645R79003
        POLYCYCLIC ORGANIC MATTER:

            REVIEW AND ANALYSIS
                    by

     J. 0. Milliken and E. G. Bobalek

           Special Studies Staff
       Office of Program Operations
Industrial Environmental Research Laboratory
    U.S. Environmental Protection Agency
     Research Triangle Park, NC  27711

               February 1979

-------
                                  11
                               CONTENTS

                                                                Page

Executive 'Summary	     iv

     Measurement of POM	     iv
     Sources of POM	      v
     Ambient Concentrations of POM .	    vii
     Exposure to POM	   viii
     Health Effects of POM	      x
     Recommendations	    xii
     Recommendations for the Short-Term	    xii
     Recommendations for the Long-Term	     xv

Acknowledgements 	    xvi

Introduction 	      1

     Background, 	      1
     Definition of POM	      1
     Overview of Report	      2

Objectives and Scope 	      4

Review and Analysis	      7

     Available Information	      7
     Chemical and Physical Properties of POMs.  ... 	     10
     Sources of POM	     12
     Ambient Concentration of POM	     17
     Health Effects of POMs	     21

Conclusions	     24

     Measurement of POM.	     24
     Sources of POM.	     25
     Ambient Concentrations of POM	     26
   .  Exposure to POM	     27
     Health Effects of POM	     28

Recommendations	     30

     Recommendations for the Short-Term	     30
     Recommendations for the Long-Term 	     32

References	  .     34

Appendix A.  Capillary-Column Gas Chromatograms of POM
             Fractions from Combustion Products	    A-l

Appendix B.  Ambient Concentration Data for POM.	  .    B-l

-------
                                  Ill
                                TABLES

                                                                Page

1.  Estimates of Total Annual B(a)P Emissions from Major
    Sources	     vi

2.  Reported Environmental Concentrations of Benzo(a)pyrene.  .     ix

3.  Estimated Daily Exposure to B(a)P, Carcinogenic PAH,  and
    Total PAH	    xi

4.  Summary of Physico-Chemical Properties of Polycyclic
    Aromatic Hydrocarbons and their Relevance to
    Carcinogenesis	     13

-------
                                  IV
                           EXECUTIVE SUMMARY



     The Glean Air Act Amendments of 1977 add Section 122,  "Listing of
          t

Certain Unregulated Pollutants," to Title I of the Clean Air Act (1).


This amendment directs the Administrator to "... review all available


relevant information and determine whether or not emissions of ...


polycyclic organic matter into the ambient air will cause,  or contribute


to, air pollution which may reasonably be anticipated to endanger public


health."


     The principal conclusion of this report is that, on the basis of


available information, emissions of polycyclic organic matter (POM) into


the ambient air cannot, at this time, be reasonably anticipated to


endanger public health.  Because current uncertainties in the POM ex-


posure assessment data and in the health effects data present a risk


that this conclusion is incorrect, development of additional POM assess-


ment data is recommended.  The rationale for this conclusion and spec-


ific recommendations for developing the requisite POM assessment data


base are set forth below.



Measurement of POM


     Because POM in the ambient air consists of a large number of spec-


ific polycyclic organic compounds at very low concentrations and co-


exists with a broader spectrum of organic compounds, POM is very dif-


ficult to detect and quantify.  Traditionally, the amount of POM has


been identified by measurement of benzo(a)pyrene [B(a)P], which is one


of many POM compounds, or by benzene soluble organics [BSD], which


represents a broader range of organic compounds.  Because the relation

-------
                                  VI
TABLE 1.  ESTIMATES OF TOTAL ANNUAL B(a)P EMISSIONS FROM MAJOR

          SOURCES3'b
Source
Current B(a)P Emissions (Mgc/yr)
Estimated Min.
. Estimated Max.
Burning Coal Refuse Banks

Coke Production

Residential Fireplaces

Forest Fires

Coal-fired Residential Furnaces

Rubber Tire Wear

Automobiles (gasoline)

Commercial Incinerators
280

  0.050

 52

  9.5

  0.85

  0

  1.6

   .98
310

300

110

127

740

 11

  3.8

  4.7
      From reference 3.

      Benzo(a)pyrene [B(a)P] is only one of many individual POM com-
pounds.  Traditionally B(a)P has served as an indicator of total POM,
but this practice is questionable.

     cmg = megagrams or metric ton.
      Includes both wildfires and prescribed burning.

-------
                                  Vll
wide ranges of estimates for B(a)P emission factors.  Although the open



burning sources contribute a significant portion of total B(a)P emis-



sions, the temporal and spatial relationship of these sources to human



receptors diminishes their probable impact on public health.




     Coal derived materials (e.g., coal tar, pitch, creosote, and soot)




and coal conversion processes (e.g., coking, gasification, and liquifi-



cation) are known sources of POM  (6, 46, 47).  Approximately 300 indi-




vidual POM compounds have been identified as products of coal carboni-



zation (coking) (48), and there are surely many more.  Coal itself is



highly aromatic in chemical structure and is known to contain B(a)P and




other POMs (49), but most POMs from coal processes are chemically formed




during high temperature (700° C to 1500° C) pyrosynthetic conversion




reactions.




     Coal gasification involves chemical reactions similar to carboni-



zation, and is also a source of POMs.  The high boiling condensation



products of coal gasification contain POMs.  Coal liquification products



and by-products, shale derived oil, and petroleum crude oil represent




additional energy related materials which contain POM (6, 26).  Proper




design, construction, and maintenance of coal processing plants is



necessary to prevent the commercialization of coal and shale conversion



from becoming a new major source of POM.






Ambient Concentrations of POM



     Trace levels of POM have been detected in air, water, and soil.  In



air, POM is thought to exist primarily in a condensed phase, either



adsorbed onto particulate matter, especially soot, or condensed with



other organic compounds as an aerosol.  Because POMs are formed in a

-------
                                 Vlll
vapor phase during high temperature combustion or pyrolysis processes

and condense or adsorb onto particulate matter, smaller particles,  which
         •
have a higher surface area to mass ratio, have a higher concentration of

POM per unit mass.  These smaller particles are more apt to penetrate

deep into the lung during inhalation.  The solubility of POMs in water

is very small, and consequently most POM in aquatic systems adsorbs on

particulates and in sediment.  Typical levels of benzo(a)pyrene in the

environment are illustrated in Table 2.

     Ambient air sampling of B(a)P or total POM is primarily by particu-

late capture on glass fiber filters.  The recognized disadvantages of

this method are (1) inability to capture vapor phase POM, (2) chemical

reaction on the filter, and (3) desorption or stripping of POM from

particulate on the filter.  All of these problems result in a loss of

detectable POM from the filter, and consequently, the reported values of

ambient concentrations probably underestimate the real situation.


Exposure to POM

     Because significant uncertainty exists for POM source inventories

and ambient air concentrations, it is difficult to provide confident

estimates of human exposure to airborne POM.  The existing data are

primarily for B(a)P emission factors and ambient B(a)P levels (e.g.,

references 1-3, 6), but the ratio of B(a)P to total POM is known to be

highly variable, ranging at least 1% to 20% (28).  However, some ex-

posure estimates have been made for occupational settings.  In addition

to cigarette smoking the greatest human exposures to airborne POM prob-

ably occur in certain occupations, viz., the top side of coke ovens,

coal tar pitch working, and hot asphalt paving (45).  The total fraction

-------
                                  IX
   TABLE 2.  REPORTED ENVIRONMENTAL CONCENTRATIONS OF BENZO(a)PYRENEc

t
Urban air
Rural aire
Groundwater
Surface water
f
Drinking water
Marine sediment

Soil
Minimum
0.02C ng/m3

N.D.g
.6 ng/1
N.D.g


1.53 ng/g soil
Maximum
2.4C ng/m3

4 ng/1
210 ng/lh
2.1 ng/1


350k ng/g soil
Mean
0.33d ng/m3
O.le ng/m3



2201 ng/g dry
sediment


      Benzo(a)pyrene  [B(a)P] is only one of many POM compounds.  The use
of B(a)P as an indicator of total POM or of carcinogenic POM is questionable.

      Urban air values are from the 32 National Air Surveillance Network
(NASN) urban sites that reported B(a)P concentrations during 1977.  Con-
centrations reported by NASN are quarterly composite averages.
     Q
      Minimum and maximum values are for the 1977 NASN B(a)P quarterly
composite averages.  The minimum is for Honolulu, Hawaii, second and
third quarter; the maximum is for Youngstown, Ohio, first quarter.

      Geometric mean of 1977 NASN urban B(a)P values.

     eThis value is from reference 53 and represents the average of
two rural locations for 1976.

      Data from reference 53.

     gN.D. = none detected at a detection limit of 1 ng/1.

      Contaminated surface water concentrations of B(a)P have been
reported as high as 12000 ng/1 (53).

      This value is from reference 54 and represents the average of two
sampling stations.  Station A was 75 ng/g dry sediment and station B
was 370 ng/g dry sediment.  Sampling location was Buzzard's Bay, MA.
     •
     •'prom reference 55.  Sample was from forest soil remote from
industry and human habitation.
     jj
      From reference 56.  Sample was from urban soil.  B(a)P concentra-
tions in.contaminated soil near an oil refinery have been observed as high
as 220,000 ng/g soil.

-------
of POM compounds in most emissions from these sources is small,  e.g.,

typically 1% of the total hydrocarbon content.  However, some of the
          t
POMs in this fraction are POM compounds which laboratory bioassays  have

proved to be carcinogenic.

     In addition to exposure to airborne POM, ingestion of water and

food also contribute to the total body burden of POM.  Estimates of the

relative contributions of air, water, and food to the total body burden

illustrate that food contributes the overwhelming exposure route (see

Table 3).  Because the digestive system presents different receptor

organs for POM, the risk to health from POM ingestion cannot be compared

directly to inhalation of POM.


Health Effects of POM

     Epidemiology studies have conclusively demonstrated that long-term

occupational exposure to coking and other processes where products  of

high temperature coal conversion processes are present, results in  in-

creased risk to certain types of cancer (47).  Because POMs are present

in the process and effluent streams of these high temperature coal

conversion process, and because many POMs, in pure form, are very potent

carcinogens in laboratory animal studies, POMs are suspected to be

significant co-factors in the causation of these occupational cancers.

     Health assessment of environmental POM is complicated by uncertain-

ties in extrapolating dose-response information from pure compound

animal bioassay tests to human exposure to the complex environmental

pollutant mixture.  Epidemiology studies are confounded by the con-

current presence of a wide spectrum of non-POM compounds, some of which

may be equally or more hazardous than POM (46).  However, epidemiology

-------
                                 XI
TABLE 3.  ESTIMATED DAILY EXPOSURE TO B(a)P, CARCINOGENIC PAH,  AND
          TOTAL PAH'
                   a,b
•
Source
Water
Food
Air
TOTAL
Estimated
Exposure, yg/day
B(a)P Carcinogenic PAH°
0.0011 0
1.1 4
0.005 0
1.1 4
.0042
.2
.027
.2
Total PAHd
0.027
27.
0.11
27.
      From reference 53.

      PAH = Polynuclear Aromatic Hydrocarbons, a major subset of POM.

      "Carcinogenic PAH" here refers to the sum of B(a)P, benzo(j)-
fluoranthene, and indeno[l,2,3-cd]pyrene  and is acknowledged to repre-
sent only a partial list of actual carcinogenic PAH.

      "Total PAH" is the sum of individual PAHs which were measured.

-------
                                  Xll
studies have demonstrated that populations exposed to polluted air,

e.g., in urban areas, are in an increased risk category for lung cancer.
           i
Because levels of B(a)P and presumably total POM parallel indices of

general air pollution, and because many POM compounds are carcinogenic

in laboratory bioassays, POM in community air is suspected as a cofactor

which contributes to the excess risk to lung cancer for urban residents.


RECOMMENDATIONS


     Implementation of the following recommendations will support a more

complete environmental assessment for polycyclic organic matter, as was

mandated in Section 122 of the Clean Air Act Amendments.  Although it

is unlikely that the issue of "whether or not ... (POM)  ... may reason-

ably be anticipated to endanger public health" will be resolved with

certainty in the near future, information derived from carrying out the

following recommendations will greatly reduce the risk that an incorrect

decision will be made.  Additionally, this information would support the

development of POM control strategies if that course of'action became

necessary.


Recommendations for the Short-Term (one to five years)

     1.  It is recommended that EPA develop and promote  the use of a

standardized procedure for total POM.  Although total POM includes both

the hazardous and non-hazardous POM compounds, preliminary statistical

calculations indicate that this total can represent an improved indi-

cator of the integrated hazard of a particular effluent  stream of am-

bient air samples (57).  That is, a measurement for total POM can cor-

relate better with the POM associated hazard than the previous indicators,

-------
                                 Xlll
B(a)P and BSD.  Several semiquantitative measurement techniques for



total POM exist; these are the sensitized fluorescence test (37) and the



gravimetric determination of material in one or several adjacent liquid




chromatography  (LC) fractions (e.g., Level I, LC fractions 3 and 4)




which contain most of the POM compounds (52).  Either of these tech-




niques, as it exists or modified, could provide the basis for a stand-



ardized procedure for total POM.  Further, the combined use of total POM




and the previously used indicators would increase the reliability of the



estimate of hazard even more.  Statistical techniques should also be



developed to determine the best such joint use of all available indi-



cators including total POM.




     2.  It is recommended that EPA evaluate the impact of residential



coal burning on ambient air quality, particularly ambient air levels of



total POM.  Although coal currently supplies only about two percent of



the total space heating energy in the residential and commercial sec-



tors, the combustion of coal in inefficient residential stoves and



furnaces can be expected to contribute significant amounts of POM to




community air (27, 29, 38, 50, 51).  There is renewed interest in resi-



dential coal-fired equipment in certain parts of the country; e.g., the



mountains and west-north central regions where this use of coal is



becoming economically competitive with alternate residential energy



sources (51).  Airborne emissions from residential coal burning are




uncontrolled, relatively rich in hazardous organics such as POMs, and



emitted at close to ground level.  Consequently, any increase in the




usage of residential coal combustion for space heating, particularly in

-------
                                  XIV
densely populated areas, could lead to exposure levels of ROMs, which

could be hazardous (50).
         t
     3.  It is recommended that sampling and analysis for total POM

become an integrated component for all emission assessment projects

involving pyrolitic or combustion processes.  The objective is to iden-

tify and quantify the major sources of POM in a systematic manner.

Development of such an improved source inventory for POM is important

for negating (or establishing) relationships between emissions of POM

and community health.  Because current emissions data for POM is focused

on relatively few large industrial or utility point sources, the con-

tribution of many dispersed area sources may be underestimated.  This is

particularly true for POM emissions, where the "large" point sources

tend to be efficiently operated fossil fuel combustors that emit rela-

tively few POMs per unit of heat input, and the small, but multiple and

dispersed area sources tend to be less efficient combustors and conse-

quently emit more POM per unit of heat input.

     4.  It is recommended that EPA expand its ambient air monitoring

program to include the sampling and analysis of total POM.  Quarterly

composite samples of benzo(a)pyrene [B(a)P] are already analyzed for at

approximately 40 sites, but the level of B(a)P alone is not a sufficient

indicator of either total POM or the hazard associated with POM.

Assuming that a standardized procedure for total POM is developed as

recommended in subparagraph 1 above, then it is recommended that this

procedure be implemented at each of the 40 National Air Surveillance

Network (NASN) monitoring sites where B(a)P is currently being

monitored.

-------
                                  XV
Recommendations for the Long-Term (five to twenty-five years)




     1.  It is recommended that EPA support a fundamental  study of the




formation -of POMs in combustion systems as a function of fuel  parameters



and combustion operating conditions.  This knowledge would support the




development of improved source emissions inventories and additionally



would provide a data base which would support the development  of POM



control technologies.



     2.  It is recommended that EPA continue and expand research pro-



grams directed at tracking the chemical fate of POMs from the  points of



emissions through the transporting media to the eventual receptor.  The




goal of these transport and transformation studies is to relate ambient



levels of POM to specific sources of POM.  POM sampled from a  flue gas



at 400  F may not have the same compositional profile as the ambient POM




resulting from the same source.  Individual POM species may degrade,




e.g.* by photo-oxidation, between the flue and the receptor.  Because




the relative toxicity of any POM derivative can be very sensitive to



even minor chemical changes, it is as important to characterize the



transformation products as well as the parent POM itself.



     3.  It is recommended that all coal conversion processes  be thor-



oughly evaluated as potential sources of polycyclic organic matter.



Evaluation of these sources should include the monitoring of both point



and fugitive emission streams for total POM.  The existing data base




suggests that coal conversion processes involving high temperatures



represent the potentially most potent sources of hazardous POMs (26, 46,



47).

-------
                                  XVI
                           ACKNOWLEDGEMENTS
     The authors gratefully acknowledge the valuable discussions with
Dr. W. G.-Tucker, SSS/IERL, and Mr. J. A. McSorley, SSS/IERL,  in devel-
oping the objectives, scope and approach to this project.

     Advice and assistance in obtaining information sources was help-
fully provided by Mr. J. A. Manning, OAQPS, and Dr. R. M. Bruce, Envir-
onmental Criteria and Assessment Office.

     We also appreciate the constructive review comments of Mr. G.  L.
Johnson, SSS/IERL, and Mr. W. W. Whelan, TIS/IERL.

     Finally, we gratefully acknowledge the assistance of Mrs. S. Milton
in assembling the many draft copies and final draft.

-------
                             INTRODUCTION



Background
         •

     The Clean Air Act Amendments of 1977 add Section 122,  "Listing  of


Certain Unregulated Pollutants," to Title I  of The Clean Air Act  (1).


This amendment directs the Administrator to  "...review all  available


relevant information and determine whether or not emissions of ...


polycyclic organic matter into the ambient air will cause,  or contribute


to, air pollution which may reasonably be anticipated to endanger public


health."  If an affirmative determination results from this review,  it


is required that polycyclic organic matter (POM)  be added to the  list  of


pollutants under section 108 (a)(l), "Air Quality Criteria and Control


Techniques," or 112(b)(1)(A), "National Emissions Standards for Hazard-


ous Air Pollutants."  Additionally or alternatively, specific categories


of sources of POM could be included in the list published under section


lll(b) (1)(A), "Standards of Performance of  New Stationary Sources."


     The regulatory impact of any of the above actions would prompt


research, development, and demonstration of  POM control systems.  The


consequences of exercising any of the aforesaid regulatory actions  on


the basis of the "available information" are examined in this report.



Definition of POM


     Strictly speaking, any carbon containing compound with two or  more


ring structures could be considered polycyclic organic matter.  Because


this definition of POM represents such an extraordinary range of  com-


pounds, a more limiting definition is necessary.  In this report  poly-

cyclic organic matter (POM) is defined to be the generic class of

-------
                                  -2-
multiple ring, aromatic compounds which have condensed rings.. The ring


structures may contain heteroatoms  (e.g., nitrogen, sulfur, oxygen)

         •
and/or have attached substituent groups (e.g., methyl, ethyl, nitrate,


cyano).  The  term polynuclear aromatic hydrocarbons (PAH) is more res-


trictive and  refers to that subset  of POMs which consists of condensed


ring aromatics containing only carbon and hydrogen atoms.  The above


definitions are  consistent with those used in earlier reviews of POM


information  (1,  2).


     The usage of some POM compounds is entrenched in our everyday


lifestyle and society.  For example, many consumer goods (e.g., cos-


metics, dyes, and foods) contain natural and/or synthetic POMs.  Cig-


arette smoke  is  known to contain many POM compounds.  Some POM compounds


are valuable  pharmaceuticals such as diazepam (a relaxant), acriflav-


inium chloride (an antibiotic), reserpine (an anti-hypertensive), and


many anti-tumor  agents.  The POM class also includes many biochemicals


that perform  an  essential role in controlling human life processes,


including mental health.


     POM compounds occur in trace quantities in effluent streams of


combustion or pyrolysis processes.  They are widely distributed in the


environment.  The ubiquity of POMs  in plant life suggests that bio-


synthesis is  also a possible origin of POMs, but available data indicate


that this is  not a major source of  POM (6, 7).



Overview of Report


     The purpose of this report is  to review and analyze the current


data base for environmental assessment of POMs, as mandated in the Clean


Air Act-Amendments of 1977.  An adequate data base is requisite not only

-------
                                  -3-
for environmental assessment but also for control strategy development,

if it is determined that POM is reasonably anticipated to endanger
         f
public health.

     The following elements of the data base are important to environ-

mental assessment of POMs and potential control strategy development:


1.   Fundamental chemical and physical properties of POMs.

2.   Sources of POM.

3.   Ambient concentrations and chemical fate of POMs in air, water,  and

     soil.


     Data on the health effects of POM have recently been reviewed by

the Syracuse Research Corporation (4).  It is not intended that this

report focus on the health effects issue.  However, several current

issues and questions relative to the health effects of POMs are dis-

cussed.

-------
                                  -4-
                         OBJECTIVES AND SCOPE




     The overall goal of this study is an independent review and anal-


ysis of available POM data with a broad view of present and probable


future emission sources, environmental fate, and potential health
                                                    *

effects.  Furthermore, EPA's Office of Research and Development, par-


ticularly its Industrial Environmental Research Laboratories (one at


Research Triangle Park, NC; the other in Cincinnati, OH), must be pre-


pared to coordinate and support the development of POM control strategies,


if POMs are recognized as a public health hazard.  The main issues to be


addressed in the review and analysis of the POM data base are:




1.   -Has all available information been identified by the independent


     contractors who were commissioned to review the literature; viz.,


     Syracuse Research Corporation (4), Energy and Environmental Anal-


     ysis, Inc. (3), and the Research Triangle Institute (6)?  If yes,


     is this information base sufficient to assess public health hazard


     of POMs?  What are the most serious gaps in the data base?  What


     factors limit the credibility of the available information?


2.   Is the existing data bank on fundamental chemical and physical


     properties of POMs adequate to support environmental assessment,


     regulation, and control of POMs?  If not, where do we need to focus


     present and future research?  For example, what fraction of POMs


     can be expected in the vapor phase at ordinary ambient temper-


     atures?  What adsorption characteristics describe the interaction


     of POMs with airborne carbonaceous and inorganic particles?  What

-------
                                  -5-
     are the solubilities of various POMs in water?  What  are  the  struc-


     tures and stabilities of POMs and their chemical derivatives?

         •
3.   What.is the accuracy and completeness of the source inventories  of


     POMs (3, 6)?  Have the recent reviews (3,  4, 6)  addressed probable


     future sources such as coal conversion processes in addition  to


     past and present sources?



     To what extent can environmental scientists correlate source  emis-


     sion profiles of POMs with chemical profiles of ambient samples?



     What is the state of the environmental material balance for POMs;


     i.e., are POMs being produced and accumulated at a rate faster than


     their removal?  For example, does the decrease noted  in ambient


     urban B(a)P concentrations in the Syracuse Research Corporation


     report (4) take into account the total environmental  balance  of


     B(a)P?  Or does it reflect a shift in the distribution of B(a)P


     from the atmosphere to the land and water?  Has B(a)P decreased


     while other POMs remained constant or increased?


4.   What is extent and quality of ambient multimedia monitoring data?


     Has the fate of emissions been sufficiently addressed?  Do we


     understand the dynamics of the "POM cycle?"


     4.1  What is our knowledge of dispersion and distribution of  POM


          materials?


     4.2  What is the persistence of POM in different media?


     4.3  What is known of the chemical transformations (e.g., degra-


          dation via biological systems, and photochemical degradation


          of POMs)?

-------
                                  -6-
     4.4  What are the limitations of collection and analytical meth-




          odologies?
         t                                • .


5.   What'do we know about the public health and ecological hazards of




     POMs?



     5.1  What differentiates between a toxic profile of POMs and a




          benign mixture?  Are "young POMs" more toxic than "old POMs?"




          What are the potential health effects of the intermediate




          derivatives?  How does the adsorption of POMs on carbonaceous




          particles affect the biological activity of POMs?



     5.2  What are the implications of synergistic and antagonistic




          health effects of various POM compounds?  For example, how




          significant is the phenomenon of co-carcinogenesis of POM with




          other widespread environmental agents, such as S0_ and NO ?




          What are the probable consequences of POMs acting as metal



          binding ligands for heavy metals (.e.g., as cadmium, mercury,



          and lead) to form biochemically significant POM-metallocenes?



     5.3  Does the evidence for POM effects on enzymatic bioreactions




          implicate a broader range of health effects?  What evidence




          exists for health effects other than cancer?






     The above framework of questions was formulated as a guide to our



review and analysis of available information on POMs.  There exists an



abundance of data on POM, but for the purpose of environmental assess-



ment, these data are often insufficient or not directly applicable for



engineering estimates of human exposure to POM or for evaluation of the



human health effects of POM.

-------
                                  -7-                  ;








                          REVIEW AND ANALYSIS





Available*Information
         y


     Essentially all readily accessible information (i.e.,  data which



had been published) has been reviewed in the 1972 NAS report  (1),  the



1978 EEA report (3), the 1978 SRC report (4), or the 1978 RTI  report



(6).  The NAS report was the first comprehensive review of POMs which



attempted to interpret and evaluate the increasing amount of data on



their carcinogenic effects.  The EEA report, a preliminary assessment of



the major sources of B(a)P, relies heavily on the 1967 study by Hangebrauck



(8).  The SRC report, a health-assessment document, includes a com-



prehensive review of the recent experimental studies of the mutagenic



and carcinogenic effects of POMs.  The RTI report emphasizes the multi-



media aspect of the POM problem.  Other notable information sources on



POMs include the IARC monographs on POMs (9), proceedings of recent



Battelle symposiums on PAHs (10, 11), and the Information Overview.on



Coal Conversion Technologies (12).



     The different perspective of recent information searches  on POMs



have helped to insure a more complete retrieval of relevant information



from existing literature.  The EEA report identifies about 15  major



emission sources and estimates possible exposures that these  sources may



present.  As a compendium of quantitative information on  B(a)P emis-



sions, the EEA report is limited by uncertainties in previous  meas-



urements of B(a)P and in subjective estimates of natural  emissions.  EEA



recognizes that many potentially important sources have not been studied



(e.g., fugitive emissions from industrial sources and small internal



combustion engines).  The probability that exposures from unknown sources

-------
                                  -8-
exist at significant levels either continuously or intermittently sug-

gests that POMs cannot be dismissed as a potential health risk.
          •
     Available source information needs further study, quantification,

and differentiation before any "best available control technology"

(BACT) can be sought, even conceptually.  In particular, a more precise

discrimination needs to be made between the relative contributions of

the controllable anthropogenic sources.  The hypothesis that anthro-

pogenic combustion is a primary source of POMs needs further study so

that contributions from this source can be compared with those deriving

from natural or uncontrolled sources.

     The SRC report (4) was mainly, but not exclusively, dedicated to

characterization of the biological activity of pure POM compounds,

namely the PAHs.  SRC classified the compounds in terms of their res-

pective carcinogenic potency as determined principally by animal test-

ing.  From this point of view, the study was relatively thorough.

Unfortunately, because PAH-induced cancer was the focus, the study may

have been narrow, as was the NAS study (1), in terms of potential over-

all health effects.  Stemmer (13) has suggested that the consequence of

the metal ligand capability of some benign PAHs can be expected to

affect enzymatic mechanisms of life processes.  Chronic cardiovascular

disease and emphysema are examples of these.  Hence, improved chemical

and bioassay characterization of the various mixtures of POM compounds

is a prerequisite to an improved health assessment for POMs.  This type

of data remains as a major unsolved problem of POM health effects research.

     Another view of the information search, the RTI search (6), began

with an indexing of chemical POM compounds not necessarily limited to

the usual potential carcinogens.   Compounds were tracked to their

-------
                                  -9-
potential sources, to quantify potential exposure from these sources.   .


Much of this information resulted from health research with a goal  of
         t

occupational safety and health.  RTI's approach identified information


missed by the other searches.  For example, the widespread occurrence


and biomagnification of POMs via the food chain was uncovered as a


distinct possibility.  An abundance of European literature on the occur-


rence of POMs in food (particularly barbecued, smoked, and fried foods)


was also reviewed in detail by the RTI group.


     The information searchs have been comprehensive, but not exhaus-


tive.  The incomplete overlap of findings by the above groups, each


approaching from A different direction, indicates that, if another  point


of view search were initiated, further contributions would supplement


the present lists of cited POM studies.  Examples of this would be  a


search to identify all available information on the occurrence of POMs


in vegetable foods, or a review of the world's literature on POMs in


mycotoxins (14).  However, literature searches on POMs are coming to the


point of diminishing returns, particularly as they apply to developing


an improved source inventory and subsequent control systems strategy.


Major gaps in existing research data can now be identified; these areas


need attention in order to refine the characterization of emissions


sources, the chemical nature of the emissions, and (to a more limited


extent) the fate of the emissions insofar as this fate relates to reas-


onable predictions of exposure hazards.  Development of this data will


optimize the benefits of future control system programs.

-------
                                  -10-
Chemical and Physical Properties of POMs


     A comprehensive data base on chemical and physical  properties of
         •

POMs is needed before a thorough environmental assessment  can be com-

pleted.  Key questions that depend on this type of data  are:

     1.  What are the relative amounts of POM compounds  in the vapor

phase, condensed phase, and adsorbed phase?  How does this distribution


change with temperature?  Can this distribution be described  not only


for ambient POM, but also for combustion flue gases?

     2.  What are the solubilities of the various POMs,  particularly in


potable waters?

     3.  What are the molecular parameters that differentiate mutagenic

or toxic POMs from biologically unreactive POMs?


     Although information relative to each of these questions exists, in

many cases the data are for pure compounds, and not complex mixtures


such as exist in environmental samples of POM.  Thus, the  major problem

is in extrapolating data obtained under highly controlled  experimental

conditions to the complex situations encountered where a multitude of

POMs coexist in air, water, or soil with many other organic and in-

organic species.


     The question of distribution of POMs between vapor, condensed and

adsorbed phases has been studied both experimentally by Murray et al.,

(15) and theoretically by Natusch (17).  Important factors in this

distribution are:  (1) the equilibrium vapor pressure of the  individual

POMs over the condensed phase and over the adsorbed phase, (2) the

kinetics of adsorbtion/condensation, and (3) the temperature.  Pupp et_

al. (16) have reported the equilibrium vapor concentrations (EVC) of

-------
                                  -11-
several representative ROMs.  Except for the highest molecular weight .

POMs, they concluded that, at normal ambient temperatures,  the EVC was
         •
equal to 'or greater than the concentrations of POMs commonly found in

air.  Therefore, in theory pure POM compounds could exist in the vapor

phase at gross concentrations significantly higher than typical measured


ambient air levels.  Because POMs are typically observed in "partic-


ulate" form (1) instead of the vapor phase, it is likely that adsorption

onto co-existing particles in combustion effluents occurs.   The de-

pressive effect of adsorption on the equilibrium vapor pressure and

hence the EVC is known to exist, but a quantitative description of this

effect has not been reported.  This type of missing data is essential to

resolving the question of relative distribution between phases.  Without

this knowledge, we cannot determine, even with accurate ambient meas-


urements of total POM, if POMs are adsorbing onto particles or if ad-

sorbed POMs are being stripped off into the unsaturated atmosphere.

     At the elevated temperatures encountered in the stacks of fossil

fuel burning plants, most or all of the POMs are expected to exist in

the vapor phase.  Although the theoretical analysis of Natusch (17)

predicts "PAH vapors present in the stack system of a coal fired power

plant will be rapidly and quantitatively adsorbed onto the surfaces of

co-entrained fly ash particles at, or close to, the stack exit," these

vapors may be so dispersed that much of the emanating PAH remains in the

vapor phase.  An improved understanding of these factors is a prereq-

uisite to exposure modelling and to the efficient design of any "add-on"

control device for POMs.

     Information on the solubility of POMs in water is important in

understanding their impact on aquatic biota and also in determining

-------
                                  -12-
their potential for transport to humans via drinking water.  The data

for POM solubilities in water are probably adequate for application to
          •
the above'problems (18, 19, 20).  Furthermore, Makay and Shiu (18) have

reported a method of estimating the solubility of a POM based on mol-

ecular structure.

     Although the aqueous solubilities of POMs are quite low, primarily

much less than 100 yg/L at 25  C, Andelman and Suess (7) point out that

POMs may be solubilized in much greater quantities by the presence of

surfactants or that they may adsorb to a wide variety of colloidal

material or biota which are abundant in most natural waters.  This

phenomenon increases the likelihoood of POM exposure to humans and all

other lifeforms which depend on water for their viability.

     The correlation of molecular structure with biological effects is a

goal that extends beyond POMs (21, 22).  For POMs, some progress has

been made by investigators such as Lehr et_ al. (23) and Hecht et^ al.

(24) in correlating qualitative structural parameters with mutagenic

activity.  Quantitative structure-activity relationships such as those

of Wishnok and Archer for nitrosamines (22) have not, as yet, been

attempted for POMs.  A summary of the relation of some basic physico-

chemical properties of POMs to carcinogenesis is given in Table 4 (25).

An improved understanding of the cause and effect relation between

bioreactiyity and chemical properties eases the problem of comprehen-

sively evaluating the biological activity of each POM compound.


Sources of POM

     It is generally agreed that incomplete combustion accounts for the

majority of airborne POM emissions (1-6, 8-12, 26-28).  Because a

-------
                                 -13-
TABLE 4.  SUMMARY OF PHYSICO-CHEMICAL PROPERTIES OF POLYCYCLIC AROMATIC
          HYDROCARBONS AND THEIR RELEVANCE TO CARCINOGENESIS (from
          reference 25)
  Property

Molecule Size
Thickness



Planarity


Conjugation



Symmetry


Substituents
                         Remarks

         optimum planar area for PAH carcinogens
          Very large PAH are not carcinogenic.  There
seems to be a limit to the size of a molecule beyond
which it cannot induce cancer.
Apparent
* 120% .
This has been studied in more detail with hetero-
cyclic compounds, but there seems to be a limit to
the thickness of a molecule for proper "fit."

PAH carcinogens are planar.  Deviations from plan-
arity reduce carcinogenicity.

Reduction (by partial hydrogenation) of the maxi-
mum number of cumulative double bonds in a PAH
structure reduces carcinogenicity.

Carcinogenic activity may be more frequent in com-
pounds with asymmetric geometry.  Exceptions exist.

Size and position of substituents modify
carcinogenicity.
Molecular Weight  Most carcinogens have molecular weights below 500.
Volatility
Solubility in
  Water
Fluorescence
Photodynamics
PAH hydrocarbons may be sublimed.  Important in
transport of particles in the atmosphere.

PAH (carcinogenic and non-carcinogenic) in lipids
varies.  As membranes of cells contain lipids, it
may be important in transport of molecules or by
virtue of interaction with membrane.

Many PAH carcinogens exhibit fluorescence.  Not
specific.

Many PAH exhibit photodynamic activity.  Not
specific.

-------
                                  -14-
significant portion of airborne POM is associated with suspended parti-

cles, transport Of POM to water and soil via atmospheric deposition can
         •
be expected (1, 6, 28).  Table 1 illustrates the range of reported

emission factors.

     Emission factors for major sources of atmospheric B(a)P have been

estimated and compiled in the recent EEA report (3).  It is acknowledged

by EEA and by most reviewers of the EEA report that these estimated

emission factors may have error bounds which span two orders of mag-

nitude.  This uncertainty can be directly attributed to:  (1) the normal

variation of fractional POM in a combustion effluent--i.e., the pro-

portion of POM in the effluent is extremely dependent on the combustion

parameters, (2) variation in the ratio of B(a)P to total POM, (3) exper-

imental errors associated with collection and analytical methodologies,

and  (4) the paucity of quantitative data on either B(a)P or total POM

emissions from individual sources.  It is probable that the data base on

POM emission factors will improve significantly in the next 1-5 years;

e.g., as evidenced by the series of Source Assessment Documents cur-

rently being developed by the EPA Industrial Environmental Research

Laboratory at RTP (33).

     A major difficulty in the development of a working "source inven-

tory" for POMs is the difficulty in differentiating between the relative

contributions of industrial point sources (e.g., petroleum catalytic

cracking, coal-fired power boilers) and fugitive area-wide sources

(e.g., residential furnaces, automobiles, open burning sources).  For

example, open burning includes forest fires—both wildfires and pre-

scribed burns, agricultural burning, peat burning, and burning coal

-------
                                  -15-
refuse banks.  The wide range of combustion conditions for these sources,




and the paucity of relevant laboratory and field data contribute to  the



"several order of magnitude" uncertainty in the estimated emission



factors. Additionally, the lack of knowledge of fuel loading;  e.g.,




total acres burned, contributes further to the uncertainties in total



POM emissions from these sources.




     Open burning sources probably contribute a very significant portion



of total POM emissions (1, 3), but the temporal and spatial relation-




ships of these sources to human receptors diminishes their probable



impact on public health.  The quality and quantity of existing data  do



not permit an accurate assessment of this effect.  In addition to the



benefits of dispersion and subsequent dilution of POMs emitted from  open



burning, the delayed atmospheric residence between source and receptor




favors increased photo-oxidative degradation of POMs (28).  Although



probably beneficial, complete understanding of this effect is clouded  by



recent theories that implicate the diol epoxide derivatives of POMs  as



the functional agents in POM-induced mutagenesis (4, 10, 11).  The diol



epoxides are known to be metabolic derivatives of the parent POM, but  it




is reasonable to expect that these and other similar derivatives may



occur as intermediate products of photo-oxidation.




     Formation of POMs during combustion processes depends on a number



of combus.tion parameters; e.g., fuel type, fuel to air ratio, temperature-



time relationship of the reaction mixture, and the catalytic properties



of coexisting reaction components.  POM compounds within the total mix



are highly dependent on these factors. In general, the amount of B(a)P



from combustion of oil, wood, or natural gas is less than the amount of

-------
                                  -16-
B(a)P emitted from combustion of an equivalent (in terms of Btu input)




amount of coal  (8, 27).  It is expected that this ranking would remain




the same for total POM emissions under similar combustion conditions;




and, with the exception of natural gas, this has been confirmed exper-




imentally by Lee, et_ aJL (27).  In terms of the size of the combustor,




it is thought that emissions of B(a)P  (and presumably total POM) in mass




per unit heat input  are inversely related to the gross heat input to the




combustor (8).  That is, POM emissions per unit heat input can be




expected to decline  as the size of the combustor increases.  This is



probably related to  the fact that increased temperature and a longer




residence time  in the combustion zone favor a decrease in POM emissions



(8, 9, 26).  Increased excess air usually tends to lower POM emissions




(29), but because this factor also decreases the average flame temper-




ature, too much excess air may be counterproductive.



     The compositional distribution of specific POM compounds resulting



from a combustion source is also very dependent on the combustion par-




ameters.  Lee j3t_ al. have described the distribution of POM compounds



resulting from  the combustion of three common fuels: coal, wood, and




kerosene (27).  Examination of the capillary-column gas chromatograms in




Appendix A illustrates the diversity of POM compounds in various com-



bustion effluents.   By constructing and studying alkyl homolog plots of




these POM mixtures and comparing the plots with those obtained from



analysis of POMs in  ambient air particles, Lee et al. were able to




correlate:  (1) distributions of coal-soot POMs with POMs in ambient air



particles from Indianapolis, and (2) distributions of kerosene-soot POMs




with POMs in ambient air particles from Boston (27).  This is signifi-



cant because coal is a primary heating fuel in Indianapolis, and

-------
                                  -17-
heating is primarily by kerosene and other petroleum based fuels  in


Boston. More studies of this type are needed to help clarify the  rela-
         •

tive contributions of specific energy sources to the total POM emission


inventory.   This is particularly important to assignment of priorities


for research and development of control system strategies.


     By analyzing the EEA (3) estimates of B(a)P emissions from major


sources, it can be concluded that even a complete elimination of  B(a)P


emissions from major industrial point sources may not achieve any sign-


ificant reduction in the total level of POM emitted.  That is, when one


considers the uncertainties in the EEA estimates, it is plausible that


B(a)P emissions from area-wide, fugitive type sources; e.g., residential


fireplaces, forest fires, coal refuse banks, and coal-fired residential


furnaces (maximum estimate = 1300 Mg/yr) may constitute greater than


2000 times the B(a)P being emitted from well defined industrial point


sources; e.g., coke ovens, coal-fired power plants and oil-fired  in-


dustrial boilers (minimum estimate = .59 Mg/yr).  However, this un-


certainty should not deter efforts to control and reduce B(a)P and other


POM emissions from those sources where epidemiological and clinical


evidence indicate that emissions of such compounds "may reasonably be


anticipated to endanger public health."



Ambient Concentration of POM


     From the review of available information on ambient concentration


data for POMs, we conclude that POMs are ubiquitous at finite levels in


air, water, and soil (6).  Appendix B illustrates the wide range of


ambient concentrations reported for POMs and B(a)P.  In addition to the


normal spatial and temporal variations expected with ambient monitoring

-------
                                  -18-
of specific pollutant species, the range of ambient concentrations

reported for POMs is further increased as a result of inconsistent
          •
sampling and analytical  techniques (6).  That is, different invest-

igators have used different sampling and analytical methods; and each

sampling method has characteristic limitations in collection efficiency

and sample representativeness.  Both the SRC report (4) and the EEA

report (3) have briefly  reviewed how these factors contribute to ex-

perimental errors in sampling and analysis of B(a)P.  A more complete

review of the state-of-the-art in analytical methodology and stationary

source sampling methodology for POMs has been published by Jones et al.

(34) .

     The notoriety of B(a)P may be an inadvertant consequence of the

development and popularity of a few select analytical methodologies that

are preferentially more  effective in detecting and quantifying this par-

ticular POM.  Improved methodology to capture and quantify other POM

compounds is presently being developed (43).  This is a prerequisite for

advancement of an adequate POM monitoring program.  The state-of-the-art

is such that even what may be technically achieved in principle is not

practical; the effort is too great and the cost is prohibitive.  For

example,  detection of even well characterized PAHs at levels below

1 ng/m  in ambient air strains the limits of even the most advanced gas

chromatography/mass spectroscopy system (34) .  Because such instru-

mentation and associated technical support is very expensive, widespread

surveillance of multimedia ambient levels of individual POM compounds at

the currently achievable levels of detection would be prohibitively

costly.  A promising semi-quantitative spot test for presence of POMs,

-------
                                  -19-
which was recently developed (37), could simplify identification of



sources and evaluation of the performance of any BACT for point sources
          *


suspected'of being major emitters of POMs.  In addition to the present



limitations of the sampling and analytical methodologies, the chemical



reactivity of POMs in ambient media further complicates the analysis of



ambient POM data.



     The chemical fate of POMs in air is complex, but the formation of



their epoxide and diol derivatives via photochemical oxidation is thought



to be a step in the process leading to mutagenesis (4, 9-11).  Although



evidence points to a lesser chemical stability of POM in air than in



aqueous suspensions, sediments, or soils (28), aerosols containing B(a)P



can survive long enough to travel 100 kilometers or farther from com-



bustion emission sources (35).  B(a)P and many other POMs are photo-



chemically unstable as vapors, but their stability is greater when



adsorbed on carbonaceous particles (soot) (41).  It is not known if



reactions between POM and other atmospheric constituents (e.g., SO ,



NO , ozone, and 0 ) affect the risk to public health.  Chemically,
  A              £•


electrophilics (e.g., S0_ and NO ) should promote the efficiency of POM
                        ^       J\


degradation and eventual sedimentary removal.  Purifying the air of



these electrophilics may have the tendency to stabilize POM aerosols.



     The particle size distribution of POM aerosols is concentrated in



the submicron range where washout by natural precipitation and col-



lection by particle control devices at the source are minimal (36),



particularly in the absence of coagulants such as electrolyte mists.



The reaction of POMs with airborne electrolytic species would enhance



the "wet removal" processes of POM from the atmosphere.  The "wet

-------
                                  -20-
removal" processes consist of capture by cloud droplets and the scrub-




bing effect of rain; and because of the enormous quantity of water which




cycles in*the atmosphere, these processes are responsible for the bulk




of the removal of airborne particles (44).  The magnitude of the above




covariant effects has not been determined.  These dynamics of the POM-




environment reaction matrix add to the complexity of interpreting am-




bient concentration data for ROMs.




     Care must be taken not to trade-off air emissions for disposal as




solid wastes.  For example, removal and collection of POMs from sta-




tionary combustion source effluent streams could create a serious soil




and wastewater contamination problem.  Soils could be contaminated by




accumulations in biota or in leachates which contaminate aqueous sedi-




ment, surface water, and even ground water.  POMs have been identified




in each of these media (6, 28).  Accumulations of PAHs and their alkyl



homologies persist for centuries, and are stable enough to serve as a



geological index of historical combustion events such as forest fires



and volcanoes (30, 32).



     The effects of POM on the normal rate.of decomposition of organic



matter in soil may be important.  Trace metals such as zinc and copper



are operative (at low concentrations) to promote and (at excess concen-



trations) to inhibit the degradation of biological materials (42).




Because POMs can act as metal-binding ligands, they may combine with



existing zinc and copper sinks to form stable POM-metallocenes.  Thus,



they may reduce the decomposition rate by removing the essential ele-




ments of zinc and copper and stabilize existing organic matter.  Fur-



thermore, POMs and POM-metallocenes may act as biocidal agents on other



lower forms of animal life which normally accelerate decomposition.

-------
                                  -21-
     Because all of the above factors may have a marked effect  on the


ambient POM data reported, we recommend a critical analysis of  the
          •

present day collection and analytical methods before proceeding with any


expansion of ambient POM surveillance programs.



Health Effects of POMs


     Because POMs have been studied extensively in the role of  tumor


initiation and promotion, chemical carcinogenesis has been the  focus of


the SRC health assessment of POMs (4).  A major conclusion of the SRC


report (4) is that some (but not all) components of the POM mixtures


which are emitted from combustion sources can initiate mutagenic events


by reacting with DNA.  A much proposed hypothesis is that the actual


species needed to induce pathological mutagenesis is the diol epoxide of


some PAHs rather than the parent PAH such as B(a)P (23).  That  this


hypothesis describes all POM/receptor interactions has not been accepted


by most investigators in chemical carcinogenesis (4).  Therefore, it is


uncertain if the diol epoxide (or other active chemical derivatives that


occur in ambient air) is the driving agent for a biological response


normally attributed to the parent PAH reactant.  Another problem is that


contaminants (such as SOO may affect the character of a reactant


substrate such as human lung tissue.  Furthermore, transient active


metabolites have a lifetime which may make them unsuitable for analysis


by present chemical or bioassay techniques.


     Generally, particulate POMs which are available for inhalation also


co-exist with metallic pollutants in the particulate phase.  Heavy


metals (e.g., thallium, mercury, lead, and bismuth) are usually present


in trace quantities in coal combustion effluent streams (38).  It is

-------
                                 -22-
uncertain if this co-existence with metals deactivates or accelerates

POM in the biological conversion reactions that form the direct car-
          •
cinogen.  'Many heavy metals and their oxides may aid in retention,  to

provide longer residence time for the POMs in,  for example,  lung tissue.

Such effects complicate the extrapolation of pure compound bioassay

results to predict the physiological fate of POM captured by any ex-

posure route, whether inhalation, ingestion, or epidermal.  The con-

sequences of such phenomena would significantly affect the development

of control systems.  For example, using the co-action agent to measure

and control emissions may be more effective than directly attacking the

POMs themselves.

     The SRC report points out that disabling toxicity was a significant

cause of mortality that complicated longer-term studies of carcinogenic

potential in sample animals.  Often cancer research attempts to cope

with this as a problem rather than appraise it as a direct effect of the

agents involved.  Some of the potentially "worse" carcinogens may

require a very toxic resistant species of animal that will survive long

enough to separate the mutagenic phenomenon from other complications.

The toxicity threshold to POMs alone (or in combination with other

inseparable components of both natural and anthropogenic emissions) may

be trivial if ambient air levels are stabilized either chemically or by

physical adsorption on particles; however, they could be severe if high

concentrations of transients occur intermittently.  Other suspected

health effects are respiratory diseases such as chronic emphyzema and

asthma (13).  Neurophysiological and other behavioral response effects

as noted for exposure to vaporous organic matter (such as from aviation

-------
                                  -23-
fuel) may also be important  (39).  Because these effects are primarily

occupational hazards, the definition of "may endanger (public health)"
         •
is an issue.  Do we apply the sliding scale test of a severe harm (e.g.,

delayed carcinogen) to few vs. a lesser harm (e.g., any adverse behav-

ioral response) to many  (40)?

     It is not known if  extrapolating occupational mortality and mor-

bidity indicators to the population at large is linear.  Risk assessment

in these extrapolations  is highly speculative, depending in part on the

level of optimism (or pessimism) of the evaluator.  Mandating zero

emissions of both POMs and other co-acting species is the only option

for the pessimist.  The  optimist would quantify concentrations as func-

tions of positions in space and relate these concentrations to proba-

bilities of increasing mortality or morbidity rates compared to a base-

line of clean air.  In approaching the problem from the view of BACT,

the pessimist's goal is unachievable; however, the optimist's goal

(accepting some finite deterioration of air quality) has a reasonable

chance of success.  Data are insufficient to permit assignments of risks

which are related to exposures by any or all modes of exposure (28);

i.e., inhalation, epidermal and ingestion.  Therefore, performance goals

for BACT cannot yet be defined as an achievable target for research,

development, and demonstration.

-------
                                  -24-
                              CONCLUSIONS




     Polycyclic organic matter (POM) is ubiquitous in the environment.
         •                                               •           ,

Both natural and anthropogenic sources emit a large number (greater than


100) of specific POM compounds.  Many of these compounds have been found


to be carcinogenic to animals.  At issue is whether emissions of POM


into the ambient air can be reasonably anticipated to endanger public


health.  Uncertainties in both POM exposure data and POM health effects


data preclude a conclusive scientific assessment of this issue.  How-


ever, existing data do indicate that there exists a potential for a POM


associated public health hazard, and, consequently, this issue cannot be


dismissed.


     The following findings have resulted from the review and analysis


of available information on POM.




Measurement of POM


     Because POM in the ambient air consists of a large number of spec-


ific polycyclic organic compounds at very low concentrations and co-


exists with a broader spectrum of organic compounds, POM is very dif-


ficult to detect and quantify.  Traditionally, the amount of POM has


been identified by measurement of benzo(a)pyrene  [B(a)P], which is one


of many POM compounds, or by benzene soluble organics [BSO], which


represents a broader range of organic compounds.  Because the relation


of B(a)P or BSO to total POM is variable, both B(a)P and BSO are limited


in their usefulness as reliable indicators of total POM.


     Recent improvements in POM sampling and analysis will result in


more efficient capture and characterization of a broader class of POM

-------
                                  -25-
compounds.  Cartridge samplers with absorbent resins can now be used tp

increase sampling efficiency, and gas chromatography coupled with mass
         •
spectroscbpy now provides a tool for improved identification of indi-

vidual POM compounds.

     Several semiquantitative procedures are available for the estima-

tion of total POM; these are the sensitized fluorescence test for poly-

nuclear aromatic hydrocarbons (37), and the combination of gravimetric

fractions resulting from a liquid chromatography separation (52).

Although these procedures for total POM do not characterize the dis-

tribution of individual POM compounds, total POM can be a better indi-

cator of the integrated amount of hazardous POM compounds than any

individual POM, e.g., B(a)P, or the collective total of organic com-

pounds indicated by benzene soluble organics (BSO).



Sources of POM

     The sources of POM are principally, but not exclusively, related to

the incomplete combustion or pyrolysis of fossil fuels and other organic

matter.  Burning coal refuse banks, forest fires (both wildfires and

prescribed burning), residential combustion of coal  and coke production

are thought to be major sources of POM released to the atmosphere (3).

Estimates of total POM emissions from these sources  are very crude, and

consequently, the prediction of human exposure by modeling the dis-

persion of POM from known sources is limited.  Table 1 illustrates the

wide ranges of estimates for B(a)P emission factors.  Although the open

burning sources contribute a significant portion of  total B(a)P emis-

sions, the temporal and spatial relationship of these sources to human

receptors diminishes their probable impact on public health.

-------
                                 -26-
     Coal derived materials  (e.g., coal tar, pitch, creosote, and soot)
                                                                      *
and coal conversion processes (e.g., coking, gasification, and liquifi-
         t
cation) are known sources of POM  (6, 46, 47).  Approximately 300 indi-

vidual POM compounds have been identified as products of coal carboni-

zation (coking)  (48), and there are surely many more.  Coal itself is

highly aromatic  in chemical structure and is known to contain B(a)P and

other POMs (49), but most POMs from coal processes are chemically formed

during high temperature  (700° C to 1500° C) pyrosynthetic conversion

reactions.

     Coal gasification involves chemical reactions similar to carboni-

zation, and is also a source of POMs.  The high boiling condensation

products of coal gasification contain POMs.  Coal liquification products

and by-products, shale derived oil, and petroleum crude oil represent

additional energy related materials which contain POM (6, 26).  Proper

design, construction, and maintenance of coal processing plants is

necessary to prevent the commercialization of coal and shale conversion

from becoming a  new major source of POM.


Ambient Concentrations of POM

     Trace levels of POM have been detected in air, water, and soil.  In

air, POM is thought to exist primarily in a condensed phase, either

adsorbed onto particulate matter, especially soot, or condensed with

other organic compounds as an aerosol.  Because POMs are formed in a

vapor phase during high temperature combustion or pyrolysis processes

and condense or  adsorb onto particulate matter, smaller particles, which

have a higher surface area to mass ratio, have a higher concentration of

-------
                                 -27-
POM per unit mass.  These smaller particles are more apt to penetrate

deep into the lung during inhalation.  The solubility of POMs in water
         •
is very small, and consequently most POM in aquatic systems adsorbs  on

particulates and in sediment.  Typical levels of benzo(a)pyrene in the

environment are illustrated in Table 2.

     Ambient air sampling of B(a)P or total POM is primarily by particu-

late capture on glass fiber filters.  The recognized disadvantages of

this method are (1) inability to capture vapor phase POM, (2) chemical

reaction on the filter, and (3) desorption or stripping of POM from

particulate on the filter.  All of these problems result in a loss of

detectable POM from the filter, and consequently, the reported values of

ambient concentrations probably underestimate the real situation.


Exposure to POM

     Because significant uncertainty exists for POM source inventories

and ambient air concentrations, it is difficult to provide confident

estimates of human exposure to airborne POM.  The existing data are

primarily for B(a)P emission factors and ambient B(a)P levels (e.g.,

references 1-3, 6), but the ratio of B(a)P to total POM is known to  be

highly variable, ranging at least 1% to 20% (28).  However, some ex-

posure estimates have been made for occupational settings.  In addition

to cigarette smoking the greatest human exposures to airborne POM prob-

ably occur in certain occupations, viz., the top side of coke ovens,

coal tar pitch working, and hot asphalt paving (45).  The total fraction

of POM compounds in most emissions from these sources is small, e.g.,

typically 1% of the total hydrocarbon content.  However, some of the

-------
                                 -28-
POMs in this fraction are POM compounds which laboratory bioassays  have •


proved to be carcinogenic.
          •

     In addition to exposure to airborne POM, ingestion of water and


food also contribute to the total body burden of POM.  Estimates of the


relative contributions of air, water, and food to the total body burden


illustrate that food contributes the overwhelming exposure route (see


Table 3).  Because the digestive system presents different receptor


organs for POM, the risk to health from POM ingestion cannot be compared


directly to inhalation of POM.



Health Effects of POM


     Epidemiology studies have conclusively demonstrated that long-term


occupational exposure to coking and other processes where products  of


high temperature coal conversion processes are present, results in  in-


creased risk to certain types of cancer (47).  Because POMs are present


in the process and effluent streams of these high temperature coal


conversion process, and because many POMs, in pure form, are very potent


carcinogens in laboratory animal studies, POMs are suspected to be


significant co-factors in the causation of these occupational cancers.


     Health assessment of environmental POM is complicated by uncertain-


ties in extrapolating dose-response information from pure compound


animal bioassay tests to human exposure to the complex environmental


pollutant mixture.  Epidemiology studies are confounded by the con-


current presence of a wide spectrum of non-POM compounds, some of which


may be equally or more hazardous than POM (46).  However, epidemiology


studies have demonstrated that populations exposed to polluted air,


e.g., in -urban areas, are in an increased risk category for lung cancer.

-------
                                 -29-
Because levels of B(a)P and presumably total POM parallel indices of


general air pollution, and because many POM compounds are carcinogenic
          »

in laboratory bioassays, POM in community air is suspected as a cofactor


which contributes to the excess risk to lung cancer for urban residents.

-------
                                 -30-
                            RECOMMENDATIONS


        •
     The .principal recommendation of this report is that the EPA Office


of Research and Development continue to develop an improved data base on


POM source inventories, ambient concentrations of POM, and health effects


of POM mixtures in the environment.  The current environmental assess-


ment programs being conducted by the Office of Energy, Minerals and


Industry of EPA/ORD are examples of programs that will provide scien-


tific and engineering data that will support a more conclusive assess-


ment of the risk of ROMs'to public health.  Several specific recom-


mendations are set forth below.



Recommendations for the Short-Term (one to five years)


     1.  It is recommended that EPA develop and promote the use of a


standardized procedure for total POM.  Although total POM includes both


the hazardous and non-hazardous POM compounds, preliminary statistical


calculations indicate that this total can represent an improved indi-


cator of the integrated hazard of a particular effluent stream of am-


bient air samples (57).  That is, a measurement for total POM can cor-


relate better with the POM associated hazard than the previous indi-


cators, B(a)P and BSD.  Several semiquantitative measurement techniques


for total POM exist; these are the sensitized fluorescence test (37) and


the gravimetric determination of material in one or several adjacent


liquid chromatography (LC) fractions (e.g., Level I, LC fractions 3 and


4) which contain most of the POM compounds (52).  Either of these tech-


niques, as it exists or modified, could provide the basis for a stand-


ardized procedure for total POM.  Further, the combined use of total POM

-------
                                 -31-
and the previously used indicators would increase the reliability of the
                                                                      i
estimate of hazard even more.  Statistical techniques should also be

developed* to determine the best such joint use of all available indi-

cators including total POM.

     2.  It is recommended that EPA evaluate the impact of residential

coal burning on ambient air quality, particularly ambient air levels of

total POM.  Although coal currently supplies only about two percent of

the total space heating energy in the residential and commercial sec-

tors, the combustion of coal in inefficient residential stoves and

furnaces can be expected to contribute significant amounts of POM to

community air (27, 29, 38, 50, 51).  There is renewed interest in resi-

dential coal-fired equipment in certain parts of the country; e.g., the

mountains and west-north central regions where this use of coal is

becoming economically competitive with alternate residential energy

sources (51).  Airborne emissions from residential coal burning are

uncontrolled, relatively rich in hazardous organics such as POMs, and

emitted at close to ground level.  Consequently, any increase in the

usage of residential coal combustion for space heating, particularly in

densely populated areas, could lead to exposure levels of POMs which

could be hazardous (50).

     3.  It is recommended that sampling and analysis for total POM

become an integrated component for all emission assessment projects

involving pyrolitic or combustion processes.  The objective is to iden-

tify and quantify the major sources of POM in a systematic manner.

Development of such an improved source inventory for POM is important

for negating (or establishing) relationships between emissions of POM

-------
                                 -32-
and community health.  Because current emissions data for POM is focused


on relatively few large industrial or utility point sources,  the con-
         t

tributioit of many dispersed area sources may be underestimated.   This is


particularly true for POM emissions, where the "large" point  sources


tend to be efficiently operated fossil fuel combustors that emit rela-


tively few POMs per unit of heat input, and the small, but multiple and


dispersed area sources tend to be less efficient combustors and con-


sequently emit more POM per unit of heat input.


     4.  It is recommended that EPA expand its ambient air monitoring


program to include the sampling and analysis of total POM.  Quarterly


composite samples of benzo(a)pyrene [B(a)P] are already analyzed for at


approximately 40 sites, but the level of B(a)P alone is not a sufficient


indicator of either total POM or the hazard associated with POM.


Assuming that a standardized procedure for total POM is developed as


recommended in subparagraph 1 above, then it is recommended that this


procedure be implemented at each of the 40 National Air Surveillance


Network (NASN) monitoring sites where B(a)P is currently being monitored.



Recommendations for the Long-Term (five to twenty-five years)


     1.  It is recommended that EPA support a fundamental study of the


formation of POMs in combustion systems as a function of fuel parameters


and combustion operating conditions.  This knowledge would support the


development of improved source emissions inventories and additionally


would provide a data base which would support the development of POM


control technologies.


     2.  It is recommended that EPA continue and expand research prog-


rams directed at tracking the chemical fate of POMs from the points of

-------
                                  -33-
emissions through the transporting media to the eventual  receptor.   The



goal of these transport and transformation studies is to  relate ambient



levels of 'POM to specific sources of POM.  POM sampled from a flue  gas



at 400° F may not have the same compositional profile as  the ambient POM




resulting from the same source.  Individual POM species may degrade,




e.g-> by photo-oxidation, between the flue and the receptor.  Because




the relative toxicity of any POM derivative can be very sensitive to



even minor chemical changes, it is as important to characterize the



transformation products as well as the parent POM itself.



     3.  It is recommended that all coal conversion processes be thor-



oughly evaluated as potential sources of polycyclic organic matter.




Evaluation of these sources should include the monitoring of both point




and fugitive emission streams for total POM.  The existing data base




suggests that coal conversion processes involving high temperatures




represent the potentially most potent sources of hazardous POMs (26, 46,




47).

-------
                                  -34-
                              REFERENCES
 1.  National Academy of Sciences, Committee on Biological Effects of
     Atmospheric Pollutants.  Participate Polycyclic Organic Matter.
     Washington, DC, 1972.

 2.  Scientific and Technical Assessment Report on_ Participate Poly-
     cyclic Organic Matter  (PPOM).  U.S. Environmental Protection
     Agency, Washington, DC.  Publication No. EPA-600/6-75-001.  1975.

 3.  Energy and Environmental Analysis, Inc., Preliminary Assessment of
     the Sources, Control and Population Exposure tp_ Airborne Polycyclic
     Organic Matter (POM) as_ Indicated by_ Benzo(a)pyrene  [B(a)P].  Nov.
     1978.

 4.  Syracuse Research Corporation.  EPA External Review Draft:  Health
     Assessment Document for Polycyclic Organic Matter.  May 1978.

 5.  Kites, R. A., R. E. LaFlamme, and J. W. Farrington.  Sedimentary
     Polycyclic Aromatic Hydrocarbons:  The Historical Record.  Science,
     lj)8_:829-831, 1977.

 6.  Research Triangle Institute.  Draft Report:  Sources and Ambient
     Concentration Data for Polycyclic Organic Matter.  June 1978.

 7.  Andelman, J. B. and M. J. Suess.  Polynuclear Aromatic Hydrocarbons
     in the Water Environment.  Bull. Wld Hlth Org.  43, 479; 1970.

 8.  Hangebrauck, R. P., D. J. von Lehmden, and J. E. Meeker.  Sources
     of Polynuclear Hydrocarbons  in the Atmosphere.  U.S. HEW, Public
     Health Service, AP-33, PB-174-706, 1967, Washington, DC.

 9.  International Agency for Research on Cancer.  I ARC Monographs of_
     the Evaluation of the Carcinogenic Risks of Chemicals tp_ Man:
     Certain Polycyclic Aromatic  Hydrocarbons and Heterocyclic Com-
     pounds .  Vol. 3.  World Health Organization.  1973.

10.  Freudenthal, R. I., and P. W. Jones (eds.).  Carcinogenesis,
     Vol. I.  Polynuclear Aromatic Hydrocarbons:  Chemistry, Metabolism,
     and Carcinogenesis, Raven Press, New York, New York.  1976.

11.  Jones, P. W., and R. I. Freudenthal (eds.).  Carcinogenesis,
     Vol. 3:  Polynuclear Aromatic Hydrocarbons, Raven Press, New York,
     New York.  1978.

12.  Braunstein, H. M., E. D. Copenhaver, and H. A. Pfuderer.  Envir-
     onmental, Health, and Control Aspects of_ Coal Conversion:  An
     Informational Overview, ORNL/EIA-94 UC-11, -41, -48, -90i, Energy
     Research and Development Administration, Assistant Administrator
     for Environment and Safety, April 1977, Vol. I and II.

-------
                                 -35-
13.  Stemmer, K. L.  Clinical Problems Induced by PAH, in Carcinogenesis,
     Vol. I:  Polynuclear Aromatic Hydrocarbons:  Chemistry, Metabolism
     and jCarcinogenesis, Raven Press, New York, New York, 1976.
         •
14.  Gould, R. F.  Mycotoxins and Other Fungal Related Food Problems,
     Adv. in Chem. Series 149, ACS, Washington, DC, 1976.

15.  Murray, J. J., R. F. Pottie, and C. Pupp.  The Vapor Pressures and
     Enthalpies of Sublimation of Five Polycyclic Aromatic Hydrocarbons.
     Can. .J. Chem., 52(4):557-563, 1974.

16.  Pupp, C., R. C. Lao, J. J. Murray, and R. F. Pottie.  Equilibrium
     Vapor Concentrations of Some Polycyclic Aromatic Hydrocarbons,
     Arsenic Trioxide and Selenium Dioxide and the Collection Effic-
     iencies of These Air Pollutants.  Atmos. Environ., IB_(9): 915-925,
     1974.

17.  Natusch, D. F. S., and B. A. Tomkins.  Theoretical Consideration of
     the Adsorption of Polynuclear Aromatic Hydrocarbon Vapor onto Fly
     Ash in a Coal-Fired Power Plant.  In:  Carcinogenesis, Vol. 3,
     Polynuclear Aromatic Hydrocarbons.  P. W. Jones and R. I. Freudenthal
     (eds.).  Raven Press, New York, New York, 1978.  pp. 145-153.

18.  Mackay, D., and W. Y. Shiu.  Aqueous Solubilities of Polynuclear
     Aromatic Hydrocarbons, J_. Chem. Eng. Data, ^2_(4)399; 1977.

19.  May, W. E., and S. P. Wasik.  Determination of the Aqueous Solu-
     bility of Polynuclear Aromatic Hydrocarbons by a Coupled Column
     Liquid Chromatographic Technique.  Anal. Chem.  5£(1)175; 1978.

20;  Schwarz, F. P.  Determination of Temperature Dependence of Solu-
     bilities of Polycyclic Aromatic Hydrocarbons in Aqueous Solutions
     by a Fluorescence Method.  J_. Chem. Eng. Data, 22_(3)273; 1977.

21.  Hansch, C.  A Quantitative Approach to Biochemical Structure-
     Activity Relationships, Ace. Chem. Res., 2_, 232; 1969.

22.  Wishnok, J. S., M. C. Archer, A. S. Edelman, and W. M..Rand.
     Nitrosamine Carcinogenicity:  A Quantitative Hansch-Taft Structure-
     Activity Relationship, Chem.-Biol. Interactions, 20, 43; 1978.

23.  Lehr, R. E., H. Yagi, D. R. Thakker, W. Levin, A. W. Wood, A. H.
     Conney, and D. M. Jerina.  The Bay Region Theory of Polycyclic
     Aromatic Hydrocarbon-Induced Carcinogenicity.  Cafcinogenesis,
     Vol. 3:  Polynuclear Aromatic Hydrocarbons, Raven Press, New York,
     New York, 1978.

24.  Hecht, S. S., E. J. LaVoie, and D. Hoffman.  Structure-Activity
     Relationships in Polynuclear Aromatic Hydrocarbons, Conference  on
     Carbonaceous Particles in the Atmosphere, Berkeley, California,
     March 20-22, 1978.

-------
                                  -36-
25.  Multimedia Environmental Goals for Environmental Assessment,
     Vol. 1, EPA-600/7-77-136a, pp. G-15.

26.  Guerin, M. R.  Energy Sources of Polycyclic Aromatic Hydrocarbons.
     Oak Ridge National Laboratory (CONF-770130-2), 1978.

27.  Lee, M. L., G. P. Prado, J. B. Howard, and R. A. Hites.  Source
     Identification of Urban Airborne Polycyclic Aromatic Hydrocarbons
     by GC/MS and HRMS, Biomed. Mass Spec. £: 182-6; 1977.

28.  Suess, M. J.  The Environmental Load and Cycle of Polycyclic Aro-
     matic Hydrocarbons.  The Science of the Total Environment, 6^ 239;
     1976.

29.  Source Assessment:  Coal-Fired Residential Combustion Equipment
     Field Tests, June 1977, EPA-600/2-78-004o, U.S. Environmental
     Protection Agency, June 1978.

30.  Blumer, M.  Polycyclic Aromatic Compounds in Nature.  Sci. Am.,
     234_(3), 35; 1976.

31.  Blumer, M., and W. W. Youngblood.  Polycyclic Aromatic Hydrocarbons
     in Soils and Recent Sediments.  Science, 188, 53; 1975.

32.  Hites, R. A., R. E. LaFlamme, and J. W. Farrington.  Sedimentary
     Polycyclic Aromatic Hydrocarbons:  The Historical Record.  Science,
     198, 829; 1977.

33.  U.S. EPA, Industrial Environmental Research Laboratory (RTP, NC),
     Source Assessment Documents, Contract No. 68-02-1874.  For example:
     Source Assessment:  Coal-Fired Residential Combustion Equipment
     Field Tests, June 1977, EPA-600/2-78-004o, June 1978.

34.  Jones, P. W., J. E. Wilkinson, and P. E. Strup.  Measurement of
     Polycyclic Organic Materials and Other Hazardous Organic Compounds
     in Stack Gases:  State-of-the-Art, EPA-600/2-77-202, October 1977.

35.  Lunde, G., and A. Bjorseth.  Polycylic Aromatic Hydrocarbons in
     Long-Range Transported Aerosols, Nature, 268, 518-519, 1977.

36.  National Academy of Sciences.  Committee on Medical and Biological
     Effects of Environmental Pollutants.  Airborne Particles.  EPA-600/1-
     77-053, November 1977.

37.  Sensitized Fluorescence for the Detection of Polycyclic Aromatic
     Hydrocarbons, EPA-600/7-78-182, September 1978.

38.  Preliminary Emissions Assessment of_ Conventional Stationary
     Combustion Systems, Vol. II, Final Report, EPA-600/2-76-046b,
     March 1976.

-------
                                  -37-
39.  Knave, B., B. A. Olson, S. Elofson, F. Gamberale, A. Isaksson,
     P. Hindus, H. E., Persson, G. Struwe, A. Wennberg, and P. Westerho-lm.
     Long Term Exposure to Jet Fuel.  Scand. J_. Work Environ. ^ Health.
     4:19*45, 1978.
     ^—   »

40.  Drechsler, T.  Public Health Endangerment and.Standards of Proof:
     Ethyl Corp. vs. EPA.  Environmental Affairs, VI_(2):227-247, 1977.

41.  Fitch, W. L., and D. H. Smith.  Analysis of Polymeric Carbon, Con-
     ference on Carbonaceous Particles in the Atmosphere, Berkeley,
     California, March 20-22, 1978.

42.  Tyler, G.  Effects of Heavy Metal Pollution on Decomposition in
     Forest Soils — III.  Statens Naturvardsverk, July 1977, pages
     1-105.  (English Summary in:  Summaries of Foreign Government
     Environmental Reports, NTISUB/C/135-012, No. 64, Dec. 1977.

43.  EPA/IERL-RTP Interim Procedures for Level 2 Sampling and Analysis
     of Organic Materials, EPA-600/7-78-016, February 1978.

44.  Esmen, N. A., and M. Corn.  Residence Time of Particulates in
     Urban Air, Atmos. Env.  5:571-578, 1971.

45.  Bridbord, K., et_ a.l_.  Human Exposure to Polynuclear Aromatic Hydro-
     carbons.  Carcinogenesis, Vol. 1, Polynuclear Aromatic Hydrocarbons,
     Raven Press, New York, New YOrk, 1976.

46.  Komreich, M. R.  Coal Conversion Processes:  Potential Carcino-
     genic Risk, MITRE Technical Report MTR-7155, March 1976.

47.  Freudenthal, R. I., et^ aK  Carcinogenic Potential of Coal and Coal
     Conversion Products, A Battelle Energy Program Report, Battelle
     Columbus Laboratories, February 1975.

48.  Anderson, H. C., and W. R. K. Wu.  Properties of Compounds in Coal
     Carbonization Products.  U.S. Department of Interior, Bureau of
     Mines, Bulletin 606, 1962.

49.  Woo, C. S., et^ ajL  Polynuclear Aromatic Hydrocarbons in Coal-
     Identification by Their X-ray Exited Optical Luminescence.  Envir-
     onmental Science and Technology, L2(2):173-174, Feb. 1978.

50.  Weber, R. C.  Impact on Local Air Quality from Coal-Fired Resi-
     dential Furnaces.  M. S. Thesis, University of North Carolina,
     Chapel Hill, NC, 1978.

51.  Cart,  E. N., et^ aK  Evaluation of the Feasibility for Widespread
     Introduction of_ Coal into the Residential and Commercial Sectors,
     Exxon Research and Engineering Report prepared for the Council on
     Environmental Quality, Aug. 1977.

-------
                                 -38-
52.  Combustion Source Assessment Methods and Procedures Manual for
     Sampling and Analysis, prepared for EPA Industrial Environmental
     Research Laboratory, Research Triangle Park, by TRW, Jan. 1977.
         •                    .                •
53.  Santodonato, J. et_ aj_.  Human Health Effects Section of the Water
     Quality Criterion Document for Polynuclear Aromatic Hydrocarbons,
     Draft Report prepared for EPA Office of Research and Development by
     Syracuse Research Corporation, Oct. 1978.

54.  Giger, W., and M. Bluraer.  Polycyclic Aromatic Hydrocarbons in the
     Environment.  Anal. Chem . 46^:1663; 1974.

55.  Harrison, R. et_ a_K  Polynuclear Aromatic Hydrocarbons in Raw,
     Potable and Waste Water.  Water Res. . £:331-9; 1975.

56.  Shabad, L. M. et^ al .  The Carcinogenic Hydrocarbon Benzo(a)pyrene
     in the Soil.  J_. NJU. Cancer Inst.  7:1179-1191; 1969.
57.  Leadbetter, M. R. , and J. 0. Milliken.  Assessment of POM Mix-
     tures, Unpublished memo to W. G. Tucker, Industrial Environmental
     Research Laboratory, U.S. Environmental Protection Agency, Research
     Triangle Park, NC, Feb. 20, 1979.

-------
                          APPENDIX   A
                  CAPILLARY-COLUMN GAS CHROMATOGRAMS
                                  of
                POM  FRACTIONS FROM COMBUSTION PRODUCTS
                          (from reference A-l)
                               CONTENTS
Table A-l   POM  Identified by Gas Chromatography/Mass
            Spectroscopy  (from reference A-l).	      A-l

Figure A-l  Capillary-Column Gas Chromatogram of POM
            Fraction from Coal Combustion Products.  For
            compound identification key, see Table A-l from
            reference A-l	      A-2

Figure A-2  Capillary-Column Gas Chromatogram of POM
            Fraction from Wood Combustion Products.  For
            Compound identification key, see Table A-l from
            reference A-l	      A-3

Figure A-3  Capillary-Column Gas Chromatogram of POM
            Fraction from Kerosene Combustion Products.  For
            compound identification key, see Table A-l from
            reference A-l	      A-4

Reference	      A-5

-------
                                   A-l
TABLE A-l.  POM IDENTIFIED BY GAS CHROMATOGRAPHY/MASS SPECTROSCOPY
Peak No.
Compound
Peak No.
Compound
    1   Me.thylnaphthalene
    2   Biphenyl

    3   Ethylnaphthalene3
    4   Acenaphthylene
    5   Methylbiphenyl
    6   Dibenzofuran
    7   Propylnaphthalene
    8   Fluorene
    9   Methyldibenzofuran
   10   C14Hg (Unknown)
   11   Methylfluorene
   12   Ethyldibenzofurana
   13   Dibenzothiophene
   14   Phenanthrene
   IS   Anthracene
   16   Ethylfluorenea
   17   Propyldibenzofuran
   18   Methylphenanthrenec
   19   4H-cyclopenta[def]-
        phenanthrene
   20   Methyl-4H-cyclopenta-
        [def]phenanthrene
                         o
   21   Ethylphenanthrene
   22   Fluoranthene
   23   Benz[e]acenaphthylene
   24   Benzo[def]dibenzothiophene
                       25   Pyrene
                       26   Ethyl-4H-cyclopenta[def]-
                            phenanthrene
                       27   Methylfluoranthene
                       28   Benzo [ajfluorene
                       29   Benzo [bjfluorene
                       30   Benzo[ghi]fluoranthene
                       31   C..H-. (Unknown)
                             lo 1U
                       32   Cyclopenta[cd]pyrene
                       33   Benz[£] anthracene
                       34   Chrysene
                       35   Methylchrysene6
                       36   Methyl eyelopenta[c
-------
TEMP (°C»
TIMC (WIN)   0
70
90 110
130
150
170
190
210
230
250
                     10
20
30
                                                               50
                                                                           60
                                                       70
                                                      60
110
             Figure A-l.   Capillary-column gas cliroiiiatojjriiins of POM  Fractions from combustion products
                           (From reference A-l).

-------
                   WOOD
                                             14
                                      9 10  11
i       v.    giwii       nn    ji

LJJLu^JjjkLJJ UMA/wVl
                                              IS   IB
                                                  18
                                                      20
TEMP (CCJ
TIME (MIN)   0
70
90
110
130
150
170
190
210
230
250
                  10
 20
30
40
50
60
70
60
90
                                                                                                      110
          Figure A-2.  Capillary-column gas chromatogram of POM fraction from wood  combustion products
                     compound identification key, see Table A-l from Reference A-l.
                                                                        l:ov

-------
              XCROSCNB

                                           10
                                             uW
                                                         10
                                                   15
                                                       18
                                                               22
                                       .125
                                                                 23
                                                                 L'J
                                                                               30
                                                                                31
                                                                                   32
                                                                     39
                                                                                               PI
                                                                                                                47
E«P CO
IME (MINI   0
70
90
110
130
150
170
190
210
230
250
                   10
20
30
40
50
60
70
80
90
110
             Figure A-3.  Capillary-column gas chromato>>rain of POM fraction from kerosene  combustion products.
                          For  compound identification key, see Table A-l from reference A-l.

-------
                                  A-5
                               REFERENCE
A-l  Lee,. M. L., G. P. Prado, J. B. Howard, and R. A. Hites.  Source
     Identification of Urban Airborne Polycyclic Aromatic Hydrocarbons
     by GC/MS and HRMS, Biomed. Mass Spec., 4:182-6; 1977.

-------
                          APPENDIX   B
                  AMBIENT CONCENTRATION DATA FOR POM
                                FIGURES

Number

 B-l   Ambient Concentration Data for POM in Rural Air in
       yg/m	     B-l

 B-2   Ambient ..Concentration Data for POM in Suburban Air
       in pg/m	     B-2

 B-3   Ambient Concentration Data for POM in Urban Air in
       yg/nr	     B-3

 B-4   Ambient Concentration Data for Benzo(a)pyrene in Urban
       Air in vg/m	     B-5

 B-5   Ambient Concentration Data for Total POM in pg/£ in
       Unpolluted Water	     B-6

 B-6   Ambient Concentration Data for POM in River Water in
       Vg/t	     B-7

 B-7   Ambient Concentration of Total POM in pg/kg in
       Unpolluted Soils 	     B-8

 B-8   Ambient Concentrations of Benzo(a)pyrene in Various
       Soils in pg/kg	     B-9

       Reference	    B-10

-------
                                                       B-l
      Naphthalene
       Anthracene!
     Phenantfirena
         Pyfena
Dibenzo(a.£)anthracen«



        Benz(e)pyrene





        Benz(a)pyrene;
Benzo{g,h,i)perylene
        Coronene
     Fluoranthen*
                                       0.00011
                                                            0.001 i
0.01
                                                             iiiiiiiiiiiiiitiiiiiii
                                                             iiiMiiiiiinniiiiiiiii
                                                                       INmillllllllllll
                                                               IIMIIIII
                                                                     IMIIIillllllllllllH
                                                                         Illllllllllllllllll
                                                                Illllllllllllllllll
                                                             Illllllllllllllllll
                                                                       • illllllllllllllll
                                                                  iiiiiiiiiiiiiiiiiini
                                                              iiiiiiiiiiiiiiiiiii
            Figure  B-l.   Ambient  concentration data  for  POM  in  rural air  in  pg/m  .
                            Data are summarized from  RTI's  1978 draft  report on POM,
                            ref.  B-l.   Each line  represents specific values  reported,
                            Dotted  lines indicate ranges.

-------
a,o (a.)a y
                        0.0001
                                     B-2
                                   0-001
O.OJ
                                                               0.7
Figure B-2.
                Ambient  concentration data for POM in suburban air in
                pg/m  .   Data  are summarized from RTI's draft report on
                POM,  ref.  B-l.   Each line represents specific values
                reported.   Dotted lines indicate ranges.

-------
                                         O.OCOOI
                                                                                 ,Q.C6\
                                                                                                                          .0*1
       Anthracene


     fhonanlhrcnc
                                                         f.^-na,	
                                                                                    I'lffirrnncn
                                                                                                          iiiiiiiiiniiiiiiiiiiiniiiiiimiiiiiiiiiiiiiiiiiiiiiiiiiimiiiiiiiiiiii
                                                               , T .-.-.r. -..i.-i i-.
                                                  . ^.ry-HH-ntl.-L.Tir
                                                                                                    ...-r,~n.«-
                                                                                  Huuauntumnuua
                                              	^	                               ———~""tni
                                              ,.,,;•» i:.;i.-i-atj:.-.T-if'<"'t1.Tni Tin iiiiiiiiiifilllnillllllllllllllll
                                              nr--1 •"••-•"

Benz(a)a»tliracenc
                      	'	     -..^........-™ I;-jirj>Mi.jj._iui.^^,_ij.j,jiiiirti,,iitii.Mi.lii,illl,llll,lllllllll,l,llllllllllllllllll,ll,l,lllulll





     Triplienylcne  -f-i—"-~'—•——-—'


                                                                                       miiiii.iiiiiiiiiiiiiiitiiiiiiiiiuituiiiiiiiiiiitii.iiiiititn

                                                                                                             miiii.iii.ii.iiiii.ii.iiii.iiiiii.iiiii.i.tiitiiiiiiiii



                                                  r.^i>iiini^n-:jnn,-;i.-|^MV.,-.-i;,iaiMj^r>.>iir^ri.ijjr,rlriTiirriijnriiiiiiiiililiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiil


                                                  iT-ar,f.l;iiiM-.-l.i^niM.ai.tMr-.^-i.r,..-i^Mr..-1-rir,.|.r||n))M..|.iiiiiii

                                                  TT-.,^.,^.,....^.-.^...^^......,,.-..^..^....,..-^..^^..^^..—	
                                                                                                      I
                                                                                                      I
                                                                           1IIIUIIIIIIIMIIUI.I.IIIIIIIMII.IIIIIII.III



                                                                                       akSjuzBittEai.il ti

                     —*.*J..*~.-r.i.,*-j,, ,^*.                                             *llVBJ.TJ»rth-|'n»i-||-|ITng,IIIIHIIIIIIIIIIIIII.lllllllllllllllllll.lll



   Bcnzo(e)pyrene  it.^'!I!^.l!^^^?^^II-'!!^^^^~.'JTl^!'U^
            H*      i..fliMiy<.cn>t>-n Bff'T"^^;[™^^;^j^l^^.«/i^^^

                                          uii/».fr.^ii>iviii.-ic.ofiri.uiiiiffi-iitii'i.irnifniM'ariM»niiiriiii--liJu».-»jmn.-»-m*i..«.i«;.».i.L~iiiimiiiiiiiiini	




                                          i'tnuniininmiiiarirurnmfimTinff-Tiii.iiiiiiiuiii.iiiiiiiHiiliiin.iliiliiiiiiiiiiiiii
                                          >«-iriniVimi:iin.-tfyiiimr-imijii imriMfniiri-nrn«-i*tiiiiMiiiniimiliilii
                                          i.-.»« .ini«it.Tu»i..-r..:.....-TlVr..^.J.^.»..«,r	^..^p^.—„„„..,	i	tintiiii	minimum.
Figure  B-3.    Ambient  concentration  data  for  POM  in  urban  air  in  ug/m  .    Data are  summarized
                      from  RTI's  1978  draft  report  on  POM,  ref.   B-l.    Each  line  represents  specific
                      values  reported.    Dotted  lines  indicate  ranges.
                                                                                                                                                                                      oo

-------
                                                                               .fJ.V<3O\
                                                                                                .V.VOI
,0.01
,0.1
.1
                                                                                                                                                  iiiiiiiiiiiiiiniiiiinti
                                                                                    UltlllllllllMIUI
      Pervlcna  I1 »irnr~>i'i-8in«MuaiiuiTji»iivrfri'T't»r.fiiBTii--irTnivn'Mii>MiiiiiiiiiiiiiHiiiiiuiiiiiiiiiiiiiiiiiiiiiii
          7       •• -••	••••"•""•-"•••"""-'J--J"!'"«IIUJ(lllllllllllilllllHIIIIKIIIIIIIIUIIIIIIIIIIIHIMIIMII«llll
                                                                 ti^iii*^MMC*UU^U^£*IU^^Bw^^J^^&i^U^&KL^itU IIII
                                                                 .i.;ijt-.[u.vtj"g-.^--:;».jVT;i^.«-;x-.CT«<.!Hiiiiiliii.M|.

                                                                 'lia'iiinur.'mnii'nTrr'iir'friTiT'iiiiTiiitriniirantronaiiiiiiiiHiiiii
                                                                                                                        I
                                                                              aniiiiiiiiimiiiimiiiiiiiiimiiiiiiiuniiiiiixiui
                      T.iiii;i»j«.ti..>.M.yfi.vKf.».».uia.ia--Mfif j-->..^«^Mr.vjjMfm^-j-utrjni	ilillllllllinilllllllllllllllllllll
                      r.,in^m-.-i.v,».-.^.w.^-.-,1T--rlr-TM.-^^u.-^..-,-^,:,. IV! HI l.-.,.;.-^^ ..-,.•>.-jjnmum
                      •rn^''"^^ff ;>'f>Jg"""' "-""•••Ml! Iff IV '"V" ••""-I.""""	niiiiiiini	(









Antlianthrpnp t•>"•"• «'-'-"-"g™-'""«T"r"r'!""'''"»"1Tr"":"""	'"""	iiiiiiiiiiiiiiiiiiiiiiiiiiiiniiiiiiiiiniii
r^lllllOlltllltrll^lr J  " ^    ^	 ._.____.„_.	_.^_.. _ _.,_._ --'~"T'hiy 111 ulttilllitlntll tit ill lilt I


                                                                                          g.fr^itti-jv	inn	llllllll	I

                                                                                          Illllll
                                                                          -•••.•i.»r«iiii	iiuinilllllllllllllliriHIIIIIIIIHIIIII
                            	                              '"•""-•"•••• •T"ri-v-'"""'>"'"x~"J"''J'"''




                   •*IT^i'r"r"TirT"*'*"*'MJ7™"l'''*""g"''";^'"*"Tr"lirtl

                                                                                                                  i.-,L.T.rm7-.-?IWT»v»^f^r«^»
                                                    t..-nTl -11 i.-«-^J.J.,.-.f.v.-i. y.;.ii.--J  -^ .-v. ...i;x crj. uj .-^..i .^i_n^ i«-.y ..,., ,. f| -pp,,-a--n gm,.
                                                                                                                           vflT.^t:-,-.,eJL.n._^^



                                                                                                                                   .I.^IM...^.^.,^,....-^^^^^
                                                                                                                                   j;jn-..-..^.-.^ic.-^-.^.rfT^..>.^-^
                                                                                                        >.^>-,<-L-p-».J.^IMlnu»^.ll.-r,^ri-^frT,^..rf,.r^.>.rJ».;....-^o:^.

                                                                                                                            »-K.-:.i»t:-AVjrw.-.^-^^~.|aB
                                                                                                               :.^.;»--»u*.^-ja..-<».-ililliillilllllllllli
                                                                                                               .»rm.mj^.i.r»«...^i>.. -.«••»i-.^.,«..
                                                                                                               MIIIIIIIIIIIIIIIUIUIIIIIIIIIIIIIIMIIIIIIIIIII
                                                                                                               ^..^u^....^^ >....,^.L.-.-.IMI.,l.	I,,...,..,,.
                                                                                                                                                                                                                                  00
                                                                                                                                                                                                                                   I
Figure   B-3.      CContinued)

-------
                                  B-5
                                                   001
                                           ififuniiinnin
                                                      t(iim.q

Figure B-4.  Ambient  concentration data for benzo(a)pyrene  in urban
             air in yg/rn .   Data are summarized from RTI's  1978  draft
             report on  POM,  ref. B-l.  Each line represents specific
             values reported.   Dotted lines indicate ranges.

-------
           0.001
                                   0.01
                      0.1
       1.0
10
        .
vjround yVotpr
       Lakes
      iliycrs
HHIMilllHIMIHIlHIIHlllHKlmimilllllltl


MkMI.*«.a**M.MfcjI-Ui*^tMiMtlllllllMIMMIIIMI|HIIIIIIHIIIIIIIIlmHIMIIllM|llHtHllllll
"(Carcinogenic PAH)
                                                            '.(Carcinogenic PAH)
                                                                                                                               CO
                                                                                                                               I
Figure B-5.   Ambient concentration  data  for total POM in  pg/fc in unpolluted  water.   Data are
               summarized from  RTI's  draft report on POM, ref. B-l.  Each line represents  specific
               values reported.   Dotted lines indicate .ranges.

-------
                          .001
                                                            .01
    .Pcnz(a)anthraccnc



               Chryjcnjj


                  Pyrgnc
.1
10
                                                           • IMfllMHllllllllllHt
                                                                               k4IHIIMnilf|MIMinilttHIMIIItllllimlllllMtltlflltll1l|t
                                                                                            • IHIIIilllltH.lMIIHIMIHlMIIIIIIMHNIIMHMIIIlllllIIHIUIItf.
       J3cnzp(ii)pyrene

                           .. .... + *~	.^^....^^ f— -_^..^ pin-fll aafcti Manmnri—i—•--— ^	—_ —..^..^_1_. l-LT|)||j




                           ««WkV«*BM»«MH«li^HMMHMMHl««l*i^HnMIM*'t*«BWBMMWB«i«MHMa««aan«MAltM till) III |




                                                       \Mt*m*mm+rm,m*mtm»mmm*i tm Hmirummm iiiiiiiiiiiiiiiiMiliiiiniiiiM

                                                       iitiinniiUiiiiinuilMihiiiiNiiiiiHiMiuiUiuuiiiiiiMitiiiiiiiMMiiuiiiiiiHliiliiiiHmilluiriiiiutiiliiuitililiiiiiliiiiiiiittlUitiliu












PQD.zo(hlfJi!Qrantlicn9  l^*:**?*^*^**^.**^*^





                                                           •MMM«^M^«UA4MM«MI^HMI|UltlUII I lilt 11(11 IlltltltlllUll

                                                           ii >*i«i«aia m t+mr* i^,h*f tmii itimniiniiiiiiinimiiHnimintt»iiMiiyiiHiiiummuiiit





                                                           ijiiiiiiiiiiiiiiiiitiiiiiiiiitiiuiiiiiiiiitiiiiiiiiinitiiiiiiiiiuiituiiiiiiiiiiiiiiiiiiiiiuiuiiiMiuiiitntiuft
                                                           *MldfcllllH»».MiHhlillMMlnil«lai»«iM-«JrtlM •^••Mi^MMllllllllMllllllllllllliniKIIIIIIIMMIUllint





                                                           tJiMaMiinilia»fti-iMM>'i innifc.amuiiiiiUHtiiiinninmniiiinminiiiii>iMti
                                                           •ljt"--«TiJTjT"i •m.-m-mr-•••"'*-*	ni...»t..i.i..t.miii..i.iitn*.tJiiiMtiitu
                                                                                                                                                                                09
Figure  B-6.    Ambient  concentration data  for  POM in  river  water  in  ug/£.    Data  are  summarized  from
                     RTI's  draft  report  on POM,   ref.   B-l.    Each  line  represents   specific  values  reported.

                     Dotted lines  indicate ranges.

-------
                  1.0
  Marino
           (Marsh)
         (Subtjcjul)



Remote Tcrrqs.tfM



             Rural



             Urban
          10.0
100.0
                                                                                                                                   1000.0
                                                      miiiittmimimiiiMiiniiiiiMiitti
iMItllitllltlMltlliinillllllllltllMMIIIIIIIIiltlillllMMIIIIilllllllllllMtllllMlMIIIIIMIiliniMlllllllltltlllHIItlt
                                                 • lllllltUIIIIMItlllMlllllllllltMilllllIIIMIIIIIIIIIIIIIIItlllltlltlllllllltlllllllimilttllllftlllllllllllltllltlllltlttllllllllllllltlUIIIflllllf
                                                                                                                                                           oo
                                                                                                                                                           i
                                                                                                                                                           oo
    Figure  B-7.   Ambient concentration  of  total  POM  in  ug/g  in  unpolluted  soils.   Data  are  summarized

                      from  RTI's draft  report on  POM,  ref.  B-l.    Dotted  lines indicate  ranges.

-------
                      1.0
                               10.0
               Arcns

  •

              Mar mo



          Forest Sg||



          Residential



Natural Background
                                                                                      100.0
1.000.0
10,000.0
                                                                  iiiiiiiiMMittiiiittiuiititiiiiMiiuitiniiUHuiiiiiiitiMiiiiiiiMtiaiiiiiiiiitiiniiiiiiuiiiiniiiiHitttnttiiiiitiiiiiiiiiiiiiiiHinititttiiiuii
MiiMiHmMimiimuMiiiiifiiiiiiiiuiiimMiMiiiitHiiimimiiiitmtiuitiimiiiiiim






IIIIIIUIIIIMMIMIIIllllllllllllltllllllllllllllllllMIIIIIIIIIIIIIMIMIItllllllllllllllltllllllllll
 .HimmilUlltimmifllllimiUllimiliMmiilllHHIlUlUllllMmillllMlimilHItHtlMltlimilllU^
                       imttmiiiimitiiimtititiiiiimiimtiiiiiiiiiiiiiiniiiiimiiiiitMiiii
                                                                                                                                                                                00
                                                                                                                                                                                 I
       Figure  B-8.   Ambient  concentrations  of  benzo(a)pyrcne  in  various  soils  in ugAg-    Data  are  summarized

                           from  RTI's draft  report  on POM,  ref.  B-l.

-------
                                 B-10
                               REFERENCE
B-l  Research Triangle Institute.  Draft Report:  Sources and Ambient
     Concentration Data for Polycyclic Organic Matter.  June 1978.

-------