers shown  as  superscripts refer to  references at  the end  of

the Report.

-------
                                           5
                   Figure 1. Typical structural formula of  POM,
                   benzo(a)pyrene .
polynuclear aromatic hydrocarbons  (PNAH,  PAH);  (2)  polynuclear

aromatics  (PNA);  (3) polycyclic  aromatic  hydrocarbons  (PAH);  or

simply,  (4) aromatic ring  compounds.   Generally,  when  any of  these

names, or variations thereof,  are  used the broad  class of POM is being

referenced.

    The  primary reason  for interest  in POM is  that  some of these

compounds have been proved to  be carcinogens in experimental  animals.

They may also act  as cocarcinogens,  or be active  in the presence of

other cocarcinogens.     (A carcinogen is  a substance or agent which

produces or incites cancer.  Agents  which induce  altered physiological

states that may increase the risk  to carcinogenesis are

.cocarcinogens).   Furthermore,  epidemiological evidence connects lung

cancer etiology with occupational  exposure to certain particulate POM

(PPOM) compounds  in specific industries.   A few POM compounds which

have been  identified as definite carcinogens are  :

-------
                         7,12-Dimethylbenz(a)anthracene
                         Dibenz(a,h)anthracene
                         3-Methylcholanthrene
                         Benzo(c)phenanthrene
                         Benzo(a)pyrene
                         Dibenzo(a,h)pyrene
                         Dibenzo(c,g)carbazole

The last compound  in the list is  an example of a carcinogenic aza-
arene and  is  illustrated in Figure  2.
                   Figure 2. Typical structural formula of an
                   aza-arene, dfbenzo(c,g)carbazole.
1.2  Scope of Problem

    Under the Clean  Air  Amendments of 1970 the EPA was provided with
alternative approaches to control strategies for candidate pollutants.
The process of looking at these  alternatives is known as a preferred
standards path analysis  (PSP), which is explained in greater detail in
Appendix A.   The PSP analysis  in this report uses BaP as an Indicator
for PPOM.  Reasons  for  limiting this analysis to BaP as an index for
PPOM are:

-------
    a.   Existing data support using BaP as a rough indicator for PPOM




         (Figure 3 and Section 5).   If effective control is achieved for




         BaP other POM compounds should also be controlled to a large




         degree because their characteristics are similar to BaP.
    b.   .Lack of available data for POM:_  _




         1.    Emissions estimates are for BaP and not for POM; some




              scientists have limited data on total POM.
         2.   Majority of ambient air measurements are for BaP only;




              selected research studies have produced limited ambient




              data for POM.




         3.   Limited dose-response information does exist for BaP but




              none is available for total POM.




    c.    BaP has received primary emphasis in measurement techniques




         development for specific POM compounds;  none are available




         for total POM.




    Thus, this report presents the preferred standards path for BaP as




a surrogate preferred standards path for POM.  The NAS report, the




ORM/NERC/RTP position paper on PPOM, extensive personal contacts, and




other literature have been used for support^ d?~the^ scenario contained




in this report.

-------
e
oi
O
I-
ce
o
o
o
a.
   180
   160
   140
   120
   100
    80
    60
    40
    20
                       10
20
30
40
                             BaP CONCENTRATION, ng/m3
Figure 3.  Annual POM concentrations (10 compounds) versus annual BaP concentrations
in Birmingham, Ala., 1964-1965.

-------
             2.   RATIONALE AND  RECOMMENDATIONS
                   FOR EPA STRATEGY FOR  BaP
2.1  Criteria under the Clean Air Amendments of 1970
     Criteria available to the Environmental Protection  Agency  (EPA) for
recommending a preferred standards path for any pollutant  from  either a
stationary or moving  source, as described in the Clean Air Amendments of
1970 (Act),  include the following factors:   (1) Presence and magnitude
of health and/or  welfare effects of a pollutant; (2)  Nature and distri-
bution of pollutant sources; and (3) Supporting data  (implied).  Those
sections of the Act in which these factors are included, along  with the
six basic standard setting options, are described and then discussed
below.  The characteristics of a candidate pollutant  are compared with
criteria for each option given (Figure A).   This chart should not be con-
sidered a strict  decision-making tool but merely a guide that provides one
of many inputs into the selection process.   Analysis  must  begin with an
estimation of health  effects since this is the common thread among all
six options.  (See block 1 of Fig. 4.)  Once the determination  is made
that a candidate  pollutant may contribute to adverse  health and/or
welfare, one or more  of the options for control should be  chosen.  The
approach to a control strategy requires different procedures depending
upon the chosen option.

-------
oo
1
Adveise Health and
Welfare Effects: or
Possibility of En-
dangerment of Health
and Welfare
3 Y"

Magnitude of Health and Welfare
Effects: Contribute to Increase
in Mortality, or Serious Irrever-
sible or Incapacitating Reversi-
ble Illness
No
7
Nature and C
of Sources:
ness - Wheth
or Diverse, '
Moving, or B
No


istribution
Ubiquitous-
er Numerous
tationary or
oth

No
4 5
< UnTCi Adequate Data to Nature and Distri- No
See NOTE 1 Show Hazardous El- 	 IiS 	 ». butjon ., sources. "° »
Yes (ects at Current Am' Ubiquitous?
bient Concentrations
No Optional,
8
Is the Pollutant
Moving Sources? lft 12
u Can Tontrnl Rp Adequate Data
_ „ N° SiSsr INn . £»** YK .
Yes * No Ac ion SS" "* ." * .Y" = =
	 *" ° Under ESFMS "'• Q3n Control Be and Welfare


Yes Regulation^ T7
9 1 	 ^ 	 1 [No

Is the Pollutant r *
* Prominent From No L.
Stationary Sources? 1& f ^ St-c NntP 7 F
Yes to Establish and Support yes
for Health and Welfare
Effects?
No N°, Consider

ACTION
2
No Action
Under Clean
Air Act
6
Section 1 12, NESHAP"
Most Likely Appropriate1
13
Section 211, ROF,"
Most Likely Appropri-
ate*
15
Sections 202 and 231,
ESFMS,** Most Likely
Appropriate*
7
Section 109, NAAQS,**
Most Likely Appropri-
ate*
18
Section 111, NSPS,~
Most Likely Appro-
priate*
                       NOTE 1:  A yes answer here simply means the pollutant is a candidate for regulatory action; however, it does not denote mandatory action.
                       NOTE 2:  The dashed line indicates that significant pollution from stationary sources may remain even after utilizing ESFMS. Land use and transportation
                                controls may need to be implemented in addition to Section 202.
                       *MOST LIKELY APPROPRIATE simply means that the indicated option  may be the most logical without considering external factors. However, other ramifications
                        may preclude its use. Also, more than one option (e.g., Sections 109 and 202; Sections 109 and 111; etc.) may be used in combination for any given pollutant in
                        order to achieve the full intent of the Act. (Also see NOTES.) Section lll(d) is applicable to non-criteria pollutants.
                      "Acronyms   «iplamed  in  text.
                                        Figure 4.  Preferred  standards  path: Guide for determination  of regulatory action.

-------
          2.1.1     National Emission Standards for Hazardous Air Pollutants




                    (NESHAP - Section 112)




     Under the Act the Administrator is allowed to use his judgment to




determine whether a pollutant is hazardous, i.e. "may cause, or contri-




bute to, an increase in mortality or an increase in serious irreversible,




or incapacitating reversible, illness" [Section 112(a)(l)]. Within 180




days after publishing a list of suspected hazardous pollutants the




Administrator must publish proposed regulations setting emission standards




together with a notice of public hearings.   "Not later than 180 days




after such publication, the Administrator shall prescribe an emission




standard for such pollutant, unless he finds, on the basis of informa-




tion presented at such hearings, that such pollutant clearlv is not a




hazardous air pollutant, [then] the Administrator shall establish any




such standard at the level which in his judgment provides an ample




margin of safety to protect the public health from such hazardous air




pollutant" [Section 112(b)(1)(B)].




     A major consideration in making a preferred standards path choice




is to assess the largest magnitude of adverse health effects, i.e.,




possible mortality or serious illness.  This requires evaluating the




candidate pollutant for the hazardous option (Section 112) initially.




Here, the effect of atmospheric emissions on health must be analyzed in




relation to  (a) current and expected ambient levels (concentrations) and




(b) what an ample margin of safety is.  In practice, this relationship




is affected by several factors, including terrain, number of sources,




extent of buildup or persistence in the environment, etc.  Setting a




national emission standard despite these variables may logically require

-------
a level of emissions to be either zero or based upon the worst signifi-




cant situation observed or anticipated.  If in applying these criteria,




specific measured or expected ambient concentrations of the candidate




pollutant are shown to be hazardous, the hazardous pollutant option mav




be the choice.  If so, a national emission standard which provides an




ample margin of safety must be proposed and promulgated.  The standard




will apply to stationary sources, new and existing.  The Act does not




specifically reciuire that the Administrator have information on control




technology to initiate action under this option.  However, from a practical




point of view, since the Administrator must oversee implementation and




enforcement of emission standards knowledge of control technologv is




required.




          2.1.2     National Ambient Air Quality Standards (NAAQS -




                    Sections 108-110)




     The Act requires the promulgation of primarv "ambient air quality




standards the attainment and maintenance of which in the judgment of the




administrator, .  . . allowing an adequate margin of safety, are req-




uisite to protect the public health" [Section 109(b)(l)].  Similarly,




secondary NAAQS are required to "protect the public welfare from any




known or anticipated adverse effects" [Section 109(b)(2)].  Ambient air




quality standards are based upon criteria which delineate "all identi-




fiable effects on public health or welfare" from a pollutant whose




"presence ... in the ambient air results from numerous or diverse




mobile or stationary sources" [Section 108(a)]. The Act further requires




each State to "adopt and submit to the Administrator ... a plan which




provides for implementation, maintenance, and enforcement of such . . .





                                  10

-------
standard in each air quality control region .  .  .  within such State" [Sec-


tion 110(a)(l)].


     In controlling pollutants under the ambient option the effect of


existing ambient concentrations on health and  welfare must first be


analyzed.  Such data must be published in a criteria document simultane-


ously with a proposed national standard for a  specific ambient concentra-


tion which can be supported.  Then the States  are left to establish the


relationship between ambient concentrations and emission levels from


sources.  This relationship is affected by such factors as terrain,


number of sources, and effect of buildup or persistence of the candidate


pollutant in the environment.  States are responsible for prescribing


and enforcing emission standards, procedures for control of number or


location of sources, etc.  However, the Administrator must issue control


techniques information simultaneously with criteria documents [Section


108(b)(l)].  Also from the practical point of  view, development of im-

                                                t
plementation plans by States and the Administrator's review and approval


of such plans require some knowledge of control technology.


          2.1.3     Standards of Performance for New Stationary Sources


                    (NSPS - Section 111)


     The Act specifies that the Administrator  include a category of sources


on a proposed list for standards of performance "if he determines it may


contribute significantlv to air pollution which causes or contributes


to the endangerment of public health and welfare" [Section lll(b)(1)(A)].


Further, within 120 days "the Administrator shall propose regulations,


establishing Federal standards of performance  for new sources within


such category" and ". .  . promulgate within 90 days"  [Section lll(b)(1)(B)].



                                   11

-------
The Act further requires the Administrator to "prescribe regulations .  .  .




under which each State shall submit ... a plan which (A) establishes




emission standards for any existing source for any air pollutant (i) for




which air quality criteria have not been issued or which is not included




on a list published under Section 108(a) or 112(b)(l)(A) but (ii) to




which a standard of performance under subsection (b) would apply if such




existing source were a new source, and (B) provides for the implementation




and enforcement of such emission standards" [Section lll(d)(l)].




     When a limited number of source categories or limited number of




predominant categories exist, the section on standards of performance




for new sources may be the most practical approach to controlling a




candidate pollutant.  If Section 111 is used, the effect of atmospheric




emissions of the candidate pollutant on health and welfare must be




analyzed first.  However, the Act does not require calculating a re-




lationship between ambient concentrations and emissions since the




standard will reflect 'the best demonstrated system of emission reduction




for the affected new source  (taking cost into account).  Thus, the level




of control is not related directly to health and welfare effects of the




candidate pollutant.  Analysis for health and welfare effects do or




could exist.  For unmodified existing sources, States must enact emission




standards using available practical approaches, subject to EPA review




and approval [Section lll(d)].




          2.1.4     Emission Standards for Moving Sources  (ESFMS - Title II)




     Title II of the Act, Emission Standards for Moving Sources  (ESFMS),




provides for control of air pollutants from moving sources. This title




presents criteria for motor vehicles, fuel regulations, and aircraft.





                                   12

-------
a.   Motor Vehicle Emission Standards (MVES - Section 202)




          Under this section of the Act "the Administrator shall by




     regulation prescribe . .  . standards applicable to the emission




     of any air pollutant from any class or classes of new motor




     vehicles or new motor vehicle engines, which in his judgment




     causes or contributes to, or is likely to cause or contribute




     to, air pollution which endangers the public health or welfare.



     Such standards shall be applicable to such vehicles and engines




     for their useful life" [Section 202(a)(l)].  The Act further



     requires the Administrator to allow time for "development and




     application of the requisite technology, giving appropriate



     consideration to the cost of compliance within such period"



     [Section 202(a)(2)].  Subsection (b) of Section 202 presents



     exceptions which apply to standards and time requirements for



     hydrocarbons (HC), carbon monoxide (CO), and nitrogen oxides




     (NO ).
        A.


b.   Regulation of Fuels (ROF - Section 211)



          "The Administrator may, ... by regulation, control or



     prohibit the manufacture, ... or sale of any fuel or fuel



     additive for use in a motor vehicle or motor vehicle engine



     (A) if any emission products of such fuel or fuel additive



     will endanger the public health or welfare" [Section 211(c)




     (1)(A)].  The Act further provides that control under the



     subsection just quoted is prohibited "except after considera-



     tion of all relevant medical and scientific evidence available



     to him, including consideration of other technologically or




                             13

-------
          economically  feasible  means  of  achieving  emission  standards




          under  section 202"  [Section  211(c)(2)(A)].  Furthermore,  in




          connection with  registration of fuels  and  additives  the  Act




          provides means whereby "the  Administrator  may  also require  the




          manufacturer  of  any fuel or  fuel additive  - (A)  to conduct




          tests  to determine  potential health  effects of such  fuel or




          additive  (including, but not limited to,  carcinogenic,  ...




          effects),"  [Section 211(b)(2)(A)].




     c.    Aircraft Emission Standards  (AES - Section 231)




              The Administrator shall issue,  based  on studies




          conducted by  him, "proposed  emission standards applicable




          to emissions  of  any air pollutant from anv class or




          classes of  aircraft or aircraft engines which  in his




          judgment cause or contribute to or are likely  to cause  or




          contribute  to air pollution  which endangers public health




          or welfare"  [Section 231(a)(2)].  Provision is made  for




          final  regulations to be issued  within 90  days  after  the




          proposal following  public hearings within  the  most af-




          fected air  quality  control regions  (AQCR)  [Section 231




          (a)(3)].  However,  the regulation shall not take effect




          until  necessary  time has elapsed to  develop and apply




          requisite  technology and consultation with the Secretary




          of Transportation [Section 231(b) and (c)].




     In considering  the nature and distribution of sources of  the




candidate pollutant,  one of the  first  questions is  whether emissions  are




from moving sources.   If so,  then additional questions must  be asked





                                   14

-------
(block 8 of Figure 4) because at least two options become available.




However, regulation of fuels (Section 211) can be employed only after




considering all possibilities of control by emission standards under




Section 202, as well as relevant medical and scientific evidence.   If




emission standards are not, or can not be, effective without fuel  re-




gulation, then this option may be used also.




     Emissions from aircraft must also be considered in connection with




moving sources.  If such emissions have a significant impact on health




and welfare effects the Administrator, in consultation with the Secre-




tary of Transportation, is authorized to set standards.  Enforcement  of




these standards is the responsibility of the Secretary of Transportation,




after consultation with the Administrator [Section 232(a)].  Standards




can be set and enforced only after allowing what is, in the Administra-




tor's  [Section 232(a)].  Standards can be set and enforced only after




allowing what is, in the Administrator's judgment, sufficient time for




development and application of requisite technology.  Cost of compliance




must be considered also [Section 231(b)].




          2.1.5     Combination of Options




     Pollutants may be regulated under more than one section of the Act




simultaneously (Figure 4).  However, this approach cannot be used  in at




least two specific situations.  These situations occur when pollutants




are potential candidates for control under Section 112 (which cannot be




applied to a pollutant for which an ambient air quality standard exists)




or Section 111, subsection (d) (which cannot be applied to any pollutant




for which ambient or hazardous standards exist).  However, Section 111




may supplement Sections 109 and 112 to control the same pollutant  if





                                   15

-------
specified criteria (as discussed in Subsections 2.1.1-2.1.4  above and




Appendix A) are fulfilled.  Once a pollutant is regulated under Section




111, Section lll(d) becomes effective for existing sources of the pol-




lutant, unless ambient or hazardous pollutant standards apply.




     Standards for moving sources, Sections 202, 211, and 231,  are more




likely to be used in conjunction with Section 109, NAAQS, than other




possible combinations (Figure 4).  For example, Section 202 of the Act




specifies emission limitations for carbon monoxide, hydrocarbons, and




nitrogen oxides from automobiles. Concurrently, ambient air quality




standards have been set under authority of Section 109 for these same




pollutants.  However, standards for moving sources should not be ex-




pected to be used solely with Section 109.  They could also be used with




Sections 111 and 112 to control the same pollutant if specified criteria




of the Act are satisfied.




          2.1.6     Total Ban




     If health data warrants, a total ban on emissions could be achieved




directly under the hazardous standard option, indirectly under the




ambient standard option through stringent levels, and directly under the




new and/or moving source options if an adequately demonstrated system




exists to achieve zero emissions (taking cost into account).




     Using the preceding mechanisms, alternatives and criteria needed in




selecting the preferred standards path (selecting a particular option)




are illustrated in Figure 4.  Decisions listed in Figure 4 are qualified




as "most likely appropriate" because the Act is subject to interpre-




tation (Appendix A).  Since the Act allows flexibility, the choice of a




particular option may create controversy because another alternative





                                   16

-------
could be just as valid.  However, the first requirement for taking any

action under the Act is the establishment of adverse health and/or wel-

fare effects.  Once this is confirmed, alternatives can be evaluated,

using Figure 4 as a guide.

2.2  The Preferred Standards Path

     Now, available information for BaP will be examined in relation to

the first block in Figure 4, i.e., establishment of the possibility of

adverse health or welfare effects.  Specifically, the following reasons

are sufficient justification to consider Federal control of BaP.

     a)   BaP is a proven carcinogen at elevated levels in animals under

experimental conditions.

     b)   Epidemiological evidence connects lung cancer etiology with

occupational exposure to BaP, and other coal tar volatiles, in specific

industries.  Data relating to each of the six options will now be pre-

sented in the order shown in Figure 4.

          2.2.1     NESHAP

     Although an apparent urban factor exists causing a 2.0 to 2.5 times

increase in lung cancer incidence, no evidence exists to link this in-

crease solely to BaP or other POM in the ambient air.  Many areas with

high lung cancer incidence (Appendix B) do not coincide with high ambient

concentrations of BaP as measured by the NASN prior to 1970 (Appendix C).

Unfortunately, many cities with high lung cancer incidence have not been

sampled for BaP.  Nevertheless, experimental animal studies and limited

epidemiological studies do suggest that BaP may cause or contribute to

lung cancer incidence.  Naturally, lung cancer results in an increase in

mortality and/or an increase in serious irreversible, or incapacitating

reversible, illness.
                                   17

-------
However, data do not support that these hazardous effects (e.g.  lung




cancer incidence) are due to inhalation of (or other exposure to) ex-




isting or anticipated atmospheric concentrations of BaP.   Neither do data




define levels at which effects occur as no threshold of effects  can be




conjectured.  In short, with current data and extrapolation techniques




BaP has not been shown with certainty to be hazardous at ambient levels.




Of course, the Act does not require that ambient levels be the deciding




factor, but from a practical point of view, effects should be related




to levels that people are likely to receive in routine activities.  If




these are occupational levels then EPA would not be the agency to establish




standards.  Also, the nature and distribution of sources (block 5) should




have a bearing on the decision reached.  If sources other than stationary




contribute a significant amount of BaP to the total then Section 112




of the Act would not be adequate remedy.  Thus, the conclusion is that the




use of Section 112, NESHAP, for BaP control cannot be justified at this




time.




          2.2.2     NAAQS




     For BaP the NAAQS option (block 7) must be examined because both




source classifications, moving (block 8) and stationary (block 9),




contribute to ambient concentrations.  The moving source classification




also satisfies the criteria of numerous or diverse.  However, available




data do not support a specific concentration for an ambient standard.




Establishment of an ambient air quality standard requires the identifi-




cation of a defensible air quality number (including an adequate margin




of safety) where adverse health effects are observed.  Arbitrarily




establishing a number without sound medical support for it would be





                                  18

-------
inconsistent with the Clean Air Act.   Thus,  the conclusion is that




Section 109, NAAQS, is not an appropriate regulatory tool for control-




ling BaP because evidence to support  an ambient air quality standard for




BaP is not available.




          2.2.3     ESFMS




     If the decision block (no. 8) for ESFMS in Figure 4 is chosen




because moving sources are a significant source of the candidate pol-




lutant, two options become available.  The first option, emission




standards, does not appear feasible because emissions from mobile sources




are insignificant as shown in Table I.  This estimate does not include




emissions from aircraft or diesel-burning moving sources.  No estimate




has been attempted for aircraft but emissions are generally believed to




be slight.  Emissions from diesel engines have been estimated to be




minimal.




     If the emission standards option were not selected, it would cause




a shift to the regulation of fuels option shown in the diagram (block




11).  Based on negligible emissions of BaP from gasoline burning




engines the ROF option can not be justified.  Thus, based on available




data, emission standards and fuel regulation do not appear feasible.




          2.2.4     NSPS




     While it is concluded that data  do not support action by any of the




previous options, the question remains whether Federal action should be




considered under Section 111 of the Act.  Data presented thus far verifies




that Section 111 would be the only feasible regulatory option at this




time if additional regulatory action  is needed.
                                  19

-------
     Section 111 of the Act  authorizes  the  Administrator to establish


standards of performance which reflect  best systems of emission reduction


for categories of new sources.   The  Act further requires States to


establish emission standards for existing sources of the same categories


for pollutants not listed  under  Sections 108 or 112. Section 111 can be


used if the source category  "may contribute significantly to air pol-


lution which causes or contributes  to  the endangerment of the public


health or welfare."  To be defensible,  performance standards under


Section 111 must be applicable to significant categories of BaP. Sources


in these categories are identified  in  Table I based on available emission


estimates (Section 3.0).


     Quantitatively, emissions of BaP  to the ambient air are minimal


(Table I).  However, since BaP is a proven  carcinogen (and/or cocar-


        Table  I.   MAJOR SOURCES OF ESTIMATED  BaP  EMISSIONS  IN  THE

      UNITED STATES BY  DESCENDING QUANTITY OF CONTRIBUTION  (1971-73)
Source type
1 . Coal refuse fires
2. Residential furnaces, coal
(hand-stoked)
3. Coke production
4. Vehicle disposal (open burning)
5. Wood burning (fireplaces, etc.)
6. Mobile sources, gasoline
7. Forest and agricultural burning
8. Tire degradation
9. Open burning
(domestic - municipal )
10. Intermediate coal furnaces
11. Petroleum refineries
12. Enclosed incineration
(apartment - municipal)
13. Other
TOTALS
BaP emissions
(tons/year)
310
300
170*
25
25
11
11
11
10
7
7
3
4
894
% of total
34.7
33.6
19.0
2.8
2.8
1.2
1.2
1.2
1.1
0.8
0.8
0.3
0.5
100.0
            *A range of emissions has been given for this value. Appendix F contains
            the details.
                                    20

-------
cinogen) under given conditions, it may be considered to contribute




significantly to air pollution.   This contribution of BaP is from about




three major categories of sources that are amenable to control,  using




either Section 111 or other means (e.g. local ordinance).  Emissions of




BaP should be reduced because it is a carcinogen.




          2.2.4.1   Alternatives to Control Under  Section 111




     Two major alternatives under this option are:




     a)   Control all major sources of BaP by §111, assuming adequately




          demonstrated control techniques, or




     b)   control only selected sources of BaP.




Either of these alternatives could be used but would result in a com-




pletely new national strategy for BaP only, thus invoking §lll(d) since




BaP is not regulated directly under §§109 or 112.   Strong evidence indi-




cates that such a specific strategy is not warranted due to other con-




trol actions in progress, as discussed in the next section.




          2.2.5     Alternative Control Strategy




     In addition to options already discussed and  dismissed, another




alternative exists:  no direct Federal action, based either on the lack




of demonstrated control technology or that existing Federal, State, or




local controls are adequate and effective.  Control technology does




exist for, and has been applied to, many of the sources listed in Table




I; for others it is non-existent (e.g., forestry burning) or very difficult




to apply (e.g., burning coal refuse.piles).   Federal, State, and local




agencies are also making progress in controlling BaP from all sources.




     Analysis of ambient BaP concentrations shows  a definite declining




trend in all cities (except one) for which data are available. Ambient





                                   21

-------
concentrations of BaP analyzed in twenty-three cities with coke ovens



show a decrease of 55% over a six-year period (1966-1972).  Furthermore,

                                                   i

trends analysis for the benzene soluble organics (BSO)  fraction of  particulate



samples (that portion which includes several POM compounds) shows a 55%



decrease for a ten-year period (1960-1970).  (See Section 5.0.)  Such



definite declines as these indicate that the following positive actions



by control agencies are having concurrent effects on ambient concen-



trations as measured at receptor sites.



          1.   Source No. 1, coal refuse fires, is regulated in four



               States which include the majority of coal refuse fires.



               Other States can control emissions from this source by



               open burning, or other, regulations.  In addition,  the



               Bureau of Mines, Department of Interior, has proposed



               Federal regulatory action for these sources and are



               requiring spontaneously started fires to be controlled



               under existing Federal regulations.



          2.   Emissions from coal-fired residential sources decreased


                                        3 30
               28% between 1968 and 1972 '   and this trend is expected



               to continue for the following reasons:



               a)   Socio-economic conditions have established a trend



                    away from coal-burning in the home.  Urban renewal



                    is underway in practically every U.S. city.  All of



                    these projects avoid the use of coal for heating



                    because it is dirty, relatively inconvenient to use,



                    and in many places cannot meet air quality emission



                    standards.  The manufacturer of units which burn



                                  22

-------
     solid fuel for home heating has declined to the




     point that an industry directory for 1973 did not




     list coal-fired units as being manufactured.




b)   All States except Arkansas have promulgated general




     particulate emission regulations for small fuel




     burning sources.  These regulations, in effect, re-




     quire particulate collectors which are not cost-




     effective for residential sources.  Five States




     limit maximum emission from these sources to 0.8 Ibs




     of particulates per million BTU; this rate is




     slightly less than estimated emission rates for




     coal-fired home furnaces.  Of the remaining forty-




     four States, twenty-two limit emissions to 0.6 Ibs




     per million BTU, and the rest require less than




     0.6 lb/10  BTU.  For example, Oregon limits particu-




     late emissions  to 0.33 lb/10  BTU and Connecticut




     has a limit of  0.2 lb/106 BTU while Alaska limits




     emissions to 0.1 lb/10  BTU.  These regulations have




     been a contributing factor in the phase-out of coal-




     fired residential furnaces.




c)   Residential coal-fired furnaces are banned directly




     in some urban areas, such as Chicago, St. Louis and




     Milwaukee.  Others have such stringent sulfur




     content regulations that coal is banned indirectly.




     For example, Detroit, Philadelphia, and New York,




     allow a maximum of 0.3% sulfur  content.





                    23

-------
          3.    Emissions from all other sources listed in Table I are




               in the process of being reduced, or are scheduled to




               be in the near future.   The remainder of the sources




               contribute less than one-third of the total quantity




               of BaP nationwide and are dispersed fairly widely




               across the U.S.  Moreover, the majority of these re-




               maining sources are amenable to control and will likely




               be controlled by various means in meeting primary




               ambient air quality standards for particulates or other




               pollutants.  For example, by-product coke ovens




               (Appendix D) are being controlled through Federal en-




               forcement actions as well as State and local actions.




               At least fourteen enforcement actions are in progress




               and more are expected.   In addition, EPA is gathering




               data through its demonstration projects in preparation




               for proposing performance standards for new sources




               (Section 111).  These NSPS's for particulate and hydro-




               carbon controls are expected within the next twelve




               months.




     These actions that are in progress and downward trends of ambient




concentrations indicate that major sources of BaP are being con-




trolled.




          2.2.6     Advisability of Federal Control for BaP




     The advisability of Federal control should be based on an




evaluation of the magnitude of effects of ambient BaP as well as




effectiveness of existing and planned actions.  Available data have





                                   24

-------
already been presented which identify those factors of importance in




deciding whe&her Federal action should be specific for BaP.  These data




are restated here for emphasis.  First, evidence to show that ambient




POM is a health problem does not exist.  Secondly, a significant nega-




tive trend in ambient BaP concentrations indicates that such a relation-




ship to adverse health effects will be even more difficult to establish.




Controls by State/local agencies and those resulting from Federal en-




forcement actions should assist in continuing this downward trend.




Finally, NSPS for particulates and hydrocarbons expected by late 1975




should also result in further control of new sources.




2.3  Recommendations



     As a result of appraising available options and supporting data




and considering the advisability of Federal action, the following




recommendations are made.




          1.   In order to minimize POM emissions EPA should continue to




               support those national and local actions that are responsi-




               ble for the downward trends in BaP levels.  However, a




               Federal regulatory program designed specifically for con-




               trol of BaP is not warranted at this time.




          2.   Monitoring of ambient concentrations of BaP in selected




               cities should be continued on a routine basis.  Specifically,




               quarterly composited NASN hi-vol samples for BaP concen-




               trations at the same 30-40 sites studied for this report




               (Appendix C, Table C-2) should be analyzed routinely.  Section




               7.0 gives additional information on current research




               under way by NERC/RTP.





                                  25

-------
3.    Special emphasis should be given to prompt enforcement of




     existing regulations for coal combustion, incinerators,




     open burning (including coal refuse piles), and coking




     operations.  The same emphasis should be placed on new




     regulations as they are promulgated for these and other




     sources of POM.
                         26

-------
                        3.  SOURCES OF  BaP

     The literature indicates that BaP usually is associated with partic-


ulate matter.   Formation of BaP generally accompanies any incomplete


comhustion of fossil fuels or organic compounds, especially coal.  The


amount of BaP produced and emitted is postulated to be directly pro-


portional to the efficiency of combustion and control techniques applied.

Of course, controlled combustion is not the only source of BaP; some


occurs from natural causes, e.g. forest fires.  Also, natural synthesis

of BaP has been postulated by several scientists.    The more common


sources of BaP are listed in Table I and will be mentioned in more


detail below.  Values given in this section are estimates based on the


best available information.  Methods used to arrive at these estimates

may be obtained from the references.



3.1  Stationary

     The category of stationary sources of BaP encompasses a wide variety


of processes that contribute to local BaP concentrations  and accounts


for 98% of the nationwide estimate.  This category is subdivided into


heat and power generation,  refuse burning (which includes forest and


agricultural burning),  and industrial activities.  The most comprehen-

sive examination of stationary sources of BaP available is an EPA pub-
                               Q
lication (AP-33)  by Hangebrauck  et al.  Other POM compounds were


measured also, but BaP received primary emphasis; this, results of


emissions are reported as units of BaP.


                                   27

-------
          3.1.1     Heat and Power Generation



     The majority of heat and power produced and used in the United



States comes from fossil fuels (coal, oil, gas, or wood) and accounts



lor nbonL '17% ol  nationwide iiaP emissions.  Of these the primary source



of BaP appears to result from coal combustion in hand-fired residential



furnaces.  Data from all four fuels, in terms of quantities of BaP



produced, may be seen in Table II.  Emissions are dependent on efficiency



of combustion rather than the type of fuel used.  This is shown by the



low emission factor for intermediate size coal units and coal-fired steam



power plants, in relation to the gross heat input.  Oil and gas have



relatively low emission factors, while the emission factor for wood is



low when compared with hand-stoked residential furnaces.  These data are



consistent with knowledge of POM formation processes, i.e., reducing




conditions caused by insufficient oxygen.  As may be expected, local



variations in quantities of emissions will exist in given areas of the



United States depending on the type fuel, number of industries, and



other local conditions.



     Total BaP emissions given in Table II must be regarded as subject



to change because of several developments.  The two main developments



relate to two residential heating practices, woodburning fireplaces and



residential heating with coal.  These type units have the highest emission



rates and are affected by trends of an affluent society.  Wood-burning



fireplaces are on the increase in pooularity in new urban areas; local



air pollution control agencies are banning, or attempting to ban, resi-

                                                           i.

dential  coal burning.  Based on available data the total BaP emissions



from heat and power generation in the United States are estimated at


                                30
approximately 340 tons annually.

                                   28

-------
             Table II.  ESTIMATED BENZO(a)PYRENE EMISSIONS

       FROM HEAT AND POWER GENERATION SOURCES IN THE U.S., 1972*
Type of Unit
Coal
Hand-stoked residential
furnaces
Intermediate units (chain-
grate and spreader
stokers)
Coal -fired steam power
plants
Oil
Low-pressure air-atomized
Other
Gas
Premix burners
Wood
Gross Heat
(BTU/hr)

0.1 x 106
60-250 x 106
1,000-2,000 x 106
0.7 x 106 ,
0.02-21 x 10b
0.01-9 x 106
-
Benzo(a)pyrene
Emission Factor
(ug/106 BTU)

1.7-3.3 x 106
15-40
20-400
900
100
20-200
5 x 104
Benzo(a)
pyrene
Emissions
(tons /year)

300
7
<1
2
2
25
*Adapted from references  3 and 30.



          3.1.2     Refuse Burning


     In refuse burning, as in coal burning, efficiency of combustion


governs BaP emissions.  Inefficient combustion occurs in all open


burning and also in most small incinerators; thus, formation of BaP


results.  On the other hand, enclosed incinerators designed and operated


for specific tasks should result in efficient combustion and less BaP

emissions.  Presently, this category contributes about 40% to the


national total.

                      Q
     Hangebrauck et al  give BaP emission factors for: (1) municipal


and commercial incinerators for certain wastes; (2) municipal and


agricultural open burning; and (3) junked vehicle disposal.  As ex-
                                   29

-------
             Table  III.   ESTIMATED BENZO(a)PYRENE EMISSIONS
                 FROM REFUSE BURNING IN THE  UNITED STATES
                                 (1968)*
      Source  of Benzo(a)pyrene
   Benzo(a)pyrene
Emissions  (tons/year)
      Enclosed  incineration
        Municipal
        Commercial and industrial
        Institutional
        Apartment
      Open  burning
        Municipal
        Commercial and industrial
        Domestic
        Forest  and agricultural
        Vehicle disposal
        Coal  refuse fires
        23
         2
         8

         4
        10
        10
        11
        50
       340
*Adapted  from  reference 3:   an  estimate based-on 1972  data indicates a total
of approximately 340 tons being emitted,   '   accounted for as follows:
   a)  coal  refuse fires - 310 tons
   b)  forest and agricultural.- lltons
   c)  vehicle  disposal - S^tfths
   d)  other open burning -  10 tons
   e)  enclosed incineration - 3 tons
The method  of  estimating emissions may have varied for the two years.
pected,  these emission factors vary widely and again show the  importance

of efficient  combustion.  Large municipal incinerators  (50-250 tons of

refuse/day) have a BaP emission rate  of  0.1 - 6 mg/lb of charged  refuse.

Commercial  incinerator (3-5 tons/day)  emission factors  ranged  from 50-

260 mg/lb.  Emission factors for municipal open burning are given an

average  value of 150 mg/lb while vehicle disposal results in about

mg/lb of refuse.  Higher emissions of  BaP from refuse burning  than

originally  estimated (20 tons/year) by Hangebrauck e£ ai^ have  been
                                    30

-------
projected recently.  These later emissions, shown in Table III,  are




based on higher estimates of nationwide refuse burning, and the  in-




clusion of coal refuse fires, rather than significantly different




emission factors.  The single largest contributor identified is  the coal




refuse pile under "Open Burning."  Such waste piles are common in coal




mining areas and are frequently referred to as culm (or gob) piles (or




banks).






     The given estimates should be regarded as an order—of-magnitude




approximation because of the uncertainty of factors used to estimate




emissions.  Therefore, approximately 340 tons/year of BaP from refuse




burning is the best available value.




          3.1.3     Industrial Activity




     Some industrial processes are conducive to direct source measure-




ment while others must use indirect means, often imprecise, for  esti-




mating emissions.  The former category includes catalytic cracking of




petroleum and air-blowing of asphalt, probably the most obvious  sources




of BaP in the petroleum refining industry.  The latter category  of




emissions estimating, indirect sampling, may be illustrated by coke




production in the steel industry; carbon black, coal-tar pitch,  and




asphalt hot-road-mix processes; and general chemical processes.   The




procedure generally used consists of sampling the ambient air as near




the expected source, or complex of sources, as possible and back-cal-




culating emissions.  This technique necessarily yields much less accurate




results because of background ambient concentrations and other factors.
                                  31

-------
     Industrial emissions of BaP are summarized in Table IV and account


for about 20% of the national total.  Experience has indicated that


those industrial processes that must be sampled indirectly contribute


insignificant quantities of BaP to the environment, with the exception


of coke production.  Estimates of BaP emissions from coke ovens range

from about 0.06 tons per year to approximately 170 tons/year (Appendix


F).  The higher value is used here because it is based on data from the


U.S.  The high BaP discharge from coke production appears to be associated


with the effluent from the charging and coking processes.  A crude emission


factor of 2.5 g of BaP/ton of coke produced was applied to estimate BaP


emissions from coke production.  A more accurate assessment of coke oven


emissions is being developed.
                Table IV.  SUMMARY OF ESTIMATED INDUSTRIAL

              BENZO(a)PYRENE EMISSIONS IN THE UNITED STATES

                                 (1972)*
                  Source of
                Benzo(a)pyrene
             Petroleum

             Asphalt air-blowing

             Coke production
Benzo(a)pyrene
  Emissions
 (tons/year)
 (0.06-) 170
       *References  31;  8;  and Appendix  F.

     Obviously, other industrial processes are sources of BaP (and POM)

but are more localized and less significant as a nationwide problem.

Thus, the current best estimate for BaP emissions from industrial sources

is about 175 tons/year in the United States.
                                   32

-------
3.2  Mobile



     Generally, when one thinks of mobile sources,  thoughts of automo-


biles immediately enter one's mind.  Of course, sources other than the


automobile contribute to this group; for this reason, many people


prefer to call this category "transportation" sources rather than


"mobile."  Because of difficulties associated with the type testing

                    O / "7 O O
required, literature ' ' '   on vehicular sources of BaP emissions is


sparse.


          3.2.1     Gasoline-powered Vehicles


     Of the various transportation sources contributing to emissions of


BaP, the automobile has been studied the most.  Contributions of gasoline-


powered vehicles can be separated into vehicular characteristics and


fuel composition effects.  The first category includes the effects of


the air:fuel ratio; emission control devices; driver operating modes;


engine deterioration; and combustion chamber deposits.  The second


category is composed of fuel aromaticity, BaP levels, additives, and


lubricants.


     As with other sources of BaP, efficient combustion apparently


results in less BaP emissions.  Air:fuel ratios greater than stoichio-


metric, i.e. use of excess air, promote efficient combustion in the


cylinders of internal combustion engines.  Recent data indicate that BaP


production decreases about 30-fold with an increase of the air:fuel

                        2
ratio from 10:1 to 14:1.
                                   33

-------
                        4 7 32
     Various researchers '  '    have shown that other factors mentioned



above also increase emission quantities of BaP.  Engine age causes



deterioration of operating parts, thus permitting more oil consumption



and concomitant BaP emissions.  Also, excessive wear creates more space



for deposits to accumulate in the combustion area.  These deposits,



through a complex process,  cause increased BaP emissions.



     Since BaP is composed of aromatic compounds, it stands to reason



that the more basic aromatics in the fuel the easier for BaP to be



formed and emitted.  However, based on recent research, higher fuel



aromaticity may be offset by using unleaded gasoline which changes the


                                         32
character of combustion chamber deposits.    Other research in progress



tends to dispute such a conclusion.  Nevertheless, one thing appears to



be certain, with the use of unleaded gasoline the distribution of BaP



emissions from mobile sources will be different.  Development of emission



control devices introduces further complicating variables in the overall


                                                                  32
effect but studies thus far indicate these devices are beneficial.



          3.2.2     Diesel-fuel-powered Vehicles



     Studies on diesel-fuel-powered vehicles have been limited mainly to



trucks and buses under laboratory conditions.  Actual on-the-road



operation of diesel-powered vehicles can result in high BaP emissions



due to engine overloading, poor maintenance, and other factors.  If



diesel engines are not overloaded, laboratory tests have indicated that



higher BaP emissions occur at idle than at any load condition.  Further-



more, these same tests indicated that half-load produced highest BaP



emissions during operation.  Emissions dropped sharply at full load,



presumably because of increased combustion chamber temperatures re-


                                  3
suiting in more efficient burning.



                                   34

-------
       lier mohi/e.  Sources,  likely  e.prrlribu'fe,  *£ J7  e/sj/ss/
-------
3.3  Natural




     Data presented thus far provide good substantiation that BaP is




primarily a product of incomplete combustion.  However, a few sources of




BaP may be defined as naturally occurring.  The first of these has




already been alluded to—it is bituminous coal.  In addition to being




analyzed for BaP content, coal has also yielded benzo(a)anthracene and




other unspecified POM.  Two of three types of asbestos used industrially




contain appreciable quantities of natural oils.  These oils have been




found to contain BaP.




     Two additional sources may be more difficult to verify but no




question of their natural occurrence should arise.  One researcher   has




identified various molds In the environment as a source and BaP.  Several




scientists have indicated that BaP is synthesized naturally in the




environment.
                                   36

-------
                    4.  CONTROL TECHNOLOGY








    The nature of emissions referenced  above  provides a broad hint to




control procedures.   Since inefficient  combustion  of fossil fuels is a




primary source of BaP,  by  far the most  important control  technique is




to burn fuels efficiently.  Fuel or process substitution  may be




directly applicable to many processes of  BaP  emissions.   For example,




gas or oil are inherently  more efficient  than coal; incineration is




much more desirable than open burning.  These two  methods have been




adopted widely recently.




    A desirable end result may be possible only by adding control




equipment in line with the regular process.   In particular, reduction




of BaP from a source may be achieved by treating the effluent gas




stream.  Bag filters,  scrubbers, or electrostatic  precipitators are




standard devices for removing particulates upon which BaP is likely




absorbed.  Afterburners, catalytic mufflers,  or condensation may be




applicable techniques to control vapor  phase  POM.






4.1  Stationary
    Since the most important contributors of BaP emissions  in




heat and power generation are hand-fired coal furnaces  and  wood




burning, the preferred choice of control would be by  alternative  fuel




sources.  One might argue that substitution is the only means of
                                  37

-------
control for these two sources as neither is amenable to better




engineering controls.




    In regard to refuse burning, efficient incinerators are being




installed for municipal, commercial, industrial, and apartment




building sources.  Relative contributions from these sources are




diminishing in comparison to coal refuse banks.  Existing burning




refuse banks are required to be extinguished if burning began




spontaneously.  Federal regulations have been proposed for all burning




coal refuse piles which will eliminate most emissions as these




regulations are implemented.  Future culm piles are being eliminated




by more specific attention to refuse accumulation practices in mining




areas.




    Benzo(a)pyrene emissions from catalytic cracking in the petroleum




industry apparently are receiving necessary action through utilizing




carbon monoxide  (CO) waste boilers for effective control (see Table




VI).  From Table VI it is clear that CO waste-heat boilers have a




signficant beneficial effect on BaP emissions, and thus on other POM




emissions.




    Much of the BaP produced at by-product coke ovens  (Appendix D)




quite likely could be removed in the by-product stream, except for




that escaping from a leaky system.  However, BaP emissions from most




coke processing still appears to be quite high, whether it originates




from charging, leaks, or directly from the gas stream.  Some possible




approaches to controlling BaP from coke ovens are under investigation




by EPA and industry groups; two demonstration projects for particulate





                                   38

-------
               Table VI.   ESTIMATED BENZO(a)PYRENE EMISSIONS
            FROM CATALYTIC CRACKING SOURCES IN THE UNITED STATES
                                   (1968)*

Type of Cracking Unit
FCC
no boiler
CO boiler
Subtotal
HCC
no boiler
CO boiler
Subtotal
TCC (air-lift)
no boiler
CO boiler
Subtotal
TCC (bucket-lift)
no boiler
CO boiler
Subtotal
Total
Petroleum Consumption
(million barrels/year)

424
1,230
1,654

14
55
69

27
118
145

17
75
92
1,960
Benzo(a)pyrene
Emission (tons/year)

0.08
0.02
0.10

'3.4
0.0
3.4

2.4
0.0
2.4

0.0
0.0
0.0
5.9
*Adapted from reference 3; reference 31 gives 1973 petroleum consumption
as 2,372 million barrels/year which yields about 7 tons of BaP annually.
control from coke ovens are in progress and BaP sampling is being done

simultaneously.

    The Control Systems Laboratory, NERC/RTP, began coke oven

particulate control demonstration projects in 1972.  These projects

were designed 'to demonstrate the feasibility of controlling

particulate emissions from coke ovens, on both the charging and

pushing sides of the oven.  Particulate control demonstration projects

are, for the most part, being accomplished on existing coke ovens.

The purpose of demonstrating the availability of control technology

was to gather background information for standards of performance for

               f\ f\
new coke ovens.   However, since the demonstration projects are on

                                     39

-------
existing sources, preliminary results indicate that particulate




controls are technically feasible for existing as well as new sources.




Since coke oven controls are complex and many questions must be




answered, the various projects underway by EPA are divided into




several phases.  Only initial phases of both charging and pushing




control demonstration have been completed.  However, from observations




during these phases, particulate control appears to be feasible.




    In the charging-phase demonstration project, source sampling for




BaP was performed in conjunction with particulate sampling.




Preliminary analysis of sampling results indicate that about 90



                                            23
percent control of particulates is achieved.    With this level of




particulate control, a similar degree of reduction in BaP emissions




was expected.  Preliminary evaluation of the BaP fraction of the




particulate from the controlled versus the non-controlled larry car




(vehicle that drops coal into the oven) indicates that approximately




85-90 percent control is achieved.^3 if these results are obtained for




other types of particulate control (e.g., pipeline charging*),  control




processes should also reduce BaP emissions.  Such results confirm




previous expectations of control engineers that particulate control




will significantly reduce BaP emissions.




    Presently, no data are available on the amount of BaP control




achievable on the pushing side of the coke oven.  The second phase of




a demonstration project is underway to obtain control estimates on the




pushing side.  Based on observations of installed systems, engineers
    * Pipeline charging is a method for pre-heating coal and charging


it into coke ovens via pipelines in a totally enclosed system;


licensed by Coaltek Associates.




                                   40

-------
estimate that control on the pushing side should work as well as


                                          on

controls on the charging side of the oven.




    Thus, we see from Tables I-IV that BaP emissions originate




primarily from stationary sources, accounting for about 97 percent of




total nationwide BaP emissions.  However, due to inadequate sampling




techniques these emissions should be considered estimates of relative




ranking.  Furthermore, since stationary sources account for the




majority of BaP, these emissions are highly localized.  Table VII




depicts some degree of localization for the three major sources.  Such




localization indicates those areas where control efforts should be




emphasized.





4.2  Mobile
    Before the current concern for reducing vehicle emissions was




instigated by the Clean Air Act, most vehicles operated with fuel-rich




carburetion to promote smooth performance and quick power response.




Preliminary modifications prior to the 1970 Clean Air Amendments




resulted in leaner fuel-air mixtures and lower BaP emissions.  This




trend is indicated in Table VIII by the emission factors estimated in




1972 for the periods indicated.  "Advanced systems" in Table VIII




refer to prototype emission control devices spurred into development




by the 1970 Act.  Generally, advanced systems utilize catalytic




converters and thermal reactors.




    Emission control devices will effect control of BaP concomitantly.




Data to support this view are based on current exhaust sampling •



           32
techniques.   If such a high degree of control is achieved in actual




practice, mobile source pollution caused by BaP will assume a much




                                    41

-------
             Table VII.  CONTRIBUTIONS TO NATIONAL TOTALS OF
                    BENZO(a)PYRENE BY SOURCE AND STATE
                                 (1972)*
Source of
benzo(a)pyrene
emission
Coal refuse burning




Total
Residential coal-
fired furnaces




Total
Coke production



Total
State
Pennsylvania
West Virginia
Ohio
Kentucky
Virginia
Alabama

Illinois
Pennsylvania
Kentucky
Ohio
Tennessee

Pennsylvania
Indiana
Ohio
Alabama }
Michigan /
W. Virginia )

Percent of U.S. total
from given source
contributed by given state
46
37

16

99
23
16
7
7
7
60
27
15
15
21
78
     *References 28; 33; and 34.


less significant role.  At any rate, one may logically expect a
reduction of BaP emissions from light-duty vehicles as a result of the

EPA motor vehicle control activity.  A higher degree of uncertainty

exists for other type transportation sources, e.g., buses,  heavy diesel
trucks, aircraft, etc.
                                   42

-------
                              Table VIII

               AUTOMOTIVE BENZO(a)PYRENE EMISSION FACTORS
                               (1972)*
                Source
Benzo(a)pyrene emission factors
   Ug/gal  of fuel  consumed)
   Uncontrolled car (1956-1964)

   1966 Uncontrolled car

   1968 Emission-controlled vehicle

   Advanced systems**
              170

             20-80

              6-24
   *Adapted from reference 3 & 32

  **Prototype emission control systems



4.3  Natural


    Control of BaP from natural sources is even more limited than

other types control because natural emissions are less well-defined.

For bituminous coal, product substitution would be the logical

approach to control.  Of course,  suitable substitutes for specific use

and application must be found.  Asbestos control under NESHAPs should

preclude BaP from the environment also.

    As for control of BaP from natural synthesis and molds, no

suggestion has been forthcoming.   Their contribution to ambient

concentrations is so negligible that it is likely not worth time and

money required for exploration.
                                   43

-------
                  5.  AMBIENT CONCENTRATIONS


    Nationwide ambient concentrations  for most pollutants of interest,
including BSO and BaP, are available through  the National Air
Surveillance Network (NASN).   Concentrations  available through this
network provide only relative nationwide comparisons because of
limited sampling sites in a given area.  An additional problem in
using NASN data for ambient BSO and BaP concentrations is that none of
these sites were selected specifically to sample for organic
pollutants.  However, representative data are available  from some
suspended particulate sites in the NASN.  In  addition, summary reports
on POM reference various special studies that have been  undertaken
since the late 1950's.
    If a detailed breakdown of NASN data by quarter were examined, a
seasonal trend for BSO and BaP would be obvious with few exceptions.
Higher concentrations would be noticeable for the fall and winter
quarters at most sites while the summer generally has the lowest
concentrations.  Studies conducted in  the late 1950's indicated this
             12 11
same pattern.  '

5.1  Measurement Technique

    To better understand the measurement  of  ambient BaP  data,  a
brief description of the process should be helpful.  Particulate
samples are collected upon tared fiber glass  (flash-fired) filters by
                                  45

-------
drawing air through the filters at a relatively high volume of air




flow (aproximately 50-60 cfm) during a 24-hour day.  These filters are




sent to a laboratory for preliminary weight analysis, after which




organics are extracted from a portion of the filter by the Soxhlet




method for about six hours using a suitable solvent  [ordinarily




benzene—thus the term benzene soluble organics (BSD)].  Subsequently,




BaP is separated from total BSD by thin layer chromatography (TLC)




followed by ultraviolet (UV) spectral analysis.




    A minimum of 50 mg of BSO material is required for a BaP




analysis using TLC and UV.  In order to obtain a sufficient quantity




of sample for this analysis several organic fractions from one station




must be pooled.  Usually each urban 24-hour hi-vol particulate sample




will contain about 25 mg of benzene extract.  Generally, samples for a




three month period, consisting of 5-7 samples, are composited to




obtain one data point.  Thus, an annual station mean would consist of




averaging four quarters of composite samples for one station (see




Appendix C).  Table IX is an annual summary of BSO,  BaP, and TSP for




the last three years of analysis from 121 selected NASN sites where




BaP was analyzed on a routine basis.  Table C-2 of Appendix C lists




those sites which were analyzed after a special request for 1971 and




1972 BaP data.  TSP is an abbreviation for total suspended




particulates, which is obtained from a weight analysis of hi-vol




filters and relating this to average air flow for the 24-hour sample




period.  Note from Table IX that BaP trends follow the average BSO




content and not TSP sampled.





                                  46

-------
              Table IX.   THREE YEAR SUMMARY OF TSPa,  BSO,

                  AND BaP* FOR 121  NASN SITES IN THE

                 UNITED STATES - 1968, 1969, AND 1970
Year
1968
1969
1970
Pollutant
TSPa
(yg/m3)b
89 (88)d
84 (81)
96 (88)
.BSOa
(yg/m3)
5.7 (5.5)
5.0 (5.0)
4.3 (4.3)
BaPa
(ng/m3)c
2.4 (1.6)
2.3 (1.6)
2.0 (1.2)
             *See Appendix C for the annual  average for each
              selected site for each year.

             aTSP = total suspended particulate; BSO = ben-
              zene soluble organics; and BaP = benzo(a)-
              pyrene.

              yg/m3 =  micrograms of sample/cubic meter of
              air.

             cng/m3 =  nanograms of sample/cubic meter of air.

              Numbers  in parens are median station values
              for the  121 sites.
    No known routine studies by State or local agencies are being


conducted on ambient levels of BSO and BaP.  Several States have


inquired about analytical procedures but have given no indication of


follow-through.  At any rate, no BaP data from State or local programs


are recorded in OAQPS's national data bank.




5.2  Ambient BaP Concentrations


    Figure 5 is a graphical representation of annual average BaP

concentrations for coke-oven and noncoke-oven cities for 1966 to 1972.


This figure is presented to show the general trend in the two type


cities.  The overall trend is decreasing in each case.  With the


exception of the dip in 1968 for the coke-oven cities, and 1967 for
                                  47

-------
                   1967
1968        1969

       TIME, year
1970
1971
1972
   Figure 5. Trends in annual BaP concentrations in cities with and without coke ovens.



noncoke-oven cities,  the trend  is  quite similar for both type cities.


In each case, the number of  cities included  in the annual average


varies from year  to year because only those  cities with a full year's


data  (4 valid quarters)  were included (Appendix C, Table C-3).


    Average BaP  concentrations  for coke-oven cities have


decreased from 4.74 nanograms per  cubic meter (ng/m  ) in 1966 to 2.14


ng/m3 in 1972  (Figure 5  and  Table  C-3).  This is a 55% decrease in a


six-year period.  The decrease  in  noncoke-oven cities is even more


conspicuous:  a  77% decline  occurred over the same six-year period,

                                            o             O
with  concentrations dropping from  2.76 ng/m  to 0.64 ng/m .


Individual trends were decreasing  for most of the 34 cities used in


computing BaP averages.   However,  some cities showed an upward


variation in some years.  One city, Chattanooga, Tenn., showed no


decline at all over the  six-year period based on a rank correlation


coefficient of zero.   Two other cities, Pittsburgh, Pa. and Ashland,


                                   48

-------
Ky., showed a weak decreasing trend pattern with the rank correlation




of BaP versus time of -0.14 and -0.40, respectively, because of




positive fluctuations in the trend line.




     Realizing  that whatever  is  happening  in  the  noncoke-oven




cities is also happening in  the coke-oven cities for the overall




trend, one can still surmise certain occurrences from information




available.  Even though limited data do indicate that coke ovens are a




major contributor to localized BaP concentrations, one must remain




aware that coke ovens are not the sole source of BaP emissions.  For




example, the stockpiling of  steel in 1967 in anticipation of a steel




strike in 1968 likely contributed to the noticeable decrease of  BaP




concentrations in 1968 for coke-oven cities  (Figure 5).  On the  other




hand, an abrupt decline in BaP concentrations occurred in noncoke-oven




cities in 1967, indicating the possibility that  less coal was




available for nonsteel industry use, especially  small consumer use.




     Again, one must  remember that  no  readily evident method




exists for identifying the specific contribution of coke ovens because




the  amount of coal delivered to individual cities is not available.




However, the circumstantial  evidence presented is quite strong.  The




trends analysis presented below establishes  that a  significant




difference exists in BaP concentrations for  cities with and without




coke ovens.








 5.3  Trends  Analysis for  BaP




     Two  statistical  techniques  were used  to  establish whether  the




 apparently decreasing trend (Figure 5)  in BaP  concentrations  was




 significant.   These techniques were a linear regression and Spearman




                                49

-------
rank correlation analyses.  The Spearman rank correlation procedure, a




nonparametric approach correlating the rank of BaP concentrations and




time, was used to avoid the assumption of a certain distributional




form for the data, and other requirements inherent in parametric




correlation and regression analysis.  This procedure was used on the




data for each of the cities studied while the regression technique was




applied to individual averages separately for the coke-oven and




noncoke-oven cities.  An analysis of variance (ANOVA) was used to




establish whether a significant difference in BaP concentrations




existed between coke-oven and noncoke-oven cities.




    The general trend for BaP is decreasing.  Regression analysis




performed on the cities with and without coke ovens for BaP




concentrations versus time verify that the trends shown in Figure 5




are significant, i.e., the decreasing pattern in BaP concentrations




over time are unlikely to have occurred by chance alone under the




hypothesis that a trend is not present in the data.  Therefore, this




suggests that the data support an alternative hypothesis of a trend




component being present in the data.  The linear regression for both




the coke and noncoke cities could be described as parallel lines,




i.e., lines with the same negative slope but being displaced by a




significant  difference  in concentration  between  the  two  sets of  lines




representing the  different cities  (see Figure 5).




     The  Spearman  rank correlation  analysis was applied  to  the




individual  cities by  looking  at  relative rankings  of annual average




BaP concentrations  versus time  (years).   Of  the  34 individual  sites




analyzed, all have  negative correlation  coefficients,  except






                                  50

-------
Chattanooga which was zero.  Of these, 12 sites were statistically




significant.  A statistically significant result (a = 0.05) implies




that an observed correlation coefficient was not likely to be derived




from a population of coefficients whose true correlation is zero.




Statistically, we can infer that a significant negative association




exists between BaP and time for about 35% (12 of 34) of the sites.




These results support the conclusion that a decreasing trend does




exist for BaP concentrations over time.




    In establishing a significant difference between cities with




and without coke ovens an analysis of variance procedure was used to




evaluate BaP concentration variations.  With no way for considering




the many additional uncertainties (other than time) which influence




BaP concentrations in cities with coke ovens, the ANOVA shows that a




statistically significant difference does exist between the two type




cities.  In fact, only a very small probability (0.0001) exists that




the difference in the observed BaP concentrations between the two sets




of cities could have occurred by chance alone, under the hypothesis




that no difference exists in BaP concentrations between the two sets




of cities.  In other words, based on the seven year's available data,




BaP concentrations are higher in cities having a coke oven, although




it is impossible to evaluate the contribution of other sources of BaP




to explain  this difference.





5.4   National Urban  and  Nonurban Trends  for  BSD





    Composite annual averages of BSD concentrations and BSO




expressed as  a percent of TSP are shown in Figure 6 for 32 urban and




19 nonurban NASN stations.  The urban composite average concentrations




                                 51

-------

o
o
o
in
CO
    12
    10
     1960
            (ONLY 18 NONURBAN STATIONS
            INCLUDED IN THIS COMPOSITE
            AVERAGE).
1961
1962    1963
1964     1965     1966

      TIME, year
1967
1968
1969
1970
Q.
KO


o
OO
CO
     1960     1961      1962     1963    1964     1965     1966     1967     1968     1969      1970

                                          TIME, year


  Figure 6.  Trends in BSD and in BSD percentage of TSP at 32 urban and 19 nonurban stations.


 of BSD  have decreased from 10.6 Mg/m3 in  1960 to 4.8 Mg/m3  in 1970.


 This represents  a  55 percent decrease.  The national trend  in BSO


 concentrations shown closely parallels  the  trends  observed  at the vast


                                         52

-------
majority of the stations included in the composite sample.
Statistically downward trends were found for 27 out of 32 stations
represented by this grouping.  (The same statistical techniques
described under BaP trends were used for BSD.)  Only in Helena,
Montana, was there some indication found of an increasing trend
pattern.  Portland, Oregon, did not exhibit a clear-cut trend pattern
and was described as having "no trend," whereas Providence, Rhode
Island, Seattle, Washington, and Milwaukee, Wisconsin, exhibited
weaker decreasing trend patterns than was the general case.  The
overwhelming consistency in trend patterns points to some general
phenomenon exclusive of geographical location within the country, city
size, or industrial sources present in a particular city.
    The decrease in the composite average concentrations  of BSO
in 1963 — approximately 1 vg/m   — together with similar decreases
in 1964 and  1970, represent  the years with the largest concentration
changes.  The nonurban average BSO concentration, which is a factor of
approximately 4 or 5 less  than the urban averages, exhibits a more
uncertain pattern.  However, since 1966, nonurban BSO averages have
decreased from  2.2 Mg/m3 to  1.2 Mg/m3 in 1970.
    The  BSO  percent  of TSP has also  decreased  for  the urban
stations over the period considered.  The pattern is characterized by
rather sudden decreases from 1963 to 1964  (8 to  6 percent) and from
1969  to  1970 (6 to 4.5 percent).  In other years, the BSO percent of
TSP remained relatively constant.  It appears  that significant
decreases  in national urban  levels of both BSO and BSO percent of TSP
have  occurred,  notwithstanding sizable, short-term, year-to-year
fluctuations.
                                 53

-------
     Nonurban trends in the BSO concentration and the BSD percent


of TSP have also been downward during the 4-year period, 1966-1970.


It is interesting to note that the averages of the BSO concentration


and the BSO percent of TSP for the nonurban stations (with the


exception of Curry County, Oregon) were lower in 1965 and 1966.


However, insufficient pre-1965 data exist to judge whether trends in
                                                       i-

nonurban BSO concentrations have paralleled the concentration


decreases at the urban sites.
                                  54

-------
                        6.   EFFECTS OF BaP




     Description of lung diseases related to dust or airborne particles




was advanced first in the sixteenth century.  Two other researchers




demonstrated more than three hundred years later that lung cancer  was  a




prominent pulmonary disease among miners.^  Now in the mind of the general




public, lung cancer is associated with cigarette smoking because of




superabundance of publicity, as well as "the warning on the label."  The




relation between cigarette smoking and lung cancer has stimulated  interest




tn tho role of air pollution in cancer.  This interest has arisen  because




some pollutants found in urban areas, including BaP, are similar to




those found in cigarette smoke.  The necessity of defining all significant




etiologic factors in lung cancer has become evident because of its




rising incidence and poor prognosis.  At this stage, primary prevention




is the most effective means of lung cancer control.




6.1  Epidemiological Studies



     Some types of comparison strongly support the hypothesis that




increased lung cancer mortality rates are related to urban pollution.




The epidemiolopic method of studying air pollution effects on lung




cancer incidence entails comparing lung cancer death rates in communities




that have markedly different pollution levels, but are similar in  other




potentially causal factors.  Comparisons between urban and rural areas




within the same country are attractive in that, generally, fewer variables




are involved and pollution level differences are maximized.  The disadvantage





                                   55

-------
is that very little data on ambient concentrations of pollution are available

for rural areas.  Thus, comparisons between urban areas offer an advantage

over comparisons between urban and rural areas but have other disadvantages

in that different variables may affect the study.  Each comparison

requires a suitable adjustment for any known extraneous variables which

may affect it.

          6.1.1     Classification of Studies

     Four groups of epidemiologic studies are used to arrive at evidence

of environmentally related lung cancer incidence in humans.   They are

studies to:

          1.   Compare urban metropolitan populations with rural

               populations in relation to lung cancer mortality rates,

               usually without examining etiologic factors.

          2.   Compare lung cancer death rates in migrants to those

               in country of origin and those in country to which they

               migrate; and examine change in rate in relation to change

               in the environment.

          3.   Compare demographic units and study the relation between

               lung cancer death rates and various indices of pollution.

               Multiple regression techniques are generally used to

               separate environmental effects from other factors.

          4.   Sample characteristics of lung cancer decedents through

               family interviews and compare these with corresponding

               characteristics of the remaining population.

Through such  studies the prospect is offered of identifying a relatively

sharp  distinction between strongly related factors and those incidentally
associated  with lung cancer.
                                   56

-------
     Urban-rural differences undoubtedly offer clues to problems of
pulmonary carcinogensis in man.   The general association between
urbanization and increased lung cancer is not questioned.   However,
characteristics of the urban environment which are primarily
responsible for lung cancer are a subject of controversy.   The problem
of identifying causal factors is amplified by difficulties of
obtaining either accurate or extensive measures of exposure for most
factors.  Further intensification of the problem occurs from the close
association of urban pollution factors with one another.  For example,
an area with high BaP levels will generally tend to have high concen-
trations of other hydrocarbons and sulfur dioxide.
          6.1.2     Consideration of Other Variables
     Special attention should be paid to imperfections or contradic-
tions in the correlation of air pollution measurements and lung cancer
incidence.  General agreement apparently exists on the increment in the
incidence of pulmonary cancer caused by smoking; interpretations differ,
however.  This is of immense interest, because urban air and cigarette
                                                                       13
smoke have carcinogenic substances, e.g. BaP, and some gases in common.
Such similarity is likely the reason frequent reference is made to
studies of cigarette smoking and pulmonary cancer.  However, complex-
ities of the problem of pulmonary carcinogensis are compounded, rather
than simplified, by that fact.
     Considering the numerous variables that are known and anticipating
that some are likely completely unknown, one should focus attention
on the target tissue and consider the air as the primary point of
contact with carcinogenic agents.  Two specific variables which must

                                   57

-------
be considered in epidemiologic studies are irritant or toxic gases


in the atmosphere and diseased tissues at the target site.  These


variables are difficult to quantify or to determine their specific


relation to carcinogenicity.  The common link between these two


variables is their potentiating action on carcinogenic properties


of BaP.  Tissues altered by disease may incur a precancerous state
                                                     *.

which appears to contribute to carcinoma susceptibility.


     No attention has been given to effects on domestic animals


because no information appears to be available.  Of course, except for


epidemiological studies, most data on the carcinogenicity of BaP


have been obtained by using experimental animals.  Various techniques


have been used to verify carcinogenicity in these animals.  A compre-

                                                              3
hensive review of these studies is presented in the NAS report


and the NERC/RTP position paper on PPOft.

          6.1.3     Conclusions from Epidemiological Studies


     Studies of lung cancer thus far have indicated the following


conclusions, generally resulting from the four broad study classifi-

                     3
cations listed above.


          1.   Lung cancer has emerged as the single greatest cause


               of cancer death in males and is a significant cause in


               females in the U. S. Its incidence has increased in the


               last 30 years.


          2.   A major etiologic factor appears to be cigarette smoking;


               however, smoking does not account completely for this


               increased incidence.
                                   58

-------
          3.    Urban dwellers  have approximately  twice  as  high  an




               incidence of lung cancer  as  those  living in rural areas




               after correcting for smoking.   Within  urban areas,  in-




               cidence hns been greater  in  areas  of industrial  pollution.




          4.    Renzo(n)pyrene, found in  cigarette smoke in high concen-




               trations, causes cancer of the lung and  other  organs  in




               experimental animals.  Also, it is present  in  the air at




               relatively high levels in industries whose  workers  have




               been found to have high mortality  rates  due to lung




               diseases, including cancer.   Finally,  it prevails at




               different levels in all urban cummunities investigated.




          5.    Generally, immigrants have an incidence  of  lung  cancer




               between that noted for their countries of origin and  that




               of the countries to which they migrate.   The higher their




               ages at migration the nearer their incidence rates  to




               their cohorts who do not  migrate.   In  studies  where the




               home country had a greater cancer  incidence rate,  rates




               of migrants decreased significantly, even if their  cigarette




               smoking increased.




6.2  Other Effects




     Other health effects of POM are related mainly to  skin disorders




from occupational exposure of man to high levels.  Studies of these




effects do not attempt to differentiate  exposures by  air from exposures




by other means, e.g., contact.  This aspect should be considered  care-




fully in attempting to extrapolate the information derived for occupa-




tional problems to the possible hazards  of ambient air  pollution.   No





                                   59

-------
documentation exists that participate materials containing POM in ambient




air have resulted in adverse skin effects.




     Welfare effects of BaP are even more inconclusive because few




specific studies have been conducted in this area.   However,  the prob-




ability of altering soil microbial populations appears to exist and




such action may upset the ecosystem.    The nature  of BaP indicates




that it may be produced in many plants or plant products; however,




these data are speculative.
                                   60

-------
        7.  PROGRAMS TO STUDY  BaP  (AND OTHER POM)


     Polycyclic organic  matter has been studied for two hundred years


among occupational xjorkers.   Descriptions  of lung disease related to


exposure to dust or airborne pollutants were said to exist as early as


the 16th century.   The first recorded description relating the burning


of fossil fuel to occupational disease was written in 1775 by Percivall


Pott.  This was related  to constant exposure to soot among chimney


sweeps who had developed cancer  of the scrotum.  Then in 1879 two German


scientists demonstrated  that lung cancer was a prominent pulmonary

                     3
disease among miners.    Thus, occupational exposures provided the first


evidence of cause-effect relations from airborne pollutants that now


have been associated with BaP.


     Since those early years many studies  have been undertaken to


define cause-effect relationships between  airborne contaminants and lung,


disease and to identify  the causal agent in airborne pollutants. In more


recent years,  beginning in the mid-  to late 1950's, specific programs


have been initiated to establish such a relationship with the frequency


and complexity of studies increasing  each  year.  Such has been possible


because more and more people and institutions, or sponsors, have become


interested in obtaining as much  information as possible concerning BaP


and other POM compounds.  However, in all  of this interest no forcefu],


common objective has been evident.  As a result, much effort has been


duplicated and much money utilized unnecessarily.



                                 61

-------
     The purpose of this secton is to give an analysis of those programs




concentrating on BaP and other POM that are underway by NERC/RTP.





7.1  Current Research Projects



     Table X is a synoptic listing of those programs which have a




direct relation to emissions of BaP and/or other POM compounds.  These




projects were extracted from the NKRC/RTP draft Position Paper on PPOM.




Items of interest pertinent to the listed subjects will be given below.




          7.1.1     Analysis Methodology




     Several approaches to the improvement of analytical techniques are




under investigation.  The primary objective of most of these projects is




to develop more rapid, and hopefully less expensive, analytical method-




ologies for various POM.  Analytical techniques that can be easily




applied are needed so that any agency can routinely monitor and measure




BaP.  Some analytical methods which are under investigation include




projects 1, 2, 4, and 5 of Table X.




          7.1.2     Effects Investigation




     Project No. 11 has been delayed for about 1 1/2 years because




exposure-room furnishings have been impeded by equipment delays.




In the meantime, characterization of aerosols, design of containment




systems, sampling techniques, etc. are continuing so that the inhalation




study can be initiated immediately after the room is operable.  The




feasibility study listed as project 10 may be one of the more critical




in relating BaP  (and other selected POM) exposure to adverse human




health effects.  This project was begun in FY 1974.
                                   62

-------
      Table X.   APPLICABLE CURRENT POM RESEARCH  PROJECTS
Project
1.


2.



3.


4.


5.

6.


7.



0.

9.

10.




11.



12.


Development of fluorescent
detector for analysis of
carcinogenic HC
High pressure (up to
1,2000 psi ) liquid chro-
matographic determination
of total PAH
PAH profile in urban air
(ratios of various com-
pounds)
Rapid, direct gas chro-
matographic determination
of BaP
Rapid GC determination of
atmospheric PAH
Synthesis & purification
of carcinogenic air
standards
Characterization & con-
trol of air pollutant
emissions from fuel
combustion
Enclosed coke pushing &
quenching demonstration
Smokeless coke &
charging demonstration
Feasibil ity study:
Determine which tissues
best reflect PPOM expo-
sures. Develop methodo-
logy for tissue analysis.
Physiologic & pathologic
responses of animals to
inhalation of carcinogenic
hydrocarbons
Development of station-
ary source sampling
procedures for PNA
Date
begun
(FY)


1972



1972


1972


1972

1972


1972



1972

1972

1970




-



1973


1973
Scheduled
Completion


1974



1974


1974


1974

1973


1975



1977

1973

1973




1975



1978


1974
Estimate
of cost?
(51000)


47



3


31


54

55


28



200

800

950




200



500


34.4
Time
required
(Yrs.)


2



2


1.5


1.5

1


3



5

1

3




1



5


1
NOTES:
             from NERC/RTP Research Objective Accomplishment Plans (ROAP)
      and NERC/RTP Position Paper on PPOM (Draft), Dec. 1972.
      p
       Estimates of  costs for multi-year projects are for informational
      purposes  only;  they are not to be construed as accurate since an
      order of  magnitude could separate estimates & actual costs.   Single
      year projects  scheduled to end in 1973 should be nearly accurate,
      discounting extension of completion time.
                                   63

-------
          7.1.3     Stationary Sources




     Four projects (#7, 8, 9, & 12) involving BaP (and other POM) from




stationary sources were in progress during FY 1973.   Three of them were




initiated earlier by the Control Systems Laboratory.   The fourth (// 12) ,




which began the latter part of FY 1973, is a project to develop an




improved source sampling procedure for BaP (and other POM).   The initial




phase will be for about one year with options to explore any problems in




more detail if funds are made available.  Project 8 was delayed approximately




six months because of a job accident.  Fuels referred to in project 7




are such things as methanol, low BTU gas, etc.




          7.1.4     Ambient Analysis




     Data available under this grouping should be of primary interest




because ambient concentrations of POM compounds are ill-defined.  Project




3 is of major importance in achieving a better understanding of ambient




levels of BaP and other compounds.  The objective of this study is to




define, as accurately as possible in selected urban areas, specific




ratios of BaP to other POM compounds and vice versa.  Project 5 could




easily fall in this category also, because standard compounds must be




available for reliable analytical results.  A limited number (about 6)




of POM compounds were scheduled to be purified under the present contract.




7.2  Recommended Research Projects




     A study which should receive early emphasis is the development of




methodology to simplify and expedite ambient measurements.  Analytical




technique appears to be the primary area of need in expediting ambient




measurements.  The need for simplified analytical methodology was force-




fully emphasized during preparation of this report when additional




                                   64

-------
ambient BaP concentrations for 1971 and 1972 were required for trends




analysis in selected cities.  A significant time delay was entailed




because of time required to analyze for BaP.  Simplified analytical




methodology will expedite availability of ambient data for tracking




trends of BaP (and other POM) concentrations.  When new methodology is




developed, it should be more easily applicable by State and local




agencies also.
                                   65

-------
                           8.  REFERENCES

 1.    Andreuzzi,  Frank C.,  A Method  for Extinguishing and Removing Burning
      Coal Refuse Banks,  Bureau of Mines, U.S. Dept. of Interior, 1C
      8485, Washington, B.C., 1970.

 2.    Begemen,  Charles R.  and Joseph M. Colucci,  "Polynuclear Aromatic
      Hydrocarbon Emissions from  Automotive  Engines," Society of
      Atomotive Engineers,  Mid-year  Meeting, Detroit, Mich., May 18-22,
      1970.

 3.    Committee on Biologic Effects  of Atmospheric Pollutants, Particulate
      Polycyclic  Organic  Matter,  National Academy of Sciences, Washington,
      B.C., 1972.

 4.    Coordinating Research Council, Inc., CRC-APRAC Status Report,
      New York, N.Y.,  Jan.  1973.

 5.    Dixon, J.R., B.B. Lowe, B.E. Richards, L.J. Cralley, and H.E.
      Stokinger,  "The  Role of Trace  Metals in Chemical Carcinogenesis:
      Asbestos  Cancers,"  Cancer Research. 30. pp. 1068-1074, April 1970.

 6.    Duncan, L.J., "Analysis of  Final State Implementation Plans -
      Rules and Regulations," U.S. EPA Publication //APTB-1334, Research
      Triangle  Park, N.C.,  July 1972.

 7.    Esso Research and Engineering  Co., "Gasoline Composition and Vehicle
      Exhaust Gas Polynuclear Aromatic Content,"  2nd Annual Report, CRC-
      APRAC Project Cape-6-68, Linden, N.J., April 1971.

 8.    Hangebrauck, R.P.,  B.J. von Lehmden, and J.E. Meeker, Sources of
      Polynuclear Hydrocarbons in the Atmosphere, U.S. BHEW, Public
      Health Service Publication  No. AP-33,  Cincinnati, Ohio, 1967.

 9.    Intersociety Committee, Manual of Methods of Ambient Air Sampling
      and Analysis. Vol.  2, January  1970; also published as Health Laboratory
      Science,  Vol. 7, No.  1 (Supplement), American Public Health Association
      Albany, N.Y., Jan.  1970.

10.    McKay, Lewis M., Coal Refuse Fires, An Environmental Hazard.
      Bureau of Mines, U.S. Dept. of Interior, Information Circular 8515,
      Washington, D.C., 1971.

11.    Olsen, Douglas and  James L. Haynes, Preliminary Air Pollution
      Survey of Organic Carcinogens, Litton  Systems, Inc., NAPCA Publi-
      cation APTD 69-43,  Raleigh, N.C., October 1969.

                                   67

-------
12.   Rawlcki,  Eugene,  "Airborne Carcinogens  and  Allied  Compounds,"
      Archives  of Environmental Health,  Vol.  14,  pp.  46-53,  American
      Medical Assoc., Jan.  1967.
13.   Sawicki, Eugene, Walter C.  Elbert.  Thomas  R.  Hauser,  Francis  T.
      Fox, and Thomas W. Stanley, "Benzo(a)pyrene Content of  the  Air  of
      American Communities," AIHA Journal.  Vol.  21, No.  6,  pp.  443-451,
      Dec. 1960.

14.   Shabad, L.M.,  Y.L. Cohan,  A.P.  Ilnitsky, A. Ya.  Khesina,  N.P.
      Shcherbak,  and G.A. Smirnov,  "The Carcinogenic Hydrocarbon  Benzo
      (a)pyrene in the Soil,"  Journal  of  the National  Cancer  Institute,
      Vol. 47, No.  6, pp. 1179-1191,  Dec. 1971.

15.   Stenburg,  Robert L., Darryl J.  von Lehmden, and  Robert  P. Hangebrauck,
      "Sample Collection Techniques for Combustion Sources  -  Benzopyrene
      Determination," AIHA Journal, Vol.  22,  No. 4, pp.  271-275,  Aug.
      1961.

16.   Task Panel, National Environmental Research Center,  "Draft  NERC/RTP
      Position Paper on Particulate Polycyclic Organic Material (PPOM),"
      U.S. EPA, Research Triangle Park, N.C.,  December 1972.

17.   U.S. Bureau of Mines, Methods and Costs  of Coal  Refuse  Disposal and
      Reclamation,  U.S. Dept.  of Interior,  Information Circular 8576,
      Washington, D.C., 1973.

18.   U.S. Dept.  of  Health, Education, and  Welfare, Air Quality Criteria
      for Particulate Matter,  Public Health Service, NAPCA  Publication
      No. AP-49,  pp. 128-144,  Washington, D.C.,  Jan. 1969.

19.   U.S. Dept.  of  Health, Education, & Welfare, Occupational  Exposure
      to Coke Oven Emissions (Criteria for  a Recommended Standard...),
      Public Health Service NIOSH,  1973.

20.   U.S. Dept.  of  Health, Education, & Welfare, United States Metro-
      politan Mortality:  1959-1961,  Public Health Service  Publication,
      AP-39, pp.v-xi and 59-87,  Cincinnati, Ohio, 1967.

21.   Unpublished Contractor's report, Preliminary Results  by personal
      communication with the CSL staff, EPA Contract //68-02-0290, Feb.
      1974.

22.   Personal communication with Robert McCrillis, CSL, 2/12/74.

23.   Personal communication with R.  V. Hendricks, CSL,  based on  preliminary
      results from EPA Contract //CPA 70-165,  2/6/74.
                                    68

-------
24.    Faber,  Paul V.,  "Pipeline Coke-Oven Charging,"  Chemical  Engineering.
      Dec.  24,  1973,  pp.  36-37.

25.    E.  Allegrini,  "Cost of Retrofitting Coke Oven Particulate  Controls,"
      EPA Contract //68-02-0299, Vulcan-Cincinnati, Inc.,  Cincinnati,
      Ohio, 1974.

26.    Sebasta,  William, "Ferrous Metallurgical Processes,"  Ch. 36,  Air
      Pollution, 2nd  ed., Vol.  3, A.  C.  Stern, ed., Academic Press, New
      York, 1968, pp.  144-145.

27.    Shreve, R. N.,  Chemical Process Industries.  2nd ed.,  McGraw-Hill
      Book Co., Inc.,  New York, 1956, p.  90.

28.    TI.  S. EPA, 1972  National Emissions Report:   National  Emissions
      Data Systems (NEDS) of the Aerometric and Emissions Reporting
      System (AEROS).  EPA-450/2-74-012,  Research Triangle Park,  N.C.,
      June 1974.

29.    Brinkerhoff, Ronald J., "Inventory of Intermediate-Size  Incinerators
      in the United States - 1972",  Pollution Engineering,  5:11, November,
      1973, Greenwich, Conn. pp. 33-38.

30.    Personal Communication, John R. O'Connor, OAQPS, Cost Analysis  Branch,
      August 27, 1974.

31.    Cottrell, Ailleen,  "Annual Refining Survey", Oil &  Gas Journal,
      72:13, pp.82-84, April 1, 1974, Tulsa,  Okla.

32.    Gross, George P., "The Effect of Fuel and Vehicle Variables on
      Polynuclear Aromatic Hydrocarbon and Phenol  Emissions,"  Society of
      Automotive Engineers, presented at Automotive Engineering  Congress,
      Detroit, Mich.,  Jan. 10-14, 1972.

33.    U.  S. Bureau of Mines, Minerals Yearbook 1972;   Metals,  Minerals,
      and Fuels, Vol.  I,  pp. 427-437, U. S. Dept.  of  Interior, Washington,
      D.C., 1974.

34.    ,U.  S. Dept. of Interior Task Force to Study  Coal Waste Hazards,  "List
      of Coal Waste Banks," Internal Document, June 15, 1972.
                                    69

-------
APPENDICES
    71

-------
 APPENDIX A  -  OVERVIEW OF A  PREFERRED STANDARDS PATH








     Generally,  six major  options for control of air pollutants are




available to the Environmental Protection Agency (EPA) under the Clean




Air Amendments of 1970  (Act):   (1) National ambient air quality stand-




ards (NAAQS) - Sections  108, 109 and 110; (2) standards of performance for




new sources (NSPS)- Section 111;  (3) national emission standards for




hazardous air pollutants (NESHAP)— Section 112; and (4) Title II, Emis-




sion Standards for Moving  Sources  (ESFMS) -  Sections 202  (automobiles);




211 (regulation of fuels); and  231 (aircraft emissions).   A seventh




option for consideration,  if data evaluation indicates it, is that of  no




Federal action.   Other  options may be identified by subdividing or




combining the above seven  options.  The process of selecting the most




appropriate option is called a preferred standards path (PSP) analysis.




However, a PSP decision may not  be possible if adequate data are not




available for determining  the preferred option.  Thus, an eighth major




option becomes available:   postponement of a decision until more data




become available.  Of course, this latter option should carry with it




recommendations of what data are needed and a suggested priority for




obtaining such data.




     A.   Purpose of a  Preferred Standards Path




          The purpose of a PSP  is  to recommend a regulatory course of




          action to develop applicable standards resulting in control  of





                                 A-l

-------
a given pollutant.  Recommendations must be based on a thorough


state-of-the-art assessment of emissions and health effects


for the pollutant as related to objectives of the Act.  Basic


objectives of the Clean Air Act are described below.


1.   Protection of health and welfare by attainment and main-


     tenance of a designated numerical ambient standard:
                                               *

     This objective is applicable to those pollutants for


     which adverse effects levels (health or welfare) may be


     determined and that exist in excessive concentrations


     which are relatively easilv measureable in the ambient


     air.  Usually, this goal is achieveable by setting NAAQS


     following procedures set forth in Sections 108-110 of the


     Act.


2.   Enhancement of air qualitv:  This goal is for those


     pollutants which may be tolerated in exiguous quantities


     but their increase is limited in order to prevent the


     creation of new problems.  As related to stationary


     sources, this goal may be achieved best by applying


     Section 111 of the Act.  Title IT—Sections 202, 211,


     and 231—will be used for moving sources.


3.   Protection of life:  This objective is the ultimate


     purpose of the Act.  It applies to those pollutants that


     are present in the ambient air and must be reduced to


     safe levels.  CTenerally, these pollutants cannot be


     readily measured and concentrations must be reduced to


     a minimum (or possibly to zero if the air pollutant is



                        A-2

-------
          hazardous as defined in the Act) to achieve the goal of




          protecting life.  Thus, Section 112 of the Act would be




          applicable to such pollutants.




B.   Necessary Analysis for a Preferred Standards Path




     Information needed to develop a standard will depend to a




     large degree on the recommended control option, i.e., which




     section(s) of the Act is (are) to be used in attaining the




     prescribed objective.  A thorough evaluation which leads to




     such a recommendation is the heart of a PSP analysis.  There-




     fore, most items which should be considered in this type




     analysis are included in the following subsections.




     1.   Summary of the existing problem - Without an exhaustive




          knowledge of the existing situation no effective analysis




          can be undertaken. Specific factors to be examined in




          summarizing current conditions are described briefly




          herein.  yFigure A-l is provided as a guide in the logic




          of analyzing a candidate pollutant for determining what




          control strategy may be applicable.









          a.   Information on the pollutant's effects:  A quanti-




               tative criteria document is not required to deter-




               mine a pollutant's effects on human health and




               welfare.  However, health effects must be quantified




               for a candidate pollutant prior to implementing




               Sections 109 and  112 of the Act.  Both these sec-




               tions of the Act  provide general guidance to the





                             A-3

-------

•Adverse Health and
Welfare Ellecls: or
Possibility ol En-
dangerment of Health
and Welfare
3

Magnitude of Health and Welfare
Effects: Contribute to Increase
in Mortality, or Serious Irrever-
sible or Incapacitating Reversi-
ble Illness
No
7
Nature and C
of Sources:
ness - Wheth
or Diverse, S
Moving, or 6
No


istribution
Ubiquitous-
er Numerous
tationary or
oth

No

ACTION
2
No Action
Under Clean
Air Act
4 5-6
<• .n-rr, Adequate Data to Nature and Distri- No
See NOTE 1 Show Hazardous El- 	 1" 	 * bution of Sources- 1 »•
Yes lecls al c""ent Am- Ubiquitous?
bient Concentrations
No Optional
8
Is the Pollutant
Moving Sources? 12
u Can Control BP Adequate Data
n irhi.u=^ RU Available to Yes
— 1' ' Hcmeveo By 	 .N , *• ,.-, D.J , — — ••
Yes * NoAcL " f™«» *» U - "° Yes ^,~
- 3; Under ESFM5 . £_,„ rjont[0| Be and Welfare


Yei Kegulationi1 [7
9 | 	 ^_ 	 1 I^No

Is the Pollutant I "
*" Prominent From No ^
Stationary Sources? I'D " See Note 7
' 	 ' Adequate Data Available
Yes to Establish and Support yes
for Health and Welfare
Effects?
N° No^ Consider
1 -
1
Section 1 12, NESHAP"
Most Likely Appropriate*
13
Section 211, ROF,"
Most Likely Appropri-
ate*
15
Sections 202 and 231,
ESFMS, ** Most Likely
Appropriate*
,7
Section 109, NAAQS,"
Most Likely Appropri-
ate*
18
Section 111, NSPS,-
Mosl Likely Appro-
priate*
 NOTE 1:  A yes answer here simply means the pollutant is a candidate lor regulatory action; however, it does not denote mandatory action.
 NOTE 2:  The dashed line indicates that significant pollution from stationary sources may remain even after utilizing ESFMS. Land use and transportation
          controls may need to be implemented  in addition to Section 202.
 'MOST LIKELY APPROPRIATE simply means  that the indicated option may be the most  logical without considering external factors. However, other ramifications
  may preclude its use. Also, more than one option (e.g., Sections 109 and 202: Sections 109 and 111; etc.) may be used in combination For any given pollutant in
  order to achieve the lull intent of the Act. (Also see NOTES.) Section lllidl is applicable to non-criteria pollutants.
**Acronyms  explained  in  text.

            Figure  A-1 .   Preferred standards  path:  Guide for determination  of  regulatory action.

-------
     nature and severity of effects needed before action




     can be taken under either section.  Some differen-




     tiation between chronic and acute effects is needed.




     The objective is to estimate the likelihood of




     occurrence and severity of adverse effects.




b.   Ambient levels:  Without a good estimate of ambient




     concentrations of the candidate pollutant little




     appreciation for its health and/or welfare effects




     is attained.  An evaluation of ambient levels




     should determine whether the pollutant is ubiquitous




     or is related only to specific point sources.  If




     sources are numerous or diverse and cause adverse




     health effects at ambient levels then Section 109




     may be applicable.  If effects are caused by limited




     point sources Section 111 may be recommended.  Of




     course, Title II must be used for moving sources.




c.   Sources of the pollutant:  A precise emission




     inventory is not required; however, the following




     information is needed:   (1) an estimate of emissions




     from various sources, along with the probability of




     resulting ambient levels;  (2) the expected growth




     and/or decline of the number and type of source




     categories that emit  the pollutant; and  (3)  the




     number of individual  sources within a given  category.




     This information determines whether observed effects




     are caused by  localized  or numerous and  diverse





                   A-5

-------
          sources.   Contributions  from mobile sources  must be




          included  also.




     d.    Technology assessment:   An evaluation of the state-




          of-the-art in monitoring and control technology  must




          be available.  Monitoring technology includes ambient




          and source sampling instrumentation.   Control tech-




          nology includes the feasibility of applying  existing




          methods to various process configurations within an




          industry.   Furthermore,  this section should  evaluate




          the effectiveness of reducing emissions  beyond those




          levels achievable with  current control methods.




     e.    Environmental impact:   A PSP should consider the




          impact of the suggested  option upon the  environment.




          The completeness of this consideration may be dicta-




          ted by the option under  consideration.  For  example,




          if a NESHAP option is being studied, cost is pre-




          cluded as an item for consideration; health  effects




          are of primary  importance. Generally, this section




          should contain  such things as the accumulation of




          the pollutant in the environment, its effect on the




          fate of other pollutants, and other impacts.




2.    Summary of existing  regulatory efforts - In this  sub-




     section, the status  of existing regulatory efforts should




     be examined in relation to how the selected PSP would




     impact these regulations.  State, local, and  Federal




     control programs should be considered in order that




                        A-6

-------
     pertinent actions  for a new pollutant  would  not  be  con-


     tradictory,  or cause  some  additional obstacle  in en-


     forcing existing regulations.   These possibilities  may be


     considered in the  following two ways.


     a.    Existing State,  local, or Federal regulations  may


          achieve partial  control for a candidate pollutant.


          If so,  the degree of  control may  be determined by
                    *

          summarizing the  number and type regulations and  by


          determining their degree of enforcement.


     b.    Some control  of  the candidate pollutant may result


          from emission reductions of other pollutants,  voluntary


          controls by the  emitter, or Federal policy. Here,


          EPA must decide  whether this method yields  the


          desired control, or if the procedure results in  the


          best control.


3.    Summary of options available - The three options available


     for controlling air pollution from stationary  sources are


     described in greater  detail in this subsection.   As pre-


     viously stated, "no control" or "postponement  of a


     decision to control"  are also valid options  that may


     result from a PSP analysis.  One of the latter options


     may be chosen if based on existing evidence, an adverse


     effect cannot be shown to result from a candidate pol-


     lutant.  The Administrator of EPA has  such broad, dis-


     cretionary powers to reach a decision under the Act that


     more attention is given below to the options.



                        A-7

-------
a.   National Ambient Air Quality Standards:  This approach




     may be used if the candidate pollutant has "an




     adverse effect on public health and welfare" and




     ambient levels result "... from numerous or diverse




     mobile or stationary sources" [Section 108(a)(l)].




          A NAAQS includes a primary and secondary standard.




     Primary standards must be at levels which, "based on [air




     qualitv] criteria and allowing an adequate margin of




     safety, are requisite to protect the public health" [Sec-




     tion 109(b)(2)].




          Since ambient levels are specified, NAAQSs are not




     directly applicable to sources.  However, the attainment




     of an ambient standard may require source control.  States




     decide how to control source emissions in order to comply




     with the ambient standards.  Each State prepares an




     implementation plan (SIP), subject to approval by the




     Administrator [Section 110(a)(2)].  SIP regulations may




     utilize emission standards, land use policies, trans-




     portation control measures, or other approaches.  The




     State has wide leeway for controlling new and existing




     sources except for moving sources, which are specifically




     exempted [Section 116].  If a State submits an unaccept-




     able plan, EPA is required to promulgate regulations that




     will achieve the standard(s) [Section 110(c)].  Likewise,




     if a State does not enforce its regulations this respon-




     sibility reverts to the Federal government [Section






     113(8)1'           A-g

-------
     The milestones prescribed by the Act for NAAQS are more complex

than for the other options and thus can result in longer delays before

control is achieved.  Significant milestones are listed below.

          MILESTONE                               ACTION
    1.  Start


    2.  Within 12 months


    3.  Within 3 more months
    4.  Within 9 more months
    5.  Within 4 more months
    6.  Within 2 more months
     7.  Within  3 more years
                            (4)
Inclusion of candidate
 pollutant on Section 108
 list.
Issue air quality criteria
 and proposed standards
 (primary and secondary).

Promulgate standards.
Submit SIP's

Approval/disapproval of
 SIP.  If-not submitted or
 disapproved, prompt proposal
 of plan by EPA.O)
Promulgate EPA plan if SIP
 remains inadequate.
Attainment of primary(5)
 standard.
      The Act  delineated  deadlines for initiating  the regulatory process
in each  option (including the other two described below) for certain
pollutants and sources  (e.g., the six pollutants with existing air
quality  criteria documents, Group I NSPS, etc.).   It also provided  for
subsequent inclusion of pollutants on relevant lists "from time to
time."   Once begun, however, the statutory  timetable binds EPA to the
described milestones.
      2
      The Administrator may extend the deadline for submitting SIP's for
secondary standards an additional 18 months.
      3
      If a State fails to revise an existing SIP within 60 days, or a
longer prescribed period,  upon notification by the Administrator, EPA
revisions are  to be proposed promptly.

      4
      If adequate technology is unavailable, an extension of up to  two

years may be granted by the Administrator for the  attainment of a
primary  standard.

      Secondary standards must be attained  within  a reasonable time as
specified within the SIP.

                                  A-9

-------
b.   New Source Performance Standards:  This option mnv




     he used if a category of sources "mav contribute




     significantly to air pollution which causes or




     contributes to the endangerment of public health or




     welfare"  [Section lll(b)(A)].  National standards




     established under Section 111 are applicable to (])




     new stationary sources for specific categories and




     (2) modifications resulting in new emissions of air




     pollution.  The Administrator can distinguish among




     classes,  types, and sizes within categories.  Section




     lll(d)(l) requires States to control existing sources




     covered by NSPS applicable to the specified category




     unless the pollutant is covered under NAAQS or




     NESHAP.   If the pollutant is covered bv Section




     108(a),  NAAQS,  the  States  can prevent  construction




     or modification of stationary sources which may




     interfere with attainment or maintenance of a NAAQS




     [Section  110(a)(A)]. With proper enforcement, SIP's




     control existing sources as necessary to attain and




     maintain  the NAAQS.  Moreover, if a pollutant has




     been designated as hazardous the standard will apply




     to all designated sources—existing, modified, and




     new.




          The  level of the standard "reflects the degree of




     emission  limitation achievable through  the  application of




     the best  system of  emission  reduction which (taking into







                    A-10

-------
               account the cost of achieving such reduction) the Adminis-

               trator determines has been adequately demonstrated"

               [Section lll(a)(l)].  Thus, the degree of control (which

               likely will not be more stringent than the standard) is

               not related directly to health and/or welfare.  States

               also may use best demonstrated control methods in

               setting standards for existing sources.

                    Enforcement of NSPS is a Federal responsibility

               initially  [Section 113(a)(3)] but may be delegated to

               States if  acceptable enforcement procedures are adopted

               [Section lll(c)(l)].  In addition, each state must

               adopt plans to regulate unmodified existing sources of

               the same category, provided the pollutant is not a

               criteria or hazardous pollutant.  These plans are

               similar to procedures required to achieve NAAQS  [Section

               lll(d)].

                    Milestones for NSPS do not allow for time extension

               as under NAAQS.  Milestones as given in the Act follow.


                       MILESTONE                 ACTION

               1.  Start                    Include category of
                                             sources on Section
                                             111 list.
               2.  Within 120 days          Issue proposed standards.

               3.  Within 90 more days      Promulgate standards.
                                             Effective upon promulgation.
     6
      Presently no guidance is available to states in meeting specific
deadlines for controlling emissions from unmodified existing sources for
a given category.  However, such guidelines have been proposed.


                                 A-ll

-------
c.    National Emission Standards for Hazardous Air Pollutants:




     This approach may be used if the pollutant "mav




     cause, or contribute to, an Increase In mortality or




     an increase in serious irreversible, or incapacitating




     reversible, illness" [Section 112(a)(l)].  Standards




     under Section 112 are applicable to new and existing




     stationary sources, with the provision of a 90 day




     delay for existing sources [Section 112(c)(l)].




     Authority to make distinctions or classifications,




     as in NSPS, is subject to question.  These emission




     standards are established at a level which "provides




     an ample margin of safety to protect the public




     health from such hazardous air pollutant" [Section




     112(b)(1)(B)].  Technology or cost of control can




     not be considered, except perhaps in deciding what




     margin of safety is ample.  In order to assure an




     ample margin of safety the worst case expected in




     practice should be considered. This will involve




     reviewing the total number of sources, ambient




     concentrations, persistence of the pollutant, build-




     up in body tissue, etc.  If an ample margin of




     safety, based on review of the above factors, re-




     quired zero emissions a total ban could be accomplished




     under Section 112.  (A similar result may be possible




     under Sections 108-110 if the ambient standards were
                   A-12

-------
                set low enough.  Zero emissions can be achieved

                under Section 111 if technology to achieve 100 per

                cent control is adequately demonstrated.)  Enforce-

                ment of NESHAP is a Federal responsibility [Section

                113(a)(3)], but the Administrator may delegate author-

                ity to States under certain conditions [Section 112

                (d)].  However, this does not preclude the Federal

                enforcement prerogative.  Again, provisions for time

                extensions are granted in the Act  [Section 112(c)(l)

                (B); (c)(2)].  Milestones outlined for NESHAP result

                in the second quickest means for attainment of standard,

                However, with provision for extension and exemption

                for two-year periods, no definite  time for attainment

                of standards can be estimated.  Milestones designated

                in the Act are as follows:

                     MILESTONE                ACTION
                1.  Start            Include candidate pollutant
                                      on Section  112  list.
                2.  Within 180 days  Issue proposed standards.

                3.  Within 180 more  Promulgate standards.  Effee-
                     days             tive for new source upon
                                      promulgation, and  90 days
                                      later for existing sources.


C.   Discussion of Various Options

     Under this section the ramifications and  reasoning for choos-

     ing a specific option should be given.   Support for the  recom-

     mended control option as decided by the  PSP analysis must be

     substantiated by items discussed in this  section.  In other


                                  A-13

-------
     words, a detailed discussion of factors outlined in "B" above




     must support the chosen control option.  Insight into legal or




     other challenges should be given in detail.   Except for those




     cases where prohibited by statute (e.g., Section 112) the




     economic impact for each option should be included.




D.   Recommendations




     The objective of this Section is to succinctly state the




     recommended control option and develop an accomplishment plan




     to implement the recommended approach.
                             A-14

-------
  APPENDIX B - LUNG CANtER MORTALITY  IN SELECTED  SMSAs





     Table  B-l  is a ranking of Standard Metropolitan Statistical Areas




(SMSAs)  by  descending order of deaths from lung  diseases.  In comparing




the principal cities of the SMSA (column 2)  in this table with sampling



sites of Table  C-l in Appendix C, one may see  that the majority of those



cities with relatively high lung cancer deaths have not been sampled for



BaP in the  NASN.




     A definitive program to sample and analyze  the ambient air for BaP




in selected cities where high lung cancer incidence is established needs



to be undertaken.  This should be considered a first step in defining




the relationship between the urban factor and  carcinogens, specifically



BaP or other POM.
                                 B-l

-------
 Table B-l.  WHITE  MALE DEATHS AND DEATH  RATES PER 100,000 POPULATION
PER YEAR BY SMSA* FOR MALIGNANT NEOPLASM  OF  TRACHEA, BRONCHUS, AND  LUNG
                                1959-1961**
Rank SMSA
1 Charleston, S. C.
i? Albany, Ga.
3 Galveston, Texas
4 Lake Charles, La.
5 New Orleans, La.
6 Newport News, Va.
7 Montgomery, Ala.
8 Jersey City, N.J.
9 Shreveport, La.
10 Baton Rouge, La.
11 Norfolk, Va.
12 Jacksonville, Via.
13 Birmingham, Ala.
14 Dubuque, Iowa
15 Baltimore, Md.
16 Jackson, Mich.
17 Tyler, Texas
18 Houston, Texas
19 Monroe, La.
20 Mobile, Ala.
21 Pensacola, Fla.
22 Charleston, W. Va.
23 Honolulu, Hawaii
24 Miami, Fla.
25 Portland, Me.
26 Toledo, Ohio
Death
rate
65
64
63
63
61
59
58
56
56
55
55
54
53
53
52
52
52
51
51
50
50
49
49
49
49
49
Total
deaths
75
23
79
65
468
85
61
556
121
74
214
206
278
55
945
92
49
536
44
111
80
141
55
683
134
313
Rank
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
, 43
1 44
i
! 46
:' 47
48
49
50
51

SMSA
Albany-Schen., N.Y.
Beaumont, Texas
Memphis, Tenn.
Omaha, Neb. -la.
Richmond, Va.
Buffalo, N.Y.
Decatur, 111.
Savannah, Ga.
Dallas, Texas
Des Moines, Iowa
Philadelphia, Pa. -N.J.
St. Louis, Mo. -111.
Atlanta, Ga.
Cincinnati , Ohio
Erie, Pa.
Flint, Mich.
New York, N.Y.
Paterson, N.J.
Syracuse, N.Y.
Tampa, Fla.
Topeka, Kansas
Washington, D.C.
W. Palm Beach, Fla.
Wheeling, W.Va.-Ohio
Wilmington, Del .

Death
:ate
48
48
48
48
48
47
47
47
46
46
46
46
45
45
45
45
45
45
45
45
45
45
45
45
45

Total
Deaths
492
134
214
280
184
831
75
61
495
162
2482
1178
377
610
155
185
7025
752
354
682
79
723
160
141
186

 *Standard Metropolitan
 **Reference 20.
Statistical  Area.
                                    B-2

-------
      APPENDIX  C -  NATIONAL AIR SURVEILLANCE NETWORK




       AMBIENT  AIR MEASUREMENTS FOR BENZO(a)PYRENE




     Data given in this appendix are arithmetic averages of the four




quarterly composites of the listed urban sampling site.  The annual



average is computed only if 3 or more quarterly composite values exceed




the minimum detectable concentration of 0.2 nanograms/cubic meter (ng/m^).



     Table C-l contains annual averages for 121 selected NASN urban



sampling sites.   These sites were selected  from the total NASN system



based on valid data for each site for all three years.  Each site had to



meet minimum requirements of the NASN for each quarter in order to have




a valid annual average for the given site.   In order  to be included in




the summary, each sampling station must have a minimum of five scheduled



samples collected and analyzed for any given quarter.




     Table C-2 is a listing of 40 sites selected to update BaP con-




centrations in cities with and without coke ovens.  Three sites were



selected in National Parks so nonurban background readings would be



available.  A limited number (40) was chosen because  of time and re-



source restrictions since routine analysis  of NASN BaP was discontinued



in 1970.



     Table C-3 gives valid annual averages  of BaP for the sites listed



in Table C-2 for  all years of record (1966-1972).  Numbers in paren-



thesis are the number of valid sites included for the year.
                                 C-l

-------
Table C-l.   NATIONAL AIR SURVEILLANCE NETWORK AMBIENT AIR MEASUREMENTS
	 1 	
Location,
state and cjty
Alabama
Gadsden
Hunts vi lie
Montgomery
Alaska
Anchorage
Arizona







Gra/id Canyon National Park
Maricopa County
Tucson
Arkansas
Little Rock




Montgomery County
West Memphis
California
Glendale
Humboldt County
Long Beach
Los Angeles
Oakland
Riverside
Sacramento
San Bernardino
San Diego
San Francisco
Colorado
Denver
Connecticut
Hartford
New Haven
Florida
Jacksonville
Tampa
Georgia
Atlanta
Hawaii
Honolulu
Idaho
Boise City
Butte County



























Benzo(a)pyrene [ng/m3(25°C)]
1968

2.37
2.74
2.93

1.68

0.19
0.47
0.67

0.89
0.23
2.24

1.55
0.31
2.09
1.83
1.62
1.29
1.43
1.00
1.22
1.83

2.25

1.42
1.38

2.94
1.46

1.83

0.55

2.01
0.17
1969

1.76
1.82
2.04

1.28

0.18
0.27
0.53

1.05
0.18
2.41

1.63
0.49
2.27
1.86
1.61
0.82
1.79
0.90
1.38
1.15

2.51

1.95
2.10

2.30
1.00

1.86

0.57

5.96
0.09
1970

2.50
1.56
1.31

0.77

0.10
0.28
0.41

0.65
0.13
0.59

0.97
0.11
1.00
1.23
0.95
0.67
0.72
0.83
0.65
0.63

2.20

1.37
1.21

1.35
0.46

0.92

0.19

1.14
0.07
                                  C-2

-------
Table C-l (continued).   NATIONAL AIR SURVEILLANCE NETWORK AMBIENT AIR
    MEASUREMENTS FOR BENZO(a)PYRENEa, 121  SELECTED SITES, 1968-1970
Location,
state and city
Illinois
Chicago
Springfield
Indiana
East Chicago
Hammond
Indianapolis
Monroe County
Parke County
Iowa
Des Moines
Kansas
Topeka
Wichita
Kentucky
Ashland
Covington
Louisiana
New Orleans
Maine
Acadia National Park
Maryland
Baltimore
Massachusetts
Worchester
Michigan
Detroit
Flint
Grand Rapids
Trenton
Minnesota
Duluth
Minneapolis
Moorhead
St. Paul
Nebraska
Omaha
Thomas County
Benzo(a)pyrene [ng/m3(25°C)]
1968
3.10
1.06
4.90
2.08
4.07
0.54
0.43
1.12
0.69
1.04
9.30
3.56
1.55
0.33
2.31
1.65
5.13
0.78
3.44
1.44
2.73
1.14
0.85
1.82
1.91
0.20
1969
3.88
1.27
6.75
3.29
5.18
0.25
0.26
0.92
0.44
0.67
10.89
4.11
1.52
1.12
2.76
1.48
3.91
1.69
1.70
1.58
2.08
1.43
1.04
1.75
1.55
0.13
1970
2.00
0.85
5.26
1.67
2.32
0.16
0.39
0.69
0.31
0.50
6.67
4.38
1.14
0.20
2.06
1.63
2.56
1.49
0.87
0.84
1.09
0.62
1.59
1.01
1.01
0.12
                                  C-3

-------
Table C-l (continued).  NATIONAL AIR SURVEILLANCE NETWORK AMBIENT AIR
    MEASUREMENTS FOR BENZO(a)PYRENEa? 121  SELECTED SITES, 1968-1970
Location,
state and city
Nevada
White Pine County
New Hampshire
Concord
Coos County
New Jersey
Camden
Glassboro
Jersey City
Newark
Paterson
Perth Amboy
Trenton
New Mexico
Albequerque
New York
• Jefferson County
North Carolina
Cape Hatteras National Park
Charlotte
Durham
North Dakota
Bismarck
Ohio
Cincinnati
Cleveland
Columbus
Dayton
Toledo
Youngstown
Oklahoma
Cherokee County
Oklahoma City
Tulsa
Oregon
Curry County
Portland
Benzo(a)pyrene [ng/m3(25°C)J
1968

0.14

0.98
0.20

1.57
1.20
2.35
2.14
1.95
1.17
1.04

1.77

0.19

.22
5.56
7.96

.86

1.77
3.00
2.21
2.36
1.80
5.64

.21
.73
.77

.14
4.13
1969

0.07

0.65
0.11

2.41
1.09
2.68
1.82
1.24
1.20
1.46

1.12

0.25

.11
4.85
3.38

.96

2.90
3.75
2.73
1.88
1.49
9.86

.18
.71
.47

.08
2.60
1970

0.12

0.61
0.14

1.92
1.18
4.65
1.53
1.20
1.01
1.10

1.06

0.24

.21
1.85
3.38

.44

2.58
2.78
1.57
1.48
1.38
7.12

.22
.88
.79

.09
2.31
                                  C-4

-------
Table C-l (continued).  NATIONAL AIR SURVEILLANCE NETWORK AMBIENT AIR
    MEASUREMENTS FOR BENZO(a)PYRENEa, 121 SELECTED SITES, 1968-1970
Location,
state and city
Pennsylvania
All en town
Altoona
Bethlehen
Clarion County
Harrisburg
Philadelphia
Pittsburgh
Reading
Scranton
Wilkes-Barre
York
Rhode Island
East Providence
Providence
South Carolina
Greenville
Tennessee
Chattanooga
Memphis
Nashville
Texas
Matagorda County
San Antonio
Utah
Ogden
Salt Lake City
Vermont
Burlington
Orange County
Virginia
Danville
Hampton
Lynchburg
Norfolk
Portsmouth
Roanoke
Shenandoah National Park
Benzo(a)pyrene[ng/m3(25°C)]
1968

1.17
17.96
2.05
.97
1.32
2.87
6.31
2.36
6.08
1.58
1.91

1.17
1.95

18.55

7.39
1.34
5.96

.16
.85

.82
.97

.71
.31

2.47
1.53
8.72
4.90
10.17
7.65
.30
1969

1.92
22.28
2.00
1.23
1.45
4.03
13.75
1.75
7.65
1.54
2.00

1.21
2.15

7.00

4.17
.74
2.80

.12
.63

.67 .
.65

.48
.28

1.79
.88
6.28
3.91
3.39
5.30
.31
1970

2.40
19.25
2.71
1.23
1.53
2.44
5.86
1.56
2.88
1.30
1.21

1.22
2.11

3.39

5.54
1.36
3.61

.27
.95

2.49
1.44

.74
.16

2.69
1.08
4.49
1.67
4.94
6.20
.21
                                 C-5

-------
Table C-l (continued).  NATIONAL AIR SURVEILLANCE NETWORK AMBIENT AIR
    MEASUREMENTS FOR BENZO(a)PYRENEa, 121 SELECTED SITES, 1968-1970
Location,
state and city
Washington
Seattle
West Virginia
Charleston
Wisconsin
Kenosha
Madison
Milwaukee
Superior
Wyoming
Casper
Cheyenne
Benzo(a)pyrene[ng/m3 (25°C)]
1968
1.97
4.57
1.41
1.33
4.65
3.30
.90
.60
1969
1.57
2.63
1.74
9.62
4.03
1.58
.60
.46
1970
1.51
2.11
1.31
1.07
2.46
1.50
.44
.43
   Annual  average.
                 '-' /-
                                   C-6

-------
Table C-2.   LISTING OF 40 NASN SITES SELECTED FOR 1971-72  BaP  ANALYSIS

*1.
2.
3.
4.
5.
6.
7.

8.
9.
10.
11.
*12.

13.
14.
15.
Coke oven
Birmingham
Gadsden
Chicago
Gary
Indianapolis
Terre Haute
Ashland

Baltimore
Dearborn
Detroit
Trenton
Duluth

St. Paul
St. Louis
Buffalo
ci
16
17
18
19
*20
21
22

23
24
25
26





ties
. Cleveland
Toledo
Youngstown
. Bethlehem
Erie
. Philadelphia
. Pittsburgh

Chattanooga
. Houston
. Spokane
. Milwaukee





Noncoke oven cities
1.
2.
3.
4.
5.
6.
*7.


8.
9.
10.
*11.

12.
13.
*14.
Montgomery
Jacksonville
Honolulu
Hammong
Baton Rouge
New Orleans
Acadia National
Park
( a 1 1\
New York
Newport News
Norfolk
Shenandoah Nat'l .
n _ ._ i .
Hark
Seattle
Charleston, W. Va.
Grand Canyon
  *Not included in Figure 5.   Birmingham
   data; Duluth had known extraneous infl
and Erie had insufficient
uences in 1971  and 1972.
          Table C-3.   ANNUAL BaP AVERAGES  FOR SELECTED CITIES

                               1966-1972
Year
1966
1967
1968
1969
1970
1971
1972
Coke
4.74 (15)*
5.34 (15)
3.75 (18)
4.41 (23)
3.02 (21)
2.18 (11)
2.14 (19)
Non-coke
2.76 ( 7)
2.29 ( 8)
2.64 ( 8)
2.14 (11)
1.41 (11)
1.22 ( 8)
0.64 (11)
          *Number of cities included in average (no.  with
           full  year's valid data).
                                  c-7

-------
 APPENDIX D  -  DESCRIPTION OF BY-PRODUCT COKE  PRODUCTION



     Coke is the carbon residue  of certain grades of bituminous coal



after destructive distillation.   It is  used as a fuel, primarily in



making pig iron which is an essential ingredient in steel.  Destructive



distillation (coking) is accomplished principally in by-product ovens



and consists of driving certain  volatile matter off coal, leaving a



residue with a high percentage of carbon and relatively small amounts of



impurities.



     A brief description of the  by-product coking process is presented



here as extracted from reference 26 of  the text.  By-product ovens are



usuallv constructed in groups called batteries.  They consist of a block



of many long, narrow firebrick ovens with heating chambers made of



similar brick located between the ovens, so that a battery of ovens is a



huge block of coking cells separated from an intricate system of combustion



chambers  (Figure D-l).  The fuel used  to heat  coke ovens may be blast   <-
                                                                          i


furnace gas, coke oven gas, or natural  gas, each fuel requiring a different



set of burner adjustments.  None of the coal being coked is burned to



provide heat for the coking operation.



     The  carbonization of the coal begins soon after crushed and sized



coal is loaded into red-hot ovens.  The gaseous products and condensates



are conveyed continuously via the large collecting mains from the coke



oven battery to an adjacent by-product  plant.  Usually each oven has a



steam jet aspirator which aids in conveying the  gaseous  carbonization


products  into the collecting main.



                                  D-l

-------
V
ro
             CROSSOVER FLUE
                                                                                  CONTROL
                                                                                   HOUSE
                                                          /^r-S
                              T=?
SECTION THROUGH
   AIR PORTS
                                                                                                                7  BLAST
                                                                                                                ^ FURNACE
                                                                                                                  GAS MAIN
          I ONG SECTION THROUGH   LONG SECTION      TRANSVERSE SECTION THROUGH
          INSPECTION HOLES AND THROUGH CROSSOVER        HEATING FLUES AND
              WASTE-GAS        AND AIR PORTS           UNDERJET GAS-DUCTS
          «£CIRCULATING DUCTS.
                        TRANSVERSE SECTION THROUGH
                        AN OVEN AND REGENERATOR
               Figure D-1.  Koppers-Becker underjet low-differential combination coke oven  with waste-gas recirculation
               (Courtesy of Koppers Company, Inc.)2?

-------
     The charging holes on the top of the battery are closed almost




immediately after charging to minimize escape of gas or dust into the




atmosphere.  Spacing of the charging holes, and sleeves on the charging




car hoppers, have been used to minimize such emissions.  Smoothness of




operation and the physical characteristics of the coal being used aid in




the reduction of these emissions.  Most operators attempt, through door




maintenance programs and operating procedures, to further reduce the




escape of gas through the charging holes.  Various methods of control-




ling particulate and gaseous emissions during charging have been inves-




tigated. More widespread application of these methods has occurred in




Europe and Japan than in the U. S.  However, recent attempts at application




have been undertaken by several U. S. companies, with varying degrees of




success.




     In the by-product plant, tar, benzene, naphtha, and other com-




mercial products are segregated and separately removed from the coke




oven gas. The by-product recovery plant has processes similar to those




in petroleum refineries and chemical plants.  The cleaned gas, having a




medium heating value, is customarily used as fuel in boilers and other




furnaces, including those of the coke oven.  Since sulfur has an unde-




sirable influence in steelmaking, the steel industry has always en-




deavored to use materials with low sulfur content.  Therefore, low




sulfur coal is used for coke making.  Coke oven gas usually contains




some sulfur-bearing compounds derived from the sulfur in the coal.




     After the coking period, the incandescent coke is pushed from the




oven into a quenching car where burning in air takes place.  This car is




a large special-type railroad car with perforated sides, which is used





                                  D-3

-------
to carry the flaming coke to an area where a huge quantity of water is




sprayed onto the coke to quench it, i.e., extinguish its incandescence.




During this operation, which stops the burning of the coke, large quanti-




ties of steam are generated.  Hot coke extinguishing is usually performed




in a quenching tower.  The quiescent coke is then conveyed to a cooling




wharf where the coke is spread out so that the excess water may drain




away.  After cooling, the coke is sized in a plant similar to that for




screening rock and ores.  As with the charging phase of coke production,




various attempts at controlling emissions from the pushing of finished




coke have been attempted with various degrees of success.
                                  D-4

-------
                APPENDIX E  -  COSTS OF CONTROL
           AND GROWTH PATTERNS FOR COKE OVENS
     Most polycyclic  organic matter (POM) is emitted  when coal  is
charged to coke  ovens, during the coking process, or  when the finished
coke is pushed out  of the  oven.
Charging Controls
     The charging process  is not generally controlled at present. Two
possibilities  for control  are described.  One is to equip the larry car
(vehicle that  dumps coal into the ovens) to prevent escape of fumes.
This process  is  now in the demonstration stage but estimates indicate
that the capital cost for  retrofit of an existing 70  oven battery would
approximate $1,500,000.  Several things could increase this cost:   (1)
the necessity  to add  additional vents to each oven; (2)  adding  holes in
the roof for vents; and  (3) strengthening the batterv structure if
necessary to bear  the weight of  the new larry car, which weighs 100 tons
compared to 50 tons for  a  non-controlled car. Operating costs have  not
been compiled  as yet. This type of information is expected to  be
available by the end  of  CY 1974.
     Another control  method already in use is pipeline charging. This
method eliminates  the larry car  and the dumping cycle.  The entire
charging operation  is confined within a pipe.  For a  new 70 oven battery
costing about  $24,000,000, the control equipment can  be added for
$8,000,000.  However, since the  coke is preheated in  the process, claims
are made that  the production capacity of the battery  is increased by
50%.  To obtain  the same capacity without pipeline charging would cost
                                 E-l

-------
$36,000,000.  Therefore, the control system results in a 10% capital

                         i
saving at the greater production figure.

     Retrofit costs will run $12 to $14 million for a 70 oven battery.

However, if a market is assumed for the additional coke produced,


installation of control equipment would cost about the same as adding


ovens.  Claims are also made that operating savings from the use of a


higher percentage of cheaper, high volatile coal; the increase in yield


of blast furnace coke; and the shorter cycles would more than compensate

for the operating costs involved.

Coking Controls

     Gases generated during the coking process are vented to a by-

products plant and thus are not released to the air.  However, con-


siderable leakage often occurs around door seals.  At present the only

remedy is careful operation and adequate maintenance, including cleaning

the seals at each cycle.


Oven Discharge Controls

     Although most of the POM evolves during the coking cycle, it is

possible that some may evolve during the discharge cycle.  One company

is testing an indexing hood connected to a scrubber to control parti-

culates.  The company is continuing its tests of this control method for

charging and discharging.  At this stage of testing, insufficient data

are available to judge the adequacy of performance or to accurately

estimate costs.  Company representatives estimated total costs would

approximate $10,000,000 to retrofit their batteries that contain a total

of about 220 slots.
                                  E-2

-------
Cost Summary

     Subject to reservations stated above, the tables below detail
                      25
retrofit control costs   for a 70 oven battery.  These costs are estimates

only.
                        Table E-l.   CAPITAL COSTS
                               (in $1,000)
              Pipeline charging
                Material and field costs           $11,858
                Modification to ovens and              546
                  installation of charging
                  main
                Coordination and start-up              388
                Subtotal                           $12,792
              Pushing
                Material and field costs             1,619
                Overhead and profit                     26
                Coordination and start-up               35
                Subtotal	$ 1.680
              Total                                $14,472
                                  E-3

-------
                       Table E-2.  ANNUAL COSTS
                             (in $1,000)
             Pipeline charging
               Maintenance at 6% Inv.              $  768
               Utilities                              318
               Labor                                  229
               Taxes, Ins. at 2% Inv.                 256
               Interest (8%)                        1,023
               Depreciation (15 yrs)                  853
               Subtotal$3,447
             Pushing
               Maintenance at 6% Inv.                 101
               Utilities                               92
               Labor                                    0
               Taxes and  Ins. at 2%  Inv.               34
               Interest (8%)                          134
               Depreciation (15 yrs)                  112
               Subtotal	$  473
             Total                                 $3,920

     Assuming that a 70 oven coke battery produces 630,000 tons of
coke annually,  the average cost per year to achieve about 85-90% BaP
control is approximately $6.25 per ton of coke.   This  compares with
                                                                   33
the cost of coke at $40.70 per ton in 1972 (latest published data),
or about 15% of the market value.  This annualized cost  of retrofit
is significantly greater than that for a new coke battery installing
pipeline charging, where the estimated cost is about $0.10 per ton
of coke.
     Capital costs for retrofitting a battery with a modified larry car
                                                               21
have been estimated at 1/5 to 1/10 those for pipeline charging.    However,
operating costs have been projected to be much greater,  even to the extent
                                  E-4

-------
of accounting for the difference in capital costs.  These operating costs


are only estimates and are not based on operational data because the first


phase of the demonstration project has just been completed.  Additional


time will be required before operating costs can be obtained.  Therefore,


no definitive conclusion can be stated for costs at this time.  Neverthe-


less, if the above cost estimates are reasonable one can immediately see


the large differences between retrofitting coke ovens and installing


controls on a new battery.


Growth Patterns


     Despite the historic growth of the steel industry, the consumption


of coke per ton of steel poured continues to drop.  In 1960, 1542 pounds


of coke were used per ton of steel; by 1970 the amount of coke dropped


to 1248 pounds; and by 1980 a further drop to 1100 pounds is projected.


Thus, even though hot metal production is expected to increase at a


normal rate, coke production will not change appreciably.



     However, because of air pollution regulations and obsolescence many


existing batteries will have to be replaced.  The table below was taken

                                                      3?y
from a bv-product coke oven survey of Koppers Company.
                                  E-5

-------
Table E-3.   SUMMARY OF PROJECTED NEW OVEN CONSTRUCTION
                     UNITED STATES
Estimated date
for operation
1974
1975
1976
1977
1978
1979
1980
Number
of
batteries
2
5
7
5
3
2
5
Number
of
ovens
139
386
455
326
196
109
465
Annual
capacity
total coke
(1000 tons)
1,881
4,440
5,260
3,470
2,370
1,465
6,230
New coke
capacity
(1000 tons)
457
1,445
810
700
270
—
—
                             E-6

-------
      APPENDIX F - EMISSION  ESTIMATES FOR COKE OVENS


     Emission data  for  BaP  from coke ovens are limited.  Estimates of


emissions which have  been made from available data vary extensively,  e.g.,


from 0.06 to 166 tons per year.   Presently, crude sample data from


uncontrolled plants in  the  United States are available.  Some data

                                                           4
have also been reported from the  U.S.S.R and Czechoslovakia.


Demonstration projects  are  in progress at several plants within the U.S.


Data on BaP emissions from  uncontrolled and controlled sources are being


gathered bv contractors;  these data should be available soon.  Until


these data are available, the following calculations, based on crude


data, are presented to  demonstrate the method of arriving at the


estimate contained  in the report. Total tons of by-product coke produced


in the U.S. are used  in the calculations because total nationwide


emissions are of interest.   However, beehive oven production is excluded


because in 1972 it  accounted for  approximately only 1% of the U.S. total.


All units are in U.S. net tons.



     A.  Estimates Based on U.S.  Data:


         1.  Given:


             a.  Ratio of BaP to Total Particulates


                   = 1.734 x 10~3


                 (Reference 1,  Smith)





             b.  Through-put loss of  particulates  in  the coking


                 operation    = 1.1  x 10   tons  of  particulate

                                  F-l

-------
              per ton of coal charged  (minus unloading and


              quenching)


              (Reference 2, AP-42)


         c.  Quantity of coal consumed per  ton of coke


             produced    3  1.45


              (Reference 3,  Bureau  of Mines)





         d.  Tons of coke produced in  1972    =  59.9 x 106


             Tons produced  at furnace  plants    = 54.2 x  10°


              (Reference 3,  Bureau  of Mines)





      2.  Find:


              Estimated BaP emissions  from  Coke  ovens
(l.734xlO-3  - tonJteP - Vi.lxio-3  ton particulateX
V              ton particulate/\              ton of coal  /

       tons. of coal\ / q       Q6  6  tons  coke\
                   M               year    )
               t     tons of coke


               166  tons of BaP/yr.
B.  Estimate Based on Soviet Data:



       1.   Given:


           Ratio of BaP/Total Particulates  = 1.05 x 10~3


           (This value is the average of the maximum ratios


           of two Soviet coal-coke plants.)


           (Reference 4, Masek, Table 2)
                                F-2

-------
      2.  Find:




               Estimated BaP emissions for U.S. coke ovens




               assuming B.I is the accurate ratio.
           ^1.05  x 10~3 \Lff  tons of BaP\   in.  ..     , ., D/
           	:H[166  	J  =101 tons of BaP/yr,

           V1.734 x 10~V\       year
C.  Estimate Based on Czechoslovak Data:





       1 .
           Ratio of BaP/Total Particulates    = 6 x 10




           (This estimate is the average value of BaP from



           dust on the upper floor of the battery.)




           (Reference 4, Table 5)





       2.  Find:




                Estimated BaP emissions from U.S. coke




                plants assuming C.I is the accurate ratio.
              6  x 10~    \/,,-/•   tons of BaP \   n », ..    c „ „/
                         1166   	 I "0.06 ton of BaP/yr.
                         1         «• A *•»•»*     I                   J
           a.734  x 10
                      -3/1         year
                               F-3

-------
                         REFERENCES FOR APPENDIX F


1.  Smith, William M.,  "Evaluation of Coke Oven Emissions," Presented
    to the 78th General Meeting, AISI, New York, N.Y.,  May 28,  1970.
    (Also presented at  the 63rd Annual Meeting of APCA, St. Louis,  Mo.,
    June 14-18, 1970.)

2.  OAQPS, Compilation  of Air Pollutant Emission Factors,  2nd Edition,
    U.S. EPA, pp. 7.2-1 - 7.2-3.

3.  U.S. Bureau of Mines, Minerals Yearbook 1972;  Metals, Minerals,
    and Fuels, Vol. I,  pp. 427-437, U.S. Dept. of Interior, Washington,
    D.C., 1974.

4.  Masek, Vaclav, "The Composition of Dusts from Work Sites of Coke
    Ovens," Staub - Reinhaltung Luft, Vol. 30, No. 5, [English edition]
    pp. 34-37, May 1970.

-------