PBS6-232774
DEVELOPMENT OF ADVISORY LEVELS FOR POLYCHLORIMATED
3IPHENYLS  (PC3S) CLEANUP
U.S.  Environmental Protection Agency
Washington, DC
May 86
               U.S. DEPARTMENT OF COMMERCE
            National Technical  Information Service

-------
                                                 P5do-<: 3277U
                                            EPA/600/6-86/002
                                            May 1986
        DEVELOPMENT OF ADVISORY LEVELS

 FOR POLYCHLORINATED BIPHENYLS (PCBs)  CLEANUP
          Exposure Assessment Group
Office of Health and Environmental  Assessment
      Office of Research and Development
     U.S. Environmental  Protection  Agency
               Washington, D.C.

-------
                                  TECHNICAL REPORT DATA
                           (Please read Inuruciions on the revene before completing)
 1 REPORT NO
     EPA/600/6-86/002
                                                          3 RECIPIENT'S ACCESSION NO
4 TITLE AND SUBTITLE
 Development of Advisory  Levels  for
 Polychlorinated Biphenyls  (PCBs)  Cleanup
             5. REPORT DATE

              Mav 1986
             6. PERFORMING ORGANIZATION CODE

              EPA/600/21
7 AUTHOR(S)
 Seong T. Hwang, James W.  Falco,  Charles  H.  Nauman
                                                          8. PERFORMING ORGANIZATION REPORT NO
9 PERFORMING ORGANIZATION NAME AND AOORESS
                                                          10. PROGRAM ELEMENT NO
 Office of Health and  Environmental  Assessment (RD-689)
 Exposure Assessment Group
 U.S. Environmental Protection  Agency
 Washington, D.C.  20460
             II. CONTRACT/GRANT NO.
 12. SPONSORING AGENCY NAME AND AOORESS
                                                          13. TYPE OF REPORT AND PERIOD COVERED
        same  as  #9
             14. SPONSORING AGENCY CODE

               EPA/600/21
15. SUPPLEMENTARY NOTES
16 ABSTRACT
        This document presents background  information used in developing advisory  level;
  of PCBs  in soil  estimated to be permissible  in  protecting public health.  The  results
  of exposure assessment and health effects studies  are combined to arrive at the
  permissible levels of PCBs.  Health effects  studies conducted using animals for  the
  duration of 10-30 days are used to determine the  10-day advisory levels for PCB  clean-
.up.   The long-term advisory levels are based on the carcinogenic risk evaluations.
        Exposure pathways considered in estimating  the 10-day and long-term average dail
  intakes  include  soil  ingestion, inhalation,  dermal  contact, ingestion of contaminatec
  food, and ingestion of water.  Exposure to drinking water contaminants is presumed to
  occur independently of other pathways, because water could come from a clean public
  water system.  The exposure pathways most pertinent to the evaluation of permissible
  PCB  levels in soil  are soil ingestion, vapor inhalation, and contaminant contact with
  human skin.
        The currently available modeling techniques  considered most appropriate  within
  the  constraints  of availability of input data are  used to estimate exposures.  PCB
  advisory levels  are presented as ranges of values  to reflect the difference in soil-ai
  partition coefficients depending on soil type, different types of commercial Aroclors,
  and  variations in the soil  ingestion rate.
17.
                               KEY WORDS AND DOCUMENT ANALYSIS
                 DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS  C.  COSATI Field/Group
13 DISTRIBUTION STATEMENT

 Release to  the  public
19 SECURITY CLASS (This Report/
  Unclassified
                                                                        21
NO OF PAGES

 216
                                             20 SECURITY CLASS (Thitpage/
                                               Unclassified
                                                                        22 PRICE
EPA Form 222O-I (»-73)

-------
                                   DISCLAIMER





     This document has been reviewed in accordance with the U.S. Environmental



Protection Agency's peer and administrative review policies and approved for



publication.  Mention of trade names or commercial products does not constitute



endorsement or recommendation for use.

-------
                                     CONTENTS



 Tables 	          v

 Foreword	       vii

 Preface	        ix

 Abstract 	          x

 Authors and Reviewers	        xl

 1.  Executive Summary	       1-1

 2.  Introduction	       2-1

 3.  Chemical  Compositions	       3-1

 4.  Production	       4-1

 5.  Uses	       5-1

 6.  Disposal	       6-1

 7.  Chemical  and Physical Properties 	       7-1

 8.  Environmental  Distribution 	       8-1

 9.  Environmental  Fate and Transport	       9-1

10.  Toxicology	      10-1

11.  Existing  Standards and Guidelines  	      11-1

12.  Exposure  Assessment Methodology	      12-1

     12.1  Estimation of Exposures for Contaminated Sites 	      12-3
     12.2  Determination of Permissible Pollutant Level in Soil . .      12-10
     12.3  Incorporation of Time-Varying Parameters 	      12-11
     12.4  PCB Advisory Evaluations	      12-14

13.  Water Quality Limits 	      13-1

14.  Leachate  Contamination of Groundwater	      14-1

15.  Soil  Ingestion Pathway	      15-1

-------
                               CONTENTS (continued)
16.  Inhalation Pathway 	      16-1

     16.1.  Intake by Air Exposure Route	      16-1
     16.2.  Emission Evaluation Scenarios 	      16-2
     16.3.  Air Dispersion Modeling	      16-8
     16.4.  Air Exposure Evaluation	      16-10

17.  Dermal Contact Pathway 	      17-1

18.  Comparison of Exposures by Soil  Ingestion, Inhalation, and
     Dermal Contact	      18-1

19.  Results	      19-1

     19.1  Derivation of Permissible Soil  Contamination  	      19-1
     19.2  Summary of Results	      19-4

20.  Limitations of Application 	      20-1

21.  References	      21-1

APPENDICES

 A.  Models Used in Air Release Rate Calculations	       A-l

 B.  Example Emission Rate Calculations for Four Studied
     Scenarios	       B-l

 C.  Summary of Computer Runs for Each Aroclor and at Each
     Value of Soil-Air Partition Coefficients  	       C-l

 J).  Health Advisories for PCBs in Soil
     (Prepared by M.L. Dourson, Environmental Criteria
     and Assessment Office, Cincinnati, OH) 	       D-l
                                       IV

-------
                                     TABLES
 1.  Permissible PCB soil  contamination  levels  (uncovered
     surface contamination)	      1-6

 2.  Permissible PCB soil  contamination  levels  (25-cm-thick
     clean cover)	      1-7

 3.  Approximate composition  of Aroclors 	      3-3

 4.  Chemical  and physical properties of PCBs	      7-3

 5.  Solubility of chlorobiphenyls  in water	      7-6

 6.  PCB monitoring in ambient air  by NYSDEC	      9-2

 7.  Existing  PCB standards and guidelines  	     11-3

 8.  Maximum lifetime risk for ingesting soil  contamination
     at different PCB levels	     15-1

 9.  Maximum daily PCB intake by ingestion  of  soil
     at various PCB concentrations  	     15-2

10.  Comparison of PCB intakes by Ingestion  and inhalation
     routes for acute effects	     16-3

11.  Comparison of PCB intakes by ingestion  and inhalation
     routes for carcinogenic  effects  	     16-4

12.  Concentration of PCBs in soil  at saturation vapor
     pressure  based on Kj  =  1,000 cm3/g	     16-7

13.  PCB emission rates from  1 yg/g PCB  soil  at different
     control levels	     16-7

14.  Values of constants for  standard deviation expression as
     a function of downwind distance  and stability  condition  ....     16-10

15.  Ambient PCB concentrations at  different locations
     (PCB in soil  = 1 yg/g, PCB-1254)	     16-11

16.  Comparison of intakes by various exposure  routes	     18-1

17.  Evaluation conditions for each Aroclor	     19-5

18.  Low and high values of  air-soil  partition  coefficient used
     in the evaluation	     19-6

-------
19.  Permissible PCB-1242 soil  contamination  levels  (uncovered
     surface contamination)	     19-7

20.  Permissible PCB-1248 soil  contamination  levels  (uncovered
     surface contamination)	     19-8

21.  Permissible PCB-1254 soil  contamination  levels  (uncovered
     surface contamination)	     19-9

22.  Permissible PCB-1260 soil  contamination  levels  (uncovered
     surface contamination)	     19-10

23.  Permissible PCB-1242 soil  contamination  levels  (25-cm-thick
     clean soil  cover)	     19-11

24.  Permissible PCB-1248 soil  contamination  levels  (25-cm-thick
     clean soil  cover)	     19-12

25.  Permissible PCB-1254 soil  contamination  levels  (25-cm-thick
     clean soil  cover)	     19-13

26.  Permissible PCB-1260 soil  contamination  levels  (25-cm-thick
     clean soil  cover)	     19-14
                                      VI

-------
                                    FOREWORD

     The Exposure Assessment Group (EAG)  of EPA's  Office  of  Research  and
Development has three main functions:   1) to conduct  exposure  assessments;  2)
to review assessments and related documents; and 3)  to develop guidelines  for
Agency exposure assessments.  The activities under each of these  functions  are
supported by and respond to the needs  of  the various  EPA  program  offices.   In
relation to the third function, EAG sponsors projects aimed  at developing  or
refining techniques used in exposure assessments,  and at  applying these techniques
to develop health-based advisory levels for contaminant cleanup.   This document
is one of these projects and was done  for the Office  of Solid  Waste and Emergency
Response.
     Polychlorinated biphenyls (PCBs), commercially known as Aroclors, consist
of mixtures of chlorinated biphenyl compounds.  Many  sites contaminated by  PCBs
remain contaminated because of PCB persistence in  the environment. Although
commercial  PCB production has been banned by the Toxic Substances Control  Act,
continued use in previously existing commercial  equipment can  result  in spills
which require cleanup.  EPA has become increasingly involved in the discovery,
assessment, and cleanup of these sites.
     The purpose of this document is to provide advisory  levels for PCB cleanup,
and to describe the detailed technical and scientific rationale and methods
used in developing these advisory levels  for PCBs  in  contaminated soil.   This
project required development of exposure  and risk  assesment  methodology related
to hazardous waste and spill sites, and analyses of health effects data.   The
advisory levels and the assessment methodology thus developed  will help EPA set
                                      VII

-------
priorities in PCB spill  and cleanup management,  and  address other PCB contaminant

problems.
                                                   Michael Callahan, Director
                                                   Exposure Assessment Group
                                     vm

-------
                                    PREFACE

     The Exposure Assessment Group of the Office of Health and Environmental
Assessment (OHEA) has prepared this development document for advisory
levels for polychlorinated biphenyls (PCBs) cleanup at the request of the
Office of Emergency and Remedial  Response.  This document summarizes the
procedures concerning multimedia  exposure assessments for PCB-contaminated
sites, and literature information on chemical  and physical properties and health
effects pertinent to evaluation of exposures to PCBs.  The purpose of this
document is to serve as a technical and scientific basis for developing health-
based advisory levels for PCBs in soil  at spill or cleanup sites.  The literature
search supporting this document is current to May 1986.
                                      IX

-------
                                    ABSTRACT

     This document presents background information  used  in  developing  advisory
levels of PCBs in soil  estimated to be permissible  in  protecting  public  health.
The results of exposure assessment and health effects  studies  are combined  to
arrive at the permissible levels of PCBs.  Health effects  studies conducted
using animals for the duration of 10-30 days are used  to determine the 10-day
advisory levels for PCB cleanup.  The long-term advisory levels  are based on
the carcinogenic risk evaluations.
     Exposure pathways considered in estimating the 10-day  and long-term average
daily intakes include soil  ingestion, inhalation, dermal contact, ingestion
of contaminated food, and ingestion of water.  Exposure  to  drinking water con-
taminants is presumed to occur independently of other  pathways,  because  water
could come from a clean public water system.  The exposure  pathways most per-
tinent to the evaluation of permissible PCB levels  in  soil  are soil  ingestion,
vapor inhalation, and contaminant contact with human skin.
     The currently available modeling techniques considered most  appropriate
within the constraints of availability of input data are used  to  estimate
exposures.  PCBs advisory levels are presented as ranges of values to  reflect
the difference in soil-air partition coefficients depending on soil  type,
different types of commercial  Aroclors, and variations in  the  soil  ingestion
rate.

-------
                              AUTHORS AND REVIEWERS



      The  Exposure  Assessment  Group  of the Office of Health and Environmental

 Assessment was  responsible for preparing this document.



 AUTHORS

 Seong T.  Hwang
 Exposure  Assessment  Group
 Office of Health and Environmental  Assessment
 U.S.  Environmental  Protection Agency

 James W.  Falco
 Office of Environmental  Processes and Effects Research
 U.S.  Environmental  Protection Agency

 Charles H. Nauman
 Exposure  Assessment  Group
 Office of Health and Environmental  Assessment
 U.S.  Environmental  Protection Agency

 REVIEWERS

      The  following  individuals provided  review comments and criticisms

 during several  peer-review processes to  which this document was subjected.


 Don R. Clay,  Director
 Office of Toxic Substances
•U.S.  Environmental  Protection Agency

 Joseph A. Cotruvo,  Director
 Criteria  and  Standards  Division
 Office of Drinking  Water
 U.S.  Environmental  Protection Agency

 Yoram Cohen
 Department of Chemical  Engineering
 University of California,  Los Angeles
 Los Angeles,  California

 Barbara Davis
 Office of Waste Programs Enforcement
 U.S.  Environmental  Protection Agency
                                        XI

-------
Karen Hammerstrom
Exposure Evaluation Division
Office of Toxic Substances
U.S. Environmental Protection Agency

Krishan Khana
Health Effects Branch
Office of Drinking Water
U.S. Environmental Protection Agency

Russell Kinerson
Office of Toxic Substances
U.S. Environmental Protection Agency

William Marcus
Office of Drinking Water
U.S. Environmental Protection Agency

Robert E. McGaughy
Carcinogen Assessment Group
Office of Health and Environmental Assessment
U.S. Environmental Protection Agency

Mary Lund Mortensen
Agency for Toxic Substances and Disease Registry
Atlanta, Georgia

Debdas Mukerjee
Environmental Criteria and Assessment Office—Cincinnati
U.S. Environmental Protection Agency

Judith A. Nelson, Director
Regulatory Coordination Team
Office of Pesticides and Toxic Substances
U.S. Environmental Protection Agency

Arnett No Id
Office of Toxic Substances
U.S. Environmental Protection Agency

Edward V. Ohanian
Health Effects Branch
Office of Drinking Water
U.S. Environmental Protection Agency

Suresh Rao
Soils Department
University of Florida
Gainesville, Florida
                                      xn

-------
Danny Reible
Department of Chemical Engineering
Louisiana State University
Baton Rouge, Louisiana

Charles Ris
Carcinogen Assessment Group
Office of Health and Environmental Assessment
U.S. Environmental Protection Agency

Jerry F. Stara
Environmental Criteria and Assessment Office--Cincinnati
Office of Health and Environmental Assessment
U.S. Environmental Protection Agency

Louis Thibodeaux
Hazardous Waste Research Center
Louisiana State University
Baton Rouge, Louisiana

Edwin F. Tinsworth, Deputy Director
Office of Toxic Substances
U.S. Environmental Protection Agency

-------
                             1.   EXECUTIVE  SUMMARY

     This report has been  prepared  in  response  to a memorandum dated April 9,
1985, from the Office of Emergency  and  Remedial  Response  (OERR),  requesting
that the Office of Health  and Environmental  Assessment  (OHEA) develop advi-
sory levels for polychlorinated  biphenyls  (PCBs) which  can  be used as guide-
lines for initiating removal  action for sites contaminated  with  PCBs.   Inter-
ested offices within EPA,  including OERR,  have  advised  OHEA that  these  advisory
levels for PCBs cleanup should be developed based on  considerations of  public
health protection from short-term and  long-term exposures.  The  advisories
presented in this report include permissible levels of  PCBs in soil correspond-
ing to 10-day and lifetime acceptable  intakes.
     Exposure routes considered  in  developing these advisory levels include
drinking water, ingestion  of  PCB-contaminated soil by children and adults, and
inhalation of ambient air  contaminated  with PCBs.  Other  exposure routes, such
as dermal exposure, food intake, and ingestion  of fish  which have bioaccumula-
ted PCBs, are considered in relation to their importance  and their relevance to
the present document.  In  view of the  high bioaccumulation  factor for PCBs, the
consideration of bioaccumulation is important in setting  PCB levels in  surface
water in which aquatic animals live.  If one of these routes is  a controlling
factor in relation to the  exposure  route or human intake  considered, the  advi-
sories need to be reevaluated.
     Commercial-grade PCBs marketed as  Aroclors in the  United States are  mix-
tures of many chlorinated  biphenyl  compounds in various proportions.  Each
PCB compound may exhibit its  own toxicological  characteristics and physical and
chemical properties.  This fact  complicates the exposure  analysis in deriving
the allowable concentrations  in  drinking water  and soil.   It is  conceivable
                                      1-1

-------
that chemical  and physical  properties reported in  the  literature  for  each  Aro-
clor designation represent  an average property for the mixture.   To define the
variability of safe levels  for contamination by different  Aroclor designations,
exposure analyses have been performed for several  Aroclors:   Aroclor  1242,
1248, 1254, and 1260.  The  steps used in developing the advisories include:  (1)
the evaluation of toxicological  effect studies, (2) exposure analysis for  PCB
intake from drinking water, soil ingestion,  air inhalation,  and dermal  contact,
and (3) risk assessment combining the toxicology studies and exposure analysis.
     Ten-day noncancer health advisories are based on  the  short-term  acceptable
intake (AI) derived from studies of animals  treated with Aroclor-1254 for  no
more than 30 days to examine noncarcinogenic effects.   This  AI value, which
forms the basis for establishing permissible levels of PCBs  in soil,  is 0.1  and
0.7 mg/day for a 10-kg child and a 70-kg adult, respectively. The permissible
PCB concentrations for each carcinogenic risk level are based on  the  potency
factor of 4 (mg/kg-day)'l rounded off from two independent evaluations based on
an Aroclor-1260 study.
     It is likely that not  all  of the PCBs ingested or inhaled by humans are
absorbed.  Proper calculations of absorption rate  and  hence  exposure  should  be
based on realistic pharmacokinetics-type models to determine intake.   Lack of
experimental data with which to estimate the parameters needed in the pharmaco-
kinetics models has prevented their applications to the analysis  for  PCB absorp-
tions through human exchange boundaries.  Future work  should consider these
models.  Although most animal studies (in rats and mice) on  the extent of
absorption in the gastrointestinal tract show absorption in  excess of 90%,
there are two experiments on monkeys reporting less than 88% absorption in one
case and less than 13% and  40% absorption for a specific congener in  another
case, based on the analysis of feces and urine.  Vehicles  used in administering
                                      1-2

-------
PCBs were not specified.  It is likely that the high adsorption characteristics
of PCBs on soil  could retard the absorption rate in the human intestinal  tract.
In the risk analysis performed in the present study, the absorption rate  for
humans after ingestion of RGB-contaminated soil  is  considered to be 30%.
     Absorption from dermal  exposure has been reported  to be as significant  as
from other routes of exposure, but little information is available for  the
quantitative evaluation of dermal absorption rates.  Five percent dermal  ab-
sorption is assumed for soil contaminants in contact with human skin.   The
dermal absorption rate of contaminants present on soil  is presumed to  be  less
than that for contaminants spilled on skin in pure  form.  Inhalation studies
using PCB aerosols show that the absorption of PCBs from inhalation exposure
readily occurs.   In the present analysis, an absorption factor of 50%  is
assumed for absorption of PCB vapors after inhalation into human lungs.
     The circumstances under which human exposure occurs are divided into three
classes depending on population distribution:  (1)  Exposure occurs on-site.
This can be further subdivided into:  (a) sites which are readily accessible
to children, and, hence, the soil from which will be subject to ingestion, der-
mal contact, and inhalation, and (b) sites for which there is no possibility of
soil ingestion,  and, hence,  exposure is only through inhalation; (2) sites
which no population is assumed to enter within a radius of 0.1 km from  the
site; and (3) sites which no population is assumed  to enter within a radius  of
1 km from the site.
     The soil ingestion rates used for Class (l)(a) evaluations are 3  and 0.6
g/day.  The former is a value based on data from a  study of an adult person
with pica, while the latter represents a long-term  average value for soil
ingestion.  If sites are not accessible to populations  at distances of  0.1
km or 1 km from the site, as in Classes (2) and (3) above, it is assumed  that

                                      1-3

-------
no ingestion of contaminated soil occurs and the exposure route is that of
inhalation.
     The emission rate of volatilized PCBs can be considerably reduced by
covering the contaminated soil  by low-porosity uncontaminated soil  or clay
material.  The reduction in the emission rate will  result in a decrease in
ambient air concentrations of PCBs by the action of blowing winds.   When PCB-
contaminated material  is directly exposed to the atmosphere, the PCB levels in
soil  required to maintain the same level of exposure will be less than those
expected when the PCB-contaminated material is covered with low-permeability
material of appropriate thickness.  The cover would also serve as a deterrent
to soil ingestion and  direct dermal  contact.
     The depletion of  PCBs from soil  caused by volatilization is accounted for
in the exposure analysis by solving a partial differential  equation simulating
PCB vapor diffusion through the soil  air-phase pores, and the distribution of
PCBs  between air and soil phases.  Boundary conditions assume that  the air-phase
resistance is relatively small  compared to the diffusional  resistance in the
soil  air-phase pores.   The available experimental  data reasonably follow the
time-emission rate relationship predicted from the models based on  this assump-
tion.  Since the depletion rate varies over time,  it is averaged over the
exposure period.  Depletion averaged over a period of time should lead to a
lesser inhalation exposure than that based on the  model assuming that depletion
does  not occur.
     The worst-case emissions would occur when the contaminated soil  is initial-
ly exposed to the atmosphere and the soil is contaminated up to the conditions
exhibiting saturation  vapor pressure.  A constant  emission rate can be assumed
if the vapor-phase concentration maintains a constant value at the  surface of
                                      1-4

-------
soil contamination for time-varying emission  rates.   Calculations  corresponding
to Classes (1), (2), and (3)  for exposure  possibilities  witn  surface  contamina-
tion are repeated at an assumed 25-cm thickness  of a  soil  cover  initially  free
from PCB contamination.  Among many factors affecting the  emission rate  (in-
cluding vapor pressure, soil-air partition coefficient,  Henry's  Law constant,
etc.), the value of the soil-air partition coefficient shows  the most wide-
ranging variation, because of the variation of the experimental  soil-water
partition coefficient available in the literature  for soil  textures ranging
from 40 to 1,000 cm3/g.
     The method for determining the permissible  PCB  levels  in soil,.which  com-
bines the routes of soil ingestion, inhalation,  and  dermal  exposure,  has been
computerized to avoid the necessity for hand  calculations.
     The results of these computer calculations  are  summarized in  Tables 1 and
2, which have been prepared using different combinations of the  following
variables:
     (1)  Surface contamination representing  a situation where the contaminated
          soil surface has been left uncovered after  removal  action.
     (2)  25-cm (10-inch) clean cover applied, representing a situation  in
          which clean soil material is used on top of the  contaminated soil
          surface.
     (3)  Two different soil  ingestion rates  (3  and  0.6  g/day) for Class
          (l)(a), corresponding to sites accessible  to children.
     (4)  Different AI levels (short-term  AI, and AIs at different cancer  risk
          levels).
     (5)  Four Aroclors (Aroclor 1242, 1248,  1254, and 1260).
     (6)  Two selected values of the soil-air partition  coefficient,  repre-
          senting the high and low values.
                                      1-5

-------
en
                                                   TABLE 1.   PERMISSIBLE  PCB  SOIL CONTAMINATION LEVELS
                                                            (UNCOVERED  SURFACE CONTAMINATION)
Permissible levels (iig
Noncancer short-term8
acceptable Intake (uq/day)b
Location and
route of human 100 700
exposure for child for adult
On the contaminated site
- Soil IngestlonC. 25-100' 510-730
Inhalation6
- Soil Ingestlond, 42-420 2100-3000
Inhalation6
- Inhalation only6 47-vs9 vs
0.1 km from vs vs
PCB/g soil) corresponding to
Cancer risk specific doses (ug/day)
0.00175
(ID'7 risk)

0.008-0.01
0.01-0.06
0.01-0.2
2.0-220
0.0175
(10-6 risk)

.08-0.1
0.1-0.6
0.1-2.0
90-2. ZxlO4
0.175
(10-5 risk)

0.8-2
1-6
1-20
7.7x!03-vs
V5
(10-* risk)

8-17
35-61
77-470
8.7x!05-vs
             contaminated site
             -  Inhalation only6

            1 km from                   vs9            vs                220-1.3xl03   2.2xl04-1.3xl05    vs         vs
             contaminated site
             -  Inhalation only6


            aShort-term 3 10-day Intake.
            bBased on average weights of  10 and 70 kg for  a  child and  an adult,  respectively.
            cCMIdren ages 1-5, with pica (consuming 3 g soil/day).
            dCM1dren ages 1-5,  without  pica (consuming 0.6 g soil/day).
            6lnhalat1on rates  are assumed to be 20 m'/da
 		 ._ __ __ ..3/day for the short-term and  longer-term noncancer  exposures;
 all other (more chronic) exposures assumed to be 10 m'/day  as  a result  of  182  days'  exposure  per year.
^Ranges result In each case because 1) four PCBs (1242,  1248,  1254,  1260) are considered, each with  a different
 vapor pressure, and 2) high and low values for soil-air partition coefficient  are used  In  the calculations.
9vs denotes no theoretical upper-bound limit.  Practical reasons require no free-flowing PCB liquids for  the  limit.

-------
                                       TABLE 2.   PERMISSIBLE PCB SOIL CONTAMINATION LEVELS
                                                    (25-cm-THICK CLEAN COVER)
Permissible levels (iig PCB/g soil) corresponding to
Noncancer short-term3
acceptable Intake (ug/day)b
Location and
route of human
exposure
On the contaminated site
- Soil IngestlonC,
Inhalation6
- Soil Ingest lond,
- Inhalation6
- Inhalation only6
0.1 km from
100 700
for child for adult

110-200? 800-1400
450-vs9 3100-vs
vs vs
vs vs
Cancer risk specific doses dig/day)
0.00175
(10-7 risk)

0.01-0.2
0.02-0.6
0.02-1.0
1-vs
0.0175
(10-6 risk)

0.1-2.0
0.2-6.0
0.2-vs
620- vs
0.175 1,75
(10-5 risk) (10-4 risk)

1-17 22-vs
1.0-48 93-vs
2.0-vs 770-vs
vs vs
 contaminated site
 - Inhalation only6

1 km from
 contaminated site
 - Inhalation only6
vs
              vs
                                 vs
                                                   vs
                                                                  vs
                                                                             vs
aShort-term a 10-day Intake,
bBased on average weights of 10 and 70 kg for a child and an adult, respectively.
cChildren ages 1-5. with pica (consuming 3 g soil/day).
dCh11dren ages 1-5. without pica (consuming 0.6 g soil/day).
elnhalatton rates are assumed to be 20 m3/day for the short-term and longer-term noncancer exposures;
 all other (more chronic) exposures assumed to be 10 m'/day as a result of 182 days'  exposure  per year.
^Ranges result In each case because 1) four PCBs (1242,  1248, 1254. 1260) are considered,  each with  a  different
 vapor pressure, and 2) high and low values for soil-air partition coefficient are used In the calculations.
9vs denotes no theoretical upper-bound limit.  Practical reasons require no free-flowing PCB  liquids for  the  limit.

-------
     (7)  Exposures for 10 days after cleanup  or spill  of  contaminants  for
          short-term advisories.
     Table 1 shows the range of values for permissible  PCB concentrations  in
soil when the soil is contaminated up to the surface in contact  with  the atmo-
sphere and is left uncovered.  Table 2 represents the case where the  contami-
nated soil left at the site, or after remediation, is covered  with  a  25-cm
(10-inch) clean soil  layer.  The ranges in both  tables  result  from  the  use  of
four Aroclors and the use of high and low values for the soil-air partition
coefficient.  Other factors reflected in the ranges are differences in  vapor
pressures and Henry's Law constants for each Aroclor.  The permissible  PCBs
levels in soil  specific to each combination of the scenarios are compiled  in
Appendix C, as  obtained from computer simulations.
     The symbol "vs"  in Tables 1 and 2 indicates that no upper-bound  limit  for
PCB concentrations in soil can be derived from the exposure evaluation, because
the PCB concentration in soil is above the vapor saturation concentration.
There are two reasons for such a result.  First, the emission  rate  cannot
exceed the upper-bound value which can be expected when the air-phase concen-
tration of PCBs at the contaminated soil surface is maintained at the vapor
saturation point.  The concentration at the vapor saturation point  corresponds
to the vapor pressure concentration.  Second,  when the  cover is  applied, not
only is the emission  rate retarded, but also the concentration of PCBs  in  soil
being ingested  is controlled by the amount of  PCBs adsorbed on soil in  equili-
brium with the  air phase being emitted.  Therefore, the concentration of PCBs
in the initially clean soil material cannot exceed the  concentration  in equi-
librium with saturated vapor.
     In actuality, the "no upper limit," or the  level above vapor saturation,
designated by vs, should be interpreted with great care.  The  assumptions  used
                                      1-8

-------
in the exposure evaluation are critical.   They  include  but  are not  limited
to:  (1) no soaking of clean  cover  by  liquid  PCBs  for the thickness  of  25 cm;
(2) no disturbance of cover material  by  construction  activities or  children
digging the ground; (3) no exposure to initial  spills when  25  cm of  clean cover
(Table 2) is assumed; (4)  no  population  enters  the area within the  respective
radius of distances from the  site;  and (5) the  cover material  is at  least
equivalent to soil material.
     From a practical point of view,  Assumption 1  is  tantamount to  requiring
the presence of no free liquids  in  the soil,  which may  otherwise result in the
phenomenon of "wieking."  Since the ranges shown  in Tables  1 and 2  are  depend-
ent upon the type  of Aroclors and the  values  of the soil-air partition  coeffici-
ent, site-specific or Aroclor-specific information should be used to establish
an appropriate level  of PCBs  for that  particular condition.  The methodology
for performing site-specific  exposure  evaluations  is  presented.   Computer
outputs for the selected Aroclors under  the ranges and  conditions of common
environmental concern are presented in Appendix C, and  can  easily be used to
find the permissible concentrations in soil suitable to particular  situations.
     Table 1, for  example, can be interpreted as  follows:
     (1) When the  site is  amenable  to  access  by children with  possibilities of
ingesting the contaminated soil  exposed  to the  atmosphere,  and when  exposure
occuring to the children by inhalation and dermal  contact is accounted  for, the
permissible PCS levels in soil should  range from 25 to  100  ug/g and  42  to
420 ug/g for prevention of noncancer  effects  from  10-day exposure for a child
with an average weight of 1U  kg  ingesting  soil  at  the rates of 3 and 0.6 g/day,
respectively.  For cancer  effects,  permissible  levels in soil  for a  lifetime
exposure to PCBs resulting from ingestion  of  and dermal  contact with contaminated
soil and inhalation of contaminated air  should  range  from 0.08 to 0.1 u9/9

                                      1-9

-------
 and 0.1  to  0.6  pg/g,  corresponding  to an upper-bound risk estimate of 10"6 at
 assumed  soil  ingestion  rates of  3 and 0.6 g/day, respectively.  The specific
 level  will  depend on  the types of Aroclor present, the likely ingestion rate,
 and the  extent  of soil-air  partitioning.  For sites in which there is no possi-
 bility of soil  ingestion, PCB levels in soil, based on the inhalation route
 only,  should  range  from 47  u9/9  to  no limit value for a 10-day exposure for a
 child  with  an average weight of  10  kg, and correspond to no limit value for an
 adult  with  an average weight of  70  kg.  The permissible levels of PCBs in soil,
 based  on the  inhalation pathway  only, range from 0.1 to 2 ug/g, corresponding
 to a lifetime AI at  a risk  factor of 10~6.  Again, the level will be dictated
 by the types  of Aroclor present  and the specific characteristics of the site
 involved.
      (2)  If  there  is no possibility of a population entering the contaminated
 site within a radius  of 0.1 km from the site, the PCB levels in soil  can remain
 at no  limit value and 90 to 2.2  x 104 ug/g, without exceeding 10-day AI and
 lifetime AI at  10-6  risk, respectively.
     Similar  interpretations can be made for the results applicable to sites
 without  affected population up to 1 km from the site, and to the carcinogenic
-risks  listed  at lO'4, 10'5  and 10'7.
     The short-term  AI  levels (100 yg/g day for a child and 700 yg/g day for
 an adult) used  in this  report to develop 10-day advisories based on noncancer
 effects  are derived  from animal  studies, which collectively indicate that
 the experimental threshold  for adverse effects of Aroclor 1254 is at or near
 a  dose of 1.0 pg/kg  body weight.  Using this dose as a No Observed Adverse
 Effect Level  (NOAEL)  and a  safety factor of 100, the 10-day AI levels for non-
 cancer effects  described above (100 and 700 u9/day) were computed and serve
                                      1-10

-------
Advisory levels for 1-day and lifetime noncancer effects cannot be derived at
this time because of the insufficiency of the available  data.   However,  in view
of the experimental duration, the 10-day advisories may  v/ell  be used for the
1-day advisories.
                                      1-11

-------
                                2.   INTRODUCTION

     The purpose of this document is fourfold:   (1)  to provide background
information compiled in the process of developing permissible health  advi-
sories for polychlorinated biphenyls (PCBs)  in  soil  and drinking water,  in
response to a request from the Office of Emergency and Remedial  Response,  (2)
to outline the procedures used in developing the advisories,  (3) to list per-
tinent input data necessary in carrying out  the exposure analyses and in set-
ting the allowable concentration limits, and (4) to  present  an outline summary
of the results obtained from computer simulations of the techniques used.
     The information and methods presented are  intended for  use in setting  safe
advisory levels to protect public health from short-term, longer-term, and
lifetime exposures to PCBs released from hazardous waste facilities or from
spills at previously contaminated sites.  Particular interests pertain to
levels of PCBs allowable in contaminated soil,  and the potential of PCB  migra-
tion to groundwater from PCB-contaminated or hazardous waste  facilities. These
advisories are not concerned with setting PCB limitations in  sediments contami-
nating surface water, which can be  a source  of  bioaccumulation of PCBs in
aquatic animals.
     The analyses presented in this report provide the basis  for deriving PCB
levels allowable in soil and drinking water, which are likely to be primary
sources of exposure pathways.  PCB  problems  that may exist in rivers  and estu-
aries because of contaminated sediments are  not dealt with in these analyses.
The analyses mainly address health  concerns  at  hazardous waste sites  or  at
sites with contaminated soil.  The  primary health impacts considered  include
adverse impacts associated with ingestion of contaminated soil,  inhalation  of
ambient air, and dermal contact with the soil.   Other exposure routes, such as

                                      2-1

-------
drinking water, food intake,  and  ingestion  of  bioaccumulated  fish  are  consider-
ed to the extent that the pathways  are  relevant.   PCBs  that have migrated  from
contaminated sites to various exposure  media are  evaluated for  short-term  and
lifetime impacts to arrive at corresponding advisory  level values.
     The total  dose of PCBs is obtained by  summing each dose  from  all  major
exposure pathways, and is compared  with acceptable intakes (AI) judged from  the
health effects  information available  in the literature.  A longer-term AI
considered most appropriate in deriving 10-day noncarcinogenic  advisories  is
used in the exposure evaluation.  Advisories for  protecting against  carcino-
genic risks are similarly derived at  various risk levels based  on  the  potency
factor.
     This report is not intended  to address the achievability of the safe
levels developed.  Although the report  contains a brief statement, taken from
the available literature on analytical  capability, control technology, and
environmental distribution of PCBs, the data base seems insufficient to make a
generalization  concerning the level of  PCB  cleanup achievable in practice.
     The Exposure Assessment  Group  distributed three  earlier  drafts  of this
document under  the title of "Development of Health Advisories for  Polychlorin-
ated Biphenyls  (PCBs)" for internal review  and comment  on May 9, 1985  and  July
25, 1985, and under the title of  "Development  of  Health Advisories for Poly-
chlorinated Biphenyls (PCBs)  Cleanup" on December 16, 1985.   As a  result of
the comments from the Office of Drinking Water, on the  December 16,  1985 draft,
the 1-day advisories have been replaced by  10-day advisories  because no data
could be found  indicating that 3,4,5,3',4',5'-hexachlorobiphenyl,  used for
development of  the 1-day advisories,  is a component in  commerical  Aroclors.
This final draft reflects changes made  to incorporate comments  from  the Office
of Toxic Substances, the Office of  Emergency and  Remedial Response,  and OHEA's
                                      2-2

-------
Environmental  Criteria and Assessment Office.
                                      2-3

-------
                           3.   CHEMICAL COMPOSITIONS

     A polychlorinated biphenyl  (PCB)  is any member in a  family of organic
compounds with two or more chlorine substitutions on biphenyl  rings,  and can
be typified by the following chemical  structure:
                          X      X      X      X
     The symbol, x, in the structural  formula  represents  possible positions  of
chlorine that can be substituted  for hydrogen,  which  is one  of the basic  ele-
ments of aromatic hydrocarbons.   Based on  the  possible distribution of substi-
tuted chlorine atoms on two benzene  rings  of biphenyl, it is calculated that
there could theoretically be 209  types (congeners)  of PCBs.
     Patents disclose that PCBs are  prepared by the chlorination  of biphenyl
in the presence of a catalyst.  The  process  yields  a  complex mixture of
chlorinated biphenyl compounds.   It  is unlikely,  however, that all combinations
of chlorinated biphenyls would  be formed  in  the chlorination process.   Although
the crude mixture is purified to  remove reaction  impurities, the  resulting
product is still a mixture of chlorinated  biphenyls in various proportions.
Their compositions depend upon  the chlorination conditions.
     Commercial-grade PCBs, consisting of  mixtures  of different composition,
are sold under the trade name Aroclors.  Impurities such  as  chlorinated dibenzo-
furans and chlorinated naphthalenes  are known  to  exist in commercial  PCBs.  The
sole producer of Aroclors in the  United States  for  the period 1957 to  1972 was
the Monsanto Chemical Company.  Their products  are  characterized  by four-digit
numbers.  The first two numbers represent  the  type  of molecule (12 = biphenyl-

                                      3-1

-------
 based;  54  =  terphenyl-based;  25,44 = blends of PCBs and chlorinated terphenyls);
 and the last two  digits  refer to the percentage of chlorine by weight.  PCB
 products are also manufactured  in other countries, including Germany, France,
 Japan,  and the  U.S.S.R.
      Table 3 illustrates  approximate compositions of individual biphenyls for
 some Aroclors (U.S.  EPA,  1976b).  Although one might expect some 140 to 150
 separate congeners in  an  Aroclor, the actual analysis of Aroclor 1248, for
 example, identified  less  than 50 peaks in the high-resolution gas chromatograph
 using a typical Aroclor  1248  sample (U.S. EPA, 1976b).  No compounds which can
 be formed  by addition  of  chlorine rather than substitution were found in a
 detailed study  of PCBs (U.S.  EPA, 1976b).  It is suspected that the conditions
 prevailing during industrial  manufacturing of PCBs do not favor the formation
 of addition  compounds, or that  these latter compounds might have been destroyed
 in the  step  used  to  purify the  Aroclor.  In constrast to the analysis shown in
 Table 3, another  publication  reports an analysis of Aroclor 1221 to contain
_12.7% biphenyl, 47.1%  monochlorophenyls, and 40.2% dichlorophenyls (Hutzinger
 et al., 1974).
      Major PCB  components in  foreign products bearing the names of Kanechlor
 and Phenoclor for Japanese and  French products, respectively, have been identi-
 fied.  The number of the  major  components separated from Kanechlor 400 is five,
 and that from Phenoclor  DP6 is  seven.
                                       3-2

-------
                 TABLE 3.   APPROXIMATE COMPOSITION OF AROCLORS
Percent by weight for Aroclcr
Chlorobiphenyl
C12H10
C12H9C1
C12H8C12
Ci2H7Cl3
C12H6C14
Cl2H5d5
C12H4C16
C12H3C17
C12H2C18
Cl2Hld9
C12C110
1221
11
51
32
4
2
0.5
ND
ND
ND
ND
ND
1242
<0.1
1
16
49
25
8
1
<0.1
ND
ND
ND
1248 1254
<0.1
<0.1
2 0.5
18 1
40 21
36 48
4 23
6
ND
ND
ND
designation
1260 1016
<0.1
1
20
57
21
12 1
38 <0.1
41 ND
8 ND
1 ND
ND
ND = non-detectable.
                                      3-3

-------
                                 4.   PRODUCTION

     Commercial  production of PCBs  from the  starting  material  benzene  was  begun
in the 1920s by  Swann Research,  Inc.,  of Annington, Alabama, which  referred  to
these products under the trade name  Aroclor.   PCBs were  manufactured at  that
location by Swann Research, Inc., and  its successor,  Monsanto  Chemical Company,
until the plant  was closed in 1971.   Monsanto  continued  production  at  another
plant at Sauget, Illinois, until  1977.  The  only  other known manufacturer  of
PCBs is Geneva Industries of Houston,  Texas, which manufactured  PCBs from  1972
through 1974.
     The domestic sale of Aroclor products peaked to  33,000 metric  tons  in
1970, and has declined since then due  to restrictions on the use of PCBs
(Hutzinger et al., 1974).  PCBs  were available commercially as mixtures  (Aro-
clors) of 20 to  75 chlorinated biphenyls, and  were marked according to the
weight of chlorine contained in  the  mixtures.  These  commercial  mixtures in-
cluded Aroclors  1242, 1254, 1248, 1260, 1262,  1268,  1221, 1232,  and 1016,  in
descending order according to domestic sales.   In the year of  peak  production,
57% of the Aroclors produced were in the form  of  Aroclor 1242  (U.S. EPA, 1980a).
The production in Japan and the  annual consumption in Finland  are estimated  at
26 million pounds and 0.5 million pounds per year, respectively. PCB-1016 (41%
chlorine) is a more recent product,  and its  sales prevailed for  the period
1972-1976.
     PCBs may be formed as side-products in  other manufacturing  processes  in-
volving the use  of chlorinated benzene or biphenyl in the reaction  step.  For
example, some of the trichlorobenzene  used as  a solvent  in the manufacture of
the dry pigment  phthalocyanine blue  is converted  to  PCBs during  the reaction.
PCBs formed can  contaminate the  pigment product at concentrations from a few
                                      4-1

-------
parts per million to as much  as  0.1%.   Similarly,  dichlorobiphenyl  is  formed
in the manufacture of diarylide  yellow pigments  as a  product  of  side  reaction
with the reactant dichlorobenzidine.   The process  of  chlorinating water which
contains biphenyl in a compound  used  as a dye carrier in  dyeing  polyester
fibers can form PCBs as a side-reaction product  which can contaminate  the
water.  No natural sources of PCBs  have been  identified.
     PCBs have been imported  into the  United  States for use in various applica-
tions.  Decachlorobiphenyl was imported from  Italy for use as a  wax filler  in
the investment casting industry  until  1976.   PCBs  imported from  France are  used
in mining machinery cooling systems.   The percentage  of imported PCBs  over  the
total domestic sales for the  period 1971 through 1975 in  the  United States  is
in the range of 1.6% to 2.7%  (U.S.  EPA, 1976b).
                                      4-2

-------
                                    5.  USES

     Products containing  PCBs  have  been used  in agriculture and industry for
decades.  Their use is  are mainly attributable to high chemical stability and
physical properties desirable  in certain applications.  These properties include
nonflammability, high dielectric constant, plasticizing capability, and ease
of volatilization under heated conditions.  Since 1930, PCBs have been exten-
sively used as dielectric fluids in  electrical transformers and capacitors,
and have also been used for  a  variety of other purposes, including use in heat
transfer and hydraulic  systems, in  the investment casting industry, and as
plasticizers and solvents in sealants and adhesives.  PCBs are also used as
flame retardants in the manufacture  of hard plastic products in which heat
resistance is desired,  as a  dye carrier in carbonless copy paper, and as a
plasticizer in paints.
     Several published  sources provide a comprehensive breakdown of uses for
different types of Aroclors  (Hutzinger et al., 1974; Nisbet and Sarofim, 1972;
Versar, Inc., 1977). The most widely used Aroclors were 1242, 1248, 1254, and
1260.  Aroclor 1016 was used after  1970 but in much smaller quantities than the
four types mentioned above.  A Monsanto marketing bulletin on PCBs, published
in the 1960s, also described their  possible use as gaskets and packing materi-
als; as vehicles in graphic  arts; as impregnation agents; as moisture-proof
coatings; as wax substitutes;  as de-dusting agents; in insecticides; in abra-
sives, lubricants, and  cutting oils; in inks; in mastics; and in tank coatings
(Monsanto Chemical Co., undated).   A number of other uses have been patented.
     In the United States, there are 17 companies that have used PCBs in the
manufacture of askarel  capacitors,  and 13 companies that have used PCBs in the
manufacture of askarel  transformers. According to one study, in 1976 approxi-
                                     5-1

-------
mately 25 investment casting foundries (out  of a total  of 135 in the United
States) used PCB-filled waxes in the manufacture of metal castings (U.S. EPA,
1976a).
     In 1971, because of environmental concerns, the manufacturer voluntarily
restricted the sale of PCB products for use  only in "closed" systems, which
include electrical  transformers and capacitors with insulating fluids that
contain PCBs.  These two applications account for all  of the current use of
PCBs in the United States.  The company was  on a schedule to phase out pro-
duction of all PCBs by 1979.  The cessation  of the production will reduce the
amount of PCBs being released into the environment, but millions of pounds of
PCBs are still being used in electrical insulation applications.  The environ-
mental contamination by existing PCBs, and their environmentally safe treatment
or disposal, continue to be of concern.
                                      5-2

-------
                                  6.   DISPOSAL

     A material balance performed on  the amount  of  PCBs  produced,  sold,  and
purchased provides an estimate of the amount  of  PCBs  lost  or  disposed  of in
the manufacturing process.   The total estimated  amount  reported  to have  been
disposed of or lost for the year 1974 is about  3.8  million pounds  (U.S.  EPA,
1976b), of which about 1.8 million pounds are estimated  to have  been  land-
disposed, and the rest to have been incinerated.
     The 1.8 million pounds of PCBs in the land-disposed wastes  generated from
the manufacturing process amounts to  only a small fraction of the  total  PCBs
sent to land disposal facilities.  The total  land-disposed amount  of  PCBs for
the year 1976 is reported to have been about  12  million  pounds (U.S.  EPA,
1976b).  The most important source of PCB waste  has been capacitors that have
failed or become obsolete,  or that are contained in obsolete  equipment.   Other
PCB wastes include solid wastes from  PCB manufacturing  facilities  and  from
operations using PCBs in non-electrical  applications.
     The data base for the WET Model, prepared  by SCS Engineers  (Undated),
shows that PCB fluids containing 50%  PCB-1254 sent  to hazardous  waste  treatment,
storage, and disposal facilities (TSDF)  amount  to about  4,500 tons per year.
This waste competes for the capacity  of  TSDF  regulated  under  Subtitle  C  of the
Resource Conservation and Recovery Act.   Water  effluents from PCB  production
and first-tier use facilties are relatively small compared with  the amounts
being disposed of in landfills.  Severe  local impacts are  evident  by  the dis-
charge into rivers of these effluents.  PCBs  are now  found in the  sediments,
water column, and biota in the rivers.  A few examples  of  current  PCB  problems
include the Hudson River and the New  Bedford, Massachusetts,  harbor.   As a
result of a strong tendency of PCBs to adsorb on sediments, and  of sediment
                                      6-1

-------
migration, PCB problems are identified  farther downstream  from the discharge



points.



     Twenty spills involving PCB  products  have been identified (U.S. EPA,



1976c).  These spills occurred in transformer installations from trucks  and



railroad cars while they were en  route  to  their  destinations, and from leaking



drums.
                                      6-2

-------
                      7.   CHEMICAL AND PHYSICAL PROPERTIES

     The chemical  and physical  properties  of  PCBs  can  be divided  into two
groups:   (1)  those relevant  to  the commercial and  industrial use  of  PCBs,  and
(2) those that are needed  in exposure evaluation and hence  in developing media-
specific safe level-advisories  for PCBs.   The latter properties will be briefly
summarized herein.
     The widespread  distribution  of  PCBs in the environment suggests that  the
major route by which PCBs  are transported  from treatment, storage, and dis-
posal facilities is  through  the atmosphere in the  form of volatilized vapor
and adsorption on particulate matter.  Vapor  pressure  is one of the  important
properties affecting volatilization.  The  available vapor pressure data for
commercial Aroclors, as  reported  in  the  literature, have been compiled and
are presented in this chapter.  Vapor pressure, as distinguished  from partial
pressure or true pressure,  refers to the maximum vapor-phase pressure achiev-
able under equilibrium conditions at the soil-air  interface.
     Experimental  data (U.S. EPA, 1980a) suggest that  PCBs  are strongly ad-
sorbed on earth materials,  including soil. PCBs adsorbed on soil, or present
in the soil mixture, will  be subject to  ingestion  if the contaminated sites  are
accessible to children or  to adults  with habitual  pica.  The bioaccumulation
factor (in the food  chain  and in  aquatic biota) is also an  important physical
parameter when there is  a  likelihood of  PCB transport  in water.
     As pointed out  previously, there are  a number of  congeners for  each of  the
Aroclors.  Thus, the properties listed herein for  Aroclors  represent averages
over the various species that constitute the  mixtures.  The observation that
environmental samples have  contained more  chlorobiphenyls with high  chlorine
levels than is characteristic of  freshly manufactured  Aroclors is attributable
                                     7-1

-------
in large part to the possible metabolism and  volatilization  of  lower  chlorine
species, coupled with enhanced sorption  of  species with  more chlorine.
     The more common types of Aroclors are  shown  in  Table  4. Thirteen Aroclors
were listed in a manufacturer's booklet  (Monsanto Chemical Co.,  undated).
These compounds range from oily liquids  to  white  crystals  and hard transparent
resins, and generally have similar chemical and biological characteristics.
     The properties and parameters commonly needed in  estimating the  environ-
mental fate and transport of a given chemical  are vapor  pressure, solubility
in water, soil-water partition coefficient, and bioaccumulation  factor.  These
properties of PCBs, and other relevant properties, are shown in  Table 4  (Burk-
hard et al., 1985; MacKay and Leinonen,  1975;  Hutzinger  et al.,  1974; Monsanto
Chemical Co., undated; Hwang, 1982; U.S. EPA,  1979a).  Information on addition-
al physical and chemical  properties such as viscosity, softening points, and
other factors, can be found in references authored by  Hutzinger  et al. (1974),
Monsanto Chemical Co. (undated) and U.S. EPA  (1980a).
     The vapor pressure of PCBs and their solubility in  water are low, and
tend to decrease as the number of chlorine  substitutions on  the  phenyl rings
increase.  Aroclors are soluble in most  aliphatic and  aromatic  solvents, and
are highly resistant to the action of strong  alkali  or acids, or high tem-
peratures.  Aroclors subjected to bomb tests  are  reported  to have shown  no
evidence of oxidation (Hutzinger et al., 1974).   PCBs  have been  shown to adsorb
relatively rapidly and strongly to various  materials,  including  soil, wood,
plastic, and glass (Hutzinger et al., 1974).
     Partition coefficients indicating a measure  of  partitioning under equili-
brium conditions between PCBs at the interfaces of air-soil, air-water,  water-
soil media are important parameters in exposure analysis.  Experimental  data
are scarce.  Data for the distribution between air and water in  the form of
                                      7-2

-------
                                TABLE 4.   CHEMICAL  AND PHYSICAL  PROPERTIES  OF  PCBs
PCB
PCB-1016
(Arochlor
1016)
PCB-1221
PCB-1232
PCB-1242
PCB-1248
PCB-1254
PCB-1260
PCB-1262
PCB-1268
PCB-1270
PCB-2565
PCB-4465
PCB-5442
PCB-5460
2, 2', 5,5'-
tetra-
chloro-
blphenyl
? 7 ' 1 & F, _
£.,£. ,O,M,J,-
penta-
chloro-
biphenyl
Molecular
weight


257.9
200.7
232.2
266.5
299.5
328.4
377.5















Specific
KQW gravity


24,000
12,000 1.182
35,000 1.266
380,000 1.380
1,300,000 1.445
1,070,000 1.538
14,000,000 1.620
1.646
1.810
1.947
1.727
1.712
1.434
1.740








Solubility*
in water
(mg/L)


0.42
15.0
1.45
0.24
5.4x10'^
1.2xlO-z -0.03
2.7xlO-3










4.6x10-2



2.2x10-2
Vapor Henry's law
press. (mmHg) constant (atm.
at 25°C m3/g mol )


4xlO"4
6.7xlO-4,
4.06x10'; Ah
4.06xlO'J 5.73xlO'4u
4.94x10^ 3.51xlO'^f
7.71X10'5 8.37x10'^
4.05xlO-5 7.13x10-3















aHutzinger et al., 1974; Monsanto Chemical  Co.,  undated.
bMacKay and Leinonen, 1975.
cHwang, 1982, and U.S. EPA, 1980c.

Bioaccumulation factor:  31,200 L/kg.

Soil-water partition coefficient (U.S.  EPA, 1980a):  22 - 1938 L/kg.

-------
Henry's Law constant and water and soil, exist for some selected Aroclors.

Experimental data measuring the distribution of PCBs between air and soil are

nonexistent.  Estimates of air-soil partition coefficient can be calculated

based on Henry's law constant and soil-water partition coefficient using one of

several empirical relationships.  The Henry's Law constant and soil-water par-

tition coefficient, in turn, are dependent on water solubility and percent

organic carbon in soil, respectively.

     The Henry's Law constants shown in Table 4 are based on information in

MacKay and Leinonen (1975) for PCB-1242, PCB-1248, and PCB-1260; and in a U.S.

EPA  research report (1980c) for PCB-1254.  Burkhard et al. (1985) recently

published a list of calculated Henry's Law constants for  PCB-1242, PCB-1248,

PCB-1254, and PCB-1260.  The value for PCB-1254 is an experimental value ob-

tained in the EPA laboratory in Cincinnati, Ohio, while others represent cal-

culated values.  A comparison of Henry's Law constants for PCB-1254 shows that

the  values in MacKay and Leinonen (1975) and Burkhard et  al. (1985) differ by

a  factor of 10, while those in MacKay and Leinonen (1975) and the experimental

-EPA  value (1980c) differ by a factor of 3.  Since MacKay  and Leinonen's value

is closer to the experimental value, Henry's Law constants for PCB-1242, PCB-

1248, and PCB-1260 are taken from MacKay and Leinonen (1975).

     In the absence of experimental data, the soil-water  partition coefficient
      •j
Kd  (cnr water/g soil), and the air-soil partition coefficient, Kas (g soil/

cm^  air) can be estimated from water solubility and percent organic carbon

 (%OC) in soil, using correlations presented by various researchers.  For exam-

ple, the values for the octanol-water partition coefficient, Kow, can be used

to estimate the values for the soil sorption coefficient  based on soil organic
                       •j
carbon content, KQC (cnr water/g organic carbon), and the bioconcentration fac-

tor  (BCF) by the following formula:

                                      7-4

-------
     log KQC = 0.544 log Kow + 1.377     (Kenaga  and  Goring,  1980)        (1)
     log KOC = 1.00 log Kow - 0.21       (Karickhoff,  1979)               (2)
     log BCF = 0.76 log Kow - 0.23       (Veith et  al.,  1980)             (3)

The Kg- and Kas values then can be estimated by
                                          %OC
                                v   =
                                 d   "OC  V100'                           (4)

                                K   = —                                (5)

where H represents Henry's Law constant.   Since the common unit for H is given
in atm m3/g mol, a conversion factor of 41 (=1/2.44 x 10~2) should be multiplied
in the right-hand side of Eq. (5) when the units for Kas,  H, and K^ are g soil/
cm3 air, atm m3/g mol, and cm3 water/g soil,  respectively.  The multiplica-
tion of Kas by the concentration of  PCBs  in soil will  provide the concentration
of PCBs in the air phase above contaminated soil of interest under equilibrium
conditions.  It should be recalled that the air-soil  partition coefficient,
                              •j
Kfls, has the unit of g soil/cm0 air.  This is equivalent to the ratio of the
air-phase to soil-phase concentration, or (mg/cm3 air)/(mg/g soil).  The esti-
mation of Kas requires the knowledge of Henry's Law constant and the soil-water
partition coefficient as given by Eq. (5).
     A listing of solubilities of each chlorinated biphenyl is shown in Table
5 (U.S. EPA, 1976c).
     There has been much speculation as to the  possible role of photochemical
reaction in the environmental decay  of PCBs.  The results  of a study using
mercury vapor (UV) sources (U.S. EPA, 1976c)  have been difficult to extrapolate
                                      7-5

-------
to environmental  conditions because the radiation  wavelength  is  not  within  the
spectrum of solar radiation at the surface of the  earth.   More recent experi-
ments have been reported using a light  source more closely approximating the
spectral distribution of solar radiation,  but the  values  for  the photochemical
reaction constants are not available.
     The Monsanto Chemical Company has  reported vapor pressure data  only for
high-temperature conditions for Aroclors 1242, 1248,  1254, and 1260  (Monsanto
Chemical Co., undated).  The temperature used in presenting the  data ranges
from 150°C to 300°C.  These vapor pressure values  may be  extrapolated to the
temperatures which are of common environmental concern,  but their accuracies
would be doubtful.
     An EPA report presents kinetic data obtained  from biodegradation experi-
ments using water-soluble Aroclor 1242  (U.S.  EPA,  1980a).  The rate  constants
are presented for biphenyl compounds with  up  to the three chlorine substitutions
present in Aroclor 1242.  The data clearly show that  many of  the chlorinated
biphenyls with four chlorine substitutions do not  biodegrade  after 48 hours of
degradation run.   Inferring from the compositions  of  Aroclors 1242 and 1254 as
given in Table 3, it is conceivable that Aroclor 1242 may biodegrade to some
extent because it contains a substantial amount of chlorinated biphenyls with
two and three chlorine substitutions.   It  appears  that biodegradation of Aro-
clor 1254 would be insignificant or may not occur  because most biphenyl com-
ponents have four or more substituted chlorine atoms.
     PCBs have several properties which make  them  toxic in the environment.
In addition, they can significantly bioaccumulate  and concentrate in the fatty
tissues of all organisms.  For example, the PCB concentration in resident fish
is often many times higher than that in the surrounding water.  PCBs are chem-
ically stable compounds that are able to persist in the environment  for long
                                      7-6

-------
               TABLE 5.  SOLUBILITY OF CHLOROBIPHENYLS IN WATER
Compound                                     Solubility mg/L  (ppm)
Monochlorobi phenyls
  2-                                                5.9
  3-                                                3.5
  4-                                                1.19

Dichlorobiphenyls
  2,4-                                              1.40
  2,2'-                                             1.50
  2,4'-                                             1.88
  4,4'-                                             0.08

Trichlorobiphenyls
  2,4,4'-                                           0.085
  2',3,4-                                           0.078

Tetrachlorobi phenyls
  2,2',5,5'-                                        0.046
  2,2',3,3'-                                        0.034
  2,2',3,5'-                                        0.170
  2,2',4,4'-                                        0.068
  2,3',4,4'-                                        0.058
  2,3',4',5-                                        0.041
  3,3',4,4'-                                        0.175

Pentachlorobi phenyls
  2,2',3,4,5'-                                      0.022
  2,2',4,5,5'-                                      0.031

Hexachlorobi phenyl
  2,2',4,4',5,5'-                                   0.0088

Octachlorobi phenyl
  2,2',3,3',4,4',5,5'-                              0.0070

Decachlorobiphenyl                                  0.015
  4,4'-Dichlorobiphenyl
  +Tween 80 0.1%                                    5.9
  +Tween 80 1%                                    >10.0
  +Humic acid extract                               0.07
                                      7-7

-------
periods.  Impurities in commercial  PCBs  could  amplify  the  PCB  problem because
of their similarity in chemical  structure and  toxicity (Monsanto Chemical  Co.,
undated).
                                      7-8

-------
                        8.   ENVIRONMENTAL DISTRIBUTION

     The release of PCBs into  the  environment through disposal on or in  land,
and through effluent discharges  into waterways, together with their high atten-
uation characteristics  and  long  half-life,  has  resulted in detectable levels in
ambient air, soil,  rivers,  sediments,  and in tissues of fish, wildlife,  cattle,
poultry, and a large portion of  the human population.  Measurable amounts of
PCBs have been found in Antarctic  ice,  showing  that atmospheric transport
over long distances does occur (U.S. EPA, 1976b).  Monitoring shows that the
soils in the rural  and  urban areas where there  is  no record of PCB disposal or
contamination, contain  detectable  amounts of PCBs  (U.S. EPA, 1976c).  One study
estimates that 70%  of the PCB  load to  Lake  Michigan is through atmospheric
transport (University of Wisconsin, 1980).
     The results of soil sampling  show that PCBs are more prevalent in urban
soil than in agricultural soil.  Data  indicate  that PCBs are rarely detected
in agricultural soil, while urban  soils'showed  PCB contaminations up to  about
12 wg/g soil, with  averages ranging from 0.01 to 0.21 yg/g  (U.S. EPA, 1976c;
Carey, undated).  Sixty-three  percent  of the soil  samples showed detectable PCB
levels.  The most prevalent PCBs in soil were Aroclor 1254, and, to a lesser
extent, Aroclor 1260.
     The PCB concentrations in air samples  over Lake Michigan taken during 1977
(University of Wisconsin, 1980)  were  lower  than in those taken in the urban
portion of Milwaukee.  The main  components  were identified  as Aroclors 1242 and
1254, while the particulate-phase  PCBs contained Aroclor 1260 in some instances.
The average concentration of PCBs  in  the air over  Lake Michigan was 0.87 ng/m^
(0.44 to 1.33 ng/m3).  The concentrations of PCBs  in the particulate samples
were similar to those in the air samples.   Air  samples taken in later years
                                     8-1

-------
from Lake Michigan showed an average concentration of  1  ng/m3.   These  concen-
trations are lower than those reported in the ambient  air in  the continental
United States.
     The ambient air concentrations  of PCBs  for urban  Chicago averaged 7.7  ng/
m3.  The average composition in the  air sampled was 86%  Aroclor 1242,  13% 1254,
and 1% 1260.  The particulate-phase  PCBs had a slightly  different composition
for the same Aroclors.   The ambient  air in Milwaukee showed an  average PCB
concentration of 2.25 ng/m3.
                                     8-2

-------
                     9.  ENVIRONMENTAL FATE AND TRANSPORT

     PCBs have been found in samples  of air, water, soil, sediments, fish,
birds, and mammals (including humans) all  over the world (U.S.  EPA,  1980a).
Once released into the environment,  PCBs persist and collect in animal  tis-
sues.  Environmental  problems caused  by PCBs were largely unreported until
1966, when PCB contamination of fish, eagles, and humans was detected.   The
best-documented incident concerning  the effects of ingested PCBs on  humans  is
the case that occurred in Yusho,  Japan, in 1968 (U.S. EPA, 1980a).   Several
other cases have also been reported  (U.S.  EPA, 1980a, U.S. EPA, 1976c).
     Sediments containing PCBs have  been reported in rivers, estuaries,  and
harbors (U.S. EPA, 1981b), in concentrations ranging from 20 to 50,000  pg/g.
Leaching of PCBs could occur, although it  will be constrained by the low solu-
bility limits.  Once PCBs dissolved  in water enter the soil medium,  it  is
possible that further migration will  be severely retarded in view of the high
soil-water partition coefficients.   The retardation factors calculated  from
these coefficients can be used to simulate the arrival time and concentration
decrease in groundwater.  This will  be further explained later.
     Despite their low vapor pressures, PCBs can volatilize into the atmo-
sphere.  The migration of PCBs through air is considered to be  one of the basic
mechanisms by which the ubiquitous presence of PCBs occurs in nature (U.S. EPA,
1980a).  The New York State Department of  Environmental  Conservation (NYSDEC)
analyzed samples of PCB-contaminated  air at several localities.  These  analyses
are shown in Table 6 (NYSDEC, 1979;  U.S. EPA, 1981b).  Concentrations of PCBs
in the ambient air as high as 300 gg/m3 were reported at the disposal site.
The average values ranged from 130 to 0.3  ug/m3.  Concentrations of  suspended
particulates in the air in the vicinity of dump sites were also monitored by
                                     9-1

-------
              TABLE 6.  PCB MONITORING IN AMBIENT AIR BY NYSDEC
                             Max. PCB cone.           Average PCB cone.
   Site                         (ug/m3)                   (ug/m3)


Caputo dump                      300                       130

Fort Miller dump                  35                        24

Remnant area                      10                         9

Moreau site                       15                         5.6

Buoy 212 site                                                0.7 (one sample)

Old Moreau Site (Summer 1979)                                0.3
                                      9-2

-------
 NYSDEC,  using  high-volume  air samplers  (NYSDEC, 1979; U.S. EPA, 1981b).  The
 geometric  annual mean concentration of  particulate matter was 36 to 63 ug/m^,
 while  the  24-hour  averages were  71 to 144 ug/m^.  The amount of PCBs adsorbed
 to the collected suspended particulates was not reported.
     Based on  mass-transfer models, MacKay and Leinonen (1975) calculated the
 half-lives of  PCBs present in solution  in a water column of 1 m depth.  Half-
 lives  provide  some indication of how fast a compound can volatilize from
 solution.   The half-lives  for Aroclor-1242, Aroclor-1248, Aroclor-1254, and
 Aroclor-1260 are reported  to be  12.1 hr, 9.5 hr, 10.3 hr, and 10.2 hr, respec-
 tively.  Half-lives will be longer when depths are greater than 1 m.  The cal-
 culated  half-lives are  for evaporation  from a calm, liquid surface.  Turbulence
 provides exchange  of contaminants between the surface layer and the bulk of the
 water  column.   This exchange results in increased emission rates, thus pro-
 viding shorter half-lives.
      In  addition to the importance of attenuation mechanisms when PCBs inter-
.act with soil, biodegradation is also suggested as a potentially important
 mechanism. Biodegradation studies using pure and mixed microbial cultures,
 and the  resulting  metabolic changes in  PCB compounds, have been summarized by
 Hutzinger  et al. (1974) and Hwang (1982).  Photochemical degradation of PCBs
 in the atmosphere  is also  of interest,  since a number of pesticide compounds
 have been  shown to break down through the photochemical route.  However, very
 little information is available  in the  literature to determine the extent of
 PCB degradation in the  atmosphere.
     The safe  disposal  and treatment of PCBs discarded after their use in
 electrical applications is a matter of  great concern with regard to human
 health.   Incineration techniques are frequently applied to PCB material at
 elevated temperatures and  high  residence times.  Several experiments involving
                                     9-3

-------
pyrolysis of commercial  PCBs have been reported (Buser and Rappe,  1979).   The
PCBs used in the pyrolysis experiments included tetrachlorobiphenyl,  penta-
chlorobiphenyl,  hexachlorobiphenyl,  heptachlorobiphenyl,  and octachlorobiphe-
nyl.  The analysis of the pyrolysis  residues showed the presence of  chlorinated
furan compounds.  However, the researchers concluded that the formation of
furan compounds  is the result of uncontrolled burning of  PCBs, and that the
emission of these compounds can be reduced by controlling the burning process.
     The high-temperature combustion of PCBs, such as in  the case of  transform-
er fires, results in the formation of polychlorinated dibenzofurans  (PCDFs) and
other toxic compounds.  In an experiment studying conditions favoring the
formation of polychlorinated dibenzodioxins (PCDDs), researchers found that the
optimum conditions for the formation of PCDFs are a temperature of near 675°C
at a residence time of 0.8 seconds or longer, with 8% excess oxygen  (Midwest
Research Institute [MRI], 1984).  No conditions for the formation of  PCDDs are
represented.  The report states that detection of PCDDs was occasional and at
low levels.
                                     9-4

-------
                               10.   TOXICOLOGY

     The advisory levels  for  PCB  cleanup  presented  in  this  document  (i.e.,
the permissible PCB soil  contamination  levels)  are  health-based  values.   These
advisories are derived through  calculations  which first  estimate human  risk-
specific (cancer end point) or  acceptable intake  (AI)  (noncancer end  point)
levels, and then determine the  exposure rates which will  effect  these intake
levels.  Risk-specific doses  are  derived  for the  cancer  end point, and  a  10-day
AI level is derived for an approximate  10- to 30-day exposure  considering only
noncancer effects.  A detailed  assessment of the  available  cancer and noncancer
health effects data for PCBs  is presented in Appendix  D.  Only a brief  overview
and major issues will  be  presented  here.
     The determination of risk-specific intake  levels  is  accomplished through
a mathematical process which  makes  use  of a  cancer  potency  factor and a
reflected risk level or levels  to estimate the  intake  level  that would  cor-
respond to such risk levels.  Cancer potency factors for  PCBs  have been
determined through an exhaustive  analysis of animal studies.   Values  have been
calculated by ORD (EPA, 1980b)  to be 4.34 (mg/kg-day)'1  and by OTS  (EPA,  1985b)
to be 3.57 (mg/kg»day)~l. An average of  these  values, or 4.0  (mg/kg«day)~l
is used in the calculations presented in  this document.   A  discussion of  the
data and methods used to  estimate cancer  potency  factors  for PCBs is  included
in Appendix D.  The determinations  made by ORD  and  OTS are  both  based on  the
same animal study (Kimbrough  et al., 1975),  with  only  slightly different
assumptions being incorporated.
     A noncancer AI level was derived for PCBs  during  the preparation of  this
report.  It must be emphasized  that this  AI  ignores the  cancer end point  and
is based on toxicity other than cancer.  The 10-day AI level of  100  ug/day

                                     10-1

-------
for a child and 700 ug/day for an  adult,  derived  for  use  in  this  document,
is based on feeding studies with  rabbits  and  rats in  which  a NOAEL  for  de-
creased reproductive rate, and effects  on thyroid and liver, were evaluated.
These studies are described briefly below.
     Villeneuve et al.  (1971)  found increased incidences  of  fetal  death,  re-
sorptions, and abortions at 12.5  mg/kg/day of Aroclor 1254  in rabbits when
exposed on days 1 through 28 of pregnancy. A dose of 1.0 mg/kg/day appeared
to be without effect.   Collins and Capen  (1980a,  b, c) in a  series  of studies
on thyroid effects in  rats, determined  that 50 ug PCB per g  of diet (~  2.5
to 5.0 mg/kg/day) for  4 weeks  was  associated  with clearly defined adverse
effects, but that doses of 5 wg PCB per g of  diet (~  0.25 to 0.5  mg/kg/day)
were not.  Carter (1983) demonstrated liver hepatomegaly  in  rats  at doses of
20 ug Aroclor 1254 per  g of diet  (~ 2 mg/kg/day)  for  14 days; such  an ef-
fect, in the absence of other  signs of  toxicity (e.g., fatty infiltration of
the liver), might not  be considered adverse.   Grant and Phillips  (1974) ob-
served increased liver  weights in  rats  at doses as low as 5  mg/kg/day Aroclor
1254 given in corn oil  for 7 consecutive  days.  Collectively, these studies
indicate that the experimental threshold  for  adverse  effects of Aroclor 1254
in studies of 30 days'  duration or less is at or  near a dose of 1 mg/kg/day.
Thus, it seems reasonable to use  this latter  dose, a  No Adverse Effect  dose,
as a basis for health  advisories  for Aroclor  1254 for short  exposure durations,
                                     10-2

-------
                    11.   EXISTING  STANDARDS  AND GUIDELINES

     The 1968 incident in which  toxic symptoms  were experienced  by Japanese
people exposed to contaminated cooking oil  gave rise to  a great  deal  of concern
in the United States with regard to hazardous  chemicals.  The  U.S. Food and
Drug Administration (FDA) started  sampling  foods for possible  contamination by
PCBs in 1969, and detected levels  of PCBs  in fish from the  Great Lakes, in milk
caused by use of materials containing PCBs,  and in chickens as a result of the
existence of PCBs in the feed.   The temporary  tolerance  levels for residues of
PCBs proposed by FDA became effective June  29,  1979 (U.S. FDA, 1984).
     In the early 1970s, EPA proposed the  establishment  of  criteria for PCBs
being discharged in industrial effluents,  but  has not so far issued effluent
limitations concerning PCBs.                      	
     The Occupational  Safety and Health Administration (OSHA)  adopted standards
for PCB exposure by industrial workers. Subsequently, the  National  Institute
for Occupational Safety  and Health (NIOSH),  after their  extensive assessment,
recommended lowering the allowable concentration of PCBs in the  workplace.
However, OSHA has not  acted on this recommendation.  The New York State Depart-
ment of Health issued  an interim guideline  for  the allowable ambient  air con-
centration of PCBs to  which individuals may  be  exposed during  the duration of
a PCB reclamation project planned  for the  Hudson River (NYSDEC,  1979).
     The EPA promulgated regulations relating  to manufacture,  processing,
distribution in commerce, use, diposal, storage, and markings  of PCBs and PCB
items.  The regulations  originally became  effective May  31, 1979, and were
later amended.  A complete presentation of  the  effective regulations  can be
found in the latest edition of 40  CFR Part  761  (U.S. EPA, 1984b).  The PCBs
referred to in these regulations include any chemical substances or their
                                     11-1

-------
mixtures containing concentrations of chlorinated biphenyls  of  50  ppm or
greater.  The regulations pertain to prohibitions on manufacturing,  pro-
cessing, distribution in commerce, and use,  and specifically apply to PCB
incinerators, chemical  waste landfills disposing of PCBs,  transformers,
pigments, electrical  and heat transfer equipment, natural  gas pipeline com-
pressors, microscopy mounting medium, capacitors, PCB containers,  and
hydraulic systems (U.S. EPA, 1984b).
     The PCB standards  and guidelines for numerical limitations of PCBs in
food, drinking water, and ambient air existing at the present time are shown
in Table 7.  Because of the complicated nature of the EPA's  regulations pro-
mulgated under TSCA, these regulations are not presented in  tabular  form.
                                     11-2

-------
               TABLE 7.  EXISTING PCB STANDARDS AND GUIDELINES
Exposure pathways                                    Maximum allowable PCBs


Food (FDA standard)3

  Milk fat and dairy products                           1.5 ug/g (ppm)

  Poultry                                               3.0 ug/g (ppm)

  Eggs                                             '     0.3 ug/g (ppm)

  Fish and shellfish                                    2.0 ug/g (ppm)

  Finished animal feed                                  1.0 ug/g (ppm)


Drinking water (New York State)                         1.0 ug/L (ppb)


Ambient air

  Populated areas (New York State guideline)15           1.0 yg/m3
  Workplace (OSHA standard)                             500 ug/m3

  Work site (NIOSH guideline)0                          1.0 yg/m3
aU.S. FDA, 1984.
bNew York State Department of Health, 1981.
CNIOSH, 1977.
                                     11-3

-------
                     12.  EXPOSURE ASSESSMENT  METHODOLOGY

     The presence of PCBs in environmental  media  poses  a  potential  health
risk to humans from the following sources  of  intake:
     • ingesting contaminated soil
     • inhaling contaminated air
     • ingesting contaminated food
     • drinking contaminated water
     • dermal absorption of PCBs in contact with  skin
Other exposure pathways affecting ecological  communities,  such  as  phytotoxicity
to plants, may also need to be considered.   If multiple-route exposures are
possible, the concentrations allowable for a  single-route  exposure should  be
adjusted to"meet the acceptable levels of  acute and  chronic health effect
exposures from all  sources of intake.   The amounts of  each medium  subject  to
human intake used in this analysis are as  follows:   daily  iintake of drinking
water, 2 L/day; daily inhalation of air, 20 m3/day  (U.S.  EPA, 1984b).
     Acute effects  result from short-term  or  long-term intakes.  Carcinogenic
effects are normally treated as resulting  from lifetime intakes.  Safe levels
of PCBs in soil corresponding to 1-day and 10-day acceptable intakes should
be based on consideration for preventing acute health  effects from short-term
and longer-term exposures.  Levels of  PCBs in  soil corresponding to acceptable
intakes for long-term effects can be derived  from the  acceptable daily intake
(ADI) based on long-term health studies for acute effects, or from the car-
cinogenic potency slope based on lifetime  exposure for  carcinogenic effects.
     The long-term risk level for ingestion of contaminated soil over a 70-year
lifetime exposure can be obtained by
                                     12-1

-------
                           69 vrs                ~kt
                    Risk = 7      (GI)  (SM)  Co  e   (POT)  (IR)
                          t=0               (BW) (LT)(6)
where Risk = lifetime risk;  GI  = gastrointestinal  tract  absorption  of  PCBs; SM
= exposure frequency over a  lifetime;  Co  =  initial  concentration  of PCBs  in
soil; k = biodegradation constant (I/day);  POT  =  potency slope  factor  (mg/kg/
day)'1 for PCBs; IR = daily  ingestion  rate  of soil;  BW  = body weight (70  kg for
adults and 10 kg for children); and  LT =  exposure time  over  a lifetime (70
years).  If the contaminant  undergoes  biological  or chemical degradation  in
soil, and follows first-order kinetics in its disappearance  under isothermal
conditions, the contaminant  concentration will  change as a function of time
according to C^"1^.  The summation  in Eq.  (6)  is necessary  in  order to add
all the rt-sks associated with the daily dosage  over a lifetime.  The initial
concentration of PCBs in the soil is calculated at an assumed lifetime risk
according to Eq. (6).  A computer is convenient to use  to sum all the  daily
risks.  The soil ingestion scenario  will  be applicable  to sites which  are
readily accessible, especially by children.  Since the  population is living
on or near the site, the exposure to PCBs due to inhalation  of  contaminated air
cannot be neglected.  In order to account for the inhalation exposure  in  deter-
mining the allowable PCB levels in soil due to  the combined  routes  of  ingestion,
inhalation and dermal absorption, the  ambient air concentrations  at the expo-
sure points are needed.  The concentration  of PCBs in the vapor phase  is  the
result of the volatilization of the  PCBs  from the contaminated  soil and their
dilution by winds.  The dilution factor for ambient air concentrations can be
defined as

                             D = Ca/Cas                                  (7)
                                     12-2

-------
where C, (ug/m3)  = the ambient  air  concentration  at  an  exposure  location,
and Cas (ug/m3) = the concentration of  PCBs  in  air at the soil surface where
the emission occurs.   The value of  Cas  continuously  increases  as the PCB con-
centration in soil increases, until  the concentration of  PCBs  in the air phase
corresponds to that of PCB vapor pressure.   Beyond this point, a further in-
crease in PCB concentration in  soil  will  have a minimal effect on the vola-
tilization rate.
     Once exposure pathways are identified,  exposure evaluation  requires infor-
mation on the levels  of concentration to which  a  given  target  population may  be
exposed.  Each pathway may require  route-specific evaluation.  A general method
for estimating exposures for contaminated sites will be described first.  The
method can be simplified by eliminating those pathways  that are  unimportant or
unrelated in the evaluation of  PCB  advisories.  The  relevant assumptions for
the simplification are described below.
12.1  Estimation of Exposures for Contaminated  Sites
     The combined human intake  of contaminants  from all exposure pathways
should not exceed the acceptable intake (AI, in mg/day) needed for preventing
adverse effects from short-term and lifetime exposures.  The intake from an
individual route when soil is contaminated  can  be expressed quantitatively as
follows:

       i)  Intake by  soil ingestion (mg/day):

                     I, = (CS)(IR x 1(T3)(GI)(SM)(F)
                                                                       (8)

     The term (CS)(IR) in Eq.  (8) represents the  daily  amount  of a contami-
nant ingested resulting from soil ingestion, in ug/day, because  Cs repre-
                                     12-3

-------
sents the contaminant concentration  in  soil  (ug/g)  or ppm (both  units  are
equivalent), and IR is the soil  ingestion  rate (g/day) and GI  is defined in  Eq.
(6).  The factor 10'3 is needed  to convert the unit from yg/day  to nig/day.
The fraction of the ingested contaminants  that will enter human  organs and
systems to cause toxicity is given as 61.   An individual  may not always be
present on the contaminated site over his  lifetime.  The frequency factor of
exposure over a lifetime, SM, represents the fraction of a lifetime that an
individual will be exposed to the contaminants under consideration.  The factor
F is necessary because soil ingestion only occurs during childhood (1  to 5
years of age), and the weight of the human body changes from childhood to
adulthood.

      ii)  Intake by air inhalation of  volatilized contaminants  (mg/day):

                   I2 = (Kas)
-------
partial differential  equation,  as  described  in  the Section  16.
     If the concentration of PCBs  in  soil  is at or above  saturation  conditions,
under which the air phase concentration  of PCBs is equal  to the  vapor  pressure
concentration for a particular  Aroclor  or  a  mixture of  Aroclors,  the further
increase in Cs will not  increase the  ambient air concentration,  assuming  that
other factors, such as temperature, remain constant.  Therefore,  the daily
intake by inhalation remains constant above  the concentration  of  PCBs  in  soil
providing saturated air  concentration.   This concentration  of  PCBs  in  soil,
or the saturated concentration  in  soil  for air  inhalation,  will  be  denoted by
      iii)  Intake by dermal  absorption  (mg/day):

                        I3 = (CS)(CR  x  1(T3)(ABS)(SM)                    (10)

     There are many occasions when  children  playing  in  the  yard  or  adults
working in the garden will come in  direct  contact  with  contaminated soil.
Dermal contact does not  necessarily constitute adverse  exposure.  The
contaminant needs to be  systemic to be  absorbed into the human body and  to
exert toxicity.  In Eq.  (10), the term  (CS)(CR) represents  the contaminant
contact rate with skin in ug/day since  Cs  is in ug/g (=ppm) and  CR  is
the dermal contact rate  of soil  in  g/day.  The factor  10'3  is used  to  con-
vert the contact rate from vg/day to  mg/day, and SM  will  be 1 when  the
short- or longer-term (10-day) exposure  is estimated,  and will be between
0 and 1 when the lifetime exposure  is estimated.
                                      12-5

-------
      iv)  Intake by drinking water (mg/day):

                             lA = (CW)(IW)(SM)
                                                                       (11)
     In Eq. (11) it is assumed that the contaminant in drinking water is
completely absorbed into the human body at  the average daily water con-
sumption rate of IW or the absorption fraction is 1.  In order to relate
the contaminant concentration in groundwater,  Cw, to the contaminant con-
centration in soil, Cs, a fate and transport model  can be used to estimate
the concentration in the leachate entering  groundwater, or

                             Cy, = CL/fg, mg/L
                                                                       (12)

where fg represents a functional relationship describing contaminant trans-
port in groundwater.  This function should  be selected to suit the most
appropriate conditions for the system.   The leachate concentration, CL,
referring to the contaminant concentration  in liquids just before entering
groundwater, should not be confused with the contaminant concentration in
groundwater, which results from mixing  of the leachate with groundwater.
Also, care should be exercised in using groundwater transport models, be-
cause some models will treat the leachate concentration as a boundary con-
dition, while others require the contaminant concentration in groundwater
as a boundary condition, which should be obtained by groundwater monitoring.
When the units of Cw and C|_ are all in  mg/L, then the function fg becomes
dimensionless.  Most leachate from hazardous waste land disposal  sites may
enter groundwater over a finite surface area,  favoring area source models for
simulating pollutant transport in groundwater.
                                     12-6

-------
     There is no reliable method of  predicting the leachate concentration
from the contaminant concentration in soil,  or vice versa.   For the exposure
evaluation, an equilibrium relationship  vetween soil  and leachate will  pro-
vide a first approximation.  Monitoring  data can also be used relating  the
concentrations between leachate and  soil.   An equilibrium condition can be
written as
where KLs is a partition coefficient  in  (mg/kg)/(mg/L).  Eqs.  (11),  (12),  and
(13) are combined to get
                            I4 =     Cs     (IW)(SM)
                                 (fg)(KLS)                              (14)
When the equilibrium condition is  not  appropriate,  it can be modified to
include transport processes between  the  soil  and leachate.

       v)  Intake by fish ingestion  (mg/day):
     At the average daily fish consumption  rate of  IF (kg/day),  and under
the assumption of complete absorption  of the  contaminant associated with
the consumption of fish, the exposure  can be  estimated as

                              15 = (CF)(IF)(SM)                        (15)

where Cp is the contaminant concentration in  fish.   The use of the biocon-
centration factor BCF, (mg/kg fish)/(mg/L water), to relate pollutant con-
centrations in fish and water, gives
                                    12-7

-------
                          I5 = (BCF)(CW)(IF)(SM)                       (16)

where it is assumed that contaminants  are  present  in  water  in  dissolved  form
and that bottom sediments or benthal deposits  on which  pollutants  may  be ad-
sorbed are not directly swallowed by fish.   Under  the condition  of equili-
brium between the pollutant-containing soil  and leachate  which is  generated
from the soil, substitution of Eqs.  (12) and (13)  into  Eq.  (14)  results  in
                                                                      (17)
     The transport functions,  fg,  in  Eqs.  (14)  and  (17)  may  assume distinct
mathematical descriptions, because one pertains to  transport in  groundwater
and the other to that in surface water.

     vi)  Intake by inhalation of  contaminants  adsorbed  on particulates
(mg/day) may be expressed by
                     I6 = (C )(IH)(CS x 10-9)(ABP)(SM)                 (18)
     Contaminant-containing soil  can be airborne  by  blowing winds.   In
addition, toxic substances volatilized from contaminated  soil  can be ad-
sorbed on particulate matter present in the ambient  air.   Exposure  to con-
taminants occurs because of inhalation of air containing  these particulates.
The exposure location could be distant from the source  of emission, or  in
the vicinity of the emission source.  Exposure concentrations  will  change
accordingly.  Another form of exposure relates to inhalation of air con-
                                     12-8

-------
taining participate matter on which volatile  constituents  are adsorbed.

The intake rate can be estimated based on the concentration  of contami-

nants in wind-dispersed soil  or on participate matter,  Cs  ug/g (=ppm),

and the concentration of the  particulates in  the  ambient air, Cp  yg/m^,

as shown in Eq. (18).  The absorption fraction, ABP,  is used because

contaminants present in or on soil  (or particulate  matter) may be bound  on

the solid material, reducing  the contaminant's absorption  rate.   Finally,

the factor 10'^ is a conversion factor to make the  units consistent.



     vii)  Intake by ingestion of vegetables  (rug/day):

     The intake rate due to ingesting IV kg/day of  vegetables, plants, or

agricultural  products containing Cv mg/L of contaminants will be



                             I? • (CV)(IV)(SM)
                                                                      (19)


If it is assumed that equilibrium is established  between the contaminant

concentrations in plant and soil, then the exposure can be modified as:



                          I?  = (KSV)(CS)(IV)(SM)
                                                                      (20)


where Ksv is a partition coefficient defined  as contaminant  concentration  in

plant/total contaminant concentration in soil  (mg/kg  plant)/(mg/kg soil).



    viii)  Intake by ingestion of food meat:

     The contaminant intake at consumption rate of  IM (kg/day) of meat con-

taining Cm (mg/kg) of pollutant is
                                     12-9

-------
                             18 = (Cm)(IM)(SM)                         (21)







Here again, an equilibrium relationship  is  assumed  between the contaminant



concentrations in the animal  body and  plants.  Therefore, the intake  rate



due to meat consumption is
                      IS = (Kv





                                                                      (22)





where Kvm and Ksv are the partition  coefficients  used  to  describe  pollu-



tant distribution between meat and vegetables,  and  the partition between



vegetables and soil, respectively.



12.2  Determination of Permissible Pollutant  Levels in Soil



     The total intake from all possible  exposure  pathways is  set equal  to



the acceptable intake (AI) for short-term and chronic  health  effects;  or







                          AI = l! +  I2 + 13 + •••                     (23)







Eq. (23) can be solved for permissible contaminant  levels in  soil  correspond-



ing to each acceptable intake.  It is possible  that some  exposure  pathways



occur independently of others.  For  example,  a  residence  which  is  located  on



a contaminated site may use drinking water from a clean public  water treat-



ment system, and may thus be free of contaminants found on the  site.   It is



also possible that domestic animals  are  not raised  for food consumption on



the contaminated site under consideration. Under such circumstances,  all



exposure pathways need not be considered.  If exposure pathways of significant



concern are related to soil ingestion, inhalation of contaminated  air,  or





                                    12-10

-------
dermal contact with soil,  as is  the case  for  development  of  PCB  advisories,
Eqs. (8), (9), and (10) can be added to solve for  Cs,

                               (AI)(1000)
       CS=	
            C(IR)(GI)(F) + (Kas)(D)(IH)(ABA x 105)  +  (CR)(ABS)]SM
                                                                       (24)

     The emission rate is  limited by the  air  phase concentration in equili-
brium with the contaminant concentration  in soil.   Once the  contaminant  soil
concentration reaches the  level  at which  the  vapor phase  concentration  in
equilibrium with the soil  is at  the vapor pressure concentration, a further
increase in contaminant concentration in  soil (Cs  > Csm)  does  not increase
the emission rate.  At or  above  this concentration, the ambient  air concen-
tration remains constant regardless of the concentration  of  the  contaminant
in soil.  Under such conditions, Cs in Eq. (9) is  no  longer  a  variable,
and therefore Eq. (24) does not  apply. This  situation can be  remedied  by
considering the intakes by the individual route of exposure  at a constant
value of Cs [Cs = Csm in Eq. (9)] for inhalation exposure, and solving  for
Cs.  The form of the equation will be slightly different  from  that for  Eq.
(24).
                  (AD(IOOO) - (Kas)(Csn|)(D)(IH x 103)(ABA)(SM)
                                           (CR)(ABS)]SM
12.3  Incorporation of Time-Varying Parameters
     The body weight of a human constantly changes  until  maturity.   The cal
culation of AIs from the safe dose level  (SL)  given in  mg/kg«day requires
                                    12-11

-------
an assumption of body weight.  For rigorous treatment, the estimation of

lifetime exposure should take into account changes in body weight.   In this

case, it is convenient to work with SL instead of AI  for exposure calcula-

tions.  For carcinogens with a potency value at POT (mg/kg-day)'^,  the

equivalent SL at an assumed risk level, R (such as 10~6, etc.),  can be

obtained by
                          (SL)eq.  =  R  ,  mg/kg-day                   (26)
                                    POT
The risk level shown represents an upper-bound estimate.   An upper-bound

estimate of risk of 10~6, for example,  means  that upon lifetime exposure

to a contaminant, a person experiences  an  increased maximum risk of devel-

oping cancer in a probability of 1 in one  million.

     Snyder (1975) presented data on the change of body weight  as a func-

tion of age.  A regression analysis on  Snyder's data for  average male weight

provides the following relationship.



                 BW = 3.14 + 3.52 (age), kg for age 0 - 18 yr         (27)


                 BW = 70, kg for age greater  than 18 yr               (28)



     To obtain the daily exposure averaged over an individual's lifetime,

intake rates given by Eqs. (8) - (10),  (10),  (17), (18),  (20),  and (24)

should be divided by the body weight, and  the daily intake per  unit body

weight should be averaged by summing the total intake per unit  body weight

over the period during which exposure occurs  and dividing the result by

LT.  For purposes of illustration, Eqs. (8) and (9) are repeated below:

                                    12-12

-------
       i)  The average daily exposure by  soil  ingestion  per  unit  body weight
in mg/kg-day can be determined  as
             _            (C0e-kt)(IRX103)(GI)(SM)
          BW " 1 day               (BW)(LT)
     Again, in Eq.  (29)  [also  in  in  Eq.  (30)],  the  contaminant  present  in
soil is assumed to  disappear by biodegradation  and  other  reactions,  accord-
ing to first-order  kinetics.  Other  processes affecting the  concentration
in soil are considered in the  exposure analyses for individual  pathways.

      ii)  The average daily exposure by  inhalation of volatilized contami-
nants in mg/kg-day  is calculated  from
          _   _            (Kas)(C0e"kt)(D)(IH  x  1Q3)(ABA)(SH)
          BW ' 1 day               (BW)(LT)
Similar expressions can be written  for other  exposure  pathways.   For  conser-
vative contaminants, the term C0e   *•  in Eqs.  (29)  and  (30)  can be replaced
by Cs.  The total  dose from all  exposures  should not exceed SL,  or (SL)eq>
           cL =  l +  2 + j3 f          for  noncarcinogenic  effects   (30)
                BW   BW   BW   •"
           /SL\    - il + il + il +      for  carcinogenic  effects     (32)
           v   'eq.    BW   BW   BW
                                    12-13

-------
As before, Eq. (31) or (32) can be solved  for  the  permissible  concentrations  in
soil, Cs.  From Eqs. (8)  and (29), one can solve for  the  factor  F  for  use  in
Eq. (8).  The use of LT = 25550 days,  and  the  assumption  that  soil  ingestion
occurs during ages 1 through 5 (t = 365 to 1825 days), yield F = 0.323.  The
factor F does not depend  on the soil  ingestion rate.   Eqs.  (8) and  (29)  use
Eqs. (28) and (27), respectively, for  BW.
12.4  PCB Advisory Evaluations
     Under normal conditions, significant  soil ingestion  is limited to
children (Lepow, 1975).  Although very limited information  is  available  on
the ranges of age subject to soil ingestion, one investigation presented a
case study of an adult with a history  of habitual  eating  of garden  soil,
which may have been associated with a  pica illness (Wedeen  et  al.,  1978).
The fraction of soil contaminant absorbed  by humans is dependent upon  the
type of compound and its  soil contaminant  adsorption  characteristics,  and
is generally smaller than that which can be expected  when contaminants are
present in food or drinking watder.
     PCBs can be removed  from surface  water, leaving  it suitable for drinking.
Well water that comes from ground water could  be a direct source of drinking
water.  The location of the drinking water exposure does  not necessarily have
to be at the site of the  contamination.  It is assumed that the  population
which may be subject to PCB contamination  in drinking water is remote  from the
RGB-contaminated sites, and the allowable  water concentration  is separately
calculated on the basis of not eating  contaminated soil and not  inhaling con-
taminated air in the immediate vicinity of the site.   The water  concentra-
tion for a single-route exposure can be calculated as
                                    12-14

-------
                                  AI
                            w " 2 L/day
where Cw = concentration of PCBs in water in mg/L,  and  AI  =  the  acceptable
intake for prevention of acute and carcinogenic  adverse health effects,  in
mg/day.  If fish caught in PCB-contaminated surface water  are eaten,  and if
the same water is the source of drinking water,  the allowable concentration
of PCBs (Cw mg/day) should be determined as
                                     AI
                         C^ =
                          W   2 L/day + F •  BCF
where F is the daily fish consumption,  BCF is the bioconcentration  factor
(31,200 L/kg) (U.S. EPA, 1980b; U.S.  EPA,  undated).   The national average of
fish consumption is 0.0065 kg/day (U.S. EPA,  1984b).   However,  it is  more
appropriate to use regional values where such data are available.
     The variabilities of input values  needed in Eq.  (24) (appropriate for PCB
exposure pathways) are wide-ranging for some  values,  and narrow for others.
The inhalation rate of air used for calculation  is 20 m3/day  for both adults
and children (U.S. EPA, 1985d).  Soil  ingestion  rates used for  evaluating
short-term exposures are 3 and 0.6 g/day,  representing conditions with and
without pica, respectively (further explained in Section 15).   One  lifetime
exposure evaluation is based on an average daily rate of 0.6  g/day  multiplied
by factors to correct for the changing  weight of the  body as  a  person grows
from a child to an adult.  This exposure is assumed to occur  from age 1 to 5
years.  However, the soil ingestion rate of 3 g/day is also used in long-term
exposure evaluation.  The absorption  factors  for PCBs through  the gastrointes-
tinal tract for ingested soil, through  the pulmonary  organs for inhaled air,
                                    12-15

-------
and through the skin for contacted soil  are assumed  to be 0.3,  0.5,  and  0.05,
respectively (U.S. EPA, 1984a;  U.S.  EPA,  1985e).   The  off-site  factor  is
assumed to be 1 for longer-term (10-day)  exposure evaluations,  and 0.5 for
lifetime exposure evaluations,  using the  carcinogenic  potency factor.  A
similar approach can be used for short-term (1-day)  and lifetime  noncancer
exposure evaluations.  However, these evaluations are  not performed  because  of
a lack of data regarding health effects.
     If all intake routes, including drinking water, soil  ingestion, air  inha-
lation, dermal contact, and intake of PCBs by means  of fish  or  other food are
of relevant importance, the allowable concentration  levels can  also  be com-
bined in similar fashion.  Since the scope of the present study pertains  to
site cleanup, the applicable formulas for combining  concentrations are not
presented, but they should be considered  as the situation "warrants.
                                    12-16

-------
                          13.   WATER QUALITY  LIMITS



     The concentration levels  of PCBs in drinking water are  based  on  single-

route exposures that are estimated to result  in  acute  and  chronic  toxic  effects.

This does not imply that bioaccumulation in aquatic  organisms  does not occur.

The assumption pertains to absence of fish  contaminated with PCBs  in  the diet.

If other routes are of concern, the allowable concentrations in  water should  be

redefined.  The following levels of PCBs in drinking water,  corresponding to

10-day AIs, can be calculated  for children  and adults:



    • 10-day health advisory:
      Safe        1 mg/kg-daylO kg
  concentration =	100-1 L/day—  = 0>1 m9/L (= 10°  PPb)  (child)
     Safe         1 mg/kg'day70 kg
  concentration =    100-2 L/day    = 0-35 mg/L (=  350 ppb)  (adult)
Similarly, the concentration levels at the various  upper-bound  cancer risks

assumed are calculated, and the results can be tabulated  as  follows:

                                              Advisory
                    Upper-bound                 level
                    cancer risk                (ng/L)

                      ID'4                      875
                      ID'5                      87.5
                      10-6                      8.75
                      10-7                      0.9
Example chronic toxicity advisory level  (at 10'^ maximum risk)

                    IP"6 risk 70 kg	 = 8.75xlo-6 mg/L (=8>75 pg/L)
                    4(mg/kg-day)~1-2 L/day

                                       13-1

-------
     As indicated previously, an Aroclor constitutes  a mixture of many con-



geners.  Since each congener compound exhibits different solubility charac-



teristics, the applicability of these limits  to individual  congeners is ion



question.  In the absence of short-term data  for non-carcinogenic effects,  the



10-day health advisory may be used as the 1-day health advisory for commercial



Aroclors.
                                       13-2

-------
                  14.  LEACHATE CONTAMINATION OF  GROUNDWATER

     Contaminated leachate will impact groundwater quality.   To date,  ground-
water monitoring data showing major contamination of  groundwater by  PCBs has
been rarely reported.  If the contaminated site  is located  above an  unsatur-
ated zone, soil  through which leachate has to migrate to  reach  groundwater
will adsorb PCBs and will greatly retard PCB  migration, as  evidenced by the
high soil-water partition coefficients.  Experimental  work  (U.S. EPA,  1980a)
has shown that the adsorption characteristics vary depending  upon the type of
soil used.  The experimental  values are comparable to the partition  coeffici-
ents estimated from the values of Kow (water-octanol  partition  coefficient)
given in Table 4.  PCBs entering groundwater  at  hazardous waste sites  could
also be retarded as They'are carried along the flow lines.
     The area-source groundwater model (Hwang, 1985)  has  been used to evalu-
ate the extent of retardation and dilution of contaminants  in groundwater. A
typical precipitation rate has been used to estimate  a leachate generation
rate which is a source term in the groundwater rate and transport model.  Two
different values of the retardation factor covering the extreme variations of
the soil-water partition coefficients were considered: Rj =  127, corresponding
             •3                                               O
to Kj = 22 cnr/g; and R^ = 5715, corresponding to K^  = 1000 cnr/g, where Kd
represents the soil-water partition coefficient,  and  Rj is  the  retardation
factor (Rd = 1 + ^b K^, Pjj = bulk density, e  = porosity).  The  results of
                 e
modeling show that when the concentration of  PCBs in  leachate is maintained
at 0.12 mg/L, the vertically averaged PCB concentration in  groundwater at
1000 cm away from the center of a disposal site after two years of release is
0.5 x 10'4 mg/L and 1.9 x 10"? mg/L for the low and high  values of the retar-
dation factor, respectively.  Other parameter values  used in  this simulation
                                      14-1

-------
were: leachate flow rate = 23.4 cm-Vs,  groundwater seepage velocity  = 5 x
10'4 cm/s, porosity of groundwater medium = 0.35,  depth of the aquifer = 300
cm, size of disposal site = 0.5 acre,  and the bulk density of the medium =
2 g/cm^.  The simulation was repeated  for a distance of 1  km away from the
site.  The concentration values at that distance were very small.
     The groundwater transport analysis back-calculated allowable leachate  con-
centrations entering groundwater below a hazardous waste facility, given the
maximum allowable concentrations at a  compliance point.  These calculations do
not account for "facilitated transport" via dissolved organics, cosolvents,
etc.  As indicated previously, the maximum allowable concentrations  were based
on the allowable daily intakes designed to prevent acute and chronic health
effects.  For the purposes of simulation, the maximum allowable drinking water
concentrations at-such distances as 1000 cm and 1  km from the contaminated  site
can be estimated.  For acute toxicity,  the drinking water concentration of  0.1
mg/L is assumed; for chronic carcinogenic toxicity, the concentration of PCB  in
groundwater assumed was 8.7 ng/L, corresponding to a 10"^ lifetime risk.
     The down-gradient groundwater concentration is a complex function of
leachate concentration, dispersion and retardation in groundwater, initial
dilution in groundwater, biodegradation (if any),  and groundwater velocity.
The functional relationship can be found elsewhere (Hwang, 1985), and takes
the form

                               CL = fg CM                                (35)

where CL represents the leachate concentration corresponding to the drinking
water concentration Cw at a point of interest, and fg is a functional relation-
ship which incorporates fate and transport of PCB  in the groundwater medium.
                                     14-2

-------
At the lower end of the  retardation  coefficient  (R
-------
                         15.   SOIL  INGEST ION  PATHWAY

                                                           i
     A literature search shows that there  is  very  limited  information  on  the
rate of likely soil  ingestion  by  children  and adults which  can  be  used in
exposure assessment.   The situation for  which the  information is derived  dif-
fers from study to study.  Lepow  (1975)  studied  the mouthing behavior  of  ten
2- to 6-year-old children in  connection  with  investigations into the principal
cause of the excessive lead accumulation in the  children.   The  total soil
ingestion rate for a 2-year-old child  based on the average  amount  of street
dirt, house dust, and soil  ingested by the child by putting his hands  and
fingers in his mouth, can be  summed as 0.6 g  of  soil per day.
     Wedeen et al. (1978) observed  the lead concentration  in blood of  a black
woman with a 12-year history  of habitual eating  of garden  soil. Using the
levels of blood lead concentration  and the concentrations  of lead  in the  soil
analyzed, they estimated the  amount of lead the  subject had consumed each year
from her garden soil.  From this  estimate, the soil ingestion rate is  estimated
to have been in the range of  between 1.96  and 3.9  g/day, with an average  value
at about 3 g/day.  The lead concentration  in  the dried  garden soil  is  reported
to be between 690 ug/g and 700 ug/g of soil.
     Investigators at the Centers for  Disease Control present the  lifetime
ingestion rate of contaminated soil according to age group  (Kimbrough  et  al.,
1984).  The paper states that  the data presented are "based on  work done  study-
ing lead uptake from contaminated soils."  However, the writers of this report
were unable to locate the original  experimental  work or its source to  cite  in
this evaluation.  The ingestion rate is  assumed  to change  at different ages,
and is given as 0 for the age  group 0  to 9 months, as 1 g/day for  the  age group
9 to 18 months, as 10 g/day for the age  group 1.5  to 3.5 years, as 1 g/day  for
                                     15-1

-------
the age group 3.5 to 5 years,  and as  0.1  g/day  for  a  5-year-old child.

     The second column of Table 8 shows the lifetime  carcinogenic risk  posed

by ingesting soil contaminated with  PCBs  at various concentrations.   This

table is prepared using Eq.  (6) at the  soil ingestion rate of 3 g/day for

children aged 1 through 6 and  an average  weight of  10 kg.   The values for

other parameters used are SM = 0.5,  GI  =  0.3,  and k = 0.   The risk values  in

the second column compare with those  in the third column,  which are  prepared

using the soil  ingestion rate  applicable  to different age  groups, as suggested

by the Centers  for Disease Control.
       TABLE 8.  MAXIMUM LIFETIME RISK FOR INGESTING SOIL CONTAMINATION
                           AT DIFFERENT PCB LEVELS
PCB level in soil (ug/g)
0.1
1
5
10
20
50

(IR = 3)
1.54 x ID"6
1.54 x 10-5
7.7 x lO-5
1.54 x lO'4
3.08 x lO'4
7.7 x ID'4
Lifetime risk
Age-dependent
1.92 x 10~6
1.92 x ID'5
9.6 x ID'5
1.92 x lO'4
3.8 x 10-4
9.6 x ID'4
aTaken from Kimbrough et al., 1984.
                                     15-2

-------
A computer program was  convenient  to  use  in  carrying  out  the  summation  of
daily intakes over a lifetime  period.   The  lifetime  risk  represents  an  upper-
bound estimate of the unit  risk  that  can  occur  as  a  result  of ingesting PCB-
contaminated soil over  a lifetime,  and  indicates the  risk posed  by  a single
exposure pathway; that  is,  soil  ingestion is the sole route for  PCB  intakes,
and other pathways, including  air,  water, fish  are assumed  to be insignificant
sources of human intake of  PCBs.    Since  the population that  will be subject
to soil ingestion resides in the area and must  breathe the  air affected by
PCB emissions, the magnitude of  PCB intakes  by  the ingestion  and inhalation
routes needs to be compared to determine  the significant  pathway.   The  compari-
sion is presented in Section 18.
     Similarly,  in order to determine the daily health advisory  levels  for a
single exposure pathway, the daily  PCB  intakes  equivalent to  ingesting  3 g of
soil in a day at various PCB concentrations  are calculated.  The results are
shown in Table 9.
           TABLE 9.  MAXIMUM DAILY  PCB  INTAKE  BY  INGESTION  OF  SOIL
                        AT VARIOUS  PCB  CONCENTRATIONS
                                      Daily  PCB intake at  30% absorption
        PCB level  in soil  (ug/g)                 (mg/day)
0.1
1
5
10
20
50
0.00009
0.0009
0.0045
0.009
0.018
0.045
                                     15-3

-------
     The exposure pathways  for soil  ingestion,  air  inhalation,  and  other  routes



must be evaluated.  If one  pathway is found to  be dominant  over the other,  the



insignificant pathway based on short-term and long-term intake  rates can  be



ignored.  If they are comparable,  the concentration levels  need to  be adjus-



ted to reflect the combined intake rates  by using Eq.  (24)  or combinations



of Eqs. (8)  through (22).
                                     15-4

-------
                           16.   INHALATION PATHWAY

16.1.  INTAKE BY AIR EXPOSURE ROUTE
     Exposure to PCBs occurs at  or near contaminated  sites  through  inhalation
of ambient air contaminated with PCB  vapors or particulate  matter on  which
PCBs are adsorbed.  The PCB vapors emitted from contaminated  soil will  be
diluted by the action of winds  before a person inhales  the  ambient  air.  When
PCBs are adsorbed on soil, the  vapor  pressure of the  PCBs above the soil sur-
face will be always less than the vapor pressure exerted by PCBs when they  are
present in pure form.  In other  words,  the adsorption phenomena depress  the
vapor pressure that can exist under saturated conditions.   This true  vapor
pressure is referred to as "partial pressure."  When  adsorption reaches  its
saturation capacity on soil, the partial  pressure will  be equal to  the pure PCB
vapor pressure.
     Estimates of PCB concentrations  in the ambient air impacting the popula-
tion at hazardous waste sites are discussed in this section,  as well  as  com-
parison with the intake rates of PCB  through soil ingestion.   In calculating
ambient air PCB concentrations,  the first task was to estimate the  emission
rates of PCBs from the bulk of  soil contaminated at various concentrations  of
PCBs.  The emission rate calculations can be rigorously performed by  the
methods summarized by Hwang (1982) for  steady state conditions, and by methods
presented in the Appendix for transient conditions.
     Based on the inhalation rate of  20 m^/day, and absorption rates  of  50% and
30% for inhaled and ingested PCBs (U.S. EPA, 1984b),  respectively,  the concen-
trations of PCBs in inhaled air  and particulates equivalent to the  dosage
causing acute and chronic toxic  effects can be estimated.   The purpose of this
exercise is to evaluate the concentrations of PCBs in the air, which  are compa-
                                     16-1

-------
rable to the ingestion dosage.   Tables  10 and  11  show this  comparison  for
intake rates corresponding to acute and chronic effects  respectively.
     Table 10, for example, shows  that  a daily intake of 3  g of  soil contain-
ing 5 yg/g of PCBs is equivalent to a daily  inhalation of air containing 0.45
yg/m3.  The concentration of PCBs  on participate  matter  must be  as  high as
7,500 ug/g  for the inhalation  of  particulates at an  assumed concentration  of
60 yg/m3 to be comparable to the ingestion of  soil  and inhalation of air
described above.  This concentration is used because  the concentration should
not exceed the primary ambient  air quality of  75  yg/m3 for  particulate matter.
Since the concentrations of PCBs on soil under consideration are in the range
which is less than this concentration,  it can  be  assumed that the PCB  intake by
inhalation of particulate matter at contaminated  sites is relatively unimpor-
tant.  Similar arguments can be made for the results  shown  in Table 11 for
long-term intakes.  The equivalent air  concentrations Ce shown in Tables 10 and
11 are calculated by the following formula:
                  ,  / 3\  _ daily intake (mg/day)  x 103 yg/mg
                e     m     20 m3/day 0.5 (absorption factor)        (36)
16.2.  EMISSION EVALUATION SCENARIOS
     Emission rates are estimated for four different  scenarios:   Case 1—as
the PCBs volatilize from the initial contaminated soil  column,  they are depleted
from the column of soil by diffusional  transfer of PCBs across  the soil-air
interface, resulting in unsteady-state emission rates,  and the  layer depleted
of PCBs acts as cover material  retarding the volatilization rate; Case 2—the
same scenario as in Case 1 except that the contaminated soil  is  initially
covered with 25 cm of cover material; Case 3—PCBs are  volatilized from the
                                     16-2

-------
     TABLE 10.  COMPARISON OF PCB INTAKES BY INGESTION AND INHALATION ROUTES
                              FOR ACUTE EFFECTS
PCB levels
 in soil
 (ppm)
Daily acute
 intake
 (mg/day)
Equiv. air
 cone, for
 acute ingestion
 (ug/m3)
Cone, of PCBs
 on particulates
     (ug/g)
 0.1

 1

 5

10

20

50
 0.00009

 0.0009

 0.0045

 0.009

 0.0018

 0.045
    0.009

    0.09

    0.45

    0.9

    1.8

    4.5
       150

     1,500

     7,500

    15,000

    30,000

    75,000
                                     16-3

-------
TABLE 11.  COMPARISON OF PCB INTAKES BY INGESTION AND INHALATION ROUTES FOR CARCINOGENIC EFFECTS
PCB levels in soil
(ug/g)
0.1
1
5
10
20
50
Lifetime risk
at IR = 3
1.54 x 10~6
1.54 x 1C'5
7.7 x 10-5
1.54 x 10-4
3.08 x 10-4
7.7 x 10-4
Average daily intake
(mg/day)
3.86 x 10-6
3.86 x ID'5
1.93 x lO-4
3.86 x 10-4
7.72 x 10-4
1.93 x 10-3
Equiv. air cone.
for the risk
(pg/m3)
0.00039
0.0039
0.019
0.039
0.077
0.19
Cone, of PCBs on
participates
(ug/g)
6.4
64.3
322
643.7
1287
3217

-------
surface of contaminated soil,  and the PCB  concentration  at  the surface  is  kept
at a constant value; Case 4--the contaminated soil  is  covered with  25 cm of
clean cover soil  to retard the volatilization rate,  and  the concentration  of
PCBs at the surface is kept at a constant  value.
     As pointed out previously, there exists  a PCB  saturation point above  which
the air-phase PCB concentration in equilibrium with  (or  partitioned with)  the
contaminated soil remains constant, and hence the emission  rate of  PCBs will
also remain essentially steady over time.   Below  this  point, a concentration
profile of PCBs across the contaminated soil  column  starting from the surface
to the depth of contamination  will be created as  volatilization progresses.
This will result  in unsteady-state emission rates which  will vary over  the
period that exposure occurs.  The period considered  includes 10 days for 10-day
advisory, and estimated lifetime (70 years) for long-term advisory.
     The concentration of PCBs in soil  corresponding to  the saturation  point
can be estimated  from the knowledge of  vapor  pressure  and air-soil  partition-
ing.  For example, since the vapor pressure of Aroclor 1254 (7.71 x 10~5 mmHg)
as reported in a  publication,  corresponds  to  the  saturation concentration  of
1,362.7 ug/m3, the PCB concentration in soil  at the  saturation point is
                  C  =
                               1.362.7 ug/m3
                   s   (Xc(9 soil/cm3 air)  x 106 cm3/m3]
                         aS
                     = 1.362.7 (2.44 x 1Q-2)(1.000)  a 4 ug/g
                         (8.37 x lO'3 x 106)
for an assumed value for Kd of 1,000 cm3/g.   The saturation concentration is
dependent upon the value of the air-soil  partition coefficient.   PCB saturation
                                      16-5

-------
 concentrations  in  air,  and  the  corresponding  concentrations  in  soil for the
 Aroclors  considered  as  part  of  this evaluation, are tabulated in Table 12,
 based on  the soil-water partition  coefficient of  1,000 cm3/g for highly ad-
 sorbable  earth  material  (U.S. EPA, 1980a), which  is used in calculating the
 air-soil  partition coefficient.  A similar table  can be prepared at the lower
 end  value of the soil-water  partition coefficient, which is approximately 40
 cm3/g for sandy material  (U.S.  EPA, 1980a).
      Case 3  and perhaps Case 4  may be unrealistic, because as volatilization
 continues, the  PCS concentration in soil decreases and the surface layer,
 depleted  of  PCBs,  should  act as an uncontarninated layer decreasing the emission
 rate.  But this estimate  should provide upper-bound values for  emission rates.
 Cases 3 and  4 would  be  applicable  in real-case situations when  the concentra-
 tion  of PCBs in soil  is  high enough so that the air-phase concentration in
 equilibrium  with soil  remains constant until  the  concentration  of PCBs in soil
 decreases to the saturation  point  (Csm, as defined on p. 12-5).  Below the
 saturation point,  the air-phase concentration will no longer remain constant,
 but  will  decrease  in  direct  proportion to the soil-phase concentration.  In the
 emission  rate calculation,  the  partial pressure of PCBs as a result of parti-
tioning between the  air  and  soil phases is used.  Since the vapor pressures and
 Henry's Law  constants are different for most PCBs, some typical PCBs are used
 for  the purpose of illustrative calculations.  Table 13 summarizes the results
 of calculations for  emission rates for soil containing 1 vg of  Aroclor-1254
 and  Aroclor-1242 per g  of soil.  The values shown for Cases 1 and 2 are the
 averages  for one day  emitted after the initial contamination at the concentra-
 tion.  This  is  for illustration only because Aroclor-1248 and Aroclor-1260 are
 also  used for emission  rate  calcualations.  The models used for the emission
 rate estimation and  necessary calculations are shown in Appendix A.  The emis-

                                      16-6

-------
    TABLE 12.  CONCENTRATION OF PCBS IN SOIL AT SATURATION VAPOR PRESSURE
                                 BASED ON Kd = 1000 cm3/g
                           PCB concen-                   Saturated
                         tration in soil                    vapor
                         with saturated                  concentration
                          vapor (gg/g)                    (ug/m3)
Aroclor-1254
Aroclor-1242
Aroclor-1260
Aroclor-1248
4
250
28.2
55.3
1362.7
5823.3
822.8
7962.8
               TABLE 13.   PCB EMISSION RATES FROM 1 yg/g PCB SOIL
                         AT DIFFERENT CONTROL LEVELS

Scenario
Case 1
Case 2
Case 3
Case 4

Emission rates
Aroclor-1254
1.03
3.67
1.13
1.67
x 10'11
x lO'13
x 10-10
x 10-13a
(g/cm2«s)
Aroclor-1242
2.7 x 10-12
2.8 x 10-14
8.57 x 10-12
1.14 x 10-l4a
aThe models for estimating emissions from landfills underpredict the emission
 rate in comparison to Case 2.   The estimates for Cases 2 and 4 are based on
 the mathematical  model  (described in Appendix A) and the empirical model
 (Farmer et al., 1980),  respectively.
                                     16-7

-------
sion rates at various concentration levels of PCB in soil  are made for evalua-
ting the impact of volatilization on the exposed population at various loca-
tions.  Table 13 shows that the emission rate of Aroclor-1254 is different
from that of Aroclor-1242 at the same soil contamination level.   The table  also
shows that the use of cover material is very effective in  reducing the emission
rate.  The average emission rates over a period of 10 days, or a lifetime,  for
Cases 1 and 2 can be similarly estimated by the rigorous mathematical  formulae
provided in Appendix A.  As PCBs volatilize, the partial pressure of PCBs at
the soil-air interface decreases, and the soil  layer, depleted of PCBs, pro-
vides the barrier for mass transfer, causing the emission  rates  given  for Cases
3 and 4 to approach values comparable to those for Cases 1 and 2, respectively.
The Thibodeaux and Hwang model (1982), originally developed for  land treatment
facilities, provides emission rates similar to those shown for Case 1  and 2 in
Table 13.
     PCBs volatilized into the atmosphere from a contamination site will  im-
pact the population in the surrounding region.   The concentrations of  PCBs  at
the point of impact need to be determined in order to evaluate the signifi-
cance of the air emissions compared with the soil ingestion and  dermal path-
ways.  Acute and chronic impacts are based on the daily concentrations and
the concentrations averaged over an annual period.  Emission rates correspond-
ing to all four cases of maintenance levels are used to estimate the concen-
trations of PCBs in the ambient air at the site and at distances of 0.1 km
and 1 km.
16.3.  AIR DISPERSION MODELING
     Dispersion modeling is used to estimate the ambient air concentrations
which may be possible for daily and annual exposures.  Dispersion modeling for
estimating the annually averaged concentrations makes use  of six stability
                                     16-8

-------
classes, six wind speed classes,  and 16 sectors,  assigning the receptor point
to one of the 16 sectors.   The wind rose data consisting of 6 x 6 x 16 = 576
elements are compiled by the National  Climatic Center in Asheville, North
Carolina.  The dispersion model  for the annual concentration sums the con-
centration contributions over the entire range of stability classes and wind
speeds for each exposure location downwind of the contamination site, and can
take the following form (Bruce,  1969):
                                   6
              C(X,k) = 2.03 x 106Q I         1      6>    fijk
                                            x      i     u             (37)
where C(X,k), the annual  concentration located in  a sector k  at  a distance
of X downwind of the site (pg/m3);(a)i,X = standard deviation of the plume
in the z-direction (vertical  direction)  at distance X for stability class i,
Q = emission rate, g/s,  Uj  =  mean wind speed  for wind speed class j, m/s  and
f^k = frequency of wind  in stability  class i, wind speed class  j, and direc-
tion in sector k.  Both  X and the standard deviation have the the units of
meters.
     The values for the  standard deviation can be  found in an air pollution
textbook (Wark and Warner,  1981), or can be determined by a curve-fitting
equation of the form

                             (az)i>x = aXb +  d                        (38)

where a, b, and d are empirical  constants varying  according to stability  i
and distance X (Wark and  Warner, 1981;  Martin, 1976).  The values for these
constants are given in Table  14  (Martin, 1976).
                                    16-9

-------
       TABLE 14.  VALUES OF CONSTANTS FOR STANDARD DEVIATION EXPRESSION
          AS A FUNCTION OF DOWNWIND DISTANCE AND STABILITY CONDITION
Stability
A
B
C
D
E
F

a
440.8
106.6
61.0
33.2
22.8
14.35
X < 1 km
b
1.941
1.149
0.911
0.725
0.678
0.740

d
9.27
3.3
0
-1.7
-1.3
-0.35

a
459.7
108.2
61.0
44.5
55.4
62.6
X > 1 km
b
2.094
1.098
0.911
0.516
0.305
0.180

d
-9.6
2.0
0
-13.0
-34.0
-48.6
     The estimation of the on-site ambient air concentration does not require
the use of air dispersion modeling presented above.   The ambient air concen-
tration is controlled by the extent of dilution before dispersion occurs down-
wind of the source.  The dilution can be estimated from the knowledge on the
rate of PCB emissions and volumetric rate of wind being mixed with PCB vapors.
The method for estimating the on-site ambient air concentrations is described
in detail in Appendix A.
16.4.  AIR EXPOSURE EVALUATION
     Table 15 summarizes the results of dilution estimation and dispersion
modeling to obtain the concentration levels of PCBs  in ambient air at various
locations considered for emissions of PCB-1254.  This table is a summary of one
set of calculations for the PCB concentration of 1 ug/g in soil for each
scenario.  The wind speed of 10 mph is used for both one-day and annual  concen-
tration averages.  The Climatic Atlas of the United  States provides information
                                    16-10

-------
on annual average wind speed.   A default  value of  10 mph  represents a typical
annual wind speed in the United States.   A site-specific  evaluation will  require
detailed wind rose information based  on  local  measurements..
        TABLE 15.  AMBIENT PCB CONCENTRATIONS AT DIFFERENT LOCATIONS
                        (PCB IN SOIL  =  1  wg/g, PCB-1254)
Concentrations (wg/ra3)
0.1 km from site

Case 1
Case 2
Case 3
Case 4
On-site
0.61
1.4xlO-3
11
0.017
Daily
0.026
5.9xlO-5
0.48
7.1xlO-4
Annual
0.0065
1. 47xlO-5
0.12
l.SxlO'4
1 km from site
Daily
0.0016
3.5xlO-6.
0.03
4.3x10-5
Annual
0.0004
8.9xlO-7
7.3xlO-3
1.1x10-5
     The standard deviation curve for  D stability  is  employed in  estimating the
ambient air concentrations at the distances  of  0.1 and 1 km from  the site as
shown in Table 15, since D stability is by far  the most frequently occurring
stability class.  Although D is the most common stability,  an exposure-weighted
average stability should be used for site-specific evaluations.   The frequency
with which winds blow toward a sector  of interest  is  assumed to be 1 for evalu-
ating the worst-case daily concentrations, while it should  be based on the most
common of the standard 16 wind directions for estimation of the average annual
concentration levels.  The concentrations in ambient  air on-site  and at dis-
tances of 0.1 km and 1 km from the site are  given  in  the table.   Calculations
are performed for the ambient air concentrations of PCB-1242, PCB-1248, and
                                    16-11

-------
PCB-1260, but these are not tabulated here.   The values in the table should  not
be construed as representing fixed ambient air concentrations  at  the location
and under the mode of exposure.   The values  are presented by way  of  illustra-
tion to compare the contributions to ambient air occurring given  the different
assumed conditions.
     The concentration of PCBs  in equilibrium with soil containing 1 ug/g
PCB-1254, which corresponds to  the partial pressure of PCBs partitioned  above
the soil, is 340 yg/m3.  This represents  the maximum vapor concentration when
PCB is emitted from the soil  surface.  Based on this concentration and the
estimated ambient air concentrations given in Table 15, one can calculate the
dilution factors of the emissions for use in Eq. (9) or (24).   For example,
the air dilution factor for the  on-site exposure corresponding to the Case  1
emission
rate would be

                            0 =  0.611/340 =  0.0018

     The values shown in Table  15 can also be used to determine the  daily in-
takes and lifetime risk levels  corresponding to breathing each ambient air  le-
vel.  For example, the daily intake from  the exposure to the ambient air con-
centrations of PCB-1254 at 0.1  km from the site for the Case 1 emission  rate
can be based on a daily inhalation rate of 20 m3/day of air and 50%  absorption
factor for inhalation.

          Daily intake (mg/day)  = C(ug/m3) • 20 m3/day • (1/1000  mg/yg)

                                = 0.026 (20)(1/1000)(0.5) = 0.00026  mg/day
                                    16-12

-------
     The same daily intake can be obtained by using Eq.  (9):

          Daily intake = Kas'Cs-D'IH'10+3(ABA)(SM)
          = (8.37xlO-3/1000)(l/2.44xlO-2)(l)(0.026/340)(20)(103)(0.5)(l)
          = 0.00026 mg/day

     Similarly, the lifetime risk associated with breathing the ambient  air can
be calculated as follows:

Risk = C(wg/m3) • 20 m3/day • 1/70 kg • 1/1000 mg/yg (4)  (mg/kg/day)-l  (0.5)(0.5)

where C is the ambient air concentrations shown in  Table  15 and the value 4
(mg/kg'day)'1 represents the potency factor for PCBs,  and an additional  factor
of 0.5 is the off-site factor under the assumption  that a resident  stays  in the
area 50% of the time.
     A series of calculations can be performed as shown above,  or the procedure
shown above can be reversed to back-calculate the PCB  contaminations in  soil
which will provide the allowable ambient air concentrations at  the  locations
considered and at the acute and chronic effect levels.
                                    16-13

-------
                         17.   DERMAL CONTACT PATHWAY

     Deposition of contaminated soil, dirt, or dust on human skin can provide
another pathway for human intake of PCBs.   PCBs can be absorbed through skin
when PCB-contaminated particulates  come into contact with skin.  Exposure
evaluation requires an estimation of the amount of the particulates on skin,
and the extent or rate of absorption.  The absorption rate is dependent upon
the type of chemicals.  Some  chemicals are readily absorbed, while others are
not.
     There are many factors affecting the amount of soil  which can be accumula-
ted on skin.  Factors include the exposed human skin area, contact time, type
of soil, soil  conditions, and type  of activities.  For example, the amount
deposited on children playing in a  contaminated area may  be different from that
on adults working in a garden.
     OHEA (U.S. EPA, 1984b) has made an estimation of the amount of soil depo-
sition on skin based on the studies by Lepow (1975) and Roels et al. (1980).
Both investigators, using adhesive  tape, measured the amount of soil and dirt
accumulated by children on exposed  areas such as hands, palm, and fingers. The
measured amount of soil ranges from 0.5 to 1.5 mg/cm2, with an average value  of
1 mg/cm2.  It  should be noted that  this is an average value over the surface  of
the exposure area, and that some parts of the body may have more accumulation
of soil than others.
     The area  of human skin that will come in contact with soil or dirt
depends upon the protective measures used during the time that such contact
occurs, as well as the age group involved.  The exposed surface area of an
adult is estimated to range from about 900 to 2,900 cm^.   The exposed surface
area of a child may be smaller in proportion to their total surface area.
                                     17-1

-------
The ranges of values associated with  the factors mentioned above makes it dif-
ficult to arrive at an average value  for the amount of soil  and dirt accumula-
ted on soil.  An assumption of an average soil  deposition at 1 mg/cm^, and
exposed surface area of about 1,000 cm^ on a daily basis, provides an average
daily deposition rate of 1 g per day.   The variability is such that this value
may be different by a factor of as much as two.
     Investigators at the Centers for Disease Control  present an estimated
daily deposition of soil on skin according to age group (Kimbrough et al.,
1984).  Their tabular presentation shows that the daily amount of soil de-
posited on skin is 0 for the age group 0 to 9 months;  1 g for the age group
9 to 18 months; 10 g for the age group 1.5 to 3.5 years; 1 g for the age group
3.5 to 15 years; and 100 mg at age 15 years.
                                     17-2

-------
18.  COMPARISON OF  EXPOSURES  BY  SOIL  INGESTION,  INHALATION,  AND  DERMAL  CONTACT

     Table 16 shows comparisons  of  PCB  intake by various  exposure  routes.
Calculations apply  for  PCB-1254  because the  emission  rate is dependent  upon
the soil-air partition  coefficient, which  is different  for each  PCB.  The
ambient air concentration  is  the on-site value based  on the  emission  rate
averaged over 1 day after  soil  is contaminated up to  the  surface without
cover.
         TABLE 16.   COMPARISON  OF  INTAKES  BY  VARIOUS EXPOSURE  ROUTES3
Exposure route
Soil ingestion

Inhalation
Dermal absorption
Dust inhalation
Contact Absorption
rate fraction
3 g/day (with pica)
0.6 g/day
20 m3/dayb
1 g/day
20 m3/dayc
0.3
0.3
0.5
0.05
0.5
Daily intake
(mg/day)
9 x 10-4
1.8 x 10-4
6.1 x lO-3
5 x 10-5
6 x 10-7
Lifetime
intake
(mg/70 yrs)
Id
0.2
786
0.64f
7.7 x 10-3e
aUsed for illustration:  PCB-1254 at  concentration of 1 ug/g in soil.
bOn-site ambient air concentration  based on 1-day average emission rate
 after surface contamination (no cover).
Concentration of suspended particular matter:   60 pg/m3.
d!83 days/year for 6 years.
eOff-site factor = 0.5.
fy\n average exposure of  1.5 mg/cm2  and the exposure surface area of 1000
 assumed; off-site factor = 0.5.
                                     18-1

-------
     Evaluation under the scenario  of  the  use  of  cover,  or  longer-term  emission
averages, may change the daily  intake  by inhalation  considerably.   For  example,
a calculation shows that the  use  of 25-cm  clean soil  cover  will  reduce  the
daily intake by inhalation to 1.4 x 10'^ mg/day instead  of  6.1  x 10~3 mg/day.
On the other hand, exposures  by soil ingestion, dermal absorption,  and  dust  in-
halation will be likely to decrease because  clean soil is used.   The ingested
soil or the soil on the surface of  the cover that may be accumulated on skin is
initially clean.  Hence, the  daily  intakes by  pathways other  than  inhalation
also become small.  However,  the  concentration of PCBs in the initially clean
cover material  could increase as  the PCBs  in the  air phase  being emitted are
adsorbed on the cover material  as the  liquid PCBs rise toward the  surface due
to capillary potential.  The  table  suggests  that  soil  ingestion  and inhalation
are two competing exposure routes for  PCB  intake. The dermal contact can also
become a contributing route for some conditions of exposure duration.   When  a
high concentration of particulate matter in  the ambient  air is  prevailing, the
comparison shown in Table 16  can  no longer apply. Consequently, the contribu-
tion to PCB intakes by inhalation of particulate  matter  will  increase.   There
are a range of other possibilities  which may result  in a comparison different
from that shown in Table 16.
                                     18-2

-------
                                 19.   RESULTS

19.1.  DERIVATION OF PERMISSIBLE  SOIL  CONTAMINATION
     Determination of the permissible  PCB  levels  in  soil  for  intake through
the combined exposure routes makes  use of  1) Eq.  (24) when the  soil concen-
tration is below the saturation point; and 2)  the summation of  Eqs. (8),  (9),
and (10), equated to the acceptable intake otherwise.   For each Aroclor under
consideration, a separate exposure  evaluation  can be made for the  following
classes of exposure location and  route:  1) Exposure occurs on-site.   This
can be further subdivided into:  (a) sites which  are readily  accessible to
children, and hence for which soil  ingestion is a possibility,  and (b) sites
for which there is no possibility of soil  ingestion, and  hence  exposure is
only through inhalation; 2)  sites which no population is  assumed to enter
within the radius of 0.1 km  from  the site; and 3) sites which no population is
assumed to enter within the  radius  of  1 km from the  site.
     Two classes of soil ingestion  rates are evaluated  when exposure  occurs on-
site (Class 1 above).  For the first class, estimates of  exposure  are calcula-
ted for a person with pica who consumes 3  g per day  between the ages  of 1-5
years.  For the second class, estimates of exposure  are calculated for soil
ingestion of 0.6 g per day between  the ages of 1-5 years. For  both classes,
frequency of exposure is assumed  to be every other day  for lifetime exposures.
For a 10-day exposure, soil  consumption is assumed to occur consecutively for
10 days.  No soil ingestion  is assumed for sites  which  are not  accessible to
population within 0.1 km or  1 km  from  the  contaminated  site.  The  route of
exposure in these cases is by inhalation only.  For  lifetime  inhalation expo-
sure estimates, it is assumed that  the population is exposed  50% of the time,
i.e., 12 hours/day, or 6 months/year.
                                    19-1

-------
     The emission rate of volatilized PCBs  can  be  considerably  reduced  by
covering the contaminated soil  by low-porosity  uncontaminated  soil  or clay
material.  The reduction in the emission  rate will  result  in a  decrease in
ambient air concentrations of PCBs by the action of blowing winds.   When PCB
material is directly exposed to the atmosphere, the PCB  levels  in  soil  required
to maintain the same level of exposure will  be  less than those  expected when
the PCB material  is covered with low-permeability  material  of  appropriate
thickness.  The cover would also serve as a deterrent  to soil  ingestion and
direct dermal  contact.
     The worst-case emissions would occur when  the contaminated soil  is initial-
ly exposed to the atmosphere and the soil is contaminated  up to the conditions
exhibiting saturation vapor pressure.  Models are  used to  estimate emission
rates which can be constant or time-varying depending  upon the  degree of soil
contamination.  The constant emission rate can  be  assumed  if the vapor  phase
concentration maintains its constant value at the  surface  of contamination.
There will be a profile along the layer of soil contamination  for  time-varying
emission rates.  The models for constant  and time-varying  emission rates are
applied with or without cover material.  Calculations  corresponding to  Cases 1,
2, and 3 for exposure possibilities are repeated at an assumed  25-cm (10-inch)
thickness of a soil cover initially free  of PCB contamination.
     The ambient air concentrations given in Table 15, and the  resulting air
dilution factors calculated, are based on an annual average wind speed  of  10
mph.  When wind speed is lower than this, it is possible that  the  daily dilu-
tion factors could be higher than the values in the calculations (less  dilu-
tion).  The combined soil concentration values  based on  Eq. (24) will be lower
when the dilution factor is higher.  More accurate considerations  of meteoro-
logical conditions will require site-specific  evaluation.
                                     19-2

-------
     Among many factors  affecting  the emission rate (including vapor pressure,
soil-air partition coefficient,  and  Henry's law constant),  the variability
associated with the soil-air partition coefficient is more  pronounced than any
other chemical  and physical  properties.   This is caused by  the wide variation
in experimental values for the soil-water partition coefficient reported in
the literature  (U.S. EPA,  1980a),  ranging from 22 to 2,000  cm3 water/g soil.
For clay and sandy materials, the  range includes about 40 to 1,000.  The
values of K
-------
    (5)  Noncancer and cancer effects  are  evaluated  at  the  acceptable  intakes
         corresponding to 10-day  exposure  on  the  first  day  of  cleanup  and
         after 10 days of elapsed time upon cleanup,  and  lifetime  permissible
         exposures.
    (6)  Two extreme values of the soil-air partition coefficients are used
         in the evaluation.  The  high  values  of Kj  (soil-water)  correspond to
         low values of Kas (soil-air).
    (7)  Area contaminated is 45  m x 45 m  or  approximately  0.5 acres.
     The combinations of these evaluation  conditions  are  presented in  tabular
form in Table 17.  The soil ingestion  rates of both  3 and 0.6  g/day are used
in the evaluation pertaining to the longer-term  (=10-day) intakes.  The
soil ingestion rate of 0.6 g/day  between ages 1 and  5 is  used  for  lifetime
exposure evaluation, and this value is averaged with  respect to  soil ingestion
and changing body weight over a lifetime.  For each  Aroclor, there are 120
different situations demanding different permissible levels of PCBs in soil
depending upon the location, route and duration of  exposure, elapsed time
after site cleanup, and the type  of health effects  to be  protected. Table
18 shows the corresponding values for  K
-------
                                         TABLE 17.  EVALUATION CONDITIONS FOR EACH AROCLOR
to
I
tn
Values for partition
Location and route
of exposure
Intake rates
Ten-day intake, Kda
child (100 pg/d) L
Ten-day intake, Kd
adult (700 Mg/d) L
10'7 risk Kd
(0.00175 pg/d) L
10'6 risk Kd
(0.0175 pg/d) L
10'5 risk Kd
(0.175 iig/d) L
10'4 risk Kd
(1.75 pg/d) L


Soil ingestion
(3 g/d)
inhalation
dermal
1,000
Ob
1,000
0
1,000
0
1,000
0
1,000
0
1,000
0
40
25^
40
25
40
25
40
25
40
25
40
25
On-site
3
coefficient (Kd cm /9)

Soil ingestion
(0.6 g/d)
inhalation
dermal
1,000
0
1000
0
1,000
0
1,000
0
1,000
0
1,000
0
40
25
40
25
40
25
40
25
40
25
40
25


Inhalation
only
1,000
0
1,000
0
1,000
0
1,000
0
1,000
0
1,000
0
40
25
40
25
40
25
40
25
40
25
40
25
and cover depth (L cm)
0.1 km
from site
Inhalation
1,000
0
1,000
0
1,000
0
1,000
0
1,000
0
1,000
0
40
25
40
25
40
25
40
25
40
25
40
25
1 km
from site
Inhalation
1,000
0
1,000
0
1,000
0
1,000
0
1,000
0
1,000
0
40
25
40
25
40
25
40
25
40
25
40
25
          aKd  =  soil-water partition coefficient in units of cm3 water/g soil (= cone, in soil/cone, in water), high
           values close to clays, low values close to sand.
          bMeans no  cover designated as "surface contamination."
          cMeans 25  cm (10") clean soil cover applied immediately after remedial action.

-------
       TABLE 18.  LOW AND HIGH VALUES OF AIR-SOIL PARTITION COEFFICIENT
                            USED IN THE EVALUATION
Kd (cm3 water/g soil)3
PCB type
1242
1248
1254
1260
High
1,000
1,000
1,000
1,000
Low
40
40
40
40
"as
Low
2.35 x
1.44 x
3.43 x
2.92 x
(g soil

10-5
10-4
10-4
10-4
/cm3 air)b
High
5.87 x 10-4
3.6 x ID'3
8.58 x ID'3
7.31 x ID'3
aK(j is the soil-water partition coefficient and has the unit of cnr* water/g
 soil  which is equivalent to concentration in soil/concentration in water.
DKflS is the soil-air partition coefficient and has  the unit of g soil/cm3
 air,  which is equivalent to concentration in air/concentration in soil.
 This  is calculated by (H/Kd) (1/2.44 x 10'z).
                                     19-6

-------
                                    TABLE 19.  PERMISSIBLE PCB-1242 SOIL CONTAMINATION LEVELS
                                                (UNCOVERED SURFACE CONTAMINATION)
                                                     Permissible levels (ug/g) corresponding to
                          Noncancer short-term3
                        acceptable Intake MQ/day**     	Cancer risk specific doses tug/day)
Location and
route of human
exposure
100
for child
700
for adult
0.00175
(10-7 risk)
0.0175
(10-6 risk)
0.175
(10-5 risk)
(10l4
risk)
On the contaminated site

 - Soil IngestlonC.       55-60f         510-690            0.008-0.01      0.08-0.1        0.8-1.0     8-13
   inhalation6

 - Soil Ingest ion*1,       92-247         2100-2800          0.03-0.06       0.3-0.6         3.0-6.0     35-61
   inhalation6

 - Inhalation only6       116-vs9        vs                 0.04-0.2        0.4-2.0         4.0-20      110-200


0.1 km from               vs             vs                 4.0-20        110-200            l.lxlO4      vs
 contaminated site
 - Inhalation only6

1 km from                 vs9            vs                 310-430         3.1xl04            vs         vs
 contaminated site
 - Inhalation only6


aShurt-term = 10-daj Intake.
''Based on average weights of 10 and 70 kg for a child and an adult, respectively.
••Children ages 1-5, with pica (consuming 3 g soji/day).
''Children ages 1-5, without pica (consuming 0.6 g soli/day).
elnhalation rates are assumed to be 20 m'/da
                                        '/day for the short-term and longer-term noncancer exposures;
 all  other (more chronic)  exposures assumed to be 10 nr/day as a result of 182 days exposure per year.
'Ranges result In each case because 1) four PCBs (1242, 1248, 1254, 1260) are considered, each with a different
 vapor pressure, and 2) high and low values for soil-air partition coefficient are used In the calculations.
9vs denotes no theoretical upper-bound limit.  Practical reasons require no free-flowing PCB liquids for the limit.

-------
                                                •TABLE 20.   PERMISSIBLE  PCB-1248 SOIL CONTAMINATION LEVELS
                                                             (UNCOVERED  SURFACE CONTAMINATION)
i
oo
Permissible levels (pg/g) corresponding to
Noncancer short-term6
acceptable Intake ug/dayb
Location and
route of human 100 700
exposure for child for adult
On the contaminated site
- Soil IngestionC, 32-80f 612-710
Inhalation6
- Soil tngest1ond. 42-330 2500-2900
Inhalation6
- Inhalation only6 47-vs9 vs
0.1 km from vs vs
Cancer risk specific doses (pg/day)
0.00175 0.0175 0.175 1.75
(10-7 risk) (10-6 risk) (10'5 risk) (10-1 risk)

0.01 0.1 1.0 8-10
0.02-0.04 0.2-0.5 2.0-5.0 37-49
0.02-0.08 0.2-0.8 2.0-8.0 87-110

 contaminated site
 - Inhalation only6

1 km from
 contaminated site
 - Inhalation only6
                                         vs
                                                        vs
2.0-8.0           90-110         8.7x103       8.7xlo5-vs


250-270       2.4x104-2.5x10*       vs            vs
               aShort-term a 10-day Intake.
               bBased on average weights of 10 and 70 kg for a  child  and  an adult,  respectively.
               cCMldren ages 1-5, with pica (consuming 3 g soil/day).
               dCMldren ages 1-5, without pica (consuming 0.6 g  soli/day).
               elnhalat1on rates are assumed to be 20 mVda
                                        3/day  for the short-term and longer-term noncancer exposures;
 all  other (more chronic) exposures assumed to be 10 m3/day as a result of 182 days exposure per year.
'Ranges result In each case because 1) four PCBs  (1242,  1248. 1254, 1260) are considered, each with a different
 vapor pressure, and 2) high and low values for soil-air partition coefficient are used in the calculations.
9vs denotes no theoretical upper-bound limit.   Practical reasons require no free-flowing PCB liquids for the limit.

-------
                                  TABLE  21.'  PERMISSIBLE  PCB-1254  SOIL  CONTAMINATION LEVELS
                                              (UNCOVERED  SURFACE CONTAMINATION)
Location and
route of human
exposure

Noncancer short-term3
acceptable Intake vg/day"
100 700
for child for adult
Permissible levels (wg/g) corresponding to
Cancer risk specific doses (pq/day)
0.00175 0.0175 0.175
(10-' risk) (10-6 risk) (10-5 rtsk)


1.75
(ID*4 risk)
On the contaminated site

 - Soil IngestlonC,       90-100f         720-730             0.009-0.01       0.09-0.1       1.0-2.0       12
   Inhalation6

 - Soil 1ngest1ond,       370-420         2980-3000           0.01-0.04        0.1-0.4        3.0-4.0       36-59
   Inhalation6

 - Inhalation only*       vsg            vs                  0.01-0.05        0.1-0.5        5.0-7.0       460-470


0.1 km from               vs             vs                  5.0-7.0         460-470        4.7xl04       vs
 contaminated site
 - Inhalation only6

1 km from                 vs             vs                  1.3xl03         1.3xl05        vs            vs
 contaminated site
 - Inhalation only6


aShort-term » 10-day Intake.
DBased on average weights of 10 and 70 kg for  a child  and  an adult, respectively.
Children ages 1-5. with pica (consuming 3 g soil/day).
Children ages 1-5, without pica (consuming  0.6 g soil/day).
elnhalat1on rates are assumed to be 20 m3/day  for the  short-term and longer-term noncancer exposures;
 all  other (more chronic) exposures assumed  to be 10 m3/day  as  a result  of 182 days exposure  per year.
'Ranges result In each case because 1) four  PCBs  (1242,  1248,  1254, 1260) are considered,  each with a different
 vapor pressure, and 2) high and low values  for sod-air partition  coefficient are used  In the calculations.
9vs denotes no theoretical  upper-bound limit.  Practical reasons require no free-flowing PCB  liquids for the limit.

-------
                                              TABLE  22.   PERMISSIBLE PCB-1260 SOIL CONTAMINATION LEVELS
                                                          (UNCOVERED SURFACE CONTAMINATION)
Location and
route of human
exposure

Noncancer short-term3
acceptable Intake ug/dayD
100 700 ,
for child for adult
Permissible levels (pQ/g) corresponding to
Cancer risk specific doses (tig/day)
0.00175 0.0175 0.175
(10-' risk) (10-6 risk) (10-5 Ms|()


1.75
(10-4 risk)
            On  the  contaminated  site

             -  Soil  Ingestlont,        25-87*         640-710            0.01            0.1     .      1.0            12-17
               Inhalation6

             -  Soil  1ngestiond,        61-360         2670-2900          0.01-0.04       0.1-0.4       1.0-4.0        40-48
£>             Inhalation6
i
o           -  Inhalation  only6        vs9            vs                 0.01-0.06       0.1-0.6       1.0-6.0        77-91


            0.1 km  from               vs              vs                 6-220           90-2.2xl04    7.7xl03-vs     vs
             contaminated  site
             -  Inhalation  only6

            1 km from                 vs              vs                 220-240         2.2x10*       vs             vs
             contaminated  site
             -  Inhalation  only6


            'Short-term a  10-day Intake.
            "Based  on average weights  of  10 and 70 kg for  a child and sn adult, respectively.
            cCh11dren ages 1-5,  with pica (consuming  3 g soil/day).
            Children ages 1-5,  without pica (consuming 0.6 g soil/day).
            elnhalat
-------
                                  TABLE 23.  PERMISSIBLE PCB-1242 SOIL CONTAMINATION LEVELS
                                               (25-cm-THICK CLEAN SOIL COVER)
Location and
route of human
exposure

Noncancer short -term9
acceptable Intake uQ/dayb
100 700
for child for adult
Permissible levels (gg/g) corresponding to
Cancer risk specific doses (gg/day)
0.00175 0.0175 0.175
(10-7 risk) (10-6 risk) (10"5 risk)


1.75
(10-4 risk)
On the contaminated site

 - Soil Ingestlon'.       170-200f       1200-1400          0.03-0.2        0.3-2.0       3-17            170-vs
   Inhalation6

 - Soil 1ngest1ond.       450-820        3100-5700          0.1-0.6         1.0-6.0       12-48          260-vs
   Inhalation6

 - Inhalation only6       vs9            vs                 0.9-1.0         9-vs          86-vs          vs


0.1 km from               vs             vs                 85-vs           vs            vs             vs
 contaminated site
 - Inhalation only6

1 km from                 vs             vs                 vs              vs            vs             vs
 contaminated site
 - Inhalation only6


aShort-term = 10-day Intake.
"Based on average weights of 10 and 70 kg for a child and an adult, respectively.
Children ages 1-5, with pica (consuming 3 g soil/day).
('Children ages 1-5, without pica (consuming 0.6 g soil/day).
elnhalat
-------
                                            TABLE 24.  PERMISSIBLE PCB-1248 SOIL CONTAMINATION LEVELS
                                                         (25-cm-THICK CLEAN SOIL COVER)
10
i
                                                               Permissible levels (wg/g) corresponding to
                                    Noncancer short -term3
                                  acceptable Intake pg/dayb
                                                                  Cancer risk  specific do&es  (u9/day)
Location and
route of human
exposure
100
for child
700
for adult
0.00175
(10-' risk)
0.0175
(10-6 risk)
0.175
(10-5 risk)
1.75
(10-* risk)
On the contaminated site

 - Soil 1ngest1onc,       160-190f       1100-1300
   inhalation6

 - Soil 1ngest1ond.       650-vs9        4500-vs
   Inhalation6

 - Inhalation only6       vs             vs
          0.1 km from               vs             vs
           contaminated site
           - Inhalation only6

          1 km from                 vs             vs
           contaminated site
           - Inhalation only6
0.01-.09


0.02-0.1


0.02-0.1


2-14



vs
                                                                                       0.1-1.0
                                                                             vs
              1-10
26-460
0.2-1         2.0-10      93-2,500


0.2-1         2.0-14      1.9x10*


1.9xl04       vs          vs
                                                                                           vs
                                                                                                       vs
          aShort-term *  10-day Intake.
          bBased on average weights of 10 and 70 kg for a child and an adult, respectively.
          GCh11dren ages 1-5, with pica  (consuming 3 g soil/day).
          ^Children ages 1-5, without pica  (consuming 0.6 g soil/day).
          elnhalat1on rates are assumed  to  be 20 tn3/day for the short-term and longer-term noncancer exposures;
           all other (more chronic) exposures assumed to be 10 m3/day as a result of 182 days exposure per year.
          fRanges result In each case because 1) four PCBs (1242, 1248, 1254, 1260) are considered, each with a different
           vapor pressure, and 2) high and  low values for soil-air partition coefficient are used In the calculations.
          9vs denotes no theoretical upper-bound limit.  Practical reasons require no free-flowing PCB liquids for the limit.

-------
                                  TABLE 25.   PERMISSIBLE PCB-1254 SOIL CONTAMINATION LEVELS
                                               (25-cm-THlCK CLEAN SOIL COVER)
I
I—•
U)
Location and
route of human
exposure

Noncancer short-term3
acceptable Intake pg/day°
100 700
for child for adult
Permissible levels (wg/g) corresponding to
Cancer risk specific doses (gg/day)
0.00175 0.0175 0. 175 1.7
(10-7 risk) (ID"6 risk) (10'5 risk) (10'4


risk)
On the contaminated site

 - Soil 
-------
                                           TABLE  26.   PERMISSIBLE PCB-1260 SOIL CONTAMINATION LEVELS
                                                        (25-cm-THICK CLEAN SOIL COVER)
vo
Permissible levels (ug/g) corresponding to
Noncancer short-term9
acceptable Intake ng/dayb
Location and
route of human 100 700
exposure for child for adult
On the contaminated site
- Soil IngestlonC, 110-184? 800-1300
Inhalation6
- Soil Ingest 1on««, 550-800 4000-5000
Inhalation6
- Inhalation only6 vs9 vs
0.1 km from vs vs
Cancer risk specific doses dig/day )
0.00175 0.0175
(10-' risk) (10-6 risk)

0.01-0.02 0.1-1.0
0.02-0.07 0.2-0.7
0.02-0.08 0.2-0.8
1-8 620-770
0.175 1.75
(10-5 risk) (10-* risk)

1.0-2.0 22-360
1.0-7.0 120-550
1.0-8 620-770
vs vs
          contaminated site
          -  Inhalation only6

         1 km from
          contaminated site
          -  Inhalation only6
vs
               vs
                                   vs
                                                   vs
                                                                vs
                                                                            vs
         aShort-term s  10-day  Intake.
         bBased  on  average  weights  of  10 and 70  kg for  a  child and an adult, respectively.
         cCM1dren  ages 1-5, with pica  (consuming  3 g soil/day).
         ^Children  ages 1-5, without pica (consuming 0.6  g  soil/day).
         elnhalat1on rates  are assumed  to be 20  mVday  for  the short-term and longer-term noncancer exposures;
          all  other (more chronic)  exposures assumed to be  10 m'/day as a result of 182 days exposure per year.
         fRanges result In  each case because 1)  four PCBs (1242,  1248. 1254, 1260) are considered, each with a different
          vapor  pressure, and  2) high  and low values for  soil-air partition coefficient are used In the calculations.
         9vs  denotes no theoretical upper-bound  Hm1t.  Practical reasons require no free-flowing PCB liquids for the limit.

-------
     The symbol  "vs" indicates that  no upper-bound  limit  for  PCB  concentra-
tions in soil can be derived from the exposure  evaluation.  This  results  mainly
for two reasons.  First, the emission rate cannot exceed  the  upper-bound  value
which can be expected when the air phase concentration  of PCBs  at the  contami-
nated soil surface is maintained at  the saturation  point.  The  concentration  at
the saturation point corresponds to  the vapor pressure  concentration.   Second,
when the cover is applied, the emission rate is not only  retarded,  but also the
concentration of PCBs in soil  being  ingested is controlled  by the amount  of
PCBs adsorbed on soil in equilibrium with the air phase being emitted. Hence,
the concentration of PCBs in the initially clean soil material  cannot  exceed  the
saturation point concentration.  The PCB concentrations in  soil corresponding to
vapor saturation concentrations are  250, 55, 4, and 28  vg/g when  K
-------
     Since the ranges shown in these tables are dependent upon the values  of
the soil-air coefficient, the site-specific or contaminant-specific informa-
tion will help find an appropriate level  of PCBs for that particular condition.
This can be done either by using the procedure outlined  in  the main body of the
report, or can be conveniently done by looking up the values listed in the
Appendix for each Aroclor at low and high values of  soil-air partition coef-
ficient.
     The results in Tables 23 through 26  for each Aroclor assume  that the  25-cm
clean cover material is placed on top of  contaminated soil.  In this case, the
intake rate by exposure to soil ingestion is calculated  based on  the estimated
concentration profile existing in the cover material. This profile exists
because of the establishment of the vapor-solid adsorption  equilibrium between
the vapors being emitted and the soil.  The concentration profile, which changes
as a function of time, is estimated by mathematical  models,  the  concentration
used for soil ingestion is the average concentration along  the thickness of the
initially clean cover material.
     If the prevailing contaminants at a  site are PCB-1242, for example, Table
19 can be interpreted as follows:
     (1)  When the site is amenable to access by children with possibilities
of ingesting the contaminated soil  exposed to the atmosphere, the permissible
PCB concentrations levels in soil should  range from  55 to 60 ug/g, and 92  to
247 ug/g for prevention of noncancer effects from 10-day exposures at soil
ingestion rates of 3 g/day and 0.6 g/day, respectively.
     When the site is accessible to children and the population has the poten-
tial of on-site exposures to the contaminated soil and air  over a lifetime, the
permissible PCB levels in soil  should range from U.008 to 0.01, 0.08 to 0.1,
0.8 to 1.0, and 8 to 13 pg/g, corresponding to the best  estimate  of an upper-

                                    19-16

-------
bound oncogenic risk at 10'7, 10'6, 10'5 and 10'4,  respectively.   The specific
level will be dependent upon the likely soil ingestion rate and the extent of
soil-air partitioning.  Because of the PCB concentration profile being esta-
blished in the soil column as volatilization occurs,  the PCB concentration
averaged over the depth will gradually decrease over  time.   Hence, if the popu-
lation is allowed to enter the site at some time after site cleanup, the per-
missible levels for preventing 10-day noncancer health effects  can change.
Again, the specific level  will be dictated by site-specific characteristics
such as the soil-air partition coefficient.
     (2)  If there is no possibility of population  entering the contaminated
site within a radius of 0.1 km from the site, the PCB levels in the soil can
remain at the no theoretical upper-bound limit value  (vs ug/g)  without exceed-
ing the 10-day AI upon inhalation exposure for 10 days; and at  110-200 ug/g
without exceeding the average daily dose corresponding to a 10~6 risk for life-
time exposure.  Similar interpretations can be made for the results applicable
to the carcinogenic risk listed at 10"4, 10~5, and  10~7, and to sites without
affected population up to 1 km from the site.
                                    19-17

-------
                        20.   LIMITATIONS  OF  APPLICATION

     It Is assumed that the  25-cm (10-inch)  clean  cover material used  remains
undisturbed in the process of human  activities  on  the  site.  At times  this
assumption may be found arbitrary, because an opportunity  could exist  that
would expose the contaminated soil surface in contact  with the atmosphere by
inadvertent disturbances of  soil  surfaces, construction activities,  utility
installation, precipitation, or children  playing on the site, to name  a  few.
In this case, additional thickness of  cover  material should be used, or  the
site should be made inaccessible to  children or should be  kept from  any  activ-
ities that would lead to disturbance of the  soil surfaces.  Spills on  top of
the clean cover will  result  in a situation equivalent  to the surface contami-
nation case, requiring a more stringent concentration  limit in soil.   In this
case, the results given for  the 25-cm-thick  clean  cover material do  not  apply.
     The tabulated results are intended to be applicable under certain specific
conditions.  Under conditions similar  to  those  used in preparing the tables,
the values can be used without additional evaluations. A  particular situation
may warrant a site-specific  evaluation which may require the use of  conditions
different from what has been assumed in preparing  the  tables.  If the  analysis
is available to show the specific type of Aroclor  contaminating the  soil, the
individual table should be used.   If the  value  for the soil-air partition coef-
ficient can be better defined, the range  of  the permissible PCB concentration
should be further narrowed.
                                      20-1

-------
                                21.   REFERENCES
Binder, S.  (1985)   Estimating the amount  of  soil  ingested  by  young  children
     through trace  elements.   Report  by  the Centers  for  Disease  Control.

Bruce, D.B.  (1969)  Workbook  of atmospheric  dispersion  estimates, U.S. Depart-
     ment of Health, Education, and Welfare.   Pub. No.:  999-AP-26.

Burkhard, L.P.; Armstrong,  D.E.; Andren, A.M.   (1985)  Henry's Law constants
     for the polychlorinated biphenyls.  Environ.  Sci. Technol.  7:590-596.

Buser, H.R.; Rappe, C.   (1979)  Formation  of  polychlorinated dibenzofurans  from
     the pyrolysis  of individual PCB  isomers.   Chemosphere  8:157.

Carey, A.  (Undated)  Chemical/element concentrations in surface soils of
     selected U.S.  urban areas.  U.S. Environmental  Protection Agency, pri-
     vate communication.

Carter, J.W.  (1983)  Onset of hepatomegaly in PCB  (Aroclor 1254)-treated rats.
     Bull. Environ. Contain. Toxicol.  31(2):183-187.

Collins, W.T.; Capen, C.C.   (1980a)  Biliary  excretion of t:hyroxine-I-125
     and fine structural alterations  in  the thyroid  glands  of  gunn-rats fed
     PCBs.  Lab. Invest. 43:158.

Collins, W.T.; Capen, C.C.   (1980b)  Ultrastructural and functional  altera-
     tions of the rat thyroid  gland produced  by polychlorinated  biphenyls
     compared with  iodide excess and  deficiency, and thyrotropin and thyrox-
     ine administration.  Virchows Arch. B. 33(3):213-231.

Collins, W.T.; Capen, C.C.   (1980c)  Fine  structural lesions and hormonal
     alterations in thyroid glands of perinatal  rats exposed i^  utero and fed
     by the milk to polychlorinated biphenyls.  Am.  J. Pathol. 99(1):125-141.

DuPont, R.R.  (1985, Nov.)  Evaluation of  air emission release rate  model pre-
     dictions of hazardous  organics from land treatment  facilities.   Presented
     at American Institute of  Chemical Engineers meeting, Chicago, IL.

Farmer, W.J.; Yang, M.-S.;  Letey, J., Dept. of Soil  and  Environmental  Sciences,
     University of  California-Riverside; Spenser,  W.F.,  Science  and  Education
     Administration, Federal Research, USDA.   (1980)  Land  disposal  of hexa-
     chlorobenzene  wastes:  Controlling vapor  movement in soil.  EPA-600/2-80-
     119.  Prepared for U.S. Environmental  Protection Agency,  Municipal
     Environmental  Research Laboratory,  Cincinnati,  OH.

Grant, D.L.; Phillips,  W.E.J.   (1974) The effect  of age and sex on  the toxicity
     of Aroclor 1254, a polychlorinated  biphenyl,  in the rat.  Bull. Environ.
     Contam. Toxicol. 12:145-152.

Hwang, S.T.  (1982)  Toxic  emissions  from  land disposal  facilities.   Environ.
     Prog. 1:46.


                                     21-1

-------
Hwang, S.T.  (1985, May)   Assessing exposure to  ground  water  contaminants
     migrated from hazardous waste facilities.   Proceedings,  Conference  on
     Management of Unconfined Hazardous  Waste Sites,  Cincinnati,  OH.

Hutzinger, 0.; Safe, S.;  Zitko,  V., eds.  (1974)   The chemistry  of  PCBs.  CRC
     Cleveland, OH: CRC Press.

Jury, W.A.; Spencer, W.F.; Farmer, W.J.   (1983)   Behavior  assessment  model  for
     trace organics in soil.  I.  Model  description.  J. Environ. Qual.  4:558-
     564.

Karickhoff, S.W.  (1979)   Sorption of hydrophonic pollutants  on  natural  sedi-
     ments.  Water Res. 13:241-248.

Kenaga, E.E.; Goring, C.A.I.  (1980)  Relationship between water solubility,
     soil sorption, octanol-water partitioning,  and bi concentration  of  chem-
     icals in biota.  In:  Aquatic Toxicology.   (Eaton, J.C.; Parnish, P.R.;
     Hendricks, A.C.; eds.).  American Society for Testing and Materials, in
     press.  ASTM STP 707.

Kimbrough, R.D.; Falk, H.; Stehr, P.  (1984)  Health  implications of  2,3,7,8-
     tetrachloro-dibenzodioxin (TCDD) contamination of  residual  soil. J.
     Toxicol. Environ. Health 14:47.

Kimbrough, R.D.; Squire,  R.A.; Linden, R.E.; Strandberg, J.D.; Montali,  R.J.;
     Burse, V.W.  (1975)   Induction of liver tumors in  Sherman strain female
     rats by polychlorinated biphenyl Aroclor 1260.  J. Natl. Cancer Inst.
     55:1453-1459.

Lepow, M.L. (1975)  Investigations in sources of lead in the  environment  of
     urban children.  Environ. Res. 10:415.

Lorenz, H.; Neumeier, G.   (1983)  Polychlorinated biphenyls:  profile  of  a group
     of substances.  MMV Medizin Verlag Miinchen.

MacKay, D.; Leinonen, P.L.  (1975)  Rate of  evaporation of low-solubility
     contaminants from water bodies to atmosphere. Env. Sci. Technol.
     9:1178.

Martin, D.O.  (1976)  The change of concentration standard, deviation  with
     istance.  J. Air Poll.  Control Assoc.  26:145.

Monsanto Chemical Company.  (Undated)  The  AROCLORS—physical properties  and
     suggested applications.

MRI.   (1984, December)  Thermal  degradation  products  from  dielectric  fluid.
     Prepared for U.S. Environmental Protection  Agency, Office of Toxic  Sub-
     stances.  EPA-560/5-84-009.

National Institute for Occupational Safety  and Health (NIOSH).  (1977)
     Criteria for a recommended standard: occupational  exposure  to  polychlori-
     nated biphenyls (PCBs).  DHEW (NIOSH)  Pub.  No. 77-225.  U.S. Government
     Printing Office, Washington, D.C.

                                      21-2

-------
New York State-Department of Environmental  Conservation.   (1979)   New York
     State air quality report, continuous  and manual  air  monitoring  system.
     DAR-80-1.  Albany, NY.

New York State Department of Health.  (1981, March  16)  Memorandum from John
     Hawley.

Nisbet, I.C.T.; Sarofim, A.F.  (1972)   Rates and  routes of transport of PCBs  in
     the environment.  Environ. Health Perspect.  1:21.

Roels, H.A.; Buchet, J.-P.; Lauwerys,  R.R.; Bruaux,  P; Cl yaeys-Thoreau, F.;
     Lafontaine, A.; Verduyn, G.  (1980)   Exposure  to lead by  the  oral  and the
     pulmonary routes of children living  in the vicinity  of a  primary lead
     smelter.  Environ. Res. 22:81-94.

SCS Engineers.  (Undated)  W-E-T model hazardous  waste data base.   Final  draft.
     Prepared for U.S. Environmental Protection Agency, Office of  Solid Waste,
     Washington, D.C.  Reston, VA.

Snyder, U.S.  (1975)  Report of the task group on reference manual.   Inter-
     national Commission of Radiological Protection  No. 23.  Pergamon Press,
     New York, NY.'

Thibodeaux, L.J.; Hwang, S.T.  (1982)   Landfarming  of petroleum wastes:
   .  modeling the air emission problem. Environ. Prog. 1:42-46.

University of Wisconsin-Madison, Water Resources  Center and Water  Chemistry
     Program.  (1980, March)  Atmospheric  chemistry  of PCBs  and PAHs.  Prepared
     for U.S. Environmental Protection Agency, Chicago, IL.

U.S. Environmental  Protection Agency.   (Undated)  Guideline for assessing  human
     exposure.  Office of Solid Waste.

U.S. Environmental  Protection Agency.   (1976a, January)   Development of a  study
     plan for definition of PCB usage, wastes, and  potential substitution  in
     the investment casting industry.   NTIS No.:  PB251842 C.I.

U.S. Environmental  Protection Agency,  Office of Toxic Substances.   (1976b,
     February)  PCBs in the United States:  industrial  use and  environmental
     distribution;  Task 1.  Prepared by Versar, Inc.   NTIS No.: PB252402.

U.S. Environmental  Protection Agency.   (1976c, March)  National Conference on
     Polychlorinated Biphenyls.  NTIS  No.:  PB253248.

U.S. Environmental  Protection Agency.   (1978, March)   Microeconomic  impacts of
     the proposed PCB ban regulations. NTIS No.: PB281881 C.I.

U.S. Environmental  Protection Agency.   (1979a)  Water-related  environmental
     fate of 129 priority pollutants.   Vol. 1.  EPA-440/4-79-029a.

U.S. Environmental  Protection Agency.   (1979b, May)   Polychlorinated biphenyls
     1929-1979:  Final report.  Office  of Toxic Substances.  NTIS  No.:  PB256559.
                                      21-3

-------
U.S. Environmental  Protection Agency.  (1980a)   Attenuation of water-soluble
     polychlorinated biphenyls by earth materials.   EPA-600/2-80-027.

U.S. Environmental  Protection Agency.  (1980b)   Ambient water quality criteria
     document for PCBs.  Environmental  Criteria and Assessment Office, Cincin-
     nati, OH.  EPA-440/5-80-068.

U.S. Environmental  Protection Agency.  (1980c,  April)   Determination of Henry's
     Law constants of selected priority pollutants. Municipal Environmental
     Research Laboratory, Cincinnati, OH.   (Authors:  Warner, H.P.;  Cohen,
     J.M.; Ireland, J.C.)

U.S. Environmental  Protection Agency.  (1981a,  April)   Evaluation of PCB
     destruction efficiency in an industrial  boiler.  EPA-600/2-81-055a.

U.S. Environmental  Protection Agency.  (1981b,  May) Environmental  impact
     statement on the Hudson River PCB Reclamation  Demonstration Project.
     Region II. New York, NY.

U.S. Environmental  Protection Agency.  (1983)  Analysis of adipose and blood
     sera samples for individual  PCB isomers.  Draft Final Report on EPA
     Contract 68-01-5915, 11/30/81 - 5/31/83.

U.S. Environmental  Protection Agency.  (1984a,  Nov.)  Risk analysis  of TCDD-
     contaminated soil.  EPA-600/8-84-031.

U.S. Environmental  Protection Agency.  (1984b)   Code of Federal Regulations,
     Title 40, Part 761.

U.S. Environmental  Protection Agency.  (1985a)   Drinking water criteria docu-
     ment for polychlorinated biphenyls (PCBs).  Environmental Criteria and
     Assessment Office, Cincinnati, OH.  Final  Draft.   EPA-600/X-84-198-1.

U.S. Environmental  Protection Agency.  (1985b,  June 12)  Personal communica-
     tion, Don R. Clay, Director, Office of Toxic Substances.

U.S. Environmental  Protection Agency.  (1985c,  Nov. 7)  Personal communication,
     Martin, P. Halper, Director, Exposure Evaluation  Division, Office of Toxic
     Substances.

U.S. Environmental  Protection Agency.  (1985d,  Aug.)  Development of statistical
     distributions or ranges of standard factors used  in exposure assessments.
     Prepared by GCA Corporation for the Office of  Health and Environmental
     Assessment.  NTIS No: PB85-242667.

U.S. Environmental  Protection Agency.  (1985e)   Personal communication with
     Office of Toxic Substances.

U.S. Food and Drug Administration.  (1984)  Code of Federal Regulations, Title
     21, Part 109.30.
                                      21-4

-------
Veith, G.D.; Macek, K.J.; Petrocelli,  S.R.;  Carroll,  J.   (1980)   An  evaluation
     of using partition coefficients  and  water  solubility  to  estimate  biocon-
     centration factors for organic chemicals in  fish.   J.  Fish.  Res.  Board
     Can. (Prepublication copy).

Versar, Inc.  (1976)  Assessment  of wastewater  management,  treatment technology,
     and associated costs for abatement of  PCB  concentrations in  industrial
     effluents.  Prepared for U.S. Environmental  Protection Agency,  Office of
     Toxic Substances.  NTIS No.:  PB251433  C.I.

Versar, Inc.  (1977)  A first-order mass  balance  model for  the sources, distri-
     bution, and fate of PCBs in  the  environment.   Prepared for the  U.S.
     Environmental  Protection Agency,  Office of Toxic Substances, Washington,
     D.C.

Villeneuve, D.C.; Grant, B.L.; Khera,  K.; Clegg,  D.J.; Baer,  H.;  Phillips,
     W.E.J.  (1971)  The fetotoxicity  of  a  polychlorinated.  biphenyl  mixture
     (Aroclor 1254) in the rabbit  and  in  the rat.   Environ. Physiol. 1:67-71.

Wark, K.; Warner, C.F.  (1981) Air pollution:  its  origin  and control.  New
     York: Harper and Row.

Wedeen, R.P.; Mailik, O.K.; Baiuman,  V.;  Bogden,  J.D.  (1978) Geophagic  lead
     nephropathy: case report. Environ.  Res. 17:409.
                                      21-5

-------
               APPENDIX A
MODELS USED IN AIR RELEASE RATE CALCULATIONS

-------
             DERIVATION OF MODELS  FOR  ESTIMATING  VOLATILE  EMISSIONS

           FROM CONTAMINATED SOIL  COLUMNS  UNDER TRANSIENT  CONDITIONS



     Because of the limited aqueous  solubility and  high  soil  affinity  of PCBs,

it has been assumed that these compounds move vertically in  soils, primarily

by diffusion in the vapor phase.   If transport of PCBs  is  by vapor phase dif-

fusion through interstitial spaces between soil particles, a mass balance over

an infinitesimal vertical element  of soil  can be  written as  follows:
AE(-Dei3C)    - AE(-DeijC)       = AAZ3C  + A&zFsaj_                   (A-l)
       3z z.t          3Z z+Az,t      8t         3t
where:
      A = cross-sectional  area of interest,  cm2
      C = concentration of PCBs in the vapor phase  in  soil  pores,  g/cm^
     Cs = concentration of PCBs in soil,  g/g
     D^ = molecular diffusivity, CIT//S
    D! = effective diffusivity, cmz/s (=Di'E1/'J)
     _E = pore porosity
     Ps = bulk density of soil = true density of  soil, Ps,  multiplied by
            -       m3
      t = time, seconds
      z = depth measured from the soil -air interface,  cm


     In Eq. (A-l), the effective diffusivity,  Dgi ,  is  used in place of

to account for the tortuosity effect in porous media.   The use of effective

diffusivity is consistent with the findings which describe emission rates of

volatile chemicals from landfills and soils (Hwang, 1982;  Thibodeaux, 1979;

Farmer et al., 1980).  The effective porosity  for dry  soil is used for sim-

plicity.  The effect of moisture can be incorporated in the porosity term as

shown by Farmer et al . (1980).

     Since changes in soil and vapor phase PCB concentrations occur slowly, it

can be assumed that vapor phase concentrations and  soil concentrations of PCBs


                                      A-l

-------
are in local equilibrium.  If PCB concentrations in soil  and in interstitial

vapors approach equilibrium, they are related by the following equation:
                                  C  = KH •  C
                                  °s   TT                              (A-2)


where

     Kd = soil/water partition coefficient

      H = Henry's constant




     Rearranging Eq. (A-l) and substituting  Eq. (A-2) into the resulting rela-

tionship yields
                              Dei  l!C  .  (1 + IsJ^d) 3C               (A-3)
                                  32-          E'H    3t
or
                         a 32C = JC                                    (A-3)
                                 3t
where
                        a =      Dei'E,                                (A-4)
                            (E+Ps'[l-E]'Kd/H)
     a can also be defined as
                                ei                                     (A-5)
                             1 + K-S
                                      A-2

-------
where
                        K = JSd • D
                            H    ps
     Eq. (A-3) can be solved to estimate PCB soil  concentration, vapor
phase concentration, and emission rate into air above soil  for the various
cases described in this report if initial  and boundary conditions are speci-
fied for each of these cases.
Case 1.  Surface is exposed to the atmosphere.  The boundary and initial
         conditions are

         1.  I.C.           C = (H/Kd)Cso, at t =  0, z > 0
         2.  B.C.           C = (H/Kd)Cso, at z =  », t > 0
         3.  B.C.           C = 0,         at z =  0, t > 0

where C$o is the initial concentration of  PCBs in  soil.  The solution to Eq.
(A-3) for the above initial and boundary conditions is
                            = (H/Kd)Cso '  erf ,
where
                                              2   n
                   erf (n) = error function = /* / exp(-n<-) dn
                                                  o
     The flux rate at the soil-air interface (N^) can be estimated as a
function of time from equation (A-6) by using the concentration gradient
                                      A-3

-------
as follows:
          NA = -E-D .  !£.
           H       ei  3z
                                 E D
         ei
z=0
                                 /Hat
              v-  so
                                                                   (A-7)
     The boundary conditions used here are superior to those used by Jury et
al . (1983), assuming that the vapor-phase boundary layer is rate-controlling.
Experiments by DuPont (1985) on emission rates from contaminated soil show
that when the emission rates for volatile organics are plotted against the
reciprocal of ^t, a straight line is obtained.  This observation is consis-
tent with the relationships given by Eq. (A-7), and Thibodeaux and Hwang
(1982).  The model derived by Jury et al . (1983), based on the boundary con-
ditions of the controllirrg boundary layer in the air phase, does not provide
a straight-line relationship between emission rate and 1/Tj..  For this
reason, the relationship derived in this report is used for exposure evalua-
tion.
     The average flux rate, NA, over an exposure interval, T, can be calculated
using Eq. (A-7).
                      T            /IlaT     Kd

or
                                NA(T) = 2 NA(T)                       (A-9)

     To estimate the total average emission rate, Q, the flux rate defined in
Eq. (A-9) must be multiplied by the area of soil contaminated.
                                      A-4

-------
                                   Q = A •  NA                         (A-10)

     Furthermore, while Eq. (17) in Section 16 can be used at  any distance
X from the site to estimate air concentrations of  PCBs,  it cannot be used  on-
site.  Although at present there is no generally accepted methodology for
estimating on-site concentrations from an area source,  on-site PCB air con-
centration was estimated based on a "box model" approach, by using the equation
                                   C =
                                       LS-V-H                         (A-ll)
where
     H = mixing height = 2 m
     V = average wind speed within mixing zone
       = 0.5 wind speed at the mixing height  = 0.5 x 4.5 meter/sec = 2.25  m/s
    LS = width dimension of contaminated area perpendicular to the wind  direc-
         tion = /A = 45 m

     A need exists for development of a more  rigorous approach to estimating
on-site ambient air concentrations.  Time constraints did not  allow development
and validation of a rigorous model.
     Estimation of ingestion of contaminated  soil  required the calculation of
an appropriate soil concentration.  This concentration was calculated by deter-
mining the average concentration of PCB in soil  to a depth of  or 25.4 cm (10
inches) for a period of 6 years beginning at  time  0.  Because  the error  func-
tion has no closed-form solution it was approximated by
                                      A-5

-------
                                    2

                           _ a(2n+l) iTt

                          e  L2      ..   cinl(2n±l)n z i                 (A-12)
                  n    L      2n+l
                      n=0
sin{(2nH)nz }
where L Is a depth which was selected such that
                                  Cs (L.t)
for all exposure durations.  In calculations reported in this report, L was


set equal to 250 cm.  Integrating Cs of the exposure duration tQ (5 years) and


depth £ (25 cm) yields an equation for average PCB soil concentration, C^


                                                                        to
             n«            Q      m                           «^^9n4>1^^^
—           ,   ,          v>\   r                              ~   \  ., ' n2tn
C  =—L-  /   /  C  dz =  7    so  I {i-CQS(2n+l n £u . fl-e     41-^      ui


                                                                                 (A-13)



Case 2.  The contaminated surface is covered with PCB-free soil material.  Let
         i = thickness of cover, cm, and L = the depth of contamination mea-
         sured from the top of cover material, cm.  The initial and boundary
         conditions become:


         1.  I.C.           C=0,   0 0


         4.  B.C.          JC_ = 0,   z = L,     at t > 0
                           3z
where Cg is the initial concentration of PCBs in the vapor phase, which can be


obtained by CQ = (H/KjjCso.  E(l- (A~3) witn these initial  and boundary condi-


tions can be solved using the Fourier Series technique.  The solution is
                                     A-6

-------
         «  -g(2n+l)2 II2t
C = 4CQ  I e4T2
    n   n=0
                             sin {  2"+l  n zi  Cos{ln±lnl}
                                 1    2    Lf     l   2    LJ
(A-14)
     The flux rate at the soil-air interface (N^)  can  be  estimated  as  a  func-

tion of time from equation (A-14)
                                              2  2
              z=0
                 o r  c n    "     a (2n+ir n  t
                 2 C0 E>Dei   v  e     4L2           cos(^IniDMl         (A~15)
                                                       i    2   LJ
                               n=0
     The average emission rate over a time period,  T,  can  be obtained  by  inte-

gration of Eq. (A-15).   The result is
NA
^8(H/Kd)'Cso-E-Dei-Ly
          air2T          n=0 (2n+l)2
                                    ,-
                                 - e
                                      a (2n+l)2 n2)t? .  cos((2n+l)  nil
                                             4-L^    J        2    L
                                                                           (A-16)
or
                                                DiE1/3(2nfl)2
                                 y     1

                                n=0 (2n+l)2
                                    (1+K-S)-L2 '4
                                                                   2  L       (A-17)
                                     A-7

-------
     The summation of terms given in  Eq.  (A-16)  can  be  conveniently  carried  out


by means of computer simulation.   The time interval  t£-ti  should  be  set  equal


to the exposure interval.   In calculating exposures, the maximum  average expo-


sure was estimated.  This  was achieved by calculating NA as  a  function of time


and determining the time at which the maximum value  of  N/\  occurred;  tj was then


set equal to this time.

                                                                     2 2
     It should be noted  that when the value of the expression  a(2n+l)  n   -jn

                                                                 4L2
the exponential term of  Eq. (A-16) is small or considerably  less  than  1, the


average of the exponential  term over  a time, t,  will be close  to  1.   In  this


situation, averaging of  the exponential  term of  Eq.  (A-15) by  the integration


formulae given by Eq. (A-16) or Eq.  (A-17) may easily result in an erroneous


answer because one has to  evaluate very precise  numbers of many decimal  points


for the values of the exponential term.  It is more  practical  to  numerically


average Eq. (A-15) than  to obtain the average value  by  using the  integration


formula given by Eq. (A-16) or Eq. (A-17):
                                                                          (A-18)
The steps of the summation and the integration with respect  to n  and t,  respec-


tively, need to be carried out by means  of a  computer.


     As in Case 1, Eqs. (A-10) and (A-ll)  are used to estimate emissions rate


and on-site air concentration of PCBs.   However,  Eq.  (A-17)  or (A-18) is sub-


stituted into Eq. (A-10) as an estimate  of flux rate.


     Also as in Case 1, the average soil  concentration  used  to estimate  inges-


tion of soil must be calculated.  This can be accomplished by noting that C$Q


                                     A-8

-------
= CQ •  Kd/H,  substituting  this  relationship  into  Eq.  (A-14)  and  integrating

the resulting equation  over  the depth  interval  z  and  over  the time  interval

tj to t} + tQ.  The result is
            1 + p ^            -      oo

      1   f     f  Cs  dz = 32L-*CSO   7 M    cos(2n+l)na,  f  -a (2n+l)2  n^t,
     ~fi  t     0S       n^ait    n=0U    COS-2~!r  e4^
(2n+l)2
                            _ e
                                                        cos  f(2n+l)nti
                                                               2     L
                                                                           (A-19)
where to = 5 years and  ?  = 25 cm.

     When the initial  PCB soil  concentration  used  in  estimating  exposures  ex-

ceeds the concentration at which the vapor pressure of  PCB  is  achieved,  a

different model  must be used in both Case 1 and  Case  2.   The vapor  phase PCB

concentration that can  be achieved in the interstitial  voids in  soil  is

limited to the concentration corresponding to the  vapor  pressure.   While this

limits the emission rate, it should be noted  that  as  the soil  zones nearest

the air-soil interface  become depleted of PCB, the emission rate decreases.

If PCB is present in soil concentrations that produce the vapor  pressure in

the vapor phase, the average emission rate may be  increased because soil near

the surface is depleted less rapidly.

     In modeling this phenomenon  it has been  assumed  that at any given  time,

the concentration profile of PCB  in soil as a function  of depth  is  a steady-

state profile.  As in the previous models, the concentration of  PCB in  the

interstitial soil void  space is assumed to be in equilibrium with  PCB soil
                                      A-9

-------
concentrations.  Given these assumptions and the initial  conditions  that
                                       H .  C
                                  Cs ' CSO
where
     Css = PCB soil  concentration at which the vapor pressure is achieved.
     A mass balance can be written to determine the rate of depletion from
soil.  If the soil concentration profile is as defined in Figure A-l, this
mass balance is
                               aC
                         E«r\  •  -
                          Dei   3z
           dz.  = 	       (A-20)
                 2 f  Css[P(l-E)+E'Kd/H]l   + Ps(l-E)(CSo-Css)
     Because we assume that any any time the soil  and vapor PCB approach
their steady-state concentrations,
                                      dz
substituting equation (A-21) into equation (A-20)  and integrating the result-
ing equation over the time interval  0 to t and the corresponding depth inter-
val 0 to z yields the result
                                      A-10

-------
o
0)
Z
III
O
z
o
o
      •88
                                                     t3>*2
                                DEPTH (z)
   Figure A-l.   Model  of chemical  vapor movement through soil when
   partial  pressure  is equal  to vapor pressure.
                                  A-ll

-------
    z _ 2 /	E '  Dei  '  Css'   t	        (A-22)

               E'CSS + 2'P'(l-E)-Cso-Kd/H - (l-E)'P'Css-Kd/H
As in the previous case, the flux rate can be calculated  as
                                       E'D'C.«.fH
                                          °
             "      ei  3x      ei  z         z    Kd                 (A-23)






or




               _ / E-Dei'Css * |2-(1-E)'P-CSO + E-Css'H/Kd  -  (1-E)-P-C5S)

            "A     	

                                2  / t'Kd/H



                                                                     (A-24)





     If the average flux is determined for the time interval  T,  it  is  easy to



show that







                                NA (T) =  2NA(T)                      (A-25)






As indicated previously, Eqs. (A-10) and  (A-ll) can be  used to estimate  emis-



sion rate and on-site air concentrations.   However, Eq.  (A-24) is substituted



into Eq. (A-10) as an estimate of  flux rate in this case.



     The average soil concentration to a  depth of 25 cm over  the exposure



duration of up to 5 years of exposure must be determined to estimate  ingested



dose of PCBs.  The equation used to estimate this average depends on  whether



the depth z in Eq. (A-22) is less  or greater than LI (25 cm)  at  the end



of the ingestion exposure periods.  The time TS at which z  =  LI  is  easily




                                      A-12

-------
calculated using the  following equation:


T5 = O^S-L^-tE-Css  -  2(l-E)P'Cso-Kd/H  -  P-(l-E)-Css'Kd/H)/(EM)el-Css)      (A-26)

     If the ingestion exposure period, T,  is  less  than  15,  the  depth,  z, will
always be less than  LI, and  the  average  soil  PCB concentration,  Cs,  can  be
calculated as follows:
                                                                           1/2
Cs = 2'(Css-2-Cso)  {E-Dei'Css-T/[E'Css+2(l-E)-P-Cso-Kd/H  -  (1-E)-P-CSS-Kd/H]}
                                                                     +  CSO
                                                                         (A-27)
     If the ingestion exposure  period  is  greater than  15,  the  depth,  z,  will  be
greater than Lj at the end  of the  exposure  period,  and the average  soil  PCB
concentration can be calculated as follows:
Cs = 2-(Css-2'Cso) {E-Dei-Css/[E-Css+2(l-E)-P-CSQ-Kd/H
                         °'5  i  t
     - (l-E)-Css-P'Kd/H]}   'Is1'5  /(3-L!-!)  + CSQ-T5/T
       2'(CSS-L1) {[E-CS5+2(1-E)'P-CSO  -  (l-E)-P'Css-Kd/H]
     [E'Dei-Css]} -5 {T°-5 - T5°-5}/T                            (A-28)
                                      A-13

-------
     When clean cover is  placed  over  contaminated  soil,  a  similar  model  can
be developed as in the case where  soil  is  contaminated to  the  surface.   For
such situations, assuming that  local  equilibrium between vapor and solid
phases and steady-state concentration distributions  at any time are attained,
the following mass balance which yields  relationships illustrated  in Figures
A-2 and A-3, which define the concentrations  of PCB  in soils as a  function of
depth

L4 = U'L^U-EJ

   + ^-(E'H/Kjj + (1-E)-P)-CSS-[2(1-E)-P'(CSO-CSS) + E'H/Kd  + (1-E) -P) -Css )]}

   / {2-[2-d-E)'P-(Cso-Css) +  [E'H/Kd+(l-E)-P]'Css]}                 (A-29)

     The time at which PCB reaches the air-soil interface, Tj,, can be estimated
by rearranging Eq. (A-22) and substituting 14 + LI for z,  as follows:

Tb = (L4+L1)2-{E-Css+2'(l-E)-P-Cso-Kd/H  -  (1-E)-P'C^-Kj/H}

     •{A-E'Dgi-C^}-1                                                (A-30)

     Integrating Eq. (A-24) over the  exposure time interval Tjj to  T+Tb  and
dividing the result by T yields  the following expression for the average flux
over the exposure period:
                                      A-14

-------
_i   CSO
o
m
O   C
UJ
O
z
o
o
     88
                                       DEPTH (z)
   Figure A-2.  Model of chemical vapor movement  through  soil  when partial
   pressure is equal to vapor pressure.
                                      A-15

-------
o
H
<
UJ
o
z
o
o
o
CO

o
o
Q.
DEPTH  OF CLEAN  COVER
                                    DEPTH
  Figure A-3.  Mass  balance  for vapor movement through soil when
  partial  pressure  is  equal  to vapor pressure.
                                 A-16

-------
      ,
NA= - /    NA dt
                  / E-Dei'Css  (Z-(1-E)-P-CSO  +  E'Css-H/Kd  -  (1-E)-P-CSS

                                         /  Kd/H
     As before, Eqs. (A-10)  and (A-ll)  can  be  used  to  estimate  emission  rates
and on-site air concentrations.
     Finally, the average soil  concentration to  a depth  of  25 cm over the
exposure of duration up to 5 years must be  determined  in order  to estimate
ingested dose of PCBs.   The  equation used to estimate  this  average is:
                                     /(T-(L1+L4)}                    (A-32)
Calculation of the Depth-Averaged Concentration  for  Uncovered  Surface
     We want to find the average concentration of  PCBs  in  soil  over the expo-
sure period.  As time progresses, the concentration  in  soil  decreases because
of volatilization.  First, we want to find  the time  when the emission rate at
any time equals the average emission rate.   We equate Eqs.  (A-7)  and (A-8).
Then
                        E'Dei    H   C   _  2E "Dei    H    C
                        - '  F '   $0  ~  - •  -J7 •   SO
                        /Hat    Kd        /naT      Kd
                                      A-17

-------
     The emission rate equals  the  average emission rate at t = T/4, where T is

the exposure time (1  day,  10 days, or 70 years), and t is any time.  From Eq.

(A-6), the vertically-averaged  concentration over a depth of 2 cm is:
                          _              z
                          C = H  . C   If  erf(_L_) dz                (A-34)
                              Kd    SO z J0	
                                                2/at
or
                         _          z
                         C  = C   If  erf(_L_) dz                     (A-35)
                           s    SO z j 0
                                           2/at
The integral  should  be  numerically evaluated at t = T/4, T being the exposure

period for developing health advisories, and at an appropriate depth within

which the concentration average  is desired.
                                     A-18

-------
                            REFERENCES  (APPENDIX  A)
DuPont, R.R.  (1985,  Nov.)   Evaluation  of  air  emission  release  rate  model  pre-
     dictions of hazardous  organics  from land  treatment  facilities.   Presented
     at American Institute  of  Chemical  Engineers  meeting,  Chicago,  IL.

Farmer, W.J.; Yang, M.-S.;  Letey,  J., Dept.  of Soil  and  Environmental  Sciences,
     University of California-Riverside; Spenser, W.F.,  Science and  Education
     Administration,  Federal Research,  USOA.   (1980)  Land disposal  of  hexa-
     chlorobenzene wastes:  controlling  vapor movement in soil.   EPA-600/2-80-
     119.  Prepared for U.S. Environmental  Protection Agency, Municipal
     Environmental Research Laboratory,  Cincinnati,  OH.

Hwang, S.T.  (1982)  Toxic  emissions from land disposal  facilities.   Environ.
     Prog. 1:46.

Jury, W.A.; Spencer,  W.F.;  Farmer, W.J.   (1983)  Behavior  assessment model  for
     trace organics in soil.   I.   Model  description.  J. Environ. Qual.  4:558-
     564.

Thibodeaux, L.J.  (1979)  Chemodynamics.  New  York,  NY:  John Wiley and  Sons.

Thibodeaux, L.J.; Hwang, S.T.   (1982)   Landfarming of petroleum wastes:
     modeling the air emission problem.   Environ. Prog.  1:42-46.
                                      A-19

-------
                         APPENDIX B
EXAMPLE EMISSION RATE CALCULATIONS FOR FOUR STUDIED SCENARIOS

-------
     Example calculations  for estimating  volatile  emissions  of  PCBs  based
on the models presented in Appendix  A  for unsteady-state  conditions, and
under steady-state conditions are  presented  below.

Case 1.  Unsteady-state emission:  no cover.
    The contaminated soil  is  exposed to the  atmosphere, and  no  clean soil
cover is applied on top of the contaminated  surface.   PCB-1254  is  used  as  an
example.  The average emission rate, which is  obtained by  averaging  the
instantaneous emission rate over a time period,  t  (sec),  can be obtained by:
                          d  cm  air
where                a =	§1
                         E + PS(1-E)
                       .   "tl
                         1 + S •  K
with S
= __ , and K = -1 Ps
Dei = 0.05 (0.35)1/3 = 0.0352 cm2/s
       1L in  9 so11  = 8.37 x IP"3 x  41  =  0  000343
                                           "
       Kd    cm3 air     100°              "        m9/9  soil
                                     B-l

-------
Hence:
                 	0-0123	  = 2.45 x 10'6 cm2/s
                 0.35 + ?.fi5(n.fiS)    1
                                   0.000343
     The average emission rate when the initial  concentration in soil,

C
-------
contamination depth, L = 200 cm,  can  be  obtained  by  Eq.  (A-13):
-  . 2(H/Kd)Cso.E.Qe1  .  i J
For an average emission over a period of  10  days  (864,000 sec)
          2(0.000342)(10"6)(0.35)(0.0352)
     NA = 	
                    (200)(864,000)
     „   864,000    2.45x10  (2n+l) n2t     m       f2n+l)n(25.4)  ,
     I  /        e       4(200)2                 l     2    200  '
    n=0  °
    = 7.6 x lO'14 g.'cm2.s

Similarly, the average emission rate over  a  period of  70 years

    = 5 x lO'14 g/cm2-s

NOTE:  The emission rate equation is programmed  in a computer, and the sum-
mation is carried out using the program.
Case 3.  Steady-state emission—no cover.
     The same scenario assumes that the contaminated surface is exposed to the
atmosphere, and its surface is maintained  at the concentration of interest over
the period of emission.  This will apply to  the  case where a large reservoir of
PCBs is available for emission until the concentration at the surface reaches
                                     B-3

-------
the saturation value.  This will  be the upper-bound emission rate for contami

nated soil with no clean cover applied.

PCB-1254


          H_        0.34 g/cm3 air/g/cm3 water

          Kd "  s   1000 mg/g soil/mg/cm3 water  s


                            mg/cm3 ai r
                  = 0.00034 mg/g S0ii    0.001  mg/g soil



                  = 3.4 x 10-7 mg/cm3 air = 3.4 x 10~7 x 10'3 x 10'6 '-g/m3


                  = 340 ng/m3



     Partitioned vapor concentration:   340 ng/m3

          Vapor pressure at 25°C = 7.71 x 10 ~5 mmHg
          C* = PMW  = 14.7 (7.71x10-5/760)  •  328.4 = 8.5x10-8 #/ft3
               RT     10.73 (460 + 77)
where F = 1.8 (25) + 32 = 77°F,
     C* = saturation concentration of PCB-1254 in #/ft3 or ng/m3,
      P = vapor pressure,
     MW = molecular weight,
      R = gas constant.
          8.5x10-8 (454) x Ip6 ug  = 1362.7 ng/m3
          (0.3048)3 m3
When PCB-1254 is greater than 4 ppm in soil,  the partial  pressure in the air

phase is equal to vapor pressure.   An average temperature of 25°C is used for

evaluating the saturation concentration in the vapor phase.  Since the vapor
                                      B-4

-------
pressure is dependent upon temperature,  a  specific  evaluation  will  require a
temperature of interest.   The temperature  in  the subsurface  soil  may not  fluc-
tuate considerably over the average value.

PCB-1242
     Concentration in air above  soil  with  1  ug/g PCB

          340 x 5.73xlO-4  = 23.3
                8.37xl03

     Saturated vapor concentration
          1362.7 4.06xlO-4  266.5   = 5823.3 yg/ra3
                 7.71xlO-5  328.4
     Note:   When the PCB concentration  in  soil  =  250  pg/g  =  5823.3   the
                                                           (  23.3 ),
            vapor phase is saturated.
     In order to calculate the emission rate, the values for  gas-phase mass
transfer coefficients are estimated using  the relationship given  by Hwang
(1982).
        PCB-1254; kg = 5.8x10-5   18   0-5
                                (328.4)    = 1.36xlO-5  g mol/cm2 •  s

        PCB-1242; kg = 5.8x10-5 .  18   °'5 = 1.51X10'5  g mol/cm2 .  s
                                (266.5)
     Mole fractions of PCBs in  the gas  phase  (y)  when  the concentrations are
retained at 340 ug/m3 and 23.3  ug/m3 for PCB-1254 and  PCB-1242,  respec-
tively, are:
                                      B-5

-------
     PCB-1254     y = 340 pg/m3 •  0.0742  ppb/(ug/m3)  x  10-9  =  2.52x10-8

     PCB-1242     y = 23.3 wg/m3 • 0.0916 ppb/(yg.m3) x 10'9 = 2.13x10-9


     The emission rate,  Q, can be  obtained from the  formula  Q  = MW-kg-y,

where MW = molecular weight (Hwang,  1982).   Thus,  the emission rate  for each

Aroclor is:



     PCB-1254     Q = 328.4 (1.36xlQ-5)(2.52xlO-8) =  l.lSxlO'10 g/s  cm2

     PCB-1242     Q = 266.5 (1.51x10-5)  (2.13x10-9)  = 8.57x10-12 g/s Cm2



Case 4.  Steady-state emission—cover applied.

     It is assumed that  the contaminated  site  is covered with  PCB-free soil

cover, and that the concentration  at the  top of the  contaminated soil is  main-

tained at 1 ug/g over the period of  emission.   The emission  rate can be obtained

from the formula (Hwang, 1982).
          Q = D1  PT4/3 C* ,  g/cm2-s
                       h
where D^ = diffusivity, cm2/s,
      PT = total porosity,
      C^ = true vapor pressure in equilibrium with  soil,  g/cnr,
       h = cover thickness, cm.


Using the diffusivity value of 0.05 cm2/s,  one can  get  for PCB-1254,  an

emission rate of
          Q = (0.05K0.35)4/3 340 ug/m3 ID"6 g/ug x 1Q-6 cm3/m3
                                        25.4 cm
            =  1.67 x 10-13 g/cm2-s


                                     B-6

-------
Similarly, for PCB-1242:
          Q = (0.05) (0.35)4/3 23.3xlQ-6 x IP"6 = 1.14xlO'14 g/cm2 •  s
                                    25.4
                                     B-7

-------
                           REFERENCES  (APPENDIX  B)
Hwang, S.T.   (1982)   Toxic emissions  from land disposal  facilities.   Environ.
     Prog. 1:46.
                                     B-8

-------
                 APPENDIX C










 SUMMARY OF COMPUTER RUNS FOR EACH AROCLOR AND



AT EACH VALUE OF SOIL-AIR PARTITION COEFFICIENTS

-------
                                                 TABLE  C-l.   PERMISSIBLE  PCB-1242  SOIL CONTAMINATION LEVELS9
                                                         (UNCOVERED  SURFACE CONTAMINATION.  Kd  =  1000)
o
 i
                                                                     Permissible levels  (ug/g)  corresponding to
                                          Noncancer  short-term3
                                        acceptable Intake
                                                                  Cancer risk specific doses  (ug/day)
Location and
route of human
exposure
100
for child
700
for adult
0.00175
(10-' risk)
0.0175
(10-6 riS|(]
0.175
1 (10-5 risk)
1.75
(10-4 risk)
On the contaminated site

 - Soil 1ngest1onc,         55
   inhalation*

 - Soil ingest1ond,         92
   Inhalation6

 - Inhalation only6         116
510


2100


vs
                                                                            0.008
                                                                            0.06
                                                                            0.2
0.08
0.6
0.8
             20
            61
            204
                0.1  km from
                 contaminated  site           vsf
                 - Inhalation  only6

                1 km from
                 contaminated  site           vs
                 - Inhalation  only*
                                           vs
                                           vs
                                                            20
                                                            428
                                  204
             l.lxlO4     vs
                                  3.1xl04      vs
                                                                                                      vs
                aShort-term 5 10-day  Intake.
                bBased on average weights  of  10 and  70 kg  for  a  child  and  an  adult,  respectively.
                cChildren ages 1-5, with pica  (consuming 3 g soil/day).
                Children ages 1-5, without pica (consuming 0.6  g soil/day).
                elnha1ation rates are assumed  to be  20 m^/day  for the  short-term and longer-term noncancer exposures;
                 all  other (more chronic)  exposures  assumed to be 10 m^/day as  a result  of  182  days  exposure per year.
                'vs  denotes no theoretical upper-bound limit.  Practical reasons require no free-flowing PCB liquid  for the limit.
                9So11-a1r partition coefficient = 2.35 x 10"5  g  soil/cm3 air  (=• H.41/Kd  = 5.73  x 10"1  (41J/1000 =
                 2.35 x 10-5).

-------
                                                 TABLE C-2.  PERMISSIBLE PCB-1242 SOIL CONTAMINATION LEVELS?
                                                          (UNCOVERED SURFACE CONTAMINATION. Kd » 40)
                                                                     Permissible levels (ug/g) corresponding to
                 Location and
                route of human
                  exposure
                          Noncancer short-term9
                        acceptable Intake iig/dayb

                             100           700
                          for child     for adult
                	Cancer risk  specific  doses  dig/day)	

                   0.00175         0.0175         0.175      1.75
                 (10-' risk)      (ID'6 risk)  (10'5 risk)   (10"4  risk)
o

(S3
On the contaminated site

 - Soil IngestlonC,         60
   Inhalation6

 - Soil ingestiond,         247
   Inhalation6

 - Inhalation only6         vsf
690


2800


vs
.01


.03


0.04
                                                                                             0.1
0.3
                                                                                             .4
             1.0
3.0
            13
35
                         110
                0.1 km from
                 contaminated site
                 - Inhalation only6

                1 km from
                 contaminated site
                 - Inhalation only6
                            vs
                                           vs
                                           vs
                                                            310
                                                                             110
                                               1.1x10*     vs
                                  3.1xl04      vs
                                                                                                      vs
                aShort-term » 10-day Intake.
                bBased on average weights of 10 and 70 kg for a child and an adult, respectively.
                cCh1ldren ages 1-5, with pica (consuming 3 g soil/day).
                ^Children ages 1-5, without pica (consuming 0.6 g soil/day).
                6Inhalation rates are assumed to be 20 m^/day for the short-term and longer-term noncancer exposures;
                 all other (more chronic) exposures assumed to be 10 m-Vday as a result of 182 days exposure per year.
                fys denotes no theoretical upper-bound limit.  Practical  reasons require no free-flowing PCB liquid for the limit.
                ^Soil-air partition coefficient - 2.35 x IO'5 g soil/cm3  air (= H.41/Kd = 5.73 x 10'* (41)/1000 =
                 2.35 x ID'5).

-------
                                                TABLE C-3.  PERMISSIBLE PCB-1248 SOIL CONTAMINATION LEVELS0.
                                                         (UNCOVERED SURFACE CONTAMINATION. Kd = 1000)
                Location and
               route of human
                 exposure
                                                                    Permissible levels (pg/g) corresponding to
                          Noncancer short-term3
                        acceptable Intake pg/dayp

                             100           700
                          for child     for adult
                	Cancer risk specific doses (pg/day)	

                   0.00175         0.0175         0.175      1.75
                 (10-7 risk)     (10-6 risk)  (10'5 risk)   (10'4 risk)
i
CO
On the contaminated site

 - Soil IngestlonC,         32
   Inhalation6

 - Soil 1ngest1ond,         42
   Inhalation6

 - Inhalation only6         47
612 •


2500


vs
                                                                           0.01
                                                                           0.04
                                                                           O.OB
0.1
0.5
0.8
            10
            49
            110
               0.1 km from
                contaminated site
                - Inhalation only6

               1 km from
                contaminated site
                - Inhalation only6
                            VST
                                           vs
                                           vs
                 8.0
                                                            270
110
8,700       8.7xl05
                                  2.5xl04      vs
                                                                                                      vs
               aShort-tenn = 10-day Intake.
               bBased on average weights of 10 and 70 kg for a child and an adult, respectively.
               cCh1ldren ages 1-5, with pica (consuming 3 g soil/day).
               ^Children ages 1-5, without pica (consuming 0.6 g soil/day).
               elnhalat1on rates are assumed to be 20 m3/day for the short-term and longer-term noncancer exposures;
                all other (more chronic) exposures assumed to be 10 m3/day as a result of 182 days exposure per year.
               fvs denotes no theoretical upper-bound limit.  Practical reasons require no free-flowing PCB liquid for the limit.
               9Soil-air partition coefficient = 2.35 x 10'5 g soil/cm1 air (= H.41/Kd = 5.73 x 10'" (41J/1000 =
                2.35 x ID-5).

-------
                                 TABLE C-4.  PERMISSIBLE PCB-1248 SOIL  CONTAMINATION LEVELS9
                                          (UNCOVERED SURFACE CONTAMINATION.  Kd =  40)
                                                     Permissible levels  (ug/g)  corresponding to
                          Noncancer short-term3
                        acceptable Intake pg/day°
                                                                  Cancer risk specific  doses (tig/day)
Locatto and
route of human
exposure
100
for child
700
for adult
0.00175
(10-' risk)
0.0175
(lO'6 risk)
0.175
(lO"5 risk)
1.75
(10-4 risk)
On the contaminated site

 - Soil 1ngest1onc.
   Inhalation6

 - Soil Ingest 1ond,
   Inhalation6

 - Inhalation only6
                            80
                            330
710


2900


vs
0.01


0.02


0.02
O.I
0.2
                                                                             0.2
2.0
             2.0
37
            87
0.1 km from
 contaminated site          vs
 - Inhalation only6

1 km from
 contaminated site          vs
 - Inhalation only6
                                           vs
                                           vs
                                                            2.0
                                                            250
                                  90
                                  2.5x10*
                              8,700
                                                                                          vs
                                                                                                      vs
                                                                                                      vs
aShort-tenn s 10-day Intake.
DBased on average weights of 10 and 70 kg for a child and an adult,  respectively.
cCMldren ages 1-5, with pica (consuming 3 g soil/day).
^Children ages 1-5, without pica (consuming 0.6 g soil/day).
^Inhalation rates are assumed to be 20 m^/day for the short-term and longer-term noncancer exposures;
 all other (more chronic) exposures assumed to be 10 m^/day as a result  of  182 days  exposure per year.
fvs denotes no theoretical upper-bound limit.  Practical  reasons require no free-flowing PCB liquid  for the  limit.
^Soil-air partition coefficient = 2.35 x 10~s g soil/cm3  air (= H.41/Kd  = 5.73 x 10'"  (41)71000 =
 2.35 x 10-5).

-------
                                 TABLE C-5.  PERMISSIBLE PCB-1254 SOIL CONTAMINATION LEVELS?
                                          (UNCOVERED SURFACE CONTAMINATION, Kd * 1000)
 Location and
route of human
  exposure
                                                     Permissible levels (ng/g) corresponding to
                          Noncancer short-term3
                        acceptable Intake ug/dayb
   100
for child
   700
for adult
                                        Cancer risk specific  doses  (tig/day)
  0.00175
(10-7 risk)
  0.0175         0.175      1.75
(10-6 risk)   (10-5  risk)   (10'4 risk)
On the contaminated site

 - Soil IngestlonC,         90
   Inhalation6

 - Soil Ingest1ond,         370
   Inhalation6

 - Inhalation only6         vs'
                 720


                 2980


                 vs
                    0.01
                    0.04
                    0.05
                 0.1
                 0.4
                 0.5
                          12
                          59
                          460
0.1 km from
 contaminated site          vs
 - Inhalation only6

1 km from
 contaminated site          vs
 - Inhalation only6
                 vs
                 vs
                    '•f
                                                   460
                                                  4.7xl04      4.7xl06
                                                   1.3xl05      vs
                                                                            vs
aShort-terai a 10-day Intake.
''Based on average weights of 10 and 70 kg for a child and an adult, respectively.
cCh11dren ages 1-5. with pica (consuming 3 g soil/day).
dCh11dren ages 1-5, without pica (consuming 0.6 g soil/day).
elnhalatlon rates are assumed to be 20 nvvday for the short-term and longer-term noncancer exposures;
 all other (more chronic) exposures assumed to be 10 m^/day as a result of 182 days exposure per year.
rvs denotes no theoretical upper-bound limit.  Practical reasons require no free-flowing PCB liquid for the limit.
^Soli-air partition coefficient - 2.35 x W5 y soil/cm* air (= H.41/Kd = 5.73 x 10'" (41)71000 =
 2.35 x lO*5).

-------
                                                 TABLE C-6.  PERMISSIBLE PCB-1254 SOIL CONTAMINATION LEVELS9
                                                          (UNCOVERED SURFACE CONTAMINATION.  Kd = 40)
                                                                     Permissible levels (pg/g)  corresponding to
                 Location and
                route of human
                  exposure
                                          Noncancer short-term3
                                        acceptable Intake ug/dayD
                             100
                          for child
   700
for adult
	Cancer  risk  specific  doses  (pg/day)	

   0.00175          0.0175          0.175       1.75
 (10-'  risk)      (10-6  risk)   (10'5  risk)   (10'4  risk)
i
(ft
On the contaminated site

 - Soil Ingest Ion*.         100
   Inhalation6

 - Soil 1ngest1ond,         420
   Inhalation6

 - Inhalation only6         vsf
   730


   3000


   vs
                                                                            0.009
                                                                            0.01
                                                                            O.dl
                  0.09
                  0.1
                  0.1
2.0
12
            36
            470
                0.1 km from
                 contaminated site          vs
                 - Inhalation only6

                1 km from
                 contaminated site          vs
                 - Inhalation only6
                                           vs
                                           vs
                                                            1300
                                                                             470
                                                  4.7x10*      vs
                                     1.3xl05       vs
                                                                                                      vs
                aShort-term s 10-day Intake.
                ''Based on average weights of 10 and 70 kg for a child and an adult, respectively.
                cChtldren ages 1-5, with pica (consuming 3 g soil/day).
                dCh11dren ages 1-5, without pica (consuming 0.6 g soil/day).
                elnhalat1on rates are assumed to be 20 m-Vday for the short-term and longer-term noncancer exposures;
                 all other (more chronic) exposures assumed to be 10 nr/day as a result of 182 days exposure per year.
                fvs denotes no theoretical upper-bound limit.  Practical  reasons require no free-flowing PCB liquid for the limit.
                9So1l-a1r partition coefficient = 2.35 x 10~s g soil/cm1  air (= H.41/Kri = 5.73 x lO'" (41)/1000 =
                 2.35 x ID'5).

-------
                                 TABLE C-7.  PERMISSIBLE PCB-1260 SOIL CONTAMINATION LEVELS9
                                          (UNCOVERED SURFACE CONTAMINATION. Kd = 1000)
                                                     Permissible levels (n9/9) corresponding  to
 Location and
route of human
  exposure
                          Noncancer short-term3
                        acceptable Intake ug/dayp
   100
for child
   700
for adult
                                        Cancer risk specific  doses  (ug/day)
  0.00175
(10-' risk)
  0.0175
(10-6
J-175
    risk)
  1.75
(10-4 risk)
On the contaminated site

 - Soil 1ngest1onc,         25
   Inhalation6

 - Soil ingest1ond,         61
   Inhalation6

 - Inhalation only6         vs'
                 640


                 2670


                 vs
                    0.01


                    0.04


                    0.06
                 0.1
                 0.4
                 0.6
           12
           48
           91
0.1 km from
 contaminated site          vs
 - Inhalation only6

1 km from
 contaminated site          vs
 - Inhalation only6
                 vs
                 vs
                                  240
                                                   91
                                                  7.7xl03      vs
                                                                            vs
aShort-term a 10-day Intake.
bBased on average weights of  10 and 70 kg for a child and an adult, respectively.
cCh11dren ages 1-5, with pica (consuming 3 g soil/day).
''Children ages 1-5, without pica (consuming 0.6 g soil/day).
6lnhalat1on rates are assumed to be 20 m3/day for the short-term and longer-term noncancer  exposures;
 all other (more chronic) exposures assumed to be 10 m3/day as a result of 182 days  exposure  per year.
fvs denotes no theoretical  upper-bound limit.  Practical  reasons require no free-flowing  PCB  liquid  for  the  limit.
«Soil-air partition coefficient = 2.35 x 10"5 g soil/cm3  air (= H.41/Kd = 5.73 x 10"1  (41J/1000 =
 2.35 x ID'5).

-------
                                                 TABLE C-8.   PEKHISSIBLE  PCB-1260  SOIL CONTAMINATION LEVELS?
                                                          (UNCOVERED  SURFACE CONTAMINATION, Kd  =  40)
                                                                     Permissible  levels  (ug/g) corresponding to
                 Location and
                route  of  human
                  exposure
                                          Noncancer short-term3
                                        acceptable Intake nq/dayb
                             100
                          for child
   700
for adult
                                                                  Cancer risk  specific doses Ug/day)
  0.00175
(10-' risk)
  0.0175
(10-6 r1sk)
i-8
    '175
  1.75
(10-4 risk)
o
i
co
                On  the contaminated site

                 -  Soil  1ngest1onc,
                   Inhalation6

                 -  Soil  1ngest1ond,
                   inhalation6

                 -  Inhalation  only6
                0.1  km from
                 contaminated  site
                 -  Inhalation  only6

                1 km from
                 contaminated  site
                 -  Inhalation  only6
                            87


                            360


                            vs'



                            vs



                            vs
   710


   2900


   vs



   vs



   vs
0.01


0.01


0.01



1.0



220
 0.1
 0.1
 0.1
 76
 2.2xl04
1.0
1.0
1.0
7600
                                                                                          vs
           17
           40
           77
           7.6x10*
                                                                                                      vs
                aShort-term s  10-day  Intake.
                bBased  on  average  weights  of  10 and 70 kg for  a  child and  an adult, respectively.
                cChildren  ages  1-5, with pica (consuming 3 g soil/day).
                dCh1ldren  ages  1-5,  without pica (consuming  0.6  g  soil/day).
                elnhalatlon  rates  are assumed to be 20 m3/da
                                        3/day for the short-term and longer-term noncancer  exposures;
 all  other (more chronic) exposures assumed to be 10 m-vday as a result  of 182 days  exposure per  year.
fys denotes no theoretical upper-bound limit.  Practical  reasons require no free-flowing  PCB liquid  for the  limit.
9So1l-a1r partition coefficient = 2.35 x 10'5 g soil/cm'  air (- H.41/Kd  - 5.73 x 10"«  (41)/1000 =
 2.35 x 10-5}.

-------
                                                TABLE C-9.  PERMISSIBLE PCB-1242 SOIL CONTAMINATION LEVELS9
                                                         (25-cm-THICK CLEAN SQIL COVER, Kd = 1000)
                                                                    Permissible  levels  (ug/g) corresponding to
                Location and
                route of human
                 exposure
                                         Noncancer short-term8
                                       acceptable Intake ug/day°
                             100
                          for child
   700
for adult
                                                                  Cancer risk specific doses (ng/day)
  0.00175         0.0175         0.175
(10-7 risk)     (10-6 rtsk)   (10-5 rtsk)
  1.75
(10-4 risk)
£->
I
IO
On the contaminated site

 - Soil ingestlonC,         200
   inhalation6

 - Soil ingest 1ond,         820
   inhalation6

 - Inhalation only6         vsf
   1400


   5700


   vs
0.2


0.6


0.9
                                                                                                         48
                                                                                                         86
                                                                                                                     170
260
                                                                                                                     vs
               0.1 km  from
                contaminated site          vs
                -  Inhalation only6

               1 km  from
                contaminated site          vs
                -  Inhalation only6
                                           vs
                                           vs
                                                            85
                                                            vs
                                                                             vs
                                                                             vs
                                                                                          vs
                                                                                          vs
                                                                                                      vs
                                                                                                      vs
               aShort-term 2  10-day intake.
               bBased on average weights of 10 and 70 kg for a child and an adult, respectively.
               Children ages 1-5, with pica (consuming 3 g soil/day).
               Children ages 1-5, without pica (consuming 0.6 g soil/day).
               elnhalation rates are assumed to be 20 m'/day for the short-term and longer-term noncancer exposures;
                all other (more chronic) exposures assumed to be 10 nryday as a result of 182 days exposure per year.
               fvs denotes no theoretical upper-bound limit.  Practical reasons require no free-flowing PCB liquid for the limit.
               9Soil-air partition coefficient = 2.35 x ID'S g soil/cm* air (= H.41/KH = 5.73 x 10"11 (41)/1000 =
                2.35 x ID-5).

-------
                                                TABLE C-10.  PERMISSIBLE  PCB-1242  SOIL  CONTAMINATION  LEVELS^
                                                         (25-cm-THICK CLEAN SOIL COVER, Kd  =  40)
o
i
                                                                    Permissible levels  (ug/g)  corresponding  to
                                         Noncancer short-term9
                                       acceptable Intake pg/dayb
                                                                  Cancer  risk  specific doses  (ug/day)
Location and
route of human
exposure
100
for child
700
for adult
0.00175
(10-7 risk)
0.0175
(10-6 risk)
0.175
(10-5 risk)
1.75
(10-4 risk)
On the contaminated site

 - Soil Ingest1onc,         170
   Inhalation6

 - Soil 1ngest1ond.         450
   Inhalation6

 - Inhalation only6         vsf
1200


3100


vs
                                                                           0.03
                                                                           0.1
                                                                           1.0
0.3
1.0
                                                                                            vs
3.0
12
                                                                                                         vs
                                                                                                                     vs
                                                                                                                     vs
                                                                                                                     vs
               0.1 km from
                contaminated site          vs
                - Inhalation only6

               1 km from
                contaminated site          vs
                - Inhalation only6
                                           vs
                                           vs
                                                            vs
                                                            vs
                                                                             vs
                                                                             vs
                                                                                          vs
                                                                                          vs
                                                                                                      vs
                                                                                                      vs
               aShort-term s 10-day Intake.
               bBased on average weights of 10 and 70 kg for a child and an adult,  respectively.
               'Children ages 1-5, with pica (consuming 3 g soil/day).
               dCMldren ages 1-5, without pica (consuming 0.6 g soil/day).
               elnhalat1on rates are assumed to be 20 m^/day for the short-term and longer-term noncancer  exposures;
                all other (more chronic) exposures assumed to be 10 m-Yday as  a result  of  182  days  exposure  per year.
               fvs denotes no theoretical upper-bound limit.  Practical  reasons require no free-flowing  PCB  liquid  for  the  limit.
               9So11-a1r partition coefficient = 2.35 x 10'5 g soil/cm*  air (= H.41/Kd  = 5.73  x 10"1  (41)/1000 =
                2.35 x 10-*).

-------
                                 TABLE C-ll.  PERMISSIBLE PCB-1248 SOIL CONTAMINATION LEVELS9
                                          (25-cm-THICK CLEAN SOIL COVER, Kd = 1000)
                                                     Permissible levels (ug/g)  corresponding to
 Location and
route of human
  exposure
                          Noncancer short-term3
                        acceptable Intake pg/dayp
   100
for child
   700
for adult
	Cancer risk  specific doses  tug/day)	

   0.00175         0.0175          0.175      1.75
 (10-7 risk)      (10-6  risk)   (lO'5  risk)   (lO'4 risk)
On the contaminated site

 - Soil Ingest Ionc,         190
   Inhalation6

 - Soil 1ngest1ond,         650
   Inhalation0

 - Inhalation only6         vs^
0.1 km from
 contaminated site          vs
 - Inhalation only6

1 km from
 contaminated site          vs
 - Inhalation only0
                 1300


                 4500


                 vs



                 vs



                 vs
                    0.09


                    0.1


                    0.1
                                                               I
                    14
                                  vs
                               10
                               10
                               14
                  19.000        vs
                                                   vs
                                                                vs
26
93
19,000
                                                              vs
                                                                            vs
aShort-tenn a 10-day Intake.
bBased on average weights of 10 and 70 kg for a child and an adult, respectively.
cCh<1dren ages 1-5, with pica (consuming 3 g soil/day).
^Children ages 1-5, without pica (consuming 0.6 g soil/day).
elnhalation rates are assumed to be 20 nvVday for the short-term and longer-term noncancer exposures;
 all other (more chronic) exposures assumed to be 10 nvvday as a result of 182 days exposure per year.
fys denotes no theoretical upper-bound limit.  Practical reasons require no free-flowing PCB liquid for the  limit.
9Soll-air partition coefficient = 2.35 x 10'5 g soil/cm* air (= H.41/Kd = 5.73 x 10'" (41)71000 =
 2.35 x 10-*).

-------
                                                 TABLE C-12.  PERMISSIBLE PCB-1248 SOIL CONTAMINATION LEVELS9
                                                          (25-cm-THICK CLEAN SOIL COVER. Kd = 40)
                                                                     Permissible levels (pg/g)  corresponding to
                 Location and
                route of human
                  exposure
                                          Noncancer short-terma
                                        acceptable Intake ng/dayb
                             100
                          for child
   700
for adult
                                                                  Cancer risk specific doses (ug/day)
 0.00175
10-7 risk)
0.0175
       _          0.175
(10-6 risk)   (10-' r)skj
  1.75
(10-* risk)
r>
i
On the contaminated site

 - Soil 1ngest1onc,          160
   inhalation6

 - Sol I Ingest Ion*1,          vs^
   inhalation6

 - Inhalation only6          vs
                                                           1100
                                                           vs
                                                           vs
                    0.01


                    0.02


                    0.02
                0.1
                                                                                             0.2
                                                                                             0.2
            1.0
                             2.0
                             2.0
                          460
                        2500
                        1.9x104
                0.1 km from
                 contaminated site          vs
                 - Inhalation only6

                1 km from
                 contaminated site          vs
                 - Inhalation only6
                                           vs
                                           vs
                                                            2.0
                                                            vs
                                     1.9xl04
                                                                             vs
                                                                                          vs
                                                                                          vs
                                                                                                      vs
                                                                                                      vs
                aShort-term 3 10-day intake.
                bBased on average weights of 10 and 70 kg for a child and an adult,  respectively.
                Children ages 1-5. with pica (consuming 3 g soil/day).
                dChildren ages 1-5, without pica (consuming 0.6 g soil/day).
                elnhalation rates are assumed to be 20 mVday for the short-term and longer-term noncancer exposures;
                 all other (more chronic) exposures assumed to be 10 m3/day as a result of 182 days  exposure per year.
                fvs denotes no theoretical upper-bound limit.  Practical  reasons require no free-flowing PCB liquid for  the  limit.
                ^Soil-air partition coefficient = 2.35 x lO'* g soil/cm*  air ( = H.41/Kd = 5.73 x 10'"  (41)/1000 =
                 2.35 x 10-*).

-------
                                                 TABLE C-13.  PERMISSIBLE PCB-1254 SOIL CONTAMINATION LEVELS9
                                                          (25-cm-THICK CLEAN SOIL COVER. K«j = 1000)
                                                                     Permissible levels (ug/g) corresponding to
                 Location and
                route of human
                  exposure
                                          Noncancer short-term9
                                        acceptable Intake pg/day"
                             100
                          for child
   700
for adult
                                                                  Cancer risk specific doses (ijg/day)
  0.00175
(10-' risk)
0.017S
                 0.175
(10-5 risk)   (ID-5  risk)
  1.75
(10-4 risk)
o
i
On the contaminated site

 - Soil 1ngestlonc.         180
   Inhalation6

 - Soil 1ngesttond,         520
   Inhalation6

 - Inhalation only6         vs'
   1300


   4000


   vs
                                                                            0.02
                                                                            0.06
                                                                            0.08
                 0.2
                 0.6
                 0.8
            14
                                                                                                                      vs
                                                                                                                      vs
                                                                                                                      vs
                0.1 km from
                 contaminated site          vs
                 -  Inhalation only6

                1 km from
                 contaminated site          vs
                 -  Inhalation only6
                                           vs
                                           vs
                                                            14
                                                            vs
                                                                             vs
                                                                             vs
                                                                                          vs
                                                                                          vs
                                                                                                      vs
                                                                                                      vs
                aShort-term a 10-day Intake.
                ''Based on average weights of 10 and 70 kg for a child and an adult, respectively.
                Children ages 1-5, with pica (consuming 3 g soil/day).
                dCh11dren ages 1-5, without pica (consuming 0.6 g soil/day).
                elnhalatton rates are assumed to be 20 m3/da
 			Vday for the short-term and longer-term noncancer exposures;
 all other (more chronic) exposures assumed to be 10 nrvday as a result of 182 days exposure per year.
fys denotes no theoretical upper-bound limit.  Practical reasons require no free-flowing PCB liquid for the  limit.
^Soil-air partition coefficient = 2.35 x 10-s g soil/cm* air (= H.41/Kd = 5.73 x 10'" (41)/1000 =
 2.35 x lO'3).

-------
                                                TABLE C-14.  PERMISSIBLE PCB-1254 SOIL CONTAMINATION LEVELS9
                                                         (25-cm-THlCK CLEAN SOIL COVER. Kd - 40)
                                                                    Permissible levels (ug/g)  corresponding to
                Location and
               route of human
                 exposure
                                         Noncancer short-term9
                                       acceptable Intake ug/day"
                             100
                          for child
   700
for adult
	Cancer risk  specific  doses  (ug/day)	

   0.00175          0.0175          0.175       1.75
 (ID'7  risk)      (10-6  HS)C)   (JO-5  r1sk)  (i0-4 nsk)
o
i
On the contaminated site

 - Soil 1ngesttonc,         140
   Inhalation6

 - Soil 1ngest1ond,         vs
   Inhalation6

 - Inhalation onlye         vs
                                                          970
                                                          vs
                                                          vs
                    0.01
                                                                           0.02
                                                                           0.02
                  0.1
                                     0.2
                                     0.2
                               10
                                                                                                                     vs
                                                                                                                     vs
                                                                                                                     vs
               0.1 km from
                contaminated site          vs
                - Inhalation only6

               1 km from
                contaminated site          vs
                - Inhalation only6
                                           vs
                                           vs
                                                            10
                                                            vs
                                                                             vs
                                                                             vs
                                                                                          vs
                                                                                          vs
                                                                                                      vs
                                                                                                      vs
               aShort-term a 10-day Intake.
               bBased on average weights of 10 and 70 kg for a child and an adult, respectively.
               cCh1ldren ages 1-5. with pica (consuming 3 g soil/day).
               ^Children ages 1-5, without pica (consuming 0.6 g soil/day).
               elnhalatton rates are assumed to be 20 m^/day for the short-term and longer-term noncancer exposures;
                all other (more chronic) exposures assumed to be 10 mVday as a result of 182 days  exposure  per year.
               fvs denotes no theoretical upper-bound limit.  Practical  reasons require no free-flowing PCB  liquid  for  the  limit.
               ^Soil-air partition coefficient = 2.35 x lO'* g soil/cm*  air {= H.41/Kd = 5.73 x 10'*  (41)/1000 =
                2.35 x 10-s).

-------
                                 TABLE C-15.  PERMISSIBLE PCB
                                          (25-cm-THICK CLEAN S
                                   -126<
                                   SO|lL
                                                                                ;60  SOIL CONTAMINATION LEVELS?
                                                                                  COVER. Kd =  1000)
 Location and
route of human
  exposure
                                                     Permissible levels (ng/g) corresponding to
                          Noncancer short-term9
                        acceptable Intake gg/dayb
   100
for child
                                                           700
                                                        for adult
	Cancer risk  specific  doses  (gg/day)	

   0.00175         0.0175          0.175       1.75
 (10-7 risk)      (10-6  risk)   (10*5  risk)   (10"4  risk)
I
t—•
en
On the contaminated site

 - Soil Ingest1onc,         184
   Inhalation6

 - Soil 1ngest
-------
                                                 TABLE C-16.  PERMISSIBLE PCB-1260 SOIL CONTAMINATION LEVELS?
                                                          (25-cm-THICK CLEAN SOIL COVER. Kd = «)
                                                                     Permissible levels (pg/g) corresponding to
                 Location and
                route of human
                  exposure
  Noncancer short-term9
acceptable Intake pg/day°

     100           700
  for child     for adult
                                                           	Cancer risk  specific doses  (gg/day)	

                                                              0.00175         0.0175         0.175      1.75
                                                            (10-' risk)      (10-5 risk)  (10"5 risk)   (10"4  risk)
o
 i
On the contaminated site

 - Soil 1ngest1onc,
   inhalation6

 - Soil 1ngestiond,
   inhalation6

 - Inhalation only6
                                             110
                                            800
                   800


                   5000


                   vs
0.01
0.02
                                                                            0.02
0.1
0.2
                 0.2
1.0
1.0
             1.0
360
550
            620
                0.1 km from
                 contaminated site           vs
                 - Inhalation only6

                1 km from
                 contaminated site           vs
                 - Inhalation only6
                                           vs
                                           vs
                                                            1.0
                                                            vs
                                                     620
                                                                             vs
                                                                                          vs
                                                                                          vs
                                                                                                      vs
                                                                                                      vs
                aShort-term 2 10-day Intake.
                bBased on average weights of  10 and 70 kg for a child and an adult, respectively.
                cChildren ages 1-5, with pica  (consuming 3 g soil/day).
                ^Children ages 1-5, without pica  (consuming 0.6 g soil/day).
                6lnha1ation rates are assumed  to  be 20 iWday for the short-term and longer-term noncancer exposures;
                 all other (more chronic) exposures assumed to be 10 m^/day as a result of 182 days exposure per year.
                fvs denotes no theoretical upper-bound limit.  Practical reasons require no free-flowing PCB liquid for the limit.
                9soil-a1r partition coefficient = 2.35 x 10'* g so11/cm3 air (= H.41/Kd = 5.73 x lO'" (41)/1000 =
                 2.35 x 10-5).

-------
                  APPENDIX D
      HEALTH ADVISORIES FOR PCBs IN SOIL
                 Prepared by

              Michael  L. Dourson
 Environmental Criteria and Assessment Office
Office of Health and Environmental Assessment
     U.S. Environmental Protection Agency
               Cincinnati, Ohio

-------
                                   APPENDIX D

                                    CONTENTS


EXISTING GUIDELINES, RECOMMENDATIONS, AND STANDARDS 	   D-2

NONCARCINOuENIC EFFECTS 	   D-4

QUANTIFICATION OF SHORT-TERM HEALTH ADVISORY LEVELS 	   D-ll

CARCINOGENIC EFFECTS	   D-14

QUANTIFICATION OF CARCINOGENIC RISK 	   D-17

SPECIAL CONSIDERATIONS	   D-20

     High-Risk Subpopulation	   D-20
     Cocarcinogenesis, Initiation, and Promotion	   D-21

SUMMARY	   D-21

REFERENCES	   D-25
                                      D-l

-------
              EXISTING GUIDELINES,  RECOMMENDATIONS, AND  STANDARDS

     The manufacture,  sale,  and  distribution  of  PCBs  have  been  restricted
under Section 6(e)  of  the Toxic  Substances Control Act  (TSCA)  (P.L.  94-469).
PCBs were restricted to sealed systems  as of  1977, and manufacture  and  distri-
bution were banned  in  1979.   Rules  for  the disposal of PCBs  were proposed  in
1978 (43 FR 7150).
     The U.S. EPA (1980) has set ambient water quality criteria for PCBs  for
the protection of humans from increased risk  of  cancer over  a  lifetime  of  10~5,
10'6 and 10'7 at 0.79, 0.079, and 0.0079 ng/L.   As a  result  of  the  large  bio-
concentration factor in fish, these criteria  apply regardless  of whether  expo-
sure occurs through consumption  of  2 L  of water  and 6.5  g  of fish/day or
through consumption of fish  alone.   The Food  and Drug Administration (FDA) has
set temporary tolerances for PCBs in food and food related products, as shown
in Table 0-1.
     Occupational exposure limits for PCBs have  been  recommended by the American
Conference of Governmental Industrial Hygienists (ACGIH, 1980), and criteria
have been set for PCBs in workplace air by the National  Institute  for
Occupational Safety and Health (NIOSH,  1977). The time-weighted average  (TWA)
and short-term exposure limit (STEL) for Aroclor 1254,  are,  respectively,  0.5
and 1.0 mg/m3, and for Aroclor 1242, 1  and 2  mg/m3  (ACGIH, 1980).   The  NIOSH
(1977) criterion is 1.0 ug/m3 for a 10  hours/day, 40  hours/week exposure.
     The National Academy of Sciences developed  a 24-hour  suggested no  adverse
response level (SNARL) for PCBs  of  350  ug/L based on  the induction  of
mixed-function oxidase enzymes in the liver of rats administered Aroclor  at
doses of 1 to 2 mg/kg (NAS,  1980).   For this  analysis,  an  uncertainty factor  of
100 was used, since only enzyme  induction was reported  in  this dose range.

                                     D-2

-------
                      TABLE D-l.  FDA REGULATIONS FOR PCBs
                                                      Temporary tolerances
     Commodity                                                (ppm)
Milk (fat basis)                                               1.5
Manufactured dairy products (fat basis)                .        1.5
Poultry (fat basis)                                            3.0
Eggs                                                           0.3
Finished animal feeds                                          0.2
Animal  feed components of animal origin                        2.0
Edible portion of fish and shellfish                           5.0
Infant and junior foods                                        0.2
Paper food packaging material                                  10.0

SOURCE:  21 CFR 109.30.
                                      D-3

-------
                            NONCARCINOGENIC  EFFECTS

     Tests of the acute lethality  of  PCB  products  in  laboratory  animals, with
the exception of the guinea  pig, suggest  that,  in  general, PCB products have
similar toxicity regardless  of  route  of administration,  species,  or  age of
animal.  The single dose oral LQ$Q value  in  rats,  rabbits, mice,  and mink
ranged from 0.5-20 g/kg bw (Grant  and Phillips,  1974; Bruckner et  al., 1973;
Kimbrough et al., 1978; Fishbein,  1974; Garthoff et al.,  1981, Aulerich and
Ringer, 1977).  Route of administration also had little  effect (less than one
order of magnitude) on lethality,  with the  lethal  dose for dermal  administra-
tion in rabbits ranging from 0.7-3 g/kg bw  (Nelson et al., 1972),  while the
lethal dose in mice administered PCBs by  intraperitoneal  injection ranged from
0.8-1.2 g/kg bw (Lewin et al.,  1972).
     There are only two indications of major differences  in  the  acute  toxicity
of PCBs.  First, there is limited  evidence  that the guinea pig may be  more
sensitive to the lethality of PCBs than other species.   Miller (1944)  observed
a 100% mortality in a small  group  of  guinea  pigs receiving two oral  doses of
PCB (43% chlorine) at levels of 67 mg/animal at an interval  of 7 days  apart;
and McConnell and Kinney (1978) reported  an  1059.30 of 0.5 mg/kg for the  PCB
isomer 3,4,5-sym-hexachlorobiphenyl (HCB) in guinea pigs.  This  indication of
possible large interspecies  differences in  sensitivity is of concern in
species-to-species extrapolation when there is insufficient  data to indicate
which experimental animal most  accurately reflects the sensitivity of  humans.
The second problem concerns the possible  large difference in toxicity  of  speci-
fic isomers of PCBs.  There are indications, on the basis of limited data
available from Biocca et al. (1981),  that four different hexa-PCBs differ in
      value from 19 mg/kg to >64 mg/kg after oral administration  to mice.   It
                                      D-4

-------
would not be unreasonable to assume that even larger differences  will  be en-
countered as more isomers are tested.   Variation in toxicity of the different
isomers is not of great concern in defining acceptable environmental  levels,
since individual isomers were not commercially made and released  to the envi-
ronment.  The analytical methods used  to measure environmental  PCBs determine
the levels, in mg, of specific Aroclors.  Since, as described above,  the
Aroclors do not differ greatly in acute toxicity, using data from the most
toxic Aroclor should be protective without overly penalizing the  other
Aroclors.
     Some data are available on the nonlethal acute toxicity of PCBs  admini-
stered by the oral route for periods of 30 days or less.  The effects described
in these studies were alterations of the liver, thyroid, and reproductive
system.  Rosin and Martin (1983) reported that a dose of 500 mg/kg of Aroclor
1254 for 14 days to CD-I mice decreased pentobarbital sleeping time,  indicating
a substantial induction of microsomal  enzymes.  At lower doses, Sanders et  al.
(1974) reported that exposure of ICR mice to diets containing 250 to 62.5 ppm
of Aroclor 1254 for 14 days resulted,  respectively, in hepatomegaly and eleva-
ted serum corticosterone (the latter presumably as a result of altered liver
steroid metabolism).  These exposures  would result in doses of 32.8 and 8.1
mg/kg bw/day, assuming that a mouse consumes 13% of its body weight per day.
     Similarly, Narbonne (1979) reported decreases in phenobarbital sleeping
time in Sprague-Dawley rats maintained fro 8 days on a diet containing 100  ppm
of Phenoclor DPS (dose of 5 mg/kg bw/day, assuming that a rat consumes 5% of
its body weight/day).  PCB-induced increases in liver enzymes were suggested as
the reason for the increase in testicular acid phosphatase observed by Dikshith
et al. (1975) in Sprague-Dawley rats fed Aroclor 1254 at a dose of 50 mg/kg
bw/day for 7 days.'  Increases in liver-to-body weight ratio appear to be one of
                                      D-5

-------
the sensitive indicators  of  PCS  exposure.   Carter  and Mercer  (1983)  reported
that 64 mg/kg bw/day  of Aroclor  1254  caused increased liver weight,  while  Grant
and Phillips (1974) observed increased  liver weight  at  doses  as  low  as  12.5 and
5 mg/kg bw/day in  male and female  Wistar  rats,  respectively,  receiving  Aroclor
1254 in the diet  for  14 days.  Carter (1983) observed hepatomegaly in  rats in-
gesting diets containing  as  little as 20  ppm Aroclor 1254  (1  mg/kg bw/day) for
14 days.  Doses as low as 1  ppm  in diet of  3,4,5,3',4',5'-hexachlorobiphenyl
(345 HCB)  for 28  days caused liver microabscesses  and an increased liver weight
in 18-20 g 5-week-old C57B1/6J mice (Biocca et  al.,  1981).  Although adverse
this study, 0.3 ppm in diet  could  be  considered  a  lowest observed adverse
effects were observed at  concentrations as  low  as  0.3 ppm  in  diet in this
study, there is no documentation indicating that commercial Aroclors contain
345 HcB.  Hence,  this study  for  345 HcB cannot  be  used  for establishing the
short-term Health  Advisory (HA), corresponding  to  an acceptable  intake  (AI)
level.
     Besides changes  in the  liver, other  effects reported  for exposure  to  low
levels of  PCBs were increased thyroid activity  in  Sherman  rats maintained  on
diets containing  250  ppm  of  Aroclor 1254  (12.5  mg/kg bw) for  14  days;  in
Sprague-Oawley rats,  administration of  Aroclor  1254  by  gavage for 21 days  at
a dose of  0.05 g/kg bw/day resulted in  weight loss and  decreased body tempera-
ture (Komives, 1979;  Komives and Alayoku, 1980).  Enlarged thyroid has  also
been found in Osborne-Mendel rats  maintained on  diets containing 5 ppm  of
Aroclor 1254 (0.25 mg/kg  bw/day) for  4  weeks (Collins and  Capen, 1980b).   This
exposure level also resulted in  increased liver enzymes in Holtzman  rats
(Garthoff  et al.,  1977).
     The toxicity of  PCBs resulting from  exposures of between 30 and 90 days
has been more extensively studied. Alterations  in liver histopathology
                                     D-6

-------
occurred at doses as low as 5 ppm in  diet  for 5 weeks  in Holtzman rats  (Kasza
et al., 1978b).  In the mouse (MNRI), a dose of Clophen A-60,  as  low as  0.025
mg/mouse (0.8 mg/kg bw/day, assuming  a mouse weight  of 0.03 kg)  for 62 days,
increased the estrous cycle, probably as a result  of PCB-induced  changes in
liver steroid metabolism (Orberg  and  Kihlstrom, 1973).  At  higher dietary con-
centrations of 167 ppm (22 mg/kg  bw)  for 6 weeks,  Aroclor 1016 and 1242  de-
creased the immunologic capabilities  of BALB/CJ mice (Loose et al., 1978).
     Species other than rats and  mice have been tested to a lesser extent for
this duration.  Rabbits exposed to diets containing  3  ppm of Aroclor 1254 (0.15
mg/kg bw/day, assuming that a rabbit  consumes 4.9% of  its body weight per day)
for 8 weeks developed hepatomegaly and immunosuppression.  In  the guinea pig,
Vos and van Genderen (1973) reported  that  diets containing  250 ppm of Clophen
A-60 (7 mg/kg bw/day, assuming that a guinea pig consumes 2.8% of its body
weight per day) for 4-7 weeks was lethal,  while diets  containing  50 ppm  of
Clophen A-60 or Aroclor 1260 (1.4 mg/kg bw) for 4  to 7 weeks produced immuno-
suppression.  Allen et al. (1974) and Allen (1975) observed comedones and
facial edema in rhesus monkeys ingesting diets containing 25 of ppm Aroclor
(1.1 mg/kg bw, assuming that a monkey consumes 4.2%  of its  body weight per day)
for 2 months.
     Studies of subchronic and chronic exposure (i.e., ^90  days)  to PCBs have
failed to use sufficiently low doses  to define a no  observed adverse effect
level (NOAEL) in rats.  In Sprague-Dawley  rats, Allen  et al. (1976) and  Allen
and Abrahamson (1979) reported that a 52-week exposure to diets containing 100
ppm of Aroclor 1248, 1254, or 1262 (5 mg/kg bw/day)  followed by a 13-week
observation period resulted in hepatomegaly and liver  necrosis.   At a lower
exposure of 75 ppm in the diet (3.75  mg/kg bw/day) for 36 weeks,  Sprague-Dawley
rats developed focal necrosis (Jonsson et  al., 1981).   Elevated liver porphy-
                                      D-7

-------
 rine levels were detected by both Kimbrough et al.  (1972) and Zinkl  (1977)
 after exposure of Sherman rats for 8 months to 20 ppm of Aroclor 1254 or 1260
 (1 mg/kg bw/day) and of CD rats for 20 weeks to 10  ppm of Aroclor 1254 (0.4
 mg/kg bw/day).  In a study employing near-lifetime  exposure (2 years), Morgan
 et al.  (1981) reported an increase in mortality (33% as compared with 8% in
 controls) in Fischer 344 rats at the lowest dose tested (25 ppm; 1.25 mg/kg
 bw/day).  The subchronic studies demonstrated increasing liver pathology over
 the dose ranges studied, 0.5-5 mg/kg bw/day; while  in the only chronic study,
 the lowest dose tested (1.25 mg/kg bw/day) resulted in early deaths.
     In mice, dietary exposure levels to Kanechlor-300, -400 or -500 or Arochlor
 1254 of between 100 and 500 ppm (13-65 mg/kg bw/day) for periods from 23 weeks
 to 11 months produced hepatomegaly (Ito et al., 1973; Bell, 1983; Kimbrough and
•Under, 1974).  Koller (1977) used groups of BALB/CJ mice which were maintained
 for 9 months on diets containing 0, 3.75, 37.5, or 375 ppm of the Aroclors 121,
 1242, or 1254 (0.46, 4.57, or 47.75 mg/kg bw/day).   The Aroclor with the lowest
 chlorine content (1221) produced no liver lesions,  while exposure to Aroclor
 1242 resulted in increased liver weight in the high-dose group.  In mice ex-
 posed to Aroclor 1254, increased mortality was observed in the high-dose group,
 mild hepatopathology was observed in the median dose group, and no liver lesions
 were detected in the low-dose group.  The no observed effect level (NOEL) in
 this study in mice of 0.45 mg/kg bw/day is nearly identical to the lowest ob-
 served  effect level (LOEL) of 0.5 mg/kg bw/ day associated with porphyria in
 rats (Kimbrough et al., 1972; Zinkl, 1977).
     The only other species tested in chronic bioassays was the monkey, and it
 proved  to be highly sensitive to the toxic effects of PCBs.  The most common
 observations in monkeys exposed to Aroclor 1248 in the diet for a period of
 from 8  to 39 months were skin lesions, palpebral edema, and erythema  (Barsotti
                                      D-8

-------
and Allen, 1975; Allen and Barsotti,  1976;  Allen et  al.,  1980;  Becker  et  al.,
1979).  These effects were observed  at the  lowest doses tested, ranging from
2.5 to 3 ppm in the diet (0.095-0.126 mg/kg bw/day).  In  addition,  Becker et
al. (1979) reported that monkeys fed  diets  containing 3 ppm of  PCBs had gastric
lesions, body weight loss, and reduced hemoglobin and leukocytes.   In  the
monkey, doses as low as 0.1 mg/kg bw/day produced frank toxic effects; no
studies have been conducted from which a NOAEL can be derived or to indicate
how close 0.1 mg/kg bw/day is to the  NOAEL  for monkeys.
     Although PCBs have not been demonstrated to be  animal  teratogens  following
oral exposure, these compounds have  been demonstrated to  adversely  affect
reproduction.  When adminstered to pregnant Wistar rats at  a dose of 100  mg/kg
bw/day on days 6 to 15 of gestation,  Villeneuve et al. (1971) observed no
adverse effects; however, using the  same treatment schedule, Spencer (1982)
reported that Sprague-Dawley rats were infertile after receiving 15 mg/kg
bw/day, that animals receiving 5 mg/kg bw/day had reduced litter weights, and
that 2.5 mg/kg bw/day was the NOEL.   Rabbits had resorptions, abortions,  and
fetuses at similar dose levels of 12.5 mg/kg bw/day  administered on days  0
to 28 of gestation; however, slightly smaller doses  of 10 mg/kg bw/day were
reported to be the NOEL.  The Hartly  guinea pig, which has  been shown  to  have
greater sensitivity to the toxicity  of PCBs than most other species, had
macerated fetuses after receiving 2.2 mg/day (6/5 mg/kg bw/day) of  Clophen A-50
on days 10 to 60 of gestation (Brunstroem et al., 1982).
     Effective doses of PCBs were lower than exposure occurred  before  and dur-
ing gestation.  In a two-generation  study,  Sherman rats maintained  on  diets
containing 20 ppm Aroclor 1254 (1 mg/kg bw/day) had  reduced litter  size,  and at
100 ppm (5 mg/kg bw/day) the pups that were born died during nursing (Linder et
al., 1974).  In this study, 5 ppm (0.25 mg/kg bw/day) was the NOEL.  Complete

                                      D-9

-------
loss of fertility was observed in male and  female Wistar rats  caged together
for 9 weeks while ingesting 6.4 mg/kg bw/day  of Aroclor 1254 emulsified in
their drinking water (Baker et al.,  1977).  Males regained normal  fertility
after removal  from treatment for 2 weeks.   When Aroclor 1254 was  administered
to lactating Holtzman rats at 32 mg/kg bw/day on days  3, 5 and 7  of lactation,
the future mating behavior of nursing male  pups was adversely  affected (Sager,
1983).  A lower dose of 8 mg/kg bw/day was  a  NOEL.
     Of the species tested, the mink and the  monkey are the most  sensitive to
the reproductive toxicity of PCBs.  Bleavins  et al. (1980) maintained mink on
diets containing 5 ppm Aroclor 1242  or 20 ppm Aroclor  1016 (doses of 0.75 and
3 mg/kg bw/day, assuming that a mink consumes 15% of its body  weight per day)
for 18 months and observed complete  reproductive failure in the Aroclor 1242
group and 25% mortality and infertility in  the Aroclor 1016 group.  A more
recent study by Aulerich et al. (1985) tested yet lower doses  fed to mink via
diet.  Aroclor 1254 at 2.5 ppm; 3,4,5,3',4',5'-hexachlorobiphenyl  (345 HCB) at
0.1 or 0.5 ppm; 2,4,5,2',4',5'-hexachlorobiphenyl (245 HCB) at 2.5 or 5.0 ppm;
or 2,3,6,2',3',6l-hexachlorobiphenyl (236 HCB) at 2.5  or 5.0 ppm  of diet were
fed to groups of 10 standard dark mink (proven breeders).  A group of 20 ani-
mals served as controls.  All of the mink fed 0.5 ppm  345 HCB  died within 60
days, while those fed 0.1 ppm showed 50% mortality after 3 months.  One still-
born kit was whelped in the Aroclor  1254 group.  245 HCB and 236  HCB did not
affect reproductive performance at either dose.
     In a limited study (8 animals/group),  Allen et al. (1980) maintained
rhesus monkeys on diets containing 2.5 or 5 ppm (0.1 or 0.2 mg/kg bw/day) of
Aroclor 1248 for 18 months.  In the  low-dose  group, increased abortions were
observed, while in the high-dose group, the mothers showed overt  signs of
toxicity and no live births occurred.  After  removal from exposure for 1 year,

                                      D-10

-------
fertility had still  not returned to normal,  and  some  pups  died  during nursing.
It is apparent that  frank effects in reproduction  were  observed in  mink  at
lower doses than in  monkeys and still  lower  than the  NOEL  in  rats,  rabbits,  and
guinea pigs following repeated exposure to PCBs.

              QUANTIFICATION OF SHORT-TERM HEALTH  ADVISORY LEVELS

     PCBs belong to  a class of chemically stable,  multi-use industrial  chemicals
that have been widely distributed in the ecosystem.   Technical  preparations  con-
sist of complex mixtures of discrete PCB isotners.  Because of. their physicochem-
ical properties, PCBs have been used as heat exchangers, dielectric, hydraulic
and lubricating fluids, plasticizers,  pesticide  extenders, adhesives, printing
inks and surface coatings.
     The physical  and chemical properties and the  chemical formulations  of  PCBs
vary considerably, depending on the amount and position of chlorine substitu-
tion.  Such properties as stability, volatility, and  water solubility are par-
ticularly important  in regard to frequency of occurrence in the environment.
The higher chlorinated biphenyls are less volatile than the lower chlorinated
biphenyls (Mieure et al., 1976).  PCBs are extremely  insoluble  in water.  The
solubility of commercial mixtures (for example,  the Aroclors) ranges from 25 to
200 ppb (25 to 200 ug/L), depending on the chlorine content (Nisbet and  Saro-
fim, 1972; Haque et  al., 1974).  The solubility  of discrete PCB isomers  has
been examined, and ranges from 1 to 600 ppb  (1 to  600 wg/l.) depending on the
degree of chlorine substitution in the biphenyl  ring  (Haque and Schmeddig,
1975).
     PCBs elicit a variety of adverse  health effects.  This is  true of even
partially well-defi.ned compositions such as  Aroclor 1254.   For  example,  toxi-

                                      D-ll

-------
cology studies on Arqclor 1254 of less  less  than 30 days'  duration  report
liver toxicity (Grant and Phillips,  1974;  Carter, 1983),  thyroid  toxicity
(Collins and Capen, 1980a, b,  c)  and reproductive toxicity (Villeneuve  et  al.,
1971), as well as other types  of  toxicity  (U.S.  EPA, 1985b).   At  first, it
would seem that this variety in elicited adverse effects  would make it  diffi-
cult to distinguish the critical  toxic  effects of PCBs.   However, it appears
that the experimental thresholds  for these effects may be similar,  at least for
studies of 30 days' duration or less.
     Villeneuve et al. (1971)  found  increased incidences  of fetal death, re-
sorptions, and abortions at 12.5  mg/kg/day of Aroclor 1254 in rabbits when
exposed on days 1 through 28 of pregnancy.  A dose of 1.0 mg/kg/day appeared
to be without effect.  Collins and Capen (1980a, b, c),  in a series of studies
on thyroid effecs in rats, determined that 50 ppm of diet (~ 2.5  to 5.0
mg/kg/day) for 4 weeks was associated with clearly defined adverse effects, but
that doses of 5 ppm of diet (~ 0.25  to  0.5 mg/kg/day) were not.  Carter
(1983) demonstrated liver hepatomegaly  in rats at doses  of 20 ppm Aroclor 1254
of diet (~ 2 mg/kg/day) for 14 days; such an effect in the absence of toxicity
(e.g., fatty infiltration of the  liver) might not be considered adverse.  Grant
and Phillips (1974) observed increased  liver weights at  doses as  low as 5 mg/
kg/day Aroclor 1254 given in corn oil for 7 consecutive days.  Collectively,
these studies indicate that the experimental threshold for adverse effects of
Aroclor 1254 in studies of 30 days'  duration or less is  at or near a dose of  1
mg/kg/day.  Thus, it seems reasonable to use this latter dose as  a basis for
health risk assessments for Aroclor 1254 for short durations.
     Utilizing a dose of 1 mg/kg/day weight as a No Adverse Effect dose, a 10-
day exposure level for PCBs may be calculated as follows:
                                      D-12

-------
      10-day exposure level  = 1 mg/kg/day x 10 kg = 0.1 rig/day for a 10-kg  child.
                                     100
where
     10 kg = assumed body weight of a child,  and
       100 = uncertainty (safety) factor chosen in accordance with the
             National Academy of Sciences guidelines,  in which a NOAEL
             from an animal  study is employed.

For a 70-kg adult the 10-day exposure level would be 0.7 mg/day.

                              CARCINOGENIC EFFECTS

     A number of short-term assays predictive of carcinogenic potential  have
been performed using the Aroclors and individual  isomers of PCBs.  Negative
results have been obtained in the reverse mutation assay using S^. typhimurium
(Schoeny et al., 1979; Schoeny, 1982; Meddle and Bruce, 1977), and in the
dominant lethal assay in rats (Green et al.,  1975). PCS products also did  not
produce chromosomal  changes  in ]). melanogaster (Nilsson and Ramel, 1974) or in
the sperm and bone marrow cells of rats (Green et al.,  1975; Garthoff et al.,
1977; Dikshith et al., 1975).  Wyndham et al. (1976),  however, observed
increases in reversion frequency in S.. typhimurium exposed to 4-chlorobiphenyl
and to a lesser extent with  Aroclor 1221, while the more highly chlorinated
2,2',5,5'-tetrachlorobiphenyl and Aroclor 1254 were negative.  The positive
response was observed in one strain, TA1538,  in the presence of a metabolic
activation system derived from rabbit liver.   In addition, a weak positive
response was observed by Peakall  et al. (1972) in an assay of chromosomal
aberration in the embryos of ring doves fed Aroclor 1254.  The variable data
observed with PCBs is consistent with the poor response and lack of correlation
                                      D-13

-------
with animal carcinogenicity  data  reported  for  many  highly  chlorinated  compounds
in short-term assay.
     Early bioassays  of PCBs were inadequate as  a  result of  small  group  size
or periods of exposure extending  for less  than 1 year.  These  studies  failed  to
demonstrate that PCBs were carcinogenic  when fed to rats or  mice  (Kimura and
Baba, 1973; Ito et al., 1973; Ito et al.,  1974;  Kimbrough  and  Under,  1974).
The study by Kimbrough and Linden (1974) in which  50 BALB/CJ male  mice were fed
diets containing 300  ppm.of  Aroclor 1254 for 6 or  11 months  was suggestive that
Aroclor 1254 was a liver carcinogen.  Of the 22  animals surviving  PCB  treatment
for 11 months, all had areas of adenofibrosis  in the liver,  and seven  had his-
tologically identified hepatomas.  In animals  surviving 11 months  which  are
maintained on PCB-contaminated diets for 6 months,  and  in  control  animals,
there were, respectively, only 0/24 and  0/58 livers with areas of  adenofibrosis
and 1/24 and 0/58 animals with hepatomas.  There were no histologically  identi-
fied hepatocellular carcinomas in.any group.
     In a later study in rats using a larger group  size and  a  longer exposure
period, Kimbrough et  al. (1975) reported an increased incidence in hepato-
cellular carcinomas in animals exposed to  PCBs.   In this study, 200 female
Sherman rats were exposed to diets containing  a  nominal 100  ppm  (range 70-107
ppm) of Aroclor 1260  for 630 days.  The  incidence  of hepatocellular carcinomas
in the treated group  compared with control animals  was  26/184  and  1/173,
respectively; while the incidence of hepatic neoplastic nodules was 144/184
and 0/173, respectively.
     In the only other chronic bioassay  performed  (NCI, 1978), 24  male and 24
female Fischer 344 rats/group were maintained  on diets  containing  0, 25, 50,  or
100 ppm of Aroclor 1254.  Although dose-related  increases  in nodular hyper-
plasia were observed, there  was no significant increase in neoplastic  lesions.
                                      D-14

-------
With respect to the tumor incidence reported by Kimbrough  et  al.  (1975),  the
number of rats/group in the NCI  study may have been too small  to  detect  a
carcinogenic response.
     Very limited information is available on the carcinogen!city of PCBs in
humans.  In a survey of 1,200 patients in Yusho, Japan, 5.5 years after  expo-
sure to PCBs, 41% of the 22 deaths were attributed to neoplasia  (Kuratsune,
1976; Urabe, 1974).  The relevance of these findings is unclear  since the
tumors were at various common sites, and comparable incidences for an unexposed
population were not presented.  In two letters to the editor,  Bahn et al.
(1976, 1977) described suspected increases in malignant melanomas in a small
group of workers exposed to Aroclor 1254.  In the 31 "heavily  exposed" workers,
there were two cases of melanoma, while in the 41 "less heavily"  exposed
workers, there was one melanoma.  The International Agency for Research  on
Cancer (IARC) estimates that only 0.04 cases would be expected in this number
of individuals (IARC, 1978).  The only epidemiologic study (Davidorf and  Knupp,
1979) examined the association between ocular melanomas and populations  resid-
ing in areas -of known high environmental levels of PCBs.  The  authors concluded
that a causal relationship between PCBs and ocular melanomas was  not demon-
strated.  The IARC (1978) considered the association between  exposure to
Aroclor 1254 and malignant melanomas described in the two  letters to the  editor
of the New England Journal of Medicine (Bahn et al., 1976, 1977)  to be sugges-
tive evidence that PCBs are human carcinogens.
                                      D-15

-------
                      QUANTIFICATION OF  CARCINOGENIC  RISK

WEIGHT OF EVIDENCE FOR HUMAN CARCINOGENICITY
     The likelihood that  a chemical such as PCB  Is  a  human  carcinogen  Is  ex-
pressed through a characterization  or  stratification  of  the "weight  of evi-
dence" (human, animal, short-term test)  and a  final indiction  of  the "overall
weight of evidence" for human carcinogenicity.
     The IARC has characterized  the evidence  for the  carcinogenicity of PCBs  in
humans as "inadequate," the evidence for carcinogenicity in animals  as "suffi-
cient," and the supportive evidence from short-term tests as "inadequate,"
(IARC, 1982).  The overall weight-of-evidence  designation for  PCBs under the
IARC scheme is 2B (probably carcinogenic in humans; evidence inadequate in
humans and sufficient in  animals).  The  EPA has  recently proposed a  similar
scheme, which is an adaptation of the  IARC scheme (U.S.  EPA, 1984a).  There are
some differences in the two schemes; however,  for PCBs the  requirements for
inadequate evidence of carcinogenicity in humans and  for sufficient  evidence  in
animals are essentially identical.  Similarly, the overall  weight-of-evidence
under the new EPA scheme  would be designated  as  B2, which has  the same require-
ments as the 2B designation of the  IARC  scheme.
     To comply with the EPA proposed guidelines  for carcinogen risk  assessment
(U.S. EPA, 1984a), any final risk estimate developed  or  used in risk charac-
terization should be coupled with the  EPA classification of the qualitative
weight-of-evidence.  Thus, those risk  levels  used in  this paper are  understood
to carry the "B2" designation; for  example,  1  x  10~5  (B2),  1 x 10'6  (B2), etc.
POTENCY SLOPE FACTORS
     The use of a potency slope  factor,  necessary in  calculating risk or back-
calculating permissible PCB soil concentrations  from  selected risk  levels,  does
                                      D-16

-------
not definitely indicate that  the  chemical  is a human carcinogen.   The  likeli-
hood that the agent is a human  carcinogen  is a function of the weight-of-
evidence described above.   The  proposed  EPA guidelines for carcinogen  risk
assessment (U.S. EPA,  1984a)  suggest  that  agents  falling  into Groups A and  B
are suitable for quantitative risk  assessments.
     The U.S. EPA Carcinogen  Assessment  Group  (CAG) has used the  data  from
female rats in the study by Kimbrough et al. (1975) to quantify the carcino-
genic risk from exposure to PCBs  (U.S. EPA, 1980).  In this analysis,  the TWA
concentration of PCB (Aroclor 1260) in the diet was determined to be 88.4 ppm,
associated with a daily dose  of 4.42  mg/kg bw/day, by assuming that an adult
rat consumes food equal to 5% of  its  body  weight  per day.  In addition,  for
this analysis, the incidence  of hepatocellular carcinomas  (26/184 in treated
animals and 1/173 in controls)  and  neoplastic  nodules (144/184 in treated
animals and 0/173 in controls)  were combined to produce tumor incidences of
170/184 and 1/173 in the treated  and  control groups, respectively. Using these
data and the linearized multistage  model,  a cancer potency value  (q^)  for
human exposure to Aroclor 1260  of 4.34 (mg/kg/day)'1 was  calculated using the
data in Table 0-2.  The U.S.  EPA  Office  of Toxic  Substances  (OTS) has  also  used
the data from the same study, but by  altering  two of the  variables (see Table
D-2), calculated a qj of 3.57 (mg/kg/day)'1 (U.S. EPA, 1985a).  The average
of these two values is 4.0 (mg/kg/day)'1,  and  this value  has been adopted by
this health advisories appendix for use  in developing advisory levels  for PCBs
clean-up.
     Using this q^, a risk-specific dose  (RSD) of Aroclor 1260 that would
result in an increased lifetime risk  level of  10~5 for a  70-kg man can be
calculated as follows:
                                      D-17

-------
                 TABLE D-2.  DATA USED AS THE  BASIS FOR THE
Species
Strain
Sex
Body weight (assumed)
Length of exposure
Length of experiment
Tumor site
Tumor type

PCB product tested
        Dose
     (mg/kg/day)
          0
         4.42
rat
Sherman
female
0.4 kg (0.35 kg)a
645 days (730 days)3
730 days
liver
combined hepatocellular carcinomas
and neoplastic nodules
Aroclor 1260
         Incidence
(No. responding/No, tested)
            1/173
          170/184
aThe data in parentheses indicate the alternative values used by OTS in
 computing a cancer potency factor.
SOURCE:  Kimbrough et al., 1975.
                                      D-18

-------
                   RSD = 1 x 10'5 x 70 kg bw *  qj (mg/kg/day)'1
                       = 0.175 yg/day.

Thus, using the q^ for humans of 4.0 (mg/kg/day)"1,  the  RSOs  corresponding to
lifetime risks of 10'4, 10'5, and 10'6 are 1.75,  0.175,  and 0.0175 wg/day,
respectively.  The adoption of these potency and  concentration levels  for all
PCBs requires a further assumption that all  PCB compounds  are carcinogenic and
that the potency of Aroclor 1260 is representative of  any  mixture of any other
PCB compound.

                             SPECIAL CONSIDERATIONS

HIGH-RISK SUBPOPULATION
     Two separate groups of high-risk subpopulations for exposure to PCBs may
be identified.  The first group includes those  persons with the  potential for
frequent or high exposure, namely, occupationally-exposed  workers and  breast-
fed infants, as PCBs are excreted in the breast milk of  lactating humans
(Miller, 1977; Rogan et al., 1980; Wickizer  et  al.,  1981;  Mes and Davies, 1979;
Kuwabara et al., 1979; Hofvander et al., 1981).  The second group includes
those persons with an inability to oxidize PCBs via  glucuronidation to facili-
tate detoxification and elimination of these toxicants,  such  as  embryos,
fetuses, and neonates (2 to 3 months old) (Calabrese and Sorenson, 1977;
Gillette, 1967; Nyhan, 1961), especially breast-fed  infants who  receive a
steroid via human breast milk that inhibits  glucuronyl transferase activity
(Calabrese and Sorenson, 1977; Gartner and Arias,  1966), children simultane-
ously exposed to the antibiotic novobiocin (Lokietz  et al., 1963; Calabrese
and Sorenson, 1977), persons with Gilbert's  syndrome or  Crigler  and Najjar
                                      D-19

-------
syndrome (Lester and Schmid,  1964;  Calabrese  and  Sorenson,  1977),  or persons
with hepatic infections such  as  infectious  hepatitis  (Calabrese and Sorenson,
1977).
COCARCINOGENESIS, INITIATION,  AND PROMOTION
     OiGiovanni  et al.  (1978)  demonstrated  that Aroclor  1254 had weak tumor
initiating activity in  the mouse two-stage  tumorigenesis models.  Promoting
activity was not indicated for this Aroclor in a  study by Berry et al. (1978).
Using other experimental  systems, Ito  et  al.  (1973) observed an increased inci-
dence of liver tumors in  rats  co-administered PCBs and benzene hexachloride as
compared with benzene hexachloride  treatment  alone.   Co-administration of other
potent live carcinogens,  3'-methyl-4-dimethyl aminoazobenzene, N-2-fluorenylace-
temide and diethylm'trosamine, with PCBs, however, has resulted in the inhibi-
tion of the tumorigenic response (Ito  et  al., 1973).  Similarly, antineoplastic
effects were reported by  Nishizumi  (1980) for pups of dams  administered
Kanechlor-500 and diethylnitrosamine.   These  studies  make it apparent that expo-
sure to PCBs can affect the carcinogen!city of other  xenobiotics.

                                    SUMMARY

     For the purposes of  setting advisory levels  for  PCBs contaminating soils,
acceptable intake (Al)  levels  have  been developed which  are based  on both the
toxicity and the carcinogenicity of PCBs.  The 10-day health advisories (HA)
for toxicity other than cancer have been  developed.   The 1-day and lifetime HAs
could not be evaluated.  Advisories for the cancer end point are expressed in
terms of 10~4 to 10"7 lifetime individual excess  risk levels.  See Table 0-3 for
a summary of these AIs and risk-specific  doses.
                                      D-20

-------
  TABLE U-3.  SUMMARY UF RISK SPECIFIC, UUSES (RSDs) FOR CANCER
     RISKS UF PCBs, OR OF ACCEPTABLE INTAKES FOR PROTECTION
          AGAINST THE NONCARCINOGEMC EFFECTS OF PCBs
Description
Value (py/day)
11T4 RSU
11T5 RSU
11T6 RSU
Short-term AI (1-day)
Longer-term AI (10-day)

Lifetime AI
    1.75
    U.175
    U.0175
    Not estimated
    10U for a child
    7UO for an adult
    Not estimated
                            U-21

-------
     The toxicity and carcinogenicity of PCBs  have  recently been critically
evaluated in a number of EPA documents (U.S.  EPA,  1980,  1983,  1984b, 1985b).
The approach taken here in developing a 10-day exposure  advisory for non-
cancer toxicity has been to select from the available literature the animal
study that addresses the critical  toxic effect and  yields the  most appropriate
no observed effect, no observed adverse effect, lowest observed effect, or
lowest observed adverse effect level.  This dose was  then divided by an appro-
priate uncertainty factor to obtain the 10-day HA.
     The calculation of a 10-day HA for noncarcinogenic  toxicity should make
use of animal data derived after an exposure  period ranging from 10 to 30 days.
The literature contains several animal studies which  address this length of
exposure.  The studies chosen as a basis for  the 10-day  HA (Villeneuve et al.,
1971; Grant and Phillips, 1974; Collins and Capen,  1980a, b, c; Carter, 1983)
yield a No Observed Adverse Effect Level (NOAEL).   These studies involved the
feeding of Aroclor 1254 to rabbits and rats.   Collectively, in these studies, a
NOAEL can be ascribed to a dose of 1 mg/kg/day).  Dividing this NOAEL by an
uncertainty factor of 100 yields a 10-day HA  of 100 wg/kg/day  for a child,
and 700 ug/day for a 70-kg adult.
     The 1-day HA could not be estimated based on  animal data  derived from
studies with an exposure duration  of 1 to 5 days.   The lifetime HA noncancer
                                                  %
toxicity, likewise, could not be estimated.
     The cancer risk specific doses (intake levels) have been  calculated by
solving for exposure in the equation

                           Risk =  Potency x Exposure

and multiplying by 70 kg.  A cancer potency factor  of 4.0 (mg/kg/day)~1 was
                                      D-22

-------
adopted for use in these calculations.   This  value is  the mean of the values



determined by OTS (3.57 (mg/kg/day)"1)  and by ORD  (4.34  (mg/kg/day)-1) for



PCBs.
                                   0-23

-------
                            REFERENCES  (APPENDIX  D)
Allen, J.R.  (1975)   Response of the non-human  primate to polychlorinated
     biphenyls exposure.   Fed. Proc. 34:1675-1679.

Allen, J.R.; Abrahamson,  L.J.  (1979)   Responses  of  rats exposed to  poly-
     chlorinated biphenyls for 52 weeks.   II.   Compositional and enzymic
     changes in the  liver.  Arch. Environ.  Contain. Toxicol. 8:191.

Allen, J.R.; Barsotti,  O.A.   (1976)   The  effects  of  transplacental and mam-
     mary movement of the PCBs on infant  rhesus monkeys.  Toxicology
     6:331-340.

Allen, J.R.; Barsotti,  D.A.; Carstens,  L.A.   (1980)   Residual effects of
     polychlorinated biphenyls on adult nonhuman  primates and their  off-
     spring.  J. Toxicol. Environ. Health 6(l):55-66.

Allen, J.R.; Carsten, L.A.;  Abrahamson, L.J.   (1976)  Responses of rats
     exposed to polychlorinated biphenyls for  52  weeks.  I.  Comparison of
     tissue levels of PCB and biological  changes.  Arch. Environ. Contarn.
     Toxicol. 4:409.

Allen, J.R.; Norback, D.H.;  Hsu, I.C.   (1974)   Tissue modifications  in mon-
     keys as related to absorption,  distribution, and excretion of poly-
     chlorinated biphenyls.   Arch. Environ. Contain.  Toxicol. 2:86.

American Conference  of  Governmental  Industrial  Hygienists (ACGIH).   (1980)
     Documentation of the threshold  limit values.  4th ed.   (Includes sup-
     plemental documentation, 1981). Cincinnati, OH, p. 486.

Aulerich, R.J.; Bursian,  S.J.; Breslin, W.J.;  Olson,  B.A.; Ringer, R.K.
     (1985)  Toxicological manifestations of  2,4,5-2',4',5'-, 2,3,6,2',
     3',6'-, and 3,4,5,3',4',5'-hexachlorobiphenyl and Aroclor 1254  in mink.
     J. Toxicol. Environ. Health 15:63-79.

Aulerich, R.J.; Ringer, R.K.  (1977) Current  status  of  PCB toxicity to mink,
     and effect on their reproduction.  Arch.  Environ. Contam. Toxicol. 6:279-
     292.

Bahn, A.K.; Grover,  P.; Rosenwaike,  I.; O'Leary,  K.;  Stellman, J.   (1977)
     PCBs and melanoma.  N.  Gig. J.  Med.  296:108.

Bahn, A.K.; Rosenwaike, I.;  Herrmann, N.; Grover, P.; Stellman, J.;  O'Leary,
     K.  (1976)  Melanoma after exposure  to PCBs.  New Engl. J. Med. 295:450.

Baker, F.D.; Bush, B.;  Tumasonis, C.F.; Lo, F.C.   (1977)  Toxicity and per-
     sistence of low-level PCB in adult Wistar rats,  fetuses, and young.
     Arch. Environ.  Contam.  Toxicol. 5(2):143-156.

Barsotti, D.A.; Allen,  J.R.   (1975)   Effects  of polychlorinated biphenyls on
     reproduction in the primate. Fed. Proc.  34:338.


                                     D-24

-------
Becker, G.M.; McNulty,  W.P.;  Bell,  M.   (1979)   Polychlorinated  biphenyls-
     induced morphologic changes  in the gastric mucosa  of the rhesus  monkey.
     Lab. Invest.  40:373.

Bell, M.  (1983)  Ultrastructural  features  of  the  murine cutaneous  microvas-
     culature after exposure  to polychlorinated biphenyl compounds  (PCBs)  and
     benzo(a)pyrene (BAP).   Virchows Arch.  B.  42(2):131-142.

Berry, D.L.; Giovanni,  J.D.;  Juchau, M.R.;  Bracken, W.M.; Gleason,  G.L.;
     Slaga,  T.J.  (1978)  Lack  of  tumor-promoting  ability of certain  environ-
     mental  chemicals in a  two-stage mouse  skin tumorigenesis assay.   Res.
     Commun. Chem. Pathol.  Pharmacol.  20:101.

Biocca, M.;  Gupta, B.N.; Chae,  K.;  McKinney, J.D.; Moore, J.A.   (1981)  Toxi-
     city of selected symmetrical  hexachlorobiphenyl  isomers in the mouse.
     Toxicol. Appl. Pharmacol.  58:461-474.

Bleavins, M.R.; Aulerich, R.J.; Ringer, R.K.   (1980)  Polychlorinated biphe-
     nyl s (Aroclors 1016 and  1242): effects on survival and  reproduction in
     mink and ferrets.   Arch. Environ.  Contam. Toxicol. 9(5):627-635.

Bruckner, J.V.; Khanna, K.L.; Cornish,  H.H.  (1973)   Biological  responses  of
     the rat to polychlorinated biphenyl s.  Toxicol.  Appl., Pharmacol. 24:434-
     448.

Brunstroem,  B.; Kihlstroem, I.; Lundkvist,  U.   (1982)  Studies  of fetal death
     and fetal weight in guinea pigs fed polychlorinated biphenyls  (PCBs).
     Acta. Pharmacol. Toxicol.  50(2):100-103.

Calabrese, E.J.; Sorenson,  A.J.  (1977)  The health effects  of  PCBs with par-
     ticular emphasis on human  high-risk groups.  Rev.  Environ. Health 2(4):
     285-304.

Carter, J.W.  (1983)  Onset of  hepatomegaly in PCB (Aroclor  1254)-treated
     rats.  Bull.  Environ.  Contam.  Toxicol. 31(2):183-187..

Carter, J.W.; Mercer, L.P.   (1983)   Prediction of  food  intake,  weight gain,
     and organ weights  in rats  as  a formulation of dietary Aroclor  1254
     (PCBs)  by the saturation kinetics  model for physiological  responses.
     Nutr. Rep. Int. 27:561-568.

Collins, W.T.; Capen, C.C.   (1980a) Biliary excretion  of thyroixine-I-125
     and fine structural alterations in the thyroid glands of gunn-rats fed
     PCBs.  Lab. Invest. 43:158.

Collins, W.T.; Capen, C.C.   (1980b) Ultrastructural  and functional altera-
     tions of the  rat thyroid gland produced by polychlorinated biphenyls
     compared with iodide excess and deficiency, and  thyrotropin and  thyrox-
     ine administration.  Virchows  Arch.  B. 33(3):213-231.

Collins, W.T.; Capen, C.C.   (1980b) Fine structural  lesions and hormonal
     alterations in thyroid glands  of  perinatal  rats  exposed in utero and
     by the  milk to polychlorinated biphenyls.  Am. J.  Pathol.  99(1):125-141.


                                      D-25

-------
Davidorf, F.H.; Knupp,  J.A.   (1979)   Epidemiology  of  ocular melanoma.   Inci-
     dence and geographic relationship  in  Ohio  (1967-1977).   Ohio  State  Med.
     J. 75(9):561-564.

DiGiovanni, J.; Slaga,  T.J.;  Berry,  D.L.;  Juchan,  M.R.   (1978)   The  effects
     of 2,3,7,8-tetrachlorodibenzo-p-dioxin  (TCDD)  and  Aroclor  1254  on  7,12-
     dimethylbenz[A]anthracene  metabolism  and tumor-initiating  activity  in
     mouse skin.  Proc. AACR, ASCO  19:110.

Dikshith, T.S.S.;  Rockwood, W.; Abraham, R.; Coulston,  F.   (1975)  Effects of
     polychlorinated biphenyls  (Aroclor 1254) on  rat  testis.  Exp. Mol.  Pathol.
     22:376.

Fishbein, L.  (1974)  Toxicity  of chlorinated biphenyls.   Ann.  Rev.  Pharmacol.
     14:139.

Garthoff, L.H.; Cerra,  F.E.;  Marks,  E.M.   (1981)   Blood chemistry  alterations
     in rats after single and multiple  gavage administration  of polychlorina-
     ted biphenyls.  Toxicol. Appl.  Pharmacol.  60(1):33-44.

Garthoff, L.H.; Friedman, L.; Farber, T.M.;  Locke,  K.K.; Sobotka,  T.J.;
     Green, S.; Hurley, N.E.; Peters, E.L.;  Story,  G.E.; Moreland, F.M.;
     Graham, C.H.; Keys, J.E.;  Taylor,  M.J.; Seal era, J.V.; Rothlein, J.E.;
     Marks, E.M.;  Cerra, F.E.;  Rodi, S.B.; Sporn,  E.M.   (1977)   Biochemical and
     cytogenetic effects in rats caused by short-term ingestion of Aroclor 1254
     or Firemaster BP6.  J. Toxicol. Environ. Health  3:769-796.

Gartner, L.M.; Arias, I.M.  (1966)   Studies  of  prolonged neonatal  jaundice in
     the breast-fed infant.   J. Pediat. 68(1):54.

Gillette, J.R.  (1967)   Individually different  responses to drugs  according
     to age, sex,  and functional or  pathological  state.  In:  Uolstenholme,
     G.; Porter, R., eds. Drug responses  to man.   Churchill, London, p. 28.

Grant, D.L.; Phillips,  W.E.J.  (1974)   The effect  of  age and  sex on  the  toxi-
     city of Aroclor 1254, a  polychlorinated biphenyl,  in  the rat.   Bull.
     Environ. Contam. Toxicol.  12:145-152.

Green, S.; Sauro,  F.M.; Friedman, L.  (1975)  Lack of dominant  lethality in
     rats treated  with  polychlorinated  biphenyls  (Aroclors 1242 and  1254).
     Food Cosmet.  Toxicol. 13:507-510.

Haque, R.; Schmedding,  D.W.;  Freed,  V.H.   (1974)   Aqueous  solubility, adsorp-
     tion and vapor behavior  of polychlorinated biphenyl Aroclor 1254.
     Environ. Sci. Technol. 8(2):139-142.

Haque, R.; Schmedding,  O.W.   (1975)   A  method of  measuring the  water solubility
     of hydrophobic chemicals:  solubility  of five  polychlorinated  biphenyls.
     Bull. Environ. Contam. and Toxicol. 14(1):13-18.
                                      D-26

-------
Meddle, J.A.; Bruce,  W.R.   (1977)   Comparison of tests for mutagenicity or
     car.cinogenicity  using assays  for  sperm  abnormalities, formation of
     micronuclei,  and mutations  in  Salmonella.   In:  Hiatt., H.H., ed.  Origins
     of human cancer: proceedings  of conference  on cell proliferation, Cold
     Spring Harbor,  NY.   Vol.  4, p.  1549.

Hofvander, Y.; Hagman, U.; Under,  C.E.; Vaz, R.; Slorach, S.A.   (1981)
     World Health  Organization collaborative breast feeding study.   I.
     Organochlorine  contaminants in individual samples of Swedish human milk,
     1978-1979.  Acta. Paediatr. Scand.  70(l):3-8.

International Agency  for Research  on Cancer  (IARC).  (1978)   Polychlorinated
     biphenyls.  In:   IARC monographs  on the evaluation of the carcinogenesis
     risk of chemicals to humans,  Vol.  18, p. 43.  Lyon, France.

International Agency  for Research  on Cancer  (IARC).  (198?) IARC monographs on
     the evaluation  of the carcinogenesis  risk of chemicals to humans.  Supple-
     ment 4.  Lyon,  France.

Ito, N.; Nagasaki, H.; Aral, M.; Makiura,  S.; Sugihara, S.; Hirao,  K.   (1973)
     Histopathologic  studies on  liver  tumorigenesis induced in mice by tech-
     nical polychlorinated biphenyls and its promoting effect on liver tumors
     induced by benzene hexachloride.   J.  Natl.  Cancer Inst.  51:1637-1646.

Ito, N.; Nagasaki, Makiura, S.;  H.;  Arai,  M. (1974) Histopathological studies
     on liver tumorigenesis in rats treated  with polychlorinated biphenyls.
     Gann. 65:545-549.

Jonsson, H.T. Jr.; Walker, E.M.; Green, W.B.; Hughson, M.D.;  Hennigar, G.R.
     (1981)  Effeft  of prolonged exposure  to dietary DDT and  PCB on rat liver
     morphology.   Arch.  Environ. Contam. Toxicol. 10(2):171-183.

Kasza, L.; Collins,  W.T.; Capen, C.C.;  Garthoff, L.H.; Friedman, L.  (1978a)
     Comparative toxicity of polychlorinated biphenyls and polybrominated
     biphenyl in the rat thyroid gland:  light and electron microscope altera-
     tions after subacute dietary  exposure.  J.  Environ. Pathol. Toxicol.
     1:587-599.

Kasza, L.; Weinberger, M.A.; Hinton, D.E.   (1978b)  Comparative toxicity of
     polychlorinated  biphenyls and polybrominated biphenyl in the.rat thyroid
     gland: light  and electron microscope  alterations after subacute dietary
     exposure.  J. Environ. Pathol.  Toxicol. 1:241-257.

Kimbrough, R.; Buckley,  J.; Fishbein,  L.;  Flamm, G.; Kasza, L.; Marcus, W.;
     Shibko, S.;  Teske,  R.  (1978)   Animal toxicity.  Environ. Health Perspect.
     24:173-184.

Kimbrough, R.D.;  Linder, R.E.   (1974)   Induction of adenofibrosis and hepa-
     tomas in the  liver of Balb/cJ mice by polychlorinated biphenyls  (Aroclor
     1254).  J. Natl. Cancer  Inst.  53:547-579.
                                      D-27

-------
Kimbrough, R.O.; Under, R.E.; Gaines,  T.B.   (1972)   Morphological  changes
     in livers of rats fed polychlorinated  biphenyls.   Arch.  Environ.  Health
     25:354.

Kimbrough, R.D.; Squire, R.A.; Under,  R.E.;  Strandberg,  J.D.;  Montali,  R.J.;
     Burse, V.W.  (1975)  Induction of  liver  tumors  in  Sherman  strain  female
     rats by polychlorinated biphenyls  (Aroclor  1260).   J.  Natl.  Cancer  Inst.
     55:1453-1459.

Kimura, N.T.; Baba, T.  (1973)  Neoplastic  changes in the rat liver induced
     by polychlorinated biphenyls.   Gann. 64:105-108.

Koller, L.D.  (1977)  Enhanced polychlorinated biphenyls  lesions  in Moloney
     leukemia virus-infected mice.   Clin. Toxicol. 11(1):107-116.

Komives, G.K.  (1979)   Body weights, food intakes, and  water  intakes in  rats
     during daily administration of closely controlled  doses  of polychlorinated
     biphenyls.   Bull. Environ. Contain. Toxicol.  22(6):761-766.

Komives, G.K.; Alayoku, G.  (1980)   Body temperatures and weights in rats
     during daily administration of closely controlled  doses  of polychlorin-
     ated biphenyls.  Bull. Environ. Contain.  Toxicol. 25(6):913-917.

Kuratsune, M.  (1976)   Epidemiologic studies  on  Yusho.   In:   Higuchi,  K., ed.
     PCB poisoning and pollution,  pp.  9-23.

Kuwabara, K.; Yakushiji, T.; Uathanabe, I.; Yoshida,  S.;  Kyoama,  K.; Kunita,
     N.  (1979)   Levels of polychlorinated  biphenyls  in blood of breast-fed
     children whose mothers are non-occupationally exposed to PCBs. Bull.
     Environ. Contain.  Toxicol. 21(4-5):458-462.

Lester, R.; Schmid, R.  (1964)  Bilirubin metabolism.   New Engl.  J. Med.
     270(15):779.

Lewin, C.; McBlain, W.A.; Wolfe, F.H.  (1972) Acute  intraperitoneal toxicity
     of DDT and PCBs in mice using  two  solvents.  Bull. Environ.  Contain.
     Toxicol. 8(4):245-250.

Linder, R.E.; Gaines,  T.B.; Kimbrough,  R.D.   (1974)   The  effect of PCB on rat
     reproduction.  Food Cosmet. Toxicol. 12:63-77.

Lokietz, H.; Dowben, R.M.; Hsia, O.Y.  (1963) Studies  on the effect of  novo-
     biocin and glucuronyl transferase. Pediatrics  32:47.

Loose, L.D.; Pittman,  K.A.; Benitz, K.-F.;  Sitknorth, J.B.; Mueller, W.; and
     Coulston, F.  (1978)  Environmental chemical-induced immune dysfunction.
     Ecotoxicol. Environ. Safety 2:173.

Mes, J.; Davies, O.J.   (1979)  Presence of  polychlorinated biphenyls and
     organochlorine pesticide residues  and  the absence  of polychlorinated
     terphenyls in Canadian human milk  samples.   Bull.  Envion.  Contam.
     Toxicol. 21(3):381-387.
                                      D-28

-------
McConnell, E.E.; Kinney, J.D.   (1978)   Exquisite  toxicity  in  the  guinea  pig
     to structurally similar halogenated  dioxins,  furans,  biphenyls,  and
     naphthalenes.  Toxicol. Appl.  Pharmacol.  45:298.

Mieure, J.P.; Hicks, 0.; Kaley,  R.G.;  Saeger,  V.W.   (1976)  Characterization
     of polychlorinated biphenyls.   National  Conference  on  Polychlorinated
     Biphenyls (Nov. 19-21,  1975,  Chicago,  IL).   EPA-560/6-75-004.   U.S.
     Environmental Protection  Agency,  Washington,  D.C.

Miller, J.W.  (1944)  Pathologic changes  in animals  exposed to  a  commercial
     chlorinated diphenyl.   Pub. Health Rep.  59:1085.

Miller, R.W.  (1977)  Pollutants in breast  milk:  PCBs  and  cola-colored  babies
     (editorial).  0. Pediatr. 90(3):510-511.

Morgan, R.W.; Ward, J.M.; Hartman,  P.E.  (1981)   Aroclor 1254-induced intes-
     tinal metaplasia and adenocarcinoma  in the  glandular  stomach of  F344
     rats.  Cancer Res. 41(12)5052-5059.

Narbonne, J.F.  (1979)  Effects  of age and  sex on liver  responses of  Pheno-
     clor DP6, a polychlorinated biphenyl,  in  the rat.   Bull. Environ.
     Contain. Toxicol. 22(l):49-54.

National Academy of Sciences (NAS).  (1980)  Drinking  water and health,
     Vol. 3.  Washington, D.C.:  National  Academy  Press,  p..  133.

National Cancer Institute (NCI).  (1978)   Bioassay of  Aroclor (trademark)
     1254 for possible carcinogen!city.  CAS No.  27323-18-8.  Carcinogenesis
     Technical Report Series No. 38.  Bethesda, MD.

National Institute for Occupational Safety  and Health  (NIOSH).   (1977)
     Criteria for a recommended  standard...occupational  exposure  to poly-
     chlorinated biphenyls (PCBs).   U.S.  Department  of Health,  Education, and
     Welfare.

Nelson, N.; Hammon, B.; Nisbet,  C.T.;  Sarafirm, A.F.;  Dewey,  W.H.  (1972)
     Polychlorinated biphenyls:  environmental  impact.   Environ. Res.  5:249.

Nilsson, B.; Ramel, C.  (1974)  Genetic tests  on  Drosophila melanogaster with
     polychlorinated biphenyls (PCBs). Hereditas 77:319-322.

Nisbet, I.C.T.; Sarofim, A.F.   (1972)   Rates and  routes  of  transport  of PCBs
     in the environment.  Environ.  Health Perspect.  1:21.

Nishizumi, M.  (1980)  Reduction of diethylnitrosamine-induced  hepatoma in  rats
     exposed to polychlorinated  biphenyls through their  dams.  Gann 71(6):910-
     912.

Nyhan, W.L.  (1961)  Toxicity  of drugs in the neonatal  period.  J.  Pediat.
                                      D-29

-------
Orberg, J.; Kihlstrom, J.E.   (1973)   Effects of long-term  feeding of poly-
     chlorinated biphenyls (PCB,  Clophen A60) on the length of the oestrous
     cycle and on the frequency of implanted ova in the mouse.  Environ. Res.
     6:176.

Peakall, O.B.; Lincer, J.L.;  Bloom,  S.E.   (1972)  Embryonic mortality and
     chromosomal alterations  caused  by  Aroclor 1254 in ring doves.  Environ.
     Health Perspect. 1:103-104.

Rogan, W.J.; Bogniewska, A.;  Damstra, T.   (1980)  Pollutants in breast milk.
     New Engl. J. Med. 302:1450.

Rosin, D.L.; Martin, B.R.   (1983)   Comparison of the effects of acute and
     subchronic administration  of  Aroclor  1254, a commercial mixture of
     polychlon'nated biphenyls, on pentobarbital-induced sleep time and
     [  C]pentobarbital  disposition  in  mice.  J. Toxicol.  Environ. Health
     11:917-931.

Sager, D.B.  (1983)   Effect of  postnatal exposure to polychlorinated biphenyls
     on adult male reproduction function.   Environ. Res. 31(l):76-94.

Sanders, O.T.; Zepp, R.L.; Kirkpatrick, R.L.  (1974)  Effect of PCB inyestion
     on sleeping times, organ weights,  food consumption, serum corticosterone
     and survival of albino mice.   Bull. Environ. Contam.  Toxicol. 12(4):394-
     399.

Schoeny, R.S.  (1982)  Mutagenicity  testing of chlorinated biphenyls and
     chlorinated dibenzofurans.  Mutat. Res. 101:45.

Schoeny, R.S.; Smith, C.C.; Loper, J.C. (1979)  Non-mutagenicity for Salmon-
     ella of the chlorinated  hydrocarbons  Aroclor 1254, 1,2,4-trichloroben-
     zene, mi rex, and kepone.  Mutat. Res.  68:125-132.

Spencer, F.  (1982)   An assessment of the  reproductive toxic potential  of
     Aroclor 1254 in female Sprague-Dawley  rats.  Bull. Environ. Contam.
     Toxicol. 28(3):290-297.

Street, J.C.; Sharma, R.P. (1975)  Alteration of induced  cellular and humor-
     al immune responses by pesticides  and  chemicals of environmental concern:
     quantitative studies  of  immunosuppression by DDT, Aroclor 1254, carbaryl,
     carbofuran, and methyl parathion.  Toxicol. Appl. Pharmacol. 32:587.

Urabe, H.  (1974)  Foreword to  the fourth  report of the Study Group on Yusho
     and PCB.  Fukuoka Acta.  Med.  65:145.   (In Japanese).

U.S. Environmental Protection Agency.  (1980)  Ambient water quality criteria
     document for PCBs.  Environmental  Criteria and Assessment Office, Cin-
     cinnati, OH.  EPA-440/5-80-068.   Available from:  National Technical
     Information Service,  Springfield,  VA;  PB81-117798.

U.S. Environmental Protection Agency.  (1983)  Quantitative risk assessment
     of reproductive risks associated with  polychlorinated biphenyl  (PCB)
     exposure.  Support Document  B,  Docket  No. OPTS-62032  PCB.


                                      D-30

-------
U.S. Environmental  Protection Agency.   (1984a)   Proposed  guidelines  for  car-
     cinogen risk assessment; request  for comments.   Federal  Register  49(227):
     46294-46301.

U.S. Environmental  Protection Agency.   (1984b)   Health  effects  assessment  for
     polychlorinated biphenyls.   Environmental  Criteria and  Assessment Office,
     Cincinnati, OH.  Final  draft.  EPA-540/1-86/004.   Available  from:  National
     Technical  Information Service,  Springfield,  VA;  PB86-134152/AS.

U.S. Environmental  Protection Agency.   (1985a,  June  12)  Personal  communica-
     tion, Don  R. Clay, Director,  Office  of  Toxic  Substances.

U.S. Environmental  Protection Agency.   (1985b)   Drinking  water  criteria  docu-
     ment for polychlorinated biphenyls (PCBs).   Environmental  Criteria  and
     Assessment Office, Cincinnati,  OH.  Final  draft.   EPA-600/X-84-198.
     Available  from:  National  Technical  Information  Service, Springfield, VA;
     PB118312/AS.

Villeneuve, D.C.; Grant, B.L.;  Khera,  K.; Clegg,  D.J.;  Baer,  H.;  and Phillips,
     W.E.J.  (1971)  The fetotoxicity  of  a polychlorinated biphenyl mixture
     (Aroclor 1254) in the rabbit  and  in  the rat.  Environ.  Physiol. 1:67-71.

Vos, J.G.; van  Genderen, H.   (1973)  Toxicological aspects of immunosuppres-
     sion.  In:  Deichman, W.B., ed.  Pesticides  in the environment: a con-
     tinuing controversy;  8th International  conference  on toxicology and
     occupational medicine,  Miami, FL.  New  York,  NY: Intercontinental Medical
     Book Co.

Wickizer, T.M.; Brilliant, L.B.; Copeland, R.;  Tilden,  R.  (1981)  Polychlor-
     inated biphenyl contamination of  nursing mothers'  milk  in  Michigan  (USA).
     Am. J. Public  Health  71(2):132-137.

Wyndham, C.; Devenish, J.; Safe, S.  (1976)   The  in vitro metabolism,  macro-
     molecular  binding, and  bacterial  mutagenicity of 4»chlorobiphenyl ,  a model
     PCB substrate.  Res.  Commun.  Chem. Pathol.  Pharmacol. 15:563.

Zinkl, J.G.  (1977)  Skin  and liver  lesions  in  rats fed a polychlorinated
     biphenyl mixture.  Arch. Environ.  Contam.  Toxicol. 5(2):217-228.
                                     D-31

-------