NCD980602163_20000525_Warren County PCB Landfill_SERB C_Air Emissions - Correspondence - Info - Monitoring - Sampling, 1983 - 2000-OCR

f
-------·-----
•
RHJfl fota 2364 lttlaed (t•II)
Mboi,£wf
f,,_.,• ISf ti ~._,;..,.,,..t,: 6'10t.c...,.
.._.J..,~~: ,~.$c.e,,.,...
ta"-~ ~ ~~ I ---"
1'~..,~:
. . herwtstry
SAMPLE# 001724 001726 N.C. DEPARTMENT OF HEALffl AND HUMAN SERVICES STA TE LABORATORY OF PUBLIC HEAL m P.O.BOX 28047, RALEIGH, N.C. 27611 GAS-LIQUID CHROMATOGRAPHY REPORT SHEET PROJECT: PCBLANDf]LL DATE OF ANALYSIS: MAYB-15,.2000 MILLIGRAMS Hydrocarbons, BP Hydrocarbons, Hydrocarbons, Unidentified 36-126 °c Aromatic Halogenated Peaks . (1500) (1501) (1003) <0.040 <0.040 <0.040 None <0.040 <0.040 <0.040 None 1' NIOSH Manual of Analytical Methods (NMAM), Fourth Edition, 8/15/94. Samples desorbed in 1.0 ml CS:z, analyzed by packed column GC/FID analysts. MDL/PQL reported are typical and may not be validated for all listed analytes in method. C:\My Documcnts\Reports\001724.doc i: ..
N.C. DEPARTMENT OF HEAL TH AND HUMAN SERVICES STATE LABORATORY OF PUBLIC HEALTH P.O.BOX 28047, RALEIGH, N.C. 27611 GAS-LIQUID CHROMATOGRAPHY REPORT SHEET PROJECT: PCB Landfill DA TE OF ANAL YSIS:_-'"'4~/2-7--2-812_0-00.._ ______ _ Nanograms SAMPLE# PCBsll 001725 <20 JL 13mm Glass fiber filter desorbed in hexane, analyzed by Gas chromatography with electron capture detector. MDUPQL not fully validated, reporting level estimated. GCREP.DOC 1/99 REV -, I: • .,,
PROJECT: SAMPLE# Chromium 001727 0.309 N.C. DEPARTMENT OF HEAL TH AND HUMAN SERVICES STATE LABORATORY OF PUBLIC HEALTH P.O.BOX 28047, RALEIGH, N.C. 27611 REPORT SHEET PCB Landfill DATE OF ANALYSIS: April 13, 2000 (ICP/MS) Micrograms1' Manganese Copper Zinc Arsenic Silver Cadmium Barium 0.0138 0.042 0.405 0.003 0.009 0.016 0.017 Lead 0.050 1'Reported results based on acid digestion ofMCEF, 25 ml final volume, ICP/MS Analysis with no blank or Trip Blank subtracted. These results may be below the MDL/PQL value(s). C:\My Documents\Templates\001727 .doc
Air Concentrations Calculations
1. Convert flow rates to exposure limit oonditions.
Standard oonditions from NIOSH guide are 68°F and 29.94 in Hg.
2. Average air flow rate.
2
3. Calculate total volume of air.
VT = vave X Runtime
4. Calculate ooncentration.
Concentration = Quantity + VT
5. Convert to NIOSH guide units.
RESULTS OF AIR SAMPLING AT LEACHATE SAND FILTER
Sample no.
Pump no.
Beginning flow rate
Ending flow rate
Runtime
Average flow rate
Beginning temp
Ending temp
Reference temp
Pressure
Reference pressure
Total volume of air
Chromium
Manganese
Copper
Zinc
Arsenic
Silver
Cadmium
Barium
Lead
NIOSH
NIOSH Limits
OSHA Limits
IDLH Limits
001727
13077
1218 cm3/min
1236 cm3/min
240 min
45 F =
60 F =
68 F =
789.5 torr =
=
Metals
Date 415/2000
= 1323 scm3/min
= 1304 scm3/min
1313 scm3/min
505 R
520 R
528 R
31 .08 in Hg
29.92 in Hg
315,209 scm3
Sample results NIOSH IDLH
ug ug/scm3 mg/m3 mg/m3 Notes mg/m3
0.3090 9.80E-07 0.00098 0.500
0.0138 4.38E-08 0.00004 1.000
0.0420 1.33E-07 0.00013 1.000
0.4050 1.28E-06 0.00128 5.000 Zinc Oxide
0.0030 9.52E-09 0.00001 0.010 ttOSHA
0.0090 2.86E-08 0.00003 0.010
0.0160 5.08E-08 0.00005 0.005 ttOSHA
0.0170 5.39E-08 0.00005 0.500 Barium Nitrate
0.0500 1.59E-07 0.00016 0.100
National Institute of Occupational Safety and Health
Part of US Department of Health and Human Services
It is responsible for recommending health and safety standards.
Recommended exposure limits are time-weighted average concentrations
for up to a 10-hour workday during a 40-hour week.
Time-weighted average concentrations that rrust not be exceeded
during any 8-hour work shift of a 40-hour week.
lrrmediately Dangerous to Life or Health
The maxirrum concentration about which only a highly reliable breathing
apparatus providing maxirrum worker protection is permitted. As a
safety margin, these limits were set to allow a worker the ability to escape
without loss of life or irreversible health effects and also consider the effect
of eye or respiratory irritation and other deleterious effects that would
prevent escape.
250
500
100
500
5
10
9
50
100
RESULTS OF AIR SAMPLING AT LEACHATE SAND FILTER
Sample no.
Pump no.
Beginning flow rate
Ending flow rate
Runtime
Average flow rate
Beginning temp
Ending temp
Reference temp
Pressure
Reference pressure
Total volume of air
PCB
NIOSH
NIOSH Limits
OSHA Limits
IDLH Limits
001725
15118
640 cm3/min
645 cm3/min
241 min
45 F
60 F
68 F
789.5 torr
=
PCBs
Date 4/5/2000
695 scm3/min
680 scm3/min
688 scm3/min
505 R
520 R
528 R
31.08 in Hg
29.92 in Hg
165.751 scm3
Sample results NIOSH
ng ug/scm3 mg/m3 mg/m3
20 1.21E-07 0.00012 0.001
Notes
Less than
detection
National Institute of Occupational Safety and Health
Part of US Department of Health and Human SeNices
It is responsible for recommending health and safety standards.
IDLH
mg/m3
5
Recommended exposure limits are time-weighted average concentrations
for up to a 10-hour workday during a 40-hour week.
Time-weighted average concentrations that must not be exceeded
during any 8-hour work shift of a 40-hour week.
Immediately Dangerous to Life or Health
The maximum concentration about which only a highly reliable breathing
apparatus providing maximum worker protection is permitted. As a
safety margin, these limits were set to allow a worker the ability to escape
without loss of life or irreversible health effects and also consider the effect
of eye or respiratory irritation and other deleterious effects that would
prevent escape.
RESULTS OF AIR SAMPLING AT LEACHATE SAND FILTER
Sample no.
Pump no.
Beginning flow rate
Ending flow rate
Runtime
Average flow rate
Beginning temp
Ending temp
Reference temp
Pressure
Reference pressure
Total volume of air
Hydrocarbons
(BP 36-126 C)
Hydrocarbons
(Aromatic)
Hydrocarbons
(Halogenated)
NIOSH
NIOSH Limits
OSHA Limits
IDLH Limits
001724
15118
Hydrocarbons
Date 4/5/2000
640 cm3/min = 695 scm3/min
645 cm3/min = 680 scm3/min
241 min
688 scm3/min
45 F = 505 R
60 F = 520 R
68 F = 528 R
789.5 torr = 31 .08 in Hg
= 29.92 in Hg
165751 scm3
Sample results NIOSH
mg mg/scm3 mg/m3 mg/m3
0.040 2.41E-07 0.2413
0.040 2.41E-07 0.2413
0.040 2.41E-07 0.2413
National Institute of Occupational Safety and Health
Part of US Department of Health and Human Services
Notes
Less than
detection
Less than
detection
Less than
detection
It is responsible for recommending health and safety standards.
IDLH
mg/m3
Recommended exposure limits are time-weighted average concentrations
for up to a 10-hour workday during a 40-hour week.
Time-weighted average concentrations that rrust not be exceeded
during any 8-hour work shift of a 40-hour week.
lrrrnediately Dangerous to Life or Health
The maxirrum concentration about which only a highly reliable breathing
apparatus providing maxirrum worker protection is permitted. As a
safety margin, these limits were set to allow a worker the ability to escape
without loss of life or irreversible health effects and also consider the effect
of eye or respiratory irritation and other deleterious effects that would
prevent escape.
RESULTS OF AIR SAMPLING AT LEACHATE SAND FILTER
Sample no.
Pump no.
Beginning flow rate
Ending flow rate
Runtime
Average flow rate
Beginning temp
Ending temp
Reference temp
Pressure
Reference pressure
Total volume of air
Hydrocarbons
(BP 36-126 C)
Hydrocarbons
(Aromatic)
Hydrocarbons
(Halogenated)
NIOSH
NIOSH Limits
OSHA Limits
IDLH Limits
001726
13077
Hydrocarbons
Date 4/5/2000
1218 cm3/min = 1323 scm3/min
1236 cm3/min = 1304 scm3/min
240 min
1313 scm3/min
45 F = 505 R
60 F = 520 R
68 F = 528 R
789.5 torr = 31 .08 in Hg
= 29.92 in Hg
315,209 scm3
Sample results NIOSH
mg mg/scm3 mg/m3 mg/m3
0.040 1.27E-07 0.1269
0.040 1.27E-07 0.1269
0.040 1.27E-07 0.1269
National Institute of Occupational Safety and Health
Part of US Department of Health and Human Services
Notes
Less than
detection
Less than
detection
Less than
detection
It is responsible for recommending health and safety standards.
IDLH
mg/m3
Recommended exposure limits are time-weighted average concentrations
for up to a 10-hour workday during a 40-hour week.
Time-weighted average concentrations that must not be exceeded
during any 8-hour work shift of a 40-hour week.
lrmiediately Dangerous to Life or Health
The maximum concentration about which only a highly reliable breathing
apparatus providing maxirrum worker protection is permitted. As a
safety margin, these limits were set to allow a worker the ability to escape
without loss of life or irreversible health effects and also consider the effect
of eye or respiratory irritation and other deleterious effects that would
prevent escape.
REMEDIATION OF
PCB CONTAMINATED SEDIMENTS
VOLATILITY AND SOLUBILITY
CONSIDERATIONS
Ronald J. Scrudato, Jeffrey R. Chiarenzelli,
J3mes J. Pagano and Michele Wunderlich
Environmental Research Center
State University of New York
Oswego, NY 13126
Summary
Through volatilization and long distance
atmospheric transport, polychlorinated biphenyls
(PCBs) have been redistributed throughout the
global environment. Over the last 70 years, these
compounds have permeated into every known
environmental niche including the remote polar
regions of the globe. In this paper, the solubility
and volatility of the PCB congeners are reviewed
relative to the remedial technologies that are
currently in use or under consideration. The
following discussion focuses primarily on the
management options for PCB contaminated,
subaqueous solids that require removal,
dewatering, drying and either treatment to degrade
the target contaminants and/or containment in
engineered facilities including constructed islands,
upland secure landfills and subaqueous pits.
Environmental mobility resulting from natural and
engineered processes is discussed in relation to
the potential for contributing to the global loading
and redistribution of PCBs. Additionally, select
emerging technologies and management options
are reviewed relative to tht:ir potential to produce
secondary environmental impacts resulting from
the soluble and/or volatile redistribution of PCBs.
Based on a lack of long term experience and
recognition that contaminants will remain
unaltered for decades, technologies involving
engineered containment structures should be
considered temporary remedial measures until
·-----
I• I
cost competitive, destructive processing of
contaminated sediments is feasible.
Introduction
Polychlorinated biphenyls (PCBs) are a family of
semi volatile, organic chemicals that exhibit
properties that made them highly desirable for use
in a wide range of industrial applications. This
class of compounds was extensively used in
processes that required low water solubility, low
vapor pressure, inflammability, high heat
capacity, low electrical conductivity, and
favorable dielectric properties. Although accurate
data on the total amount of PCBs produced by the
Monsanto Industrial Chemicals Co., sole
manufacturer in the United States, are not
available, it is estimated that more than 1.2 billion
pounds were manufactured during the period
1929-1977 (Durfee, et al, 1976). These
compounds were used in electrical insulating
fluids, lubricants requiring tolerance to high
pressures and temperatures, to improve heat and
fire resistance in transfer and hydraulic fluids and
as a constituent in elastomers, inks, adhesives,
paints, lacquers, varnishes, pigments, garment
coatings, pesticides, carbonless copy paper and
waxes (Fuller, et al, 1976).
In addition to the United States, other known
producers of PCBs include the United Kingdom,
France, Italy, Germany, Spain, Czechoslovakia,
and up to I 972, Japan. It is estimated that PCB
production in the United States accounted for
about one-half of the total world production. By
mid-1971, Monsanto terminated sales of PCBs
and polychlorinated terphenyls (PCTs) except for
use in closed electrical systems. In 1976 the
Toxic Substances Control Act was passed by
Congress and the U.S. production of PCBs was
soon halted because of growing concerns over the
environmental persistence and the adverse effects
attributed to this class of compounds. Although
production in the United States ceased in 1977,
PCBs continue to be used in a wide range of
industrial applications in the United States and
I •
abroad. In 1976, the USEPA estimated that of the
total amount of PCBs purchased by U. S.
industries for the period 1930-1975, about 150
million pounds were distributed in the general
environment contaminating water, air, solids and
biota (Durfee, et al, 1976).
As can be seen from Exhibit 1, the PCB molecule
consists of the biphenyl ring with one to as many
as ten chlorines attached at the ortho, meta or para
positions. These compounds have been
internationally marketed under a number of trade
names including Aroclor (U.S.), Clophen
(Germany), Phenoclor (France), K.anechlor
(Japan), Cloresil (Italy) and others. The Aroclors
manufactured in the U .S by Monsanto included a
range of chlorinated compounds with 21 to as
much as 70 percent chlorine. Although an
Aroclor can theoretically contain as many ·as 209
individual compounds or congeners, in practice,
environmental samples contain far fewer
identifiable chromatographic peaks.
PCB Solubility and Volatility
Each PCB congener has unique physico-chemical
properties dependent on the number and position
of the chlorine substitution on the biphenyl.
Although there are exceptions, with increasing
chlorination, congeners become less soluble and
less volatile and are consequently more
environmentally stable and immobile. For
example, the water solubility and vapor pressure
(25°C) of a PCB compound with 42 percent
chlorine, marketed in the United States as Aroclor
1242, is 0.016 mg/I and 5.7x 10-4, whereas the
more chlorinated Aroclor 1260 is 9.5 x 10-4 mg/I
and 1.3 x 10-5, respectively (Hatzinger, et al,
1974). In addition to being less soluble, the more
chlorinated PCB compounds have the greater
affinity for solid particles and are also the most
resistant to chemical and reductive biochemical
degradation.
PCBs are generally noted for their low water
solubility, yet based on the chlorine content of the
compound, the solubility ranges from several
parts per million to sub part per billion for the
more chlorinated compounds (Ruell, et al, 1993;
Exhibit 2). As can be seen from Exhibit 2, the
solubility of the range of congeners comprising
Aroclors can vary by more than seven orders of
magnitude. The solubility of individual PCB
compounds varies by orders of magnitude
depending on the degree of chlorination. For
example, the water solubility of a PCB congener
with one chlorine attached to the biphenyl, 2-
monochlorobiphenyl, is 5.9 mg/1 whereas the
octachlorinated biphenyl solubility is 0.007 mg/1.
The discharge of a specific Aroclor to the aquatic
environment can therefore result in the selective
partitioning of the lower chlorinated congeners to
the aqueous phase. Bush et al. (1985) reported
that 65 percent of the PCBs contained in upper
Hudson River water consisted of three, lower
chlorinated congeners. The more chlorinated
compounds are more readily sorbed to particulates
and therefore become concentrated in the
suspended and bottom sediments.
Within water bodies, microbial processes can
have a profound affect on the solubility and
therefore the availability of PCBs. Anaerobic
microbial dechlorination of PCB-contaminated
bottom sediments does not degrade compounds
with chlorines attached at the 2 or 6 positions
(ortho) on the biphenyl. As a consequence, lower
chlorinated, orthosubstituted compounds are
enriched in the partially degraded sediment. The
residual, partially transformed PCB compounds,
are more soluble and therefore more available to
aquatic organisms. Exhibit 3 depicts the
microbial reductive dechlorination of PCBs with
time. As can be seen from this illustration, about
25 percent of the total PCBs were degraded within
about three months. Dechlorination plateaus after
about three months and no further degradation
was observed during the remaining 11 o+ week
period (Pagano, et al, 1995, 1997). Furthermore,
florisil traps used to collect the gaseous products
from the reactor used in these experiments
indicated the lower chlorinated, primarily
, ..
orthosubstituted congeners escaped the reactor as
volatile PCBs.
Increased solubility considerations also become
an availability factor with physico-chemical and
biochemical degradation process'es that result in
the production of lower chlorinated intermediate
products. For example, hydroxylated PCBs are
produced as intermediate products during
photocatalysis (Wang, et al, 1998), and the
produced compounds may be more soluble and
therefore more available than the suite of parent
compounds.
Although relatively insoluble in water, PCBs are
lipophilic and readily concentrate in fats and oils.
The accumulation of PCBs in the fatty tissues of
Great Lakes fish was one of the growing
environmental concerns that led to the limited use
of these compounds to closed systems during the
early to rnid-1970s and to the eventual ban on its
production and wide use in the United States.
PCBs persist in Great Lakes' fish at levels
requiring fish advisories, which recommend that
children and women of child-bearing age eat no
fish and all others limit consumption to one meal
per month.
In addition to carcinogenic concerns, there is
growing research evidence that the lower
chlorinated PCBs may be implicated in a variety
of health effects including breast cancer,
hormonal disruption and neurobehavioral impacts
(Arcaro and Gierthy 1997; Carpenter, et al, 1997;
Seegal, et al, 1997). Natural and engineered
processes contributing to the increased
bioavailability of the lower chlorinated PCB
compounds may therefore have significant health
implications which to date have been overlooked
or disregarded.
The potential and concern for volatile losses,
atmospheric transport and redistribution of PCBs
were recognized as early as the mid-1970s
(Bidelman and Olney 1974; Harvey and
Steinhauer 1974; Haque, et al, 1974; Fuller, et al,
'.
1976). In reports prepared for the USEPA, vapor
phase PCBs were recognized as a major
component of the total PCB budget (Durfee, et al,
1976). More than twenty years ago, the USEPA
estimated that as much as 13 percent of the total
PCB input to Lake Michigan was from
atmospheric fallout. Recent research indicates
PCBs and other semi-volatile compounds readily
volatilize from contaminated soils and sediments
and as much as 75 percent of the total PCBs can
be lost as water is evaporated from subaqueous
sediments at ambient temperatures (Chiarenzelli,
et al, 1997a and b, 1998b).
Although classified as semivolatile, vapor phase
PCBs are recognized as a primary contributor to
the total loading of persistent organic pollutants
(POPs) that have been globally redistributed to the
higher latitudes (Wania and Mackay 1993, 1996;
Wania, et al, 1996; Muir, et al, 1996). The lower
chlorinated compounds are also the more volatile
and partition to the higher altitudes where they
have been found in Canadian alpine snows and
meltwaters (Blais, et al, 1998). Vapor phase
partitioning of PCBs will also result in the lower
chlorinated compoW1ds being atmospherically
redistributed while the more chlorinated
congeners remain as residuals in the originally
contaminated material.
PCB-Remedial Technologies
During the period of 1930-1975, an estimated 290
million pounds of PCBs were deposited in
landfills and dumps. By 1975, an additional 150
million pounds were released to the environment
providing a source of contamination to the air,
water, soil, sediment and biota (Durfee, et al,
1976). There are commercially available
technologies to manage the highly contaminated
PCB oils including secure landfills, incineration,
pyrolytic processing, and other destructive
processes. Since 1976, and passage of the Toxic
Substances Control Act (TSCA), the United States
has imposed strict regulatory guidelines governing
the transport, storage and disposal of highly
contaminated PCB substances which carry stiff
penalties for violations. Material containing less
than 50 mg/kg total PCBs can still be disposed in
landfills. Since the mid-l 970s, however, the
highly contaminated fluids contained in
decommissioned transformers and capacitors are
not being released to the environment or being
disposed in municipal landfills. PCB oils, fluids
and material containing greater than 50 mg/kg are
classified as hazardous wastes and must meet the
TSCA regulatory standards. Because they are
strictly regulated, these substances do not
currently pose as significant a problem to the
environment as other, less contaminated and more
broadly dispersed material. However, there are
large global stores of PCB contaminated liquids
and solids in the waters and in the bottom and
suspended sediments of lakes, harbors and rivers
of the world. Additiona11y, PCBs continue to be
released from environmental stores including soils
and contaminated water bodies as the atmospheric
fugacity changes. Although long regarded to be
relatively insoluble and nonvolatile, large --
quantities of PCBs sorbed to particulates and in
the vapor phase are present in the world
atmosphere.
ln aquatic systems that have been impacted by
direct and indirect discharges of PCBs, the size
and character of the receiving body of water
determines whether the majority of the total PCBs
will be in the aqueous phase or contained in the
suspended and bottom sediments. Whereas the
bottom sediments in a water body impacted by
PCB discharges may often contain mg/kg
concentrations, the associated aqueous phase
concentrations will normally be 2-3 orders of
magnitude lower. Nevertheless, depending on the
size and concentration of the water body, the
aqueous phase mass of PCBs may be greater in
large lake and river systems than the amount
stored in the suspended bottom sediments. For
example, the total estimated amount of PCBs in
solution in Lake Michigan in 1976 was 1.0xl05
pounds, whereas the total in the bottom sediments
was estimated to be l.7x104 pounds (Durfee, et al,
1976).
Remedial Processes
Aqueous Phase PCBs. Isolation and degradation
of aqueous phase PCBs is readily achievable
using a wide range of commercially available and
developing technologies including, but not limited
to, activated carbon or resin sorption, biofiltration,
ultraviolet irradiation, photocatalysis, ozonation,
peroxidation and a range of electrochemical
processes. The more chlorinated compounds are
more difficult to desorb from associated organic
and inorganic solids whereas the lower
chlorinated fractions are more difficult to remove
from contaminated liquids because of their
relative higher solubilities. Technologies
designed to degrade the aqueous phase PCBs,
including photochemical, electrochemical and
other forms of advanced oxidative processes,
preferentially degrade the lower chlorinated
congeners and may result in a relative increase in
-----
the more chlorinated compounds (Chiarenzelli, et
al., 1996). Advanced oxidative treatment of
aqueous phase PCBs can also lead to the
. production of more mobile products including
hydroxylated intermediates which may be more
soluble than the original compound (Wang, et al,
1998). The effectiveness of advanced oxidative
technologies to degrade aqueous phase PCBs is
also dependent on the physical and chemical
characteristics of the contaminated water. Water
hardness, for example, directly affects
effectiveness of the process because the radicals
are scavenged by dissolved carbonates.
Contaminated Solids and Solid Suspensions.
Although there are a number of technologies that
are effective in degrading aqueous phase PCBs,
remediation of solids and solid suspensions
present unique challenges. Tightly sequestered
compounds sorbed to particulates, mixing
effectiveness, quenching effects of associated
suspended or dissolved compounds and reduced
thermal, radiant and electrical energy transmission
achievable in solid suspensions, negatively affect
treatment efficiencies of PCB-contaminated solids
and slurries. The PCBs are more readily degraded
when they are physically and/or chemically
removed and isolated from solids. Separation
technologies to remove and isolate contaminants
sorbed to associated solids involve physical,
thermal, and chemical partitioning. Low
temperature thermal extraction has been
commercially used to separate PCBs from soils.
In this process, indirect heating of the
contaminated solids vaporizes the semi-volatile
compounds and with condensation, the PCBs are
concentrated in the aqueous phase. Once
separated from the solids, the liquids can be
isolated or degraded utilizing thermal, biological
or physico-chemical processes.
Soil washing technologies have been used to
remove contaminants adsorbed to solids in slurry
suspensions. A wide range of ionic and non-ionic
surfactants has been used in solid suspensions or
slurries to improve desorption of the contaminants
from the particulates. Additionally, a range of
organic and inorganic solvents, including steam,
has also been effectively used to extract PCBs
from contaminated solids. Bench scale steam
extraction experiments conducted on PCB-
contaminated soils reduced the total concentration
of an Aroclor 1260 from 193 to 76 mg/kg at a
liquid to solids ratio of 3: l (Chiarenzelli, personal
communication, 1998). With the use of steam
extraction, the aqueous solubility was exceeded
and three separate phases formed including the
condensed liquid, the extracted solids and a
yellow grease-like substance. Analysis of the
three phases indicated more than 95 percent of the
PCBs were contained in the grease-like material.
The liquid condensate contained 122 ug/L and the
solids retained 39 percent of the original
concentration.
Containment Technologies
Solids containing up to several hundred mg/kg
concentration of PCBs represent one of the most
challenging, contentious and politically charged
environmental issues involving PCB contaminated
materials. The extensive, worldwide
contamination of riverine and harbor sediments
contaminated by PCBs and similar semi-volatile
compounds and the large potential costs involved
with the remediation of these materials, has
heightened debate amongst regulators, responsible
parties, the scientific community and community
members.
In the United States, 200 miles of the Hudson
River have been impacted by PCB discharges by
the General Electric Corporation facility at Fort
Edwards, New York. Millions of cubic yards of
contaminated dredge material must be removed
from the New Yorlc/New Jersey harbor area
annually in order to maintain shipping channels.
Thousands of cubic yards of PCB-contaminated
sediments are planned for removal from the St.
Lawrence River contaminated by PCB and P AH
discharges by the Alcoa, Reynolds and General
Motors Corporations. Contaminated soils must be
excavated and removed from the residential
neighborhoods of Pittsfield, Massachusetts and
thousands of cubic yards of highly contaminated
sediments will be removed from the New Bedford
Massachusetts Harbor and stored in engineered
containment structures.
Long term storage of PCB-contaminated solids in
designed containment facilities does provide a
relatively inexpensive remedial option. As long
as the system is kept dry and the integrity of the
containment structure is maintained, the
contaminants should remain isolated from the
surrounding environment. Containment facilities
will require continuous monitoring and
maintenance for decades, perhaps centuries.
Because there are no long term examples to
provide insights to the effectiveness of
containment facilities to isolate contaminated
solids, in 1975 the USEPA expressed concerns
about the responsibility, liability and long term
integrity of landfills used for storage ~f PCBs.
---
Containment facilities are designed to prevent the
infiltration of surface and groundwater to
minimize leachate production. The PCBs
sampled from contaminated sediments dredged
from the Hudson River remained unchanged for
more than a decade after being deposited in
containment fill sites (Rhee, personal
communication, 1998). As long as the
contaminated sediments are kept dry, physico-
chemical and microbial processes are inhibited
and the contaminated sediments will remain
unchanged as long as the integrity of the
containment facility is intact.
Because of the low solubility and strong
partitioning to solids of the more chlorinated
congeners, the bottom sediments of water bodies
that have received direct PCB discharges serve as
vast repositories of the contaminant. These
contaminated sediments consist of admixtures of
organic and inorganic solids and when dredged
from aquatic systems, varying amounts of water
must be removed or separated in order to remove
or treat the target contaminants. Additionally,
much of the PCB-contaminated sediments found
in harbors and rivers are also impacted by other
organic and inorganic contaminants. Dioxins, a
range of trace metals and other organic
contaminants including, but not limited to,
polyaromatic hydrocarbons (PAHs) also impact
the New York/New Jersey Harbor sediments.
Remediation of subaqueous contaminated
sediments therefore presents unique problems and
challenges. The bottom sediments must be
removed without redistributing large quantities of
the contaminated material down current. Once the
saturated material is removed, it is dewatered and
if the removed material is to be thermally treated,
it must be dried.
PCBs in subaqueous environments present other
unique challenges and debate continues on
whether more environmental damage is done by
physically removing the contaminated material
from the impacted body of water. The General
Electric Corporation has argued for years that
natural biodegradation is effectively reducing the
total concentration and toxicity of the PCBs in the
Hudson River. Reductive dechlorination of
Hudson River PCBs is an active phenomena and
there is convincing evidence that the total quantity
has declined with time. However, research
conctucted on reductive microbial degradation
indicates this process is limited in that a minimum
concentration of about 40 mg/kg total PCBs is
required in order for the process to be initiated
(Sokol, et al, 1998). Furthermore, reductive
degradation is not able to completely degrade
PCBs and produces a host of lower chlorinated
compounds which are more readily available
within the aquatic environment.
Various technologies have been proposed and
assessed to remediate large quantities of
contaminated material that will be removed from
the world's waterways. The range of containment
technologies include long term storage in
engineered upland disposal facilities, containment
in excavated, subaqueous pits, and creation of fill
sites or islands near the source of the
contaminated dredge material similar to the
Miller-Hart Island (Hamons, et al., 1997) used to
contain sediments removed from the Chesapeake
Bay area near the Baltimore Harbor.
Containment is normally favored over treatment
technologies primarily because of the relatively
lower costs. Depending on the selected
containment option and transportation costs, this
management option is relatively inexpensive
compared to treatment technologies designed to
degrade the contaminant. However, recent studies
suggest significant quantities of PCBs volatilize as
water evaporates from contaminated solids
(Chiarenzelli, et al, 1997a, b; 1998a, b ). Although
these studies were conducted on small samples at
the laboratory scale, there is sufficient evidence to
indicate PCBs and similar semivolatile organic
compounds are being globally redistributed
atmospherically f'N ania and Mackay 1993, 1996;
Muir, et al, 1996; Tenenbaum 1998).
The PCB concentration of the St. Lawrence River
sediments used in the series of experiments
illustrated in Exhibits 4 and 5 was about 63
mg/kg. Exhibit 4 schematically illustrates the
relative volatile losses of a PCB-contaminated
sediment in relation to initial moisture content.
As can be seen from this exhibit, volatile losses
were directly related to the initial moisture content
and maximum losses occurred when the sample
was completely saturated. Exhibit 5 depicts the
relative volatile losses of Aroclors spiked onto
quartz sands. Note also that the sediment
collected from the St. Lawrence River (StLR)
behaved more like Aroclor 1260 even though the
original Aroclor consisted of 1248. The St.
Lawrence River sediment used in the experiments
depicted in Exhibit 5 was subjected to more than
twenty years of reductive dechlorination and
therefore enriched in the lower chlorinated
congeners. The difference in volatility is
attributed to the age of the sediment and the tight
sequestering of the PCBs to the solids (Hatzinger
and Alexander, 1995). It is evident from these
data that the lower chlorinated Aroclors more
readily volatilize at ambient temperatures and
relative humidity. In a series of similar
experiments conducted on contaminated St.
Lawrence River sediments, the volatile loss of
PCBs at ambient temperature and relative
humidity was found to be directly related to the
evaporative loss of water (R2 >.998).
The potential for volatile losses of PCBs from
drying sediments at ambient temperatures and
relative humidity suggests that effective
monitoring is required to ensure that volatilization
does not contribute to the global burden of
atmospheric organic contaminants. Safeguards
may be required to prevent volatile losses during
the dewatering and drying of contaminated dredge
material. This is particularly true for those
sediments that have been microbially degraded
resulting in the production of lower chlorinated
and therefore more soluble and volatile
compounds.
One containment technology currently being
considered for the New York/New Jersey dredge
material is to transport the material from the
dredge site by barge and then by rail to the
Pennsylvania coal fields where it will be mixed
with cement and then used to fill abandoned
mines. Solidification of contaminated dredge
material by addition of a cementing agent is also
being considered for other containment facilities.
Addition of a cementing agent to dredge material
can increase the temperature of the mixture and
thereby increase the evaporative loss of water,
potentially resulting in the volatile losses of semi-
volatile compounds. In order to assess the effects
ofJime additions to PCB contaminated material,
the USEP A sponsored a research project designed
to determine whether the compound affected the
temperature of the admixture and resulted in the
evaporative loss of the PCBs (Constant, et al.,
1995). Quicklime additions resulted in significant
increases in the temperature of the contaminated
is active in most subaqueous environments.
Anaerobic microbial degradation, however,
cannot totally degrade the PCBs and produces
lower chlorinated, orthosubstituted PCBs,
enhancing the solubility and availability of the
contaminant to aquatic organisms. In order to
reduce the exposure of PCBs, it is therefore
imperative to remove the source from the
impacted water body and either isolate or degrade
the contaminant.
Since the early to mid-1970s, it has been
recognized by the regulatory agencies and others
that the lower chlorinated PCBs are soluble and ----
volatile and consequently, select compounds will
readily partition to the aqueous environment and
to the atmosphere. Despite this recognition,
remedial technologies that are invariably
contributing to the global loading and
redistribution of PCBs and other similar,
semivolatile compounds are being used and
aggressively promoted. The major remedial
challenge, therefore, is to develop cost effective
----
treatment processes to degrade the PCBs
associated with the millions of cubic yards of
contaminated soils and sediments that need to be
removed from the aquatic environment while
minimizing volatile losses. Containment
alternatives being considered for this class of
material include pre-treatment involving the
mixing of cementing agents with the contaminated
solids or vitrification. Stabilized or solidified
contaminated solids will require some form of
environmental isolation and long term monitoring
and maintenance to ensure the contaminants
continue to remain isolated and immobile.
Containment management options are currently
favored over processes involving destruction of
the contaminants because of the costs required to
destroy or effectively isolate and remove the
PCBs from the solid particles. Effective
containment will not only require continuous
maintenance, but will need to be maintained to
ensure the engineered system continues to prevent
offsite migration as well as guard against water
----
intrusion. Without water, however, the
contaminants will remain virtually unchanged for
decades possibly centuries. Although cost
competitive treatment options are not presently
available, engineered containment technologies
should be considered temporary solutions.
Containment facilities wi11 require additional
remedial attention at some time in the future and
should be designed and constructed with this
recognition. Designs of containment structures
should incorporate features that will enable
effective recovery of the contaminated sediments
when advanced, cost effective degradation
technologies become available.
Bench scale research results indicate PCB
volatilization is directly correlated with
evaporative water losses. Contaminated
sediments subjected to periodic wetting and
drying, including inter-tidal fluctuations and
periodic flooding, may well be contributing to the
volatile loss of PCBs. Dewatering and drying
during the dredging and handling of contaminated
sediments may be contributing to the overall
global loading of PCBs. Technologies that
involve low temperature extraction and/or
exothermic reactions that elevate the temperature
of the sediments that promote the evaporative loss
of water, result in the volatile loss of PCBs.
Aerobic microbial degradation of PCB-
contaminated sediments that involves mixing,
aeration and watering may also be contributing to
the global redistribution of PCBs.
It is evident that the global redistribution of PCBs
and similar semi-volatile compounds result from
the volatile losses and atmospheric transport of
this class of compounds. To reduce continued
exposure to the aquatic environment, the sources
must be removed and isolated from the impacted
water bodies. However, currently employed and
proposed remedial technologies that involve
dewatering and drying of contaminated sediments
may contribute to the loading of PCBs and other
semivolatile compounds to the global
environment. To ensure against this possibility,
--
---
-----
effective monitoring programs should be designed
and implemented at remediation sites involving
dewatering, drying, thermal processing and
chemical and/or microbial degradation of PCB-
contaminated sediments.
References Cited
Arcaro, K. F. and J. F. Gierthy, 1997, Estrogenic
responses of human breast cancer cells to
combination of hydroxylated PCBs, pesticides and
l 7b-estradiol. The Toxicologist, Abst. No.
1499,39:295.
Bidleman, T. F. and C. E. Olney, 1974,
Chlorinated hydrocarbons in the Sargasso Sea
atmosphere and surface water. Science, 183 :516-
518.
Blais, J.M., D. W. Schindler, C. G. Muir, L. E.
Kimpe, D. B. Donald and B. Rosenberg, 1998,
Accumulation of persistent organochlorine
---
compounds in mountains of western Canada.
Nature, 395:585-588.
Bush, B., K. Simpson, L. Shane and R. Koblintza,
1995, PCB congener analysis of water and
Caddisfly larvae (lnsecta:Trichoptera) in the
Upper Hudson River by glass capillary
chromatography. Bull. Environ. Contam. Toxicol.
34:96-105.
Carpenter, D. 0 ., 1997, New dimensions in our
understanding of the human health effects of
environmental pollutants. In Proceedings of the
1996 Pacific Basin Conf. On Hazardous Waste,
Kuala Lumpur, Malaysia. CONF.-961 I 157:404-
418.
Chiarenzelli, J. R., 1998, personal
communication.
Chiarenzelli, J. R., R. J. Scrudato, K. Jensen, T.
Maloney, M. Wwiderlich, J. Pagano and J.
Schneider, 1998a, Alteration of Aroclor 1248 in
foundry waste by volatilization? Water. Air and
Soil Pollution, l 04: 112-124.
Chiarcnzelli, J. R., R. J. Scrudato, B. Bush, D.
Carpenter and S. Bushart, 1998b, Do large-scale
remedial and dredging events have the potential to
release significant amounts of semivolatile
compounds to the atmosphere? Environmental
Health Perspective, 106(2):47-49.
Chiarenzelli, J. R., R. J. Scrudato and M. L.
Wunderlich, 1997a, Volatile loss of PCB Aroclors
from subaqueous sand ES&T, 31(2):597-602.
Chiarenzelli, J. R., R. J. Scrudato, M. L.
Wunderlich, G. N. Oenga and 0. P. Lashko,
1997b, PCB volatile loss and the moisture content
of sediment during drying. Chemosphere,
34( 11 ):2429-2436.
Constant, W. D., L. J. Thibideaux and P.
Erickson, 1995, Volatilization of polychlorinated
biphenyls from soil following application of
quicklime. American Environmental Lab. 5195: l,
7-11.
Durfee, R. L., G. Contos, F. C. Whitmore, J. D.
Barden, E. E. Hackman, and R. A. Westin, 1976,
PCBs in the United States Industrial Use and
Environmental Distribution. USEPA Technical
Report, EPA 560/6-76-005, Versar, Inc.,
Springfield, VA.
Fuller, B., J. Gordan and M. Kornreich, 1976,
Environmental assessment of PCBs in the
atmosphere. USEP A Technical Report, EPA
450/3-77-045, MITRE Corp., McLean, VA.
Hamons, F. L., D. Bibo, M. Hart, W. Young, C.
Donovan, M. Vargo and L. Walsh, 1997, Hart-
Miller Island:from remnant island through
containment facility to park and natural resources
area. In Proceedings: International Workshop on
Dredged Material Beneficial Uses, M. C. Landin,
ed., Baltimore.
Haque. R., D. W. Schmedding and V. H. Freed,
1974, Aqueous solubility, adsorption and vapor
behavior of polychlorinated biphenyl Aroclor
1254. £S&T8(2):l39-142.
Harvey, G. R. and W. G. Steinhauer, 1974,
Atmospheric transport of polychlorinated
biphenyls to the North Atlantic, Atmospheric
Environment, 8:777-782.
Hatzinger, P ., and M. Alexander, 1995, Effects of
aging of chemicals in soil on their
biodegradability
and extractability. ES&T 29(2):537-545.
Muir, D. C. G., A. Omelchenko, N. P. Grift, D. A.
Savoy, W. L. Lockhart, P. Wilkinson and G. J.
Brunskill, 1996, Spatial trends and historical
deposition of polychlorinated biphenyls in
Canadian midlatitude and Arctic lake sediments.
ES&T. 30(12):3609-3617.
Pagano, J. J., R. Scrudato, G. M. Sumner, 1997,
Utilization of anaerobic bioreactor and sampling
systems for long-term biorernediation research.
In: Proceedings I 1" International Symposium of
Chlorinated Dioxins and Related Compounds,
Indianapolis, (31 ):420-424.
Pagano, J. J., R. J. Scrudato, R. N. Roberts and J.
C. Bemis, 1995, Reductive dechlorination of
PCB-contaminated sediments in an anaerobic
bioreactor system,. ES&T., 29:2584-2589.
Rhee, Y ., 1998, personal communication.
Ruell, P., M. Buchmann, H. Nam-Tran and U. W.
Kesselring, 1993, Application of the mobile order
theory to the prediction of aqueous solubility of
chlorinated benzenes and biphenyls. ES&T,
27(2):266-270.
Seegal, R. F., J. F. Gierthy, K. F. Arcaro and K.
0. Brosch, 1997, Neurochernical and
neuroendocrine effects on non-planar (NCP) and
coplanar (CP) PCBs. Toxicologist, 36:332.
Sokol, R., C. M. Bethony and Y. Rhee, 1998,
Effect of Aroclor 1248 concentrations and rate
and extent of polychlorinated biphenyl
dechlorination. Environ. Toxic. and Chem.,
17: 1922-1926.
Tenenbaum, D., 1998, Northern overexposure.
Environ. Health Persp., 106(2):A64-69.
Wang, Y., C-S. Hong and B. Bush, 1998,
Decomposition of 2-chlorobiphenyl in aqueous
solutions by UV irradiation with the presence of
titanium dioxide, J. Adv. Oxids. Technol., (in
press)
Wania, F. and D. Mackay, 1993, Global
fractionation and cold condensation of low
volatility organochlorine compounds in polar
regions. Ambio. 22(1 ): l 0-18.
Wania, F. and D. Mackay, 1996, Tracking the
distribution of persistent organic pollutants.
ES&T, 30(9):390-396.
Wania, F., J. Haugen, Y. D. Lei and D. Mackay,
1996, Temperature dependence of atmospheric
concentrations of semivolatile organic
compounds. ES& T, 32(8): 1013-lo2 I.
. ' ' .
Author Background
The authors are affiliated with the Environmental
Research Center (ERC) of the State University of
New York, Oswego. R. Scrudato received his
Ph.D. in geology from the University of North
Carolina, Chapel Hill. In 1977 he became director
of tfi.e ERC and has been involved with inactive ·
hazardous waste site remediation and remedial
technology development for more than 20 years.
He has been a member of the New York State
Superfund Management Board since 1986 and is
on the Board of Directors of the New York Sea
Grant Institute. In 1989, he was appointed by the
governor of New York to the Great Lakes
Protection Fund and also serves on the Board of
the Environmental Advocates. He has served as a
technical advisor on three Technical Assistant
Grants (TAGs) on NPL sites in upstate New York.
During the past 25 years, Scrudato has worked on
environmental research projects in West Virginia,
Texas, Lake Ontario, the Adirondacks, Tierra del
Fuego, Argentina, Chile, Paraguay, Uruguay and
Brazil.
List of Exhibits
Exhibit l. The PCB molecule illustrating the
dichlorobiphenyl and chlorine bonding at the
ortho, meta and para positions on the biphenyl.
Exnibit 2. PCB solubility relative to gas
chromatographic retention time (chlorine content).
Exhibit 3. Anaerobic dechlorination of Aroclor
1248-contaminated sediments during a 124 week
experiment conducted in a recirculating, upflow
reactor maintained at 35°C. Note the relatively
rapid reduction in the chlorine per biphenyl
concentration occurred during the initial 20 to 25
week period followed by a plateau which
persisted for the remaining 100+ weeks.
Exhibit 4. Evaporative loss of PCB-contaminated
sediments at various initial moisture content.
Polychlorinated Biphenyls (PCBs)
dlchloroblphenyl
Meta Ortho Meta
Para -0-0-Para
Meta Ortho Meta
• 4 • , PCB Solubility in H20 (ppm)
1 ()() Solubilty HtifflllH lroffl:
0.01
0.0001
1E-06
Ruelle II 11. (1993)
Increasing -
Chlorination • • 1 E-08 ~__._ _ _.__..____,...___.
0 10 20 30 40 50
Retention Tme (min.)
t·
" ' I •
4.3
>. 4.1 • ~
.c:1 3.9 Cl. e 3.7 ~ = ·c J.s
0 :a U 3.3
t, 3.1 OIi ... ~ 2.9 < 2.7
2.5
[-<>-SF 4 ■ SF 4 (sediment) -Ir-SF 5 -6-Aroclor 12481
A A A A A
i i
0 6 12 18 24 30 36 42 48 54 60 66 72 78 84 90 96 102 108 114 120
Weeks
16-.--------.....-------------.------~-----
--■-1242 -•--1248 ··•· 1254 ---8--· 1260 --0--StLR
~10-t-----+-----~.......:::...._=--..:q.-----+-----~
-~ I s-+------lf---~~"---+------1-------+-----__.
~ ... ------·-~ J? 6 u --------·-· .... --------------0 ~ 4+--7"=';,,C---,f------+----=---=i-==----=-·~-=--···~----~
., .. -~_:::_:..----·-·-£1 ____ __,::::::. -----------····-··•· ----·•···•-···-···-·· 2-+--e----£;;~!,;==------+--_-_-_____ -,, __ =--i--=:=.:;_---+------..j
-#-· --·-
0-............ --.----.------....... -.-...... _____ i--____ ~------'
0 5 10 15 20 25
Evaporation Time (hours)
· _,olid Waste Management
~rrund Section
.,a~l"dous Waste Section
.-l 'solid Waste Section
Organics Lab: )('.' ---Inorganics Lab: --
CHAIN OF CUSTODY RECORD
Project Name: Pc 6 ~J_J Sampled by: L+K~.~lu,,.;t.~
Site ID # (NCD#) Sampler ID ;;,-9, '
Location: w~ ~ . Telephone: Cia._) 2 J j -:;.._ I ]~
Address: a.ff' ~ ,j_~ :jO I,., 2 Date Sampled: 3L~ 1 i.1.'i. l Time Sampled: ;
<1J : Jo -//:Jo
Sample Types: Soil Water Waste Other _\,
Remarks: /.JV\ f,~
Field Sample ~ ~£? $ J
Numbers LI . ':4
c;).tt.). ~5
Relinquished by: -~---~ v.-~~ '(·,:--Time: --, .. Date: • ,;;':.-' ·---
(Signature) C ,,,:;...,,,--
. "
Received by:---· ?Ji;~ Q }{(b.:?h: ·--------------Date:3-2D-9t'--Time:---/!) ·_,., 3 S----------
(Signature) '
Relinquished by: Date: Time:
(Signature)
--------------------------------------------------------------------------------------------------------------------Received by: Date: Time:
(Signature)
Relinquished by: Date: Time:
(Signature)
---------------------------------------------------------------------------------------------------------------------------Received by: Date: Time:
(Signature)
Results Reported: \ _ _£ {? N Date: V/1/ V/4 ·•
Time:
;:.,..._J-.\ (S1J!>lj4HU•..,,
, (
V
"CR.FRM)
N.C. Dept. of Environment, Health and Natural Resources
Division of Laboratory Services
State Laboratory of Public Health
... P.O. Box 28047, Raleigh, N.C. 27611
Environmental Sciences Analysis Report
Name of Owner, Patient ~J.? 'nj:
or Supply: f C... r3 ~, JX
Address: (,JU~, NC
County:
Report Io: f~ ~
Address: "101 0~ JO., 5~ 15°
·R~k N~ ~-,c 0
~·
Laboratory Sample Sample Description or Remarks
Number Number
981019 016 7'ii 5 F~· ~y f' I.A( -:;; 1-$"
(!) ;J J,,,..
981020 0/C?8Ll • I (9 :J/25'
981021 Olt7S5 r-J
Results In
·; ell" P. F
/V,[), JV./).
/V. o. ;V. /).
II/,() ;V.[).
Date Recei ~-1.:i Z-f/2_
llate Extra,-.~~ l/-3-9';( Y5JJv'
Date Reported ,,,..,...--;,c:--=--.1..~--__,_~~-h=~~
DEHriR For:-rn 2364 l{ei;ised (S-89)
hii06ralo-rr
-:,...-,e;L-.<~----L..;!:.,,__--4-C-~~-</:..,, B D
, r
tar Organic Chemistry
3-22-1 S:➔SJ7 3: 2 8At-1 Fl?Ot1
October 21, 1997 .................................................................................... by FJ\..,X -
To: Mike Kelly
From: Joel Hirschhorn
Re: Methane air monitoring at PCB Landfill
In the overnight package received today there were two memo reports on methane testing, one
for sampling in Jan_/Feb. and the second for sampling in early October. According to the first
repon~ there was to be monthly testing.
If there is data for the period between these times, would you please provide me with those
internal reports.
SEP-03-97 13 ,34 FROM ,REGION 4 PESTICIDES ID ,404 562 8973
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
REGJON4
Science and Ecosystem Support Division
980 College Station Road
Athens, Georgia ~2720
August 22, 1997
MEMORANDUM
SUBJECT: Results of Air Study at Warren County Landfill, North Carolina
Project Number 97-0345
TO:
FROM:
/
/
Beverly Hudson ,//
Remedial Project Man~
/24P Tim Slagle t7,,,, /·?;t1_
Air Team ,;JI
/ Jon Vail ~ 4~,z_
/"'°'---Supert~tim
PAGE
We have attached the final copy of the results of the August 5 and 6, 1997 air
study that SESD conducted at Warren County Landfill. If you have any questions
please feel free to give Tim at [phone], or Jon at (706) 355-$611 a call.
cc: Steve Hall
Bill Bokey
Archie Lee
Craig Brown
/
Ol'TIONAJ. FORM 99 (7--90)
FAX TRANSMITTAL 7
Fax II
GENER.OJ.. SERVICES AOM!N!STRA TION
INTRODUCTION
1u ,q111q bb:.<' t::S::C,1 ..:l
AIR STUDY
WARREN COUNTY LANDFILL
NORTH CAROLINA
AUGUST 1997
On August 5 and 6, 1997, Tim Slagle and Jon Vail of the United States
Environmental Protection Agency (EPA), Region 4, Science and Ecosystem
Support Division (SESD) conducted an air quality study at Warren County Landfill,
North Carolina. The study was requested by Waste Management Division, North
Remedial Branch, to determine if PCBs {polychlorinated bi-phenyls) were being
emitted into the ambient air from the landfill.
STUDY AREA
The six air monitoring sites utilized during the study were located within the
fenced boundaries of the landfill (Figure 1) for security purposes. Two sites,
designated A and B, were established adjacent to the vent pipe at the center of
the landfill. Meteorological data consisting of wind speed and direction was
recorded at this sampling location for the duration of the air study. Meteorological
data was recorded at the top of the land fill because the surrounding terrain was
relatively free of obstructions. Site C was located at the northwest corner of the
landfill. Site D was located on the west side between sites E and C. Site E was
located at the southwest corner of the landfill. Site F was located near the
southeast corner of the landfill. The upwind measurements at Site C were used to
determine if any of the pollutants being measured were transported into the area
from other sources.
SAMPLING PROCEDURES
.
The Air PCB samples were collected by the high volume PUF/XAD method. The
sampling methodology conformed to T0-4 of the Compendium of Methods for the
Determination of Toxic Organic Compounds in Ambient Air. The high volume
sampler utilizes a glass fiber prefilter with polyurethane foam {PUF) and XAD
absorbent cartridge for collection of the compounds in ambient air. Approximately
210 M 3 of air was sampled during the 24-hour sampling period.
1
INTRODUCTION
AIR STUDY
WARREN COUNTY LANDFILL
NORTH CAROLINA
AUGUST 1997
On August 5 and 6, 1997, Tim Slagle and Jon Vail of the United States
Environmental Protection Agency (EPA}, Region 4, Science and Ecosystem
Support Division (SESD) conducted an air quality study at Warren County Landfill,
North Carolina. The study was requested by Waste Management Division, North
Remedial Branch, to determine if PCBs (polychlorinated bi-phenyls) were being
emitted into the ambient air from the landfill.
STUDY AREA
The six air monitoring sites utilized during the study were located within the
fenced boundaries of the landfill (Figure 1) for security purposes. Two sites,
designated A and B, were established adjacent to the vent pipe at the center of
the landfill. Meteorological data consisting of wind speed and direction was
recorded at this sampling location for the duration of the air study. Meteorological
data was recorded at the top of the land fill because the surrounding terrain was
relatively free of obstructions. Site C was located at the northwest corner of the
landfill. Site D was located on the west side between sites E and C. Site E was
located at the southwest corner of the landfill. Site F was located near the
southeast corner of the landfill. The upwind measurements at Site C were used to
determine if any of the pollutants being measured were transported into the area
from other sources.
SAMPLING PROCEDURES
The Air PCB samples were collected by the high volume PUF!XAD method. The
sampling methodology conformed to T0-4 of the Compendium of Methods for the
Determination of Toxic Organic Compounds in Ambient Air. The high volume
sampler utilizes a glass fiber prefilter with polyurethane foam (PUF) and XAD
absorbent cartridge for collection of the compounds in ambient air. Approximately
210 M3 of air was sampled during the 24-hour sampling period.
1
SEP-03-97 13 ,35 FROM ,REGION 4 PESTICIDES ID ,404 562 8973 PAGE
SAMPLE CUSTODY
At the start of the sampling event, an initial flow rate and an initial elapsed time
reading was recorded from each high volume sampler internal clock. The samplers
were then sealed with custody seals. During the 24-hour sampling event a guard
was posted and the samplers were checked periodically. At the conclusion of the
sampling event, the integrity of the custody seals was verified, .and a final flow
rate and elapsed time were recorded. The sampling cartridges were then returned
to individual sampling jars and sealed with custody tape.
QUALITY ASSURANCE
Co-located duplicate sets of PUF/XAD samplers were operated at sites A and 8-
They were ,used to show the precision of the monitoring method.
A field blank PUF/XAD cartridge was carried to the field but not exposed. It was
analyzed to document lack of contamination from field transport. Initial and final
flowrates were recorded for the samplers to verify flowrate stability.
ANALYTICAL PROCEDURES
The samples were analyzed by the EPA, Region 4, SESD laboratory in Athens,
Georgia, in accordance with the Analytical Support Branch Laboratory Operations
and Quality Control Manual, September 1990.
The media designed to collect PCBs in air was analyzed by the T0-4 procedure. A
surrogate mixture was added to each PUF/XAD cartridge before extraction. The
particulate filter and PUF/XAD cartridge was Soxhlet extracted for approximately
24 hours using a 5% ether in hexane solution. The sample extract was analyzed
for the target analytes listed in Table 1 by using a Gas Chromatograph equipped
with an Electron Capture Detector.
AIR RESULTS
No PCBs in ambient air at the landfill were detected at sub-microgram per cubic
meter concentrations. The air sample analysis data is attached as Table 2. The
prevailing wind was from the north northwest with an average speed of 2. 7 miles
per hour. Meteorological data is attached as Appendix 1.
2
...
~ ~ ::;
<:-, ;;,
r .. -..
•c
1u ,qioq bb 2 8973
-... -.. --· ~ _ ........... --·
-.. -.. ----··-··-··J.
0 50 i00 175 200
SCALE
(feet)
FALJ::: Lj
400
-"'l..L..L--------------------------~---------~
~EPA
FIGURE 1
WARREN COUNTY LANDFILL
WARREN COUN1Y, NORTH CAROLINA
AUGUST 6, 1997
LEGEND:
Ae S,AJ~PLER LOCATION
-. MET STATION
9 VENTPIPE
----------··--·---··-··~··-,
I
@
es eA
.
; I
l :
._ •• -•• -•• -•• -•• -•• -•• -•• .l
.. ~ 0 50 100 175 200 400
::: ,, SCALE
i ~~ ~-------=~-
6zEPA
FIGURE 1
WARREN COUNTY LANDFILL
WARREN COUNlY, NORTH CAROLINA
AUGUST 6, 1997
LEGEND:
Ae SArv'IPLER LOCATION = MET STATION
8 VENTPIPE
Site location Analyte PCB-1016 (AROCLOR 1016) PCB-1221 (AROCLOR 1221) PCB-1232 (AROCLOR 1232) PCB-1242 (AROCLOR 1242) PCB-1248 (AROCLOR 1248) PCB-1254 (AROCLOR 1254) PCB-1260 (AROCLOR 1260) AIR VOLUME (M3) A ug/M3 0.00341 U 0.01136 U 0.00341 U 0.00341 U 0.00341 U 0.00341 U 0.00341 U 220 TABLE 2 Air PCB Results Warren County PCB Landfill Warren County, North Carolina 8 C ug/M3 ug/M3 0.00342 U 0.00350 U 0.01142 U 0.01168 U 0.00342 U 0.00350 U 0.00342 U 0.00350 U 0.00342 U 0.00350 U 0.00342 U 0.00350 U 0.00342 U 0.00350 U 219 214 D ug/M3 0.00346 U 0.01152 U 0.00346 U 0.00346 U 0,00346 U 0.00346 U 0.00346 U 217 U-material was analyzed for but not detected, the number is the minimum quantitation limit ug/M3 = micrograms per cubic meter of air E F ug/M3 ug/M3 0.00302 U 0.00302 U 0.01008 U 0,01008 U 0.00302 U 0.00302 U 0.00302 U 0.00302 U 0.00302 U 0.00302 U 0.00302 U 0.00302 U 0.00302 U 0.00302 U 248 248 (! tr ,: I (S r.,. I u ' .. r.,. r.,. O' 'T ~ C :3 ~ tr ' .. C :z ,I: ,: tr (! .. .. r .. C tr rJ .. C ,I: (S ,I: CJ cr II: II u " r.,. ,: J: ( tr 'T
SEP-03-97 13 ,36 FROM ,REGION 4 PESTICIDES ID ,404 562 8973
Table 1
Target PCB Analytes
PCB -1242 (AROCLOR 1242)
PCB -1254 (AROCLOR 1254)
PCB -1221 (AROCLOR 1221)
PCB -1232 (AROCLOR 1232)
PCB -1248 (AROCLOR 1248)
PCB -1260 (AROCLOR 1260)
PCB -1016 (AROCLOR 1016}
3
PAGE 5 _.,7
Site Location A ug/M3 Analyte PC8·1016 (AROCLOR 1016) 0.00341 U PCB·1221 (AROCLOR 1221) 0.01136 U PCB-1232 (AROCLOR 1232) 0.00341 U PCB-1242 (AROCLOR 1242) 0.00341 U PCB·1248 (AROCLOR 1248) 0.00341 U PCB-1254 (AROCLOR 1254) 0.00341 U PCB-1260 (AROCLOR 1260) 0.00341 U AIR VOLUME (M3) 220 TABLE 2 Al r PCB Results Warren County PCB Landfill Warron County, North Carolina 8 C ug/K,3 ug/M3 0.00342 U 0.00350 U 0.01142 U 0.01168 U 0.00342 U 0.00350 U 0.00342 U 0.00350 U 0.00342 U 0.00350 U 0.00342 U 0.00350 U 0.00342 U 0.00350 U 219 214 D ug/M3 0.00346 U 0.01152 U 0.00346 U 0.00346 U 0.00346 U 0.00346 U 0.00346 U 217 U-material was analyzed for but not detected, the number is the minimum quantitalion Hmit ug/M3 = micrograms per cubic meter of air E F ug/M3 ug/M3 0.00302 U 0.00302 U 0.01008 U 0.01008 U 0.00302 U 0.00302 U 0.00302 U 0.00302 U 0.00302 U 0.00302 U 0.00302 U 0.00302 U 0.00302 U 0.00302 U 248 248
SEP-03-97 13,36 FROM,REGION 4 PESTlGJUES
APPENDIX 1
WIND SPEED AND DIRECTION
RM YOUNG CO. TRAVERSE CITY, Ml
26700 SERIES TRANSLATOR
---------------------------------------------------------------------DATE DATE TIME TJ:ME WS:.AVG WS:MAX l'f"D:AVG WD:SDV T:AVG T:MIN T:MAX
MON DAY HR MIN MPH MPH DE:G DEG DEG F DEG F DEG F
---------------------------------------------------------------------
.s s 14 0 3.S 12 354 48********.************
8 s 15 0 2.5 10 21 64*********************
s 5 16 0 2.0 10 15 59********•-~T**~******
8 s 17 0 5.6 24 293 85*****************WWWT
8 5 J.8 0 2.5 10 62 90*********************
s 5 19 0 1.5 5 296 65*********************
8 5 20 0 2.6 7 332 41*********************
8 5 21 0 1.1 3 312 67****************•****
" 5 22 0 0.9 3 351 77********************* ~
8 5 23 0 2 .9 7 324 20****~~*-~************
0 6 0 0 3.l 7 336 16********************* ..,
3 6 l 0 l.O 3 153 lOO*********************
8 6 2 0 2.7 s 318 17*********************
8 6 3 0 2.S 5 335 12********•************
s 6 4 0 2.5 5 329 15******•**************
8 6 5 0 l.9 5 328 19*********************
8 6 6 0 3.5 7 337 ll********************* ,., 6 7 0 1.S 6 315 27********************* 0
s 6 8 0 2.0 6 313 29*********************
3 6 9 0 2.8 9 358 32***********~*~~-t****
8 6 10 0 2_8 8 355 30*********************
8 6 11 0 3.2 9 358 ~4*••-~*S~*************
8 € 12 0 3.9 9 338 37****************••***
8 6 l3 0 '.3-9 12 349 55********~~·•*********
8 6 14 0 4.1 13 340 43*********************
t' Ali J:::
Sep 4 '97 16:43
T.2.2 Check condition of remote Fax. [phone]
T.2.2 Check condition of remote Fax. [phone]
T.2.2 Check condition of remote Fax. [phone]
DIVISION: OF WASTE MANAGEMENT
919-733-4996,Ext.201
Fax::919-71S-l60S
FA:k TRANSMISSION COVER SHEET .i
TO: q;;/' ~_,.~ . .]OJ. 9,/9.1:,.37
/1,,.d .;If~~ -.tjl)rf~ 89t-lf~
"7;:J,77~· ~W.-<..cE -?, 9 9P/=-otf'lo
FAX: ~ $ufe&od -11 r~ ~Sz, IO<>l>
':::7/.htA7~~-1-~ -1tdf---JIJ-3C>t,()
FROM: J;rf_ 1+r-~----
COMMENTS: -------'
, . . .
TO:
FAX:
FROM:
DIVISION OF WASTE MANAGEMENT
919-733-4996, Ext. 201
Fax: [phone]
FAX TRANSl\flSSION COVER SHEET
~---'/47-8%,--lf~;;J-,
at?~, ~--« = -7,. ?/ 'N'/ c (t/o
~ 7/..u,ThJ;nd -119-;iS?-/ouo
-:::;::J/41 ~,.__o, ... -o-/-J..,\_ -'JI 1 -1 /.T-3 Dt. 0
COMMENTS: ---------------------
TOT AL NUMBER OF PAGES INCLUDING COVER SHEET: J ~-
DATE SENT: q I 1/ 17 I l
' ,
l♦I Environment
Canada
Atmospheric
Environment
Service
4905 Dufferin Street
Downsview, Ontario
M3H 5T4, Canada
Environnement
Canada
Service
de l'environnement
atmospherique
Mr. William L. Meyer, Director
Division of Solid Waste Management
Department of Environment, Health
and Natural Resources, State of North Carolina
P.O. Box 27687
Raleigh, North Carolina 27611-7687
Dear Dr. Meyer:
Your file
Our File
CANADA'S GREEN PLAN
LE PLAN VERT DU CANADA
Votre reference
Notre reference
February 11, 1997
In early December you sent me a package of information about PCB air monitoring
at the Warren County landfill .and asked for my review. I apologize for taking so long to
reply, and hope that the comments below are not too late to be of value. My background
in PCBs goes back to the early 1970s and my research group has published several articles
on aspects of their air sampling techniques, analysis, physicochemical properties and
atmospheric behavior. A short CV is attached.
I've known Bob Lewis for many years. Throughout the 1980s his branch, AREAL,
supported much of our developmental work on air sampling methods for PCBs and
pesticides while I was at the University of South Carolina. Bob was one of the first
individuals to use polyurethane foam (PUF) as a sampling medium for PCBs and later
developed a "sandwich" cartridge of PUF and granular sorbents to collect more volatile
compounds. He also recognized, very early, that semivolatile organics are speciated
between the particle and gas phases in the atmosphere and that glass fiber filters alone
were inadequate for collecting these compounds from air. In short, Bob is one of the leading
experts in the area of sampling methods for PCBs and other organics. In years of
interaction with Bob, I have found him to be totally honest and straightforward in his
dealings.
There are several issues raised by Joel Hirschhorn about the study carried out at the
Warren County landfill and reported by Lewis et al. in Environmental Science and
Technology (1985). In essence, Mr. Hirschhorn has accused EPA of mis-representing the
air concentration data collected at the landfill and for not carrying out a more detailed
investigation following the initial study.
Canada This paper contains a minimum of 50% recycled fibres,@ including 10% post-consumer fibres.
Ce papier con11en1 un minimum de 50% de fibres recyclE!es.
dont 10% de hbres recyclE!es apres consommation
.,
To put the problem into perspective, it is helpful to consider the origins of PCBs in
ambient air. Present-day sources of PCBs to the atmosphere include volatilization of PCBs
from landfills, leakage of electrical equipment, and "recycling" from contaminated water, soil
and vegetation . Very few measurements have been made on PCB release from landfills.
Recently published mass balance estimates for PCBs in the environment (Harrad et al.,
1994; Johnson et al., 1992) quote only two papers, the one by Lewis et al. (1985) and
another by Murphy et al. (1985) on emissions of PCBs from sanitary landfills in the Chicago
area. The database on emissions from the other sources is also not large, but from the
available information release of PCBs from large areas of land and water is substantial.
From a long-term inventory of PCBs in the water column of Lake Superior, Jeremiason et
al. (1994) estimated that 26,500 kg have been volatilized from the lake over a twelve-year
period, or an average of 221 O kg/y. Net volatilization of PCBs from lakes Superior, Michigan
(Hornbuckle et al., 1994, 1995) and Ontario (Hoff et al., 1996) is also implied by fugacities
(partial pressures) of PCBs in surface water which exceed those in the atmosphere. By
comparison, a report to Ontario Ministry of Environment and Energy (Johnson et al., 1992)
estimated that only 1000 kg/y of PCBs was emitted into the air within all of eastern North
America by leaking transformers (440 kg/y), stationary fuel combustion (450 kg/y), landfills
(69 kg/y) and incinerators (42 kg/y).
Harrad et al. (1994) estimated that 93% of the contemporary environmental burden
of PCBs in the U.K. is held in soil. Their inventory of current PCB sources to the
atmosphere were: volatilization from soils = 88%, transformer and capacitory leaks = 9%,
refuse-derived fuel= 2% and other industrial sources= 1 %. A model of PCB loss from soils
in the U. K. indicated that concentrations peaked in the late 1960s and have since declined
with a half-time of 10-20 years (Harner et al., 1995). Jones et al. (1995) estimated that
25,000 tonnes of PCBs have been lost from U.K. soils between 1970 and 1990. Clearly re-
emission of PCBs from contaminated water bodies and soil is large compared to releases
from landfills and other point sources.
I have spent a lot of space on this because it is important to recognize that PCBs in
ambient air within Warren County are likely to come from a variety of sources, including
volatilization from soils. Mr. Hirschhorn makes a case about ambient levels of PCBs in four
out of thirty-nine air samples. The positive four were 11, 12, 50, and 71 ng/m 3 Aroclor 1260.
Of these, only the last two can really be considered as higher than the LOO of 1 O ng/m3.
Were these values actually due to emissions from the landfill? Maybe, and maybe not.
a) PCBs were dumped over a large area of central North Carolina. Even though
efforts were made to clean up the mess, it is likely that some PCBs migrated away
from the immediate sites of contamination by the "grasshoppering" process of
volatilization and redeposition. Indeed, air samples which were taken at a site
where Aroclor-1260 contaminated soils were excavated showed levels of PCBs after
the cleanup operation in the range of 20-90 ng/m3 (Macleod and Lewis, 1980). Is
it possible that the roadway spill sites (even after cleanup) and other soils in the
region contributed to the observed PCBs in ambient air?
b) The levels of PCBs found in ambient air in the vicinity of the Warren County
landfill were not too far above the LOO, and Lewis suggests that they may, in fact,
be "false positives". The analytical work on these samples was done in the early
1980s, by packed-column GC-ECD . PCBs are not well resolved by packed
columns, giving a sort of lumpy chromatogram such as the one displayed by Lewis
et al. (1985), Figure 3. Given the difficulty in quantifying such a chromatogram, it
is possible that other ECO-sensitive compounds, toxaphene and chlordane in
particular, contributed to apparent PCB levels. Both of these organochlorine
pesticides are multicomponent mixtures whose GC retention times overlap those of
Aroclors 1254 and 1260. They can be separated out before analysis by silica gel
chromatography, but this was apparently not done for these air samples.
In the late 1970s we did quite a lot of air sampling in Columbia, South
Carolina for PCBs and pesticides. A paper published by us (Billings and Bidleman,
1980) shows packed and capillary chromatograms of PCBs and toxaphene in
ambient air, after silicic acid separation of the two chemicals. Figures 1 and 3 of
that article clearly show the overlap of the PCB and toxaphene profiles by packed-
column GC. During these years and afterward (toxaphene was banned in 1982 and
remaining stocks were used through 1986), toxaphene was a very abundant
pesticide in southern air. Mean concentrations of toxaphene in Columbia were
about 10 times higher than those of PCBs. Chlordane was in about the same
concentration range as Aroclor 1254. In my view, it is quite likely that PCBs at the
Warren County sites were overestimated because of these analytical interferences.
These arguments show that there is considerable ambiguity about the source --and
even the identity --of the apparent PCBs in ambient air near the Warren County landfill.
On p. 4 of his first statement (Nov. 27), Mr. Hirshhorn compares the concentrations
of PCBs at the main vent of the ToSCA landfill with levels found at uncontrolled landfills:
"However, the paper focused on results from measurements at 120-180 cm for
the other three landfills, which is somewhat analogous to the main vent pipe
condition at the Warren County landfill."
Not really, because vertical profiles of PCBs change greatly with height. This can be seen
from the data in Lewis et al. (1985) Table 3, where PCB concentrations drop by a factor of
about 100 between 2 cm and 180 cm height. In his response letter, Lewis states that the
vents were sampled by placing the intake of the air sampler within the vent pipes, not in the
ambient air around the pipes. This will certainly get the biggest whiff! So if a comparison
is going to be made, it should be the in-pipe measurements at the ToSCA landfill vs . the 2-
cm height measurements at the uncontrolled landfills. In this comparison the main vent pipe
showed 120 µg/m3 (main vent) vs . 271-520 µg/m 3 (uncontrolled landfills, October).
Further down on p. 4, and also on p. 4 of his Dec. 10 statement, Mr. Hirschhorn
states that the vent pipe may have been emitting higher levels of PCBs in the summer, and
cites the summer-fall differences seen at the uncontrolled landfill site as support. I don't
. l
think this is likely. Surface emissions at the uncontrolled site will vary with temperature
according to a well-established relationship for semivolatile organics: Log C = m/T + b,
where C is the concentration in ambient air, T is the ambient temperature (Kelvin) and m is
related to the heat of vaporization of the compound. The heat of vaporization of PCBs is
such that C is expected to increase by a factor of three for each 10°C rise in temperature.
So there will be large seasonal differences in PCB levels in ambient air at the uncontrolled
landfills. However the interior of the To SCA landfill is probably much less sensitive to these
seasonal variations, for the same reason that a cellar maintains a relatively constant
temperature. Thus I wouldn't expect large summer-winter variations in the concentration of
PCBs emitted from the main vent.
Moreover, the emissions from vent pipes may become less with time. Landfills
"breathe" by production of methane and also by the in-and-out movement of air due to
changes in barometric pressure (Murphy et al., 1985). It is, of course, the "out" movement
from both processes that pumps PCBs into the air. So as decay of vegetation tapers off,
one of the processes (methane emission) will be of less consequence.
The main vent pipe was releasing PCBs out of one small point source, whereas PCBs
from the uncontrolled landfills were volatilizing over a wide area. This is evident from the
description in the Lewis et al. (1985) paper of the uncontrolled sites as being strewn with
leaking capacitors with visibly leaking askarels. Thus the emissions from the uncontrolled
landfills were surely much greater than from the ToSCA landfill. At the emission rate of
the main vent (12 ng/s), a dispersion model predicted very low downwind concentrations of
4 x 1 o-s to 1 x 10-7 ng/m3 --far lower than any of the measured ambient air concentrations.
To me, this is an indication that PCBs found in ambient air near the ToSCA site were
probably emanating from an area source, such as volatilization from contaminated soils.
Mr. Hirschhorn is speaking for a community of citizens who have a right to have their
concerns taken seriously. We know a lot more about the sources of PCBs and how to
measure them than we did in the early 1980s. Clearly, it would be beneficial to carry out
a subsequent investigation of PCBs in the region using today's greatly improved analytical
methodology. Such a study should include on-site and off-site measurements of PCBs in the
landfill vicinity, along the roadsides where PCBs were initially dumped, and in other regions
of the state for comparison. If such an investigation does indeed implicate the landfill,
corrective action should be swift. This is the best way to respond to the concerns of the
community, and I believe that Mr. Hirschhorn would agree. It is unfortunate, however, that
he seems more intent on fixing blame than getting on with it.
f?.
se rch Scientist
''
References
Billings, W.N. and T.F. Bidleman, 1980. Field comparison of polyurethane foam and Tenax-
GC resin for high volume air sampling of chlorinated hydrocarbons. Environ. Sci. Techno/.
14:, 679-683.
Harrad, S.J., A. Sewart, R.E. Alcock, R. Boumphrey, V. Burnett, R. Duarte-Davidson, C.
Halsall, G. Sanders, K.S. Waterhouse, S.R. Wild and K.C . Jones (1994). Polychlorinated
biphenyls in the British environment: sinks, sources and temporal trends. Environ. Po/fut.
85: 131-146.
Lewis, R.G., B.E. Martin, D.L. Sgontz and J.E. Howes, Jr., 1985. Measurement of fugative
atmospheric emissions of polychlorinated biphenyls from hazardous waste landfills. Environ.
Sci. Technol. 19:, 986-991 .
Macleod, K.E. and R.G. Lewis, 1980. Measurement of contamination from PCB sources.
Sampling and Analysis of Toxic Organics in the Atmosphere. ASTM Special Pub. 721, 56-
69.
Murphy, T.J., L.J. Formanski, B. Brownawell and J.A. Meyer, 1985. Polychlorinated biphenyl
emissions to the atmosphere in the Great Lakes region . Municipal landfills and incinerators.
Environ. Sci. Technol. 19:, 942-946.
Harner, T. D. Mackay and K.C. Jones, 1995. Model of the long-term exchange of PCBs
between soil and the atmosphere in the southern U.K. Environ. Sci. Technol. 29:, 1200-
1209.
Hoff, R.M., W.M.J. Strachan, C.W. Sweet, C.H. Chan, M. Shackleton, T.F. Bidleman, K.A.
Brice, D.A. Burniston, S. Cussion, D.F. Gatz, K. Harlin and W.H. Schroeder, 1996.
Atmospheric deposition of toxic chemicals to the Great Lakes: a review of data through
1994. Atmos. Environ. 30: 3505-3527.
Hornbuckle, K.C., J.D. Jeremiason, C.W. Sweet and S.J . Eisenreich, 1994. Seasonal
variations in air-water exchange of polychlorinated biphenyls in Lake Superior. Environ. Sci.
Technol. 28: 1491-1501.
Hornbuckle, K.C., C.W. Sweet, R.F. Pearson, D.L. Swackhamer and S.J. Eisenreich, 1995.
Assessing annual air-water fluxes of PCBs in Lake Michigan. Environ. Sci. Technol. 29:
869-877.
Jeremiason, J.D., K.C . Hornbuckle and S.J . Eisenreich, 1994. PCBs in Lake Superior,
1978-1992: Decreases in water concentration reflect loss by volatilization. Environ . Sci.
Technol. 28: 903-913.
. '
Johnson, N.D., M.T. Scholtz, V. Cassaday, K. Davidson and D. Ord, 1992. MOE Toxic
Chemical Emission Inventory for Ontario and Eastern North America, final report to Air
Resources Branch, Ontario Ministry of the Environment, report no. P.92-T61-5429/0G.
Jones, K.C., A.E. Johnston and S.P . McGrath, 1995. The importance of long-and short-
term air-soil exchange of organic contaminants. lnternat. J. Environ. Anal. Chem. 59: 167-
178.
. '
TERRY F. BIDLEMAN
Atmospheric Environment Service
4905 Dufferin Street, Downsview, Ontario, Canada M3H 5T 4
Phone: [phone] FAX: [phone]
EDUCATION
B.Sc. 1964. Chemistry, Ohio University, Athens, OH, U.S.A.
Ph.D. 1970. Analytical Chemistry, University of Minnesota, Minneapolis, MN, U.S.A.
Postdoctoral 1970-72. Dalhousie University, Halifax, Nova Scotia, Canada.
POSITIONS
1994-present: Adjunct Professor, Dept. of Chemical Engineering & Applied Chemistry,
University of Toronto, Toronto, Ontario, Canada .
1992-present: Research Scientist, Atmospheric Environment Service, Environment
Canada, Downsview, Ontario, Canada.
1975-92: Professor (1989-92), Associate Professor (1981-89), Assistant Professor
(1975-81 ), Department of Chemistry and Biochemistry, and Marine Science Program,
University of South Carolina, Columbia, SC, U.S.A.
1983-84: Visiting Scientist, Swedish Environmental Protection Board, Solna, Sweden.
1972-75: Research Associate, Dept. of Food Science, Technology, Nutrition and
Dietetics, University of Rhode Island, Kingston, RI, U.S.A.
PROFESSIONAL SOCIETIES
American Chemical Society (Div. of Environmental Chemistry and Agrochemicals),
Society for Environmental Toxicology and Chemistry (SETAC), International
Association for Great Lakes Research .
SELECTED PUBLICATIONS (out of 109)
Bidleman, T.F., Olney, C.E., 1974. Chlorinated hydrocarbons in the Sargasso Sea
atmosphere and surface water. Science 183: 516-518.
Bidleman, T.F., Matthews, J.L., Olney, C.E., Rice, C.P., 1978. Separation of
polychlorinated biphenyls, chlordane and p,p'-DDT from toxaphene by silicic acid column
chromatography. J. Assoc. Offic. Anal. Chem . 61: 820-828.
Billings, W.N., Bidleman, T.F., 1980. Field comparison of polyurethane foam and Tenax-
GC resin for high volume air sampling of chlorinated hydrocarbons. Environ. Sci. Technol.
14:, 679-683.
Bidleman, T.F., 1985. High volume collection of organic vapors using solid adsorbents.
In: J.F. Lawrence (ed.), Trace Analysis, Vol. 4, Academic Press, NY, 51-100.
Bidleman, T.F., 1988. Atmospheric processes: wet and dry deposition of organic
compounds are controlled by their vapor-particle partitioning . Environ. Sci. Technol. 22:
361-367.
Foreman, W.T., Bidleman, T.F., 1990. Semivolatile organic compounds in the ambient
air of Denver, Colorado. Atmos. Environ. 24A, 2405-2416.
Hinckley, D.A., Bidleman, T.F ., Foreman, W.T., Tuschall, J.R., 1990. Determination of
vapor pressures for nonpolar and semipolar organic compounds by capillary gas
chromatography. J. Chem. Eng. Data 35, 232-237.
Bidleman, T.F., Patton, G.W., Hinckley, D.A., Walla, M.D., Cotham , W.E., Hargrave, B.T.,
1990. Chlorinated pesticides and polychlorinated biphenyls in the atmosphere of the
Canadian Arctic. In: Kurtz, D.A. (ed.) Atmospheric Transport of Pesticides, Lewis
Publishers, Chelsea, Michigan, 332-387.
Falconer, R.L., Bidleman, T.F ., 1994. Vapor pressures and predicted particle/gas
distributions of polychlorinated bi phenyl congeners as functions of temperature and ortho-
chlorine substitution. Atmos. Environ. 28, 547-554.
Falconer, R.L., Bidleman, T.F., 1995. Preferential sorption of non-and mono-ortho
polychlorinated biphenyls to urban aerosols. Environ. Sci. Technol. 29: 1666-1673.
Cotham, W.E., Bidleman, T.F., 1995. Polycyclic aromatic hydrocarbons and
polychlorinated biphenyls in air at an urban and rural site near Lake Michigan. Environ.
Sci. Technol. 29 : 2782-2789.
Harner, T. and Bidleman, T. F., 1996. Measurement of octanol-air partition coefficients
for polychlorinated biphenyls. J. Chem. Eng. Data 41 : 895-899.
02 :l2 :9i 11 : 33 '5'416 i39 5i08
Environment
Canada
Atmospheric
Environment
Service
4905 Dufferin Street
Downsview, Ontario
M3H ST 4, Canada
Environnement
Canada
Service
de :·environneme:1t
atmospherique
Mr. William L. Meyer, Director
Division of Solid Waste Management
Department of Environment, Health
DOE: AES, ARIJP
and Natural Resources, State of North Carolina
P.O. Box 27687
Raleigh , North Carolina 27611-7687
Dear Dr. Meyer:
C...o\~~AfJ."\"~; l";R[[~; PL\\,;
L£ ['LA/\ \'ERT l)t_: (.-\;-.:,DA
February 11, 1997
In early December you sent me a package of information about PCB air monitoring
at the Warren County landfill . and asked for my review. I apologize for taking so long to
reply , and hope that the comments below are not too late to be of value. My background
in PCBs goes back to the early 1970s and my research group has published several articles
on aspects of their air sampling techniques, analysis, physicochemical properties and
atmospheric behavlor. A short CV is attached _
I've Known Bob Lewis for many years. Throughout the 1980s his branch, AREA.L ,
supported much of our developmental work on air sampling methods for PCBs and
pesticides while I was at the University of South Carolina. Bob was one of the first
individuals to use polyurethane foam (PUF) as a sampling medium for PCBs and later
developed a "sandwich'' cartridge of PUF and granular sorbents to collect more volatile
compounds. He also recognized , very early, that semivolatile organics are speciated
between the particle and gas phases in the atmosphere· and that glass fiber filters alone
were inadequate for collecting these compounds from air. In short, Bob is one of the leading
experts in the area of sampling methods for PCBs and other organics. In years of
interaction with Bob, I have found him to be totally honest and straightforNard in his
dealings.
There are several issues raised by Joel Hirschhorn about the study carried out at the
Warren County landfill and reported by Lewis et al. in Environmental Science and
Technology (1985). In essence, Mr. Hirschhorn has accused EPA of mis-representing the
air concentration data collected at the landfill and for not carrying out a more detailed
investigation following the initial study.
Canada 7r.\; =-~.CM: cc:,:;ln~ ~ n1i•,irr,1:r,: u• ~C~4 r::c:~·:.'~ l':r:::.@
1r1~:.:~1nJ iQ"',; :o::l·t:~ll~u•'.,'!: !.~:;~. ·
C1.:< r.,~rii .. , ~r,llftril un m1nr:,·n.::-:. ;:s~ .~.~!Ro~ !1!lr.;~. IK\:.:c: ,
.:Gm 1u,; O~ /11_,r,:,~ 10:v-::!:c::. ::.::>r:::: co.-:::::~:.1i,·.
02 :12:9i 11 :33 '6'416 i39 5i08 DOE ,, .:1.ES i ARljP
To put the problem into perspective, it is helpful to consider the origins of PCBs in
ambient air. Pr-esent-day sources of PCBs to the atmosphere include volatilization of PCBs
from landfills, leakage of electrical equipment, and "recycling" from contaminated water, soil
and vegetation. Very few measurements have been made on PCB release from landfills.
Recently published mass balance estimates for PCBs in the environment (Harrad et al.,
1994; Johnson et al., 1992) quote only two papers, the one by Lewis et al. (1985) and
another by Murphy et al. (1985) on emissions of PCBs from sanitary landfills in the Chicago
area. The database on emissions from the other sources ls also not large, but from the
available information release of PCBs from large areas of land and water is substantial.
From a long-term inventory of PCBs in the water column of Lake Superior, Jeremiason et
al. (1994) estimated that 26,500 kg have been volatilized from the lake over a twelve-year
period, or an average of 2210 kg/y. Net volatilization of PCBs from lakes Superior, Michigan
(Hornbuckle et al., 1994, 1995) and Ontario (Hoff et al., 1996) is also implied by fugacities
(partial pressures) of PCBs in surface water which_ exceed those in the atmosphere. By
comparison, a report to Ontario Ministry of Environment and Energy (Johnson et al., 1992)
estimated that only 1000 kg/y of PCBs was emitted into the air within all of eastern North
America by leaking transformers (440 kg/y), stationary fuel combustion (450 kg/y), landfills
(69 kg/y) and incinerators (42 kg/y).
Harrad et al. (1994) estimated that 93% of the contemporary environmental burden
of PCBs in the U.K. is held in soil. Their inventory of current PCB sources to the
atmosphere were: volatilization from soils = 88%, transformer and capacitory leaks ;::;; 9%,
refuse-derived fuel= 2% and other industrial sources= 1 %. A model of PCB loss from soils
in the U.K. indicated that concentrations peaked in the late 1960s and have since declined
with a half-time of 10-20 years (Harner et al., 1995). Jones et al. (1995) estimated that
25,000 tonnes of PCBs have been lost from U.K. soils between 1970 and 1990. Clear·1y re-
emission of PCBs from contaminated water bodies and soil is large compared to releases
from landfills and other point sources.
I have spent a lot of space on this because it is important to recognize that PCBs in
ambient air within Warren County are likely to come from a variety of sources, including
volatilization from soils. Mr. Hirschhorn makes a case about ambient levels of PCBs in four
out of thirty-nine air samples. The positive four were 11 , 12, 50, and 71 ng/m3 Aroclor 1260.
Of these, only the last two can really be considered as higher than the LOO of 10 ng/m3.
Were these values actually due to emissions from the landfill? Maybe, and maybe not.
a) PCBs were dumped over a large area of central North Carolina. Even though
efforts were made to clean up the mess, it is likely that some PCBs migrated away
from the immediate sites of contamination by the "grasshoppering" process of
volatilization and redeposition. Indeed, air samples which were taken at a site
where Aroclor-1260 contaminated soils were excavated showed levels of PCBs after
. the cleanup operation in the range of 20-90 ng/m3 (Macleod and Lewis, 1980). Is
it possible that the roadway spill sites (even after cleanup) and other soils in the
region contributed to the observed PCBs in ambient air?
02 :12 :9i 11 :3-1 '5'-116 i39 5i08 DOE :AES /ARG!P
b) The levels of PCBs found in ambient air in the vicinity of the Warren County
landfill were not too far above the LOO, and Lewis suggests that they may, in fact,
be "false p·ositives". The analytical work on these samples was done in the early
1980s, by packed-column GC-ECD. PCBs are not well resolved by packed
columns, giving a sort of lumpy chromatogram such as the one displayed by Lewis
et al. (1985), Figure 3. Given the difficulty in quantifying such a chromatogram, it
is possible that other ECO-sensitive compounds, toxaphene and chlordane in
particular, contributed to apparent PCB levels. Both of these organochlorine
pesticides are multicomponent mixtures whose GC retention times overlap those of
Aroclors 1254 and 1260. They can be separated out before analysis by silica gel
chromatography, but this was apparently not done for these air samples.
In the late 1970s we did quite a lot of air sampling in Columbia, South
Carolina for PCBs and pesticides. A paper published by us (Billings and Bidleman,
1980) shows packed and capillary chromatograms of PCBs and toxaphene in
ambient air, after silicic acid separation of the two chemicals. Figures 1 and 3 of
that article clearly show the overlap of the PCB and toxaphene profiles by packed-
column GC. During these years and afterward (toxaphene was banned in 1982 and
remaining stocks were used through 1986), toxaphene was a very abundant
pesticide in southern air. Mean concentrations of toxaphene in Columbia were
about 10 times higher than those of PCBs. Chlordane was in about the same
concentration range as Aroclor 1254. In my view, it is quite likely that PCBs at the
Warren County sites were overestimated because of these analytical interferences.
These arguments show that there is considerable ambiguity about the source --and
even the identity --of the apparent PCBs in ambient air near the Warren County landfill.
On p. 4 of his first statement (Nov. 27), Mr. Hirshhorn compares the concentrations
of PCBs at the main vent of the ToSCA landfill with levels found at uncontrolled landfills:
"However, the paper focused on results from measurements at 120-180 cm for
the other three landfills, which is somewhat analogous to the main vent pipe
condition at the Warren County landfill."
Not really, because vertical profiles of PCBs change greatly with height. This can be seen
from the data in Lewis et al. (1985) Table 3, where PCB concentrations drop by a factor of
about 100 between 2 cm and 180 cm height. In his response letter, Lewis states that the
vents were sampled by placing the intake of the air sampler within the vent pipes, not in the
ambient air around the pipes. This will certainly get the biggest whiff! So if a comparison
is going to be made, it should be the in-pipe measurements at the ToSCA landfill vs. the 2-
cm height measurements at the uncontrolled landfills. In this comparison the main vent pipe
showed 120 µg/m3 (main vent) vs. 271-520 µg/m3 (uncontrolled landfills, October).
Further down on p. 4, and also on p. 4 of his Dec. 1 O statement, Mr. Hirschhorn
states that the vent pipe may have been emitting higher levels of PCBs in the summer, and
cites the summer-fall differences seen at the uncontrolled landfill site as support. I don't
[4]ou-1
02 :12:97 11 :35 '5'416 7,39 5708 DOE :AES :ARQP
think this is likely. Surface emissions at the uncontrolled site will vary with temperature
according to a well-established relationship for semivolatile organics: Log C = m/T + b,
where C is the concentration in ambient air, Tis the ambient temperature (Kelvin) and m is
related to the heat of vaporization of the compound. The heat of vaporization of PCBs is
such that C is expected to increase by a factor of three for each 10°C rise in temperature.
So there will be large seasonal differences in PCB levels in ambient air at the uncontrolled
landfills. However the interior of the To SCA landfill is probably much less sensitive to these
seasonal variations, for the same reason that a cellar maintains a relatively constant
temperature. Thus I wouldn't expect large summer-winter variations in the concentration of
PCBs emitted from the main vent.
Moreover, the emissions from vent pipes may become less with time. Landfills
"breathe" by production of methane and also by the in-and-out movement of air due to
changes in barometric pressure (Murphy et al., 1985). It is, of course, the "out" movement
from both processes that pumps PCBs into the air. So as decay of vegetation tapers off,
one of the processes (methane emission) will be of less consequence.
The main vent pipe was releasing PCBs out of one small point source, whereas PCBs
from the uncontrolled landfills were volatilizing over a wide area. This is evident from the
description in the Lewis et al. (1985) paper of the uncontrolled sites as being strewn with
leaking capacitors with visibly leaking askarels. Thus the emissions from the uncontrolled
landfills were surely much greater than from the ToSCA landfill. At the emission rate of
the main vent (12 ng/s), a dispersion model predicted very low downwind concentrations of
4 x 1 o..e to 1 x 10-7 ng/mJ --far lower than any of the measured ambient air concentrations_
To me, this is an indication that PCBs found in ambient air near the ToSCA site were
probably emanating from an area source, such as volatilization from contaminated soils.
Mr. Hirschhorn is speaking for a community of citizens who have a right to have their
concerns taken seriously_ We know a lot more about the sources of PCBs and how to
measure them than we did in the early 1980s_ Clearly, it would be beneficial to carry out
a subsequent investigation of PCBs in the region using today's greatly improved analytical
methodology. Such a study should include on-site and off-site measurements of PCBs in the
landfill vicinity, along the roadsides where PCBs were initially dumped1 and in other regions
of the state for comparison_ If such an investigation does indeed implicate the landfill,
corrective action should be swift. This is the best way to respond to the concerns of the
community, and I believe that Mr. Hirschhorn would agree_ It is unfortunate, however1 that
he seems more intent on fixing blame than getting on with it.
[4)005' ,
02/12,.9i 11 :35 "B'.:116 i39 5i08 DOE .: AES /ARIW
References
Billings, W.N. and T.F. Bidleman, 1980. Field comparison of polyurethane foam and Tenax-
GC resin for high volume air sampling of chlorinated hydrocarbons. Environ. Sci. Technol.
14:, 679-683.
Harrad, S.J., A. Sewart, R.E. Alcock, R Boumphrey, V. Burnett, R. Duarte-Davidson, C.
Halsall, G. Sanders, K.S. Waterhouse, S.R. Wild and K.C . Jones (1994). Polychlorinated
biphenyls in the British environment: sinks, sources and temporal trends. Environ. Po/fut.
85: 131-146.
Lewis, R.G., B.E. Martin, D.L Sgontz and J.E. Howes, Jr., 1985. Measurement of fugative
atmospheric emissions of polychlorinated biphenyls from hazardous waste landfills. Environ.
Sci. Technol. 19:, 986-991.
Macleod, K.E. and R.G. Lewis, 1980. Measurement of contamination from PCB sources.
Sampling and Analysis of Toxic Organics in the Atmosphere. ASTM Special Pub. 721, 56-
69.
Murphy, T.J ., L.J. Formanski, B. Brownawell and J.A. Meyer, 1985. Polychlorinated biphenyl
emissions to the atmosphere in the Great Lakes region. Municlpal landfills and incinerators.
Environ . Sci. Technol. 19:, 942-946.
Harner, T. D. Mackay and K.C. Jones, 1995. Model of the long.term exchange of PCBs
between soil and the atmosphere in the southern U.K. Environ. Sci. Technol. 29:, 1200-
1209.
Hoff, R.M., W.M.J. Strachan, C.W. Sweet, C.H. Chan, M. Shackleton, T.F. Bidleman, K.A.
Brice, O.A. Burniston, S. Cussion, D.F. Gatz, K. Harlin and W.H. Schroeder, 1996.
Atmospheric depositlon of toxic chemicals to the Great Lakes: a review of data through
1994. Atmos. Environ. 30: 3505-3527.
Hornbuckle, K.C ., J.D. Jeremiason, C.W. Sweet and S.J. Eisenreich, 1994. Seasonal
variations in air-water exchange of polychlorinated biphenyls in Lake Superior. Environ . Sci.
Technol. 28: 1491-1501 .
Hornbuckle, K.C., C.W. Sweet, R.F. Pearson, D.L. Swackhamer and S.J. Eisenreich, 1995.
Assessing annual air-water fluxes of PCBs in Lake Michigan. Environ. Sci. Technol. 29:
869-877.
Jeremiason, J.D., K.C. Hornbuckle and S.J. Eisenreich, 1994. PCBs in Lake Superior,
1978-1992: Decreases in water concentration reflect loss by volatilization. Environ. Sci.
Technol. 28: 903-913.
@006
02 112 197 11 :36 '5'416 739 5708 DOE /AES .:ARQP
Johnson, N.D.1 M.T. Scholtz, V. Cassaday, K. Davidson and D. Ord, 1992. MOE Toxic
Chemical Emission Inventory for Ontario and Eastern North America, final report to Air
Resources Branch, Ontario Ministry of the Environment, report no. P.92-T61-5429/0G.
Jones, K.C., AE. Johnston and S.P. McGrath, 1995. The importance of long-and short-
term air-soil exchange of organic contaminants. lnternat. J. Environ. Anal. Chem. 59: 167-
178.
[4)00 7
t
'B'-116 i39 5 i 08 DOE: AES 1AR!jP
TERRY F. BIDLEMAN
Atmospheric Envlronment Service
4905 Dufferin Street, Downsview, Ontario, Canada M3H 5T4
Phone: [phone] FAX: [phone]
EDUCATION
B.Sc. 1964. Chemistry, Ohio University, Athens, OH , U.S.A.
Ph .D. 1970. Analytical Chemistry, University of Minnesota, Minneapolis, MN, U.S.A.
Postdoctoral 1970-72. Dalhousie University, Halifax, Nova Scotia, Canada.
POSITIONS
1994-present: Adjunct Professor, Dept. of Chemical Engineering & Applied Chemistry,
University of Toronto, Toronto, Ontario, Canada.
1992-present: Research Scientist, Atmospheric Environment Service, Environment
Canada, Downsview, Ontario, Canada.
1975-92: Professor (1989-92), Associate Professor (1981-89), Assistant Professor
(1975-81 ), Department of Chemistry and Biochemistry, and Marine Science Program,
University of South Carolina, Columbia, SC , U.S.A.
1983-84: Visiting Scientist, Swedish Environmental Protection Board, Solnal Sweden.
1972-75: Research Associate, Dept. of Food Science, Technology1 Nutrition and
Dietetics, University of Rhode Island, Kingston, RI , U.S.A.
PROFESSIONAL SOCIETIES
American Chemical Society (Div. of Environmental Chemistry and Agrochemicals),
Society for Environmental Toxicology and Chemistry (SETAC), International
Association for Great Lakes Research.
[4Joos
02 112:97 11 :36 'a'-116 739 5708 DOE :AES .:ARQP
SELECTED PUBLICATIONS (out of 109)
Bidleman, T.F., Olney, C.E., 1974. Chlorinated hydrocarbons in the Sargasso Sea
atmosphere and surface water. Science 183: 516-518.
Bidleman, T.F., Matthews, J.L., Olney, C.E., Rice, C.P., 1978. Separation of
polychlorinated biphenyls, chlordane and p,p'-DDT from toxaphene by silicic acid column
chromatography. J. Assoc. Offic. Anal. Chern . 61: 820-828.
Billings, W.N., Bidleman, T.F., 1980. Field comparison of polyurethane foam and Tenax-
GC resin for high volume air sampling of chlorinated hydrocarbons. Environ. Sci. Tech no!.
14:, 679-683.
Bidleman, T.F., 1985. High volume collection of organic vapors using solid adsorbents.
In: J.F. Lawrence (ed.), Trace Analysis, Vol. 4, Academic Press, NY, 51-100.
Bidleman, T.F., 1988. Atmospheric processes: wet and dry deposition of organic
compounds are controlled by their vapor-particle partitioning. Environ. Sci. Technol. 22:
361-367.
Foreman, W.T., Bidleman, T.F., 1990. Sernivolatile organic compounds in the ambient
air of Denver, Colorado. Atmos. Environ. 24A, 2405-2416.
Hinckley, D.A., Bidleman, T.F., Foreman, W.T., Tuschall, J.R., 1990. Determination of
vapor pressures for nanpolar and semipolar organic compounds by capillary gas
chromatography. J. Chem. Eng. Data 35, 232-237.
Bidleman, T.F., Patton, G.W., Hinckley, D.A., Walla, M.D., Cotham, W.E., Hargrave, B.T.,
1990. Chlorinated pesticides and polychlorinated biphenyls in the atmosphere of the
Canadian Arctic. In: Kurtz, D.A. (ed.) Atmospheric Transport of Pesticides, Lewis
Publishers, Chelsea, Michigan, 332-387.
Falconer, R.L., Bidleman, T.F., '1994. Vapor pressures and predicted particle/gas
distributions of polychlorinated bi phenyl congeners as functions of temperature and ortho-
chlorine substitution. Atmos. Environ. 28, 547-554.
Falconer, R.L, Bidleman, T.F., 1995. Preferential sorption of non-and mono-ortho
polychlorinated biphenyls to urban aerosols. Environ. Sci. Techno/. 29: 1666-1673.
· Cotham, W.E., Bidleman, T.F., 1995. Polycyclic aromatic hydrocarbons and
polychlorinated biphenyls in air at an urban and rural site near Lake Michigan. Environ.
Sci. Techno/. 29: 2782-2789.
Harner, T. and Bidleman, T.F., 1996. Measurement of octanol-air partition coefficients
for polychlorinated biphenyls. J. Chem. Eng. Data 41: 895-899.
141 009
I
February 11, 1997
MEMORANDUM
TO: PATRICK BARNES
JOEL HIRSCHHORN
FROM:
WORKING GROUt ~
MIKE KELLY .i~l>, .,
SUBJECT: AIR SAMPLING/ CHANGE IN MOB DATE FOR S&ME
As you are aware, we began pulling air samples yesterday at the landfill. Due to the
heavy medium we are collecting in, we are only able to get about 1 liter per minute or less
through the pumps. Due to the limitations on the pumps and the necessary re-charge time, we
will need the rest of the week to complete the 1500 liters required.
I just got off of the phone with S&ME and asked that they mobilize on Friday, and begin
work on Monday in order to allow us sufficient time to complete the sampling prior to their
breaching the liner.
We have also had some rain this week which has slowed our progress as well.
I have completed a work schedule for DWM folks to be at the landfill for the next few
weeks, using a mix of engineers, hydro's and geologists.
Remember, we have the bench scale pre-bid meeting on Monday, here at 10:00 am. I
have not heard much response from companies (actually, only 4, and one was a notice to "no bid"
this RFP. I hope some companies show up!!!
r
6 ~-" '\:-
\_Q_ ~ ~~~
~o~
~ r\ A-
... '
DATE TIME
02 .1[1 14:25
02 .10 14 :30
02 .10 14 :31
02.1[1 1 5 :45
02 .10 16:24
02 .1 1 [17 :41
[12 .11 08 :57
02 .11 09 :45
02 .11 10 :25
0 2 . 11 1 I): 34
0 2 . 11 1[1 : 39
(12 . 11
0 2. 11
I) 2 , 1 1 1 2: 1 3
0 2. 11 12 : 4 1
I) 2 , 11 12 :49
0 2 . 11 15 :08
0 2 . 11 1 5 : 3 (1
0 2 . 1 1
(12 . 1 1 16: 26
I) 2 , 1 1
0 2. 11 16:28
ACTIUITV PEP OPT
[1 2 . 1 1
( T ':-:' ··, "• I I ••
1 ,::i97 ■ •' • I
9197153605 SOLID WASTE DIU
DURATION F;EMOTE ID MODE F'A 6ES
(10 ·' 3 4 " 9197150166 ECM
97150166 63 0
0 1' 1 8 ·"·" 91'37150166 ECM
9715(11 E,E, 133
9715[1166 G3
0 1 ·'41 " 9197151)166 EC M 2
804 743 5788 EC M
01 ·' 1 0 ·"·" 74222(11 ECM
01 '01 ·'·' 97154020 ECM 2
00 ·' 24 '·" 9197150166 ECM
00' 00" 87048271313 63 0
01)' 00 ,., 87048271313 63
00 ' 00" 87048271313 63
00 _. 23 )) 919 821 (144 7 ECM
704:::271313 ECM 2
89198325980 63 0
03 -'52 .') [phone] ECM 7
0 1 ' 1 O '' 5(12 588 (1725 ECM 2
8803779:::538 ECM
00 ·' 46 )) 191 (12291)204 133
(11 '46 '' 919 :3325980 ECM 3
0(1 1 07)) 91971501t,5 ECM 0
919715(1165 ECM 0
1 4 17 358 2P::,36 ECM
8479319127 133 9
00' 4 7 '·' '31'3 715 571'3 ECM
03 ·' 1 8 ·"' 919 790 9827 ECM 4
0 0' 4 1" 84 0 7 8961822 ...-.,. G3
00' 36" 83019491237 1"'J.,;. ECM
0 1 ' 1 4" 9 1 9 2 5 7 1 (1 o 0 #t-ECM
RESULT
0 . K .
I ~HERRUPT
ItHEF.:RUF"T
ItHERRUF·T
0. f<.
0 . K.
0. f<.
O.K.
o. r:.
I tHEF;RU PT
E:usv
BUSV
0 . K.
0. K.
I tHERRUPT
O.K.
0. K .
0. K.
0. K.
0. K.
t-LG.34
t-L G. 34
0. K .
0. K .
t-L G. 39
0. K.
0, K . ··, •1
O.K.
0 . K .
Fax :'Jl'::l-733-5317
Analysis of Ambient Air Samples
1. Triangle Laboratories foe.
Research Triangle Park, North Carolina
(919)-544-5729
Client Services Contact: Debbie Hage
Technical Conlad: Gene Riley
PCBs: '"' $250
dioxins and furans ,= $1250 with a 21 day turnaround
2. ALTA Analytical Laboratory
Sacramento, CA
(9 l 6)-933 -1640
Contact: Bob Mitzd
PCBs: "' $250
dioxins and furans;: $1275
3. Huntington Analytical (formerly Maxim Technologies)
Minneapolis, Minnesota
(612)-659" 7600
Contact: Charles Suepcr
dioxins and furans ,= $900
PCBs: another IIuntington Lab would have to be contacted
4. ESE, lnc.
Gainesvil le, Florida
(904)-333-1605
Contact: Paul Oeiszkr
PCRs: ::: $200 to suppl y PUF m1d analysis
dioxins and furans: ::: $1250 (subcontracted)
5. Lancaster T Jaboratori es
Lancaster, Pcnnsy l vanin
(7! 7)-656-2301
Contact: Dick Entz
PCBs: =~ $225
no capabilities for dioxins a11d funm:s
6. Froehling and Rober son Inc.
Richmond, Vi rginia
( 804)-264-2 70 I
Contact: Grace T .aRosc
PCBs: "" $200
no capabilities for dioxins and furans
Jan 28 '97 11 :32 P·. 01/04
-.... -----------•----·-•--
METHODS FOR DETERMINATION OF TOXIC ORGANIC COMPOUNDS IN AIR EPA Methods by William T. Winberry, Jr., Norma T. Murphy Engineering Science Cary, North Carolina R.M. Riggan Battelle-Columbus Laboratories Columbus, Ohio NOYES DATA CORPORATION Park Ridge, New Jersey, U.S.A.
Winbewr, Jr. and Norma T. Murphy of Engineering-Science and R.M. Riggan of 8atLelle-Columbus Laboratories for the U.S. Environmental Protection Agencv,June 1988. Advanced composition and production method, de~elope.d b~· Nov<cs Data Corporation are employed to bring this durably bound book to ~,au in a rninirnwn ot ti•ne, Special techniciue~ are used to dose the Rap between ··rnanu,cript'' and ''cc,mpleted book." In order to keep the prke of the bool< w a reasonable level, it has b€en partially reproduced by photo-offset dir<,,etlv from the original report and tile cost rnving passed on to the reader. Due to thi~ method of publishing, certain portions of the r,ook may be less legible than desired. NOTICE The materials in this book were prepared as accounts of 1.vork sponsored by the U.S. Erwiroormmtal Protection Agency. This inforrr,ation has been subject to the Agency\ peer and administrati,1e review and h<1s been appro11ed for flublication. On this basis the Puh!isher a~sumes no responsibilit-; nor liabiiity 1or errors or any consequences arising from the u~e of the information contained herein. f'l'lention ot trade names or com-mercial products does not consfrtute endorsement or recornmenda,ion for use by the Agency or the PulJ\isher. Tt1e boo!< is ;nlended for informational purposes only. The n~ader h cautioned to obtain expert advice before implernentil,;on of any procedures described, since ma,erials ro be detected are ·toxic organic compounds and testing methods could be potentiallv hazardous. Final determtnation of ille suitability of any information or procedure for use by any user, and the manner of ttiat use, is the sole ,esponsibility of the tiser. Conl1ents and Subject Index Introduction. Method T01: Determination of Volatile Organic Compounds in Ambient Air Using Tenax@ Adsorption and Gas Chromatography/Mass Spectrometry (GC/MSI. . . . . . . . . . . . . . . . . .......... . Method T02: Determination of Volatile Organic Compounds in Ambient Air by Carbon Molecular Sieve Adsorption and Gas Chromatography/ Mass Spectrometry IGC/MS) ... Method T03: Determination of Volatile Organic Compounds in Ambient Air Using Cryogenic Preconcentration Techniques and Gas Chromatog-raphy with flame Ionization and Electron Capture Detection . Method T04: Determination of Organochlorine Pesticides and Polv-chlorinated Biphenyls ln Ambient Air . Appendix A-EPA Method 608 Method T05: DeterminatLon of .~1dehydes and Ketones in Ambient Air Using High Perfmrnm1ce Liquid Chrom<1togr<1µhy (HPLC) .. Method TOG: Determination of Phosgene in Ambient Air Using High Performance Liquid Chromatographv _ Method T07: Determination of N-Mftrosodirnethylamine in Ambien1 ALr Using Gas Chromatography .. Method TOB: De1e1·mina1ion of Phenol and Methylphennls (Creso!sl in Ambient Air Using High Performance Liquid Chromatography. vii --' '7 o., . I >< 1 • .D f-', 1.J) I --,J 7:<J • -I 4L . 7! en (.•.J f-', -,J '--1 Ill :::J 101 &; 12 13 ~ f-', f-', C►-J 15 (,.j n:,:i (J-J '•. ( ,;;) . . 1. .!::,
viii Contents and Subject Index Method T09: Determination of Polychlorinated Dibenzo-p-Dioxins (PCDDs) in Ambient Air Using High-Resolution Gas Chromatography/ High-Resolution Mass Spectrometry (HRGC/HRMS) ................ 221 Method T010: Determination of Organochlorine Pesticides in Ambient Air Using Low Volume Polyurethane Foam (PUF) Sampling with Gas Chromatography/Electron Capture Detector (GC/ECD) .............. 257 Method T011: Determination of Formaldehyde in Ambient Air Using Adsorbent Cartridge Followed by High Performance Liquid Chromatography (HPLC) .............. : .................... 294 Method T012: Determination of Non-Methane Organic Compounds (NMOC) in Ambient Air Using Cryogenic Preconcentration and Direct Flame Ionization Detection (PDFID) .......................... 332 Method T013: Determination of Benzo(a)Pyrcne [B(a)P] and Other Polynuclear Aromatic Hydrocarbons (PAHs) in Ambient Air Using Gas Chromatographic (GC) and High-Performance Liquid Chromatographic (HPLC) Analysis ............................ 370 Method T014: Determination of Volatile Organic Compounds (VOCs) in Ambient Air Using SUMMA® Passivated Canister Sampling and Gas Chromatographic Analysis ............................ 467 Appendix A-Availability of Audit Cylinders from United States Environmental Protection Agency (USE~A) Programs/Regional Offices, State and Local Agencies and Their Contractors .......... 561 Appendix B-Operating Procedures for a Portable Gas Chromatograph Equipped with a Photoionization Detector ................... 563 Appendix C-lnstallation and Operation Procedures for U.S. Environmental Protection Agency's Urban Air Toxic Pollutant Program Sampler .................................... 576 INTRODUCTION This Compendium has been prepared to provide regional, state, and local environmental regulatory agencies, as well as other interested parties, with specific guidance on the determination of selected toxic organic compounds in ambient air. Recently, a Technical Assistance Document (TAD) was published which provided guidance to such persons (1 ). Based on the comments received concerning the TAO, the decision was made to begin preparation of a Compendium which would provide specific sampling and analysis procedures, in a standard-ized format, for selected toxic organic compounds. The current Compendium consists of fourteen procedures which are consid-ered to be of primary importance in current toxic organic monitoring efforts. AJditional methods will be placed in the Compendium from time to time, as such methods become available. The original methods were selecten to cover as many compounds as possible (i.e., multiple analyte methods were selected). The additional methods are targeted toward specific compounds, or small groups of compounds which, for various technical reasons, cannot be determined by the more general methods. Each of the methods writeups is self contained (including pertinent liter-ature citations) and can be used independent of the remaining portions of the Compendium. To the extent possible the American Society for Testing and Materials (ASTM) standardized format has been used, since most potential users are familiar with that format. Each method has been Identified with a revision number and date, since modifications to the methods may be required in the future. Nearly all the methods writeups have some flexibility in the procedure. Consequently, it is the user's responsibility to prepare certain standard operating procedures (SOPs) to be employed in that particular laboratory. Each method indicates those operations for which SOPs are required. Table 1 summarizes the methods currently In the Compendium. As shown in Table 1 the first three methods are directed toward volatile nonpolar compounds. The user should review the procedures as wel 1 as the background material provided in the TAO (1) before deciding which of these methods best meets the requirements of the specific task. Table 2 presents a partial 1 isting of toxic organic compounds which can be determined using the current set of methods in the Compendium. Additional compounds may be determined by these methods, but the user must carefully evaluate the applicability of the method before use. ( 1) Riggin, R. M., "Technical Assistance Document for Sampling and Analysis of Toxic Organic Compounds In Ambient Air", EPA-600/4-83-027, u. S. Environmental Protection Agency, Research Triangle Park, North Carolina, 1983.
,,
J State of North Carolina
~ , , · r Department of Environment, 1h1Y-,c Health and Natural Resources
f~--t, , k Division of Air Quality
(i-"1\ _ James B. Hunt, Jr., Governor j_ ~(;. I 1
-Jonathan B. Howes, Secretary
"J'v· Alan W. Klimek, P.E., Director xl;u MEMORANDUM January 27, 1997
TO:
THROUGH:
FROM:
Lee Daniel, Chief
Technical Service_s ~e,,,9ion
Alan Klimek, P.E.w
Laura S. Butler, P.E., Chief~
Air Permits Section
SUBJECT: PCB Air Emissions
Air Sampling at PCB Landfill
(Mike Kelly fax memorandum -
January 24, 1997: Joel Hirschhorn Recommendations)
Warren County
This memorandum is a follow-up to my voice mail messages to you on January 24-
and 27, 1997 in which I requested the response of the Technical Service Section to Mike
Kelly's questions concerning Joel Hirschhorn's (PCB Working Group Science Advisor)
recommendations regarding the State's plan for testing the Warren County PCB Landfill
for PCB Air Emissions. Another copy of Mike Kelly's memorandum is attached for your
convenience.
Mike Kelly's questions concern level of detection and the volume of air samples to
be taken. Your earliest response and assistance is requested. Please contact Mike Kelly
at [phone], extension 203. Your written response is also requested.
I have also copied Mike Kelly's memorandum to Jim Roller, of the Air Quality
Analysis Branch (AQAB). I am requesting that Jim contact Mike Kelly to determine if
AQAB assistance is needed for collecting accurate, complete, and reliable meteorological
data (Hirschhorn recommendation #9).
LSB:bw
c: Bill Pate
Bill Meyer
George Murray
Ernie Fuller
Jim Roller
Attachment
P.O. Box 29580, Raleigh, North Carolina 27626-0580
Voice [phone] FAX [phone]
An Equal Opportunity/Affirmative Action Employer
50% recycles/10% post-consumer paper
I ,
iROM 9197153605 SOLID WASTE DIU 01.24.1997 10:59 p. 1
~AX MEMORANDUM
J+uary 24> 19 7 '.:~_~,. rn® -~--~ ~~
; \Ii~<, ;-·~;~·;·~-~91 TO: LAURA~ -. --~·
FROM:
'!
SUBJECT: AIR SAMP~lllNG AT pt LANDFILL : . :
lease note the attached ~ o from Joe~ Hirschhorn in refere~ce to the pro os~d air sampling at
e PCB Landfill. While I llI contJnue to be supportive ofin'irolvement b the;scienc= advisors,
is is another example of J{obloms assdciated with seeking thbir 11approval I on' everythi118 we
O, I am not sure ifwe can r.ver get Hi~chhom's "approval" o4 this event O an}' other such
?ing as it would seem to ~~-~;.his intent, to, :'road block" everythi~g we do in rde~ to insure he has
pme f!Vdnue to find fault W1ith our activities. !
. I I • !
, I •
· y question to you involve the levels o~ detection he insists ~e go for-0. nai:tograms per
ilbic meter vapor and 0.0J anagrams per cubic meter particulate phase. D y9u know of any
ttb in the_ state t~at can reac these level~ of detection? I assunie that there f~lks out there
· · at can do it, ~ut at what ,co t? This raipes another question-\fhY look at 1 vel~ this low if there
te absolutely not health tis s associated~with these concentratibos? I woul an~icipate finding
anogrant levels from the _to of our offi e building, We have ~roposed us g S~uthem Testirig
· : Wilson, and they have_tol -us their L -~ is .025 p,.icrograms!per cubic m r.: I can hear Joel
ow-"they used a lab with LOD 2500*mes what I recomme,idedf" We ave ,asked the lab to
: what they can do to Iowe this and th seem to think they oan get down o S: nanograms. but
• en this Is much higher t Joel's reco mendation. ; ;
. , I \ ' .
I ~eed some suppon from Ai Quality on!this. Give me some r~asons (1) w I ~hould, or,
:) why it. doesn't make any ense. Aren'f there some comparati;ve numbers ut there? Air
~dards? Standard LOD~s?I i . l
. ! ! 1 i
9es the l.500 cubic meJe~s lf air make s~nse? You use the flo~ and time t caJpulate anyway,
:d subsequently get the vol me. I . ·-. : ;
e will be using the multi~a ered tube, sb separating th: ~ariou~ C-Ongeners ouid not be a
p pble111, · 1 ! l ·
I eed to bear back from you SAP. HoJowlly~-today. Our cotjlputers shut own about the time
I mi shed this, so I have had ome delay Ih g~tting it to you. T}lANKS f
. I ! !
'
01.24.1997 10:59 OLID WASTE DIV I I
·. ~ ,:,,~ I I i :• ~ ~,:1~'"'.''.36t s. ! I . I ; i
~ ~ I ·I ' ,,-. I i ' •
I 1anwuy 23, 1997 ................................. y ................................. ~y FAX (2 p sea) . I ' ; . . . ' '
I i : ·o:;~♦I<ell)".'. i i
Working Group I I i
.l'at Barnes f i .
row: Joel Hlr1chhoro . ! · I · , :
j°' Rcoommendatious tClll.f~g &tote•~ f !au 1hr testing tor re+ ait Cllllsai s
X
followup on 1111r disou~f ns last 'ln~day, r 081,, tho fullo~ reco ~ions 1hr onsit,, t ing by the •tato. . : I . . ! :
I Your ol!ioo should tbJio\v be re~ by 1.aun 81Jt\er 1n hc.t m os ofl>«en,ber 23,
I <i;96 and January IO, 1997 :th t tho Sci!irujeAdviaor, bo si•oa ll\• oPJIOJtUni to provide
o,l sight <>f tho afr monitorin your o1!u,e.porftirmo or hao PodWwod for lt.
2. • o voJ~tno or lllr samp!\I s aken fbr thJ P~ose ef meaRJriag'. PC.a levels oqld be at least
I, 0 onbio mot.,.. r otoph~ that lt la ~le 'VOlwue not tho VOloolty of th sainp]or th$! is mo inlpona.ut w:lth tespeci:tc obtafning l~w detection limits.
. i )
3. e la1llJIJlne lll6!hod usOd ould J>rov!ac tho OIIJ)abillty ofinolo,uring both
PC ~ attlloh~ to ~.utiCtllatoi,. I !!hould ai_, include a ll!Ollllll .fur v~ tl,e
~r• .=::::::n~=~=Y:+::::::1: :ll
s. ie test ~ts should be j,r~dod ln t~' of total PCB oo.no~trations. u •Ile « lhtco Aroclo!l (1260, np4, and 124 ). , ·
I ; 6. 1'je lest reauJts should he iep rted;,. temjs ofl'CB vapor COlloejiltatiQDs .ac
CB vapors and
xt~t to whioh tioh medium,
phase, lwell. i ,
I I
?, ; Wlllhlllation. of sample' ' and .,,.,Jo.al mothods should re£Wt in a cap llit)i to re!l.ol>]y
detoo1 and report :PCB leva!..,i .w .. o~-por cubic ntotp1°vapor, en o.q1 n.anos,am., p., ou ic motor pattlonlate jlhase. ! ! : i
1
8. Th ~lllplh(g and testJng l>'Q1olol lbo1~d i~olude •everal 1lold anJ lab blanks, Well as SJ>ibd 118mplf to dotormUlo rcoovery rat~. • i
9. Aodur.to, ••"'!'Joto, and reliibJ meteoroJ.loaJ data should bo oJIWl~dfor aD air ia.tt,plcs so
that lei,1 conditioAs aro knoWilrer the en;. Period Of time that j,lr ""'1Jplcs O cOD&ctod.
i j i I 1 ,
! I ! .
'
p. 2
P. 1
I I I I ' ·-·Ii,, ..
}
I
/ I I •
!
LID IJASTE: DIU FROM 9197153,05 SO 01.24.1997 10,59 , J H~',...\ , . ,, I ' I . ! : ~ .,
.. I i
I ; ~ ln wJ.d directi~a <ativc to •~ loo.atlOll are paitJoularo/ onant ObviousJ,y,
$amp.ling events should be biA~ed 1n fa.v~r oflooadons and tini~s wheii the o~tiona 8.l"o downwind fth~ ltadfUl. J I : i
0. Soll or •"4iwnt sanjp~ taken for tcB analy,w, shollld b~ lo locatloas dial can reasonab\y bo nsidored places with PCB, au· release,jll3bY have been deposi~ed.
1. All PC8 laboratory ti>st!results and ltoaumonto should be .Provided cienoe Advisots and the ;st,te. , i ; ·
i. Any pubHo sw..nent ·~ the 61111!• ~cQming !he test reJ. !mist pay licit all"'1don to the
*ssibility of current false lle~tive fin · s with :ri,gp~ct to PC~ !eve.ta that · t havo been
. er at .Previous times and t pa.$t qr ~ture periods When amqient 1eMper~ es ~ ~er.
I ; • A n~ber of aample• />ll oted pmnb for lnfflDrlng methane level,; ould also fflcludo
oient ak san1ples for t•1s fbr PCB ~els. !
i .
I I . I
i
I • I I I ,
i i
I i
;I
I I I
I i
2
...J,JJ.1 . .i.., ,. J .L,.,d ,.1 il.i,1., ...... ,t
P. l
. ,
Analysis of Ambient Air Samples
1. Triangle f:aboratories Inc.
Research Triangle Park, North Carolina
(919)-544-5729
Client Services Contact: Debbie Hage
Technical Contact: Gene Riley
PCBs: :::: $250
dioxins and furans:::: $1250 with a 21 day turnaround
2. AL TA Analytical Laboratory
-Sacramento, CA
(916)-933-1640
Contact: Bob Mitzel
PCBs: :::: $250
dioxins and furans:::: $1275
3. Huntington Analytical (formerly Maxim Technologies)
Minneapolis, Minnesota
(612)-659-7600
Contact: Charles Sueper
dioxins and furans:::: $900
PCBs: another Huntington Lab would have to be contacted
4. ESE, Inc.
Gainesville, Florida
(904)-333-1605
Contact: Paul Geiszler
PCBs: :::: $200 to supply PUF and analysis
dioxins and furans: :::: $1250 (subcontracted)
5. Lancaster Laboratories
Lancaster, Pennsylvania
(717)-656-2301
Contact: Dick Entz
PCBs::::: $225
no capabilities for dioxins and furans
6. Freehling and Robertson Inc.
Richmond, Virginia
(804)-264-2701
Contact: Grace LaRose
PCBs::::: $200
no capabilities for dioxins and furans
Post-it" Fax Note 7671
Co./Dept.
Phone # -, '
Fax # _'J -3 G CS
..
j .
Phone # '?IC( 73.3 IL///
Fax # _3. /e, I --z_
Analysis of Ambient Air Samples
1. Triangle Laboratories Inc.
Research Triangle Park, North Carolina
(919)-544-5729
Client Services Contact: Debbie Hage
Technical Contact: Gene Riley
PCBs: "'$250
dioxins and furans "' $1250 with a 21 day turnaround
2. ALTA Analytical Laboratory
Sacramento, CA
(916)-933-1640
Contact: Bob Mitzel
PCBs: "'$250
dioxins and furans "' $1275
3. Huntington Analytical (formerly Maxim Technologies)
Minneapolis, Minnesota
(612)-659-7600
Contact: Charles Sueper
dioxins and furans"' $900
PCBs: another Huntington Lab would have to be contacted
4. ESE, Inc.
Gainesville, Florida
(904)-333-1605
Contact: Paul Geiszler
PCBs: "' $200 to supply PUF and analysis
dioxins and furans : "'$1250 (subcontracted)
5. Lancaster Laboratories
Lancaster, Pennsylvania
(717)-656-2301
Contact: Dick Entz
PCBs: "'$225
no capabilities for dioxins and furans
6. Froehling and Robertson Inc.
Richmond, Virginia
(804)-264-2701
Contact: Grace LaRose
PCBs: "'$200
no capabilities for dioxins and furans
7. Southern Testing and Research Laboratories
Research Triangle Park, North Carolina
(919)-237-4175
PCBs: "'$100
no capabilities for dioxins and furans
•l," ~,' L, 'f~\ in:.· .. \\i.., --r~ ~~~
~\\-:.BF A Environmental Consultants
Barnes, Ferland and Associates, Inc.
MEMORANDUM
TO: Mike Kelly
Working Group
Joel Hirschhorn
FROM: Pat Barnes
DATE: January 23, 1997
SUBJECT: Proposed Air Sampling for PCB Impacts
To formalize what we discussed Tuesday, prior to the Working Group meeting, I feel that more
focus should be given to the need for immediate sediment samples. As you know, the air is a very
dynamic media and the results of any analysis performed, unless positive, will be inconclusive.
Moreover, a negative sample will yield no new information on potential past releases. Perhaps the
best reason for performing air analysis now is to begin to establish a baseline of air quality data
within the vicinity of the landfill, which will be a very valuable tool for future impact assessments.
With that in mind, it is therefore very important that the locations and analytical approach
established today be well thought out and as comprehensive as possible, because future changes
to the plan, unless additive, should be minimal.
In our conversation, I recommended, and you concurred, that an additional location be added to
cover potential discharges in the vicinity of the leachate treatment systems. Like the main gas
vent, the septic tank vaults and effluent discharge line have a direct connection to the landfill
waste material. It may be prudent that the lids to the concrete vaults be redesigned to be air tight
or fitted with gaskets to minimize the potential for PCB vapor discharges. A filter may also be
necessary on the effluent outlet pipe, or any other gas release point of the treatment system.
Concern over the environmental impact of past releases at this point can be best addressed by the
collection of sediment samples at surface water flow choke points. There should be several such
areas within close proximity of the facility.
Mike, you have agreed that this effort makes good sense, however you suggested that it be
addressed as a part of the full sampling. Since well drilling will not begin until early next month,
and because samples will not be collected until after the new wells are installed, which is likely 30
to 45 days later, I recommend that we consider collecting sediment samples concurrent with the
proposed air samples. The locations can be selected after one field day's work within the coming
week.
Please give me a call to discuss this matter.
PAB/psgll-23MKJ.doc
3655 Maguire Boulevard • Suite 150 • Orlando, Florida 32803
Office ( 407) 896-8608 • Fax ( 407) 896-1822
6 -24-I f➔S:➔6 Cl : I ClAH Fi:;.:Ot1
cc•J·rl
,-p_:! ,;c.
r-p~ 1' /1.nuary 23, 1997 .................................................................... by FAX (2 pages)
To: Mike Kelly
Working Group
Pat Barnes
From: Joel Hirschhorn
Re: Recommendations :regarding state's plan for testing for PCB air emjssions
To follow up on om· discussions last Tuesday, I offer the following recommendations for on.site
testing by the state.
l . Your office should follow the recommendation by Laura Butler in her memos of December 23,
1996 and January 10, 1997 that the Science Advisors be given the opportunity to provide
oversight of the air monitoring your office perfonns or has performed for it.
2. Tile volume of air samples taken for the purpose of measwing PCB levels should be at least
1500 cubic meters. I emphasize that it is sample volume not the velocity of the sampler that is
most imponant ·with respect to obtaining low detection limits.
3. The sampling method used should provide the capability of measuring both PCB vapors and
PCBs attached to particulates. It should also include a means for verifying the extent to vv-hich
breah.ihrough of some PCB vapors may have occUITed with respect to the collection medium.
4. The analytical methods used should be capable ofjdentifying the full range of PCB congeners.
5. The test results should be provided in terms of total PCB concentrations, as well as in terms of
at least tbxee Aroclors (1260, 1254, and 1242).
6. The test results should be repo1ted in tem1s of PCB vapor concentrations as well as particulate
phase levels.
7. ·ne corn.bination of sample size and analytical methods shonld result in a capability to reliably
detect and report PCB levels as low as 0.1 na.nograms per cubic meter vapor, and 0.01 nanograms
per cubic meter particnlate phase.
8. The sampling and testing protocol should include several field and lab blanks, as well as spiked
samples to determine recovery rates.
9. Accurate, complete, and reliable meteorological data should be obtained for all. air satnples so
that ambient conditions are known over the entire period of time that air samples are collected.
1
P. I
6-24-1 S)S➔6 [1: 1 1 At 1 FPOH
Changes in wind direction relative to sampling location are particularly important. Obviously,
sampling events should be biased in favor oflocations and times v,hen the locations are downvvin.d
of the land.fill.
10. Soil or sediment samples taken for PCB analysis should be in locations that can reasonably be
considered places with PCB air releases may have been deposited.
11. All PCB laboratory test results an..d documents should be provided simultaneously to the
Science Advisors and the state.
12. Any public statement by the state concerning the test results must pay explicit attention to the
possibility of cun-ent false negative findings with respect to PCB levels that might have been
higher at previous times and at past or future periods when ambient temperatures are higher.
13. A number of samples collected primalily for measuring methane levels should also juclude
sufficient air samples for testing for PCB levels.
2
State of North Carolina Department of Environment,
Health and Natural Resources
Division of Solid Waste Management
James B. Hunt, Jr., Governor
Jonathan B. Howes, Secretary
William L. Meyer, Director
MEMORANDUM:
January 21, 1997
AVA
DEHNR
TO: WARREN COUNTY PCB WORKING GROUP
FROM: MIKE KELLY
SUBJECT: ACTIVITIES AT THE LANDFILL
Attached are copies of an overview on three planned activities the Division of Waste
Management plans for the next couple of months at the PCB landfill:
1) AIR MONITORING
2) METHANE GAS MONITORING
3) WEATHER STATION INSTALLATION
As discussed with this group last month, we had planned to do some air monitoring at the
landfill the week of January 6th; however, we were unable to get the materials to the science
advisors quickly enough to allow them to review it in as great as detail as they wished. As of
now, we have not rescheduled this event, but hopefully will get it done in the next 3-4 weeks. A
copy of the protocol is attached and we have incorporated many of the suggestions from Joel
Hirschhorn, and plan to include an additional sampling site in the sand filter bed based on a
recommendation from Patrick Barnes. Patrick also suggested that we do additional sediment
samples in the area, which we will incorporate in the major sampling event we hope to conduct
next month.
We also are gridding off the area of the landfill to check for methane leaks in the liner of
the landfill. This testing will be aspart of our check on the integrity of the top liner and provide
us with safety information to be utilized in the digging and well construction in the landfill.
We are likewise working on the installation of a weather station at the landfill to record
and provide us real time data to be used in a variety of ways as we push towards detoxification of
the landfill.
All of these events will be done exclusively by Division personnel and with Division
funds. Participation by members of the Working Group and science advisors is welcome.
P.O. Box 27687,
Raleigh, North Carolina 27611-7687
Voice 919-733-4996 f@ffiaji/4M5
FAX [phone]
An Equal Opportunity Affirmative Action Employer
50% recycled/10"/o post-consumer paper
State of North Carolina
Department of Environment,
Health and Natural Resources
Division of Solid Waste Management
James B. Hunt, Jr., Governor
Jonathan B. Howes, Secretary
William L. Meyer, Director
MEMORANDUM:
FROM: MIKE KELLY 1 \ 'i
SUBJECT: UPDATE ON RFP'S
January 21, 1997
.AVA
DEHNR
The following is the current status of Requests for Proposals:
1. MONITORING WELLS: The contract for monitoring wells is being sent to Environmental
Investigations at a cost of $69,615. They can be ready to start digging within two weeks of
signing the contract, and based on my conversation with them we should be putting the wells
down by mid-February.
2. DIOXIN FURAN TESTING: The contract will be awarded to Southwest Laboratories of
Oklahoma at a cost of $28,560. They have a representative in the Raleigh area that will
participate in the sampling event and take custody of the samples. I will speak with them in the
next couple of days to finalize the time frames.
3. BORE HOLE/EXCAVATION WELLS : It is anticipated that the contract will be awarded to
S&ME at a cost of $38,576. I anticipate working closely with them in the next few days to begin
submitting the necessary documents to EPA for approval. I have spoken with EPA this morning
and they will try to expedite the permitting of this activity.
4. BENCH SCALE PILOT STUDY: On January 2, Joel Hirschhorn submitted his plan for this
RFP, which we put into the state format. There has been an initial review of our first draft and
the second draft is available for review. Contracts and purchasing is also involved in this loop to
help expedite the RFP. EPA has supplied a list of their approved vendors for BCD. The RFP
will state that only BCD and gas phase chemical reduction technology will be considered. We
should have this RFP finalized next week and anticipate getting it out by February 1. As of
today, we have not set up the schedule of events to coincide with this study.
P.O. Box 27687 ,
Raleigh, North Carolina 27611-7687
Voice 919-733-4996 •@ffiiJl@!jAf.i
FAX [phone]
An Equal Opportunity Affirmative Action Employer
50% recycled/10% post-consumer paper
PROJECT:
PROJECT PARTICIPANTS:
PROJECT DATE:
PURPOSE/OBJECTIVE:
Methane gas monitoring as a measure of liner integrity
on the Warren County PCB Landfill
Larry Rose
Wendy Peacock
Bill Sessoms
Open
Gases, primarily methane, are being generated from microbial decomposition of vegetative
matter in the PCB Landfill. Methane concentrations of 5% the lower explosive limit, LEL, of
the air vent gases have been measured. The presence and flow of methane indicates that the gas
pressure in the landfill cell is greater than ambient pressure. The PVC liner was designed to
contain gases and provide venting only through the existing air vent. If the PVC liner system is
functioning as designed, there should be no gas leakage through the liner. Methane measurements
can indicate the status of liner integrity. If methane is measured, then the source and cause can be
determined and options for remediation proposed. The purpose of the project is to implement
methane monitoring as a means of determining surficial liner integrity
MATERIALS AND METHODS:
The PCB waste disposal site in Warren County as-built plans are by Sverdrup & Parcel of 1981
and will be used to identity the landfill disposal cell. During the week of January 13-17 the staff
will establish baselines on the landfill site. These baselines will be used to locate grid patterns on
the surface of the landfill cell at some future date.
For proposed methane testing the landfill surface will be marked with temporary wire flags on a
50 foot grid interval. This will provide between 50 to 80 methane sampling sites.
At each of the sampling sites, a sliding hammer probe will be used to drive a three-fourths inch
diameter by twelve inch deep void space in the landfill topsoil layer. The void space will be
plugged with a rubber stopper for 12 hours to allow accumulation of methane. After the 12 hour
accumulation the void space will be sampled for methane with a infrared gas analyzer.
The gas analyzer is a Landtec GA90 model. The analyzer has an internal vacuum pump to
extract gas from the void space through a flexible hose into a detection chamber. An infrared
beam is projected into the chamber and through the gas. The gas concentration is determined by
the degree of absorption of the infrared beam at the detector. An internal microprocessor
calculates gas concentration and the result is displayed as a digital readout on the instrument. Gas
concentrations are measured and displayed as percent by volume, parts per million and lower
explosive limits (LEL). The instrument can also measure CO2, 02, and Static Pressure. The range
and resolution for methane concentrations are:
Range: 0-100% CH4
0-60% CO2
0-25% 02
Resolution: 0.1 %
0.1%
0.1%
The readouts will be logged at each grid location. Grid locations that indicate methane will be
used for further investigations. The grid spacing will be reduced to 5 feet intervals around the
initial grid location for additional methane monitoring. If no methane is detected the grid spacing
will be reduced to 1 foot intervals. This process will identify specific areas of leakage. If methane
is detected at the 5 foot grid intervals a second, third or more 5 foot intervals will be monitored to
identify the area of leakage.
All methane monitoring results will be presented to the Working Group for discussion and
recommendations. Oversight of this project by the Working Group is welcome.
c:wpfiles/pcblfi'misc/ gas-mon. pro
State of North Carolina
Department of Environment,
Health and Natural Resources
Division of Waste Management
James B. Hunt, Jr., Governor
Jonathan B. Howes, Secretary
William L. Meyer, Director
MEMORANDUM
DATE: January 17, 1997
To: Mike Kelly
FROM: Bill Sessoms
Wendy Peacock
Larry Rose
REF: PCB Landfill
AVA
DEHNR
We have established baselines at the PCB landfill to facilitate obtaining gas samples.
The outline provided by Bill Meyer was generally followed.
The end risers of the irrigation system were used to establish a centerline. Using this
centerline, stakes were placed at 238-feet from the center vent. Corner stakes were set
116-feet at 90-degrees from the centerline. To aid in re-establishing these points, stakes
were set at the fence in line with the corner stakes. Appropriate measurements to these
stakes were recorded.
Flags were set at 50-foot intervals along the exterior baselines established by the
four staked corners. The intervals were measured from the centerline stakes in the east-
west axis and from the south end in the north-south axis.
These baselines yield 55 to 77 sampling locations depending if the north-south axis
baseline points are utilized.
Two sketches are attached showing the layout that was established.
C:\SESSOMS\PROJECTS\MISC\PCBMEM0.1
P.O. Box 27687,
Raleigh, North Carolina 27611-7687
Voice 919-733-4996 ffffitJIMMfM
FAX [phone]
An Equal Opportunity Affirmative Action Employer
50% recycled/ l O"lo post-consumer paper
.. , --ALANrMC
LANDFILL CONTROL TECHNOLOGIES
~
lo~Z
T Providing Environmental Compliance Solutions for Landfill Gas Management. (800) LANDTEC Nationwide.
111', 0 !
D West Coast Sales [phone] • [phone] FAX D East Coast Sales [phone] • [phone] FAX
. . -A IANlTEC
LANDFILL CONTROL TECHNOLOGIES ..._..
..., Providing Environmental Compliance Solutions for Landfill Gas Management. (800) LANDTEC Nationwide.
___ i l
.µt,
i <"' ~ ,,
i'
./'~. ,1'
"-'fit -io I' @I ) JO / &'Zi
#
I',. J I ,. I ' •· ., .. •
I'. 0 -• }I
+-· •
I,, .. • .)
t., ,. ,# 41 ,A
I ~
1-. .,A
ll\t~l-1,
• • • ,A
~ • ·• !4
• •
':P "i'Yf·
r
-6(}
I' ,A t
•
D West Coast Sales [phone] • [phone] FAX D East Coast Sales [phone] • [phone] FAX
PROJECT:
PROJECT PARTICIPANTS:
PROJECT DATE:
PURPOSE/OBJECTIVE:
Meteorological Station at Warren County PCB Landfill
Ed Mussier, P.E.
Open
Provide site specific rainfall, temperature, barometric pressure for monitoring landfill
environment. Site specific data for evaluating recharge and discharge events including response
of groundwater monitoring wells. This data will also be valuable during detoxification of the
landfill.
MATERIALS AND METHODS:
It is proposed that instrumentation be installed for continuous monitoring of ambient conditions,
internal and external, at the PCB landfill. The proposed equipment would be a weather station
consisting of an ambient temperature probe, barometer, and tipping bucket rain gauge. Readings
would be collected by a CRl OX date logger. The data logger is battery powered and the
information may be down loaded into laptop computers for transport to the office. It is proposed
to install the data logger and power supply in the existing maintenance shed, with the measuring
devices located outside and attached to the shed.
To monitor the conditions in the landfill it is proposed to install vibrating wire strips piezometers
in the leachate line and the measuring tube in the gas vent. These devices measure the internal
temperature of the landfill, the internal pressure, and the water level in the monitoring points. The
data can be collected continuously and is stored in the same data collection device described
above. The line from the instrument in the air vent will be run under the sod layer of the cap, to
prevent damage from routine maintenance of the landfill.
The Division currently owns a weather station. It is proposed that the data recorder, barometer,
and rain gauge from this station was borrowed. Additional equipment will need to be purchased
by the division.
The following is the equipment proposed for use at the landfill:
1 Campbell Scientific CRl OX data recorder
1 Campbell Scientific A VW4 4-Channel Vibrating Wire Interface
1 Met One Instruments 8" 370-Tipping Bucket Rain Gauge (0.10 inch sensitive)
1 Met One Instruments 26/32-1 Barometric Pressure Sensor
1 Met One, Campbell Scientific, or equivalent, temperature probe and radiation shield
2 Slope Indicator VS Piezometer (vibrating wire strip) (0-50psi water sensitive) plus
associated signal cables
2 24 amp/hr jell cell rechargeable batteries, battery trickle charger (Sears or equivalent),
miscellaneous cable, alligator clips, wire etc. [ Available locally]
All data will be submitted to the Working Group on a monthly basis. The frequency of reporting
may be changed if needed by the Working Group. The Working Group is welcome to oversight
this project.
c:wpfiles/pcblfi'misc/met-sta.pro
State of North Carolina
Department of Environment,
Health and Natural Resources
Division of Solid Waste Management
James B. Hunt, Jr., Governor
Jonathan B. Howes, Secretary
William L. Meyer, Director
AVA
DEHNR
January 17, 1997
MEMORANDUM
To: Mike Kelly
Deputy Director
Division of Waste Management
From: Pierre Lauffer ,0;z(}
Health and satetf' c(oordinator
Re: PCB Landfill Sampling/R.A. Hite's Study Review
I read R.A. Hite's study titled, "Long-Term Measurements of Atmospheric Polychlorinated Biphenyls in the
Vicinity of Superfund Dumps," and I generally have concluded that there is really no problem at all in
addressing Joe Hirschorn's concerns regarding our sampling strategy for monitoring ambient air conditions
at the PCB landfill, Warren County.
If Dr. Hirschorn wishes the Division of Waste Management to have the samples analyzed for the congeners of
PCBs (ie. dichlorobenzene, etc.) as well as for PCBs, we can easily perform that monitoring at a cost that not
much above the current cost projections. Dr. Hirschorn also stated that the analysis in Hite's study was much
more sensitive than the EPA methods of 1982. While this is correct, we will not be employing those sampling
methods and analysis as we will be employing sampling and analysis methods of 1997.
For example, at that time, the typical procedure was to employ polyurethane foam (PUF) filters with a filter
head containing glass-fiber filters. The procedure required the sampler to manually load the PUF into the
filter. This process presented many quality control/quality assessment issues. Today the sampling
procedures for PCBs state that ORBO-60 tubes are to be used in combination with glass-fiber filter
cartridges. The ORBO-60 tube is, in fact, a manufacturer loaded PUF filter combined with florisil filter
plugs. ORBO-60 tubes consist of two 30/48 mesh florisil sections. The first section is proceeded by glass wool
and contains 100 mg and the backup section contains 50 mg. There are two PUF sections: between the florisil
sections and after the florisil backup section. These filter tubes will be manifolded to glass-fiber filter
cartridge heads as required under NIOSH sampling and analysis methods.
There was also a concern by Dr. Hirschorn regarding the air-flow volume rate we will be employing as part
of our study. The flow rate is according to NIOSH sampling method #5503. We could increase that air-flow
to as high as 5 liters/minute (air flow volumes of 2.5-5.0 liters/minute are considered high volumes), but we
would not be sampling according to any established sampling methods now recognized, but using a study
done by a nationally recognized professional. The sensitivity ofHite's study is also not a major problem. The
laboratory we picked for the analysis is an American Industrial Hygiene Association accredited laboratory
and is subject to the quality control and analysis sensitivity requirements for such accreditation.
P.O. Box 27687,
Raleigh, North Carolina 27611-7687
Voice 919-733-4996 IGPl:1'@&€1
FAX [phone]
An Equal Opportunity Affirmative Action Employer
50% recycled/10% post-consumer paper
SAMPLING PLAN SUMMARY FOR:
MEASUREMENT OF FUGITIVE ATMOSPHERIC EMISSIONS OF
POLYCHLORINATED BIPHENYLS FROM THE PCB LANDFILL
WARREN COUNTY, NORTH CAROLINA
Sampling Plan Date: January 9, 1997
Sampling Plan Preparer: Pierre Lauffer
Project Participants:
Health and Safety Coordinator
Division of Waste Management
Raleigh, North Carolina 27605
Pierre Lauffer, HWS-Health and Safety Coordinator (Project Manager and Sampler)
John Kirby, HWS-Environmental Chemist (Project Chemist and Sampler)
Projected Sampling Dates: February, 1997
Site History:
Between June, 1978 and August, 1978, over 30,000 gallons of industrial waste material identified
as polychorinated biphenyls (Arochlor 1260 and 1262) were discharged deliberately along the
shoulders of approximately 210 miles of North Carolina highways. In June, 1979, EPA approved
a tract ofland (previously used for agriculture) in Warren County, North Carolina as the disposal
site for the PCB-contaminated roadside soil. The landfill ( constructed in 1982-1983 and
permitted under the Toxic Substances Control Act (TSCA) contains about 40,000 cubic yards of
soil contaminated with PCBs. The concentration of PCBs in the landfill range from 150 to almost
900 part per million (ppm), averaging about 350 ppm (1), based on 1994 subsurface soil sampling
results (these were retrieved from the bottom of the vent).
Polychlorinated Biphenyls (PCBs):
PCBs are a family of aromatic compounds consisting of two benzene nuclei bonded with
two or more chlorine molecules. The PCBs of concern in this study (the type existing in the
landfill) are the Aroclors.
Aroclors are characterized by four digit numbers. The first two (the number 12) indicate
that the mixture consists ofbiphenyls. The second two digits state the percentage by weight of
chlorine in the mixture (2).
They tend to be colorless to light yellow oily substances with a specific gravity of 1. 4-1 . 5
(3). Due to their stability and nonconductive properties, PCBs had many industrial uses including
use as insulation coating in electrical capacitors. PCBs possess, as a whole, high open cup
flashpoints (348-356° F.-Aroclor 1242, none for Aroclor 1254 and 1260), but do readily distill at
temperatures above 325°F. The distillation range for Aroclor 1242 is 325-366°F; for Aroclor
1254, 365-390°F; Aroclor 1260, 385-420°F (4).
Toxic effects from exposure to PCBs in human include chloracne, pigmentation of skin
and nails, excessive eye discharge and swelling of eyelids, and gastrointestinal disturbances. PCBs
are considered carcinogenic (5).
Because of their high toxicity, detrimental harm to the environment and stable molecular
structure (structure remains intact in the environment for long periods of time) PCB manufacture
was discontinued in 1976.
Objective:
The objective of this study is to determine ifthere are uncontrolled PCB emissions
originating from the Warren County PCB Landfill. This study will involve ambient air sampling
to determine if PCB contaminated air particulate matter and vaporized PCB (Aroclors 1242,
1254, and 1260) emissions are present. Retrieved samples will be analyzed by an independent
American Industrial Hygiene Association accredited laboratory. Southern Testing and Research
Laboratories, Inc. of Wilson, North Carolina has been selected for conducting the sample analysis.
The results of the analysis will be submitted directly to the workgroup and DWM. A risk
assessment will be requested from the DEHNR Epidemology Division to determine if the PCB
emissions (if found) are a risk to the surrounding community.
Materials and Methods:
Air sampling will be performed with low-volume (L V) and high-volume (HV) constant
air-flow sampling systems. The components of the sampling systems consists of battery operated
constant air-flow pumps (L V or HV) (Gilair Pump by Gilian™, Models: Gilair3(L V) and
Gilair5(HV)), 13 mm Gelman Swinney filter cartridges with 13mm, lµm pore-sized glass fiber
filters manifolded to ORBOTM-60 100150mg, 6 x 70mm florisil sampling tubes.
The pumps will be calibrated by the Gilian Gilibrator (digital calibrator) prior to and after
each sampling period. The glass fiber filters will be precleaned prior to loading filter cartridges by
the North Carolina Public Health Laboratory and loaded at the laboratory by laboratory
personnel. The sampling head will consist of glass fiber filter cartridge followed by the florisil
tube. The two will be manifolded together by tubing. The purpose for this sampling system is to;
1) catch any possible PCB contaminated particulates from the air, 2) to filter PCBs which may
have vaporized. This system will then be manifolded to the sampling pump with plastic tubing.
There will be five sampling periods of five hours. The sampling sessions will begin at
roughly 10:00am, 3:00pm, 8:00pm, 1:00am, 6:00am. Each sampling period will be separated by
one day. The reason for this delay between sampling events is to provide time to recharge the air-
flow pumps ( they require 16 hours of recharge time). Each sampling period will consist of seven
air-flow pumps and sampling units. One unit will be located inside the vent on top of the landfill.
Two more units will be located two meters downwind from the vent ( these will be seperated from
each other by one meter). Two other units will be located diagonally downwind at the landfill
fence line. One unit will be located 200 meters straight downwind from the vent. The air-flow
rate will be calibrated to NIOSH guidelines. NIOSH Sampling Method #5503 : Sampling for
Polychlorobiphenyls states that air-flow rate should be 50-200cc per minute. The flow-rate per
unit will correspond to its distance from the vent-the further the pump is from the vent, the
greater its flow-rate. Another sampling unit will be located 200 meters upwind and will act as a
background sampler. Approximately fifty samples will be obtained. Prior to each sampling event,
the wind direction and temperature will be indicated by a portable weather station established on
the east side of the landfill.
Funding from the DWM operations budget rather than the PCB detoxification budget will be used
for this project.
NOTE: Please see attached diagram of the landfill showing sampling locations and a copy
of the NIOSH lab procedure.
c:wpfiles/pcblfi'misc/sarnplere. wpd
----------------------------
------------·-----------·-·
------------------
----------------------------------•--·--------J 6 7;~ ,.~~--------------------------··------·-··-------·
__________________ _____ _ _______________ /R3f_1J..'_1 -------
-· ________________________ ,-__ -__ -_-__ --_-__ -_-__ --~ ... ., s5o'=====::::::::===~~---------------------------------
-----------------------------------·-·-----------····· -----~7::;~JV
I
\
\ I
i
I I I .
!
--·
.---
---
FAX COVER SHEET
FROM:
I
PHONE: 7~3-4996
COMMENTS:
TOTAL NUMBER OF PAGES INCLUDING COVER SHEET: ;.¢
F:on Hi t es ID :812-855-1 076 JRl·•l 1 7 '97 15 :06 No .002 P.01
RONALD A. HITES
School of Publi c & Environmental Affairs
Jndi~tna University
FAX: [phone]
TEL: [phone]
ro: m ~ k~ \-<. r..-u.. 'i
FAX: [phone](,0b
Number of Pnges in this transmission: __ J_.L._ ___ _
(including this Cover Sheet)
Date: l / I 1 Lq'l r i
.I) ,· -zo Time: er J _ __c______;;c.._ __
Comments
F:on Hite s ID:812-855-1 0 76 JAl··l 1 7 '97 15 :0E, No.002 P .02
. '
' I I '• '
; it't,,r1i~;flr~ixii ':t°NVIR6NMENTAL SCIENCE & 'l'lWHNOLOGY, Vol. 28, Pa,e 12.68, O~tober 19~9
Copyright ® 1989 by the American Chemlc11l Society and rc·printed by permlttlon or the copyright owner.
Long-Term Measurements of Atmospheric Polychlorlnated Blphenyls In the
l •Ii . ' ,. . I •
. Vicinity:. of Suparfund Durttps ·
. , • :• 1 -1 -· -'l
Mark H. H•rm~n,ont and Ronald A. Hites•
' ,.,... ,,;
School or Pµbllc' iiiiif~nvrronmc;ntaL(ffe~s· and Department of Cham!stry, Indiana University, Bloomington, lnd~na 47405
.• . . . . ~ ' . . . . ' ' . . ·-' . ..,
,·:1p;;t,;;fit a!idrt14\<i:$6tef'fotf0n\t'1'.i.&;; Studies, University of
Wisconsin-¥llweukee, l'dUwaukee; WI 58204.
t ·''-"'.' ·•··:· ... ,.: ·.:
•· .. ( "
two (9, 13, l 2) or thr~ (6; 10)' oolninerblal · trilxtur&s; 'en'd
some studies have ide11tffled tepreaants.t1vetP0B10011genera~
(1 , 3, 11 ). Other studiee have reported PCB ooncentrations'
in both vapor and particulate phaeas (1, JO, 13), while
otherfl have reported only a "PCB"1Val\.ii'{2~/1i.:!9);· Tne'
second problem relatea to' seasonal ·varlablllty; \ln {j\lr
earlier studies on Iale·Royale in Lab ·Superior U);:we
oh11erved a large seasonal variation in -atmospherlo POB
levels. For example,'. th& PCB oonoontt-ationa ·itt the 1i.lm••:
mer were 5 times higher than .the' concentrtitionB in the·
winier. If there ia a large aeasonai variatiCin;thie•furtha1':
complicates the compariaon·of the:data dttid above~' wi;·,
therefore, thought %.wu · important•'t;() deterin'hte the ·
seasonal variability of PCB concentr11tion11 at o:ne loc.atlon;'
If it were poeaible w understand the·fun~tion&lrn)iitiohshltf
bet.ween atmospheric concentration ind temperature, it
would then be po&liible ,to i•'COtteot tfuture ·atmosph&rio:
measurements to a· rori1n1on •:fiird~tiifa~t11fU)fdet to . achieve comparabilltyl 1 ,,..--. ·. _., .• , ... ,-~ ··:·•.·~,.,--~ •·:·,~-i1t1 .. ,,,i-
. Ot!r study ueed l?~,~te~•~Jj~ln_g ~t·n\~ltlp!e• s1:tesi
within a local area, employid, quantltation -0(:a barge ,
number of PCB congener& in both.vapor and-particulate
phases, and included tl,mpt1tat\ll'ei>b4!}'Vat}Ol1f '9J)'.eb~J'•
cally, we sampled airborne vapor and particula'te•adsorood
PCBs at three aiteli in the Bloomington1-I'.N, area,from
October 1986 to Augµst_ 1~88 .. L?~•~rrn ~PlinS' ~kl<wi ·
us to observe 11easonal ~NU'Ul(I& an 1rtdrvldual oongener and
to~l vapor and partlcu1ate 1:>Clloonotintrationa;.chang~• ·
in vapor-to-particulaW partlttonlng,-futd:vart.abtll~iof P.CB ;
~~t~d;~~:tt:t ::iu!~t::J~1:~::i~ttt\.~~::
for the Bl()Otnington arM,,\vhi~e1a.!traali:.fuel&d h'lclhe'tat.of
is proposed for thernutl<d8'truet1on :of :600000 int-of
F:Dn Hites ID::3 12-:355-1076 JAN 17'97 15 :07 ND .002 P .03
t .
I
.·. ~ . "'·
) . • ' t ' • ~
-l'io'ur.-·t ''81ooml~oW,''lN,-·area ak 'aampllng loeatlons, Sup8rfund dump sit&&, propos&d lncl~~ator site, and Interim 1tor1ci-facilty alt& .
PCB-oorttaminated ttl'atertale.froMthree Superfund dump
.,. · •Si* lUld mtn en ·interlm ·P<:'B etorage facility (see Figure
l' r. :.1).-:~' ·1• · h. 1·; ~•-,\'.~••/~"! ·,\1-~·· •·~,;I ~•,;
'"' l ,: i/ ' 4, : ' I ·, ~~ -,. ''J_l • t , ,; • • i I •• i , ;. I ' , , ,
;.;:; fExpe~imentat ,Sei,~ion . ':; ·•( ;, •., .. ,,,_
-· ···, . ' (iariipUn,r,Sitet,~Welooated three samplers on rooftops
ln the Bl()t)mintton atea•to' correspond to locations where
, .. • · )theatano1phereiis'expectM·to•be affected moat and least
·-by staok emlasionsfr6tri'thelpropoaed incinerator. These
, .. ;_;. :areaamereidentifledlri :ariltk:a'aae11mentmodel (17). A
,· :'. .. uaite .iI\·dawntownBloomi~n,fon ·the Monroe County
. ''.' :Courthouse (1ee Figure l}/wae·cloeegt to the Superfund
:., : ; , ,dwn!)!I, but it was in the area &xpeot.ed to be affected least
. h: _bv .the II tack 'emieaion. Theb'ther altes, on the Batchelor
. , . ;,, .. Middle School and;on,the-former--Sanders School, were in
. ,, •. _ _,uo,.a.ofleaa urban development tiut neArer the proposed
·, · .. -, 1incinerator ,and in ,arou of1i'roater eatirnated effect of the
, , ·.-,~tack-emilaiono.•,'r.he-Batoheloralte was also ~l km from
the interim atorage·faclllty,1.,..liJo'b was being filled with
, :•, i. · ,PCB-oon~a~d rna~rll:y,'durlns part -of our sampling
,_ ·· , . · perlod {April to-Octobet·:l987),· '. All Superfund sites, the
, . · ·lnterit11•tor11gefl(oility, the prQpoe&d incinerator site, and
-: ,n; .Q\U'..threetwiiplinirlocationa •~ro.focated within an area
!. ,1. -Of .12&Jkm$·(~-1nil). T''', _,.,. ~-·· l
.. ,;, . ~ S_.mpi, Oolltot.lonJ We collected wnplea with Hi-Vol
i ·1•·:·;ai,;;_11·~plo"'·(~_t_~"-·•·lrMfo~o,• 13orkeley, CA; Model 650)
· , .... ···'lnQ(lit\ed-iors~illtlneous~eulate and vapor collection
..•. ~Ul,ji iyetem:,vtl'luai:.ct ,bf:~~·et al. (18). Particles
'· 'f ~larterithap 0.1 µm-i,n -di~etft\Ver,·eollected on a 20.3 cm
, . l .. , :X U5,4-cm-a-lase fiber filter (OFF). (Gelman Science, Ann
,:,,1;i ,J.rbor.,Ml);;the:fitat1Uap.!n~o•ampler al.ntream. The
· · . ' • ,fllttrd -We~ heated at 160 9C ·before 1ampllng to clean
l,, · · · , th&tni7. Partlo"1ate ~PllnJ ~eluded J?avimetric mea·
.
surement of to~ sJ.1,8.pel)ded partlculate loadh\i& (TSP).
Vapors were ad,11orb~d ,biw. a .9,6 cm x 10,-0:cm poly-
urethane foam plug ·(PUF~ cartridge behind,tbe GFF in
the sampler air .atream ~1.2). Be·tore aamplirlr, ·the PUF
cartridges were deaned !>ytSoxhlet uttaetton 1n a-ootone,
dichloromethane,·and petroleum ·ether·(eactdor:24 h).
Samplers were calibrated ·with a ivertturi calibrator (Sier-
ra-Misco; Model-1080) -every 6 lli'Onths or wheti~ver 88Ill·
plers were moved or replilred: ' Flow was lieHo a 'known
amount, appromiiately-o:5 m8 µili1~1, before the' OOllection
ofeach&ample;••' ,' ·,, .. _,.,'·--: · .... ,. ·:··••· ··.
Out sample volume&"rci.nt~.ft'ol:11'7,00'mt hf 81,untner to
1500 ;I18 in Mnter.' W~_.~l~ ~eie_'~olunitie ~ti,e they
proV1ded ,iufflel~~t _a~m, 1~j \l~.·o~~~~tl~~ .?,f many
congeners at a signal 't<fblanii tabo of at t~ut 6. Break-
through. of di-'Jhro\igli ·ij#ac}µpf,il '.~~g~z:ie,rf ~~ a po-
tential problem becauae the'8 ,Mm,ol~!~ -~,.~~-er make
up 94% of Arodor 1242 (19), which waa preeent in the
Bloomington dumps. Some dichloro ®P.Be.~,~ ~-~artic•
ular would be pr~ne ,to break~_rousn, N _ l;lP.&ei'ved by
Burdick 811d Bi~man'.,(20);:blquatof~eir bi. gh~~ vapor
pre88ure11 (21). ~e-'te&ted 'tot brHk:through ·~otir times
~uring the pr~ieet_ hr coll~ting r~~B ori::a_ PP~ plug cut
into upper and 1Q~er'h$lv~. Th~'~ve1-~r_e,'~eparatec\
after sampling and -~ for PC&· 'l'lie up11trwn half
should contain''dlci6t'or .all' of th~'PC:Bs"it 'b'r~~oueh
has not occurred: 'rlitik,'~u~~~ ~•12tte-r ratio, would be
high if no breaktbfo.~"Bas'hedin&d_!'. A(fa'tlH-eqU'al to or
less than 1 inl;llc•t.fbt#a1ttlu61igh;·'Rta\ilt, bf our ex-
periments &how moasunible'Quantities of only a few di-and
trichloro congeners in the lower PUF pltJi .. Upwr to lower
~~~;:no!~~ss thm, _ 3 .w,~re'J?~d ~~l~'f~r· a f~w (1\~hloro
.· . • _ •. (.,l, •. · .•
I .
F:on Hite ::; ID:812-855-1076 15 :08 No .002 P .04
Table I. Total Vapor and Particulate PCB Concentration• and Air Temperatures for 'fhree Bloomlnlton Bite,
Courthou5o Batehelor Sandera
Julian ~ ,vapor, ni/tA~ · partic,
year day Courtho~ ija\che.or Sander• ng/m8
86 800 •· . 0.47 ·:. , ~~\t:· nm 0.079
86 816 0.40 '~--, nm nm 0,053
86 S88 • nm, · 0.68, nm nm
87 •·· 15 • 0.81 nm• nm 0,066
87 '.28 •. . nm 0.83 nm nm
87 ,29 '· · 1nm ·· ·· · nm .... 0.88 nm
87 48 o.~ , run nm 0.032
Jl,-.:~~--~--~·-·t!t.\'. .;':.:. •·i·' g_~8 ~~5
; _87 " ~,. _nil)<·•. , . , ,. 1,1 .. , nm nm
·. 87 : -~; ~v8 .. ,.: :· P .. ,1 ,, •• 0,65 .. ,.:.. 2.8 0.078
,87 , , 138 ,• . , .5,6,., ,. . -~1 )ii 1.8 0.087 87 .. 166 . 6,8 . ';•· ·' 1.9 ' , 3.0 0.18
87 194 6,9 1.7 2.8 0.0048
, 87 ~~6 ,.e . ·' . . 210 nm 0.048
1 81'~ "0·i&l·"··"":.,'8:s'1''"••i ••~ .. f ''' 0.83 0.0075
' 87 ·· 287 . "2;9 ' '· . ' ! 2:4 ' 0.SOt 0.11
••87 · ·820··, ' · 0.67 ,,.: O;Ml ' ·-· 3.6 0.0:32
! _87.-.,. 848 :2.0 .... · . ,, 0,,24 .', 0.32 0.063
88 25 0,30 I " I 0,0313 • · 0,18 0,081 . as ,68 o.se . , 012i . . 0.2a o.on ·• as •u · '16 ,,. -:'tiJ1'' .... , ••. ,. 0.1~·''"·' o.oo o.o:~6
'·88 ' ·· t10 ·• Lf'•·: ·1···nm ··,.,,'. nm o,or,4
88 111 U ' ,:· '·0.60 · ·' 0.26 .o.or,2
88 144 . 2.6° ' • l,l I 1.1 0.029'
; 88 168 · · , 8.4° .t 2,8 ·, · 3.5 0.079'
,88 193 . 9.3 ., , , 1,7 .. 1.6 0.075
: ~\I ·.:·· f~6 .. ·· 20,:;. ..,,, ·~8 ·. 8.0 0.11
: ,f POB-O:On,oentratl~m :,,erpaui of total 1uapended partlcloQ (TSP).
,tutemen~. ·. ,; .• ... 1-,; , • •. :· , •. -. .,
Sample Anlly1e1,1 After being welghed for TSP, the
·,gl.au nber filters were extracted ~th 'benzene in a Soxhlet.
lapparatua fo-j .24 h. ,;p{JF .. waa utraottld in petroleum ether
.,for '24 h in-aSoxhletlapparfit.usbuilt~specially to nccorn-
•modate i the' ,large -c:artridgei · • ;A .. ·known amount of
2,2',3;4,4\5;6;6'-octa'Chloroblphenyl [IUPAC no. 204 (22);
,Ultra•).SdJentific, Hope, Rl)'wu·added to each GFF and
PUF be(ore sample e:rtractlon .to ~l'Ye as a quantitation
. .standard .. .-. ,,·: ; .... ::.i : .. : ,;,• ,,.,.: ,;
· FolloWllli e,:tractl-ontboth·the PUF and GFF extracts
·were'.teduced in volum&l'they1were·then cleaned with a
lQ.cm-oo:lwmrof 8% deactivated 1ill011 gel (100-200 mesh).
,The sample idvas eluted withllexan\;, 10% dichloromethane
in heiane, and dichloti.nh$thane, The hexane and 10%
idiohl<>roUlethane fraotlonswere:oomblned and used for the
.arnilyti.Bof~.O&;•The requlred·volwnes of the11e fractions
·were determined ·experimerluilly •t ~e beginning of the
·project when we ·spiked PUF:wlth the internal standard
,Md with an Aroolor>t242 f!Wldai'd:-
, ·· Quan.dtatlont;:We analyied -all ~amplea with a Hew-
·tett:P.c~d :~90~_'-IWI chrdm~togia_ph equipped with an
~Ni elactron :ti&ptute detector .(2CD), a 30-m DB-5 ca-
pillary c~lumrr(J&W Scl-entlfibrFolsom, CA), and an HP
7671A autosam.pler.· Etfoh ,vapor-sample wae analyzed in
triplioa~.' ·Particulate ,,lhples w.,-re ,anal~ed only once
·b&oause-they were 1nu<ih•ieaa conoentrated than vapor
-samples:~. '1;'., .. -. : .\ ... ·,
·· The ~ temperatut-e ,pr~cun·us&d a 1-min hold at 70
0 0, then a ramp to 180 11C at 804 Ctnin-1, then at l.n °C
1tnin-~to;e60 qe;•and·«t 10.4'0 niiliT~to 280 110. We used
· aplltle111!lnjectlon,at 226 °-0;-the EGO -was held at !i2fi °C.
,Hellum~areas.flow was l,9,n'L mtn-1 at 70 °C. A 95%
vton· fll1t! 6'fo",n:ot1t'limS ft:ltitutb"'!W«&'\irled for the makeup
1ib,• ,&'otal<flow •Wa&'il:-imb,mln~•h'-'
··· ,Tb~ tnstrWJ)tiiitt\VU'~lbr~ft~tlve to the quantl-
tatlon ahuldard ~(PC~ .e9J~i"f()r➔oonsener specific mea-
partic,• partic,
µg /g ng/m'
nm nm
nm nm
nm 0.043
nm nm
nm 0.018
nm nm
0,26 nm
2.1 nm
nm nm
mn 0.097
1.6 0.0018
0.98 0.0047
1.6 0.030
run 0,0071
0.91 0.013
0.21 . 0.012
0.98 0.21
0.fl5 0.039
nm 0.026
2.4 0.01_5
0.79 0.039
O.!Jll 0.0069
J.O nm
0,72 0.054
0.63 0.010
1.6 0.029
1.3 0.097
0.82 0.085
partio,• pa.rtlc/ µi/g ni7m
nm nm
nm nm
nm nm .
nm nm
0.48 nm
nm '· 0,0046 ..
nm nm
nm __ ;,-....• ,nm .·,
~Jltil~~_.•.._~ H•r•o•m ., . .Z-6 ;. · . .,.n~---o.oo . . ., . O.Ot:2 .
0.08 0.M7
0.16 0.11
nm 0.41
0.40 nm
0.3~ 0.0,2
·2,4 ., 0.083'
·1.0 0.069
0.98 . ' 0.01&
0.42 ··0.027 ·
0.15 0.0034 o.sa o.m1a·
nm
0.86
0.30
0.74
:0.49
0.88 ·
nm
·0.069 . 0.027 .
0.4'l '
0.01,6
: 0.088
p&rtic,' av
141/g temp, •c
nm 16.1
nm . ...e.o
nm 4.2
nm •' .. 5.8
nm ,6.l
0,18 8.7 nm 6.8
nm 6.(1.,. ·,:.:o.n .. .._ : ' ,U.~.-
6::s · 18.8 . l~.Q -, u 27,IS
0.67 82.4
nm 26.9
nm 24.2
O.IS9 20.8
0.59 18.4
2.0 19.6
0.46 ,.2
' 0.21 -7.9
-·nm 12.~
1.3 1.6
nm 11.7
2.2 17,8 MO 19.8
9.6 21,8
· 0.47 , 24.4
0.715 35.8.
b nm, not meuured b6cause no ~a?Qi~ie-we.s tllk~n. (' Duplioare ~~-.. . .. . ' . .
Sltrements with a mixture' of Aroolo~ 1282, 1248, and 1282
in a 25:18:18 ratio. The wide range ofthi8 mixture inoludM
most environmentally -important PCB oongener11: The
concentrations of the irtdfvldual ·oongenets tn these tbroo
Aroclors has been previoualy determined {28).-·The cali-
bratio? mixture wa1,:anal~ 1lfter each ,group o! e~ples
to verify that the trultrurnent could r~produt» the Ofll.l•
bration. Our sampln'O'O'nSt&Wrttlyshowed 61of.the·ca1J •
brated GC peaks (78 oongenert). ·: . · ,,
Field blanks were collected ll.lld analyied with eaeh·i&t
of samples and the"1amount'of thetblank waa1JUbtract&d
from the sample on aoongerier~by-congener basla; •The·st,,e
of this correction >waurMU,'tiwllly ;.;;,5% ofthe':89.tnPle.
Occasionally, for partlculate $aulplee taken· ln ·th~'winter,
the correction waa u larJe' a.e 40%:, The.'lin&ar range,of
the instrument responee\G~d·at()a-,ng, whfoh Wat at
loost 10 times th& in$irum6ntal •llinit of dlreotfon. -Ate. level
of 0.1 ng, the relative itahditfd·:devlatlon ·wgs ·,...,5%, '·
Results and Discussion, :
'l'he PCB concen'tratlons·tn the vapor phase and ln the
particulate phase. (relatlv~ 'to both· the-air:voh:1me mld to
the particulate weight) a:re,g1v~1Ht1'Table I elonfwlth th&
average atmoephedc tem:p&rature during the··u~ptlng
period. This table haa been eubd.Mded·by the thre~ream•
piing sites. Samplaswere taken 28 tunes duringith.e period
October 1986 to August 1988; · · · · · . :, : :
Overall· Dlstrlbutloh~ -Con~tl'Atforttl -ot the'\'Ap6t end
particulate sample~w~ri{av&taked•t:,ver··a1l loM~~()ll.% ,~a
temperatures on a ·oong~n~M1pecif1c ,,bul,. · lf·eirhoma
PCBr. in Bloomi.ngton were derived only fron1.·the A.tool.or
1242 present in the Supertund dumpa;-these· a-Veragot
should ·oorrelate well wtth:thl'~lhf,bMttol'fof thlj ~10?
mixture. On a congoh1>tliy-Mnfen~tbtil•J~•~oula~
a correlation coeffl_olert, (r):~f Q.679.fot'th~~aj)or:-phale
and 0.672 for the particulate pha1e;<see Figure-2).> Por
F:on Hites ID:812 -855-1076
_,_;_,
, .. . ·.
~.-· .. ,.
1 V,'
• -ij'ui1 1
(l.000.•; ::
(l.0001 .·
;. o:Tinn.1 · •. ,, 1 · ,., IE-<)', ;..
·· 1 f· 0 (I.I lf).f ., 0.,1 · J ,! ,) I I/ .lo
111,. l\r.oolor .I .24~ no Mi. (Ill).
Flgur♦;2'. Regreap!QK.'pf me atmbspherlo oonoentretlons or 27 PCB
congei:iera in the vhi)Qr phase (le,lt a,cla) end In the particulate phase,
(rlght,4:k,l~) ver•utt~;oonoe~Jl'ijlon of ~ae congeners Jn Aroclot
12,2. ,. . ·•.,:..;; r.: .J ..
-' j;lf1 ,.---. ---., _ __.......,.__ ______ --, na~r1r
··'·. . .. , ~'· •• ~ I>
-·1 '_,,. =~ ··1·· .\. = \.
: (l ~ f ,: ..
r' ,t -(),I -~-' ~ h.1
~ ~ .. ::,
0:0q()J (1 0) ·• ,, 0 .. ·1q, lO ., JO ,. . Ave1~aie temperature,, deg C
Courthnu~" ···•····
Si1n,..kr:1
"
Figure•;• Aegreaalon~f the lcspeflthm of~ total (vapor plus portlc-
r · ulate) J'CB conc;enl71itfon versus the average air temperatute !or the
. thrH $1oomlnqton area aampllng .$!tee, ,Tho oorreletlon CO!!fflClents
· llre· 1111 tol10w11: ·Batotie'k:ir 0:890, COurthouse 0.852, and Sanders
,0.803,: .. ,., ......... -.~--.. ,.,, .. ,, . ··-" ., ---.. , ..
:1, : : ,,25 tle~®ll offr~otn~.the significance level i8 0.487 at the
.,:; .. : Jl9%JovQ.l!~C:s:omldance, .,Our~:vllues.are, thus, highly
. .., ,. ..,c.orrelated., 'Although-~eteiamore scatter than we would
..... : ... Jib, it is cl!l.ar .. that ,the;atmmpheri~ PCB composition is
· · · "limil~t tQ -that 't>f IAl'.bolot -1t42;cthe major PCB mixture
...... , ,,,pr~etit,jn_:thll d~PII,.; Note. that:the particulate concen-
~i, ... , 1, •• .tr~t1o;ns.(r.igl\'t;axl.8 otF~.2) are,a.fact.or of 100 less than
11, · .. • . !the vaPor•ph111H1 oonoentrations ~left·axis). Thus, the PCBs
· in Bloomington'&.•tmoapheH)Ql'tt,prhnarily in the vapor
•. :,PhllSe'f, .. · .. , 1 · 1· ·:-:.,,;! :_,.,,,· .. ;/
. '., .. · '" ,.;L;, S('.iJQJla.L.aAd •Site1C0Jilparbon11. The effects of
: • ,_., •·chang(ng•\lt,_'°ons or.-ln -1'ffectt·chahging air temperature,
· . are! apparent in'total vtpor PCB concentrations at all three
, 1,,; .•. sites, whe~:hi1tl10riconcehtratfons occur in warmer month.~
.,\ .,, and lowettn th(;wlnter,11.'ln~fa.ct;rthe most concentrated
, 1· •. ,,, • iVapor 188lllple ob&ewed during,the project was coUected on
1 ,,,.,, ., • , the wanneetday,at:the GoUl'thouae'elte. The sample taken
on .thb oold~t da:y. lro.m the: Bat<:helor site showed the
lowest vapor concentration in this study. A regression on
a congener-by-congener bai'ia lridlea.ted that 56 out of 61
congeners were elgnificantly <iorrelat&d (5% level) with the
· • ,,., -.avorage'lltmoaph~m.perature>On the other hand, the
,;::,, ;: ,p~ioµla~ oonoontrat~na.idid'n'c)texh.ibit a S88.IJ0ll!ll trend;
,only 18 congener& ,were &.ignifiCAl}tly correlated with at-
·. . • @ospl,leric tetnPOl'llture. ;: :. : .. . .
' , .,:.:-.,:,,, ·t,; l'-0.lmply ,ehQSvthiieft~t. we pombinedthe vapor o.nd
particµlate concenttattonsof:,aohiwnple into a total PCB
•· ;c ·., ,yalue end~gr:eaeed,the logaritharof these values versus
·. ,;;,. } : ; '1,the ,av,o~qe ait;t.e.DU>eraturt, rirllia ,analysis showed that
·. ·:th6,:'P(l'J3~?!lbft~J~oes:tli:t:tQe.~th increasing air tem-
·r)l .·y , Ml'.Gt.,.re ~tilltt,nN1e~1tet ·(eee rtgure 3). It is apparent
-,,·::, ··thaH~e ~urth'ounitte;tamote concentrated by a factor
...... h.~~.: ~f tf'J' 8\tho.tt mthor <>f.the1)theftwotites. Closer proximity
r:Y>;i';''.J>Mhej.~1:1lltlloue~-.c,it$,~.~~,~uperfund sites may be a
,J);, .. ,1~:, ,I.OJJ,i:~hf th11 dlfferen~. frliete ,may aleo be other, ns yet
1 ·• (,,.,, • •'~{'flcqlftli.ied,-90Ul'Ce•:near tlie-Oourthouae site. It is also . . ··1 . ' .. .. l . I
-:~1.r .:.1,~11 :·:F!l..v,lrqo.-;&~ •. T.totir1Q1.;·:vo1• 2$,:Np; 1Q, 1oe9
JAt·.J 1 7 '97 15:09 No .002 P.05
1.000 ---.-. -.-.. -, -.--.-., .. -.-.• ~ .. ~~-...._--, ('<)()
1-+-
(.l.1'1 ···•·-·
_, ,'
~ ; •. ' • I \ ,I, • l •
possible that this ai~ is simply exhibiting an elevated level
due to its relatively more urban charaoteristlcs. The
Ratche;or eite s~~wed the lowest totaf;vapor'.fCB con-
centrllt10ne deap1te the prtisenoe of tl{e interlrp storage
facility ~l kn1 .away (see tiibre 1). / ·:;:.
Pa.rtitulate PCitconc.entr{\tlons are V.fi,:Y low and are not
as si~ specific ~.WJlpore. Sanders and Courthouse values
were ll8ually the highest, as observed with:t}le vapor values,
but the differences'betwoon thr three sitell were:sometimea
~mall. We pre~e~~ particul~~ concentfe.tione'ln ,Table I
m two forms (m.l'lg m~s and 1n µ.g g-1 '.t~P). While tern•
poral differences in the form.er concentrations vary over
2 orders of magnitude In th~ et.tieme cases, 1in~' tsP-hased
nwnbers are mo.re .consistent .. This indic!te's'thltt" changes
in t-Otal atmospheric P~ic\llate c,o_nce,ntr.ati~nf.,COntti!>ute
t,o the scatter observed 1n'the voltune-balted ~ficentrat1ons.
Vapor-to-Partfoula:~:U..ttos, The obiiervilfon of low
partitulate PCB oon~ntratl6hs:~oriip11rt:!i, Wit:H'~ampled
vapors in atmospheri"d0sarnples made ·here ~d oy others
(1, 2, 11) s~gest.s that't,b:4 ·v~poHo·p~!fµ,Ial:.e,, {YIP) ratios
in our samples wou:td:oe hjgh;;·Jn fact;•W~ qbseryed widely
scatt.ered, but high (:►.160), -V/ P-.ratios eapoolally for the
lower chlorinated tS6fiJel'i~\'s." · ., ... ,., · ·1• ,:,·-·•,. w · '·,
The functional relationship betwoon ~he V/f' ratio and
atmospheric t.emp~rati.lA lias_been. reviewed by Bidleman
(?4); he tell~ ua that ~he~~¥.at~thpi io~ t~&:: V[P ,~atio is a
linear fun_ct1on of thfl il~il)~b~ ,o~ the _aunb!p~~ri_c tem-
perature (m degrOOB K&lvin),-'I~ :rn0$t cases, th6 particulate
concentrations in•th!s 1t~lat~?l)S~ip ~r, ~xp~e,~eij:·rela~ive
to the TSP. In our:caee, ho~ev~r,·\'Ve·uwionly'a shght
improvement in th~•r1f~ak,t'i! 'ifTSP\fas'U'ic\\ldedi thus,
for simplicity, w~ri'egtll'M'(lt;i 'Ut~·ldgai'lt~ ijf the'-V / p ratio
versus the reciprocal o·f the averag~ teii)peratute. The
correlation coefficient& betWeet) thes!l tw9.·parawe~n were
significant (at th&~% lev-el)f6t 5' o6ng&n()rii ;,ul of 61. As
examples, Fig';11e'4 ehowe l)lo~'o~ logJY!f>s;.rps 1/T
for congeners with~~lJ.PA:-C n;~;~~,. 149, ,nd 18.Q, one each
from the pent.a•; hlld-,>&'!dhept.«ohlorol!io'inol◊g\le ola&e6s.
All three congeners ehowe<1 sirnilaHren'd5 witll:tempera·
ture. Congener 180 e~~ited V / P ratio~ 6' tlriiae 1$'l8s than
the others, some •of.th'enf falling -below 1, 'Ntite that our
regresaion analysis revealed no differenooa'b'~!ween the
three sampling 4l~;'f«gg'eatbil'lliat 'Uie'"V'/.P. ratio is
consistent at a1glventlnfo'/?' ;: . !.• .;,,·~ : .. ,;· 1 ,: ..
Comparison-to Olbo,'Sltet • .'-rtie-~hno~~ijeric PCB
?once~trationa' from:~(.~ 15~1.lt!r 4titl,f'.!.ott! ... · 8~,t,t 00the~11, mcludmg contarn1rlaled 'afttl •.r~:Moffi ·sitee, ar41wen m
Table II. Givan th'tflitfe'll'ff~'t'ths.t'tefil~'~i'atllle Hems
to have on the atm:01p~lU'i~'<:<>1i~tmttb.'tl6fiij htlsttlven the
(largely unknown) dl~~~riMl!l i~'•. ltnpling te~peratures
represented by'thlflle.taln ~i61e tVlhdalfffoiilttto com-
I
I
\
F:o n Hit es ID:812-855-1 076
. .
Tabl,.JJ~ Qp~p.a.-l,o~ P~,M.U1m>he.rlo :-pee Co noon trn tioau1
• iie. 1-W-)~.~) ,'\;. .. ,, .... ,4\, .. \•,:.i....,:: i. •..• ,.,;,"' •. , .,J ••• ,.;. .. \~:• · ,\ ' •
• ,·.,-. I ! . ' ' ' ': .. ,~ . , . ,v4por
Bloomington, IN._, ''.·.·•.~·· ·•· ...... ,q;•.·.,J, ·
Monroe County Courthouse
partlc•
ulate ref
&Ummer 8.8 0.049 g
winter 0.68
Batchelor Middle School
e11mmer
winter
Banderi School
1.7
0.27
aummer 1.9
winter 0.88
Nea:J'e Landfill (a PCB dump)~
upwind 8l'i
hot.spot 8650
downwind 1800
aft.er cleanup
upwind 225
ho~pot 8900
downwind 925
Ontario, Canada
PCB dump during :cleanup
0.06
0.021
0.027
0.074
0.018
8
workelte 123 117 13
6 1n from source 92 61
·· ~"'99 tti fron'!''i!otlr~ · 17 23
>100 m from eouroe 4 8
Milwaukee, WI<
'Winter 2.25 0,37 JO
Madieon, WI•
summer
'Mfonliapoll11, MN
Chicago, Il.
Lake Michigan~
eumrner
Lake Superior
8•year m6M BWlllller
Isle Royale, MI1
summer
wlnt.er
Gulf of Mexico
mean of five off-ehore eltee
mean of three on-!lhore elt.e11
Antarctica
mean of four off-ahora elWi
Indian, Ooe11n..
30-70° &0uth latitude
7;49 0.2
7,5d
5.6'
0.87 0.13
2.8 0.06
0.59 0.025
0.111
0.0861'
10
9
5
10
0
1
4
4
it Reported as geometric mean of several observations.
biuipomd.ea Aroolor ,1242. ,~Sum of three Aroclon. dComblned
vapor e.nd pa:rtioulate· con~ntrations. •Vepor-to-partlculuto par-
titionlnr not Nported. 11/apor oonoentratfona only. 'Thi~ study.
pare thaae values. Nevertheleu, it ia interesting to note
. that the atm01pheric PCB values in Bloomington are lower
than in some other cities around the Midwest. This is true
despiui the widespread PCB cont:aminatlon in Blooming-
ton. The highest mean Bloomington vapor value is similar
to those :from·Madison; Minneapolis; Chicago, and the
relativelY, remote Isle Royale. Thia euggeste that there is
little difference in atmospheric PCB concentrations as a
function of location .. In faot, the Batchelor and Sanders
eitM in Bloomington show lower,vap0r concentrations than
at tale, .R()yale, but these tilte& ehow comparable values to
anothe.r o"8ervation on Lalte Superior. Lake Michigan and
Bhomllne ,-reas aroWld the Oul.fof Mexico show vapor PCB
concentrations eomewhat lower than the Batchelor and
Sanders sites, but·thl_s may not be significant. Bloom-
ington p~ticulatti values, although widely 1cattered, fire
aimllar t<f llle Royal&, ,but much less than summer values
· repo~ for Lake Miclilgim and Milwaukee and Madison,
WI. The1poncentratioM In the viohilty of Neal's Landfill
•.are •m~ch !higher than the :oonoontratlop at the Courthouse,
16 kmto~he~t. :crhls may~uggeetthat the atmosphere
JAH 17'97 15 :11 Ho .002 P.06
is effectively diluting vapor~phase PCBe emitted from this
and other Superfund si~. ·., ' · ··
Major differenooe aleo appear'ln Table IL: 'Wi~ter eon•
c~ntrations for Mµ~~~~ ~.Jl [a~~.o!_?-""l~,~~e,r ~~ wmter concentrationil .m·13looinlngton. Th& vapor con-
centrations from the Ontario oleanup site aria 'itll 1mr•
roundings are less than observed at Neal's Landfill. The
particulate values at this Ontario site, however, are un-
usually high in comparison with vapor conoentrations.
High conoentratlone of airbome particulates during
clfianup of contaminated aoil may have influenced thia
result. Other major differenclt8 are the concentrationa
roported for the moat remote sites in Table II: Antarctica
and the Indian Ooean. These concentratione e.M an order
of magnitude lower than the siteB in Bloomington; This
may be an indication of PCB stratification between the
Northern and Southern hemispheres.
Acknowledgments
We are grateful to I. Brum for technical assistance
throughout the project. We are indebted to Monroe
Cormty, IN, the Monroe County Community School Corp.,
and Mr. Jerry Chasteen for allowing us to locate and op-
crnto our air samplera on their ·properties. ·
Regl1itry No. Aroohlor 1242, 53469-21·9,
Literature Cited
(1) Swackhamer, D. 1,,; MoV60ty, B. D.; Hit.ea, R. A. Envil"On,
Sci. Techriol. lt88, 22, 664--672.
(2) Giam, C. S.; Atlaa, E.; Chan, H. S.; Neff, 0. S. Atmo,.
Environ. 1980, 14, 6H9.
(3) Jan, J.: Tratnik, M. Chemosphere 1988, 17, 809-818.
(4) Tanabe, S.; Hidaka, H.: Tateukawa, R. Cherrwsphere 1983,
12, 277-288.
(5) Murphy, T. J .; Formanaki, L. J.; Brownawell, B.: Meyer,
,J. A. Environ. Sci. Technol. 1985, 19, 942-946.
(6) Hollod, G. J. ThBBls, Univer&lty of Minneeot:a. Minneapolla,
MN, 1979.
(7) Lewis, R. G.: Martin, B. E.; Sgontz, D. L.; Howea, J. E.
Environ. Sci. Technol. 1985, 19, 986-991.
(8) Sgonh, D. L.; Howes, J. E. Ambient Monitoring for PCB
After Remedial Clean~p oJ Two Land.tllla Jn the Bloom-
ington, Indiana. Area. U.S. ·Environmentel Protection
Ageney Report, EPA/600/4-86/018 (NTIS No. PB86-
1775S2); 1986,
(9) Eisenreich, S. J.; Looney, B. B.; Hollod, G. J. In Phy1ical
Behavior of PCB in the Great Lake.,; Maokay, D.,,Pi11.ttereon,
8., Eiaenreich, S. J., Simmone, M. 8., Eds.; Ann Arbor
Science: Ann Arbor, MI, 1983; Chapter 7,
(10) Doskey, P. V.; Andren, A. W. J. Great Ldkes Res. 1981, 7,
15-20.
(11) Wittlinger, R.; Balltchmltter, K. Chemo1phere 1987, 16,
2497-2513.
(12) Bidleman, T. F.; Olney, C. E. Bull. Environ, Contam.
'l'oxicol. 1974., 11, 442-460.
(13) HOBein, H. R.: Gray, L.; McGuire, J. J, Air Poltut. Control
Assoc. 1987, 37, 176-178.
(14) Locating and E11timating Air Emi98ions from 8ourcea of
Polychlorlnated Biphenyla (PCB), U.S. Environmental
Protection Agency Report, EPA•450/4-84-007n (NTJS No.
PB87•209640); 1987.
(1 5) Murphy, T. J.; Rzeszutko, C. P. J. Great Lake; ReB, 1977,
-~. 805-312.
(16) Stachan, W. M. J.; Huneault, H.J. Great Lakes Res, 1979,
5, 61-68.
(17) AppliC11.tio11 for an Air Quality Permit to Con&truot a
Propoaed Bloomington Incineri11.tor Facility. We&tinghouse
Electric Corp., 1986. ·
(18) Lewia, R. G.; Brown, A. R.: Jackaon, M. D. Anal. Chem.
1977,49, 1668-1672.
(19) Albro, P. W.; Parker, C. E. J. Chromatog. 1979, 169,
161-166. .
Po n Hites ID:S:12-855 -1 076
1 .: d :.:1 , (20) ,Burdfok;· Ni1¥1/{8fdl~rriarl;r'l'}1F; Anal. Chem. 1981, 53,
. !· , . 1926--1929, · 1· ' 't~.Ht' t•'t,;~·~r'.'''
~• ,(21) F:oreman,,W.,T., BJdleman,·T,;F,J. Chromatos. 1985, 330,
:,,i,I'.• 'to .•ai• .,,2~7~l.~1~ .. ,(· • ,/~,: ._.,.;, .. , :,: ' • ·; '. .. •. _)2;Jqtj_,.~i:~~•:~~',~~ll,
1
M,:t~a~~ Z. Anal. Chem. 1980,
·.t
;.1, {1
•I
Jl ·•,fi
: ', '., l • , ~ ! , ' ..
., ,-;
' ~-• i i :; I I . I
I J ,,1 1
,. ••'
t:'., ., ...
r1 -:1•, • ii n -, .. ~-•-·r-J,:d 1.1'.::~-1 ;
. l . : ~.: ,l ' ... ~ _. ..
~·1 .::;~,'-. ~! .r r " ::,~'. .: ~!
.,,..;~.,!.,; .. ;;,.;, . ..:·,&.:.,( -~\} .. ··•'.i : ~;~•\<;
.:(,., :•(•1(: I (.i.,f .>'.•.,1 t-·-,,._'lr"'vr-•'-''
L :f ◄-.-..,,~; ,.;h•: ·i, .J ~•1 .. J'.) ·-:
.>. .. f
-,-.:.-
. '
··•!''" f
,\, ;, ~ t , i. I
:,
·-.•J•
'/ ·f•"' .. ,: ,. ., ...
.--.1':fI·,'t:
.··•···•,
,, ' I ' •• , • ,. • ' ., ,. lt. . ~ .
'j ·'. ·-\ f-'. \ : ~ •.
I,-\, .I f'-'"'••"-' i.~,
~---,_...-,,•::,}· -~ ;'--,'~(/~ l ,;I. · •• ;i•_,,L, ,iit_,f
,.
Ji. .•
,. : f' .... _ •• :.~'<i-~-
1;~ 1, • -,. ..
•1• -.·
JA N 17'97 15 12 No .002 P.07
-· , . •· •· :, :. , .. " -t1 ,.. P ,.: .. :i:•:· ·-,. ; • { .... ,~~ , .• · r
(2:J) Mullin, M. n;'PCli Woikiiiiop, Grop,hJl,t, ¥J, 1985 .
(24) Bidleman, T. F. Enuiron. ScL Technol.1988,'22, 881-'-387.
Receh>ed for reukw Nouember 1$,.1988. Accepted April 25, 1989.
This work was supported by th(!. ~nvironmental Technology
Division of the West~~~!?~~.JFl~~~ric Corp. ,:,:
"!·•·'"
.,. ,,-_,
; ... ·,;
State of North Carolina
Department of Environment,
Health and Natural Resources
Division of Air Quality
James B. Hunt, Jr., Governor
Jonathan B. Howes, Secretary
Alan W. Klimek, P.E., Director
MEMORANDUM
TO: Alan Klimek, P.E.
NA
DEHNR
January 17, 1997
Lee Daniel, Chief, Technical Services Section
FROM: Laura S. Butler, P.E., Chief ~~/
Air Permits Section 1/
SUBJECT: PCB Air Emissions
Hirschhorn Recommendation January 13, 1997
PCB Landfill
Warren County
Attached please find Joel Hirschhorn's January 13, 1997 recommendation for
"Preferred testing for PCB air emissions; peer review". By way of this memorandum, I am
requesting your review, comment and recommendation on Mr. Hirschhorn's report.
Your earliest review is requested. The Warren County PCB Landfill Working Group
continues to be concerned about potential human health effects and environmental
impacts from PCB emissions.
LSB:bw
c: Bill Pate
Bill Meyer
Lee Daniel
George Murray
Ernie Fuller
Attachment: ( 1 )
P.O. Box 29580, Raleigh. North Carolina 27626-0580
Voice [phone] FAX [phone]
An Equal Opportunity/Affirmative Action Employer
50"/o recycles/10% post-consumer paper
FROM NC DEHNR SO LID WA STE SE CTI ON 73 3 -4810
p . 1
t•. I
de~ ..J6 .. ,.1
7),.. t
f'l-'J ?honef
Fad Fax,
Janutiry 1,, 1997
To: M"ike Kelly ·
From: Joel Hirschhom
Subject: Pl·efetted testing for PCB air emissions; peer review
As to the prefetred teSting m~hod$ fot detQmmti\,g ''wl\eth~r the:i:e is a problem11 at the Wui·en
Couuty Laud:611. X now have had time 'to research the i,me:. My reoommendation to the state is
that jf any testing for PCB air emissions is oonduoted by th@ state that it &hould follow the
~1ethods ei:aployod by Indiana University in its wc,rk "Lt,ng-Term Measurements of Atw.osph01ic
Po.Jyrhlorinated Biphenyls in th~ Vicinity ofSuperfimd DtW)l)S,'' (M.H. Hemianson and R.A.
H'nes, F.:nvil'on. Sc;. &, 'roohnoJ,, v.231 n. l 0, 1989. pp, l2543-12S8).
TI1c experimental approaoh llsed in this relevant work was far more sonsitive and oomprehensive
than the EPA work in 1982, PCB level~ io air at three community loontions, miles ft·om
Superfimd dump sites, werQ mAdo.
The Indiana work correctly pAid attentlc,n to the 111.rge effect of ambient tempei·atun on PCB
levels,, Co1Tcetly rneaw.red PCBs in botb vapor and particulate forms, used high volume sampling,
had a deteoticmlimit very muoh lower thm in EPA's work, and measured and reported total PCB
congcne1 levels rather than only two commercial mixtures (i.e_, Art>clor$).
'Noto that Professor Hites, one of the authors ofthe above study, i& one of the two pe.l'sons Bill
Meyer asked for a review of the work cc,ndueted by EPA and my commCIJts on it, However, I
stro1tgly object to th~ way Bill Meyer $ought external r~icws. For example~ I would want t()
know th~ extent t() which Prof. Hites has funding from IWA. And fr<-m whom di.d Bill Meyer
obtain the names of the two persous h~ ohose to solicit reviews fi'om7 Nor do we. know what
kind of statcim:nts may have been made to Prof Hites by BID Moyer. In any event, there is a clear
published record of the .method, used by Prof. lft.tos vo.rsus those used by EPA. Intere~io,gly,
althouih the 1985 paper by Dr, Lewis et al of EPA wai. referenced in the Indhnl\ paper, not OJLC
piece of data was oned, even though a larg& tunount of publish~ dllta on PCB air levels was
presexi.tod by .Pro£ Hites in hi6 paper,
Based on my professional experienoo, J would not e>qJect Prof: Hites to say anything that Dr.
Lewis of.EPA m,ight find disagreeable, because Ptof Hit~i: wou.ld h1we a lot more to lose than to
gain pmfessionally. The Hme is true for the Canadian regulatory official asked for a review. In
fact+ most professionals placed in the position that Bill Meyer c1·eated would bo biaGed in fqvor of
a senior EPA person . In orde.r to nvoid this si,guiaoant OQ».straint, someone in Bill Meyer's
position would have to CICAte safeguards~ such as by assurlug OODJJ)lete oon:SdentitHty of the
H7Vic:wcr's ic:l~ntify, A,.d of course. if Bill M~yer wantod to be fair, l10 would. have psked me to
nmuinau, penon~ for ;:ucl1 a peo.r ~v.
cc: Technical Committee
i ' , ....... ,,, I I.:\.; • •
6-1 3-1 S:➔S)6 9 : 3 6Pt 1
cc.. ~(./
rc....-r
(-.'I-' J
January 13, 1997
To: Mike Kelly
From: Joel Hirschhorn
Subject: Preferred testing for PCB air emissions; peer review
As to the preferred testing methods for detennining "whether there is a problem" at the Warren
Couuty Landfill, l uow have had ti.me to research the issue. My recommendation to the state is
that if any testing for PCB air emissions is conducted by the state that it should follow the
methods employed by Indiana University in its work "Long-Term Measurements of Atmospheric
Po1ychlorinated Bipheny1s in the Viciruty of Superfund Dumps," (M.H. Hermanson and R.A.
Hites, Environ. Scj. & Techn.ol, v.23, n. 10, 1989, pp.12543-1258).
The experimental approach used in this relevant ,..-ork was far more sensitive an.d comprehensive
than the EPA work in 1982. PCB Jevels in air at three community locations, miles from
Superfund dump sites, \Vere made.
The Indiana work correctly paid attention to the large effect of ambient temperature on PCB
}eyels, correctly measured PCBs in both vapor and particulate forms~ used high volume sampling,
J,ad a detection limit very much lower than in EPA's work, and measured and reported tota1 PCB
congener levels rather than only two commercial mhtures (ie., Aroclors).
Note that Professor Hites, one of the authors of the above study0 is one oftl1e tv..10 persons Bill
Meyer asked for a review of the work conducted by EPA and my comments on it. However, I
strongly object to the way Bill Meyer sought external revjews. For example, I would want to
know the extent to which Prof Hhes has fun.cling from EPA. And from whom did Bill. Meyer
obtain the names of the two persons he chose to solicit re\iews from? Nor do \-Ve know what
kind of statements may have been made to Pro£ Hites by Bill Meyer. In any event, there is a clear
published record of the methods used by Prof. Hites versus those used by EPA Interestingly,
although the 1985 paper by Dr. L~is et al of EPA was referenced in the Indiana paper, not one
piece of data was cited, even though a large amount of published data on PCB air levels was
presented by Prof Hites in his paper.
Based on my professional e>...-perience, I would not expect Prof. Hites to say anythlng that Dr.
Lewis of EPA might find disagreeable, because Prof lfaes would have a lot more to lose than to
gafo professionally. The same is true for the Canadian regulatory official asked for a review. In
fact, most professionals placed in the position that Bill Meyer created would be biased n1 favor of
a senior EPA person. In order to avoid this significant constraint, someone in Bill Meyer's
position would have to create safeguards, such as by assuring complete confidentiality of the
reviewer~ s identify. And of course, if Bill Meyer wanted to be fair, he would have asked me to
nominate persons for such a peer review.
cc: Technical Committee
P . 1
. State of North Carolina
Department of Environment,
Health and Natural Resources
Division of Air Quality
James B. Hunt, Jr., Governor
Jonathan B. Howes, Secretary
Alan W. Klimek, P.E., Director
MEMORANDUM
TO: Alan Klimek, P.E.
January 10, 1997
FROM: Laura S. Butler, P.E., Chief ~
SUBJECT: PCB Landfill
Warren County
.AVA
DEHNR
Attached are three documents prepared by the Division of Epidemiology concerning the
review of recent reports on air emissions from the Warren County PCB Landfill
(Attachments 1, 2 and 3). The Division of Epidemiology has recommended that additional
PCB sampling be performed at the Warren County PCB Landfill.
Both the Division of Air Quality Raleigh Regional Office with the assistance of the
Technical Services Section and the Division of Waste Management have agreed to
conduct air monitoring at the Warren County PCB Landfill (Attachments 4 and 5).
I recommend that the air monitoring be conducted with oversight of the Department to
ensure that the air monitoring is accomplished with the corporation and coordination of all
interested parties, including the Division of Air Quality, the Division of Waste Management,
the Division of Epidemiology and the Warren County PCB Landfill Working Group and its
science advisors, Dr. Joel Hirschhorn and Patrick Barnes.
Please advise me if you need additional information.
C: Bill Pate
Bill Meyer
Lee Daniel
George Murray
Ernie Fuller
Attachments (5)
P.O. Box 29580, Raleigh, North Carolina 27626-0580
Voice [phone] FAX [phone]
An Equal Opportunity/ Affirmative Action Employer
50"/o recycles/10"/o post-consumer paper
State of North Carolina
Department of Environment,
Health and Natural Resources
Division of Solid Waste Management
James B. Hunt, Jr., Governor
Jonathan B. Howes, Secretary
William L. Meyer, Director
SAMPLING PLAN SUMMARY FOR:
MEASUREMENT OF FUGITIVE ATMOSPHERIC EMISSIONS OF
POLYCHLORINATED BIPHENYLS FROM THE PCB LANDFILL
WARREN COUNTY, NORTH CAROLINA
Sampling Plan Date: January 9, 1997
Sampling Plan Preparer: Pierre Lauffer@<,
Health and Safety Coordinator
Division of Waste Management
Raleigh, North Carolina 27605
Project Participants:
Pierre Lauffer, HWS-Health and Safety Coordinator (Project Manager and Sampler)
John Kirby, HWS-Environmental Chemist (Project Chemist and Sampler)
Projected Sampling Dates: February, 1997
Site History:
Between June, 1978 and August, 1978, over 30,000 gallons of industrial waste material
identified as polychorinated biphenyls (Arochlor 1260 and 1262) were discharged deliberately
along the shoulders of approximately 210 miles of North Carolina highways. In June, 1979,
EPA approved a tract of land (previously used for agriculture) in Warren County, North Carolina
as the disposal site for the PCB-contaminated roadside soil. The landfill ( constructed in 1982-
1983 and permitted under the Toxic Substances Control Act (TSCA) contains about 40,000 cubic
yards of soil contaminated with PCBs. The concentration of PCBs in the landfill range from 150
to almost 900 part per million (ppm), averaging about 350 ppm (1), based on 1994 subsurface
soil sampling results (these were retrieved from the bottom of the vent).
P.O. Box 27687, Raleigh, North Carolina 27611-7687 Telephone [phone] FAX [phone]
An Equal Opportunity Affirmative Action Employer 50% recycled/ 10% post-consumer paper
Polychlorinated Biphenyls (PCBs):
PCBs are a family of aromatic compounds consisting of two benzene nuclei bonded with
two or more chlorine molecules. The PCBs of concern in this study (the type existing in the
landfill) are the Aroclors.
Aroclors are characterized by four digit numbers. The first two (the number 12) indicate
that the mixture consists of biphenyls. The second two digits state the percentage by weight of
chlorine in the mixture (2).
They tend to be colorless to light yellow oily substances with a specific gravity of 1.4-1.5
(3 ). Due to their stability and nonconductive properties, PCBs had many industrial uses
including use as insulation coating in electrical capacitors. PCBs possess, as a whole, high open
cup flashpoints (348-356° F.-Aroclor 1242, none for Aroclor 1254 and 1260), but do readily
distill at temperatures above 325°F. The distillation range for Aroclor 1242 is 325-366°F; for
Aroclor 1254, 365-390°F; Aroclor 1260, 385-420°F (4).
Toxic effects from exposure to PCBs in human include chloracne, pigmentation of skin
and nails, excessive eye discharge and swelling of eyelids, and gastrointestinal disturbances.
PCBs are considered carcinogenic (5).
Because of their high toxicity, detrimental harm to the environment and stable molecular
structure (structure remains intact in the environment for long periods ohime) PCB manufacture
was discontinued in 1976.
Objective:
The objective of this study is to determine if there are uncontrolled PCB emissions
originating from the Warren County PCB Landfill. This study will involve ambient air sampling
to determine if PCB contaminated air particulate matter and vaporized PCB (Aroclors 1242,
1254, and 1260) emissions are present. Retrieved samples will be analyzed by an independent
American Industrial Hygiene Association accredited laboratory. Southern Testing and Research
Laboratories, Inc. of Wilson, North Carolina has been selected for conducting the sample
analysis. The results of the analysis will be submitted directly to the workgroup and DWM. A
risk assessment will be requested from the DEHNR EpidemologyDivision to determine if the
PCB emissions (if found) are a risk to the surrounding community.
Materials and Methods:
Air sampling will be performed with low-volume (L V) and high-volume (HV) constant
air-flow sampling systems. The components of the sampling systems consists of battery operated
constant air-flow pumps (L V or HV) (Gilair Pump by Gilian™, Models: Gilair3(L V) and
Gilair5(HV)), 13 mm Gelman Swinney filter cartridges with 13mm, lµm pore-sized glass fiber
filters manifolded to ORBO™-60 100150mg, 6 x 70mm florisil sampling tubes.
The pumps will be calibrated by the Gilian Gilibrator (digital calibrator) prior to and after
each sampling period. The glass fiber filters will be precleaned prior to loading filter cartridges
by the North Carolina Public Health Laboratory and loaded at the laboratory by laboratory
personnel. The sampling head will consist of glass fiber filter cartridge followed by the florisil
tube. The two will be manifolded together by tubing. The purpose for this sampling system is to;
1) catch any possible PCB contaminated particulates from the air, 2) to filter PCBs which may
have vaporized. This system will then be manifolded to the sampling pump with plastic tubing.
There will be three sampling periods of eight hours. The sampling sessions will begin at
roughly 10:00am, 6:00pm and 2:00am. Each sampling period will be separated by one day.
The reason for this delay between sampling events is to provide time to recharge the air-flow
pumps (they require 16 hours ofrecharge time). Each sampling period will consist of seven air-
flow pumps and sampling units. One unit will be located inside the vent on top of the landfill.
Two more units will be located two meters downwind from the vent (these will be seperated
from each other by one meter). Two other units will be located diagonally downwind at the
landfill fence line. One unit will be located 200 meters straight downwind from the vent. The
air-flow rate will be calibrated to NIOSH guidelines. NIOSH Sampling Method #5503:
Sampling for Polychlorobiphenyls states that air-flow rate should be 50-200cc per minute. The
flow-rate per unit will correspond to its distance from the vent-the further the pump is from the
vent, the greater its flow-rate. Another sampling unit will be located 200 meters upwind and will
act as a background sampler. Approximately fifty samples will be obtained. Prior to each
sampling event, the wind direction and temperature will be indicated by a portable weather
station established on the east side of the landfill.
Funding from the DWM operations budget rather than the PCB detoxification budget will be
funding this project.
NOTE: Please see attached diagram of the landfill showing sampling locations and a copy
of the NIOSH lab procedure.
PKL/H&S/HWS/Dec96.
PCB LANDFILL-WARREN COUNTY
I
I i
Qc;AQ.te pond
I I l__ad
L.__J
I
I
* background
16 7 .5 meters
183 yards
wea1her station
sampling point -*
200yds* ➔
I
I
I
I
-. POL YCHLOROBIPHENYLS 5503
mixture: C,2H,0-,CI.
[where x = 1 to 10]
MW: ca. 258 (42% Cl ; C12H7CI~;
ca. 326 (54% Cl ; C12H5Cl5)
GAS: Table 1 RTECS: Table 1
METHOD: 5503, Issue 2 EVALUATION: PARTIAL Issue 1: 15 February 1984
Revision #1: 15 August 1987
Issue 2: 15 August 1994
OSHA: 1 mg/m3 (42% Cl);
0.5 mg/m3 (54% Cl)
NIOSH: 0.001 mg/m3/10 h (carcinogen)
ACGIH: 1 mg/m3 (42% Cl) (skin)
0.5 mg/m3 (54% Cl) (skin)
PROPERTIES: 42% Cl : BP 325 to 366 °C; MP -19 °C;
d 1.38 g/ml@ 25 °C;
VP 0.01 Pa (8 X 10'5 mm Hg;
1 mg/m3) @ 20 °C
54% Cl: Cr .365 to 390 °C; MP 10 'C:
d 1.54 g/ml@ 25 °C; VP
0.0004 Pa (3 x ,o•e mm Hg;
0.05 mg/m3) @ 20 °C
SYNONYMS: PCB; 1,1'-biphenyl chloro; chlorodiphenyl, 42% Cl (Aroclor 1242); and 54% Cl (Aroclor 1254)
SAMPLING
SAMPLER: FILTER + SOLID SORBENT
(13-mm glass fiber + Florisil,
100 mg/50 mg)
FLOW RATE: 0.05 to 0.2 L/min or less
VOL-MIN:
-MAX:
1 L @ 0.5 mg/m3
50 L
SHIPMENT: transfer filters to glass vials after sampling
SAMPLE
STABILITY: unknown for filters;
2 months for Florisil tubes (1)
BLANKS: 2 to 10 field blanks per set
ACCURACY
RANGE STUDIED: not studied
none identified
not evaluated
not determined
BIAS:
OVERALL PRECISION (~rT):
ACCURACY:
MEASUREMENT
TECHNIQUE: GAS CHROMATOGRAPHY, ECO t 3Ni)
ANALYTE: polychlorobiphenyls
DESORPTION: filter + ·front section, 5 ml hexane; back
section, 2 ml hexane
INJECTION
VOLUME: 4-µL with 1-µL backflush
TEMPERATURE-INJECTION:
-DETECTOR:
-COLUMN:
CARRIER GAS: N2, 40 ml/min
250 to 300 °C
300 to 325 °C
180 °C
COLUMN: glass, 1.8 m x 2-mm ID, 1.5% OV-17/1.95%
OF-1 on 80/100 mesh Chromosorb WHP
CALIBRATION: standard PCB mixture in hexane
RANGE: 0.4 to 4 µg per sample (2]
ESTIMATED LCD: 0.03 µg per sample [2]
PRECISION (Sr): 0.044 (1)
APPLICABILITY: The working range is 0.01 to 10 mg/m3 for a 40-L air sample (1). With modifications, surface wipe samples
may be analyzed (3,4).
INTERFERENCES: Chlorinated pesticides, such as DDT and DOE, may interfere with quantification of PCB. Sulfur-containing
compounds in petroleum products also interfere [5).
OTHER METHODS: This method revises methods S120 [6) and P&CAM 244 (1). Methods S121 (7) and P&CAM 253 (8) for
PCB have not been revised.
NIOSH Manual of Analytical Methods (NMAM), Fourth Edition, 8/15/94
POLYCHLOROBIPHENYLS: METHOD 5503, Issue-2, dated 15 August 1994 -Page 2 of 5 -----------------------------------•. ~. ·-
REAGENTS:
1. Hexane, pesticide quality.
2. Florisil, 30/48 mesh sieved from 30/60 mesh.
After sieving, dry at 105 °C for 45 min. Mix
the cooled Florisil with 3% (w /w) distilled
water.
3. Nitrogen, purified.
4. Stock standard solution of the PCB in
methanol or isooctane (commercially
available).*
* See SPECIAL PRECAUTIONS.
EQUIPMENT:
1. Sampler: 13-mm glass fiber filter without
binders in a Swinnex cassette (Cat. No. SX
0001300, Millipore Corp.) followed by a glass
tube, 7 cm long, 6-mm OD, 4-mm ID
containing two sections of 30/48 mesh
deactivated Florisil. The front section is
preceded by glass wool and contains 1 oo mg
and the backup section contains 50 mg;
urethane foam between sections and behind
the backup section. (SKC 226-39, Supelco
ORBO-60, or equivalent) Join the cassette
and Florisil tube with PVC tubing, 3/8" L x
9/32" OD x 5/32" ID, on the outlet of the
cassette and with another piece of PVC
tubing, 3/4" L x 5/16" OD x 3/16" ID ,
complete the union.
2. Personal sampling pump, 0.05 to 0.2 L/min,
with flexible connecting tubing.
3. Tweezers.
4. Vials, glass, 4-and 7-ml, with aluminum or
PTFE-lined caps
5. Gas chromatograph, electron capture
detection ~ 3Ni), integrator and column (page
5503-1).
6. Volumetric flasks, 10-mland other convenient
sizes for preparing standards.
7. Syringe, 10-µL.
SPECIAL PRECAUTIONS: Avoid prolonged or repeated contact of skin with PCB and prolonged or
repeated breathing of the vapor [9-11].
SAMPLING:
1. Calibrate each personal sampling pump with a representative sampler in line.
2. Break the ends of the Florisil tube immediately before sampling. Connect Florisil tube to
Swinnex cassette and attach sampler to personal sampling pump with flexible tubing.
3. Sample at an accurately known flow rate between 0.05 and 0.2 L/min for a total sample size of 1
to 50 L.
NOTE: At low PCB concentrations, the sampler was found to be efficient when operated at flow
rates up to 1 L/min, for 24 hours [4]. Under these conditions, the limit of detection was
0.02 µg/m3•
4. Transfer the glass fiber filters to 7-ml vials. Cap the Florisil tubes with plastic (not rubber) caps
and pack securely for shipment.
SAMPLE PREPARATION:
5. Place the glass wool and 100-mg Florisil bed in the same 7-ml vial in which the filter was stored.
Add 5.0 ml hexane.
6. In a 4-ml vial, place the 50-mg Florisil bed including the two urethane plugs. Add 2.0 ml
hexane.
7. Allow to stand 20 min with occasional agitation.
NIOSH Manual of Analytical Methods (NMAM), Fourth Edition, 8/15/94
POlYCHlOROBIPHENYlS: METHOD 5503, Issue 2, dated 15 August 1994 -Page 3 of 5
CALIBRATION AND QUALITY CONTROL:
a. Calibrate daily with at least six working standards over the range 1 O to 500 ng/ml PCB.
a. Add known amounts of stock standard solution to hexane in 10-ml volumetric flasks and
dilute to the mark.
b. Analyze together with samples and blanks (steps 11 and 12).
c. Prepare calibration graph (sum of areas of selected peaks vs. ng PCB per sample).
9. Determine desorption efficiency (DE) at least once for each lot of glass fiber filters and Fiorisil
used for sampling in the calibration range (step 8). Prepare three tubes at each of five levels
plus three media blanks.
a. Remnve and discard back sorbent section of a media blank Florisil tube.
b. lnj0ct known amounts of stock standard solution directly onto front sorbent section and onto
a media blank filter with a microliter syringe.
c. Cap the tube. Allow to stand overnight.
d. D2snrb (steps 5 through 7) and analyze together with working standards (steps 11 and 12).
e. Prepare a graph of DE vs. µg PCB recovered.
1 o. Analyze three quality control blind spikes and three analyst spikes to ensure that the calibration
graph and DE graph are in control.
MEASUREMENT:
11 . Set gas chromatograph according to manufacturer's recommendations and to conditions given
on page 5503-1 . Inject sample aliquot manually using solvent flush technique or with
autosampler.
NOTE 1: Where individual identification of PCB is needed, a procedure using a capillary
column may be used (12].
NOTE 2: If peak area is above the linear range of the working standards, dilute with hexane,
reanalyze and apply the appropriate dilution factor in calculations.
12. Sum the areas for five or more selected peaks.
CALCULATIONS:
13. Determine the mass, µg (corrected for DE) of PCB found on the glass fiber filter (W) and in the
Florisil front 01'/,) and back 0Mb) sorbent sections, and in the average media blank filter (B) and
front (B,) and back (Bb) sorbent sections.
NOTE: If Wb > Wtf10, report breakthrough and possible sample loss.
14. Calculate concentration, C, of PCB in the air volume sampled, V (l):
c = ( w + w1 + wb - B -B1 -Bb) 3 V , mg/m .
EVALUATION OF METHOD:
This method uses 13-mm glass fiber filters which have not been evaluated for collecting PCB. In Method
S120, however, Aroclor 1242 was completely recovered from 37-mm glass fiber filters using 15 ml
isooctane [8, 13, 14]. With 5 ml of hexane, Aroclor 1016 was also completely recovered from 100-mg
Florisil beds after one-day storage [1 ). Thus, with no adsorption effect likely on glass fiber filters for
PCB, 5 ml hexane should be adequate to completely extract PCB from combined filters and front
sorbent sections. Sample stability on glass fiber filters has not been investigated. Breakthrough volume
was >48 l for the Florisil tube at 75% RH in an atmosphere containing 10 mg/m3 Aroclor 1016 (1].
NIOSH Manual of Analytical Methods (NMAM), Fourth Edition, 8/15/94
POLYCHLOROBIPHENYLS: METHOD 5503, Issue 2, dated 15 August 1994 -Page 4 of 5
REFERENCES:
[1] NIOSH Manual of Analytical Methods, 2nd ed., V. 1, P&CAM 244, U.S. Department of health,
Education, and Welfare, Pub!. (NIOSH) 77-157-A (1977).
[2] User check, Southern Research Institute, NIOSH Sequence #4121-U (unpublished, January 25,
1984).
[3] Kominsky, J. Applied Ind . .t!Yg. 1 (4), R-6 {1986).
[4] NIOSH Health Hazard Evaluation Report, HETA 85-289-1738 (unpublished, 1986).
[5] Hofstader, R. A., C. A. Bache, and D. J. Lisk. Bull, Environ. Contam. Toxicol., 11, 136 {1974).
[6] NIOSH Manual of Analytical Methods, 2nd ed., V. 4, S120, U.S. Department of Health, Education,
and Welfare, Publ. (NIOSH) 78-175 (1978).
[7] Ibid, V. 2, S121, U.S. Department of Health, Education, and Welfare, Pub!. (NIOSH) 77-157-8
(1977).
[8] Ibid, Vol. 1, P&CAM 253
[9] Criteria for a Recommended Standard ... Occupational Exposure to Polychlorinated Biphenyls,
U.S. Department of Health, Education, and Welfare, Pub!. (NIOSH) 77-225 (1977).
[10] Current Intelligence Bulletin 7, Polychlorinated Biphenyls (PCBs), U.S. Department of Health and
Human Services, Pub!. (NIOSH) 78-127 (1975).
[11] Occupational Diseases, A Guide to Their Recognition, revised ed., -255-256, U.S. Department of
Health, Education, and Welfare, Pub!. (NIOSH) 77-181 (1978).
[12] Dunker, J. C. and M. T. J. Hillebrand. Characterization of PCB Components in Clophen
Formulations by Capillary GC-MS and GC-ECD Techniques, Environ. Sci. Technol., 17 (8), 449-
456 (1983).
[13] Backup Data Report for S120, prepared under NIOSH Contract 210-76-0123, available as "Ten
NIOSH Analytical Methods, Set 2," Order No. Pb 271-464 from NTIS, Springfield, VA 22161.
[14] NIOSH Research Report-Development and Validation of Methods for Sampling and Analysis of
Workplace Toxic Substances, U.S. Department of Health and Human Services, Pub!. (NIOSH)
80-133 (1980).
[15] Hutzinger, 0., S. Safe, and V. Zitko. The Chemistry of PCBs, CRC Press, Inc., Cleveland, OH
(1974).
METHOD REVISED BY:
James E. Arnold, NIOSH/DPSE; S120 originally validated under NIOSH Contract 210-76-0123.
Table 1. General Information.
Compound
Polychlorinated Biphenyls
Chlorobiphenyl
Aroclor 1016 (41% Cl)
Aroclor 1242 (42% Cl)
Aroclor 1254 (54% Cl)
CAS
1336-36-3
27323-18-8
12674-11-2
53469-21-9
11097-69-1
RTECS
TQ1350000
DV2063000
TQ1351000
TQ1356000
TQ1360000
NIOSH Manual of Analytical Methods (NMAM), Fourth Edition, 8/15/94
' ·. ' \
• POLYCHLOROBIPHENYLS: METHOD 5503, Issue 2, dated 15 August 1994 -Page 5 of 5
. .
Table 2. Composition of some Aroclors [15).
Major Components Aroclor 1016 Aroclor 1242 Aroclor 1254
Biphenyl 0.1% <0.1% <0.1%
Monochlorobiphenyls 1 1 <0.1
Dichlorobiphenyls 20 16 0.5
Trichlorobiphenyls 57 49 1
Tetrachlorobiphenyls 21 25 21
Pentachlorobiphenyls · 1 8 48
Hexachlorobiphenyls <0.1 1 23
Heptachlorobiphenyls none detected <0.1 6
Octachlorobiphenyls none detected none detected none detected
EEC, Inc.
. 107 WIND CHIME COURT, RALEIGH, NC 27615
January 8, 1996
Mr. Piere Lauffer
Division of Solid & Hazardous Waste
Solid Waste Section
Raleigh, NC
Re: PCB Analytical Cost
Dear Mr. Lauffer:
9 I 9-846-I O I 6
FAX [phone]
As per our discussions, we will be glad to provide analysis of PCB air samples
using the NIOSH 5503 method. The cost of analysis for each sample will be $
70.00/sample. This cost is for a 5-7 days tum-around time of analytical serv1ces.
All samples will be analyzed by an AIHA accredited laboratory. We understand
that there may be as many as 17 samples per day to be analyz_ed. After you collect
the samples, we can pick up the samples at your office o·r you can drop the
samples by our office.
If you need additonal information, please call.
Thank You.
Sincerely,
('(½e .... _
Mike Shrimanker, CIH, CSP, PE
President
Post-I~ Fax Note
To
Fu#
7671 Date I-S _ ')7
#of ► pa~s /
. ~-c 0 l"
• '.I>_.\ 4, c., .., ~'-
,,. A I fl ~
Soittliern Testing & Researclz Laborc,tories, Inc.
3809 Airport Drive (919) 237--H75 • Fax: [phone] Wihon, NC 278%
FAX TRANSMITTAL
Date Januarv 10, 1997 -~~--...---'-i.,____..........,. ___ _
The follovving pages are for:
Name Pierre Lau ff er
Location NC Hazardous \-Vaste Section
Fax No 919-715-3605
From Kim BauQ:hman -----.:c-__o=:.=.=_.=--,-----,===------
Total Number of Pages Sent (Including Cover Sheet)_~l __
Pierre,
Our LOD for NIOSH Method 5503 is 0.025 µg. Our analysis fee for this method is
$75 /sample. This includes five point standard calibration curves and two analyse s for every
sample (The backup Florisil section is analyzed separately to evaluate brea.1.."through.:) \Ve hope
to work with you on this project.
i...,.L_, _ _. __ , ••• ,1 ,.. • • • •
Sincerely,
!J n ,,.1 ,. /~)Q!l/V ,, rytVJ
~aughruan, PhD-; H
Technicai Director
01 /07/97 1 8:03 '6'9 19 541 3527 NERLIAAIRD /AfB
From:
To:
Date:
Subject:
Ronald Hites <[e-mail]>
RTPAREAL01.HERMES(LEWIS-
B
O
B
-
O
R
)

1/7/97 3:41pm
N>C> PCB Landfill -Reply
I am really sorry that I cannot be very helpful with this
request. It has been my experience that anytime I get
near stuff in which there are (or might be) lawyers
involved, l regret it.
I continue to have the highest respect for you and your
work, but I just do not have the time to get in the
middle of this controversy.
Ron Hites
School of Public and Environ. Affairs
Indiana University
Bloomington, IN 47405
voice 812-855-0193
fax [phone]
[e-mail]
. t I mo 7671 # al pages ► /
Post-It'• brand fax transmit a me . . .
To From tf3 oj ~
Co. Co.
Dept. Phone #
Fax # 7 s--J ~ OJ-Fax F
@ 00 11 001
01/0i:9i 12 :13 "B'919 5-11 0239 APRD .'~ERL-RTP !410 01:00 2
Post-It'" brand fax transmittal memo 7671 if 01 page,, •
From: Terry Bidleman <[e-mail]>
To:
Date:
BOB-DR LEWIS <LEWIS. [e-mail].~ Fax /t I _ ;J () z / Fax # 7 ; J-_ 3 b O)
1 n/97 10:50am .___-=~---"--'---';.......--'---~-----~---
Subject: Re: N. C. PCB Landfill -Forwarded
Hello Bob,
Yes, I did get the package from Bill Meyer shortly before Christmas. Be
assured that l will pick up the gauntlet and respond to it. Bill did not
give a time frame in which he expected a response, but I hope that this can
wait until February. January is the month in which we at AES submit
proposals for next year's work and I am also gone for a week to an arctic
workshop. Finally, I am writing a report on long-range transport of
pesticides in North America for a NAFTA group called Commission for
Environmental Cooperation, and this is also due at the end of the month!
A quick look at Hirschhorn's comments and re-reading of your paper indicates
that several issues are involved. One is the quality of the data and its
interpretation in the ES&T paper, and a separate one is the follow-up (or
lack of it) by the state in continuing to monitor the landfill. I see these
as two separate issues, although Hirschhorn wants to link them.
Essentially your paper reports measurements of PCBs around the Warren Co .
landfill which were made with LV samplers. Because of the low air volumes.
limits of detection were high and in fact most of the measurements were
below the LOD .. There were a few hits, and Hirschhorn is placing a lot of
emphasis on them.
Here is why I think a lot of Hirshhorn's statements are a bunch of hooey:
a) The analyses in the 1985 paper were done by packed-column GC and
apparently without silicic acid: or florisil separation of PCBs from OC
pesticides:-Thus it is quite likely that there could have been interference
from other OCs, especially toxaphene and chlordane. Hirschhorn makes a big
case of differences between 1242 and 1260, but from the chromatogram in Fig.
3 it is clear that 1242 would have been difficult to quantify except in
high-level samples --a good portion of it is masked by the solvent front.
For these reasons I can't see making a big deal out of the differences in
the two PCB mixtures, and I agree that the ambient air concentrations
reported are upper limits. The "false positive" results may well have been
due to interfering pesticides.
b) Hirshhorn's comments about the landfill being a source do not take into
account other potential sources of atmospheric PCBs. The biggest source -·
and the likely reason why ambient levels exceeded those predicted from the
dispersion model --is volatilization from contaminated soils in the region.
Kevin Jones argues that outgassing of PCBs from soils is probably the most
important source to the atmosphere in the U.K. (lnternat. J. Environ. Anal.
Chem. 59: 167-178, 1995). Also, a large area of NC was contaminated by the
PCB spills. Although a lot of cleanup was done, surely all the contaminated
soil was not removed. The measurements by Kathy Macleod show levels of 1260
in the air at the contaminated roadside sites before and after cleanup
(paper by Kathy and yourself in the ASTM volume 721). Concentrations at
roadside areas after cleanup were in the range of the highest levels found
in ambient air which were reported in the ES&T paper. So -it is it likely
that the ambient PCBs came not from the landfill but from contaminated soils
in Warren and surrounding counties?
c) On one issue I will agree with Hirshhorn: A follow-up study of the
landfill and indeed of ambient air PCB levels in the region is needed. We
0l:0i:9i 12: 1-1 '6'919 5-ll 0239 APRD ::\'ERL-RTP
know a lot more about PCB properties now and congener-specific analysis is
now routine. Sophisticated statistical techniques such as principal
component analysis can be used to discriminate PCB patterns in soil and air
(see Grundy et al., ES& T_ 30, 2661 , 1996). It is sad that Hirshhorn is more
intent on assigning blame than on investigating the problem.
I need to write this up in a more organized manner and respond to other
issues raised by Hirshhorn, but as explained in the first paragraph, I am
having difficulty finding time to do it right now. If Bill Meyer really
needs a response before the end of the month, I will provide one. But if he
can hold off until early February, it would certainly lighten my load.
Please let me know the situation.
We are continuing work on OCs in air, with a focus on the Great Lakes and
the Arctic. Some current projects are:
a) investigating sources of OC pesticides to the Great Lakes. This involves
air sampling over Lake Superior and also in suspected source regions such as
Alabama and South Carolina. Present-day levels of toxaphene are much higher
in these states than over the Great Lakes!
b) Using enantiomers of pesticides to discriminate racemic and non-racemic
(soil) sources, working with Renee Falconer of Youngstown State. So far
this has been most successful for chlordane and heptachlor epoxide. From
enantiomeric patterns we can distinguish chlordane in soils from chlordane
that arises from termiticide use. Also, the heptachlor epoxide in air
appears to come from ~oil emissions rather than oxidation processes in the
atmosphere.
c) Ai r-water gas exchange and distribution of pesticides in the Arctic. We
now have good coverage of both the eastern and western Arctic Ocean and have
made estimates of gas exchange for HCHs and toxaphene. Concentrations these
pesticides are higher under the ice cap of the Canada Basin than anywhere
else in the world!
d) Finally, Tom Harner is finishing his work on partitioning of PCBs and
other semivolatiles to Chicago aerosols. This is actually a project that
was supported by you guys under a cooperative agreement. Tom finds that the
octanol/air partition coefficient may be an improvement over vapor pressure
for describing partitioning to urban particles. Also, we found substantial
concentrations of polych\oro-naphthalenes (PCNs) in Chicago air --up to a
third of PCB levels!
Best wishes,
Terry
>Content-Type: text/plain
>Content-Disposition: inline
>
> Terry,
>I especially need your help on this problem since Ron
>Hites returned the package to the State of North Carolina
>saying that he could not help. Attached is an e-mail I sent
>to Ron asking him to reconsider.
>
>Bob
14)0 0 2 :002
State of North Carolina
Department of Environment,
Health and Natural Resources
Division of Waste Management
James B. Hunt, Jr., Governor
Jonathan B. Howes, Secretary
William L. Meyer, Director
MEMORANDUM
TO: Linda Rimer
Henry Lancaster
FROM: Bill Meyer {JAi\
DATE: January 6, 1997
SUBJECT: PCB Landfill
AVA
DEHNR
The staff has proposed an air sampling event for the PCB Landfill air vent. The
vent is the sole source of air emissions from the landfill. There are existing differences of opinion
as to the threat of exposure from past air emissions. The vent has been filled with a carbon filter
system to sorb constituents and prevent future exposure. The purpose of the proposed sampling
is to provide additional data on air emissions to determine risk, both past and present.
In addition, the staff has proposed to monitor the surface liner for gas leaks
(methane). If methane is detected the specific location ofleaks can be identified. If the landfill
surface liner is not leaking gasses, then the probability of water leaking into the landfill is very
unlikely. This is based on the significant difference in permeability ofliners to water and gases. It
is a very simple but effective means of determining the status of the liner.
The staff is prepared to perform both sampling events. Both sampling events
would add factual data rather than the current speculative nature of air emissions and liner
leakage. However, in addition to approval of sampling protocol and questioning the need for
sampling, it has become apparent that the Working Group's real concern is with the state's role as
owner/operator, state sampling, and state analysis of samples. Based on this concern, it is
apparent to the Division of Waste Management that only independent sampling and analysis may
satisfy the Working Group.
We have several choices. One is to proceed with a state sampling effort. Another
is to have the state sample with the Working Group oversight and use independent labs for
analysis. The third will be for complete independent sampling and analysis. Either of the latter
two options will require Department funding. The compromise option is state sampling with
independent analysis. The division does not think that the continued state sampling and state
analysis will be accepted by the Working Group.
P.O. Box 27687,
Raleigh. North Carolina 27611-7687
Voice 919-733-4996
N!JC f #M)Ji/4M€1
FAX [phone]
An Equal Opportunity Affirmative Action Employer
50% recycled/ l 0% post-consumer paper
Page 2
This experience is having a very frustrating effect on the staff and the belief that ·
the Department and Administration do not recognize nor support their efforts.
In order to resolve this issue the Division of Waste Management would like for
Department approval and support to plan, sample and analyze air and methane on the landfill. The
Division of Waste Management has the staff and equipment to accomplish this and is willing, with
Department support, to address concerns raised by the Working Group on the lack of
independence of this effort. If this is not the appropriate process, the Division will need financial
commitments for the Department for at least independent analytical activities. We need a
resolution from the Department on these issues as soon as possible.
c:wpfiles/pcblf7smpling.mem
(C6-
From:
To:
Subject:
Ron,
BOB-DR LEWIS
internet."[e-mail]"
N. C. PCB Landfill
Post-It '" brand fax transmittal memo 7671 # of pages ► /
To From ,d'7 / / .. !06 0 f...-e<.,._J;J
Co. Co.
Oept. Phone i:
Fax # 6 / -J :) Z / Fa x" 7 / ;--.J b ())~
I was sorry to hear that you are unable to help us refute the criticisms lodged against me and the EPA in
connec:ion with the PCB monitoring we did at the Warren County landfill. Perhaps the package that the State sent
to you was a bit overwhelming, especially at this time of the year.
In brief, we monitored the landfill shortly after it was closed in 1983 using low-volume PUF air samplers and
GC/ECD both for within vent measurements and ambient air measurements. We found only very low PCB
concentrations in the vents, except for the main vent, where we found 120 ug/m3 of Aroclor 1242 and 2 ug/m3 of
Aroclor 1242. Over two days of monitoring, all ambient air samples were found to be negative for 1242 (LOD 6
ng/m3) and below or near the limits of detection for 1260 (10 ng/m3), except for 4 of 39 samples, including two at 50
and 70 ng/m3 at the fence line (98 m from the main vent) and two (11 and 12 ng/m3) near the main vent (one
upwind).
We published this data, along with results from the Bloomington landfill in ES&T in 1985 (19, 986-991 ). In
the paper I stated that these measurements did not appear to correlate with proximity to the main vent. (Note: In
Table II of the paper, we erroneously listed one of the measurements near the vent as being at a nearby house.
Location A-7 for the main vent was adjacent to A-8 for the house on the data--sheet. In fact, nothing was ever
detected at the house, 1000 m away.) We concluded that emissions of PCB from a well-designed chemical waste
landfill were found to be negligible.
Dr. Hirshhorn has harshly criticized our conclusions , and somehow equated what we found in Warren
County to the high emissions we found in Bloomington. In my rebuttals, pointed out that since the PCB detected in
ambient air did not correlate with proximity to the main vent (and main source), i.e., concentrations were higher at 98
m than 1 m, and since the more volatile PCB mixture (1242) was the dominant component in the vent as contrasted
with 1260 being the only component detected in air, the ambient air detects may have been false positives. Given
that we were using ECO and pattern recognition, this likelihood is high.
Do you agree that (1) based on our results , PCB emissions from the landfill appear to have been negligible,
(2) little or nothing was found in ambient air, (3) the ambient air results were likely false positives, and (4) the main
vent was not likely the source of the PCB found in air even if the air measurements were real?
If you could respond to these questions in a brief letter to William Meyer, I would be grateful. You might
also add that you know of my work and have confidence in the quality of my interpretations.
Bob
. A--\t~Ghl'V\.t.nt 1 • . State of North Carolina
Department of Environment,
Health and Natural Resources
Division of Epidemiology
James B. Hunt, Jr., Governor
Jonathan B. Howes, Secretary
Michael Moser, M.D., M.P.H.
MEMORANDUM
January 6, 1996
TO: Laura S. Butler, P.E., Chief
NA
DEHNR
Air Permits Section tvO --...
William J. Pate, P.E., C.I.H., Head ~ 01 FROM:
Medical Evaluation and Risk Assessment Branch
Occupational and Environmental Epidemiology Section
SUBJECT: Warren County PCB Landfill
In your December 16, 1996 memo, you requested that the Occupational and
Environmental Epidemiology Section (OEES) provide guidance and input on concerns about
potential human health risks posed by the Warren County PCB Landfill. Luanne Williams, a
toxicologist in this Section, reviewed the documents that you sent to me and based on requests
from Bill Meyer has prepared two memos addressing review of acceptable dioxin cleanup levels
and risk assessment of PCB air sampling data collected in 1983 by Robert G. Lewis. Copies of
· these memos are enclosed. There is some uncertainty associated with the risk assessment of the
air sampling data because of the small amount of data available and because this data was
collected more than ten years ago.
The OEES would be glad to assist you in evaluating additional sampling data if it is
collected. Please call me at 715-6432 if you want more information.
WJP:lp
Enclosures
cc: Dr. Luanne Williams
Dr. Stan Music
P.O. Box 27687, Raleigh,
North Carolina 27611-7687 ,,.,.,,. An Equal Opportunity Affirmative Action Employer
50% recycled/10% post-consumer paper
01/06/97 14:40 '5'212 620 3600 U.S.DEPT . ENERGY
Department of Energy
201 Varick Street, 5th Floor
New York, NY 10014-4811
Dr. William L Meyer, Director
State of North Carolina
Department of Environment, Health
and Natural Resources
P.O. Box 27687
Raleigh, North Carolina 27611-7687
Dear Dr. Meyer:
January 6, 1997
1 received your request for reviewing a PCB issue on January 3. As yo~1 can see by this letterhead, I
have today begun a new job. As such, I will be too busy for at least a month to devote sufficient time to
properly review the materials you submitted.
If you still need my expert assistance after February 15, 1997 with this review please contact me to
discuss my fees. Tel. No.: [phone], Fax: [phone] and e-mail: [e-mail].
-lo -1(_,/4' /(-(
(}:_,-!
-/~µ-
Sincerely,
Mitchell D. Erickson, Director
Environmental Measurements Laboratory
@ P,;,,lad w!lt1 SOf ink on rocyclod paper
l4]001
-c ~ - ---Natural Resources
Michael A. Kelly, Deputy Director
Solid Waste Management Division
Please:
_ Draft a reply for my signature
DaJe: I ( 3 \57
v;;;;::information
-Take appropriate action _ See me about attached
A rove u -_ nandle and report to me
_ Note and return attached material to me
Remarks:
~ ,~J-----~ ~-\ .
~-N\~c_~'\A,~
s:: 7 \,~ ~ "-'
@_
Cc:i~v
c~--
, .Q ~ ~t\-V-t--i'<..~
:ro ~ ~~v ~ c_~ ~~---,"-1
l_~0 ~ t"'-\¼_v
(,_____.:;_~ ~ ~\.'-\
f \r 1,'v ~ --I;__ \--.__ ~"---1;2_ \/'
.L 0 '--\ r r ". ~ ----c. ....,-'< ~\ ~ ~ ~---(__~\
L_0v ~ ~ . -0~~x:.. ..-s.
NC-RESIDENT INSPECTOR Fax:[phone]
state of North Caroliha Department of Envlr6nment,
Health and Natural ~.:esources
Division of Solid Waste M:::Jnagement
Jan 3 '97 11:02 P.02106
James B. Hunt, Jr., Govei:nor
Jonathan B. Howes, Sec~etary
William L. Meyer. Directo;
SAMPLING PLAN Slf.MMARY FOR: 1
~
AVA
DEHNR
MEASUREMENT O:f FUGITIVE ATMOSPHERIC EMISSIONS OF
POLYCHLORINA1i'ED BIPHENYLS FROM THE PCB LANDFILL
WARRJ}N COUNTY, NORTH CAROLINA
Sampling Plan Date:;
Sampling Plan Preparer;
1
~~eccmbcr 18, 1996
~iette Lauffe~
Health and Safety Coordinator
I)ivision of Waste Management
F.~eigh, North Carolina 2760S I •
Project Participants: .
Pierre Lauffer, HWS-~~alth and Safety Coordinator (Project Manager and Sampler)
John Kirby, HWS-Env%:romnental Chemist (Project Chen1ist: and Sampler)
Projected Sampling Dates:. C?/06-10/97
Site History: ,
Between June, 197& and Augui1t, 1978. over 30,000 gallons of industrial waste material
identified as polychorinated bii,henyls (Arochlor 1260 ~d 1262) were discharged deliberately
along the shoulders of approximately 210 miles ofNorth'Carolina highways. In June, 1979,
EPA approved a tract ofland (J)reviously used for agriculture) in Warren County, North Carolina
as the disposal site for the PCB-contamina1ed roadside soil. The landfill (constructed in 1982-
1983 and permitted under the }.'oxic Substances Control Act (TSCA) contains about 40,000 cubic
yards of soil contaminated with PCBs. The concentration of PCBs in the landfill range from 150
to almost 900 part per million tppm), averaging about 350 ppm (1), based on 1994 subsurface
soil sampling results (these we;~e retrieved from the bottom of the venti
< I P.O. Box 27687, Roleigh, North Corollna 27611-7687 Telephone [phone] FAX 919-71~3605
An Equal Opportunity Affir~ative Action Employer 50'J. recycled/ l O'J. post-consumer poper
I !
NC~RESIDENT INSPECTOR Fax:[phone] Jan 3 '97 11:02 P.03106
' Polychlorinated Biphenyls (J.!CBs):
PCBs are a family of aiomatic compounds consisting of two benzene nuclei bonded with
two or more chlorine molecule~. The PCBs of concern in trus study (the type existing in the
I landfill) are the Aroclors. i
Aroclors are character(~ed by four digit numbers. The first two (the number 12) indicate
that the mixture consists ofbi_t;henyls. The second two digits state the percentage by weight of
chlorine in the mixture (2). ,:
They tend to be colorl~ss to light yellow oily substances with a specific gravity of 1.4-1.5
(3). Due to their stabiJity and r:10nconductive properties, PCBs had many industrial uses
including use as insulation coa}ing in electrical capacitors. PCBs possess, as a whole, high open
cup flashpoints (348-356° F.-.Aroclor 1242, none for Aroclor 1254 and 1260), but do readily
dh,1:ill at temperatures above 3~~5°F. The distillation range for Aroclor 1242 is 325-366°F; for
Aroclor 1254, 365-390°F; Aro~lor 1260, 385-420°F (4).
Toxic effects from exp{)sure to PCBs in human jnclude chlorncne, pigmentation of skin
and nails, excessive eye dischatge and swelling of eyelids, and gastrointestinal distmbanccs.
PCBs are considered carcin.ogt;nic (5).
Because of their high t~~xicity, detrimental harm to the environment and stable molecular
structure (structure remains in*ct in the environment for long pc:riods of time) PCB manufacture
was discontinued in 1976. ;
Objective:
Generally the vapor pr~ssure of PCBs under normal conditions in the PCB landfill in
Warren County is too low for jolatilization and cm.mot be emitted directly into the air
surrounding the landfill. PCB:\ emissions carried into the air through the vapomation of more
volatile substances (ic. mcth~e, CO2, H2S) from the vent at the top of the landfill is improbable.
Though improbabk:! fh'.ere remains a slim possib_ility that PCB contaminated air
particulate matter may be tram:ported through the air in the area of the landfill. Due to the nature
of the landfill, however, it is n'._~t believed that there are uncontrolled PCB emissions.
The objective of this stu.dy is to test this premise by determining if there are uncontrolled .I
PCB emissions originating from the Warren County PCB Landfill The premise will be tested by
.I conducting ambient air sampling to determine if PCB contaminated air particulate matter and
vaporized PCB (Aroclors 124~l, 1254, and 1260) emissions fil1? present. The samples will be
analyzed by the North Carolmi Public Health Laboratory. In addition. there \\ill be a risk
assessment conducted to deterjline if the PCB emissions (if found) are a risk to the surrowiding
comn11.mjty. ;
Materials and Methods: .
Air sampling will be p1:rfonned with low-volum~ (L V) and high-volume (HV) constant
air-flow sampling systems. Th;e components of the sampling systems consists of battery operated
constant .air-flow pumps (L V 9r HV) (Gilair Pump by Gillan-™, Models: Gilair3(L V) and
Gilair5(HV)). 13 mm Gehnan::Swinney filter cartridges with 13mm, lµm pore-sized glass fiber
filters manifolded to ORBOn.1;60 100150mg, 6 x 70mm florisil sampling tubes. ··
The pumps will be calibrated by the GiUan Gilibrator (digital calibrator) prior to and after
each sampling period. The gl~tss fiber filters will be precleaned prior to loading filter cartridges
by the North Carolina Public Health Laboratory and loaded at the laboratory by laboratory
\ ~
NC-RESIDENT INSPECTOR Fax:[phone] Jan 3 '97 11:03 P.04106
personnel. The sampling head ;will consist of glass fiber filter cartridge followed by the florisil
tube. The two \\iill be manifol(ed together by tubing. The purpose for this sampling system is to;
1) catch any possible PCB contaminated particulates from the air, 2) to filter PCBs which may
have vaporized. This system ,;v-ill then be manifolded to the sampling pump with plastic tubing.
There will be three Sarrlpling periods of eight hours. The sampling sessions will begin at
roughly 1 0:O0am, 6:00pm and;2:00am. Each sampling period ~ill be separated by one day.
The reason for this delay betwf:en sampling events is to provide time to recharge the air-flow
pumps (they require 16 hours ;,f recharge time). Each .sampling period will consist of seven air-·
flow pumps and sampling uni~;. One unit will be located inside the vent on top of the landfill.
Two more units will be locatec.l two meters downwind from the vent (these will be seperated
I from each other by one meter); Two other units will be located diagonally downwind at the
landfill fence line. One m1H w.~11 be located 200 meters straight downwind from the vent. The
air-flow rate ¥\ill be calibrated1to NTOSH guidelines. NIOSH Sampling Method #5503:
Sampling for Polychlorobiphe::;iyls states that air-flow rate should be 50-200cc per minute. The
flow-rate per unit will correspbnd to its distance from the vent-the further the pump is from the
vent, the greater its flow-rate. :Ancither sampling unit will be located 200 meters upwind and will
act as a background sampler. )?rior to each sampling event, the wind direction and temperature
will be indicated by a portable!weather station established on the east s:ide of the landfill.
NOTE: Please see attached d.iagram of the landfill showing sampling locations and a copy
of the NIOSH lab procedure;.
PKUH&S/HW S/Dec96.
NC-RESIDENT INSPECTOR Fax:[phone] Jan 3 '97 11:03 P.05106
----------------. ···----·-· ··---··----··----· ----. --·---· . ---····-·-------
-----. ----·· ··---
-----·--·-··-------·-···-·--------------·-·----·-•-~1-1 .
-·· ·-·--·-··-··-··· ...... ---·----......... · ···--·-·-·--·-------. -----···-·-·-··-. ...... --.. ---. --·. -·-··. . -·· .
• • i .
NC-RESIDENT INSPECTOR Fax:[phone] Jan 3 '97 11:03 P.06106
mixture: CuH1a,,C1,
[where x • 1 10 10]
MW: ca. 2!:8 (42% Cl ; C12H7Cl.z):
ca. 320 (54-:0 Cl i C11H5Cl5}
CAS: Table 1 RTECS: Table 1
EVALUATION: PARTIAL Issue 1: 15 February 1984
Revision #t; 15 lw9u:it 1957
Issue 2: 15 Auau:t t 994
OSHA:
NIOSH:
ACG!H:
1 mg/mi (42% Cl);
0-~ mg/m' (Sol% Cl)
0.001 mg/m'/10 h (carcinc'::enj
1 mg/m1 (42% Cl) (skin) '.
0.5 mg/m' (54% Cl) (Skin) 1 \
PROPERTIES: 42% 0: BP 325 to 3el:i "C; MP -19 •c:
d US g/ml@2S °C;
VF a.01 Pa (8 x 10·5 mm Hg;
1 mg/ml) @ 20 GC
ep :365 to 390 °C; MF' 10 °C;
d 1.54 g/mL@ 25 °C; VP
0.0004 Pa (3 x 10..a mm Hg:
o.o5 mg/m~ @ zo ~c
SYNONYMS: PCB; 1,1'-bii:;henyf cht:~ro: chlorod:phenyl, 42% Cl (Aroclot 1242)~nd 54% Cl (Aroclor 1254)
SAMPLING ;
SAMPLER: F:bT~~ ~--$_OU.Cl, ~O.~ENT
(7&ir\rri ·gIass 1Iber -t ·~orisil, ... ;100~1i;so me).,.· ~·· .: : ....... -~-
FLOW RATE.:,0.05 10 0.2 L/min or le~s ~ ... ··-. ,. _ .. _ .... , ......
l
SHIPMENT: transfer filters 10 glasS:'.vials after sami:lir.g
SAMPLf
STABILITY: unknown for filters: .
.2 mo1t:is for Florisil tt'.bes [·]
Bl..ANKS: 2 to 10 fjQld blanks pe·r set
RANGE STU□IED:
BIAS:
'
ACCURACY; j
j . n~tt stud.ad
OVERALL PRECISION (Srr}:
nsne ioentified
n~ evaluated ·,
ACCURACY: n~! determined
MEASUREMENT
;TECHNIQUE: GAS CHROMATOGRAPHY, E80 r iNi) : ,, . ·. .
ANAL'ITE: polychlorobipheriyls
OESORl"TIOt~: filter + front section, ::l ml h~;,;i:n~; bad~
sedori, .2 ml hexane
INJECTION
VOLUME: 4.µL with 1-µL ba:::kflus:i
iEMPERATURE-INJECTI0N:
-OETSCTOR:
-COLUMN:
CARFIIER GA$; N1, 40 ml/min
250 to 30C "C
300 to 325 °C
180 °C
COLUMN: glass, 1.S m x 2-mm ID, 1.5% OV-17/1.95%
OF-1 on 80/100 mesh Chcomosor';) WH?
CALIBRATION: standard PCS rnbc:ture In hexane
P.ANGE: 0,4 to 4 µg p11r t.arnple [2]
ESTIMATED LOO: 0.03 µg per sanple [2]
PRECISION (5,): 0.044 [~]
------------'''---------I"---------------------f
APFlLJCABIUTY: The working range'Jls 0.01 to 10 m,/m: for a 40-L air sample [1 I, With modifications, surface wipe samples
may be analyzed [3,4]. ;
j
!NiERFERENCES: Chlorinated pest'ddes, such as DDT and DDE, may interfere with quantification of PCB. Sulfur.;;:ontaining
eompounds in 1=etroleurn praauc:s a:.;so fnter1ere [:iJ.
OTI-IER METHOOS: This method re;iise.s metho:ls S120 [GJ and P&CAM 244 {1]. Methods S121 [i] and P&CAM 253 [8) for
PCB have not been ravised. ·1
NIOSH ~\nua! Of A.1alytical Methods (NMAM}. Fo1.1rth Edition, S/15/94
< ', ~
Draft for Pub!tc Comment
t
.:/:
L'J
?~·-
\',
:-~;:
-..!..--:"·
;
,, ,,.
w
~.r.;. :,;.~. r::. :r..· ..... , ~·--
5. POTEN";:AL FOR H',)~.'AN EX."OSURE
TABLE 5-2. Atmospheric Concentrations of Polychlorinated Biphenyls
Location Year
Boston, MA 1978
Columbia, SC 1978
Columbia, SC 1985 '•··
College Station, TX 1979-1980
Newport News, VA 1988
Bloomingt6'n, IN 1986-1988
Chicago, IL 1989-1990
Adirondack, NY 1985
Chesapeake Bay 1990-1991
Lake Superior 1986
Rural Ontario, Canada 1988-1989
Antarctica 1981-1982
Arctic 1986-1987
avalues are given as mean concentrations.
bValues at three different sites.
Concentrationa
(ng/m3)
7.1
4.4
2.3
0.29
0.11-0.48
0.39
(0.185-D. 794)
Summer: 1.74-3.84b
Winter: 0.31-0.62
13.5
(7 .55-20.26)
0.95
(0.339-1.359)
0.21
(0.017-0.508)
1.25
0.2
(0 .55-0.823)
{0.02-0.18)
0.02
Reference
Bidleman 1981
Bidleman 1981
Foreman and
Bidleman 1987
Atlas and Giam
1987
Knap and Binkley
1991
Hermanson and
Hites 1989
Holsen et al. 1991
Knap and Binkley
1991
Leister and Baker
1994
Baker and
Eisenreich 1990
Hott et al. 1992
Tanabe et al. 1983
Ba ke r and
Eisenreich 1990
The ranges are given in parentheses.
..
'
' f~CBs
5. POTENTIAL FOR HUMAN EXPOSURE:
degrading orpnisms or, alternatively, by adding a genetically engineered strain that combines the
activities of mixed cultures (Unterman et al. 1989). Since PCB degrad:uion is a co-metabolic
process, Lhe ~ddition of biphenyl or monochlorobiphenyls as growth substrate:c; lo supply the
nutritional requirements and to induce the catabolic pathway is required to sustai n the growth of the
degrader population for biodegradation of PCBs in soil (Guilbeault et al. 1994; Hickey et al. 1993).
In addition, the presence of surface active agents has been shown to increase the bi oavai lability of
PCBs to the microorganisms. However~,. enriched cultures were unable to bi ode grade PCB
congeners containing five 'or higher chlo'rine substitution (Guilbeault et al. 1994). It has been
' '
reported that the mono-, and di-chlorobenzoate, and possibly other higher chlorobenzoates formed
from aerobic degradation of PCBs act as inhibitors towards further degradation of higher chlorinated
PCBs (Guilbeault et al. 1994). Therefore, the efficiency of PCB degradation is not only controlled
by the enzyme substrate selectivity pattern, but also by the metabolite production pattern.
5.4 LEVELS MONITORED OR ESTIMATED IN THE ENVIRONM~NT
5.4.1 Air
The atmospheric concentrations of PCBs in various locations are given m Table 5-2. The range of
atmospheric concentrations of PCBs in urban areas is 1-10 ng/m3 with a mean of 5 ng/m3
(Eisenreich et al. 1992). The atmospheric concentrations of PCBs in two rural areas are in the range
0.2-0.95 ng/m3 with a mean of 0.6 ng/3 and in two remote areas are in the range of 0.02-0~ 8
ng/m3 "with a mean of <0.1 ng/m3 (see Table 5-2). The range and mean atmospheric PCB
concentrations in other locations are as follows: 0.01-0.7 ng/m3 _ and 0. I ng/m3, respectively, m
marine/coastal areas; and 0.2--4.0 ~g/m3 and 1.0 ng/m3, respectively, over the Gre~kes
(Eisenreich et al. 1992). With the available data, it i§ difficult to establi sh the trend in atmospheric
--~---;--;------;-----;------;;-;-;---:---;-----:------;:-:::·-------........ .
PeB-'cmfcentrations over the last two decades following the cessation of PCB production. This is
because monitoring data indicating the levels of PCBs in air at the same locati on over this time
period are still lacking (levels from one location cannot be compared wi th levels from another
because of differing emission sources), and the recent studies (Schreitmueller and Ballschmiter
1994) generally report the atmospheric concentrations of the congeners and not the total PCBs or
Aroclors. On the basis of typical atmospheric concentrations of PCBs in pre-1980 samples
(Eisenreich et al. 1981) and the levels in more recent years (see Table 5-2), it can be concluded the
PCB concentrations in air may have shown a slightly decreasing trend from the pre-1980 lo post-
State· of North Carolina Ati-eL-h~r\t ~
D\?partment of Environment,
Health and Natural Resources
Division of Epidemiology
James B. Hunt, Jr., Governor
NA
Jonathan B. Howes, Secretary
Michael Moser, M.D., M.P.H. DEHNR
December 10, 1996
MEMORANDUM : i . i: ...
i· l JAN 81997
TO: Bill Meyer, Director
Division of'Waste Management : , : ... .
THROUGH: Stanl~-y Music, M.D., DTPH (LOND.), Chief ~-.:-__:__::___:_~---··
· Occupational and Environmental Epidemiology Sectio~ .
William J. Pate, Head ~r l3f
Medical Evaluation and Risk Assessment Branch
·, · Occupational and Environmental Epidemiology Section
FROM: Luanne K. Williams, Pharm.D., Toxicologistc;{;{ (,Ar"
Medical Evaluation and Risk Assessment Branch
Occupational and Environmental Epidemiology Section
SUBJECT: Review of Dioxin Cleanup Levels for the Warren County PCB Lan~ffill
Proposed by Hirschhorn & Associates
I have reviewed the document prepared by Dr. Joel Hirschhorn titled "Cleanup Levels
for Dioxin Contaminated Soils." My recommendations with regard to the derivation of
cleanup levels for dioxins and furans and sampling are as follows:
1. Dr. Hirschhorn has proposed 2 to 4 parts per trillion (ppt) as the residential cleanup level
based upon a target excess cancer risk of 1 x 10-s (one in a million). The recommended
target cleanup level for dioxin will be dependent upon the current or future use of the
site, use of the groundwater, and background concentrations. Guidance for determining
the target cleanup level for 2,3,7,8-TCDD (dioxin) is provided as follows:
RESIDENTIAL SOIL CLEANUP LEVEL
If the site is, or may be in the future, a residential area or in an area where activities of
sensitive human receptor populations occur (e .g., schools, day-care facilities, and
retirement centers), then the soil target concentration should be based on residential
exposure. The recommended soil cleanup level for 2,3,7,8-TCDD in a residential area is
4 ppt (USEPA Region Ill Risk-Based Concentration Table, April 1996). The oral slope
factor used to calculate 4 ppt was obtained from the USEPA 1995 Health Effects
Assessment Summary Tables. The oral slope factor was determined by US EPA based
P.O. Box 27687, Raleigh,
North Carolina 27611-7687 N!iC
11Tf t-'@tffl
An Equal Opportunity Affirmative Action Employer
50% recycled/10% post-consumer paper
...
_Bill Meyer
December 10, 1996
Page Two
upon the 1984, 1985, and 1989 review of the study Kociba RJ , Keyes DG, Bower JW et
al., 1978. "Results of a Two-year Chronic Toxicity and Oncogenicity Study of 2,3,7,8-
Tetrachlorodibenzo-p-dioxin in Rats." Toxicol Appl Pharmacol. 46(2):279-303.
In 1984, the Centers for Disease Control (CDC) staff released a paper that contained
recommendations of 1 ppb or 1000 ppt as a level that would not likely result in adverse
effects in a residential area and according to CDC would correspond to an excess
I ... cancer risk of 1x10·6. Using the oral slope factor in the USEPA 1995 HEAST, a
residential soil dioxin concentration of 1000 ppt would correspond to a 2.5 x 10-4 excess
cancer risk. It is recommended to use 4 ppt instead of 1000 ppt as a cleanup level in a
residential area.
INDUSTRIAL/COMMERCIAL SOIL CLEANUP LEVEL
If the site is, or may be in the future, an area where adult worker exposure occurs, then
the soil cleanup level should be based on industrial/commercial exposure. The
recommended soil cleanup level for 2,3,7,8-TCDD in an industrial/commercial area is 40
ppt (USEPA Region Ill Risk-Based Concentration Table, April 1996).
SOIL-TO-GROUNDWATER CLEANUP LEVEL
If the groundwater in this area is being used for drinking, then transport modeling may
be necessary to determine the maximum allowable dioxin soil concentration that would
not result in exceedance of the dioxin groundwater quality standard. If groundwater is
being used for drinking, then the soil cleanup level would be the lowest of the soil-to-
groundwater cleanup level; or the residential or industrial/commercial cleanup level
(whichever is applicable).
SOIL BACKGROUND CONCENTRATIONS
If the TCDD soil cleanup level is determined to be less than the soil background
concentrations, then it is recommended to use the soil background concentration as the
TCDD soil cleanup level. I have enclosed guidance for collecting background samples
which was obtained from USEPA 1989 Risk Assessment Guidance for Superfund
Volume I Human Health Evaluation Manual (Part A) (EPA/540/1-89/002).
2. Dr. Hirschhorn & Associates are proposing higher toxicity equivalents factors (TEFs)
than those recommended by EPA for chlorinated dioxin and furan congeners. This
would result in lower cleanup levels. Higher TEFs are proposed to be used because of
the likelihood of synergistic effects from exposure to dioxins and polychlorinated
biphenyls (PCBs). I caution the use of higher TEFs because of the uncertainty in
synergism between dioxins/furans and PCBs. Also , I caution the use of higher TEFs
because conservative assumptions have already been considered in deriving the TCDD
carcinogenic slope factor and in deriving the exposure parameter values used to
f
Bill Meyer
December 10, 1996
Page Three
generate the TCDD cleanup level. The TEFs recommended by EPA and other state
should be used to determine cleanup levels for dioxin and furan congeners. A list of the
TEFs recommended by EPA is provided in Table 1 (USEPA 1995 Supplemental
Guidance to RAGS: Region 4 Bulletins Human Health Risk Assessment).
The soil cleanup level for ead-i'dioxin and furan congener found at the site can be
calculated by div1'd'ing the cleanup level for 2,3,7,8-TCDD by the appropriate TEF. If Dr.
'Hirschhorn has scientific evidence to support different TEFs, then I would like the
opportunity to review it.
Table 1. Toxicity Equivalents Factors (TEF) for COOs and COFs*
,.
Dioxin Compound TEF Furan Compound TEF
2,3,7,8-TCDD 1 2,3,7,8-TCDF 0.1
2,3,7,8-PeCDD 0.5 1,2,3,7,8-PeCDF 0.05**
2,3,7 ,8-HxCDD 0.1 2,3,4,7,8-PeCDF 0.5**
2,3,7,8-HpCDD 0.01 2,3,7,8-HxCDF 0.1
OCDD 0.001 2,3,7,8-HpCDF 0.01
Other CDDs 0 OCDF 0.001
Other CDFs 0
* Source: EPA, 1995. Supplemental Guidance to RAGS: Region 4 Bulletins Human Health Risk Assessment.
** Correction noted per telephone conversation with EPA Region 4 on November 27, 1996.
3. ,• I have discussed the issues pertaining to this site with Dr. Renate Kimbrough with the
Institute for Evaluating Health Risks in Washington, D.C. She has expressed an interest
in reviewing the sampling protocol. She can be reached by phone at 202-289-8721 or
fax [phone]. Her address is as follows: Institute for Evaluating Health Ris ks, Suite
402, 1629 K Street N.W., Washington, D.C. 20006.
Please feel free to call me at any time . I can be reached at 715-6429. Thank you for
~he opportunity to review the report.
LKW:lp
Enclosures
:· J ~'?.l:::ite of No rth Coro lino l\ttu.h.,,,,._"' t 3
.. Depa·rtment of Environment,
Health and Natural Resources
Di vision of Epidemiolo gy NA
James 8. Hunt, Jr., Govern or
Jonathan 8. Howes, Secretary
M ichael Moser, M.D., M.P .H. DEHNR
December 9, 1996
MEMORANDUM
TO: i ~---·· ..-------<-?-.,~,,,/
Bill Meyer, Director
Division of,Waste Management
,f
THROUGH: Stanley Music, M.D., DTPH (Land.), Chief ----z__,;)
FROM:
Occupational and Environmental Epidemiology Section
Luanne K. Williams, Pharm.D., Toxicologist ;;(J:'w-
Medical Evaluation and Risk Assessment Branch
.,. Occupational and Environmental Epidemiology Section
At your request, I have provided a risk assessment following my review of the air
sampling results of the Warren County PCB Landfill reported in the 1983 USEPA study
"Measurement of Fugitive Atmospheric Emissions of Polychlorinated Biphenyls from Hazardous
Waste Landfills" and the December 3 letter from Dr. Robert G. Lewis, a co-author of the study.
RISK ASSESSMENT
1. The ambient air concentrations reported of 11 , 12, 50 , and 71 ng/m3 most likely exceed
the actual PCB concentrations present at the site for the following reasons:
'(a) the method used is not specific for PCBs but detects all chlorinated compounds
(b) ambient air concentrations were reported to be higher at 98 meters downwind
(50 and 71 ng/m3) than beside the main vent (11 ng/m3) and
(c) aroclor 1260 was the only analyte identified in ambient air even though aroclor
1242 was reported at much higher concentrations in the main vent.
2. It is my opinion that the ambient air concentrations reported are w orst-case estimate of
the concentrations that may be present at the site. Therefore, a worst-case risk
estimate is provided based upon the concentrations reported at the following locations:
Locations
beside main vent
nearby house
fence line, downwind
P.O. Box 27687, Raleigh,
North Carolina 27611-7687
(cone. detected)
11 ng/m3
12 ng/m3
50 and 71 ng/m3
'N!iC
11Mtttffl
Excess Cancer Risk
1x10-5
1 x10-5
5 to 7x10·5
An Equal Opportunity Affirmative Action Employer
50% recycled/ 10% post-consumer paper
. Bill Meyer
December 9, 1996
Page Two
If a million people were exposed to the concentrations reported at these three locations
over a lifetime, then one to seven of those individuals could possibly develop cancer as
a result of their exposure to the PCB concentrations reported. This cancer risk is a
worst-case estimate and is in addition to the existing cancer risk rate of 333 ,333 out of a
million expected cancer cases in a lifetime. ,.
The risk associated with exposu-re to the concentrations reported at the fenceline does
exceed the acceptable target excess cancer risk of 1 x1 o-6• However, the actual risk at
the site is most likely lower than 7x10-5 because the PCB concentrations present at the
site are most likely lower than reported. The PCB air concentration associated with a
1x10-5 excess cancer risk (target risk) is 10 ng/m3• The fenceline exceeds this level by
7.
3. The concentrations reported near the PCB landfill exceed typical background
concentrations in rural and urban areas. I have attached background concentrations
reported in the February 20, 1996 ATSDR Tox profile for PCBs. The highest reported
background concentration was 20 ng/m3 in Chicago in 1989-1990.
RECOMMENDATIONS
Contrary to Dr. Lewis' statement in his letter, it is my opinion that it is uncertain as to
whether or not PCBs are present at the site. Because PCBs were detected using a method
that is not specific for PCBs and since the excess cancer risk estimated at the fenceline
exceeds the acceptable target excess cancer risk of 1 x 10-5 , it is recommended to collect
additional vent and ambient air samples and analyze for aroclor 1242 and aroclor 1260 using a
more specific method. A carbon filter may also be used as suggested by BFA Environmental
Consultants to minimize PCB emissions from the landfill.
Please call me if you have any questions at 715-6429.
LKW:lp
Attachments
cc: Dr. Stan Music
Mr. Bill Pate
. ' State of North Carolina
Department of Environment,
Health and Natural Resources
Division of Air Quality
James B. Hunt, Jr., Governor
Jonathan B. Howes, Secretary
Alan W. Klimek, P.E., Director
JvffiMORANDUM
Air Permits Section
December 23, 1996
To: Lee A. Daniel, Chief
Technical Services Section
Through:
From:
Subject:
Alan Klimek ,)}/'··
.'/t1/ Laura S. Butler, P.E., Chief/
Air Permits Section
PCB Air Emissions
Warren County PCB Landfill
Warren County
AVA
DEHNR
Ernie Fuller, Regional Air Quality Supervisor, Raleigh Regional Office, has agreed to
conduct air monitoring at the PCB Landfill in Warren County. Ernie will be available to conduct
the monitoring after the first week in January 1997. I have suggested to Ernie, that the
monitoring be conducted with the cooperation and over-sight of the Joint Warren County/State
PCB Landfill Working Group, Science Advisors, Patrick Barnes and Joel Hirschhorn, and
representatives of the Division of Waste Management.
I am requesting that the Technical Services Section provide technical assistance to the
Raleigh Regional Office for this monitoring effort. You have stated that the Division of Waste
Management has the necessary monitoring equipment. As indicated to you earlier in my
memorandum of December 16, 1996, the Warren County PCB Landfill Working Group is
concerned about potential human health effects and environmental impacts from PCB emissions.
Your earliest response is requested.
cc: Ernie Fuller
Bill Meyer
Bill Pate
George Murray
Lori Cherry
P.O. Box 29580. Raleigh, North Carolina 27626-0580
Voice [phone] FAX [phone]
An Equal Opportunity/Affirmative Action Employer
50% recycles/10% post-consumer paper
January 3, 1997
MEMORANDUM:
TO: PCB WORKING GROUP
FROM: MIKE KELLY
SUBJECT: AIR SAMPLING AT THE PCB LANDFILL
Today I faxed out copies of a sampling plan to be used next week at the landfill to pull a
series of air samples from inside the landfill and around it as well. If you recall, I talked about
this sampling event at the meeting on December 17th and said that the projected start date was
January 6.
I have had members of my staff working on this event for several.weeks, ordering
equipment and preparing the necessary items, as well as coordinating with the laboratory. At the
meeting in December, I was asked to do two things: 1) not have this event or similar things done
by the Division detract from the main project of detoxification, and 2) to send the science
advisors a copy of the plan. I clearly stated that we did not plan to do an elaborate plan because
this was an initial one time event, but I would send what I could to the advisor's as soon as
possible, but I did not plan to hold up the event for any "approvals" because it was (is) not
designed to be anything more than a "look" at the air coming from the landfill. There are many,
many things to consider for the future in long term monitoring, this is only a start, a snap-shot, if
you will.
The plan was completed at approximately 11 am this morning, and was faxed via a
broadcast fax beginning around 11 :30. For some reason the fax did not go through to Joel
Hirschhron on the first try and I re-faxed at approximately 12:30.
I have discussed this plan with Joel following receipt of his memo this afternoon. This
sampling event is not designed to be a one time deciding event or "catch all", nor is it supposed
to be in response to the "recommendations of the EPA study". Rather, it is designed to pull air
from the landfill and surrounding area and measure for PCB's and to determine if there are
airborne PCB's emitting from the landfill now. Joel has opposed it. I agree with him that we are
not using the most sensitive sampling methods available, however, we will have detection limits
which will coincide with published air standards to be used as a point of reference. I will
instruct the lab to achieve the lowest possible level of detection (LOD) for this method. Any
future, long term, air sampling and monitoring will follow a detailed implementation plan.
As implied in the plan, we have not had any reason to suspect uncontrolled PCB
emissions from the landfill, however, the science advisors have cast doubt on this reasoning, and
we wanted to pull some samples as soon as possible in response to the raised questions, just as
we responded by putting a carbon filter on the vent pipe. Unlike some comments, we are not
trying to cast doubt on everything, but continue ahead with testing and actions deemed necessary
as time goes by.
We will continue to deal with facts, not conjectures and miscellaneous comments.
Technically, I have no reason not to continue with the planned sampling event. Several scientific
people have been working on this for weeks, and it is certainly adequate for what we want to do
at this point in time. It is disappointing that comments like "absolutely no reason to trust the
study" and "biased state study" are being made. No one from the state is trying to be
"discourteous and disrespectful". I would appreciate the same type of professional behavior from
others as I think we could all get a lot more productive work done.
Copy: J. Hirschhorn
P. Barnes
B. Meyer
6-[14-1 S:➔96 1 : CISAM
January 3, l 997 .................................................................................. by FAX
To: Technical Committee aud full Working Group
From: Joel Hirschhorn
Through: Doris --for immediate distribution
Subject: Response to state's Afr Sampling Plan
Today, at 12:30 PM I received that attached cover letter and Plan. Note that the Plan is dated
December 18 and that testing is scheduled to begin this Monday, Jan.nary 6. I find the state's
behavior io. providing the Plan for review at this late tin1e both discourteous and disrespectful.
I r~ject the entire Plan and advise the Working Group to r~ject the study for the following
reasons:
l1Je key stated premise of the Plan is "it is not believed that there are uncontrolled PCB
emissions." And the stated objective of the study "is to test this premise by determining if there
are uncontrolled PCB emissions." In other words, the "scientific" position of the state is that
there are no emissions and their study will prove this. In fact, the Plan states that "GeneraJly the
vapor pressure of PCBs under no1n1al conditions in the PCB landfill in Wauen County is too low
for volatilization and cannot be emitted directly into the air surrounding the landfill." If this
position was scientifically correct, then why did EPA conduct its study to measure PCB emissions
and why did EPA actually measure PCB emissions? The Working Group and concerned citizens
should reject this biased state study from the outset. There is absolutely no reason to tmst the
study.
With this kind of bias the public should not accept a state agency ao.d state personnel conducting
the sampling and tb.e laboratory testing. An independent contractor and an. iudependeut
laboratory should be used, and the Science Advisors should be given adequate time to fully
examine and review the details of the sampling and testing methods.
Technically, I do not believe that the planned sampling methods are acceptable, nor do they fully
co1Jform with th.e recommendations of the EPA study.
It seems as if the state is committed to use any tactic to attack the previous work on PCB air
emissions in the past. Even if sound sampling and testing found no PCB emissions today, it
would not in any logical or scientific sense prove that there were no significant PCB air emissions
in the past.
TIJe planned study is to include a risk assessment, but this too should not be considered objective
or reliab.1e by the public. . ..._
( cc: Mike Kelly ·----..__.
---------.. __ -·-•-•. -··-·'. ·-··. ' -· .. --·-·----_,,,,)
State of North Carolina
Department of Environment,
Health and Natural Resources
Division of Air Quality
James B. Hunt, Jr., Governor
Jonathan B. Howes, Secretary
Alan W. Klimek, P.E., Director
MEMORANDUM
January 2, 1997
TO: Doris Fleetwood, Secretary
Warren County PCB Landfill Working Group
FROM: Laura S. Butler, P.E., Chief J:;J0->
Air Permit Section
SUBJECT: PCB Air Emissions
Additional documents produced
by Joel Hirschhorn
Warren County
On December 17, 1996, I received a fax from Joel Hirschhorn in which he
indicated that he had produced other documents apparently concerning PCB air
emissions from the Warren County PCB Landfill (copy attached). On December 18 and
19, 1996, I called Mr. Hirschhorn in order to obtain copies of the documents. On both
days I left messages on Mr. Hirschhorn's voice mail, but did not receive a return call.
On December 23, 1996, I again called Mr. Hirschhorn to request the documents.
Mr. Hirschhorn told me to obtain the documents from the Secretary of the Working Group.
I called the Working Group office several times on December 23, 1996, but there was no
answer.
By way of this memorandum , I am requesting copies of the additional documents
produced by Mr. Hirschhorn. Please call me at 919-715-6236 if you have any questions
or comments concerning my request.
LSB :bw
c: A an Klimek
ill Meyer
Bill Pate
Ernie Fuller
George Murray
Lee Daniel
Attachment (1)
P.O. Box 29580, Raleigh, North Carolina 27626-0580
Voice [phone] FAX [phone]
An Equal Opportunity/ Affirmative Action Employer
50% recycles/10% post-consumer paper
' ' '. . . 5'-1 7-1 996 1 ld: 02PM FROM
December 17, 1996 ................................................................... byFAX
To: Laura S. Butler, Chief Air Permits Section, DEHNR
From: Joel Hirschhorn
Subject: Request for document
In my capacity as a Science Advisor for the Warren County PCB Landfill Working Group, I
request a copy of the document "Dispersion Modeling for Warren County PCB Landfill," by Tom
Anderson, Dec. 10, 1996.
Apparently you have distn"bnted this document for review by certain parties.
If at all possible, would you please fax me the above document as soon as possible.
I also want to let you Irnow that in addition to the two documents l have provided the Working
Group on the PCB air emissions issue I have produced some others that you should also provide
your reviewers.
;
! . ... --' ......... . ! tit',~ i--. . ·: 1· \. '.,c :_, . '.. --',. I I l .; ,:,U ,.:) Ui\(;.
. '
P. 1
State of North Carolina
Departl'(lent of Environment,
Health and Natural Resources
Division of Air Quality
James B. Hunt, Jr., Governor
Jonathan B. Howes, Secretary
Alan W. Klimek, P.E., Director
:MEMORANDUM
Air Permits Section
December 23, 1996
To: Lee A. Daniel, Chief
Technical Services Section
Through:
From:
Subject:
Alan Klimek ~'
l@ ·'8'7/ Laura S. Butler, P.E., Chief/
Air Permits Section
PCB Air Emissions
Warren County PCB Landfill
Warren County
.AVA
DEHNR
Ernie Fuller, Regional Air Quality Supervisor, Raleigh Regional Office, has agreed to
conduct air monitoring at the PCB Landfill in Warren County. Ernie will be available to conduct
the monitoring after the first week in January 1997. I have suggested to Ernie, that the
monitoring be conducted with the cooperation and over-sight of the Joint Warren County/State
PCB Landfill Working Group, Science Advisors, Patrick Barnes and Joel Hirschhorn, and
representatives of the Division of Waste Management.
I am requesting that the Technical Services Section provide technical assistance to the
Raleigh Regional Office for this monitoring effort. You have stated that the Division of Waste
Management has the necessary monitoring equipment. As indicated to you earlier in my
memorandum of December 16, 1996, the Warren County PCB Landfill Working Group is
concerned about potential human health effects and environmental impacts from PCB emissions.
Your earliest response is requested.
cc: Ernie Fuller
Bill Meyer /
Bill Pate
George Murray
Lori Cherry
P.O. Box 29580. Raleigh, North Carolina 27626-0580
Voice [phone] FAX [phone]
An Equal Opportunity/Affirmative Action Employer
50% recycles/ l 0% post-consumer paper
12/19/96 10:30 '0'919 541 3527 NERL/AMRD/MB
/),1:1,1 ~ Cl, , o J.,.
U.S. ENVIRONMENTAL PROTECTION AGENCY
NATIONAL EXPOSURE RESEARCH LABO RA TORY
DATE:
TO:
AIR MEASUREMENTS RESEARCH DIVISION (MATI., DROP 44)
RESEARCH TRIANGLE PARK, NORTH CAROLINA 27711
TELEPHONE [phone] FACSlMILE 919-541-3527
19 December 1996
Bill Meyer/Mike Kelly
E-MAIL lewis. [e-mail]
FAX MESSAGE
NUMBER OF PAGES:
(including cover)
Fax No.: 715-3605
Telephone:
FROM: Dr. Robert G. Lewis
USEPA (MD-44)
Research Triangle Park, NC 27711
SUBJECT: Response to Hirschhorn Report
MESSAGE:
Since Hirschhorn has made an issue of the quality of our analytical measurements of
PCBs, I'd like to suggest an additional reviewer:
Dr. Mitchell D. Erickson
Analytical Chemistry Laboratory
Argonne National Laboratory
9700 South Cass Avenue, CMT/205
Argonne, IL 60439-4837
Tel. [phone]
Mitch wrote the "bible" on PCB analysis (M. D . Erickson, Analytical Chemistry of PCBs,
Lewis Publishers, Buca Raton, FL, 1st. ed ., 1986, 2nd. ed., 1997). He knows me and my work
well. I'd send him the same package that you sent to Bidleman and Hites, except the last item
(my fax of December 11).
[41001
State of North Carolina
Department of Environment,
Health and Natural Resources
Division of Solid Waste Management
James B. Hunt, Jr., Governor
Jonathan B. Howes, Secretary
William L. Meyer, Director
Dr. Mitchell D. Erickson
Analytical Chemistry Laboratory
Argonne National Laboratory
9700 South Cass Avenue, CMT/205
Argonne, IL 60439-4837
Dear Dr. Erickson:
NA
DEHNR
December 19, 1996
It has been brought to my attention that you are an expert on analytical measurements of
PCBs and familiar with the work of Dr. Robert G. Lewis. Dr. Lewis authored an article published
in Environmental Science and Technology, October, 1985, pp. 986-991, titled: Measurement of
Fugitive Atmospheric Emissions of Polychlorinated Biphenyls from Hazardous Waste Landfills.
The quality and scientific interpretations of the monitoring study have recently been questioned. I
would like to respectfully request your review of both the article and recent responses to
questions concerning the quality and conclusions by Dr. Lewis and Dr. Joel Hirschhorn. Your
response to this request is appreciated and will be presented to a Joint Warren County/State PCB
Landfill Working Group established for detoxification of the landfill.
Thank you for considering this request on our behalf and assisting our efforts to address
complex scientific and technical issues as we make progress toward ultimate detoxification of the
PCB Landfill.
If you need clarifications, information or assistance please contact me at 919-733-4996,
ext. 202.
Cc -~ ~wv.; -11/-w/'1& (1 h'/
P.O. Box 27687,
Raleigh, North Carolina 27611-7687
Voice 919-733-4996
Sincerely,
~
' I '
i .. __/ ~ ::;1-JJLur~J
William L. Meyer, Director
■e•a•~¾fflf:M
FAX [phone]
An Equal Opportunity Affirmative Action Employer
50% recycled/10"/o post-consumer paper
.J.. .;.. I .1. •".I / V U L V • U V ... V.J..U U"t.1. ,Jo,:;..1 ,11....J.,1...,1 .:-1.1111\.UI i!J.U
DATE:
TO:
U.S. ENVIRONMENTAL PROTECTION AGENCY .
NATIONAL EXPOSURE RESEARCH LABORATORY
AIR MEASUREMENTS RESEARCH DIVISIOK (MAIL DROP 44)
RESEARCH TRIANGLE PARK, NORTH CAROUNA 277 11
TELEPHONE [phone] F ACS[MlLE 91 9-541-3527
19 December 1996
Bill Meyer/Mike Kelly
E-MAIL [e-mail]
FAX iv1ESSAGE
NUl\IBER OF PAGES:
(including cover)
Fax No.: 715-3605
Telephone:
FROM: Dr. Robert G. Lewis
USEP A (l\lID-44)
Research Triangle Park, NC 27711
SL:BJECT: Response to Hirschhorn Report
MESSAGE:
Since Hirschhorn has made an issue of the quality of our analytical measurements of
PCBs, I'd like to suggest an additional reviewer:
Dr. lvlitchell D. Erickson
Analytical Chemistry Laboratory
Argonne National Laboratory
9700 South Cass Avenue, CMT/205
Argonne, IL 60439-4837
Tel. [phone]
I'v1itcb wrote the "bible" on PCB analysis (M. D. Erickson, Analytical Chemistry of PCBs,
Lewis Publishers, Boca Raton, FL, 1st. ed., 1986, 2nd. ed., 1997). He knows me and my work
well. I'd send him the same package that you sent to Bidleman and Hites, except the last item
(my fax of December 11).
~\J\11
State of North Carolina
Department of Environment,
Health and Natural Resources
Division of Solid Waste Management
James B. Hunt, Jr., Governor
Jonathan B. Howes, Secretary
William L. Meyer, Director
AWA
DEHNR
December 19, 1996
TO: Dr. Terry Biddleman
Prof Ronald A. Hites
FROM: Bill Meyer, Director . .-,1" ). ; I) , .J \
Division of Waste Managemeit../~ 61, ~J,V
SUBJECT: Additional Material for review
Please add this latest report from Joel Hirschhorn to the enclosures mailed with the letter of
December 12, 1996.
sh
P.O. Box 27687,
Raleigh. North Carolina 27611-7687
Voice 919-733-4996 f@M2Ji/§N5
FAX [phone]
An Equal Opportunity Affirmative Action Employer
50% recycled/ l 0% post-consumer paper
Enclosures:
Measurement of Fugitive Atmospheric Emissions of Polychlorinated Biphenyls from Hazardous
Waste Landfills, Robert G. Lewis and Barry E. Martin, Environmental Science and Technology,
October 1985, pp. 986-991.
Draft Final Report on Ambient Monitoring for PCB's at the Warren County (North Carolina)
Landfill, D. L. Sgontz, et.al., Battelle Columbus Laboratories, April 8, 1983.
PCB Air Emissions and Health Risks From the Warren County PCB Landfill, Dr. Joel Hirshhorn,
November 27, 1996.
Air Emissions of PCB and Associated Health Risk, Patrick Barnes, December 2, 1996.
Hirschhorn Report on Warren County PCB Landfill, Dr. Robert G. Lewis, December 3, 1996.
Response to EPA letter of3 December, previously provided; Dr. Joel Hirschhorn, December 10,
1996.
Hirschhorn Report of 12/16/96 -Review of original EPA contractor report on PCB air emissions
c:wpfiles/pcbWerickson.ltr
State of North Carolina
Department of Environment,
Health and Natural Resources
Division of Solid Waste Management AVA
James B. Hunt, Jr., Governor
Jonathan B. Howes, Secretary
William L. Meyer, Director DEHNR
MEMORANDUM
TO: Luanne Williams
FROM: Bill Meyer~
DATE: December 18, 1996
SUBJECT: PCB -Air Sampling Results
Thanks for your 12/9/96 response.
Section 3 -1st sentence -"typical background"
I would delete this sentence since Section 5.4-lofthe attachment states, "levels
from one location cannot be compared with levels from another because of
differing emission sources".
With regard to your voice mail of 12/16 with the new/latest EPA report on
tox. equivalent of PCB's. There are two issues here:
(1) source of PCB's (fingerprint) unless the separate AROCLOR(s) is identified,
or ratios determined it will not allow identification of source( s) ( or elimination of
sources) and we could end up making the wrong decision. This issue is not a
health issue but one of common sense and we will continue to identify PCB' s by
specific AROCLORs and use totals for exposure determinations, if that is the
standard;
(2) if totals (PCB's) only are used for health determinations, I assume
that all the AROCLORs would still be determined (since they occur as different
peaks) and then totaled? It is also interesting that toxicity of all AROCLORs
would be considered equivalent since 1260 can be biologically degraded
anaerobically but not aerobically and the reverse is true for the lower chlorine
saturated PCBs. Even bug physiology tells us they are different. Then we might
not be as smart as the bugs.
C:wpfiles/pcblf71-willia.mern
P.O. Box 27687,
Raleigh, North Carolina 27611-7687
Voice 919-733-4996 1¢MA■&A5
FAX [phone]
An Equal Opportunity Affirmative Action Employer
50% recycled/10% post-consumer paper
State of North Carolina
Department of Environment,
Health and Natural Resources
Division of Solid Waste Management
James B. Hunt, Jr., Governor
Jonathan B. Howes, Secretary
William L. Meyer, Director
SAMPLING PLAN SUMMARY FOR:
MEASUREMENT OF FUGITIVE ATMOSPHERIC EMISSIONS OF
POLYCHLORINATED BIPHENYLS FROM THE PCB LANDFILL
WARREN COUNTY, NORTH CAROLINA
Sampling Plan Date:
Sampling Plan Preparer:
Project Participants:
December 18, 1996
Pierre Lauffe~
Health and Safety Coordinator
Division of Waste Management
Raleigh, North Carolina 27605
Pierre Lauffer, HWS-Health and Safety Coordinator (Project Manager and Sampler)
John Kirby, HWS-Environmental Chemist (Project Chemist and Sampler)
Projected Sampling Dates: 01/06-10/97
Site History:
Between June, 1978 and August, 1978, over 30,000 gallons of industrial waste material
identified as polychorinated biphenyls (Arochlor 1260 and 1262) were discharged deliberately
along the shoulders of approximately 210 miles of North Carolina highways. In June, 1979,
EPA approved a tract of land (previously used for agriculture) in Warren County, North Carolina
as the disposal site for the PCB-contaminated roadside soil. The landfill ( constructed in 1982-
1983 and permitted under the Toxic Substances Control Act (TSCA) contains about 40,000 cubic
yards of soil contaminated with PCBs. The concentration of PCBs in the landfill range from 150
to almost 900 part per million (ppm), averaging about 350 ppm (1), based on 1994 subsurface
soil sampling results (these were retrieved from the bottom of the vent).
P.O. Box 27687, Raleigh, North Carolina 27611-7687 Telephone [phone] FAX [phone]
An Equal Opportunity Affirmative Action Employer 50% recycled/ l 0% post-consumer paper
\,
Polychlorinated Biphenyls (PCBs):
PCBs are a family of aromatic compounds consisting of two benzene nuclei bonded with
two or more chlorine molecules. The PCBs of concern in this study (the type existing in the
landfill) are the Aroclors.
Aroclors are characterized by four digit numbers. The first two (the number 12) indicate
that the mixture consists of biphenyls. The second two digits state the percentage by weight of
chlorine in the mixture (2).
They tend to be colorless to light yellow oily substances with a specific gravity of 1 .4-1.5
(3). Due to their stability and nonconductive properties, PCBs had many industrial uses
including use as insulation coating in electrical capacitors. PCBs possess, as a whole, high open
cup flashpoints (348-356° F.-Aroclor 1242, none for Aroclor 1254 and 1260), but do readily
distill at temperatures above 325°F. The distillation range for Aroclor 1242 is 325-366°F; for
Aroclor 1254, 365-390°F; Aroclor 1260, 385-420°F (4).
Toxic effects from exposure to PCBs in human include ehloracne, pigmentation of skin
and nails, excessive eye discharge and swelling of eyelids, and gastrointestinal disturbances.
PCBs are considered carcinogenic (5).
Because of their high toxicity, detrimental harm to the environment and stable molecular
structure (structure remains intact in the environment for long periods of time) PCB manufacture
was discontinued in 1976.
Objective:
Generally the vapor pressure of PCBs under normal conditions in the PCB landfill in
Warren County is too low for volatilization and cannot be emitted directly into the air
surrounding the landfill. PCBs emissions carried into the air through the vaporization of more
volatile substances (ie. methane, CO2, H2S) from the vent at the top of the landfill is improbable.
Though improbable, there remains a slim possibility that PCB contaminated air
particulate matter may be transported through the air in the area of the landfill. Due to the nature
of the landfill, however, it is not believed that there are uncontrolled PCB emissions.
The objective of this study is to test this premise by determining if there are uncontrolled
PCB emissions originating from the Warren County PCB Landfill. The premise will be tested by
conducting ambient air sampling to determine if PCB contaminated air particulate matter and
vaporized PCB (Aroclors 1242, 1254, and 1260) emissions are present. The samples will be
analyzed by the North Carolina Public Health Laboratory. In addition, there will be a risk
assessment conducted to determine if the PCB emissions (if found) are a risk to the surrounding
community.
Materials and Methods:
Air sampling will be performed with low-volume (L V) and high-volume (HV) constant
air-flow sampling systems. The components of the sampling systems consists of battery operated
constant air-flow pumps (LV or HV) (Gilair Pump by Gilian™, Models: Gilair3(LV) and
Gilair5(HV)), 13 mm Gelman Swinney filter cartridges with 13mm, lµm pore-sized glass fiber
filters manifolded to ORBO™-60 100150mg, 6 x 70mm florisil sampling tubes.
The pumps will be calibrated by the Gilian Gilibrator (digital calibrator) prior to and after
each sampling period. The glass fiber filters will be precleaned prior to loading filter cartridges
by the North Carolina Public Health Laboratory and loaded at the laboratory by laboratory
personnel. The sampling head will consist of glass fiber filter cartridge followed by the florisil
tube. The two will be manifolded together by tubing. The purpose for this sampling system is to;
1) catch any possible PCB contaminated particulates from the air, 2) to filter PCBs which may
have vaporized. This system will then be manifolded to the sampling pump with plastic tubing.
There will be three sampling periods of eight hours. The sampling sessions will begin at
roughly 10:00am, 6:00pm and 2:00am. Each sampling period will be separated by one day.
The reason for this delay between sampling events is to provide time to recharge the air-flow
pumps (they require 16 hours of recharge time). Each sampling period will consist of seven air-
flow pumps and sampling units. One unit will be located inside the vent on top of the landfill.
Two more units will be located two meters downwind from the vent (these will be seperated
from each other by one meter). Two other units will be located diagonally downwind at the
landfill fence line. One unit will be located 200 meters straight downwind from the vent. The
air-flow rate will be calibrated to NIOSH guidelines. NIOSH Sampling Method #5503:
Sampling for Polychlorobiphenyls states that air-flow rate should be 50-200cc per minute. The
flow-rate per unit will correspond to its distance from the vent-the further the pump is from the
vent, the greater its flow-rate. Another sampling unit will be located 200 meters upwind and will
act as a background sampler. Prior to each sampling event, the wind direction and temperature
will be indicated by a portable weather station established on the east side of the landfill.
NOTE: Please see attached diagram of the landfill showing sampling locations and a copy
of the NIOSH lab procedure.
PKL/H&S/HWS/Dec96.
l I I I -
!
·--·-·-------·--·-----------
·--
mixture: C,2H,o-,CI,
[where x = 1 to 10] ·
MW: ca. 258 (42% Cl ; C12H1Cli.);
ca. 326 (54% Cl ; C12H5Cl5)
GAS: Table 1 RTECS: Table 1
EVALUATION: PARTIAL Issue 1: 15 February 1984
Revision #1: 15 August 1987
Issue 2: 15 Auaust 1994
OSHA : 1 mg/m1 (42% Cl);
0.5 mg/m1 (54% Cl)
NIOSH: 0.001 mg/m1/10 h (carcinogen)
ACGIH: 1 mg/m1 (42% Cl) (skin)
0.5 mg/m1 (54% Cl) (skin)
PROPERTIES: 42% Cl: BP 325 to 366 °C; MP -19 °C;
d 1.38 g/mL@ 25 °C;
VP 0.01 Pa (8 x 10·5 mm Hg;
1 mg/m1) @ 20 °C
54% Cl: BP 365 to 390 °C; MP 10 °C;
d 1.54 g/mL@ 25 °C; VP
0.0004 Pa (3 x 10·5 mm Hg;
0.05 mg/m1) @ 20 °C
SYNONYMS: PCB; 1, 1 '-biphenyl chloro; chlorodiphenyl, 42% Cl (Aroclor 1242);~nd 54% Cl (Aroclor 1254)
SAMPLING
SAMPLER: FILTER + SOLID SORBENT
(i3-mrri "glass flbe~ + Fiorisil,
,;100 mg/50 mg) . --···, . \, ,; .
FLOW RATE:)0.05_ \o 0.2 L/miri __ o_r_ ~ess
VOL-MIN: 1'"tJ J ~ 0.5 07g/!'T11
-MAX: 50 L . -·.
SHIPMENT: transfer filters to glass vials after sampling
SAMPLE
STABILITY:
BLANKS:
unknown for filters;
2 months for Florisil tut::es (1]
2 to 10 field blanks per se:
ACCURACY
RANGE STUDIED: not studied
none identified
not evaluated
not determined
BIAS:
OVERALL PRECISION (S,T):
ACCURACY:
;TECHNIQUE:
ANALYTE:
DESORPTION:
INJECTION
VOLUME:
MEASUREMENT
GAS CHROMATOGRAPHY, E:.:J r 'Ni)
p:lych lorct:ichenyls
filter +_ front sec:ion , 5 ml hexane: t::ack
sec:ion,_ 2 ml hexane
4-µL with 1-µL backflush
TEMPERATURE-INJECTION:
-DETECTOR:
250 to 300 °C
300 to 325 °C
180 °C -COLUMN:
CARRIER GAS: N:, 40 ml/min
COLUMN: glass , 1.8 m x 2-mm ID. 1.5% OV-17 /1.95:~
OF-1 on 80/100 mesh Chromoscrb WHP
CALIBRATION: standard PCB mixture in hexane
RANGE: 0.4 to 4 µg per sample [2]
ESTIMATED LOO: 0.03 µg per sample [2]
PRECISION (S,): 0.044 [1]
APPLICABILITY: The working range is 0.01 to 10 mg/m3 for a 40-L air sample [1 ]. With modifications, surface wipe samples
may be analyzed [3,4].
INTERFERENCES: Chlorinated pesticides, such as DDT and ODE, may interfere with quantification of PCB. Sulfur-containing
compounds in petroleum products also interfere [5].
OTHER METHODS: This method revises methods Si20 [6] and P&CAM 244 [1]. Methods S121 [7] and P&CAM 253 [8] for
PCB have not been revised.
NIOSH Manual of Analytical Methods (NMAM). Fourth Edition, 8/15/94
5-17-1996 10:02PM FROM P. 1
HIRSCHHORN
ffe)JASSOCIATES .
~ A Division of 1-/ygienetics Environmental Services. Inc. Suite 41.1 Phone: [phone]
2401 Blueridge Avenue rax: [phone]
Wheaton. MD 20902
December 17, 1996 ................................................................... byFAX
To: Laura S. Butler, Chief Air Pennits Section, DEHNR
From: Joel Hirschhorn
Subject: Request for docum~t
In my capacity as a Science Advisor for the Warren Co1mty PCB Landfill Working Group, I
request a copy of the document "Dispersion Modeling for Warren County PCB Landfill.," by Tom
Aoderson, Dec. 10, 1996.
Apparently you have distributed this document for review by certain parties.
If at all possible, would you please fax me the above document as soon as possible.
I also want to let you know that in addition to the two documents l have provided the Working
Group on the PCB air emissions issue I have produced some others that you should also provide
your reviewers.
. '
·-4 December 16, 1996 ............................................................. by FAX (5 pages)
To: Technical Committee
From: Joel Hirschhorn
Through: Doris (for distribution prior to WG meeting as muc.h as possible, and to all Work.ing
Group members at meeting of Dec. 17)
Subj,:d: Review of orieinal EPA contractor report on PCB air emisaiow
In response to my request~ I received the Final Report on Ambient Mo.uitoxing For PCBs At The
Warren Cowity (North Carolina) Landfill by Battelle from Dr. Lewis and have prepared the
following comments on it so that the Working Group can better understand the situation. lu
general I have found even more problems and reasons to be concerned about the entire hi.story of
this situation.
M.Y latest review, analysis and conc;lusioo! strongly support the position of th.~ll
Committ~in its re(l~nition that it is imperative that full disclosure of tht dan~t.f..LQf.~
landfill be openly communicated to the public and the state eovernment th~ a11other
press conference, letters to the Governor and Secretao· How~s, appropri31Uttions bein_g
taken by the state (:iu.d!.JJs carbon filter installation and nuv monitoring), and a well
planned r.omnrnnity meeti,n~ that the Governor is invited to participate in. I wavJ--1!!
e.mp__hasizc that the most important recommendation presented hy E.U~..oo.r i!!
!283, calling f9.r future periodic monitorin2 for PCB air releases. was NQT implcmenJed...h):
tb~ state.pr EPA. In my professional opinion this was a callous and scientific;1ll.Y.M.~..d
decisio11 by the state and EPA that sas:rificed protection of public health to de~.l
t.Q..!!Serns, like wantide to avoid eettine data that would conclusively demQD£t7.:.alfU..O
\Varnm County residents .and the broader public that the Warren County FCB~fW
was dangerous, un~af e. and a source of uncontrolled environmental rel~'l.._T_he
recommend,11,non for future periodic monitorin1 and the test results from J 983~il~t
J.Qci,cally resul!ld in the state takipg the proper precautionary step of installing tll.!:.Ml!
tilters on all landfill vent,, Thi~ matter de,,~un!ll the dronge.d pos~ihle action I)~" th~
\Vorking Group to ufe2.U,ard the interests of Warren County residentu.v__t!rnt effutiv,;
IW.oxification of the landfiU is actually accomplished as soon as possible, The GoverQ.QI
should be requested to determine why the recommendation of EPA's (on tractor to coruJ.~Kt
futw:e.monitotio2 for PCB emission~ wa:, not implemented,
The report was dated August 16, 1983,.indicating tbat complete data were available to EPA
rela.tively soou after the onsite testing had been done iu January and February 1983. Oftb.e 40+
page report, only page 2, Section 2, Conclusions had been found .in the state files provided me,
and which I previously conunented on.
1
Page 3 of the report was Secµon 3 Recommendations and included the following: "In the futpre.,
it is recommended that periodic monitoring be performed to determine the trend in PCB
emission rates from the gas vents and lcuhate access ports on the landfill site .. " ( emphasis
added) The single paragraph also included a recommendation to use somewhat different sampling
technology, called high volume PCB samplers, because they offered ''significantly lower detec.,iion
limits." It appears that these recommendations were never implemented by eithe.r EPA or the
state. The recommendation for future testing is particularly significant because there were
relatively few positive detections of PCBs in ambient air, indicating that the contractor recognized
that in later times PCB emissions and releases. offsite were plausibl~. In other words, the
contractor ,vas correct in recognizing the appropriateness of experimentally, th.rough further field
testing, determw.ing the "trend" of PCB emissions. This implicitly recognized the positive
detections of PCBs in vents and ports, as sources of environmental releases, as well as initial
detections of PCBs in ambient air. lt also recognized the reasons why the study had been
conducted.
The contractor report contafoed no information on the design or construction of the landfill,
which the published paper had presented. But the contractor report did note th.at: "Local
residents and the Warren County Health Department have expressed cone-em about the possibility
of airborne PCB emissions from the landfill being transported to neighboring areas, thus
threatening the public welfare." It also noted that "The study was performed at the request of
the North Carolina Division of Health Services ... " This makes it even more questionable why the
fuJl contractor report was not found in state files. Five specific study objectives were presented,
including dete.rmining if PCBs were present at the house approximately one-half mile away. Of 42
specific measurements in ambient air, however, only two were for the nearby house.
Th.ere is absolutely no statement iu th.e entire report that in any way supports the contention
recently communicated by EPA that the positive findings of PCB air emissions were false
positives.
There js a discrepancy between the details of the positive findings as preseuted iu the published
paper versus the contractor repo.rt. The pubJished paper indicated four positive detections in
ambient air and clearly gave the locations of three of the.mas: beside main vent, fence line
down\\-ind, and nearby house. The coil.tractor report gave: beside main vent, two for fence line
downwind, aud onsite upwind of main vent. The footnote in the published paper iudicating that
one of six measurements at the nearby house was above detection limit is inconsistent with data in
the report that indicates that only two measurements were taken at the nearby house. What is
most perplexing at this time is, therefore, that the published paper showed a positive finding of
PCBs at th.e .nearby house, while the contractor report showed all findings as less th.an the
detection limits. But the procedure used by the contractor was to consider less than dctectiou
limit findings as equal to the detection limit. Following the contractor~s procedure the average
concentration for Aroclor 1242 and 1260 for the nearby house would be 6 and l O ng/c~
respectively. In my previous comments I used what was reported in the published paper, namely
a positive finding of 10 ng/cm for .1260, which is still supported by the actual flndings in the
2
contractor report. Let me emphasize that the contractor's methodology of assuming a finding of
less than the detectioo limit is equal to the detection limit is far more appropriate than assuming
that such a nondetect finding is equal to zero, which seems to be EPA's position. Nevertheless,
there remains a discrepancy between the published paper's clear notation that a positive finding of
Aroclor J 260 was found at the nearby house versus the contractor report that in<licates that all
four readings on the two days of testing for the two Arodors were nondetects. 'What deserve$
attention by .EPA is why it reported, in the published paper, the positive finding of PCB at the
nearby house.
The variation. in detection limits was made understandable in the contractor report, because the
data given showed that sampling times and, hence, sample volumes varied somew-b.at, with
unanticipated lower volumes resulting in higher detection limits in some cases. In all cases, where
positive detections in ambient air were found, the detection limits were the lowest, making the
findings more reliable.
Although the published paper presented key PCB data in terms of ranges and averages, the data
presentation in the contractor report was different and, in my opinion, more useful and
informative. One problem in the published paper i.s that some data were reported as "ND" which
is standard notation for non-detected, while other data were reported as "<'' meaning less than the
following numerical numbe.r which normally would be the actual detection limit (DL). ln fact~
there was no reason to .report NDs at all in the published paper, and in the contractor report data
were reported either as positive detects or as <DL for PCB concentrations. The point is the NDs
represented a specific test in which no reliable data were obtained. But the published paper said
that ND represented ".nondetectable." This definition misrepresented the actual meaning of some
data. Fo.r example, for the lower leachate access port, the publishe.d. paper indicated the low end
of the two ranges were ND, but this actually referred to a test where the data were not usable .
The illference was that a ND low end represented a zero level rather than no reliabJe data. The
published paper reported averages of 0.04 and 0.05 ug/cm for 1242 and 1260, respectively, v,,Jlich
were calculated by assuming that ND = 0 and that <DL = DL. But using the actual data in the
report, and eliminating the ND values results in the averages being .06 and .08 ug/cm,
respectively. Clearly, the corrected averages portray a significantly higher level of PCBs from the
lower leachate access port.
Another systematic source or widerreporting PCB levels was the unusual procedure of not
rounding off numbers follo\\-ing the normal practice that if a digit is 5 or greater, than the
preceding digit is increased by one. For example •. 056 should become .06. But in the published
paper, some averages and other figures are too low, because rounding was not done properly.
For example, the average PCB levels for small vent E were reported as 0.24 and 0.4, but should
be 0.24 and 0.5.
There is an important lack of details about the weather conditions existing at the time of sampling.
For example, for Jan. 30 the notation was ''no sampling due to weather c~ditions," \\.ithout,
however, explaining what the weather conditions were. The relative humidity for the following
3
two days on which testing wa.s done indicated very high level~ suggesting that perhaps rain or
snow was a factor d'uring the testing period. In general, the collection of meteorological data was
incomplete, and th~ report presented no detailed analysis of bow weather conditio.us might have
affected the testing results, both positive and negative detections. Io my opinion, the weather
conditions were probably not the most conducive for transporting PCBs offsite which could exist
at other times. This supported the study's recommendation for additional future testing. For both
periods during which monitoring was done for the nearby house the wind was .not blowing from
the landfill in the direction of the house. This fact and the very limited number of measurements
taken at the house indicate, in my opinion, poor study design.
The contractor report made an important observation about the finding of Aroclor 1260 but not
1242 in. ambient air: "lt is puzzling that Aroclor 1242, the major component of the vent emissions,
was not also detected in these samples. 11 lhis is the issue raised recently by EPA However, the
contractor does not attempt to explain why the noted observation occWTed or how it can be
explained, and certrunly never raises the possibility of false positives.
In eight cases reliable data was not obtained because sample pumps malfunctioned, wfrh the only
reason g.ivel) as "due to the low temperature and high humidity conditions that prevailed during
the sampling period. 11 This supported monitoring at a later time under different climatic
conditions.
Tue published paper did not note that the two highest ambient air detections were for the same
location at the same time, but for two heights of 4 aud 15 feet, with 71 and 50 ng/scm,
respectively. The level of PCBs at the higher height is significmt.
The contractor report made it clear that gas flow rates from vents and leachate access ports were
measured by EPA .Personnel on ''March 2, 1982." Presumably this was an enor (not caught in
peer review), and was really in 1983, but it was definitely after the contractor performed the
major portion of the work. And while the published paper some data on the gas flow from the
maiu vent, it did not present the detailed data in the most understandable and sign.ificant \-vay. The
contractor report's presentation oftlte data showed that gas flow rates increased siguHicautly
from 9:45 AM to 3:00 PM, with a 32% increase over that period. This would logically suggest
that gas flow increased as temperature increased (no climatic data was given for this day in the
report)~ supporting my contention. that ~ir releases of PCBs increase with increasing temperature
and also supports the report's recommeudatio.n for subsequent monitoring at the landfill. The
modeling work by the co.o.tractor cited by EPA 011ly used the average of the data obtained for
different times, which does not, therefore, produce results indicative of maximum PCB release
rates.
The study design was deficient because the sample volumes for the small vents and the leachate
access ports was the same as for the main vent, which resulted in higher detection limits for the
fonnet versus the latter, because the quantities of PCBs in the former were substantially smaller.
Such a situation could have been predicted, and at the very least this problem \:\ould have been
4
\
I
I
I
corrected in subsequent monitoring by increasing the sampling rate and volume (as was done for
th.e ambient monitoring).
1 found it unusual th.at several significant statements made in the Conclusions section were not
contained in any form ·within the body of the report, particularly Section S Results and Discussion.
This is highly unusual, because normally there is a more extended discussion in the body of the
report that supports summary conclusions. For example, the comments in the conclusions about
PCB emission rates being reduced in the future are not supported by any information or analysis
given in the body of the repo.rt. Nor are the comments completely consistent with the report's
major recommendation for future periodic monitoring. Other statements in the conclusions are
inconsistent with the details given in the report, namely that ambient PCB levels were only found
"at or below minimum detection limits'' while, in fact, the report's data showed positive detections
as high as seven tim.es the detection limit. Based on my professional experience, these kinds of
unsupported and inconsistent statements m conclusions result from the contractor's dient (EPA)
making changes during a final review of the draft report in order to protect or serve its ov.n
interests.
As to peer review that Dr. Lewis has claimed is so significant, I want to point out that the peer
review process has many limitations, including the fact that a primary report that forms the basis
for a published paper is not provided peer reviev,rers, so that some shortcomings in the paper
cannot be properly assessed.
5
..,. •Jtate of North Carolina
Department of Environment,
Health and Natural Resources
Division of Air Quality
James B. Hunt, Jr., Governor
Jonathan B. Howes, Secretary
Alan W. Klimek, P.E., Director
December 16, 1996
MEMORANDUM
TO:
THROUGH:
FROM:
SUBJECT:
Bill Pate
Division of Epidemiolog~
1
/
Alan Klimek, Director V"
Division of Air Quality
Laura S. Butler, P.E., Chief j/fY
Air Permits Section
PCB Air Emissions
Warren County PCB Landfill
Warren County
AVA
DEHNR
We have received a report concerning air emissions from the Warren County PCB
Landfill. The Warren County PCB Landfill Working Group is concerned about potential
human health effects and environmental impacts from PCB emissions. By way of this
memorandum I am requesting your guidance and input on these issues. Copies of the
following documents are attached:
1. Final Report On Ambient Monitoring for PCB's at the Warren County (North
Carolina) Landfill, D.L. Sgontz, W.E. Bresler, L.A. Winker and J.E. Howes, Jr.,
Battelle Columbus Laboratories, August 16, 1983.
2. Measurement of Fugitive Atmospheric Emissions of Polychlorinated Biphenyls from
Hazardous Waste Landfills, Robert G. Lewis and Barry E. Martin, Environ. Sci.
Technol., Vol. 19, No. 10, 1985.
3. PCB Air Emissions and Health Risks from the Warren County PCB Landfill, Joel
Hirschhorn, Warren County PCB Landfill Working Group, November 27, 1996.
4. Air Emissions of PCB and Associated Health Risks, Patrick Barnes and Joel 0 .
Kimrey, December 2, 1996.
P.O. Box 29580. Raleigh, North Carolina 27626-0580
Voice [phone] FAX [phone]
An Equal Opportunity/Affirmative Action Employer
50% recycles/lO"k post-consumer paper
5. Hirschhorn Report on Warren County PCB Landfill, Dr. Robert G. Lewis, December
3, 1996.
6. Dispersion Modeling for Warren County PCB Landfill, Tom Anderson, December
10, 1996.
I
C: ~/William Meyer -Division of Waste Management (w/o attachments)
Ernie Fuller -Regional Air Quality Supervisor (w/o attachments)
Lee A Daniel (w/o attachments)
George Murray (w/o attachments)
State of North Carolina
Department of Environment,
Health and Natural Resources
Division of Air Quality
James B. Hunt, Jr., Governor
Jonathan B. Howes, Secretary
Alan W. Klimek, P.E., Director
MEMORANDUM
December 16, 1996
TO: Lee A Daniel, Chief
Technical Services Section
George Murray, Chief
Ambient Monitoring Section
THROUGH: Alan Klimek, Director ~
Division of Air Quality
FROM: Laura S. Butler, Chief ~
Air Permits Section
SUBJECT: PCB Air Emissions
WarrenCounty PCB Landfill
Warren County
AVA
DEHNR
On December 2, 1996, I sent you and Lori Cherry a copy of the report PCB Air
Emissions and Health Risks from the Warren County PCB Landfill for your review and
comment (copy attached). Additional reports have been received and are attached for
your review and comment. Please advise me on your opinion of the necessity for
performing air monitoring for PCB's at the Warren County PCB Landfill.
The Warren County PCB Landfill Working Group is concerned about potential human
health effects and environmental impacts from PCB emissions. Your earliest review is
requested.
C: Bill Pate -Division of Epidemiology (w/o attachments) )
\/William Meyer (w/o attachments)
Ernie Fuller (w attachments)
Lori Cherry (w/attachments)
Attachments: (6)
P.O. Box 29580, Raleigh. North Carolina 27626-0580
Voice [phone] FAX [phone]
An Equal Opportunity/Affirmative Action Employer
50% recycles/ l 0% post-consumer paper
1. Final Report On Ambient Monitoring for PCB 's at the Warren County (North
Carolina) Landfill, D.L. Sgontz, W .E. Bresler, L.A. Winker and J.E. Howes, Jr.,
Battelle Columbus Laboratories, August 16, 1983.
2. Measurement of Fugitive Atmospheric Emissions of Polychlorinated Biphenyls from
Hazardous Waste Landfills, Robert G. Lewis and Barry E. Martin, Environ. Sci .
Technol., Vol. 19, No. 10, 1985.
3. PCB Air Emissions and He~lth Risks from the Warren County PCB Landfill, Joel
Hirschhorn, Warren County PCB Landfill Working Group, November 27 , 1996.
4. Air Emissions of PCB and Associated Health Risks, Patrick Barnes and Joel 0 .
Kimrey, December 2, 1996.
5. Hirschhorn Report on Warren County PCB Landfill, Dr. Robert G. Lewis, December
3, 1996.
6. Dispersion Modeling for Warren County PCB Landfill, Tom Anderson, December
10, 1996.
Fax:[phone]
MEMORANDUM
TO: Bill Pate
Division of Epidemiology
THROUGH: Alan Klimek, Director
Division of Air Quality
DRAFT
FROM: Laura S Butler, P.E., Chief
Air Permits Section
SUBJECT: PCB Air Emissions
Warren County PCB Landfill
Wa1Ten County
Dec 13 '96 14 :26 F·. 01 /03
'· ·-·•·· 1/ l ;·,
I
'·-·._ ,i "• ' ..... ' r.
We have received a report concerning air emissions from thE, Wairen County PCB
Landfill. Tr1e Warren County PCB Landfill Working Group is concerned about potential
human health effects and environmental impacts from PCB emissions. By way of this
memorandum I arn requesting your guidance and input on these issues. Copies of the
following documents are attached:
~. Measurement of Fugitive Atmospheric Emissions of Polychlorinated Biphenyls from
Hazardous Waste Landfills, R.obert G. Lewis and Barry E. Martin. Environ. Sci.
Technol., Vol. 19, No. 10, 1985.
2. PCB Air Emissions and Health Risks from the Warren County PCB Landfill , Joel
Hirschhorn, Warren County PCB Landfill Working Group.
3. Air Emissions of PCB and Associated Health Risks. Patrick Barnes and Joel 0 .
Kimsey, December 2, 1996.
4 Hirschhorn Report on Warren County PCB Landfill, Dr. Robert G. l ewis, December
3, 1996.
5. Dispersion Modeling for Warren County PCB Landfill. Jim Roller, December 10,
1996.
C: William Meyer -Division of Waste Management
Ernie Fuller -Regional Air Quality Supervisor
Lee A Daniel
George Murray Post-it'' Fax Note 767 l D<1I<, / ~i /, 1:~.sLs ► ,--__:.---,--,----'-----+::----:--:---..__ ___ -j
To ,:. .. · ' ; I ' From
,1 .J f I j i :· • :~-'· ,>: I .
Co.
Pl1one II
Fax 11 / 1 1 ,::-Fax~
r
Fax:[phone]
DRAFT 3
MEMORANDUM
TO. lee A Daniel. Chief
Technical Services Section
George Murray, Chief
Ambient Monitoring Section
THROUGH: Alan Klimek, Director
Division of Air Quality
FROM La1.1ra S. Butler, Chief
Air Permits Section
SUBJECT: PC l3 Air Emissions
·warren County PCB Landfill
Warren County
Dec 13 '96 14:26 F·. 02 /03
On December 2, 1996, 1 sent you and Lori Cherry a copy of the report PCB Air
Emissions and Health Risks from the Warren County PCB Landfill for your rev iew and
comment (copy attached). Additional reports have been received and are attached for
your review and comment. Please advise me on your opinion of the necessity for
performing air monitoring for PCB's at the Warren County PCB Landfill
The Warren County PCB Landfill \/\forking Group is concerned about potential human
health effects and environmental impacts from PCB emissions. Your earliest review is
requested
C Bill Pate -Division of Epidemiology
\Nilliam Meyer
Ernie Fuller
Attachmants: (5)
Fax:[phone] Dec 13 ·~3 14:27 F·. 03 /03
i _ Measurement of Fugitive Atmospheric Emissions of Polychlo:-inated Biphenyls from
Hazardous Waste Landfills, Robert G. Lewis and Barry E Martin, Environ Sc i.
TechnoL , Vol. 19, No 10, 1985.
2 PCB Ai r Em issions and Health Risks from the Warren County PCB Landfill Joel
Hirschhom, Warren County PCB Landfill Working Group.
3 Air Emissions of PCB and Associated Health Risks, Patrick Barnes and Joel 0 .
Kimsey. December 2, 1996.
4. Hirschhorn Report on Warren County PCB Landfil l, Dr. Robert G Lew is , December
3 , 1996.
5. Dispersion Modeling for \Narren County PCB Landfill , Jim Roller, December 10,
1996.
J State of North Carolina
Department of Environment,
Health and Natural Resources
Division of Solid Waste Management
James B. Hunt, Jr., Governor
Jonathan B. Howes, Secretary
William L. Meyer, Director
December 12, 1996
Dr. Terry F. Bidleman
ARQP
Environment Canada
4905 Dufferin Street
Downsville, Ont., Canada M3H 5T4
Prof Ronald A. Hites
School of Public Health & Environmental Affairs
Indiana University
10th & Fee Lane
Bloomington, IN 47405
Dear Prof Hites and Dr. Bidleman:
It has been brought to my attention that you are an expert on PCB air monitoring and
familiar with the work of Dr. Robert G. Lewis. Dr. Lewis authored an article published in
Environmental Science and Technology, October, 1985, pp. 986-991, titled: Measurement of
Fugitive Atmospheric Emissions of Polychlorinated Biphenyls from Hazardous Waste Landfills.
The quality and scientific interpretations of the monitoring study have recently been questioned. I
would like to respectfully request your review of both the article and recent responses to
questions concerning the quality and conclusions by Dr. Lewis and Dr. Joel Hirschhorn. Your
response to this request is appreciated and will be presented to a Joint Warren County/State PCB
Landfill Working Group established for detoxification of the landfill.
Thank you for considering this request on our behalf and assisting our efforts to address
complex scientific and technical issues as we make progress toward ultimate detoxification of the
PCB Landfill.
If you need clarifications, information or assistance please contact me at 919-733-4996,
ext. 202.
w
P.O. Box 27687,
Raleigh, North Carolina 27611-7687
Voice 919-733-4996
Sincerely,
~cf~~ ~ William L. Meyer, Director
ff mffili:Jldk4id
FAX [phone]
An Equal Opportunity Affirmative Action Employer
50% recycled/10"/o post-consumer paper
,
Enclosures:
Measurement of Fugitive Atmospheric Emissions of Polychlorinated Biphenyls from Hazardous
Waste Landfills, Robert G. Lewis and Barry E . Martin, Environmental Science and Technology,
October 1985, pp. 986-991.
Draft Final Report on Ambient Monitoring for PCB's at the Warren County (North Carolina)
Landfill, D. L. Sgontz, et.al., Battelle Columbus Laboratories, April 8, 1983.
PCB Air Emissions and Health Risks From the Warren County PCB Landfill, Dr. Joel Hirshhorn,
November 27, 1996.
Air Emissions of PCB and Associated Health Risk, Patrick Barnes, December 2, 1996.
Hirschhorn Report on Warren County PCB Landfill, Dr. Robert G. Lewis, December 3, 1996.
Response to EPA letter of 3 December, previously provided; Dr. Joel Hirschhorn, December 10,
1996.
Response to Hirschhorn's latest comments, Dr. Robert G. Lewis, December 11 , 1996.
copy without enclosures:
Joel Hirschhorn, Science Advisor to PCB Working Group
Patrick Barnes, Science Advisor to PCB Working Group
Doris Fleetwood, Joint Warren County/State PCB Landfill Working Group
Dr. Robert G. Lewis
c:wpfiles/pcblf/bid-hite.ltr
1
12/11/96 1~:~9 _ '5'919 541 3527 NERL/ .'\.MRD /~IB
DATE:
TO:
U.S. ENVffiONMENTAL PROTECTION AGENCY
NATIONAL EXPOSURE RESEARCH LABORATORY
AIR 1v1EASUREI'v1ENTS RESEARCH DIVISION (:rvlAIL DROP 44)
RESEARCH TRIANGLE PARK, NORTH C.-\ROLINA 27711
TELEPHONE [phone] FACSr:MILE 91 9-541-3527
11 December 1996
Bill Meyers
E-:tvlAIL lewis. [e-mail]
FAX 1v1ESSAGE
NUMBER OF PAGES: 2
Fax No.: 715-3605
Telephone:
FROM: Dr. Robert G. Lewis
USEPA (MD-44)
Research Triangle Park, NC 27711-2055
SUBJECT: Response to Hirschhorn' s Latest Comments
MESSAGE:
14] 001 /0 02
I will not dignify Dr. Hirschhorn by responding to his latest attack on the quality of the EPA
monitoring study at the Warren County landfill and my scientific interpretations. It is obvious that nothing
will persuade him to abandon the faulty reasoning that he has committed himself to in order to convince
the citizens ofWarren County that they are at risk.
I will, however, provide you -with my biographical sketch in support of my credentials and suggest
that you ask him to do the same. I can send you my full C.V. (28 pp), if you wish. I believe that I am
vvidely recognized throughout the world as an expert in environmental monitoring, especially for organic
compounds in air. I wrote the defining chapter on sampling for organic chemicals in air for the American
Chemical Society's Professional Reference Book, Principles of Environmental Sampling (L. Keith, ed.),
and am currently writing a book on the subject for the ACS. I have also authored or coauthored eight
ASTM standards and two ISO standards on monitoring methods, including one on the methodology
(updated) used in Warren County (A.SIMD 4861). Furthermore, Battelle Memorial Institute, which did
the field v:ork at the landfill for us, is one of the leading environmental research institutions in the U.S. I
would suggest that you get the opinions of two of the world's leading PCB air monitoring experts as to
the validity of our conclusions and Dr. Hirschhom's arguments. They are both very familiar with my work.
Dr. Terry F. Bidleman Tel. [phone]
ARQP Fax. [phone]
Environment Canada E-mail: [e-mail]
4905 Dufferin Street
Downsville, Ont., Canada M3H 5T4
and
Prof Ronald A Hites (current editor of ES&T)
School of Public Health and Environmental Affairs
Indiana University
10th. & Fee Lane
Bloomington, IN 47405
Tel. [phone]
Fax. [phone]
E-mail: [e-mail]
,.
'
'8919 541 3527 NERL/AMRDOIB 14] 002/0 02
Biographical Sketch of Robert G. Lewis
ROBERT G. LEWIS is a Senior Scientist in EPA's National Exposure Research Exposure
Laboratory, Research Triangle Park, North Carolina. He is responsible for overseeing research
focused on new sampling and analytical methods for pesticides, PCBs, dioxins and related
semivolatile organic chemicals and for their application to total human exposure assessment.
Dr. Lewis received bis Bachelor of Science in Chemistry from the University of North Carolina,
Chapel Hill, North Carolina in 1960 and bis Ph.D . in Physical-Organic Chemistry in 1964 from the
University of Wisconsin, Madison, where he was a National Science Foundation Fellow and a
:-Jational Institutes of Health Fellow.
F ollm:ving six years with industry, Dr. Lewis joined the EPA in 1971 and held the positions of Section
Chief from 1971 to 1978 and Branch Chief from 197 8 to 1993. He has done substantial research on
monitoring and exposure assessment methods for organic chemicals in air and other environmental
media :md in human and biological tissues and fluid:;. He established and headed for 12 years EPA' s
first ambient air methods development program. He was a member of EPA' s first task force on PCBs
in 1972 and in 1983 was appointed by the Governor of North Carolina to the Intergovernmental
Working Group on PCB Detoxification.
Dr. Lewis has authored or co-authored more than 200 journal articles, books, book chapters, and
other scientific articles and reports on various matters of environmental concern. He has been
awarded the EPA Bronze medal for Commendable Service twice and has received four Scientific and
Technological Achievement Awards for outstanding journal articles.
Dr. Lewis is a member of the American Chemical Society (since 1958), Air and Waste Management
Association, American Association for Aerosol Research, American Society of Testing and Materials
(Fellow), International Standards Organization, Phi Beta Kappa, Sigma Xi, Alpha Chi Sigma
(Chemistry), Phi Lambda Upsilon (Chemistry) and Delta Phi Alpha (German). He has been listed in
American Men and Women of Science since 1966. He has served on numerous committees and work
groups concerned with environmental monitoring, exposure assessment, and the cuemical profession.
He is currently chairman of ASTM D22.05.02 on Organic Chemicals in Indoor Air, U. S.
Representative to ISO on Indoor Air, Research Area Manager for Environmental Measurement
Methods for the EPA Residential Pesticides Exposure Research Program and an advisor. to the EPA
National Human Exposure Assessment Study (NHEXAS), NCI/EPNNIEHS Agricultural Health
Study, the Lower Rio Grande Valley Environmental Study (NAFTA), the Great Lakes Air Deposition
Project and the Pesticide Spray Drift Task Force.
Air ()uality Division
December 10, 1996
~:fEMORA\lDUM
TO: I ,aura Burler, Assistant Chief: Pem1ining Section
$;,
FROM: Tom Anderson;lvfeteorologist, AQAB
THROUGH:~ Roller1 Supervisor, AQAB
SUBJECT: Dispersion Modeling for \Varren County PCB Landfill
Warrenton, NC Warren County
Per your request, I have conducted a screening-level dispersion modeling analysis for emissions of
PCBs at the Warren County PCB Landfill. Potential emissions of PCBs occur from a single gas
vent located in the center of the landfill. Maximum modeled impacts ,.vere less than 0.1 %i of tl-le
annual AAL specified in ~CAC 2D . ! 100.
E:nissions of PCBs occur fron: a 4-inch diameter PVC standpipe protruding through the cap of
the hmdfill which is used to vent biogenic emjssions from the facility. Emissions arc expected to
occur at or near ambient temperature with an assumed exit velocity of .01 m/s (due to the
J'-, .,,.vv ... , J i ,•,iMil\i'b !!U(l. A.P. !!l!~iuuiu.-. 11ruuo of lJ. l o D gmmcfcoo . (obtai.n1ild froni thi Fr Qi
research study paper) was used in the analysis.
SCREE~3 (96043) was used to evah.:ate simple terrain impacts from the vent. Since the l:.u1d
around the landfill is generally of lower elevation and due to the lov.: release height and ambient
namre of the emissions: complex terrain was not modeled. Receptors were placed at t.he closest
distance to the fence line ( 45 meters) and extended outward to approx.imat•~ly 3,000 meters. The
maximum impact of 7.4 e-7 mg/m3 occurred at the Lenee line and is well below the annual AAL
o f 8,3 e-5 mg/m3.
Pos:-it•· Fax Note 767 1
1
11
,1 I i! \1-1 •> ~~' . ./ ./'
Ph,Jc,o #
',•:·· . . , .....
December 10, 1996 ........................................................................... by FAX (five pqes)
To; Technical Committee
From: Joe.I Huschhorn
Subject: llespon~ to EPA Jetter of 3 ~er, previously provid~d
In examining the EPA 1ett~ to Bin Meyer I focused on the main points and I have
con~l1Jde-<! that I have no r~a=on to abandon or change my original findings and concluaon&. I
have faxed :a request to the ~uthor of the EPA letter for a cop>· oftht brigi,a} contractor (Battelle
Memorial Institut~) report, which, :f provided, m.,y re\leal other interestini infonnation It is
impona.nt to emphuize that only two documents have sarfaced regarding the original 1983
measurements b:y EPA, the published paper from 198.5 and one pag; of conclusions from what
cPA no·w says was a contractor report 11repued for it. Only the latter would have the testing
de~s allo.,..'Dl.g the. most complete anaJysii of what actually happened dwing r.he testing.
£PA has said that "our statements that PCB cmit,Sions from the land.fill were negligible at
the :-irne cf the 5t.idy were suppc,tted by the highest lC\'Cl oft~bn.i~al review." The problem is
that th~ paper itse~ as r noted pJ"e'\-iously b.ad a logi"1 inconsistency betweeu two of its
statements, o.ne s.aying th" emiuion, were negligible and the other that the :miS&O!Hi were at low
lev~is: Cle.arty, PCB emidion~ were m.u£:1.aed md reported int.he paper. Any re-.iewer of the
paper, how""9··er. xmy h•ve missed the importance of the two statell\eftts tad, moreover, may have
mtetpretcd the statement about emissions from the Wa?Te!l County landfill being negligible as not
ne,essanly related to health eff'cc:ts. ID fact, the paper presented no ill.formation about PCB levels
and heal-J:.. effects. Tht paper wa.s about PCB fflli1sions from different land disposal situations.
oo, potential health impacts.
EPA is nr..w defending itself by aayin& that the 1ct11aJ measurements orPCBs 11wt.re likely
f",1-sc-positive results.'' This .is a .rcnwkable statement. It wa.s not tnlldc m the original paper, nor
is th~re any C'-idrocc t•:> support th.ii c-0ntmtion. Now, EPA clA.iros that 11lt would be cleat to 1uy
annlyti~al chemiit or llnyone with environm=tal monitoring e,q,ertist that these were likely false-
posifrve relAlhs." Notice the q,.ulmer "likely." If £Cientisu have a veiy sound technical reason for
be?icvmg that e~erimmtaJ data are fal.5e•positive result, (i.e., a &ding of 2 chemical ,\ihen none
is re~lty pr~sent), they either would not repon the thta in a publication or they would clearly
utirulate why the data should b~ vie\\·~ as false positives. Any objective analysis oft.be EPA
published paper ~ill &how that there is no 9U.Uestion or ptooft!tat the reported ambient air PCB
lc:vels were ftlsc--pow:es. 'Fo.r EPA to now claim that ;u data were likclJ false-positives is
cisi:agennous at best. ~d for it to attempt to create explanations now but n~t iD the publisbed
paf et is n.ot normal profession&l conduct.
A pvt of the Cllltent :.onteution about false-positives is that if the data were "rea1" •• and
1
I •
FEV. :919-257-1000 Dec 10 '96 16:15 P·. 05
they c;-erta.icly were ral enough for the EPA pm-sous to report and vublisb them d the:o the more
vo!atile Atoclor 1242 rat.her the the ff'~as:ured 1260 would have been measured. But there are
various pcssi'blt e~taDationa of-....hy the less vo~ti1e and not the mote vob.tile Atoclor could be
found in :i.mbiezit air l!lOI1itorillg. These possible reasons include: variable t;,pe.s of PCB emi.uions
&om th.t landfill Jt different time1, making a small number of mumremenu in ambient air not
reflective of the cumulative types of emissions over time; a greater loss of the more volatile
Arc,;lor prior to me time of the measurement especially because ofthe mny months that the
wastes were e,q>osed to the atmosphere when they were first dumped on the roads and continuing
through their diaposal in the open pit; different environmental interaa.ions dwing the air trau.sport
of the-vapors. le•dmg to ditTermt variations of the PCBs reach.mg a given point ofme,uurem.ent;
the greater water solubility of 1242 versu, 1260, by• factor of about JOO, that .in the very wet
nwm.ils 1Mide the landfill could explain why considerable 1242 could be trapped in the water and
not relea~ to the air (i. e . the difference in water solubility is .much greater than the difference in
v.tpor pres..~re); mofe of the 1242 may have been biodegraded over th~ ~tirt time the wanes
exiited prior to th.e gas v..porization.
The letter from EPA no:e, that the more normal higher lellel of 1242 over 1260 wss found
in th~ m.1m and upper leacb.a.te vent pipes. But of ooursc it docs not ootc: that tb.e relationship
herv..·een the t\\-'O Aroclors was not the non:nal t;pe m three other categories reported in the
orig,in;.l published paper (lower leachate port and two small vents). Most interestingly, for
~xample, one of the mw1 vents (E) had Aroclor 1260 levels twice th.at of 1242, and at levels tliat
the detection i~sue is inelevant. Clearly the~ arc er.r.iisrion.s me.sured in the small veot whe1e a
bubble h11d formed in the top plastic liner Cf more 1260 than 1242 clearly c&me ftom this source
~flandfilJ gas, then wh}· does EPA make such a issue cffin.dmg more 1260 than 1242 at my
off.site location.? EPA also m.axes the mistake of a~ing that a non•detect finding i~ !qu.at tn
zero concco.tration,. when in tact that is a scientlncaJJy illcorrect and improper uwmptioo to
nah At ill the other arcl>ient testing loca.tioM where no positive ltvels of PCBs wa-e reported,
tbe values m•y in f.act be ~mething less t1w1 the detection limit.
The letter from EPA also attempts to mak! a i,oint that the ambient sir levels found "did
o,.,t appe3J' correlate (.sic) with proxi:m¼ry to the vents." J.n actual fact. there ii problem in EPA 's
mtetpreta.tion be.cause immediat~ly after the statement cited, the paper tlsc said "The two
m.lXJ!nl.l.111 levels were found QS rn down.wind of the ma.in vent. '1 In other words, of the four
posit~·c finding, of PCBs m ambient air reported as .. 11, 12, .SO, and 71 ng/mJ\ iccordmg to EPA
the teadings of SO and 71 (now claimtd as not ru!, but inaly%.ed nevertheless by EPA) were net
at th~ ho.isc. which was, according to the paper, some 1000 m away. According to the paper, the
highest level mea&Uted was at the fenceline and logically S() wa11. the second highe6:t level f'ound .
Tht.t leaves two very siimlu but lower PCB levels foWld At two other locationt, which according
to Ttb!e TI oftbe papet were at the nearby house an.d beside the main vent. The only
meuurement, ther~forc, that is somewhat in'egular is the lower level fo\llld beside the main vent.
because it seem~ much too low (i.e., ii there are a .mfficietitly luge number of mururemmts one
expects to find the highest concentrations at the ~uroe, with steadily lower concentrations at
.increasing -ii~r.c.es a."'t) 5-om the source). But the anomaly is the very low reading next to the
2
Fax:[phone] Dec 10 '96 16:16 F·. 05
lll2ll2 vent, especially becauM much hlgher leveli were .musured mlide the ~in .... en, and EPA
uid that there was a pct;itive gas fiow with PCBs being emitted .&-om the mam vent However,
su,li low readings where higher levels are expected cu be explamed by, for example, vari~le
cwission rates, variable wmd conditions, and poH1"bly ditferent time, ofmetMements at different
locstions.
Indeed, EPA w a big problem trying to e,cpwn why the high le\lels tound inside the main
v!ftt did iot uanslate into much higher level& ''beside main vent" ~ticb pretumably was.
a«:~rding to the figure showio.g th.e sampling locations, only l tn away. Tbe problem of course is
that variable ~ind eonditions will lead to vastty different levels measured at any given location,
unl~s, many more meuure~5 arc taken over time. The original paper noted that wind speed
\'aried n-om 0.04 to 6.6 mis, and that the ambient ai.r monitorir..g was not done on the same days
thit testin~ took place in the vents. The latter fact suu~.sts that highly variable emissions from
the land.6li wete occurrins, expltining why a very low reading might obtained near the main vent.
AJso. the paper indicates that the ambient air monitoring ocCUJTed on different days, explaining
why J>CB levels would chll)ge signm.cant)y for ditfc:rent locations.
The original EPA paptt said chat ambient air <hta were given in its Table lI and th.at .,o»ly
four oftb~ 39 ambient ili sample~ ~-zed contained detectable quantities of PCB" and that
"threfZ ~mples were posirive for Arodor 1260 only (at 11, 12, 50, ad 70 og/mJ)." The problem
is that in Tablf' II there are three footnotes for etch of the three data showing poiitive L?vels of
PCBs detected. as follows:
b~side ~ vent: ,., One of~ measurements above detection limit."
fenc~ line, downwind: "h Two of' 13 mea~ements above detection limit (0.05 and 0.07
"'g,'m') .. ,
neuby house: •• One of six measurements above detection limit."
There is a discrepancy betwCC11 these footnotes and the body of the paper's text. The total
~umber ofmeasur~mcnt SHms like 25, not 39. Another problem is thit the tabulated data
indicate t.hit the detection limit was 10 a.g/m3• but the ieCOnd £ootnote indicates that the detection
lhnr~ was from ~O to 70 ng!ni', and both the original paper and the recent EPA letter said that the
d~te,.,,-tion limits wete froo:i 10 to 20 ng!m'· while the page of c;onclusious from EPA's conuactor
repo.'1 said the detection limits were 6 to 10 ng/m.1. Only cJ01e examination of the original
aboratory d;.ta meets that might be in the Battelle ccmuactor report coitld resol'\'C these
discrepancieG.
EPA's letter cltims wt "the vent pipes were not the souru of the PCBs detected m air,
even 1f the m~asured air values were real. 1' First, EPA is ac'kDowledging that the data may
acr.ially be accurate positives and, second. EPA is speculating that some other source nf PCBs
existed aroW1d the landfill. In fact, m its original publis.hed paper, EPA said that ,.,ir 1evcls were
.tor ne.s.r bade.round" at the Wure:n Cowtty ludfill, which &c.:ienti5¢ally is a very di1ferent
arg-Jment than its present one based on claiming th.at tbe 6.ttdings were false-positives. In other
3
Fax :91'3-257-1000 Des 10 '96 16:17 F.07
words, originally EPA disregarded the .levels of PCBg it meuured iD ambient air by arguing they
w:re "at or near back11owd" and now by arguing that they were not reall)' positive findings. But
in fact, evi:u the EPA ar~t that the measu.red (real) PCB level& were onl)' bacqroUDd does
not sund up under clo,e scrutiny.
EPA cites the general data for the U.S. about hish}y vlriable amhimt &it levels ot"PCBs,
\\ith the maximum being 10 n,/m', which is actually less thlll all four positive findings reponed in
tht i,apu (11, 12, SO ud 71 nglm') More importantly; the range given iD EPA's letter (and
grvec origm.,ally in EPA's contnetor report, but not the published paper) does not distinguish ihe
typ~s ofiocations at which PCBs have been meamred In &ct, the lugbl!'J' level$ of unbient PCB
Je,,·els COJTespond to more industrial :and urban areas, and areas near any type of .incineration. of
waste . not the rural area at which the Wa.rren CoWlty landfill u located. Lcgicall)-. the low ez.d of
the: range report~ namely S ng/m3, might be approprute for mch I rural looatloo, but the ~orrect
scimtific method in rucb u mvutigation is ac:tuaDy to uke measurements at a substantial di£tance
from the lfic.ely point source (i e., the landfill) but within the local geographic area. Then, this
experimental bai;kg:rnund le-vet appropriate for the fflldy should be used. But assullling that the
low en.d of the MtiOtlal tU>gc may be appropriate, the four positive findings are significantly
higher, undumining EPA's qumeot.
The letter from EPA cwms that the 't'fbe Warren County site was brand new and should
b:.ve b~ emitting ~t a ma,;;imimi rate." This is sheer nonsense and perhaps more than any other
EPA .statement d:mo.ostntes that EPA pcnon1 are trying very hard to cover up the truth. My?
The:-e are vario"s sound. sdcntific reasons why 1and6ll cznis..~n rates were .oot n~ssarily the
highest at the time of the .EPA measurements, inc;ludmg: th~ tempmture at the time was low (-1
to • l .ibC) relative to mm.mer periods ~-hen high e.mis:gioo rates would occur; the complex mi:<rure
of materials llctually pbced in the 11lldfill and the disposil procei.s nece,sarily traps PCB vapors
that take time to find triM-port pathw2ys; the 1.atge amount of water present ii:l the materials
buried 1J'\ tbt landfill from heavy rainfalls would slowly sink and accumulate to the bottom of the
bndfill -nd the internal cb)ing process (for some of the buried waste) would more easily allow
PCB vap,Jrs to escape over time.
ustly, EPA is arguing that its computer modeling (the details of~ch were not given in
the published paper but m~y be in the contractor report) sbowcd that calculated {not measurtd)
le..-el.s of PCBs at some distuoe away from the landfill would be 11far below the detection
c1pabilicy of the samp~r employed." But such dispersion modeling is based on mmy technical
assu.mptions about the source of emissions and many cfuiu.tic conditions that cannot be assessed
at this p..)int. Impli~itly:. in rt& origina] published paper, EPA was arguing that the theoretical
moddmg .3howed dau below the PCB levels actually 1tH1UMHd 2t distan.ces •"'-'IY from the
landfill. But EPA did n.1Jt actually say in its published paper tht the modeled data werr below
background ~el&. Mi,r~ importantly, th~ eore basis for EPA', modelin.g were only three
measurements ,,(the gas flov. from the main vent and 5'.ltne wu.tated data for what was measured
in tie .n'.Wtl veot. leading to what EPA descnbed u an •"erage gaseous PCB emission rite. Toe
prnbJem, of c-our~. is that there were only a small number of meuuremeau dwin.g a winter
F CB LJOF'.K I NG GROUP Fax:[phone] Dec 10 '95 16 :18 F·. 08
period. reducing the reliability of the spe=c Sgure used by EPA m its modeling and undcmrining
ULY objective confidence in the re$UltS of the modeling . In tzuly objective and fair work, field
me~--urements o{ air levels of a toxic substance ate used to evaluate whether the modeling is
accurate and reliable. But in this case, there was not even an explicit analysis hy EPA ofiu o\\n
positive measu..~ts and it, model.~g Tesults.
buo.nclu,ion. tht mu>Qt&auc deficiency of the EPA work w11 da•t they oa:cr
d.u!d~d....tbu then :w11 3yfflricot uuo.n. to perform much more complctt tcstio& for PCB
emiuions from the llmcR Coupty IJodfill, EPA. of course, bed I conflict of intcrc,t,
hC:"-ausc it had approved Jhe»'.m:ea County PCB h1ndfllJ ,and the objective of the 1983
,u,dy ~ccm., to b,ue l)_ten to prove that IP~tt.omcnt u.n.ttioncd PCB 1,ndfills were "safe"
i'ciatiye to utt.(D.atr__oUtd toxic wastt 1itll with PCB1 tb1t regulttd dflDWJ, Sinc;e the
\ta..(T..e..tt County PCB J.andOP was ibrr cleanup for tbc ati1iP1• PCB road dnmniar aimttiOJ)
tha.t F..PA.11111ravcd and fundro.1t clyrb: w11 not in EPA'a igtereat co re.port &i&nificaot
.aruS potcntiaJll.'. daniarous 1,vth of PCB 1\r tmissiana from iu controQcd landfill. Alto,
£.PA '., i:nibliah~d Pll>tt, 3ubm,ittcrl for puhljutiop ;a Pcctmbtr 1984.Jbout two Ytars after.
the Jaodfill w11 condru~e:d. described the Warren Coonty landfill has h1vi111 the
pcrforawd Pill& sxu1m that we now );,.now was 1u>t installed. EPA 's 4cu;tigti@ of lhc &anann .u.JiS,ato-of•the--art was consisteot with what EPA wsnte.d. to portral!, namdv a
'1lD.trolle-d landfill that b,-s.~on would show how bad the uucooa:olled Indiana PO
.s.itu~t. But EPA •s ducriptton "'" inac;curacc and. 1>crbu11, iatcutiaa•lll' so if, a, the
wtc nvw dairu, tPAJu,:tf appr:o...,ed a £b111c jp the daign of the Warran Couon: landfill
«Jut wo1t1d have had to ~cur in 1981 and that allowed the state to om.it v1in1 the
pctfot.ated Jeacbate collection oioe system at tht bottom or tile landfill.
5
f1
I
I
I
r
DEHNR I ENVIR. EPI. TEL:[phone]
State of North Carolina Department of Environment,
Health and Natural Resources
Division of Epidemiology
James B. Hunt, Jr., Governor
Jonathon B. Howes, Secretary
Michael Moser, M.D., M.P.H.
MEMORANDUM
December 9, 1996
Dec 09,96 13:27 No.002 P.02
TO: Bill Meyer, Director ~
Division of Waste Management •. _ i ,,._-:.,------·· ·
THROUGH: Stanley Music, M.D., DTPH (Lond.), Chief ,_ .. :·t> -·
Occupational and Environmental Epidemiology Section
FROM: Luanne K. Williams, Pharm.D., Toxicologist P(~W-
Medical Evaluation and Risk Assessment Branch
Occupational and Environmental Epidemiology Section
At your request, I have provided a risk assessment following my review of the air
sampling results of the Warren County PCB Landfill reported in the 1983 USEPA study
"Measurement of Fugitive Atmospheric Emissions cf Polychlorinated Biphenyls from Hazardous
Waste Landfills11 and the December 3 letter from Dr. Robert G. Lewis, a co-author of the study.
RISK ASSESSMENT
1. The ambient air concentrations reported of 11 , 12, 50, and 71 ng/m3 most likely exceed
the actual PCB concentrations present at the site for the following reasons:
(a) the method used is not specific for PCBs but detects all chlorinated compounds
(b) ambient air concentrations were reported to be higher at 98 meters downwind
(50 and 71 ng/m:i) than beside the main vent (11 ng/m3) and
(c) aroclor 1260 was the only analyte identified in ambient air even though aroclor
1242 was reported at much higher concentrations in the main vent.
2. It is my opinion that the ambient air concentrations reported are worst-case estimate of
the concentrations that .!J1sY be present at the site. Therefore, a worst-case risk
estimate is provided based upon the concentrations reported at the following locations:
Locations
beside main vent
nearby house
fence line, downwind
P.O. Box 2/687, Raleigh.
{cone. detected)
11 ng/m3
12 ng/m3
50 and 71 ng/m3
f;x~ess Cancer Sisk
1x10·5
1x10·6
5 to 7x10-G
An Equal Opportunity Atfirrnative Action Empie
50% recyclod/10% post-consurnc:r paper
t
DEHNR I ENVIR. EPI. TEL:[phone] Dec 09,96 13:27 No .OQ? F' r17 -.... .._.,
Bill Meyer
December 9, 1996
Page Two
If a million people were exposed to the concentrations reported at these three locations
over a lifetime, then one to seven of those individuals could possibly develop cancer as
a result of their exposure to the PCB concentrations reported. This cancer risk is a
worst-case estimate and is in addition to the existing cancer risk rate of 333,333 out of a
million expected cancer cases in a lifetime.
The risk associated with exposure to the concentrations reported at the fenceline does
exceed the acceptable target excess cancer risk of 1x10·8• However, the actual risk at
the site is most likely lower than 7x10-0 because the PCB concentrations present at the
site are most likely lower than reported. The PCB air concentration associated with a
1 x1 o-s excess cancer risk (target risk) is 10 ng/m3• The fenceline exceeds this level by
7.
3. The concentrations reported near the PCB landfill exceed typical background
concentrations in rural and urban areas. I have attached background concentrations
reported in the February 20 , 1996 ATSDR Tox profile for PCBs. The highest reported
background concentration was 20 ng/m3 in Chicago in 1989-1990.
RECO,MMENPATIONS
Contrary to Dr. Lewis' statement in his letter, it is my opinion that it is uncertain as to
whether or not PCBs are present at the site . Because PCBs were detected using a method
that Is not specific for PCBs and since the excess cancer risk estimated at the fenceline
exceeds the acceptable target excess cancer risk of 1 x 1 o-s , it is recommended to collect
additional vent and ambient air samples and analyze for aroclor 1242 and aroclor 1260 using a
more specific method. A carbon filter may also be used as suggested by BFA Environmental
Consultants to minimize PCB emissions from the landfill.
Please call me if you haye any questions at 715-6429.
LKW:lp
Attachments
cc: Dr. Stan Music
Mr. Bill Pate
Draft for Public Comment
I
DEHNR /EN VIR ,. EPI.
( t.. •. :· PCBs
TEL:[phone]~5~ ~' ._) Dec u~a a -:::), _i b
5. POTENTIAL FOR HUMAN EXPOSURE
13:2~ N ~ ~ o · U02 F'. 0 5
de~,rading organisms or, altcmaLively, by adding a gcm:tically engin eered strain th at combines the
activities of mixed cultures (Untennan et al. 1989). Since PCB dl~gradution is a C<.J-111e1abolic
process, the ~tddition of biphenyl or monochlorobiphenyls as growth substrates to supply the
nutritional requirements and to induce the catabo!ic pathway is required to sustain th e r..rowth of rh c
degrader population for biodegr,tdation of PCBs in soil (Ciuilbeault (:t al. 1994; Hickey et al. I 993 ).
In addition, the presence of surface active agents has hl~en shown to increase the bioavailabilit.y of
PCBs to the microorganisms. However, enriched cultures were unahlc to biodegrade J>CH
congcnl~rs containing five or higher chlorine substitution (Guilbeault et ,1I. 1994 ). lt ha s been
reported that the mono-, and di~chlorobenzoate, and possibly other higher chlorobenzoates fonrn~d
from aerobic degradation of PCRs act as inhibitors towards fu11he r degradation of higher chlorinated
PCBs (Guilbeault ct al. 1994 ). Therefore, the efficiency of PCB degradation is nnt only eontrnlled
by the enzyme substrate selectivity pattern , but also by the metabolite produl:tion pat!e.rn.
5.4 LEVELS MONITORED OR ESTIMATED IN THE ENVIRONMENT
,
. .
5.4.1 Air
The atmospheric concentrations of PCB s in various locations :m: given in Table 5-2. The range of
atmospheric concentrations of l'CBs in urban areas is J. -IO ng/m3 with n mean of 5 ng/m1
(Eisenn:ic;h et al. 1992). The atmospheric concentration~ of PCBs in two rural areas arc in th e ran ge
0.2-0.95 ng/m 3 with a mean of 0.6 ng/-and in two remote areas are in the rang~cl0:£)2.~0J 8
ng/111 3 with a mean of <0.1 ng/m1 (sec Table 5-2). Th e range a,-;j-r~;.;~···;~,·sphc:ri ~ PCB
-•-·----
concentrations in other locations arc as follows: 0.0 J--{).7 ng/nr and 0 . J 11g/m 1, respectively, in
marihc7coastal ,1rcas; and 0.2-4.0 ng7rn 3 and J .0 ng/m~respcctively. over the Grc'.at Lafos
-·----·-----·--.
(Eiscnl'elcnet ar.7992). With the availab·~-d~1~.~ it i~ difficult to cstabli~h rhc trend in atmos pli-~ric
Peft-{;nrfc"t>J1tralio1is-over the last two decades . following th e c~s;ation of PCB produ ction . Thi s is
ht:cause monitoring data indicating the levels of PCBs in air at the: same lo cation o ver this tim e
period are. still lacking (levels from one location cannor be compared with levels from an oth er
because of differing cmissiou sources), and the recent studies (Schreitrnucllcr and Dallschmite.r
1994) generally report rhc atmospheric con<:entration.s of th e con geners and not th e total PCB s or
Aroclors . On the basis of typical atm ospheric concentrations of PCBs in pre -1980 sa111ple :-.
(Eisenrcieh et nl. 1981) and the )eve.I.~ in more recent years (see Table 5-2 ), it can be u ·,n;;l ud ed th~:
PCB concentrations in air rnay have shown a slightly d ecreas ing lfl'11<I fro n1 th e pre --1980 l (i po!st-
·-DEHNR I ENVIR. EPI. TEL:[phone] Dec 09,96 I ?
I 1
1
1 :.
,:: ., ,;,
.. --
5. ron,NTIAL H)k HUMAN E);POSUnE
TABLE 5~2. Atmospheric Concentrations of Polychlorinated Biphenyls
·----···----~--·--·--·--·-·-----·· ------·---~---· .. -
Location Year
Concentration~
(ng/m.'.l) Reference --------~----·----···--·-----------------
Boston, MA 1978 7.1 Bidleman 1981
Columbia, SC 1978 4.4 Bidleman 1981
Columbia, SC 1985 2.3 Foreman and
Bidloman 1987
College Station, TX 1979--1980 0.29 Atlas and Giam
0.11--0.48 1987
Newport News, VA 1988 0.39 Knap and Binkley
(0.185-0. 794) 1991
Bloomington, IN 1986-1988 Summer: 1.74-3.84ti Hermanson and
Winter: 0.31--0.62 Hites 1989
Chicago, IL 1989-1990 13.5 Holsen et al. 1991
(7.55-20.26)
Adirondack, NY 1985 0.95 Knap and Binkley
(0.339-1 .359} 1991
Chesapeake Bay 1990-1991 0.21 Leister and Baker
(0.017--0.508) 1994
Lake Superior 1986 1.25 Baker and
Eisenreich 1990
Rural Ontario. Canada 1988-1989 0.2 Hoff et al. 1992
(0 .55--0.823)
Antarctica 1981-1982 (0.02-0.18) Tanabe et al. 1983
Arctic 1986-1987 0.02 Baker and
Eisenreich 1990
'Values are given as mean concentrations. The ranges are given in parentheses.
L>Values at three different sites.
. . . ~ Ml~ C1,(l {f)bM '{fr' '{11~ ~ cL~-w;uwm L. Meyer, Director r Sofia Waste Management Division To:Tu · ~\ '(~ N\Cc6fl__ Date: /2l~ / 9&; p,,,,.,_Tlfi5 l? A M@ /J.i6[J,/1 ?5Uf:-f" yµ b .W· J/J __ Draft a reply for my signature __ Take appropriate action __ For your infomzation __ See me about attached __ Approve __ Handle and report to me _ Note and return attached material to me Remarks: M1K6-6f!u®J>Pfl.,{;, LJ c:\(lll)K_~ ~e02-~,~ ~CB ~,1t,E::,~\\§10~-S Ot-\ V../_hRnf:,t\.l ~b~ fcb ~'~ . .£:I~\~L-. ""~ l' ~~" Aro~ -ro 1tie-W,//2/2BJ ~ . worzlz(~ 0-~~ -smlE:-~'fr: .. <;;cP, "'tfr Ni '.pl Dk tl@.UI-lcl.sh. /ZEfil lti /vs r/1-, 1t"JE;W1" (TH:SN\~ ~d> &\t. ~) ., ~~ 1X='-ti:s,3t1~t~'~criPr h<S'tl'il, tie" ,, c12~ ~6f'fl,1) t<JIL ~ -s/'rtEl'T q~ C'it~ ~ 4/Joii' \~ ~10'1'5 ~~Tu \1,je l.B.u1s. .J~ 1q/f, ~a~ ~q_.L~.J,.:n<;'s ~~ \2-\~\q~ ~ tv\.t ~~"1-~~ E.uALu.li'i-e-~ v\\-¼ ~ ~0,\Js ~~ ol'\. ~clflt:-~ ~\l.8 f\\:'.~ ~ ~Sl.llt.-": -'J-J ,tA-1 ,;fl6IJ,IJ IN~® \.o E:"-bu..ai.~ °"~~\i.v ~
..
Published 1n .tnvironmental ::;c1ence ana 1ecnno1ogy, vciuoer, 1;:,ou, i-'.IJ· :100·::1::11, uy u1t: n.u.11:::u\;:c111 yuc1.u1.\..c11. t.Ju1..:1.c:1.._y
Measurement of Fugitive Atmospheric Emissions of Polychlorinated Biphenyls
from Hazardous Waste Landfills
Robert G. Lewis• and Barry E. Martin
U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711
Donald L. Sgontz and James E. Howes, Jr.
Battelle Columbus Laboratories, Columbus, Ohio 43201
■ Four landfills known to contain large quantities of po-
lychlorinated biphenyls (PCBs) were monitored for at-
mospheric emissions: Three of these were uncontrolled
and contained large numbers of electrical capacitors, many
of which were scattered on the surface and leaking PCB
askarel fluids. The other is a state-of-the-art PCB waste
landfill designed to exceed the requirements of the Toxic
Substances Control Act of 1978 (ToSCA) for PCB disposal.
Both high-volume and low-volume air sampling equipment
were utilized at each landfill to monitor air levels of PCBs
on site, upwind and downwind. In addition, vent ports
were monitored at the controlled site. Simultaneous,
collocated sampling was performed for quality assurance
purposes and to obtain information on sampling perform-
ance and comparability. High atmospheric PCB concen-
trations were measured at the uncontrolled sites, while air
levels were at or near background at the ToSCA-designed
landfill. PCBs were detected at low levels in gas vents at
the latter site.
Introduction
The disposal of polychlorinated biphenyls (PCBs) is
strictly regulated under the Toxic Substance Control Act
(ToSCA). Specifically designed chemical waste landfills
with impermeable liners are required and are subject to
approval by the U.S. Environmental Protection Agency.
The only access to the external environment in the case
of ToSCA landfills is through open vents and ports re-
quired for leachate monitoring and for relief for gases
generated within the landfill from decaying organic matter.
Prior to 1978, however, many PCB-containing articles were
disposed of in municipal landfills or at uncontrolled
chemical waste disposal sites.
PCBs possess sufficiently high vapor pressures (10--4-10-5
kPa) to be emitted directly into the air surrounding haz-
ardous waste disposal sites through volatilization from
contaminated surfaces (J, 2). They also may be released
from controlled landfills through vents along with more
volatile gases (3). To test this premise, four PCB landfills
were monitored in this study. One was a ToSCA-designed
facility in North Carolina which was studied to determine
if PCBs were being emitted into the surrounding atmos-
pheric environment from gas vents and leachate access
ports. The other three consisted of two uncontrolled
private landfills and one municipal landfill in Indiana, each
of which contained large numbers of PCB-containing ca-
pacitors.
Materials and Methods
Air sampling was performed with low-volume (L V)
and/ or high-volume (HV) sampling systems previously
described (4-6). The components of the LV sampling
system consisted of a battery-operated, constant flow
sampling pump (Du Pont Model P-4000A) and a glass
cartridge containing a 22-mm diameter X 7.6-cm long
986 Environ. Sci. Technol., Vol. 19, No. 10, 1985
cylinder of polyurethane foam (PUF). The HV sampler
consisted of a conventional Hi-Vol shelter and pump
modified by addition of an inlet head to accommodate a
PUF sampling cartridge. The sampling head was com-
prised of an aluminum housing which held a 10-cm diam-
eter particulate filter (Pallflex 2500 QAST quartz) followed
by a glass sampling cartridge containing a 62 mm diameter
X 7.6 cm cylindrical PUF plug. The General Metal Works
(Village of Cleves, OH) Model PS-1 sampler is essentially
identical with the HV sampler used in these studies.
The Du Pont sampling pumps were calibrated with a
Du Pont calibrator pack (catalog no. 66-242-f-l) before and
after each sampling period. The HV samplers were cali-
brated once each week by means of a calibrated venturi
tube which was attached to the inlet. Flow audits were
conducted in the field prior to and at the termination of
monitoring activities by an independent team using two
laminar flow elements calibrated by the National Bureau
of Standards (NBS). Clean filters and PUF plugs were
used to simulate actual-'initial operating conditions.
The PUF plugs were precleaned by Soxhlet extraction
with acetone as previously described (4), vacuum dried,
and loaded into the appropriate glass sampling cartridges
under clean laboratory conditions. The cartridges were
wrapped in hexan,e-rinsed aluminum foil and stored in
carefully cleaned glass jars padded with clean PUF for
transport to and from the sampling sites. Exposed filters
from the HV samplers were completely wrapped in pre-
rinsed foil and placed in the jars with the PUF cartridge.
Disposable latex surgical gloves and prerinsed tongs were
used for handling the sampling cartridges when PUF plugs
were loaded and unloaded in the laboratory and for atta-
ching them to the sampling systems in the field.
Exposed PUF plugs and filters were Soxhlet extracted
together with 5% diethyl ether in hexane following the
procedure of Lewis et al. (4, 6), the extracts reduced to 10
or 1 mL in Kuderna-Danish concentrators according to
analytical needs, and the PCBs determined by electron
capture gas chromatography following EPA Method 608
(7). Identification and quantification of Aroclors 1242 and
1260 in the samples were performed by the technique
originally described by Webb and McCall (8). All solvents
were pesticide quality or analytical reagent grade. NBS
Standard Reference Material 1581 (Aroclors 1242 and 1260
in motor oil and transformer oil) were used for calibration
purposes. Recovery of Aroclors 1242 and 1260 from PUF
plugs fortified with SRM 1581 at three levels (0.06, 0.6,
and 6 µg/plug) averaged 93%. One laboratory blank and
one laboratory "spiked" PUF plug were analyzed with each
20 samples. In addition, ca. 10% of all samples analyzed
were field blanks which had been transported to and from
the monitoring sites.
Continuous measurements of wind speed, wind direc-
tion, ambient air temperature, and relative humidity were
obtained with Meteorology Research, Inc., portable
weather stations. Since the landfill sites were subject to
j
□ CLEAN EARTH/TOPSOIL m ARTIFICIAL LINER
~ LEACHATE COLLECTION SYSTEM
~ CLAYLINER WATER TABLE
GROUND
-...._/ SURFACE -----
Figure 1. Cross-sectional drawing of controlled PCB landfill in North Carolina.
the Comprehensive Environmental Response, Compensa-o ■H0USE A MAIN VENT
B SMALL VENT
E SMALL VENT tion and Liability Act ("Superfund"), all samples were
collected, handled, and transported under standard
chain-of-custody procedures.
Results and Discussion
Controlled Landfill. In 1978, an unprecedented spill
of PCBs along 387 km of roadway occurred in central
North Carolina. Some 40 000-120 000 L of transformer
fluid consisting of Aroclors 1260 and 1242 in chloro-
benzenes were illegally and surreptitiously dumped along
the shoulders of the roads. Over 4 years of litigation were
required before a disposal site for the 30 000 m3 of con-
taminated soil could be located in a sparsely populated
area of the state. Because of the great amount of public
concern over the safety of the disposal site, a state-of-
the-art landfill designed to exceed the requirements of
ToSCA was constructed in late 1982. The EPA-approved
(Superfund) site, located on a 4 ha of land in a rural area,
measures 75 m X 145 m and has a maximum depth of
about 7 m. The contaminated soil is encapsulated within
0.6-m (top) to 1.5-m (bottom and sides) thick layers of
highly impervious, compacted clay, augmented with 10-mil
(0.25-mm) and 30-mil (0.75-mm) plastic liners on the top
and bottom, respectively. The landfill is properly sloped
and equipped with sump pumps and a leachate collection
system for monitoring purposes. It is also provided with
a gas vent which protrudes 1.5 m into the landfill for relief
of methane and other gases generated by bacterial decay
of organic matter contained in the soil. A cross-sectioned
drawing of the landfill is shown in Figure 1.
Several months after closure of the landfill,_local resi-
dents voiced concerns that gases emanating from the vent
pipe and (as yet uncapped) leachate collection pipes may
have been introducing PCBs into the surrounding atmos-
phere. In response, a study was undertaken in Jan and
Feb 1983 to monitor these emissions and the ambient air
at the site.
Only the LV samplers were used in this study. The
vents and leachate access ports were sampled by placing
the PUF cartridge inlets into the pipes or ports and sealing
the openings with plastic bags to assure maintenance of
positive pressures. Sampling was performed for 8 h at
reduced flow rates of 1.2-1.4 L/min so as not to exceed
the volumetric flow rates of the vent pipes. The sampling
efficiency of the L V sampler employing PUF cartridges
had been previously shown to be essentially quantitative
for collection of Aroclor 1242 in natural gas, which is
97-98% methane (9). Soil temperatures within the landfill
were presumed to be nearly constant at 4-5 °C (thermic
temperature regime); therefore, emission rates were es-
sentially independent of ambient air temperatures.
1 1000m
C UPPER LEACHATE ACCESS PORT
0 LOWER LEACHATE ACCESS PORT
0 SAMPLERS
0--...... ---l -._ --°'-'-,---..L...--------,-•s_m--, ----o___ . 8
75m ---130m 36m 36m -0 315m 160m o, , o---o~•~o----+---o
__ ..--o---_:::o-i'm c~ -
_ol-....C:C--------------' LEACHATE
POND
WEATHER STATION N----Figure 2. Sampling arrangement for monitoring at the controlled landfill
when winds were from the north. Array was rotated with wind direction
so as to monitor at the perimeter of the landfill and at half the distance
for the main vent to the perimeter.
Ambient air sampling was performed with an array of
samplers operating at 3.8 L/min and located from 1 to 200
m upwind and downwind of the main vent as shown by
the example presented in Figure 2. Air samples were
collected at 1.2 m aboveground at each of the 13 locations
designated by open circles. Simultaneous samples were
also collected at 4.6 m above ground at the perimeter of
the mounded landfill (72 m downwind of the main vent
in the arrangement shown in Figure 2) so as to be on
line-of-sight with the opening of the main vent. The spatial
arrangement of the samplers was alternated as necessary
at the start of each sampling period to reflect average wind
direction. That is, downwind samplers were placed in a
fanned-out array at half the distance from the main vent
to the perimeter, at the perimeter, and at the tree line
(100-130 m downwind). Upwind samplers were placed
accordingly. One sample was taken at the same location
in the yard of the nearest house (1 km away) during each
sampling period. Sampling was performed from 0900 to
1700 Eastern Standard Time on 3 days and from 2100 to
0500 on the fourth day. Ambient air temperatures ranged
from -1 to +14 °C, wind speed from 0.04 to 6.6 m/s, and
relative humidity from 46 to 95%. No vent sampling was
performed on these days, and all vents were left open.
Analytical results from the gas samples collected from
five vents are presented in Table I. It should be noted
that the lower leachate removal pipe extends below the
landfill. The two small pipes were installed temporarily
to release gas bubbles in the upper plastic liner. PCB
concentrations in the gases (principally methane) ema-
nating from the main vent ranged from 105 to 141 µg/m3
measured as Aroclor 1242 and from 1.8 to 2.1 µg/m3
measured as Aroclor 1260. A typical gas chromatogram
of the PCB mixture found in the main gas vent is shown
Environ. Sci. Technol., Vol. 19, No. 10, 1985 987
Table I. Controlled PCB Landfill Vent Monitoring Results
PCB concentration, µg/m3
Aroclor 1242 Aroclor 1260
location range av range av
main vent (A) 105.7-141.5 120.2 1.8-2.1 2.0
upper leachate access 0.8-2.8 2.6 0.3--0.6 0.5
port (C)
lower leachate access ND--0.09 0.04 ND--0.08 0.05
port (D)
small vent (B) ND--0.07 0.05 <0.02-<0.3 <0.02
small vent (E) <0.02--0.67 0.24 <0.02-1.3 0.4
• Letters in parentheses refer to position identification in Figure
2. b Average of three to four measurements; 7-8-h samples.
1:1 AROCLOR 1242 ANO 1260
PCB IN GAS VENT
Figure 3. Gas chromatograms of standard mixture and sample from
main gas vent at controlled landfill.
in Figure 3 along with the 1:1 Aroclor 1242-Aroclor 1260
standard used for quantification. Collocated samples
collected in the main vent agreed within 6-9%. Concen-
trations found in the leachate collection ports were much
lower, ranging from nondetectable (ND) to 2.8 µ.g/m3
Aroclor 1242 and from ND to 0.5 µ.g /m3 Aroclor 1260.
Detection limits were 0.01-0.02 µ.g/m 3• Analyses of
transformer oil and soil samples prior to interment showed
a 4:1 ratio of Aroclor 1260 to Aroclor 1242. The greater
volatility of the latter mixture apparently accounted for
the relatively higher concentrations found.
Ambient air data are given in Table II. Only four of
the 39 ambient air samples analyzed contained detectable
quantities of PCB (MDL = 6 ng/m3). Three samples were
positive for Aroclor 1260 only (at 11, 12, 50, and 71 ng/m3)
and did not appear correlate with proximity to the vents.
The two maximum levels were found 98 m downwind of
the main vent.
Gas flow measurements made at three different times
showed flow rates of 4.8-6.4 (average 5.7) L/min from the
main vent. No flow was measurable from the leachate
ports or small vents. The average gaseous PCB emission
rate from the landfill, therefore, was estimated to be 12.1
ng/s. By use of these emission parameters and meteoro-
logical conditions that prevailed during the study, standard
dispersion models were applied to calculate downwind
PCB concentrations for comparison with field measure-
ments. Two EPA models (10, 11) were employed to cal-
culate estimates of maximum hourly concentrations under
a full spectrum of meteorological conditions and estimates
of the range of hourly ambient concentrations that would
occur at downwind distances of 50, 100, and 150 m under
the meteorological conditions that probably controlled
988 Environ. Sci. Technoi., Vol. 19, No. 10, 1985
Table II. Ambient Air Monitoring at Controlled Landfill
location
beside main vent
on site, downwind
on site, upwind
fence line, downwind
off site, downwind
off site, upwind
nearby house
air concentration, µg/m3
Aroclor 1242 Aroclor 1260
<0.006
<0.006
<0.006
<0.006
<0.006
<0.006
<0.006
<0.01--0.01"
<0.01
<0.01
<0.01--0.07b
<0.01
<0.01
<0.01--0.0lc
• One of six measurements above detection limit. b Two of 13
measurements above detection limit (0.05 and 0.07 µg /m3). ·'One
of six measurements above detection limit.
dispersion during the field monitoring. The predicted
maximum 1-h concentrations downwind of the vent under
prevailing and worst case conditions ranged from 4 X 10-6
ng/m3 at 14 m to 1 X 10-7 ng/m3 at 50 to 150 m. The 8-h
concentrations would be expected to be about 10-s ng/m3•
These values are far below the detection capability of the
sampler employed (or that of any known sampler).
Uncontrolled Landfills. Three disposal sites in the
vicinity of one city in Indiana have been the subject of
recent concern because they contain large numbers of
PCB-containining transformers which were dumped there
prior to 1972. One of the sites, designated here as site 1,
is about 8 ha in size and located on an abandoned farm.
Site 2 is a small, rural site of about 0.2 ha, surrounded by
mobile homes. Site 3 was a municipal landfill, some 3 ha
in area, located in a .suburban area. At each of these sites
capacitors were stre\vn across the ground surface (some-
times in mounds) arid were visibly leaking askarels con-
taining Aroclor 1242 into the soil and nearby streams. The
total quantity and depth of burial of the capacitors is
unknown.
Both L V and HV samplers were employed to monitor
PCB atmospheric emissions at the three uncontrolled
landfills. The air measurements at the sites were per-
formed at localized areas (hot spots) where leaking capa-
citors were evident and at other locations to determine
upwind (background) and downwind levels. Three dif-
ferent sampling approaches each using PUF cartridges for
collection of PCBs were used. The L V samplers were set
up at hot spots to sample during 8-h daytime periods at
3.8 L/min with intakes positioned 120 cm above ground.
The HV systems were set to sample at 226 L/min for
8-24-h periods at hot spots and at upwind and downwind
points. The intakes at the HV sampler were 180 cm above
ground. Where line power was unavailable, gasoline-
powered generators placed downwind of the HV samplers
were used. Arrays of five L V sampling systems placed
from 2 to 180 cm above ground were used to determine the
vertical concentration profiles at hot spots. Detection
limits were ca. 10 ng/m3 for the LV sampler and ca. 50
pg/m3 for the HV sampler.
Three or four days of monitoring was performed at each
site during June and July 1983. Results from single-level
(120 or 180 cm above ground) monitoring are presented
in Table III. The following summarizes these results:
At capacitor disposal site 1, airborne PCB concen-
trations measured at 120-180 cm above five hot
spots during the day (0900-1700 central daylight
savings time) ranged from 0.4 to 18 µ.g/m3• Levels
along the downwind perimeter of the site ranged
from 0.2 to 1.8 µ.g/m3• Upwind PCB concentrations
ranged from <0.05 to 0.09 µ.g/m3•
Ambient air PCB concentrations measured at
120-180 cm above two hot spots at site 2 ranged from
Table III. Range of PCB Concentrations (µg/m3) in Air at Uncontrolled Sites, June-July 1983
site 1
location daytime
hot spots 0.4-18.0
downwind 0.3--0.5
upwind <0.05--0.10
meteorology
temperature range, °C
wind velocity range, m/s
relative humidity range, %
Table IV. Vertical Profile Air Measurements at
Uncontrolled Landfills
distance
air concentration," µg/m3
above site la site lb
ground, cm October July July
2 271-520 577-1053 602-1108
30 27-33 56-120 111-157
60 8.6-18 30-58 4o-62
120 2.9-5.7 17-30 15-21
180 1.3-2.3 6.4-1.3 8.6-10
• Average daytime levels over a 4-day period.
24 h
8.3-13.0
0.60-1.3
0.08--0.09
19-42
0-2.3
62-83
site 3
July
367-955
53-159
28-69
16-33
6.8-21
0.6 to 19 µg/m3 during the day. Near residences
adjacent to the site, levels ranged from <0.04 to 0.2
µg/m3.
At the municipal landfill (site 3), daytime PCB air
levels measured at 120-180 cm above three hot spots
ranged from to 193 µg/m3• Upwind concentrations
were fairly constant at ca. 0.05 µg/m3, and levels
measured downwind of the landfill ranged from 0.3
to 0.8 µg/m3. There were a number of residences
around the perimeter of this site.
Generally, the airborne PCB levels measured at hot spots
correlated with the quantity of exposed capacitors.
Vertical emission profiles at hot spots are given in Table
IV. Air concentrations decreased with distance above
contaminated surfaces. Levels at 2 cm above ground level
were from 40 to 100 times higher than the levels at an
elevation of 180 cm, while those at 120 cm were twice as
high. PCB emission rates during July 1983 were probably
maximized by the unseasonbly warm (19-42 °C), dry
weather which prevailed throughout the study period.
Sampling data from site 1 during Oct 1982, when tem-
peratures ranged from 13 to 30 °C, showed PCB levels
about half of those obtained during the summer.
Typical gas chromatograms of downwind and upwind
samples from site 1 are shown in Figure 4. Chromato-
grams of hot spot air samples taken at several heights are
presented in Figure 5.
Both the LV and HV samplers (including one com-
mercial version of the latter) performed well throughout
the study. Sampler pumps were calibrated before and after
each sampling period. Flow calibrations ranged from + 1.9
to-7.9% for the LV samplers and from +8 to-7% for the
HV samplers. Pre-and postflow readings agreed within
±5% in over 90% of the cases. Independent flow audits
indicated average flow accuracies of -2.8 % for 20 L V
pumps and +5.2% for seven HV samplers. Field blanks
analyzed with ambient air samples collected at each landfill
site showed no PCB (measured at Aroclor 1242) above the
minimum detectable level of 0.02 µg per plug. The blank
sampling cartridges were carried through all field handling
operations except attachment to the sampling devices.
During the field study, pairs of LV and HV samplers
were operated for the same time period at the same sam-
site 2 site 3
daytime
0.6-33.8
<0.04--0.07
24 h daytime
6.3-193
0.08--0.20
0.08--0.20 <0.04--0.05
19-38
·0.1-2.7
37-83
AROCLOR 1242
DOWNWIND
AIR SAMPLE
STANDARD
24 h
21.5-77.4
0.3--0.8
0.08--0.09
22-42
0-2.2
24-70
UPWIND
AIR SAMPLE
SITE 1
Figure 4. Gas chromatograms of standard and ambient air samples
at uncontrolled landfill.
AROCLOR 1242
STANDARD
AIR SAMPLE AT
2cm ( 1 :40 DILUTION I
AIR SAMPLE AT
60cm
AIR SAMPLE AT
120 cm
Figure 5. Gas chromatograms of standard and air samples at several
heights above hot spots at uncontrolled landfill.
pling location to estimate the reproducibility of the mea-
surement methods. Measurements were also made with
collocated L V and HV samplers to compare results ob-
tained by the two methods. As can be seen from the data
in Table V, good agreement between pairs was obtained.
The average difference calculated from all the paired L V
sampler measurements was 7.2% while that for the HV
Environ. Sci. Technol., Vol. 19, No. 10, 1985 989
Table V. Collocated Sampler Comparisons
paired LV samplers,• µ,g/m3 paired HV samplers,b µ,g/m3 paired LV and HV samplers,b.c µ,g/m3
% difference % difference % difference
sampler A sampler B in pair sampler A sampler B in pairs sampler A sampler B in pairs
9.4 11 15.7 12 14 15 8.7 7.1 +20.3
8.8 8.6 2.3 11 12 8 7.0 6.5 +7.4
11 6.8 47.2 37 61 49 8.5 11.6 -30.8
11 12 8.7 98 89 9 5.2 9.9 -62.3
5.1 5.3 3.8 30 47 44 11.0 12.3 -11.2
7.9 9.1 14.1 20 23 14 8.5 13.5 -45.5
18 19 5.4 45 45 0 18.5 28.0 -40.9
77 72 6.7 11.0 21.0 -62.5
6.2 6.0 3.3 40.0 49.0 -20.2
85 89 4.6
11.2d 20d -27.3d
• Located 30 cm apart. b Located 1 m apart. 'Data corrected to account for difference in height of intake above ground. d Average.
Table VI. Comparison of Active and Passive Sampling Data
air concentration, µ,g/m3
distance above site la
ground, cm active passive active
2 1060 980 670
30 120 53 90
60 37
120 19 14 17
180 5.1/5.2
• Average.
samplers was 20%. Since the inlets of the HV and LV
samplers were not located at the same heights above
ground, it was necessary to apply a correction factor in
order to achieve reasonably accurate comparisons between
values obtained from collocated pairs of HV and LV sam-
plers. For this purpose, corrected HV values (C180) were
calculated by
where C180 and C120 were the PCB air concentrations
measured by the collocated HV and L V samplers, re-
spectively, and V180 and V120 were concentration values
obtained from vertical profile measurements with the LV
samplers at 120 and 180 cm above ground, respectively.
After these adjustments, the L V /HV sampler comparison
averaged -27.3%, with the LV sampler generally giving
lower results.
A limited numer of experimental passive devices (12)
were also exposed at hot spots on site 1 on 2 of the 4 days
of sampling. The passive sampling devices (PSDs) were
3.8 cm diameter x 1.3 cm stainless-steel cylinders con-
taining a series of diffusion screens and plates on each end.
Tenax GC (0.4 g) was used to collect PCB gases diffusing
into the devices. The sampling rate for PCBs was esti-
mated at 30 cm3 /min on the basis of trichlorobiphenyl.
The devices were transported to and from the exposure
sites in sealed cans, which were placed in a larger can
containing activated charcoal. Cyclohexane was used for
extraction of the PCBs from the PSDs. Collocated expo-
sures with the L V samplers showed reasonably good
agreement, as can be seen from the data presented in Table
VI. The detection limit for the PSDs was only ca. 5 µ.g/m3
for 8-h exposures, however. Efforts are currently under
way to improve sensitivity by means of supercritical fluid
extraction and concentration.
Temporary remedial actions were taken during
1983-1984 at sites 1 and 2 to remove all exposed capacitors
990 Environ. Sci. Technol., Vol. 19, No. 10, 1985
site lb average
passive active passive % difference P / A
340 865 660 -24
125 105 89 -16
55/100 37 77 +48/108
18 14 -22
5/16 5 10 +100
+16.8°
Table VII. Range of PCB Concentrations• (µ,g/m3) in Air at
Uncontrolled Sites after Temporary Cleanup, Aug 1984
hot spots
downwind
upwind
location
meteorology
site 1
3.1-4.6
0.4-1.4
0.2-0.3
temperature range, °C 14-32
wind velocity range, m/s 0-1.3
relative humidity range, % 48-92
• 24-h samples.
site 2
2.7-3.1
0.H).2
0.1-0.1
Table VIII. Vertical Profile Air Measurements at
Uncontrolled Landfills after Temporary Cleanup, April
1984
distance above
ground, cm
2
30
60
120
180
air concentration,•
µ,g /ma
site 1
2.3-3.2
1.1-1.8
0.9-1.4
0.7-1.4
0.4-0.6
site lb
11.5-21.3
4.1-5.8
1.7-3.1
1.7-3.1
1.5-2.5
• Average daytime levels over a 4-day period.
and obviously contaminated surface soil. Following this
cleanup, limited additional air monitoring was performed
during a 4-day period in Aug 1984. The results of HV
sampling at both sites are presented in Table VII. Vertical
profile measurements (using the LV samplers) at two
previously monitored hot spots on site 1 are given in Table
VIII. Significant reductions (by an order of magnitude)
of PCB air levels at the hot spots were noted postcleanup.
However, downwind levels at both sites 1 and 2 appeared
unchanged, suggesting that the landfill proper is still
contributing PCBs to the surrounding atmosphere.
Conclus(ons
The results of these studies demonstrate that fugitive
emissions of PCBs into the atmosphere can occur at un-
controlled landfills. At the three sites, PCB air levels
measured at hot spots on the landfills greatly exceeded
ambient background levels, thus indicating that PCBs
from the leaking capacitors were being emitted into the
air. Concentrations that exceeded background levels were
also observed at sampling locations downwind of the
landfills, even after removal of exposed capacitors and
obviously contaminated surface soil. By contrast, air
emissions of PCB from a well-designed chemical waste
landfill were found to be negligible.
All PCB sampling systems were found to perform well.
The LV samplers offered an advantage over the HV sam-
plers when electrical power was not available (as was the
case at most of the sites monitored). However, limited
battery life would not permit 24-h sampling with the LV
pumps. The experimental passive sampler, which can
readily operate unattended for 24 h, shows much promise
if its sensitivity can be increased by 100-fold through im-
proved extraction and analysis methods.
Acknowledgments
We than Donald E. Johnson of Southwest Research
Institute, San Antonio, TX, for valuable laboratory support
in preparation and analysis of PUF cartridges, Jack C.
Suggs of the U.S. Environmental Protection Agency, Re-
search Triangle Park, NC, for modeling assistance, William
F. Barnard and Jack A. Bowen of the U.S. Environmental
Protection Agency, Research Triangle Park, NC, for per-
forming field audits, James Gray of EPA Region IV,
Athens, GA, for field support at the controlled landfill, and
Ralph Riggin of Battelle for analysis of the passive sam-
pling devices.
Registry No. Aroclor 1242, 53469-21-9; Aroclor 1260, 11096-
82-5.
Literature Cited
(1) MacLeod, K. E.; Lewis, R. G. In "Sampling and Analysis
of Toxic Organics in the Atmosphereft; American Society
for Testing and Materials: Philadelphia, PA, 1980; Pub-
lication STP721, pp 56-69.
(2) Weaver, G. Environ. Sci. Technol. 1984, 18, 22A-27A.
(3) Murphy, T. J.; Formanski, L. J.; Brownawell, B.; Meyer,
J. A. 184th National Meeting of the American Chemical
Society, Kansas City, MO, Sept 1982; American Chemical
Society: Washington, DC, 1982; ENVR 70.
(4) Lewis, R. G.; Brown, A. R.; Jackson, M. D. Anal. Chem.
1977, 49, 1668-1672.
(5) Lewis, R. G.; Jackson, M. D. Anal. Chem. 1982, 54, 592-594.
(6) Lewis, R. G.; MacLeod, K. E. Anal. Chem. 1982, 54,
310-315.
(7) Fed. Regist. 1979, 44, 69501-69509.
(8) Webb, R. G.; McCall, A. C. J. Chromatogr. Sci. 1973, 11,
366-373.
(9) Jackson, M. D.; Hodgson, D. W.; MacLeod, K. E.; Lewis,
R. G. Bull. Environ. Contam. Toxicol. 1981, 27, 226-229.
(10) U.S. Environmental Protection Agency "NTIS Tape of
User's Network for Applied Modeling of Air Pollution
(UNAMAP)ft. EPA, 1980, Version 4, NTIS No. PB81-
164600.
(11) Turner, D. B.; Busse, A. D. "User's Guides to the Interactive
Versions of Three Point Source Dispersion Programs:
PTMAX, PTDIX, and PTMTPft 1973, U.S. Environmental
Protection Agency Report EPA/DF /OOH (NTIS No.
PB81-164667).
(12) Lewis, R. G.; Mulik, J. D.; Coutant, R. W.; Wooten, G. W.;
McMillin, C. R. An~l. Chem. 1985, 57, 214-219.
Received for review December 19, 1984. Accepted April 5, 1985.
Although the research described in this article was funded wholly
or in part by the U.S. Environmental Protection Agency through
Contract 68-02-3745, it has not necessarily reflect the views of
the Agency, and no official endorsement should be inferred.
Mention of trade names for commercial products does not con-
stitute endorsement or recommendation for use.
Environ. Sci. Technol., Vol. 19, No. 10, 1985 991
\
-State of North Carolina
Department of Environment,
Health and Natural Resources
Division of Solid Waste Management
James B. Hunt, Jr., Governor
Jonathan B. Howes, Secretary
William L. Meyer, Director
AVA
DEHNR
December 3, 1996
MEMORANDUM
TO: Patrick Barnes, Science Advisor
Joint Warren County/State PCB Landfill Working Group
FROM: DWM Staff for the Joint Warren County/State PCB
Landfill Working Group
SUBJECT: BF A 12/2/96 Memorandum Air Emissions of PCB
and Associated Health Risk
Please provide scientific and epidemiological substantiation for BF A's statements that, with
respect to the Warren County PCB Landfill, "so-called landfill represents potentially significant
health risks" and the State's "apparent disregard for the safety of the citizens of Afton". Please
also provide any data on air quality modeling or other technical basis for the statement that the
"PCBs which have been deposited by air emissions may have accumulated in these surface
drainage features and thus may still present a threat to the environment." It would also be helpful
to the staff if BF A submitted the qualifications, expertise and experience of personnel providing
the statements and response to our request.
P.O . Box 27687,
Raleigh, North Carolina 27611-7687
Voice 919-733-4996
FAX [phone]
An Equal Opportunity Affirmative Action Employer
50% recycled/10% post-consumer paper
12:03 /96 13:01
ft
~
DATE:
TO:
'6'919 5-11 352i NERL/AMRD : )IB
U.S. ENVIRONMENTAL PROTECTION AGENCY
NATIONAL EXPOSURE RESEARCH LABORATORY
AIR l'vfEASUREMENTS RESEARCH DMSION (MAIL DROP 44)
RESEARCH TRIANGLE PARK, NORIB CAROLINA 27711
TELEPHONE 919-S41-306S FACSIMILE 919-541-3527
E-MAIL lewis. [e-mail]. epa. gov
FAX l\1ESSAGE
3 December 1996 NUMBER OF PAGES: 3
(including cover)
Bill Meyer Fax No.: 715-3605
Telephone:
FROM: Dr. Robert G. Lewis
USEPA (MD-44)
Research Triangle Park, NC 27711
SUBJECT: Hirschorn Report on Warren County PCB Landfill
MESSAGE:
[4]001:003
I was out until today and have just reviewed the above-referenced report that you faxed to
me yesterday.
The 1983 EPA paper (on which I was lead author) that Dr. Hirschom attacked in his
report was published in Environmental Science & Technology, a peer-reviewed journal widely
recognized as the leading journal in the world on environmental monitoring. Prior to publication,
it underwent rigorous technical review within the EPA and by two or more leading experts
outside the Agency. Therefore, our statements that PCB emissions from the landfill were
negligible at the time of the study were supported by the highest level of technical review.
It is clear from the data presented in Table II of our paper that no significant PCB
concentrations could be measured in the air surrounding the landfill, even at one meter distance
from the main vent pipe. As the paper states, "only four of39 ambient air samples analyzed
contained detectable quantities of PCB.'' These were positive only for Aroclor 1260 (the least
volatile of the two .Aroclors monitored) and "did not appear to correlate with proximity to the
vents.)' It would be clear to any analytical chemist or anyone with environmental monitoring
expertise that these were likely false-positive results. The levels detected were 0.01 to 0.07
µg/m3, at or very near the method detection limit (0.01 - 0.02 µg/m3). Had they been "real," the
more volatile Aroclor 1242 should have been found at much higher concentrations, as it was in
main and upper leachate vent pipes (see Table I). The lack of correlation of the measured ambient
air levels with proximity to the vent pipes, taken with the fact that Aroclor 1242 was the only
analyte identified in the air even though Aroclor 1260 was at much higher concentrations in the
vent gas, conclusively demonstrates that the vent pipes were not the source of the PCBs detected
in air, l!ven if the measured air values were real. It should also be noted that PCBs were
12/()3 /96 13:01 '5'919 5-H 352i NERL/ AMRD : )IB 14)002 :003
i
ubiquitous in the ambient air in the United States, typically at 0.005 to 0.01 µg/m3, at the time of
the Warren county air monitoring effort.
Our air monitoring results were also consistent with both the air emissions models applied
to the vent measurements. As we stated in the paper, the models predicted that no measurable air
concentrations of PCBs would result at 14 meters from the main vent and beyond.
In his analysis of the data presented in our paper, Dr. Hirshom refers to measurements of
"A.rochlors" (sic) obtained "at" the main vent (120.2 µg/m3 for Aroclor 1242). The vent pipe
measurements shown in Table I were made within the vent pipes, not in the air at or near the
vents. Hence, they cannot be compared with the ambient air measurements made at the
uncontrolled sites in Indiana and reported in our paper. There were other major differences
between the Warren County and Indiana sites. The Warren County site was brand new and
should have been emitting at a maximum. rate. The Indiana disposal sites were old (before 1972,
as stated in our paper). The PCB-contaminated soil in Warren County was underground, in the
thennic temperature regime, and should not have been affected by ambient air temperature (i.e.,
emission rates should have been the same whether it was summer or winter at the time of
monitoring). In Indiana, PCB-contaminated surface soil was abundant; thus, emissions were
greate: in summer when the sun heated the soil. As we said in our paper, the PCBs measured in
the vent pipes in Warren County were be transported by methane from decaying organic matter
(primarily grass excavated along with the roadside soil) and should have declined greatly once
that matter had decayed. Therefore, Dr. Hirshorn's statement on p. 4 cifhis report that the "high
levels found at the Warren County Landfill were quite comparable, and perhaps even greater than
the levels found above the other three uncontrolled landfills" is absolutely without merit.
As lead author of the ES& T paper, the interpretations of our findings were principally
mine. However, the report provided to EPA by Battelle Memorial Institute, who did the air
monitoring for the Agency under contract, reached the essentially the same conclusions ( see
attachment).
Anachment
. )
12/ 03,: 96 13 : 02 ~•9u1 5-ll :.i5:!7 1\ t.KL : .-\.}lKll : ~ltl
SECTION 2
CONCLUSIONS
lfiJ l_J I.I,.): l_J l_J ;_)
The conclusions drawn for results of this study are sul'Tlllarized below.
(1) The principal source of emissions from the landfill originate from
the main vent pipe. The average PCB concentrations of Aroclor 1242
and Aroclor 1260 measured in the main vent emissions during the
study were 123,000 ng/scm {""""100 ppb) and 2,000 ng/scm (~2 ppb),
respectively. These concentrations are substantially lower than
the current occupational standards for workplace atmospheres which
range from 0.4 to 0.8 ppm.
(2) Ambient air PCB levels on and surrounding the landfill site (even
as close as one meter from the main vent) were found to be at or
below minimum detection limits (6 to 10 ng/m3) for the sampling
method. PCB levels generally present in \he atmosphere throughout
the U.S. are in the range of 5 to 10 ng/m.
(3} Mathematical modeling predicts that ambient air PCB concentrations
on and in the vicinity of landfill resulting from the main vent
pipe emissions may be approximately 106 to 109 times lower than the
detectio~ limits for sampling method used in this study. PCB
contributions to the ambient air from the landfill based on the
model predictions are insignificant when compared to general
ambient air PCB levels in the U.S.
(4) It is anticipated that the low PCB emission rate from the landfill
will be reduced still further as decay of organic matter producing
the methane and other gases emanating from the vents subsides. The
reduction of hydraulic pressure by removal of water from the site
should also reduce ·emission rates substantially.
2
BF A Environmental Consultants
~;ttltz.._:.s --.. --------~ Barnes_, Ferland and Associates, Inc.
MEMORANDUM
TO: PCB Landfill Working Group
FROM: Patrick Barnes, Science Advisor
Joel 0 . Kimrey, P.G., Senior Hydrogeologist
DATE: December 2, 1996
SUBJECT: Air Emissions of PCB and Associated Health Risks
Bf A #95•017
We have performed a cursory review of the U.S . EPA research repon on "fugitive Atmosphere
Emissions of PCB• s from Hazardous Waste Landfills'', &! well as the review of that repon
prepared by Joel Hirschhorn, and, in general, it appears to me that the investigators made up their
minds that the Warren County PCB Landfill was going to be the control site regardless of the
testing results, They failed to draw the most important conclusion of their study, which is that
even so-called controlled landfill reprcsenu potentially significant health risks .
It is difficult to believe tha.t the facility was not designed to include gas filters at the main vent
opening . I believe that the community should demand an immediate explanation from the State of
this apparent disregard for the nfi:ty of the dtizens of Afton. Moreover, I recommer;d that
community leaders demand, in no uncertain terms, that:
1. The main vent or any uncapped opening to the landfill be fitted with an activated carbcin
adsorption type filter within 72 hours. The filters should allow for influent and e:flluent
sampling .
2. In addition to the health related sampling recommended by Joel Hirschhorn, additional
sediment sampling should be performed by the El> A or the State which include$ deposits at
the mouth of each major surface .drainage feature within 1/2 mile of the site. The PCB ' s
which have been depo,ited by air emissions may have accumulated in these surface drainas1:
fearures and thus may still present a threat to the environment.
The Holli1tt!'lr Building• 3535 Lawton Road· Suite 111 • Orlando. Florida 32803
Office(407)896-8608• Fax[phone]
DJ "d 10:6 8t 61-lSl-616 :XEJ
PCB LANDFILL ENVIRONMENTAL SECURITY
SUMMARY OF FINDINGS
By: Patrick A, Barne, -Science Advisor
Joel 0. Kimrey, P.G. -Senior Hydro1eologiat
Bad Site for Landfill
• The facility siting investigation failed to approprjately consider the criticaJ nature of th~
geolo@ical setting in locating the land~U. No in-depth geological work was perfonued to
determine actual subsurface flow characteristics
Bad Enginetrilfg Co1ttrols
• The system to remove leachate failed to properly consider the type of materials dep05ited and
subsequently does not function.
• Improper starmwa.ter mamtiernent during construction has allowed a significant amount of
waler to enter the landfill.
• Pressure frotn the water in the landfill has resulted in leakage through thb bottom hnr:r
• The pootly dc~igned/installed top liner is also allowing additional water to enter the landfill
• Water entering and leaving the landfill represents a real threat to sroundwater and surf!ce
water supplies of the area .
• Significant quantities of PCB's have apparently discharged into the air through the rnAin
landfill vent, and immediate 11.ctian should be taken to install a properly sized carbon
absorption filter .
Improper Man;tori.ng
• The State has failed to maintain compliance with the operating/monjtoring requirements of the
landfill.
• The existing monitoring wells are poorly designed and positioned, and will not properly detect
possible relea:se of contaminants from the landfill . The process of establishing !ldditional
locations is ut'lderwa.y.
• The existing surface water and sediment sampling locations are abo poorly located an·d· will
not intercept potential releases at the earliest stage, Here to, the est,bli~hment of additional
locations i! underway
The Hollister Building• 35J5 Law,on Rof!ld • Sulle 111 • Orlando, Florida 32803
Otfiee (407) ege-aeoa • Fa:x (-407) 896-1822
£0 'd T0:6
Recomm~ndations
• Redesign the environmentll) monitoring network.
• Move immediately towards complete detoxification of the landfill contents.
• Install a carbon adsorption filter on the main landfill vent
More deta.iled dis1;ussion of these items Arc available through the PCB Landfill Office m
Warrenton -[phone] .
l )-1S11.m.~
The Hollister Building• 3535 Leiwton Road• Suite 111 • Orlando. Florida 32803
Office(407)89e-se08• Fax(407)896·1822
t'O "d 81761-lSZ-616:XEj
pea AIR EMISSION§ ANP t!§ALTH RISKS
fBQM THE WAAAl!N QQYNYY
PCII b4 NDFILL
Joel S. Hirschhorn, Scimce Advisor
Warren County PCB Landfill Working Group
No\'ember 27, 1996
Data obtained from a 1983 EPA sni.dy showed conclusively that uncontrolled .releases of
PCBs into the air were occurring. Neither EPA or the state analyz:d the data properly, and EPA
madv incurrect statements indicating there was no problem, even though no annly&is iru.pported
the statements. In filct, the levels of PCBs fotmd by EPA .in the air ne3I the landfill and in the yard
of a residential house more than a half mile from the landfill were .several t.imes gre.ner than the
level of health s.ignific:mce found in EPA 's ov.n risk assessments. The PCB levels found io the
winter of J 983 were significantly above the one in one million excess cancer death risk based
oon.centration presented in EPA databsses. PCB emission levels in warm~ periods and in later
times during the past 14 years smce the PCB wastes were buried in the Lui~ could luve been
significantly higher. Exposure over long times to relatively low levels of PCB~ could also cause
non-can~er health effects, especially in children. Tue PCBs rele:tsed into the air could also result
in PCBs bcing deposited on nearby lands and, therefore, contaminate crop~: local vegetable
gardens, lild dairy and meat products ftom cattle grazing on local lands, leading to e~osure
routes other than inhalation.
Arr analysis of r/ae on{y :sta~ documents r-efe"ing to the J 983 study by EPA a,id lht!
on(~ informati()fl gi),)t11 to tht public has sh,:,wn tnat tht stat~ intentionally misrtpreunted tht
findiltgs of the 1983 tests/~ PCJJ air r~fca.ses from dr~ land.fill For a.ampl,; the highest
levels of PCBs fou·nd at the landfiU's main vent wer~ not reported by the atat~, and the state
indicated that no mt!a.mrabie amounts of PCBs /rad been/ ound in the air arownd tire site,
which wa.s nc,t the CIUe.. Th~ 5tat~ has persistently deceived the public about PCB air 7eJeas~
and, more imponandy, tht significant public health ri$ks r~ulti"g from th~m.
The results reported here .mpply still more support for the st~te takiDC seriously its
commitment to detoxify the Warren County PCB Landfill and to make .a commitment for
,up plying the nece:,,ary funding very quickly. Tb ere u now abuodaot proof th~t the
htndflll bas not been safe and secure, that the criticslly important bottom cay 2nd pl:astic
liner system lesked almoat immediately, th.at the lcschate tollectiou system never wor.ked
effectively, aud that PCB:, b11ve luked direcdy into th~ air and leachate into the
surrounding soils. A .number of n~ tests are reconu:nmd~ to assess health risks.
I
~.1r,....., I 1.1 .. u:'::r,"11 • I I C:W::.:: I -1 7 -17
C·h~ T -·' ,:.~ -•~ Ti~ • YI:• J -1n1To.JC1 c,t,. IT \r.i1-:ffl ,:i·-·, .-i
My revi¢w of the PCB fil~s provided by the state indiclted very little attention had been
given over the years si:oce the landfill construction was completed in 1983 to the potential for
human exposure to PCB air emissions from the landfill. Many people may have believed that
PCBs do not pose haurds because they are not volatile organic compounds (VOCs). However,
although PC:Bs are not classified as VOCs, they do pos.sess ~fficicntly high vapor pressures to
rclea$C potentially significant vapors into the air, if no pollution (;Qntrol techuology is used to
control svch emissions.
Tne one, most important document found in the state files was a copy of a professionil
published paper aititled "Messurement of Fugitive Atmo~llcric Emissions of Polychlorinated
Biphenyls from Hazardous Waste Landfills" (R.G. Lewis~ il, Envirownental Science and
Technolog), vol.19, no.10, 1Q85). The two lead authors were affiliated y,,i_th EPA's Research
Triangle Park facility. No document was found m the state filts that provided auy analysis of the
infomation and results in this published paper, especially with regard to the Warren County PCB
Landfill This was especially significant, becau~ the paper provided data O!l the Warren County
faciliry, probably the only data ever obtained 011 PCB air rel~ses from the landfill.
The published peiper noted in its introduction that "PCBs poi&ess suffi;iently high vapor
pressurc ... to be emitted directly into the air surrounding hazardous wa.Ste disposal site~ through
volatiliz.ation from cont.tr:ninated surfaces. They also m1y be rclea~d from controlled andfills
through ve:ats, along 'With more volatile g,scs."
The paper presented tile resultb of field testing at four PCB landfills. including the Warren
County facility, which "w~s studied to determine if PCBs were being emitted into the surrounding
atmospheric environment from gas vents and leachate access ports." The paper characterized the
Warre-..:i County Landfill as "ToSCA-designed'' and referred to it as i "statc-of-the~art widfill
designed to exceed the requirements of To SCA'' Th.esc statements referred to the federal Toxic
Substan~s Control Act, the key federal law covering pCBs. But the paper indicated that the
landfill had a perforated pipe leachate collection syste~ which ill fact was not installed, a.nd rud
not tecognize that the state had received waivers from. c:rtain landfill requirements, rahing doubts
about the paper's accuracy in dcscn"bing the lao.dtill as ~ate-of-tb~~art. The authors clearly
wanted to use the Warren C0unty Landfill as a control landfill against which to compare data
obt.ained for other less s.ophisticated PCB la.ndfills ("in the vicinity of one city in .Indiana") that
were desrnbed as "uncontrolled landfills" where PCB materials were dump~ or disposed and
were probably Superfund cleanup sites.
It should be ooted that the experimental .field work used soph.iiticated and accurate
methods for collecting and measuring PCB levels in air. Also, the field work was conducted in
January and February 1983, ~ch was some months after the PCB wastes were buried in the
landfil~ before the leadlate collection system pumps were first turned on, ~d before the landfill
CQnstruction was officially considered complete.
2
.ft is also important to note an imponant inconsistency in the statements made m the
published paper, because non-professional people who may read the paper could be mimfonned
or confused. There were two statcmc:o.ts about the significauce oftbc findings for the Warren
County facility. The paper's initial SlllilIWlIY said " ... air levels wt:re at or near background at the
T oSCA".designed landfill. PCB.s were detected at low levels in ga& vents at the latter (Warren
County] site." In the paper'$ section ·with conclusions, in referring to the Warren County facility,
it was said that "air emimons of PCB from a well-designed chemical waste landfill were found to
be negligiole." Tt is important to note these two ,tatement3 are not equivalent, because low·
i8 not the :,ame as ne:ti;iblc. Neglip"ble means inconscquenri.al or insignificant, but a level that
is measured may be low without being either iiicouiequential or insignificant ¥.'ith respect to some
important potential t!ffcct of PCB emissions. Moreover, drawing a conclusion that measured
PCB emissions were "negligi"ble" requires much more analysis than concluding they are low,
because one must answer the question: negligiole for what effect'? Ihe easier task is to conclude
that th~ emissions were low, because that could be logically based on a comparison with similar
data for the other landfills studied. However: to conclude th.it a level was negligible implies use
of som= uthc:r criterion or effect, which normally would be for health risks. 1n fact, the published
paper presented no analysis whatsoever in justifying any conclusion in reUltion to health risks.
The present r~ort provides such all analyiiis.
It should be noted that the paper paid particular attention to the history of the Warren
County Lmdfill and included reference to "the great amount ofpu.blic concern over the safety of
the disposal site." Moreover, the paper &aid: "Several months ifter closur~ of the landfill, local
residents voiced concems that gases emmating from the ve::i.t pipe and ( as :yet uncapped) leachate
collection pipes may have been introducing PCBs :into the surrounding atmosphere. In response,
a study \.Vas undertaken in Jan and Feb 1983 to monitor these emissions and the ambient air at the
site." Thu.s. it is fair to interpret the paper's statement about "negligible" air emissions to have
referred to health effects and public concerns about health eff'ects. It is also reason.able to believe
that the aJl the data from the EPA study and a. highly objective, professional evaluation of the data
would have been given to local residents. This Scicn.ce Advisor, therefore, also has closely
examined what information was comtITUrucated by the state to th~ public.
Analysi3 of Dau
Data were obtained for PCB emissions (for Arochlors 1242 and 1260) at th~ main vcnt1
the upper and lower leachate access poqs, and two small vents ( created by placing pipes in
surfac~ bubbles m. the thin plastic liner on top of the land.fill). Measurements were also made ''.in
the yard of the nearest house (J. km away).". This is a little over onehalfmile away from the
landfin.
The follo,..-ing are the most important observations about the data presented in the
published paper:
■ The highest levels measured were at the main vent. The highest level was 120.2
3
... , .... _,,,...., ,-. -
.1.....• •• • I
micrograms/cubic meter (ug/cm) for Arochlor 1242. In trying to compa!e this level with
the otha highest levc-ls found a·t the other three PCB land.£ills is ctifficuh, bec.a.u~ the g1udy
used different modes of measurement. A big difference is the height at which PCB s were
measured. For the W a1Ten County landfill, the measurements at thC' mait.l vent ue
somewhat difficult to compare to ones made at different hei~ts above an opeu landfill
location. Di:ffor~t heights mean that the potential for diluting PCB vapors ill.creases with
incre:ising height above: the PCB waste. The published paper never made any explicit
analysis of the data from the different landfills. HQw~•~r, the paper focused on results
rrom measurements at 120 to 180 cm for the other three Lmdfill$, which is some"hat
analogous to the main vent pipe condition at the Warren County Landfill. For this
comparison, the ma."mllWJl level found at the main vent was actually higher than ID.Ost of
the ma.'rimum levels reported over hot spots (i.e.: areas of high PCB levels) at the otl1er
three PCB Lmdfill.s eumined in the study, with the o~er, corresponding valu¢s being
18.0, 33.8, and 193 ug!cm. But these other levels were found for field tests conducted in
the summer of 1983, and the paper showed that PCB levels were very much higher in
June/July thau in October, "1i:th the average increase being about 200%. This ~_ge~s
t~t the maxmmmlevel of 11.0 .2 ug/cm found in Jan./Feb. in Warren County could easily
be 200°.11 or more greater if measured in the 5Ullllller, or perhaps some 3.50 ug/cm. In
otlfe.r words, t/re high kvelsfou.nd 11.t tht!. mailt vl!nt at the Warren Courrty La.ndfill
~~,, quif.t! comparabl.e, and p~rhaps e11en gr~at~r than the [e'l,·el.s[<Hlnd above the other
three u.ncontroll~d landjtlls, contrary to the s1«tement made in tht,published fXlPU that
the /~5 were! "low." ·
• It is itnportanc to not; tut the srudy also found that only the main vent was releasing a
positive gas .flow, but there waj no measurable gas .flow for the leach.ate collection pons
and ~ vents. This means that only the main. vent wis serving as a source of PCB
releases into the afr during the Muter period of the study. This f.act does not change the
accuracy of the mcasuremcuts of PCBs at the other locations: but only that there was no
measurable n.atural fl.ow of gas out of those other openings at that time. The study
measured the acrual mass .flow or flux rate for PCB releases from the main vent as 12.1
nanograms/second (ng/s). In otht:r wards, any staume.nn by governm~nt officials that
no "PCBs wr:r~ being rt!~as~from th,: Wa"en County Landfill were to1aUy
contradictory to tlu acrualfle.ld data obtained in ~arty 1983. The EP.-1. sl"dy meas11r~d
u11contr-olled rdeases of PCBsfrom the l@dfdl which probably would b~ significantly
highc during warmu puiods.. ·
■ The nex-t b.igb.est levels wer~ found at the upper leachate access port (2% of 1242 and
25% of 1260). Tbis is con~tent with contaminated leachate residing in the upper
leachate collection system at the time.
• Mnch lower levels were found at the lower le:schate access port (.03% of 1242 and 2.5%
of 1260). For both PCBs, however, there were positive readings above the very low
detection limits reported. In othu words_. thtrt were reliable futdings of PCB 11apors
from the lower leachate t:ollection syst~m b~low thl! main day and plastic bottom lin~r
system. This indica~ that PCB ctmtaminated kadr.a~ had readr~d th~ /Ok/er
leacht:t~ collection s_vst~m ur early 1983, because there is 110 other pu,u.sible
explanatiOlt for fi11di1rg PCBs in tlr~ air drawn from thl! a.cuss pip~ EPA 's fuiding
wa.s cons/.st~t with leaking of contaminaud l~dtatL through both the day layer and
plastic bottom liner into the Jowq leacluzk r,iu,zitoring and collection layer and sump.
■ Even higher levels of PCBs were found at the two smaller vents, with th(: vent with the
highest readings showing l~els in the range found for the upper leachate: access pipe.
• Although most of the measuxem~ts at the closest house were below detection limits, one
of the six measurements was at 0. O l ug/cm for Arochlor 1260 (the more toxic PCB). The
key question is: is th.is concentration of PCB ofhealth significance? To answer this
question one can consult .several EPA risk asse&smenrtype databas=s. Bo1h EPA. Region,
3 and 9 maintain such databa$eS and they contain the levels of PCBs in air that pose an
~xccss cancer death nte of one in a million ( or 10·6) for residential exposure. 1 Those
concentrations are 0.00081 and 0.00087 ug/cmin the two EPA databases. Since the
measuted level at the house was 0. O I, the data show that the cancer risk kvel at the house
,va:i about 10 times greater, or approxim.a:rely 10·5 risk. In other words, the 'W.llltcr-time
high readin~ at the house that was over onchalf mile a1>;·ay was of considerable
significance. Also , the other findings that were below the detection limits of0.006 and
0.01 ug/cm for Arochlors 1242 and 1260, respectively, are al.so significant. The reason is
that the appropriate EPA methodolo!)· is to use onehalfthe detection limits when
ass~SSJn! health risk, rather than a.$.SUme th.at a nondetect is 1 zero concentration. Thus, it
is proper to assume that most of the readings at the house would have been 0.003 and
0.005 ug/cm, and these are .also above the 10"" risk l~vel of 0.00081 and 0.00087 and
correspond to ~s of3 to 6 x 10·". In otlur words, all ofthtfUtdinzsfrom the
published pa~r shr.,w that lellels of PCBs found at the resid(!nc~ clost!$1 to the w~.,
County I.mid/ill in the winter of l98S were of l«ea/tJ: signif~ance and, lllfact,pcs«i
an wnacctprable /.ong term cancu risk. In all probcibiluy tht PCB levels found offsite
would have l>ettr considuably hirher in warm~, puiods, and maJ' also have increased
ov,r time.
• The study also found measurable PCB lev$ls iu ambient ajr at various other l~ations,
particularly a o.umbet ofreadings at th~ fi:nce line and downwind, with the maximWll level
fouod at 0,07 uglc~ v-.-h.ich is som.e 100 times greater than the 10-6 cancer risk level, or a
risk of 1 O"', which is a very high risk level. This find.mg was even more e-vidcie¢ that PCB
relel\seS from the landfill were occiJtllllg.
■ TI1e paper did not pay dose attention to the differences found in measured levels of
1The 10·' cancer risk level is the b1::it criicriou to use: because it is the IJ')()st used basis for
cleanup decisions in the federal Superfund program when residential i:xposures are appropriate.
5
Arochlor 1242 versu:i .Arochlor 1260. It is kno\\/11 th.it the vaporization rates decrease
sign.incantly with increasing chlorine content and the vaporization rate of 1242 is about 10
times higher thm for 1260. Because the water solubility of 1242 is about l 00 times
g:reata" than for 1260, one would also expect m.ore 1242 when leachate is the source of
PCB vapors. For the most pan, th.is rela.tionship existed in the data rcponed. For
example, for the main vent data the average level was 120,2 ~cm for 1242 but only 2
ug/cm for 1260. What merits some considenltion, how~er, is that over the longer term,
the .[J]Or-c slowly vaporizing but more toxic Arochlor 1260 will be expected to represent a
higher fraction of all the PCBs cmi.ncd ftom the landfill
lnfarmatian Provided Bx The State Hu Beu Wrone
Only three pertinent doc-iurumt& ha"·e been found in the files provided by the state. First1
the state's description of the ltndfill and its chronology contains a statement that in Jmuuy 1983
"EPA monitors gas venting from landfill and reports no significant :missions of PCBs. 11 Use of
th~ t~nn significant is fil:c use of the ten:n negligible by EPA, discussed earlier in this r¢i'on. The
secmi:D.gly simple statement is in fact incorrect and misleading.
Second, the s::.cond page from what apparently was only a t'wo paie actual rep on from
EPA about the testin! has bcoi. found in the file~ provided by the; ~~te. 2 The; conclusions
presented in this one page soction r:::veal a supc:r.6cial analysis of incomplete data and a clear
a.ttempt to downplay any health risk issue related to uncontrolled air releases of PCBs. lt
c~rtainJy """ould have be~ normal for EP ~ persollllel to prepare some type of report for the state,
especially since the study of PCB emissions was reportedly done because of citizen concerns
10uc undated page wa.s found; it is titled Section 2 Conclusions and the page number is 2
and pre~ted four conclusions, indicatin~ that the first and only other page may have been a brief
des...-ription of the field work by .EPA. Haud written statements at the top of the page: a.re: ''Ail
Quality, Monitoring Data from Jan. '83, Air Vent (6" PVC) Wan-en Co. PCB Landfill" Most
likely the brief report was prepared by EPA durin~ the latter part of 1983 or early 19&4, perh1ps
before all the data was fully assessed. The first conclusion noted ~t the main vent was ''the
principle i.ource of emis&ions " The average values given for the two Ar0chl0Ts match tho$e in
the published p~er. The first condusio·n also s..aid "These conccntratjons are subitmti~lly lower
than the current occupational standards for workplace atmospheres ... " But workplace standards
are set fo:r sh.ort time exposure and acute health effects. The second conclusion referred to
amb1~t PCB levels, but did not correctly refer to levels found above detection limits. The th:ird
conclusion referred to the results of 1nathematic~l modelias, md stated that levels would not be
significmt, but no mc:ution was made of the positive findiog at the neirby house. The fourth
conclusion predicted tha.t the low PCB emission rate would be reduced stiD further because of less
decay of organic matter and production of methane and by "removal of water from the site" that
should "reduce emission rates ~stantially. 11 But the state says that methane is still being
produced and the water was not removed.
6
I,·-·, • I
8 d
Thira, a page titled "pCB l.ANDFILL MlSCELLANEOU~ ~AMr'.L.1::.::,·· preparea oy tne
• state'·s .Division of Solid Waste: Management Division, and given in a doCUIIldl.t entitled Sampling
Analysis, md Leachate Removal Activity 1982-1993 (delivered to the Working Group in April
1994 ), contains four results of gas measurements in January l 983, which is the EPA work. The
probiem 1s that the data do not match the data in the published paper or the information in the one
page of e-0nclusions apparently prepared by EPA 3 Some especially significant discrepancies ue:
■ The state reponed the high:::st level of PCBs found at the "gas vent exhaust'' (surely the
n:wn vent) as ''3.0 PPB" (three parts per billion), which eq'll.ltes to 3 ug/cm, in comparison
to the .acnal maximum level reported by EPA of 120 ppb in the published paper and 123
pp b in. the one page of conclusions. Even if the state argu:::d that it nevet closely examined
the published paper it had in its possession for many years., it surely had the short report
prepared by EPA for the state so it could address public concerns.
• The state reponed for "ambient air samples" the resuh of "none detected, .. but the state
failed to r~or1 the finding of PCBs in tbe air at the location of the house about a haJf a
mile from the landfil11 which is a very significant distance to find such measurable levels,
nor the positive .findings at tbe fct1.cr;line, as reponed m the published paper. 111e:
statement in the one page conclusions page from the EPA report that ambient PCB levels
11were found to b::: at or below minimum detection limits" was de.finitely not in a.greem.-nt
with the data reported in the published paper, that inchlded four readings significantly
ab\1Ve the detection licit. The only plausible, acceptable ex:plmatio11 might be that the
brii:f EPA rep on was pr~ared p.rior to completion of EPA ' s analysis of .ill of its field
dau. This would be consistmt with the desire by the state to receive the findings oftbe
EPA study as soon as possible in order to address the consider.able public oppo:.i.tion of
Warren County residents to the landfill. In any event, the state had the published paper
that clearly rcv~led tbe positive findings of PCBs at the foceliu.e ind the nearby hous~.
• The state reported for the ''leachate CQUcction pipe exhaust" a result ofle~ thaz1 on~ pa.rt
per billion. Since EPA mQmred PCBs at the both the upper and lower leschate wllection
acc:ss pon$, the ~atc)s information is at best incomplete a.nd at worse JlliJiJ.cadi:llg. The
State would logically have had a problem with cxplailllllg any positive find.mg of PCBs
from the lower a.ocess port. The one page of EPA conclusions did not incllde any data
for the leachate collection acceSi ports, raising the question of why the state teported less
than one part per billion, especially since the state had the publi&hed paper that inclurlc:d
the maximum value of 2. 6 ppb given for the leachate access ports.
1Jt may be significant that the data report~ were given for January 6 and 12 only, but that
the published paper S4id that the study h.t.d been unde1taki:n in Januazy and February, suggesting
that not all the dtta became re;ogn.ized by the state, ~en though it was published in the paper by
the EPA scientists in 1985 and the state office had th.at paper for IIWlY yesrs.
7
HO~..:! HdV L · L l 966 L -L~-"r
Because the-rt! is no evide11ce that the adual EPA rqx,rt or the published paper had
bun provide.d w the genaal public or the Working Group by the state, these discrepancies in
the only information made available tkmonstrate tJuzt the S'late inuntionally 1nisr~presaitd
the findings of the 1983 rests for PCB air releaJes from th, landfllL This wtt.s done most
rt!centlJ in 1994. Se~a/ ~oplu of the 1985 publlshtd paptr Wtrtfou11d in stal~filu o,i thtt
landfill witlt a marki,ig that tlu paper had hem rt!ceived in 1986. ,4ll tl«c facts show rhat tlit!
stau. has )teodfastl)' <kceived the p14bli~ alxHlt the data and, mort! i,nportantly, the
significan~e oftlsefuu/ings with respect to public health risks. All PCB t:missimu could have
hu,c prev~,tt~d at mi.Jf.imal ~ost b1 using some form of carbon ad.sorptiort de~ice at ail landfill
gas ~it ports.
Copchpiou.,
The concerns of Warren County residents about uncontrolled releases ofPCBs into the air
emanating from the Warren County PCB Landfill were well founded and, in fact, proven by the
1983 measurements made by EPA, but incorrectly disregarded by EPA as well as the state. In
fact, the record shows that the state has explicitly misrepresented the facts of the EPA study in an
attempt to undermine the concerns of Warren County residents about the health risks of PCB air
emissions. Although UUlDY re,idcnu have be-ca very concerned about. drinking wit.er being
contamin.ued, all the avaibble datJl indicate tlu1t public health rhks resulting from PCB air
emi.ssions have probably been the most sii:nificant threat for resident, lhin: relatively close
to the landfill..
It is disturbing that EPA scientiSts made exp licit st.:itements designed to counter the
concerns of citizens and that those statements were not supported either by the facts or any
am.l)·sis. Of course, EPA itselfh.:id giv~ the state not only lpproval to construct the landfill, but
also financing for it, so it was not a completely impanial party, In fad, the levels of PC/Js
mea:surttd in 198.S were ndtlier low or n~gligibl~, as EP.◄ claimed. Th~ rwo chief
cottst!quQSas of raiching the wrong conclu.sio,u was that no ~mission control tecJrnofogy was
used to pr~ve/J.t PCB llir rele11ses a11d no additicmal mcJn.itoring WIU' ca"ie.d out If EPA had
come to the conclusion that sig,riflcanr /.ev,:ls of PCBs were being e.mi'tted from th~ landful,
the St!ri<Hls conurns of local residmts would hav~ b~~n fully supported by ti,.~ fedeNJl
govo-nment. fl would have been approprim~ ftw EPA, <>n th~ bast", of Its initial findi,,gs, to
ltav~ co,iducud another round of more e:,;;ten:si~e air monitoring in the summer of 1983,
especially at h oniesi'ks withilf one to mil~ of th~ landfill.
As to the issue of PCB health effects and monitoring, it should be noted that the federal
go" ernment had examrned the issue of whether exposure to the origin!l PCB spill materials on
North Carolina roads rl':sulted in increaS1;$ in PCB levels in breast milk. The study was completed
in 1982 and publishedin 1983 (W.J. Rogan et al, Chromatographic Evidence of Polycbloruated
Biphenyl Expo~e From a Spill, JoW"llal American Medical Association, voL249, no.8, pp. 10,1-
8
1058). The researchers at the National Institute ofEnviron.mental Health Sciences concluded that
the data for 12 exposed women that had been part of a larger study indicated that some part of
the PCB levels found in their bodies correlated with the types of PCB& spilled on the roads. Toe
women had been exposed to PCB vapors from spill locations along roads. It would have be~
very useful to monitor the PCB levels in breast milk in women living near the Warren County
landfill, but that did not happen.
The levels of PCBs measured during the 1983 ~ter period around the landfill and,
especially, in the area of the residC"ni:.e closest to the landfill, were of health significance and
con,-em. There is every re.a.son to believe, on the basis of scientific principles, that emissions of
PCBs from the Warren County PCB Landfill have been .significant for over 15 years. Release
rates would probably have been greater in wanner periods and may have changed over time as
more time was aYailable for vaporization of PCBs within the landfill, as compared to the early
1983 period, just a few months after the wastes were buried in the landfill. Although local
residents report having asked the state over the years to we some type of afr pollution control
system. such as carbon adsorption: the state: never implemented any control method.
In addition to cancer risks, however, attention must also be ~en to non-cancer health
effects that could result from long periods of exposure to relatively low PCB levels. Also,
exposure during pregnancy is a. threat. Research 3t Wayne State University and published in the
New England Journal of Medicine iD September 1996 reported developmental effects in eleven-
year old children wilosc mothers had consumed PCB contaminated fish in.the l 980s while
pregnant. At birth children had smaller heads and lower wei!hts, and later: children had lower IQ
scores, poor reading comprehension. memory problems and shorter attention span. Similar results
have been found in several animal studies and in studies of Taiwane~ children accidentally
exposed to PCBs. The damage to children was deemed similar to the effects oflead poisoning in
children. ·
The air release of PCBs also raises questions about other exposure routes. For example,
air relea~s of PCBs implies that some PCBs would be deposited on the local surrounding lands,
some of which are used for agriculrural purposes, suggesting that crop~, vegetables gro-wn in
housch.old gardens, and dairy and meat products from cattle grazing on local lands could be
sources of PCB exposure throu~ ingestion, not only for local residents but perhaps for others.
There is clearly a need for the st~te or EPA to conduct several types of tests. including the
following:
■ Current PCB emissions rates from the main vent and any other points open to the
atmosphere should be deternrlned using the most sensitive and reliable testing methods.
■ Testing of PCBs iD human tissues among adults and children who have lived near the
landfill for long times should be conducted.
9
,_,,..,...., • 1.1~1 : I I ~J-/r.-T:
-:::-h;T _ .10::7-i;ri; • YP .J .Jn,-,--'1c, c,t-.1r··,.u,-11r1 G-, .J
■ Health effects surveys should be conducted for long tcnn nearby residents, especially
children home by mothers that liv:d near th.c sne during pregnancy.
• There should be some t~sting of locally grou.n fruits and vegetables (and perllaps C3.Illl.~
locally grown produce knov.n to have been grown some years ago).
If PCBs are still being rele:is..-d from the lmdfill, then an engineering study ofusmg some
type of carbon adsorption 11-ystem. should be immediately in±tuted.
Finally, the ("e.,ulU reported here supply still more support for the state t»king
,i::riously its commitment to detoxify the Warren County PCB Landfill and to make 1
commitment for supplying the nece:,s•ry funding very quickly. There is oow abundant
proof that the hlndfill has not been safe and 3ecure, that the critically important bottom
d.3~ .ind plutic liner ,ystem leued, that the lesch1te collection system nevei-wor1'td
effectively, ind that PCBs h2ve leaked directly ioto the 2lir and lt,.ch:ate into the
!lurrounding soils.
1,.ir"!:.l.J W..·H 1 • I I '¥;F, l -/ c.--V
Ora: T _ ·' ,-7 _,: T ,: • y p .J .J(·,,T'-1C, Cit,. IT '··f .J1-,111 ,::r-, .J
Publishe4ci in Environmental Science and Technology, October, 1985, pp. 936-!:l~l. by the American (:hemical Society
Measurement of Fugitive Atmospheric Emissions of Polychlorinated . Bipheny!s
from Hazardous Waste Landfills
Robert G. Lewis• and Barry E. Martin
U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711
Donald L. Sgontz and James E. Howes, Jr.
Battelle Columbus Laboratories, Columbus, Ohio 43201
■ Four landfills lmown to contain large quantities of po-
lychlorinated biphenyls (PCBs) were monitored for at-
mospheric emissions: Three of these were uncontrolled
and contained large numbers of electrical capacitors, many
of which were scattered on the surface and leaking PCB
askarel fluids. The other is a state-of-the-art PCB waste
landfill designed to exceed the requirements of the Toxic
Substances Control Act of 1978 (ToSCA) for PCB disposal.
Both high-volume and low-volume air sampling equipment
were utilized at each landfill to monitor air levels of PCBs
on site, upwind and downwind. In addition, vent ports
were monitored at the controlled site. Simultaneous,
collocated sampling was performed for quality assurance
purposes and to obtain information on sampling perform-
ance and comparability. High atmospheric PCB concen-
trations were measured at the uncontrolled sites, while air
levels were at or near background at the ToSCA-designed
landfill. PCBs were detected at low levels in gas vents at
the latter site.
Introduction
The disposal of polychlorinated biphenyls (PCBs) is
strictly regulated under the Toxic Substance Control Act
(ToSCA). Specifically designed _chemical waste landfills
with impermeable liners are required and are subject to
approval by the U.S. Environmental Protection Agency.
The only access to the external environment in the case
of ToSCA landfills is through open vents and ports re-
quired for leachate monitoring and for relief for gases
generated within the landfill from decaying organic matter.
Prior to 1978, however, many PCB-containing articles were
disposed of in municipal landfills or at uncontrolled
chemical waste disposal sites.
PCBs possess sufficiently high vapor pressures (lo-4-10-5
kPa) to be emitted directly into the air surrounding haz-
ardous waste disposal sites through volatilization from
contaminated surfaces (1, 2). They also may be released
from controlled landfills through vents along with more
volatile gases (3). To test this premise, four PCB landfills
were monitored in this study. One was a ToSCA-designed
facility in North Carolina which was studied to determine
if PCBs were being emitted into the surrounding atmos-
pheric environment from gas vents and leachate access
ports. The other three consisted of two uncontrolled
private landfills and one municipal landfill in Indiana, each
of which contained large numbers of PCB-containing ca-
pacitors.
Materials and Methods
Air sampling was performed with low-volume (L V)
and/ or high-volume (HV) sampling systems previously
described (4-6). The components of the LV sampling
system consisted of a battery-operated, constant flow
sampling pump (Du Pont Model P-4000A) and a glass
cartridge containing a 22-mm diameter X 7.6-cm long
cylinder of polyurethane foam (PUF). The HV sampler
consisted of a conventional Hi-Vol shelter and pump
modified by addition of an inlet head to accommodate a
PUF sampling cartridge. The sampling head was com-
prised of an aluminum housing which held a 10-cm diam-
eter particulate filter (Pallflex 2500 QAST quartz) followed
by a glass sampling cartridge containing a 62 mm diameter
X 7 .6 cm -cylindrical PUF plug. The General Metal Works
(Villaue of Cleves, OH) Model PS-1 sampler is essentially
identical with the HV sampler used in these studies.
The Du Pont sampling pumps were calibrated with a
Du Pont calibrator pack (catalog no. 66-242-f-l) before and
after each sampling period. The HV samplers were cali-
brated once each week by means of a calibrated venturi
tube which was attached to the inlet. Flow audits were
conducted in the field prior to and at the termination of
monitoring activities by an independent team using two
laminar flow elements calibrated by the National Bureau
of Standards (NBS). Clean filters and PUF plugs were
used to simulate actual initial operating conditions.
The PUF plugs were precleaned by Soxhlet extraction
with acetone as previously described (4), vacuum dried,
and loaded into the appropriate glass sampling cartridges
under clean laboratory conditions. The cartridges were
wrapped in hexane-rinsed aluminum foil and stored in
carefully cleaned glass jars padded with clean PUF for
transport to and from the sampling sites. Exposed filters
from the HV samplers were completely wrapped in pre-
rinsed foil and placed in the jars with the PUF cartridge.
Disposable latex surgical gloves and prerinsed tongs were
used for handling the sampling cartridges when PUF plugs
were loaded and unloaded in the laboratory and for atta-
ching them to the sampling systems in the field.
Exposed PUF plugs and filters were Soxhlet extracted
together with 5% diethyl ether in hexane following the
procedure of Lewis et al. (4, 6), the extracts reduced_ to 10
or 1 mL in Kuderna-Danish concentrators according to
analytical needs, and the PCBs determined by electron
capture gas chromatography following EPA Method 608
(7). Identification and quantification of Aroclors 1242 and
1260 in the samples were performed by the technique
originally described by Webb and McCall (8). All solvents
were pesticide quality or analytical reagent grade. NBS
Standard Reference Material 1581 (Aroclors 1242 and 1260
in motor oil and transformer oil) were used for calibration
purposes. Recovery of Aroclors 1242 and 1260 from PUF
plugs fortified with SRM 1581 at three levels (0.06, 0.6,
and 6 µg/plug) averaged 93%. One laboratory bl~k and
one laboratory "spiked" PUF plug were analyzed with each
20 samples. In addition, ca. 10% of all samples analyzed
were field blanks which had been transported to and from
the monitoring sites.
Continuous measurements of wind speed, wind direc-
tion, ambient air temperature, and relative humidity were
obtained with Meteorology Research, Inc., portable
weather stations. Since the landfill sites were subject to
j
3-STAGE
LEACHATE FILTER
POND
. . . ----PCB -
TAMINATED
SOIL ..., , • / . _,J. ORIGINAL
GROUND
-/ SURFACE □ CLEAN EARTH/TOPSOIL
~ ARTIFICIAL LINER fill! LEACHATE COLLECTION SYSTEM
9 CLAY LINER WATER TABLE
----
Figure 1. Cross-sectional drawing of controlled PCB landfill in North Carolina.
the Comprehensive Environmental Response, Compensa-Q ■HOUSE A MAIN VENT
8 SMALL VENT
E SMALL VENT tion and Liability Act ("Superfund"), all samples were
collected, handled, and transported under standard
chain-of-custody procedures.
Results and Discussion
Controlled Landfill. In 1978, an unprecedented spill
of PCBs along 387 km of roadway occurred in central
North Carolina. Some 40 000-120 000 L of transformer
fluid consisting of Aroclors 1260 and 1242 in chloro-
benzenes were illegally and surreptitiously dumped along
the shoulders of the roads. Over 4 years of litigation were
required before a disposal site for the 30000 m3 of con-
taminated soil could be located in a sparsely populated
area of the state. Because of the great amount of public
concern over the safety of the disposal site, a state-of-
the-art landfill designed to exceed the requirements of
ToSCA was constructed in late 1982. The EPA-approved
(Superfund) site, located on a 4 ha of land in a rural area,
measures 75 m X 145 m and has a maximum depth of
about 7 m. The contaminated sbil is encapsulated within
0.6-m (top) to 1.5-m (bottom and sides) thick layers of
highly impervious, compacted clay, augmented with 10-mil
(0.25-mm) and 30-mil (0.75-mm) plastic liners on the top
and bottom, respectively. The landfill is properly sloped
and equipped with sump pumps and a leachate collection
system for monitoring purposes. It is also provided with
a gas vent which protrudes 1.5 m into the landfill for relief
of methane and other gases generated by bacterial decay
of organic matter contained in the soil. A cross-sectioned
drawing of the landfill is shown in Figure 1.
Several months after closure of the landfill,local resi-
dents voiced concerns that gases emanating from the vent
pipe and (as yet uncapped) leachate collection pipes may
have been introducing PCBs into the surrounding atmos-
phere. In response, a study was undertaken in Jan and
Feb 1983 to monitor these emissions and the ambient air
at the site.
Only the L V samplers were used in this study. The
vents and leachate access ports were sampled by placing
the PUF cartridge inlets into the pipes or ports and sealing
the openings with plastic bags to assure maintenance of
positive pressures. Sampling was performed for 8 h at
reduced flow rates of 1.2-1.4 L/min so as not to exceed
the volumetric flow rates of the vent pipes. The sampling
efficiency of the L V sampler employing PUF cartridges
had been previously shown to be essentially quantitative
for collection of Aroclor 1242 in natural gas, which is
97-98% methane (9). Soil temperatures within the landfill
were presumed to be nearly constant at 4-5 °C (thermic
temperature regime); therefore, emission rates were es-
sentially independent of ambient air temperatures.
0-._ ----
l
1000m
l
C UPPER LEACHATE ACCESS PORT
0 LOWER LEACHATE ACCESS PORT
0 SAMPLERS
-._ • B --o._ __ ___
75m
130m 36m O 36m -0 36m 160m -01------A•~o----+----0
....-,...::.:.0-i'm o• A
_.,......0-....-c•~ W,
__d......::C-------------' LEACHATE
--PONO 0---
WEATHER STATION N----Figure 2. Sampling arrangement for monitoring at the controlled landfill
when winds were from the north. Array was rotated with wind direction
so as to monitor at the perimeter of the landfill and at half the distance
for the main vent to the perimeter.
Ambient air sampling was performed with an array of
samplers operating at 3.8 L/min and located from 1 to 200
m upwind and downwind of the main vent as shown by
the example presented in Figure 2. Air samples were
collected at 1.2 m aboveground at each of the 13 locations
designated by open circles. Simultaneous samples were
also collected at 4.6 m above ground at the perimeter of
the mounded landfill (72 m downwind of the main vent
in the arrangement shown in Figure 2) so as to be on
line-of-sight with the opening of the main vent. The spatial
arrangement of the samplers was alternated as necessary
at the start of each sampling period to reflect average wind
direction. That is, downwind samplers were placed in a
fanned-out array at half the distance from the main vent
to the perimeter, at the perimeter, and at the tree line
(100-130 m downwind). Upwind samplers were placed
accordingly. One sample was taken at the same location
in the yard of the nearest house (1 km away) during each
sampling period. Sampling was performed from 0900 to
1700 Eastern Standard Time on 3 days and from 2100 to
0500 on the fourth day. Ambient air temperatures ranged
from -1 to +14 °C, wind speed from 0.04 to 6.6 m/s, and
relative humidity from 46 to 95%. No vent sampling was
performed on these days, and all vents were left open.
Analytical results from the gas samples collected from
five vents are presented in Table I. It should be noted
that the lower leachate removal pipe extends below the
landfill. The two small pipes were installed temporarily
to release gas bubbles in the upper plastic liner. PCB
concentrations in the gases (principally methane) ema-
nating from the main vent ranged from 105 to 141 µg/m3
measured as Aroclor 1242 and from 1.8 to 2.1 µg/m3
measured as Aroclor 1260. A typical gas chromatogram
of the PCB mixture found in the main gas vent is shown
c ..... : ,......, c,..; ,-.hnnl \/nl 10 "'''"' 1n 1QRI:;; QA:'7
Table I. Controlled PCB Landfill Vent Monitoring Results
PCB concentration, µg/m3
Aroclor 1242 Aroclor 1260
location range av range av
main vent (A) 105.7-141.5 120.2 1.8-2.1 2.0
upper leachate access 0.8-2.8 2.6 0.3-0.6 0.5
port (C)
lower leachate access ND-0.09 0.04 ND-0.08 0.05
port (D)
small vent (B) ND-O.o7 0.05 <0.02-<0.3 <0.02
small vent (E) <0.02-0.67 0.24 <0.02-1.3 0.4
0 Letters in parentheses refer to position identification in Figure
2. b Average of three to four measurements; 7-8-h samples.
1:1 AROCLOR 1242 AND 1260
PCB IN GAS VENT
Figure 3. Gas chromatograms of standard mixture and sample from
main gas vent at controlled landfill.
in Figure 3 along with the 1:1 Aroclor 1242-Aroclor 1260
standard used for quantifitation. Collocated samples
collected in the main vent agreed within 6-9%. Concen-
trations found in the leachate collection ports were much
lower, ranging from nondetectable (ND) to 2.8 µg/m3
Aroclor 1242 and from ND to 0.5 µg/m3 Aroclor 1260.
Detection limits were 0.01-0.02 µg/m3• Analyses of
transformer oil and soil samples prior to interment showed
a 4:1 ratio of Aroclor 1260 to Aroclor 1242. The greater
volatility of the latter mixture apparently accounted for
the relatively higher concentrations found.
Ambient air data are given in Table II. Only four of
the 39 ambient air samples analyzed contained detectable
quantities of PCB (:MDL= 6 ng/m3). Three samples were
positive for Aroclor 1260 only (at 11, 12, 50, and 71 ng/m3)
and did not appear correlate with proximity to the vents.
The two maximum levels were found 98 m downwind of
the main vent.
Gas flow measurements made at three different times
showed flow rates of 4.8-6.4 (average 5.7) L/min from the
main vent. No flow was measurable from the leachate
ports or small vents. The average gaseous PCB emission
rate from the landfill, therefore, was estimated to be 12.1
ng/s. By use of these emission parameters and meteoro-
logical conditions that prevailed during the study, standard
dispersion models were applied to calculate downwind
PCB concentrations for comparison with field measure-
ments. Two EPA models (10, 11) were employed to cal-
culate estimates of maximum hourly concentrations under
a full spectrum of meteorological conditions and estimates
of the range of hourly ambient concentrations that would
occur at downwind distances of 50, 100, and 150 m under
the meteorological conditions that probably controlled
Table II. Ambient Air Monitoring at Controlled Landfill
air concentration, µg/m3
location
beside main vent
on site, downwind
on site, upwind
fence line, downwind
off site, downwind
off site, upwind
nearby house
Aroclor 1242
<0.006
<0.006
<0.006
<0.006
<0.006
<0.006
<0.006
Aroclor 1260
<0.01-0.01°
<0.01
<0.01
<0.01-0.07b
<0.01
<0.01
<0.01-0.01'
0 One of six measurements above detection limit. b Two of 13
measurements above detection limit (0.05 and 0.07 µg /m3). 'One
of six measurements above detection limit.
dispersion during the field monitoring. The predicted
maximum 1-h concentrations downwind of the vent under
prevailing and worst case conditions ranged from 4 X 10...;;
ng/m3 at 14 m to 1 X 10-7 ng/m3 at 50 to 150 m. The 8-h
concentrations would be expected to be about 10-s ng/m3•
These values are far below the detection capability of the
sampler employed (or that of any known sampler).
Uncontrolled Landfills. Three disposal sites in the
vicinity of one city in Indiana have been the subject of
recent concern because they contain large numbers of
PCB-containining transformers which were dumped there
prior to 1972. One of the sites, designated here as site 1,
is about 8 ha in size and located on an abandoned farm.
Site 2 is a small, rural site of about 0.2 ha, surrounded by
mobile homes. Site 3 was a municipal landfill, some 3 ha
in area, located in a /lUburban area. At each of these sites
capacitors were stre·wn across the ground surface (some-
times in mounds) and were visibly leaking askarels con-
taining Aroclor 1242 into the soil and nearby streams. The
total quantity and depth of burial of the capacitors is
unknown.
Both L V and HV samplers were employed to monitor
PCB atmospheric emissions at the three uncontrolled
landfills. The air measurements at the sites were per-
formed at localized areas (hot spots) where leaking capa-
citors were evident and at other locations to determine
upwind (background) and downwind levels. Three dif-
ferent sampling approaches each using PUF cartridges for
collection of PCBs were used. The LV samplers were set
up at hot spots to sample during 8-h daytime periods at
3.8 L/min with intakes positioned 120 cm above ground.
The HV systems were set to sample at 226 L/min for
8-24-h periods at hot spots and at upwind and downwind
points. The intakes at the HV sampler were 180 cm above
ground. Where line power was unavailable, gasoline-
powered generators placed downwind of the HV samplers
were used. Arrays of five L V sampling systems placed
from 2 to 180 cm above ground were used to determine the
vertical concentration profiles at hot spots. Detection
limits were ca. 10 ng/m3 for the LV sampler and ca. 50
pg/m3 for the HV sampler.
Three or four days of monitoring was performed at each
site during June and July 1983. Results from single-level
(120 or 180 cm above ground) monitoring are presented
in Table III. The following summarizes these results:
At capacitor disposal site 1, airborne PCB concen-
trations measured at 120-180 cm above five hot
spots during the day (0900-1700 central daylight
savings time) ranged from 0.4 to 18 µg/m3• Levels
along the downwind perimeter of the site ranged
from 0.2 to 1.8 µg/m3• Upwind PCB concentrations
ranged from <0.05 to 0.09 µg/m3•
Ambient air PCB concentrations measured at
120-180 cm above two hot spots at site 2 ranged from
Table III. Range of PCB Concentrations (µg/m3) in Air at Uncontrolled Sites, June-July 1983
site l
location daytime
hot spots 0.4-18.0
downwind 0.3-0.5
upwind <0.0S-0.10
meteorology
temperature range, °C
wind velocity range, m/s
relative humidity range, %
Table IV. Vertical Profile Air Measurements at
Uncontrolled Landfills
distance air concentration,• µg /m3
above site la site lb
ground, cm October July July
2 271-520 577-1053 602-1108
30 27-33 56-120 111-157
60 8.6-18 30-58 40-S2
120 2.9--5.7 17-30 15-21
180 1.3-2.3 6.4-1.3 8.6-10
• Average daytime levels over a 4-day period.
24 h
8.3-13.0
0.60-1.3
0.08-0.09
19--42
0-2.3
62-83
site 3
July
367-955
53-159
28-69
16-33
6.8-21
0.6 to 19 µg/m3 during the day. Near residences
adjacent to the site, levels ranged from <0.04 to 0.2
µgj m3.
At the municipal landfill (site 3), daytime PCB air
levels measured at 120--180 cm above three hot spots
ranged from to 193 µg/m3• Upv.,ind concentrations
were fairly constant at ca. 0.05 µg/m3, and levels
measured downwind of the landfill ranged from 0.3
to 0.8 µg/m3• There were a number of residences
around the perimeter of this site.
Generally, the airborne PCB levels measured at hot spots
correlated with the quantity of. exposed capacitors.
Vertical emission profiles at h6t spots are given in Table
IV. Air concentrations decreased with distance above
contaminated surfaces. Levels at 2 cm above ground level
were from 40 to 100 times higher than the levels at an
elevation of 180 cm, while those at 120 cm were twice as
high. PCB emission rates during July 1983 were probably
maximized by the unseasonbly warm (19-42 °C), dry
weather which prevailed throughout the study period.
Sampling data from site 1 during Oct 1982, when tem-
peratures ranged from 13 to 30 °C, showed PCB levels
about half of those obtained during the summer.
Typical gas chromatograms of downwind and upwind
samples from site 1 are shown in Figure 4. Chromato-
grams of hot spot air samples taken at several heights are
presented in Figure 5.
Both the L V and HV samplers (including one com-
mercial version of the latter) performed well throughout
the study. Sampler pumps were calibrated before and after
each sampling period. Flow calibrations ranged from + 1.9
to-7.9% for the LV samplers and from +8 to-7% for the
HV samplers. Pre-and postflow readings agreed within
±5% in over 90% of the cases. Independent flow audits
indicated average flow accuracies of -2.8 % for 20 L V
pumps and +5.2% for seven HV samplers. Field blanks
analyzed with ambient air samples collected at each landfill
site showed no PCB (measured at Aroclor 1242) above the
minimum detectable level of 0.02 µg per plug. The blank
sampling cartridges were carried through all field handling
operations except attachment to the sampling devices.
During the field study, pairs of LV and HV samplers
were operated for the same time period at the same sam-
site 2 site 3
daytime 24 h daytime 24 h
0.6-33.8 6.3-193 21.5-77.4
0.08-0.20 0.3--0.8
<0.04--0.07 0.08--0.20 <0.04--0.05 0.08-0.09
19--38
·0.1-2.7
37-83
AROCLOR 1242
DOWNWIND
AIR SAMPLE
STANDARD
22-42
0-2.2
24-70
UPWIND
AIR SAMPLE
SITE 1
Figure 4. Gas chromatograms of standard and ambient air samples
at uncontrolled landfill.
AROCLOR 1242
STANDARD
AIR SAMPLE AT
2cm 11 :40 DILUTION)
AIR SAMPLE AT
60cm
AIR SAMPLE AT
120 cm
Figure 5. Gas chromatograms of standard and air samples at several
heights above hot spots at uncontrolled landfill.
pling location to estimate the reproducibility of the mea-
surement methods. Measurements were also made with
collocated L V and HV samplers to compare results ob-
tained by the two methods. As can be seen from the data
in Table V, good agreement between pairs was obtained.
The average difference calculated from all the paired L V
sampler measurements was 7.2% while that for the HV
Table V. Collocated Sampler Comparisons
paired LV samplers,• µg/m3 paired HV samplers,b µg /m3 paired LV and HV samplers,b,, µg /m3
% difference % difference % difference
sampler A sampler B in pair sampler A sampler B in pairs sampler A sampler B in pairs
9.4 11 15.7 12 14 15 8.7 7.1 +20.3
8.8 8.6 2.3 11 12 8 7.0 6.5 +7.4
11 6.8 47.2 37 61 49 8.5 11.6 -30.8
11 12 8.7 98 89 9 5.2 9.9 -62.3
5.1 5.3 3.8 30 47 44 11.0 12.3 -11.2
7.9 9.1 14.1 20 23 14 8.5 13.5 -45.5
18 19 5.4 45 45 0 18.5 28.0 -40.9
77 72 6.7 11.0 21.0 -62.5
6.2 6.0 3.3 40.0 49.0 -20.2
85 89 4.6
11.2d 20d -27.3d
• Located 30 cm apart. b Located 1 m apart. 'Data corrected to account for difference in height of intake above ground. d Average.
Table VI. Comparison of Active and Passive Sampling Data
air concentration, µg /m3
distance above site la
ground, cm active passive active
2 1060 980 670
30 120 53 90
60 37
120 19 14 17
180 5.1/5.2
• Average.
samplers was 20%. Since the inlets of the HV and LV
samplers were not located at the same heights above
ground, it was necessary to apply a correction factor in
order to achieve reasonably accurate comparisons between
values obtained from collocated pairs of HV and LV sam-
plers. For this purpose, corrected HV values (C180) were
calculated by
where C180 and C120 were the PCB air concentrations
measured by the collocated HV and L V samplers, re-
spectively, and V180 and V120 were concentration values
obtained from vertical profile measurements with the LV
samplers at 120 and 180 cm above ground, respectively.
After these adjustments, the L V /HV sampler comparison
averaged -27 .3 % , with the L V sampler generally giving
lower results.
A limited numer of experimental passive devices (12)
were also exposed at hot spots on site 1 on 2 of the 4 days
of sampling. The passive sampling devices (PSDs) were
3.8 cm diameter x 1.3 cm stainless-steel cylinders con-
taining a series of diffusion screens and plates on each end.
Tenax GC (0.4 g) was used to collect PCB gases diffusing
into the devices. The sampling rate for PCBs was esti-
mated at 30 cm3 / min on the basis of trichlorobiphenyl.
The devices were transported to and from the exposure
sites in sealed cans, which were placed in a larger can
containing activated charcoal. Cyclohexane was used for
extraction of the PCBs from the PSDs. Collocated expo-
sures with the L V samplers showed reasonably good
agreement, as can be seen from the data presented in Table
VI. The detection limit for the PSDs was only ca 5 µg/m3
for 8-h exposures, however. Efforts are currently under
way to improve sensitivity by means of supercritical fluid
extraction and concentration.
Temporary remedial actions were taken during
1983-1984 at sites 1 and 2 to remove all exposed capacitors
site lb average
passive active passive % difference P / A
340 865 660 -24
125 105 89 -16
55/100 37 77 +48/108
18 14 -22
5/16 5 10 +1 00
+16.8°
Table VII. Range of PCB Concentrations• (µg/m3) in Air at
Uncontrolled Sites after Temporary Cleanup, Aug 1984
hot spots
downwind
upwind
location
meteorology
site 1
3.1-4.6
0.4-1.4
0.2-0.3
temperature range, °C 14-32
wind velocity range, m/s 0-1.3
relative humidity range, % 48-92
• 24-h samples.
site 2
2.7-3.1
0.1-0.2
0.1-0.1
Table VIII. Vertical Profile Air Measurements at
Uncontrolled Landfills after Temporary Cleanup, April
1984
distance above
ground, cm
2
30
60
120
180
air concentration,•
µg/ma
site 1
2.3-3.2
1.1-1.8
0.9-1.4
0.7-1.4
0.4-0.6
site lb
11.5-21.3
4.1-5.8
1.7-3.1
1. 7-3.1
1.5-2.5
• Average daytime levels over a 4-day period.
and obviously contaminated surface soil. Following this
cleanup, limited additional air monitoring was performed
during a 4-day period in Aug 1984. The results of HV
sampling at both sites are presented in Table VII. Vertical
profile measurements (using the LV samplers) at two
previously monitored hot spots on site 1 are given in Table
VIII. Significant reductions (by an order of magnitude)
of PCB air levels at the hot spots were noted postcleanup.
However, downwind levels at both sites 1 and 2 appeared
unchanged, suggesting that the landfill proper is still
contributing PCBs to the surrounding atmosphere.
Conclusions
The results of these studies demonstrate that fugitive
emissions of PCBs into the atmosphere can occur at un-
controlled landfills. At the three sites, PCB air levels
measured at hot spots on the landfills greatly exceeded
ambient background levels, thus indicating that PCBs
from the leaking capacitors were being emitted into the
air. Concentrations that exceeded background levels were
also observed at sampling locations downwind of the
landfills, even after removal of exposed capacitors and
obviously contaminated surface soil. By contrast, air
emissions of PCB from a well-designed chemical waste
landfill were found to be negligible.
All PCB sampling systems were found to perform well.
The LV samplers offered an advantage over the HV sam-
plers when electrical power was not available (as was the
case at most of the sites monitored). However, limited
battery life would not permit 24-h sampling with the LV
pumps. The experimental passive sampler, which can
readily operate unattended for 24 h, shows much promise
if its sensitivity can be increased by 100-fold through im-
proved extraction and analysis methods.
Acknowledgments
We than Donald E. Johnson of Southwest Research
Institute, San Antonio, TX, for valuable laboratory support
in preparation and analysis of PUF cartridges, Jack C.
Suggs of the U.S. Environmental Protection Agency, Re-
search Triangle Park, NC, for modeling assistance, William
F. Barnard and Jack A. Bowen of the U.S. Environmental
Protection Agency, Research Triangle Park, NC, for per-
forming field audits, James Gray of EPA Region IV,
Athens, GA, for field support at the controlled landfill, and
Ralph Riggin of Battelle for analysis of the passive sam-
pling devices. ·
Registry No. Aroclor 1242, 53469-21-9; Aroclor 1260, 11096-
82-5.
Literature Cited
(1) MacLeod, K. E.; Lewis, R. G. In "Sampling and Analysis
of Toxic Organics in the Atmosphere"; American Society
for Testing and Materials: Philadelphia, PA, 1980; Pub-
lication STP721, pp 56-69.
(2) Weaver, G. Environ. Sci. Technol. 1984, 18, 22A-27A.
(3) Murphy, T. J.; Formanski, L. J.; Brownawell, B.; Meyer,
J. A. 184th National Meeting of the American Chemical
Society, Kansas City, MO, Sept 1982; American Chemical
Society: Washington, DC, 1982; ENVR 70.
(4) Lewis, R. G.; Brown, A. R.; Jackson, M. D. Anal. Chem.
1977, 49, 1668-1672.
(5) Lewis, R G.; Jackson, M. D. Anal. Chem. 1982, 54, 592-594.
(6) Lewis, R. G.; MacLeod, K. E. Anal. Chem. 1982, 54,
310-315.
(7) Fed. Regist. 1979, 44, 69501-69509.
(8) Webb, R. G.; McCall, A. C. J. Chromatogr. Sci. 1973, 11,
366-373.
(9) Jackson, M. D.; Hodgson, D. W.; MacLeod, K. E.; Lewis,
R. G. Bull. Environ. Contam. Toxicol. 1981, 27, 226-229.
(10) U.S. Environmental Protection Agency "NTIS Tape of
User's Network for Applied Modeling of Air Pollution
(UNAMAP)". EPA, 1980, Version 4, NTIS No. PB81-
164600.
(11) Turner, D. B.; Busse, A. D. "User's Guides to the Interactive
Versions of Three Point Source Dispersion Programs:
PTMAX, PTDIX, and PTMTP" 1973, U.S. Environmental
Protection Agency Report EPA/DF /OOH (NTIS No.
PB81-164667).
(12) Lewis, R. G.; Mulik, J. D.; Coutant, R. W.; Wooten, G. W.;
McMillin, C. R. Anal. Chem. 1985, 57, 214-219.
Received for review December 19, 1984. Accepted April 5, 1985.
Although the research described in this article was funded wholly
or in part by the U.S. Environmental Protection Agency through
Contract 68-02-3745, it has not necessarily reflect the views of
the Agency, and no official endorsement should be inferred.
Mention of trade names for commercial products does not con-
stitute endorsement or recommendation for use.
. :>11:i"A ----~~...a--..c-:IL Environmental Consultants -
Barnes, Ferland and Associates, Inc. #0
TECHNICAL MEMORANDUM BFA #95-017
TO: Technical Committee
FROM: Patrick Barnes, P .G:, Science Advisor
DATE: October 21, 1996
SUBJECT: Potential Source of Water in the Landfill
The attached figure is a hydrograph showing water levels in and immediately around the PCB
Landfill for a 3 year period. The bottom four lines represent the water levels in the monitoring
wells MW-I through 4 which are generally situated east, north, west and south of the landfill,
respectively.
Hydrographs are very useful in determining how water levels change with the season and by
assessing how the individual monitoring points ·relate to each other, the general direction of flow
can be ascertained.
As can be seen, with the exception of MW-2 which appears to be in the predominant down
gradient direction, groundwater flow is largely radial. Specifically, as it relates to the region
covered by MW-I, 3 and 4.
Typically, well water levels peak in the months of March, April and May, and are at their lowest
in the months of October, November, December and January. This cyclical pattern of raising and
falling water levels is typical of natural groundwater systems. It is for this reason that we have
stated in previous documents that MW-I, which has not responded to recent increases in water
levels, has apparently failed.
The graph at the top of the attached figure represents the fluctuation of water level as measured in
the leachate access pipe and the central vent observation well. The rise and fall of this water level
was thought by the State to be directly related to the heating and cooling of the landfill materials.
We agree that heating and cooling may result in some fluctuation; however, the heating and
cooling process in a system with very little organics (less than 2%) should not result in a long-
term increase in water levels as shown by the green line. Moreover, the frequency of the peaks
and valleys align very well with that of the monitoring well hydro graphs. In our opinion, this is a
strong indication that the landfill is functioning as a natural system that is receiving and releasing
water. This pattern is not in phase (the peaks of the water in the wells don't match with the peaks
of the water in the landfill) with the surrounding area because it takes the water several months to
flow through the composite liners. The delayed yield shown is very similar to that which you
The Hollister Building• 3535 Lawton Road• Suite 111 • Orlando, Florida 32803
Office ( 407) 896-8608 • Fax ( 407) 896-1822
TECHNICAL MEMORANDUM
October 21, 1996
Page 2
would expect in a semi-confined, two aquifer system. It is the result of the time it takes water to
seep through the upper composite liner. For the period of record, it appears to be fairly constant;
however, it is important to note that the rate of leakage will increase over time. The leakage rate
is directly related to the permeability of that material, flaws in the liner system resulting from
pinholes and holes formed during seam welding (Banaparte and Gross, "Field Behavior of
Double-liner Systems"). Another potential source of failure is stress cracking or brittle fracture
(Lee and Jones, Municipal Solid Waste Management In-lined, "Dry Bomb" Landfills).
The average increase of the water level by approximately 1 foot over the four year period of
record is in line with seepage rates used by the USEP A for flexible membrane liners. It is our
opinion that this increase is a good indication that water may be seeping into the landfill. The fact
that the increase is not a steady incline but varies seasonally is an indication that the system is also
discharging water through the bottom liner. There is a net increase in the landfill water level
because more enters than.leaves the system.
Additional data such as monthly precipitation and average temperatures will be collected and
reviewed; however, it is not anticipated that the information will change the general conclusion.
In summary, if no water was entering the system given the low percentage of organic material the
water level would remain flat, and if no water was leaving the system the water levels would not
decrease then increase in a cyclical pattern.
PAB/psg/10-21 TC I.doc
The Hollister Building• 3535 Lawton Road• Suite 111 • Orlando, Florida 32803
Office [phone] • Fax [phone]
Jones-Lee and Lee, 1993 "Groundwater Pollution by Municipal Landfills: Leachate Composition,
Detection and Water Quality Significance."
While the latter paper focuses on municipal solid waste landfills, similar issues arise and
situations occur for hazardous waste landfills. Whether lead or some other hazardous/deleterious
chemical is in a "dry tomb" type municipal solid waste landfill or "dry tomb" type hazardous
waste landfill does not, for many constituents, affect the overall period of time that the constituent
will be hazardous in that type of landfill.
Deficiencies in Landfill Groundwater Monitoring Systems
While typically landfill applicants and their consultants assert that the groundwater
monitoring system for a landfill will detect groundwater pollution by landfill leachate before
widespread pollution occurs, in fact, when critically examined, it can be readily ascertained that
the proposed groundwater monitoring system will not be effective in detecting pollution of
_groundwater by landfill leachate before widespread pollution occurs. Dr. Cherry (1990) of the
University of Waterloo was the first to point out that typical groundwater monitoring systems
involving vertical monitoring wells spaced at buodreds of feet apart around a landfill have a low
probability of detecting leachate leakage from the · landfill that can pollute groundwater before
widespread pollution occurs. -
Dr. Jones-Lee and the author published a review article on this topic in which they have
utilized Cherry's findings to point out that minimum Subtitle C (hazardous waste) and D
(municipal solid waste) landfills that utilize typical vertical monitoring wells will not be reliable
for monitoring landfill pollution of groundwaters by the landfill before widespread pollution
occurs (Lee and Jones-Lee, "A Groundwater Protection Strategy for Lined Landfills," 1994b).
__This situation is easily understood by the fact that the initial leakage through the flexfuk
membrane liner of the composite liner for a minimum design Subtitle C landfill or the equiyalent,
can initially leak through holes, rips, tears or points of deterioration within the flexible membrane
liner. Such leaks will produce finger plumes of limited dimensions compared to the spacing of
_ groundwater monitoring wells. Groundwater monitoring wells of the type typically used will have
~nes of capture of about one foot on each side of the well.
The senior author has found that many landfill applicants space the primary groundwater
monitoring wells at about 400 feet apart at the down groundwater gradient edge of the waste
management unit. This means that the finger-like leachate plumes produced from the initial
leakage through the flexible membrane liner of a landfill could readily fail to be detected by the
monitoring wells. There is a space of about 398 feet between the monitoring wel]s through which
_ the finger plumes of leachate. which could be on the order of a few feet wide at the location of
the monitoring wells, could pass and never he detected.
The deficiencies in groundwater monitoring in lined landfills are not new. In addition to
the reports by Cherry (1990), Parsons in Davis (1992) discussed these issues in an ASTM
19
z MONITORING WELL ELEVATION (FT) 0 -I N N w w ~ ~ w r.i (0 (0 0 0 N 0 U1 0 U1 0 U1 0 § 0 0 0 0 0 0 0 -I 0 0 0 0 0 0 0 :,-11/24/92 -ui' s t:!l 12/23/92 co C: -, ~ CD 1/26/93 ui' ~ c:,j;) 2/26/93 I('[) 6. 3/26/93 -b c::=a ::, ~ 4/23/93 -e ('[) t:. 5/20/93 ~ 5/21 /93 -(D r a. ll> 6/25/93 .... ll> "O 7/23/93 -.,., ~ a: 8/30/93 Cb a. or 9/27/93 'l l 10/22/93 en ~ 11/18/93 --(ti 11 /19/93 -
12/17/93 -
1/25/94 -
+ 2/24/94
s:
1 3/25/94 - -
+
4/28/94 -G)
::0
s: 5/18/94 0
~ C
2 ~ 5/19/94 -C "tJ
+
~ (")
C 6/27/94 --.l> m
l> _, r -I m ► ~ m 7/26/94 -::0 2 ~ me w 8/26/94 -r -n mr t <r 9/26/94 .)> _,
s: 0 ~ 10/24/94 -2 l,.. en
t 11/13/94
11/16/94 -r s: ! 12/19/94
+
1/25/95 -
2/23/95 -< m z --t 3/29/95 -
4/27/95
5/24/95 -L-/ ·1 5/25/95 -
6/22/95
7/21 /95 -
8/28/95 -
9/25/95 -
10/24/95
11/20/95 -
12/20/95
1/26/96 -
2/23/96 -
w w w w w w w w N N N w w w w w
"" Ol 0, 0 N "" Ol 0,
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
ELEVATION LMW-1 & VENT (FT)
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
APR 2 0 JQ83
DATE: APR I 8 !003
SUBJECT: Draft Final Report on Ambient Monitoring for PCB' s at the Warren County
(North Carolina) Landfill
FROM: Barry E. Martin, Chief cd'~ ~
Field Monitoring Section, EMTB, fo:.MD, EMSL/RTP (MD-76)
TO: Doyle Brittain
EPA, Region IV College Station Road
Athens, GA
Enclosed is a copy of the Draft Final Report on Ambient Monitoring
for PCB's at the Warren County (North Carolina) Landfill submitted to
me by Battelle Colmnbus Laboratories, Columbus, Ohio. Please review
and provide your comments to me so we can finalize this task with
Battelle.
If I can be of further assistance, please call me at FTS: 629-3076.
Enclosure
cc: T. Hartlage (MD-76)
' Form 1320-6 (Rev. 3-76)
J
]
]
]
]
I
DRAFT FINAL REPORT
ON
AMBIENT MONITORING FOR PCBs AT THE
WARREN COUNTY (NORTH CAROLINA) LANDFILL
by
D.L. Sgontz, W.E. Bresler, L.A. Winker
and J.E. Howes, Jr.
Battelle Columbus Laboratories
Columbus, Ohio 43201
I
Contract No. 68-02-3745
Work Assignment No. 10
Project Officer
Barry E .. Martin
Environmental Monitoring Systems Laboratory
U.S .. Environmental Protection Agency
Research Triangle Park, North Carolina 2ffil
April 8, 1983
,
J
]
j
SECTION 1
INTRODUCTION
Approximately 40,000 cubic yards of PCB-contaminated dirt excavated from
along roads the central piedmont area of North Carolina has been disposed of
in an approved hazardous waste landfill in Warren County (NC). Local
residents and the Warren County Health Department have expressed concern
about the possibility of airborne PCB emissions from the landfill being
transported to neighboring areas, thus threatening the public welfare.
In answer to this concern, a study was performed to monitor airborne PCB
emissions from the landfill and ambient air levels on and surrounding the
site. The specific objectives of the study were:
o to determine if PCBs are being emitted from vent pipes on the
landfill.
o to determine if PCBs are present in the ambient air downwind of
the vent pipes. I
o to determine if PCBs are present in the ambient air in the
vicinity of the nearest residence, approximately one-half mile
away.
o to quantify the actual concentration of PCBs, if any, being
emitted from the vent pipes.
o to quantify the actual concentrations of PCBs, if any, being
transported off the landfill.
The study was performed at the request of the North Carolina Division of
Health Services and was conducted according to a plan developed by EPA,
Region IV and EPA/EMSL personnel. Battelle-Columbus Laboratories personnel,
assisted by Jim Gray of EPA Region IV, performed the field sampling program.
J
]
]
]
1
l
]
]
l
l
1
---··--·--•-····------·-·--------
Southwest Research Institute performed the PCB analysis of the polyurethane
foam sampling cartridges. The following sections describe the landfill site,
the sampling and analytical procedures that were used, and present the
results of the study.
SECTION 2
SITE DESCRIPTION
\
The landfill is located in Warren County, North Carolina on an
approximately 20 acre tract of land owned by the State of North Carolina.
The landfill proper covers an area of approximately 75m x 145m. In the
construction, plastic pipes were installed to vent gases 1and aqueous leachate
from landfill. The locations of the vents on the landfill are shown in
Figure 1. The main vent is 4 in. in diameter, extends approximately 1.2m (4
ft.) above the ground, and is located in approximately the center of the
landfill. Two leachate vents are located near the northeast corner of the
landfill. Two small vents have been added after construction to relieve gas
pressure under the plastic cover on the landfill. One of the vents is
located approximately du~ west of the main vent and the other is located near
the leachate vents.
'f I . _L.
""' I
j
i I \ _,
I
l .i..
l ;.
"'l /2 mi. ~ from landfill
House
B
0
Vent Identification
A-Main Vent
8-Small Vent
J I
N
A
0 .
C-Upper Leachate Vent
0-Lower Leachate Vent
E-Small Vent
\ 0 Leachate
Q OE
D Q
C
Figure 1. Vent locations on Warren County (NC) Landfill
Pond
I
1·
1·
I
1
l
]
]
]
]
]
]
]
]
]
]
]
l
l -
l -
\
FIELD MONITORING
SECTION 3
EXPERIMENTAL PROCEDURES
The field monitoring program was conducted _over the -period January 26 -
February 1, 1983. Sampling was performed to determine PCB emissions from the
5 landfill vents and ambient air PCB levels on and in the vicinity of the
landfill. The monitoring schedule was as follows:
January 26 -vents and ambient air (daytime)
January 27 -vents only
January 28 -vents only
January 29 -ambient air only (daytime)
January 30 -no sampling due to weather conditions
January 31 and
February 1 -ambient air onfty (night time)
All PCB monitoring was performed with DuPont P-4000A battery-operated
pumps equipped with sampling cartridges consisting of a 20rran i.d. x 10cm long
borosilicate glass tubes into which was fitted a 22mm dia. x 7.6cm long plug
of polyurethane foam plug. The PUF sampling cartridges were connected to the
pumps with a short section of Tygon tubing as shown in Figure 2. Sampling
was performed according to the procedure described in Appendix A.
The vents which were sampled are identified in Figure 1. Sampling was
performed by placing the PUF cartridges into the vent pipes and sealing the
vent openings with a plastic bag or tape to restrict gas flow. Nominally,
sampling was performed for 8 hours at a flow rate of 1.2 -1.4 L/min.
Ambient air sampling was performed with an array of samplers located as
shown in Figures 3 and 4. Figure 3 shows the sampler placement for the
ambient air monitoring conducted on January 26 from approximately 1000 to
r ,L
r
r
l
I
I
]
1
J
l
I
]
I
1
1 l
..
I I
SAMPLING CARTRIDGE
115V ADAPTOR/
CHARGER PLUG
LED INDICATOR LIGHTS
=-
(TIMING. FLOW, BATTERY) -.a...r--s:S::,P'!
TIMING SWITCHES
FLOW RANGE VALVE
DRIVE IELT
OFF-ON SWITCH
Figure 2. DuPont P-4000A pump and sampling cartridge.
=-
I J \
1· I 0 N
1· 8
] 0 0 0
]
]
7
1
]
]
2A Q 28 0
]
©
] . MRI
~. Weather
Station
] □ I
3 4A 48 4C
J
J
il i •
1 SA 5B SC
J •
~
6A 68 6C
Figure 3. Ambient air sampling locations on January 26, 1983.
-·-·
r
l
J
]
]
]
]
J 6A
J
J 68
J
J □ 3
J
J
J 6C
J
J
1 •
-l Figure 4 • •
..
I ..
:.
J I 0 N
0 0 0
SA
4A
2A
58 048 0 7
28
4C •-
SC
Ambient air sampling locations on January 29,
8
1983.
©
MRI
Weather
Station
-•
_·]
]
_]
_]
·_]
]
_J
LJ
~
~_l
J
J
L t
1700 hrs. EST. Two samplers were located upwind of the main vent; one on-
site approximately midway between the vent and the north fencelirie .(Location
7) and the other off-site (Location 8). Two samplers were located near the
main vent (Locations 2A and 28); one on each side at a distance of 1 meter.
An array of 12 samplers were located downwind of the main vent in
approximately a 90° quadrant. Samplers were placed at three locations (4A,
48, and 4C) approximately midway between the main vent and the south fence
line. Along the south fence line, samplers were placed at three locations
(SA, SB and SC). At each location, sampling was performed at 4 ft. and 15 ft.
above ground level. The sampling locations (6A, 68, and 6C) were off-site
approximately 200 meters from the main vent. ·one sampler was placed near the
residence which is approximately 1/2 mi. west of the l~ndfill. With exception
_of the downwind fence line points, ambient air sampling at all other
locations was performed at 4 feet above ground level. The sampling pumps were
operated at a nominal flow rate of 3.8 L/min.
' Sampler pla~ement for the ambient air monitoring conducted on January 29
is shown in Figure 4. Samplers were deployed in the same general pattern as
used on January 26, however a change in wind direction required a shift in
the specific sampling points as shown. Sampling on January 29 was performed
from approximately 0900 -1700 hrs. EST.
Night time ~am.pling was pdrformed on January 31 -February 1 using the
same sampling pattern and sampling locations as on January 29. Sampling was
started at approximately 2100 hrs., January 31, and was to be terminated at
approximately 0500 hrs., February 1. However, most of the DuPont pumps
failed after 2-3 hours of operation due to the low ambient temperature
(""3o 0F) and the high relative humidity (...,95%). Consequently, very few valid
samples were obtained.
PCB ANALYSIS
Analysis for PCBs in the PUF cartridges was performed according to the
procedure given in .Appendix B. The steps in the analysis procedure included;
1) Soxhlet extraction of the foam plugs with 5% ether in hexane; 2)
concentration of the extract to 1 ml and 3) determination of PCBs in the
]
'
]
l
]
l
~]
]
]
J
;
-1
l
.l
l
]
J
.J
1
ii
1
'
extr~ct by electron capture-gas chromatography using EPA Method 608(l).
Identification and quantification of Aroclor 1242 and 1260 in the samples was
performed by the technique described by Webb and McCall(2).
METEOROLOGICAL MEASUREMENTS
Continuous measurements of wind speed and wind direction were performed
during the field monitoring period with a MRI portable weather station. The
ambient temperature sensor on the unit•'did ·not ·furrction, thus continuous
ambient temperature data were not obtained. The weather station was located
east of the landfill in an unobstructed area. Ground level elevation at the
weather station location was approximately the same as the center of the
1 andf i 11 •
Ambient temperature, relative humidity, and barometric pressure readings
were taken approximately hourly during sampling periods.
1
I 1 .
I
1·
]
1
]
l
l
]
]
J
l
J
.,
i:,::;-": !:~i~ ~ .. -
1~;;~} :·
PCB MONITORING DATA
SECTION 4
RESULTS AND DISCUSSION
The results of the PCB monitoring at the Warren County landfill are
presented in Tables 1 through 6. Tables 1 through 3 give the concentrations
of Aroclor 1242 and 1260 (in nanograms/standard cubic meter*) measured in the
vent emissions on January 26, 27, and 28, respectively. The results show
that the main vent is the predomiriate source of PCB emission from the site.
During the study period, the average concentrations of Aroclor 1242 and 1260
. observed in the main vent emissions were 123 and 2 µg/scm, respectively. The
PCB emission rates from the main vent based on these average concentrations
and the average flow rate·(measured by EPA) are: 12 ng/sec of Aroclor 1242
and 0.19 ng/sec of Aroclor 1260 •. · PCB emissions from other vents on the site
were significantly lower than f~om the main vent.
The ambient air monitoring results obtained during the study are. shown
in Tables 4,5, and 6. On January 26, vent and ambient air monitoring were
performed concurrently. Thus, the ambient air monitoring data for this date
(Table 4) are probably not representative since the flow from the vents was
restricted during the sampling period.
The daytime ambient air monitoring data for January 29 is shown in Table
5. PCBs were not detected at any sampling location downwind of the main
vent. Aroclor 1260, at a concentration near the minimum detection limit was
found in one of the upwind samples (Location A-7).
* Standard conditions -2s0c, 76Cmn Hg
~ :-~ ~ . ~ ......, J.-J 1,-J L-1 1-J a..., 1--1 .__, I...J ..... ..... .... ...l ..... ~ ~ .._. ~ TABLE 1. VENT MONITORING RESULTS -WARREN COUNTY (NC) LANDFILL(a) Sampling Date January 26, 1983 Sampling Location Samellng Per1od1 Hr EST Sampling Avg Sampling Total Sample guant itf2 PCBs In PUF t ng Code and Oescr1pt1on 5tarE Enil Time, mlns. Rate, sec/min Volume, scm Aroclor -42 Aroclor 1260 V-A Ma1n Vent 1010 1700 410 1408 0.58 82,100 V-B Small Vent west of Ma1n Vent 1010 1701 411 1237 0.51 35 -V-C Upper Leachate Vent 1005 1701 416 1424 0.59 1270 V-0 Lower Leachate Vent 1005 1701 (b) (b) (b) NO V-E·Small Vent near Leachate Vents 1010 1701 411 1460 0.60 400 a) ND -PCBs were not detected 1n sample. M1n1mum detectable levels of Arochlor 1242 and Arochlor 1260 1n the cartridges are estjmated to be 10 ng and 15 ng, respectively. b) Sample pump malfunctioned dur1ng samp11ng period. 1200 NO 360 ND 780 PCB Cone. In Alr1 ng/scm Aroclor 1242 Aroclor 1260 141,552 2,069 69 <29 2,153 610 667 1300 /' ,,
,1; ~ ! ' .. : - --' --i i; ___ .._ t.-..-11 ~ '----4 ~ t..-J '---' TI\BLE 2. VENT MONITORING RESULTS -WI\IIREN COUIITY (NC) LANDFJLda) Sampling Date January 27. 1903 Sampling Location Samel lniJ Period, fir [Sf Sampl Ing Avg--Samp 1 i ng Total Sample guantltf PCBs In PUF1 ng Star[ £nil Aroclor 12~2 Aroclor 1260 Code and Description Time, mlns. Rate. sec/min Volume, scm V-A Main Vent 0900 1700 480 1359 0.65 76700 V-8 Small Vent West of Main Vent 0900 1700 400 "1-359 0.65 ND V-C Upper Leachate Vent 0900 1700 480 1410 0.68 1920 V-D Lower Leachate Vent 0900 1700 480 1330 0.64 NO V-E·Small Vent near Leachate Vents 0900 1700 400 1259 0.60 NO a) ND -PCBs were not detected In sample. Minimum detectable levels of Arochlor 1242 and Arochlor 1260 In the cartridges are estimated to be 10 ng and 15 ng, respectively. 1380 ND 320 51 ND PCB Cone. In Rir1 ng/scm Aroclor 12~2 Aroclor 12~0 118,000 2123 <15 <23 2824 471 <16 80 /' <17 <25 11
~i~l!' .... ~ .~'!':~:,,......., }l(-,::1 · . ......., .. __.. '(-.". ..., ---...., ...... ...... .... .... ....., .... TABLE 3. VENT MONITORING RESULTS -WARREN COUNTY (NC) LANDFILL(a) Sampling Date January 20, 1983 ... Sampling Location Samellng Perlod1 llr EST Start End Sampling Avg Sampling Total Sampll? 911,1nt Hy PCDs In PUF I ng Aroclor 1242 Aroclor 1260 Code and Description Time. mlns. Rate. scc/rnln Volume, scrn -V-A-1 Main Vent(b) 0900 1700 400 1254 0.60 69300 V-A-2 Main Vent(b) 0900 1700 400 1347 0.65 68700 -V-8 Small Vent West of Main Vent 0900 1700 480 1326 0.64 42 V-C Upper Leachate Vent 0900 1700 400 1388 0.67 510 V-0 Lower Leachate Vent 0900 1700 480 1324 0.64 61 V-E Small Vent Near Leachate Vent 0900 1700 400 1279 0.61 18 a) NO -PCBs were not detected In sample. Minimum detectable levels of Arochlor 1242 and Arochlor 1260 in the cartridges are estimated to be 10 ng and 15 ng, respectively. b) Co-located s&nplers. 1130 1150 NO 206 54 18 ...... ....... _. ..__. '--J t' PCB Cone. In Air1 ng/scm Aroclor 1242 Aroclor 1260 115,500 1883 ,.. 105,692 1769 66 <23 761 307 95 84 ,, 30 30
:1.iJiffl;:;=;i~f.,~;:; ..... , ................. , ........ w...-.-...-......................................... __. ~i,,~;t: ·:t:·:;,v1lf{!l:.1t'l~>1~}f '. ·. · · ·· · -"·•-·-·--·-TABLE 4 Sampling Locdtlon Sau~,J lny Sam~l 111~ rcrlod1 llr[ST Code and Description Height, 111 tart En-a--·· ---------------Bes.1~e_!1,a_!.!L!en.t 2A 1 m west of main ventf 1 4 1005 1700 2R-1 Im east of malq vent c 4 1005 1700 211-2 1 u1 c~st of ioaln vent c 4 1005 1700 On-Site llownwJ!!_~ 4A 49m from main vent I 230°f l 4 1005 1710 40-1 49,n fro,n aia In vent (I JOO" ~ 1 1010 1711 4B-2 49m from main vent I )80° 4 1010 1:1 4C 49111 from main vent I 140° 4 1005 Fencellne Downwind SA 98m from main vent 9 230° 4 1007 1700 SA 98m from main vent 9 230° 15 1000 1702 so 98m from main vent 9 180° 4 JOOS 1706 5B 911,n from nia In vent I 180° 15 1000 1708 SC 98111 fro,n Pld In vent I 140° 4 1010 1710 SC 98m from main vent I 140° 15 1000 1713 Off-Site Downwind 6A 200m from main vent I 230° 4 1014 1702 6B 200m from main vent 9 180° 4 ioo0 (b) 6C 200m from main vent 9 140° 4 1020 1708 A-3 House west of landfill 4 1025 1730 A-7 49m upwind of main vent I 360° 4 1000 1700 A-8 Offslte, upwind of main vent 4 1015 1702 9360° Alf!IENT AIR ffiNITORING RESULTS • WAIUIEN COUNTY (NC) LANDFILL (a) SAMPLING DATE -JMUARY 26, 1903 S,1111111 ln!J AviJ. Sau~1l In'] Total Sample QuantllY. PCDs In PUF, n~ Time, 11lns Rate, scc:/1111 n Volume, SC■ Aroi:lor l 24f-Aroc l o·r-liill 415 390!1 1.62 ND 20 415 3799 1.60 NO NO 415 3865 1.60 NO NO 425 3844 1.63 NO NO 421 3811 1.60 ND NO (b) (bl 1:1 ND NO (b) (b ND ND -413 3744 1.55 ND 110 422 3822 1.61 NO 80 418 3811 1.59 NO ND 428 3883 1.66 ND NO 420 3824 1.61 ND ND 433 3826 1.66 NO NO 408 3844 1.57 NO NO (b) (b) (b) NA NA 408 3904 I.S9 ND NO 425 3845 1.63 NO NO 420 3894 1.64 ND NO 407 3894 1.58 NO NO •• -· -.----r-w-.-.-........................... ._ ----·•••• •••• •••••·••-•·•·&r•·r• •·• ... .-.-.. , ... ..._.,.,._, -•• • • -.-..-•·• --•• •· (1) ND -PCBs were not detected 1n sauple. Hlntmum detectable levels of Aroclor 1242 and Aroclor 1260 In the PUF cartridges are estimated to be !Ong and 15 ng, respectively. NA -PUF cartrldqe was not analyzed for PCBs. (b) Sample pump nialfunctloned during sa~llng period. (cl Co-located samplers. ,. ~CB _co.n\ ln .. M.r., .ri71.scrr,21 _ roclor 242 Aroc or 1 60 ·6 12 <6 <10 <6 <10 <6 <10 <6 <10 <6 71 <6 so <6 <10 <6 <10 <6 <10 <6 <10 <6 <10 --<6 <10 <6 <10 <6 <10 <6 <10
llwl-;sij·::1.,.,t.a .. 1. -1~ -... ----·-·• i,!I . ~;~,tr•i ~-!ft • I •~! ,i~1; •;~.r ••• ( /,,!~ ., 1(dic,,J,l!~l -__ !; ~ ('. ,.;'11,,J,i ,, l,:;,,\;lt , . \tj"j,;',:'' . ·: -_' <,°/tf~tf:: ~:: :' 'fi,.; _.., . . . . _,...... _. ... -.,'"1 ... , t'I TABLE 5 ---__ ,.,.,,.... ••"1'1rs··""' .,.,,...,,.. ....,. .. ,. --AtlBIENT AIR l«lNITORING RESULTS -WARREN COUNTY (NC) LANDFILL(a) SAMPLING DATE -JANUARY 29, 1983 ., ...... ··-·--·· .. -.. ..,._ ... .._.., _... •----•--.--.-._.., ........ ..,.........,..~-~--,,_.....,._. ....... _ __,_., ____ ._,,..,,,._:r■ --•----•• --Sampling Location Sampling Sa~ling Perlod1 HrEST Sampl Ing Avg. Sall'Clllng Total Sall'C)le iuantiti PC8s In PUF1 nf Code and Oescriptlon Height, 11 tart End Time, mins Rate, sec/min Volume, scm roclor 1242 Aroclor 260 Beside Hain Vent 2A lm·north of main venti J 4 0905 1713 488 3725 28-1 lm south of main vent~ l 0905 1715 490 3830 28-2 lm south of main vent 0905 1714 489 3818 On-Site Downwind 4A 43m from main vent I 310°1 l l 0905 1712 487 3685 48-1 33m fr0111 main vent I 270° c 0905 1110 485 3735 48-2 33m from main vent I 27D0 c 0905 17D1 482 3753 4C · 40m fr0111 main veni I 225° 4 09D5 1705 480 3722 Fencellne Downwind SA 85m from main vent I 310° 4 0905 1715 490-3770 SA 85m from main vent I 310° 15 0905 1717 492 3853 58 66m from main vent I 270° 4 0905 1708 483 3773 58 66m from main vent I 270° 15 D905 1711 486 3870 SC 8Dm from main vent I 225° 4 0905 1705 480 3792 SC 80m from main vent I 225° 15 0905 1707 482 3778 Off-Site Downwind 6A 134m from main vent I 280° 4 0905 1705 480 3752 68 132m from main vent I 270° 4 0905 1709 484 3823 6C J52m from main vent I 240° 4 0905 171 l 488 3839 A-3 House west of landfill 4 0930 1705 455 3366 A-7 On-site, 32m upwind of main 4 0905 1708 483 3790 vent I 100° A-8 Offslte, 74m upwind of main 4 0905 (b) (b) (b) vent i 100° ---~---· .... ---·T'··----·---------~-.,------·--------········-· .................. ._ ...... -............... _ ........... ~---~--(e) ND -PCBs were not detected In sanple. Hlnlll'llm detectable levels of Aroclor 1242 and Aroclor 1260 In the PUF cartridges are estimated to be lOng and 15ng, respectively. NA -PUF cartridge was not analyzed for PC8s. (b) Sample pump malfunctioned during sampling period. (c) Co-located samplers. 1.82 NO NO 1.88 NO ND 1.87 NO NO 1.79 NO ND 1.81 NO ND 1.81 NO ND 1.79 ND ND 1.85 ND ND 1.90 ND ND 1.82 NO ND 1.88 ND ND 1.82 ND ND 1.82 ND ND 1.80 ND ND 1.85 ND ND 1.87 ND ND 1.53 ND ND 1.83 ND 20 (b) NA NA ------......... /" ----■--·---.-.-........................... PCB Cone. In Alr1 ng/scm Aroclor 1242 Aroclor 1260 <6 <10 <6 <10 <6 <10 <6 <10 <6 <IO <6 <ID <6 <10 <6 <10 <6 <10 <6 <10 <6 <10 <6 <10 <6 <10 <6 <10 <6 <10 <6 <10 <6 <10 ND 11 -.., •••• , ...... .,1.
•••
.~ .. ~ ~ ~ ~ ._. ~ ~ .._.. TADLE 6 ~ .... ..... i....t ~ ......e PCO HONITOIIING RESULTS -WIIRRlN COUNTY {NC) ll\NOFILL (a) SAMPLING DATE -JANUARY 31 -FEDRUARY 1, 1983 -.II ...... ~-~; \....Al. '--4. 1...-.. ··•· ......... -···---··· ,
., '' ..
........
........ , •• It •• ,' ..... ,,
.
.
.......
'
............
.
. ~ •• "'•·r••·····-··· .... ·--·····•t"t ·•·r· , __ ....... ,...._ .............. _., ... _________ ----··· ••....• I •••• , I. I OT,_ Sampl Ing Location Samp11 ng Samslinl Period, HrEST Sa11111l Ing Avg. Sampling Code and Description Height, 11 tar End Time, 1111 ns Rate, sec/min Beside Hain Vent 211 lm north of main vent 4 0117 (2/1) 0530 (2/1) 253 3896 On-Site Downwind 4A 43m from main vent 9 310° 4 2100 !1/31) 0530 (2/1) 510 3998 48 33m from main vent 9 270° 4 0120 2111 0530 (2/1 ! 250 3882 4C 40m from ina1n vent 9 225• 4 0120 2/1 0530 (2/1 250 3944 "···-----· .... _ _., .... _____ _. ...... _._ .. ._. __________________ ,... ___ .... --~-~---·--.... -(a) NO -PCBs were not detected In sample. Hin1nun detectable levels of Aroclor 1242 and Aroclor 1260 in the PUF cartridges are estimated to be !Ong and 15ng, respectively. NA -PUF cartridge was not analyzed for PCBs. (b) low flow indicated during sampling period; sample volume questionable. Total Sample iuantitt PCBs in PUF1 nf PCB Cone. in A1r1 n~m Volume, scm roclor 1242 Aroclor260 Aroclor 1242 Aroclor -m 0.99 NO NO <10 <15 (b) NO NO <5 <8 2.04(bl 0.97(b NO NO <10 <15 0.99 NO NO <10 <15 ,.
J
J
l
I
I
I
i
t
I
I
I
J
The limited data obtained during nighttime sampling on January_ 31 -
February 1 is shown in Table 6. PCBs were not detected in any ambient air
samples. However, most of the data are questionable because of malfunction
of the DuPont pumps due to the low temperature and high humidity conditions~
METEOROLOGICAL DATA
The results of wind speed, wind direction, ambient temperature, relative
humidity, and barometric pressure measurements performed at the Warren County
landfill during the period January 26 -February 1, 1983 are presented in
Appendix C.
VENT FLOW RATE MEASUREMENTS
Gas flow rates from the vents on the landfill site were measured by
EMSL/EPA/RTP personnel on March 2, 1982. The measurements were performed by
sealing the vents and determining the volumetric f1ow of the exit gas with a
bubble meter. Flow data for the main vent are given below. No flow was
detected from the other vents on the landfill.
Time
0945 hrs.
1200 hrs.
1500 hrs.
I
Main Vent Exit Gas Flow
F1cpw, sec/min
(25 C, 760mn Hg)
4854
6000
6400
Avg. 5751
]
1·
I.
I
J
J
I
,)
I
1)
I
I
I
I
'
DISPERSION MODELING
In order to obtain confirmation of the ambient air concentrations
measured during the field monitoring, standard dispersion models were used to
calculate downwind concentration of PCBs using emission parameters and
meteorological conditions that prevailed during the field monitoring program.
Two EPA UNAMAP models, PTPLU and PTDIS, respectively, were employed to
calculate 1) estimates of maximum hourly concentrations under a full spectrum
of meteorological conditions and 2) estimates of the range of hourly ambient
concentrations that would occur at down wind di -stances of 50TI, lOOTI, and 150TI
under the meteorological conditions that probably controlled dispersion
during the field monitoring on January 29.
The following par·ameters were either used explicitly in the modeling or
provided guidelines from which maximum and minimum limiting concentrations
could be calculated.
PCB emission rate from main vent* --12.1 ng/sec
Main vent gas exit velocity** --0.012 m/s
Vent gas temperature (estimated) --288.2K (15°c)
Vent diameter --0.102m (4 inches)
Height of vent above ground --i/.2m
Observed ambient temperature during
monitoring (1/29/83) --minimum= 274.2°K (1°c)
maximum= 286.2°K (13°c)
8-hr. average= 282.2°K (9°c)
Observed wind speed during
monitoring (1/29/83) --minimum= 0.72 m/s (1.6 mph)
maximum= 2.3 m/s (5.1 mph)
8-hr. average= 1.6 m/s (3.5 mph)
* Average of emission measurements made on January 26, 27, and 28.
'k* Average of flow rate measurements made by EPA on 3/2/83.
Since the terrain sloped downward from the location of the vent pipe to
the ambient monitoring locations, each model was run for two scenarios to
bracket the expected actual concentrations. One scenario treated the vent as
standing 1.2 meters above a flat terrain while the second scenario placed the
vent exit at a height of 5.2 meters above a flat terrain.
The one-hour concentration predictions of the model were converted to
eight-hour averages by multiplying by a factor of 0.6. This factor was
selected after a review of the EPA publication, Workbook of Atmospheric
Dispersion Estimates (AP-26) by D.S. Turner (pp 37-38).
The maximum one-hour PCB concentrations predicted by the PTPLU model
under the two scenarios are shown in Table 7.
TABLE 7 PREDICTED MAXIMUM ONE-HOUR DOWNWIND PCB CONCENTRATIONS
FOR VARIOUS VENT HEIGHTS
Scenario
Vent 1.2m
above flat
terrain
Vent 5.2m
above flat
terrain
Maximum
1-hr. Conce~tration
(ng/m)
4.0 X 10-G
1.4 X 10-7
Distance
to Maximum
Concentration,m
14
75
Conditions Producing
Maximum Concentrations
Wind Speed Atmospheric
(mps) Stability Class
0.3 4 (neutral)
0.5 4 (neutral)
Using the PTDIS model, the average wind speed observed during the
monitoring period on January 29, Class 4 stability, and the 0.6 conversion
factor, the estimated ranges for 8-hour average ambient concentrations at the
three downwind distances calculated. The results are shown in Table 8.
\
]
J
11-
l
]
I
I
]
J
I
I
I
\
TABLE 8 PREDICTED 8-HR. DOWNWIND PCB CONCENTRATIONS FOR JANUARY 29, 1983.
Distance
Downwind from
the Vent (m)
50
100
150
Range of 8-Hour
PCB Ambient Co~centration
(ng/m)
1.5 -8.0 X 10-B
1 -8 2.0 -2.5 X 0
-8 1.25 -1.35 X 10
The concentration range limits were taken from the model output for the
two vent height scenarios.
A wider concentration range estimates can be obtained by using the
maximum and minimum one-hour concentrations calculated for the two scenarios
under all combinations of the three wind speeds (minimum, maximum, and
average) and the six stability classes. The concentration limits predicted
by the model for the three downwind points for these conditions are shown in
Table 9.
TABLE 9 RANGg OF ONE-HOU~ CONCENTRATIONS FOR VARIOUS WIND SPEEDS
AND ALL STABILITY CLASSES
Monitor Distance Minimum 1-Hour Maximum 1-Hour
Downwind from PCB Concentration PCB Concentration
Vent (m) (ng/m3) (ng/m3)
50 1.5 X 10-lQ 4.5 X 10-7
100 7.0 X 10-9 2.0 X 10-7
150 3.5 X lO_g 1.0 X 10-7
]
: r
l
\
For each downwind location the conditions which yielded the maximum 1-
hour PCB concentration were; a 1.2m vent height, a wind speed of 1.6 m/s and
Class 6 (very stable) stability. Under this set of conditions the PTPLU
model predicted that the maximum PCB concentration would be 4.8 x 10-7 ng/m3
and would occur at 37m downwind of the vent.
In summary, the dispersion models predict that downwind PCB levels under
prevailing and worst case meteorological conditions should be significantly
lower than concentrations that could be detected by the monitoring techniques
employed in this study. Thus, the monitoring data for January 29 are
consistent with the modeling predictions in that PCBs were not detected in
any downwind ambient air samples.
1
I
I -
I
I
I
I
I
I
I
I
I
I
I
I
\
SECTION 5
QUALITY ASSURANCE DATA SUMMARY
PUF CARTRIDGE CLEAN-UP CHECKS
All PUF cartridges were pre-cleaned before being used for PCB sampling.
One cartridge from each batch of 20 clean cartridges was ~e-extracted and
analyzed for PCB contamination. The batch of cartridges was considered
acceptable for sampling if the PCB level in the check sample is <10 ng.
FLOW RATE CALIBRATIONS
The flow rate of the DuPont pumps was calibrated with a bubble meter
before and after each sampling period using a DuPont Calibrator system. The
flow rate calibration data are summarized in Table 10. Average flow rates
for the sampling period were calculated from the pre-and post-sampling
calibration data.
PERFORMANCE AUDIT
A flow rate audit of the DuPont sampling pumps used during the study was
performed by W.F. Barnard, EMSL/EPA/RTP.
QUALITY CONTROL SAMPLES
A set of 18 quality control samples consisting of PUF cartridges spiked
with various quantities of Aroclor 1242 and Aroclor 1260 were analyzed with
the vent and ambient air samples. The QC samples were prepared by BCL using
NBS/SRM 1581 (Aroclor 1242 and Aroclor 1260 in oils). Results of the
analysis of the QC samples are given in Table 11.
':..
l
\
I TABLE 10. DUPONT PUMP FLOW CALIBRATION DATA
I
I Date Pump Sampling Calibrated Flow Rate, sec/min · Avg Flow
S/N Location Before Sampling After Sampling Rate,scc/min
[ 1/26/83 A-083 V-A 1370 1446 1408
A-080 V-B 1230 1243 1237
f A-118 V-C 1370 1478 1424
4789 V-D 1397 (a)
A-062 V-E 1388 1531 1460
t A-089 2A 3748 4069 3908
A-121 2B-1 3710 3888 3799
A-088 2B-2 3722 4007 3865
l 4803 4A 3760 3927 3844
A-038 4B-1 3728 3894' 3811
A-032 4B-2 3752 (a)
' A-127 4C 3710 (a)
A-061 SA{4) 3705 3782 3744
A-125 SA(lS) 3736 3908 3822
t
4696 58{4) 3728 3894 3811
A-092 58(15) 3710 4056 3883
A-087 SC ( 4) 3748 3901 3825
A-037 SC(lS) 3751 3901 3826
A-126 6A 3788 3901 3845
5136 6B 3794 (a)
A-120 6C 3800 4007 3904
A-143 A3 I 3782 3908 3845
A-094 A7 3754 4034 3894
4779 AS 3801 3987 3894
1/27/83 A-118 V-A 1301 1416 1359
A-079 V-B 1306 1411 1359
A-062 V-C 1301 1519 1410
A-120 V-D 1276 1383 1330
9806 V-E 1209 1309 1259
1/28/83 9806 V-A-1 1206 1301 1254
A-062 V-A-2 1252 1441 1367
A-037 V-B 1245 1408 1326
A-118 V-C 1283 1492 1388
A-083 V-D 1242 1405 1324
5138 V-E 1242 1305 1274
a) Pump malfunctioned during sampling period.
11 \
I TABLE 10. DUPONT PUMP FLOW CALIBRATION DATA (Cont'd.)
11 Date Pump Sampling Calibrated Flow Rate, sec/min Avg Flow
I S/N Location Before Sampling After Sampling Rate,scc/mir:i
I 1/29/83 A-126 2A 3692 3758 3725
A.;037 2B-1 3710 3949 3830
A-079 28-2 3787 3848 3818
I 4803 4A 3681 3688 3685
A-088 48-1 3698 3771 3745
4779 48-2 3710 3795 3753
I A-092 4C 3692 3753 3722
A-080 5A(4) 3721 3820 3771
A-118 5A(15) 3704 4002 3853
I 9806 58(4) 3669 3877 · 3773
A-062 58(15) 3681 4058 3870
5138 5C(4) 3687 3896 3792
r A-083 5C(15) 3692 3864 3778
A-127 6A 3779 3724 3752
A-061 68 3768 3878 3823
I A-094 6C 3768 3910 3839
A-089 A3 3710 3022 3366
A-121 A7 3779 3802 3790
I A-032 AS 3687 (a)
1/31/83 5117 2A 3731 4060 3896
( 2/1/83 4696 4A I 3738 4258 3998
4789 48 3772 3992 3882
5942 4C 3743 4144 3944
I All other pumps used on this date malfunctioned during
sampling period.
11
I a) Pump malfunctioned during sampling period.
' l
ll
a.
1
l
]
I TABLE 11 QC SAMPLE ANALYSIS RESULTS(a)
I Sample No. Aroclor 1242 Aroclor 1260
I Added,ng Found,ng % Recovery Added,ng Found,ng % Recovery
I 1 60 64 107 0 37
6 60 52 87 0 32
I 5 600 680 113 0 ND
13 600 530 88 0 85
' 8 6000 3700 62 0 190
9 6000 3200 53 0 ND
10 0 ND 60 70 117
14 0 ND 60 180 300
4 0 ND 600 515 86
7 0 ND 500 600 100
2 0 ND 6000 4700 78
12 0 ND 6000 4000 67
15 120 90 75 60 120 200
17 120 85 71 60 70 117
11 400 260 65 200 170 85
16 400 460 115 200 260 130
3 4000 1820 46 2000 1530 77
18 4000 1320 33 2000 2950 148
a) ND -Not detected. Minimum detectable levels of Aroclor 1242 and
Aroclor 1266 are estimated to be l0ng and 15ng, respectively.
]
J.
).
I
I
I
I
I
' l
FIELO ,BLANKS
Eight field blanks were analyzed with the ambient air samples. The
blanks were PUF cartridges that had been carried through all field operations
except sampling. PCBs were not detected in any of the blanks above the
minimum detectable level i.e. 10 ng for Arochlor 1242 and 15 ng for Aroclor
1260.
CO-LOCATED MONITORING
Co-located monitoring of the main vent was performed on January 28. (See
Table 3 for results). Concentrations determined from the co-located monitors
differ by 9% for the Aroclor 1242 and 6% for the Aroclor 1260. Co-located
ambient air monitoring data cannot be evaluated since PCB levels were below
minimum detectable levels in all paired samples.
-
f •
r
REFEHDiCES
1. Federal Register, Vol. 44~ No. 233, Prrcmday, Dec~r 3, 1979, Pgs. 69501-
69509.
2. Webb, "R.G. ·and McCall, A.C • ., •Quantitutive PCB Standards for Electron
Capture Gas Chromatography", Journal of Chromatogr.~ic Science, 11, Pgs.
366-373, July 1973. -
]
J
]
II
)
I
I
I
I
I
I
11
I
,I
,f
' ,,
l .
(
APPENDIX A
PROCEDURE FOR PCB SAMPLING WITH
DUPONT P-4OOOA PUMPS AND PUF CARTRIDGES
(1) Calibrate the flow rate of the DuPont pumps before sampling with a
DuPont Calibrator system.
(2) At the field site, place pumps at designated sampling locations. Record
pump S/N and corresponding sampling location I.D.
(3) Using latex glove·s, remove a clean PUF cartridge from its sample bottle,
carefully unwrap the aluminum foil from the cartridge. Fold aluminum
foil, replace in sample bottle, and tightly close the bottle cap.
Connect the PUF samp 1 i ng cartridge to the DuPont pump samp 1 i·ng in 1 et
using a short piece (12-18 in.) of Tygon tubing. (Note: Clean latex
gloves must .be worn _ at all times when handling the PUF cartridges).
(4) Using metal three-prong clamps that have been rinsed with ·s&J hexane,
mount the PUF cartridges orl the sampler support rod in a vertical
position with the inlet pointing downward. Record cartridge height
above ground.
(5) Turn pumps on and begin sampling period. Record starting clock time.
During the sampling period check pumps at least every 2 hours for proper
operation. Record any abnormal conditions.
(6) After sampling for the specified time, terminate sampling period by
turning pumps Dff. Record clock time that pump was turned off. Just
before turning pumps off, push test button on pump and check low flow
light and the elapsed time indicator lights. If low flow light comes
on, it indicates that a low flow condition existed during the sampling
\
]
I
]
I
I
I
I
I
I
I
l
t
\
period, e.g., Tygon tubing crimped, cartridge plugged, pump stopped,
etc). Record results of the low flow check. Record elapsed time from
the pump timer as a check on the clock time.
(.7) As soon as possible after termination of sampling, remove the PUF
cartridge from the Tygon sample line (using latex gloves), wrap
cartridge in its original aluminum foil wrapping, and place in the
original sample bottle. Cap tightly, label bottle with sampling data
and sample I .D. and seal the bottle cap with a strip· of "Evidence Tape".
(8) Re-calibrate the flow rate of the DuPont pumps after completion of
sampling.
I
I
[
r
' t
. -..
,
'
APPENDIX B
PROCEDURE FOR ANALYSIS OF PCBs
IN PUF CARTRIDGES
I. Equipment and Reagents Required for PUF Sample Extraction
1. Glass ware
500 ml boiling flasks
300 ml capacity Soxhlet extractors
3 ball condensers
500 ml Kuderna-Danish apparatus
15 ml receiver tubes
Snyder columns
Filter tubes (Corning 9480-32)
Pre-scored (1 ml, 5ml) amber glass vials with teflon-lined caps
9" long dispos~ble transfer (Pasteur) pipets
Wash all glassware with Alconox; rinse with deionized water,
acetone, hexane, and deionized water; then fire in kiln (500 C)
2. Equipment
Extraction Apparatus, Multi-Unit Heater (CMS 119-362)
Blunt-end forceps
Surgical tongs (approximately 12")
Steam bath
Nitrogen blow-down evaporator
Glass wool (Heater overnight at 350 C in muffle furnace)
Boiling granules (Heater overnight at 500 C in kiln)
Teflon wash bottles
]
l
l
I
r
f
t
\
3. Reagents
Burdick and Jackson, Distilled in Glass Solvents:
Acetone
Hexane
Ethyl Ether (Preserved with Ethanol)
Sodium Sulfate, 12-60 mesh, Anhydrous (Baker 5-3375) (Heated
overnight at 5oooc in kiln).
II. Sample Receipt and Extraction
1. Log samples in log book. Note any damage to sample or
irregularities (i.e., EPA chain of custody tape broken).
2. Prepare 5% ethyl ether in hexane. Prepare by case lot of hexane.
Remove 200 ml of hexane from freshly opened bottle and add 180 ml
of freshly opened ethyl ether (preserved with ethanol).
3. Rinse condenser towers with 5% ether/hexane.
4. Wipe off lab bench wit~ 5% ether/hexane.
5. Add 300 ml of 5% ether/hexane to 500 ml boiling flask. Add boiling
granules (no more than 3 granules).
6. Dim lights in laboratory before removing first sample. Rinse a
large sheet of aluminum foil with 5% ether/hexane. Be sure to use
waste rinse container. Place foil, rinsed side up, on lab bench.
Use this for forceps and tongs. Rinse forceps and tongs with 5%
ether/hexane.
7. Carefully remove sampling cartridge from jar-and unwrap aluminum
foil. Handle cartridge minimally, placing it on its own aluminum
foil wrapping.
8. Note in project log book any breakage or damage to sampling
cartridge.
]
]
]
I
I
I
I
\
9. With pre-rinsed forceps, carefully remove the foam plug (PUF) from
the sampling cartridge.
10. Place the PUF in the Soxhlet, and connect the Soxhlet to the 500 ml
boiling flask. (If hi-vol sample, also place corresponding
particulate filter in Soxhlet with PUF plug). Wet the joint with
5% ether/hexane. Place the forceps on _ the aluminum foil wrapping.
Label the boiling flask with sample I.D.
11. Taking the pre-rinsed tongs, adjust the PUF in the Soxhlet to wedge
it midway along the length of the siphon. Rinse the tongs into the
Soxhlet with the 5% ether/hexane. Rinse the forceps, glass
sampling cartridge, and aluminum foil wrapping with 5% ether/hexane
into the Soxhlet. Place the forceps and tongs on the aluminum foil
sheet. Dispose of the aluminum foil wrapping and place the glas
cartridge aside for washing and recycling.
12. Connect the Soxhlet to the condenser, wetting the glass joint with
5% ether/hexane for a good seal.
13. Repeat the process for the day's samples being sure to include a
solvent blank, field blank, and a control sample.
14. Check water flow to condenser towers, and turn on heating units.
15. As samples begin to boil, check Soxhlets making sure they are
filling and siphoning properly (4 cycles/hour). Allow samples to
cycle overnight or for a minimum of 16 hours.
16. Turn off heating units and allow samples to cool to room
temperature. Be sure the lights are dim.
17. Set up Kuderna-Danish (K-D) with receiver tubes. Add one boiling
granule to each set up. Label the K-D's with the sample I.D.
18. Pack filter tubes with glass wool and sodium sulfate. Place tube
in neck of K-D.
19. Carefully remove Soxhlet and boiling flask from condenser tower.
Drain remaining solvent into boiling flask.
I \
I
I
I
I
I
I
I
I
I
I
I
I III.
I
I
I
I
I
r t
-:..
20. Carefully pour sample through filter tube into K-D. Rinse boiling
flask 3 times with hexane. Swirling hexane along sides of boiling
flask. Once sample has drained, rinse down filter tube with
hexane.
21. Attach Snyder column to K-D and rinse Snyder column to wet joint.
22. Place K-D on steam bath and evaporate sample to approximately 5 ml.
Do not let sample go to dryness.
23. Remove sample from steam bath, rinsing Snyder column with a minimum
of hexane. Allow sample to cool.
24. Remove sample from K-0, making sure to label receiver tube.
25. Rinse nitrogen blow down spouts with hexane and place samples so as
to further concentrate. Transfer samples to pre-scored vials using
transfer pipets. Rinse receiver tube 3 times making a quantitative
transfer. Concentrate samples to 1 ml or per instruction from
analyst.
26. Make a master list of all samples prepared, date received, and
processed.: Give the l~st and sample extracts to the GC analyst.
GC Analysis (EPA Method 608)
1. Analyze samples using the following GC operating conditions.
Column: Supelcoport 100/120 mesh coated with
1.5% SP-2250/1.95% SP-2401 packed in
glass (180 cm x 4 mm ID)
Carrier:
Column
Temperature:
Detector:
5% methane/95% Argon at 60 ml/min
200 C, isothermal
ECO
2. Calibrate the system daily with a minimum of three injections of
calibration standards which hav~ been referenced to NBS/SRM 1581
(Aroclor 1242 in oils)
3. Inject 2-5 µl of the sample extract using the solvent-flush
technique. Smaller (1.0 µL) volumes can be injected if automatic
devices are employed. Record the volume injected to the nearest
0.05 µland the resulting peak size, in area units.
4. If the peak area exceeds the linear range of the system, dilute the
extract and reanalyze.
IV. Quality Control (QC)
1. Analyze one laboratory blank per each batch of 20 samples.·
2. Analyze one laboratory spike per each batch of 20 samples.
]
]
]
1
I
I
I
I
I
I
I .
APPENDIX C
METEOROLOGICAL DATA
.,_
The results of the meteorological measurements performed at the Warren
County Landfill during the PCB monitoring period are given in Tables C-1
through C-6.
l
J _·
],
J
I. I
'1
I
I
I
I
(
I
.
-:..
TABLE C-1. METEOROLOGICAL DATA FOR FIELD MONITORING PERioo(a)
Date: January 26, 1983
Time Wind Speed Wind Direction Ambient Rel. Humidity Bar. Press.,
hrs. EDT mph Deg. (Compass) Temp, °F % in Hg
0000-0100
0100-0200
0200-0300
0300-0400
0400-0500
0500-0600
0600-0700
0700-0800
0800-0900
0900-1000
1000-1100
1100-1200 58 29.95
1200-1300
1300-1400 46 29.90
1400-1500 3.3 150
1500-1600 3.0 120 48 29.89
1600-1700 2.8 135 49 29.89
1700-1800 1.6 150
1800-1900 0.1 150
1900-2000 0.7 80
2000-2100 0.2 105
2100-2200 1.3 95
2200-2300 1.9 75
2300-2400 1.9 60
a) Ambient temperature, relative humidity, and barometric pressure data are
single readings taken during the time period. Wind speed and direction values
are hourly averages calculated from continuous monitoring data.
•
1
1
1
1
I
I
I
I
I
I
I
TABLE C-2. METEOROLOGICAL DATA FOR FIELD MONITORING PERIOD~a)
Date: January 27, 1983
Time Wind Speed Wind Direction Ambienl Rel. Humidity Bar. Press.,
hrs. EDT mph Deg. ( Compass) Temp, F % in Hg
0000-0100 1.7 90
0100-0200 0.8 90
0200-0300 1.8 15
0300-0400 1. 7 30
0400-0500 1.3 30
0500-0600 1.9 15
0600-0700 2.6 30
0700-0800 2.7 90
0800-0900 2.8 75
0900-1000 4.1 45 43 88 29.93
1000-1100 6.9 45 44 81 29.93
1100-1200 8.5 60 47 66 29.90
1200-1300 8.2 60 45 71 29.87
1300-1400 9.0 60 46 68 29.86
1400-1500 6.8 60 46 65 29.86
1500-1600 8.0 60 45 67 29.86
1600-1700 8.7 45 44 78 29.84
1700-1800 7.8 45
1800-1900 9.3 45
1900-2000 7.8 45
2000-2100 8.5 45
2100-2200 11.1 45
2200-2300 9.4 45
2300-2400 9.2 45
a) Ambient temperature, relative humidity, and barometric pressure data are
single readings taken during the time period. Wind speed and direction values
are hourly averages calculated from continuous monitoring data •.
\
J
.1. ..
l
1
1
'I
I
I
I
I
I
.,_
\
TABLE C-3. METEOROLOGICAL DATA FOR FIELD MONITORING PERIOD~a)
Date: January 28, 1983
Time Wind Speed Wind Direction Ambieit Rel. Humidity Bar. Press.,
hrs. EDT mph Deg. (Compass) Temp, F % in Hg
0000-0100 10.9 45
0100-0200 12.7 45
0200-0300 12.9 45
0300-0400 12.9 45
0400-0500 12.6 45
0500-0600 11.4 30
0600-0700 12.0 30
0700-0800 11.5 30
0800-0900 10.4 25
0900-1000 9.5 15 37 87 29.82
1000-1100 10.3 360 38 83 29.84
1100-1200 10.9 15 40 75 29.85
1200-1300 11.9 15 46 64 29.82
1300-1400 11.6 360 49 54 29.80
1400-1500 11.5 360 50 49 29.80
1500-1600 10.9 366 50 43 29.80
1600-1700 7.5 360 48 43 29.81
1700-1800 14.8 360
1800-1900 2.3 360
1900".'2000 1.5 315
2000-2100 1.2 300
2100-2200 1. 7 30
2200-2300 1.1 30
2300-2400 0.8 50
a) Ambient temperature, relative humidity, and barometric pressure data are
single readings taken during the time period. Wind speed and direction values
are hourly averages calculated from continuous monitoring data.
r-
]
J
]
]
)
I
I
I
I
I
I
I
I
l
scmxaF· ·x rt a . ..; .. .. ---. . ~.' . .. I-·.:a · 1a--~-1 ·I ·i• · -·~·~r■-a•· 1··~--.... ·:i -i
\
TABLE C-4. METEOROLOGICAL DATA FOR FIELD MONITORING PERIOD{a)
Date: January 29, 1983
Time Wind Speed Wind Direction Ambie'1,t Rel. Humidity Bar. Press.,
hrs. EDT mph Deg. {Compass) Temp, F % in Hg
0000-0100 1.2 345
0100-0200 1.2 315
0200-0300 1.3 300
0300-0400 {b) {b)
0400-0500 {b) {b)
0500-0600 {b) {b)
0600-0700 {b) {b)
0700-0800 {b) {b)
0800-0900 {b) {b)
0900-1000 {b) {b) 34 86 29.99
1000-1100 1.6 90 45 57 29.99
1100-1200 2.1 180 49 44 29.97
1200-1300 3.6 150 55 33 29.92
1300-1400 3.5 180 56 32 29.89
1400-1500 3.9 18~ 55 39 29.86
1500-1600 5.0 195 54 42 29.84
1600-1700 5.1 195 50 49 29.84
1700-1800 3.7 180
1800-1900 2.3 180
1900-2000 2.2 180
2000-2100 1.8 195
2100-2200 1.4 210
2200-2300 1.6 210
2300-2400 0.4 180
a) Ambient temperature, relative humidity, and barometric pressure data are
single readings taken during the time period. Wind speed and direction values
are hourly averages calculated from continuous monitoring data.
b) Data missing due to instrument malfunction.
1·
1 .·
. ].
I
11
I I
I t t
. \
TABLE C-5 • METEOROLOGICAL DATA FOR FIELD MONITORING PERIOD{a)
Date: January 31, 1983
Time Wind Speed Wind Direction Ambient Rel. Humidity Bar. Press.,
hrs. EDT mph Deg. ( Compass) Temp, 0f % in Hg
0000-0100 1.1 330
0100-0200 1.1 345
0200-0300 0.5 355
0300-0400 1.6 315
0400-0500 0.9 345
0500-0600 0.7 330
0600-0700 0.3 345
0700-0800 0.7 325
0800-0900 0.3 315
0900-1000 2.2 360
1000-1100 5.1 5
1100-1200 4.8 5
1200-1300 4.5 360
1300-1400 5.3 300
1400-1500 3.5 345
1500-1600 3.0 2s1s
1600-1700 2.5 270
1700-1800 0.3 270
1800-1900 0.8 240
·1900~2000 1.2 360
2000-2100 1.0 180
2100-2200 0.6 240 40 87 29.84
2200-2300 0.3 165 39 92 29.84
2300-2400 0.9 180 36 95 29.84
a) Ambient temperature, relative humidity, and barometric pressure data are
single readings taken during the time period. Wind speed and direction values
are hourly averages calculated from continuous monitoring data.
Time Wind Speed Wind Direction Ambie'bt Rel. Humidity Bar. Press.,
hrs. EDT mph Deg. (Compass) Temp, F % in Hg
0000-0100
0100-0200
0200-0300
0300-0400
0400-0500
0500-0600
0600-0700
0700-0800
0800-0900
0900-1000
1000-1100
1100-1200
1200-1300
1300-1400
1400-1500
1500-1600
1600-1700
1700-1800
1800-1900
1900-2000
2000-2100
2100-2200
2200-2300
2300-2400
0.4
0.7
1.0
1.0
0.5
0.5
150
195
180
360
5
5
36 91 29.83
32 95 29.84
31 89 29.84
30 29.91
a) Ambient temperature, relative humidity, and barometric pressure data are
single readings taken during the time period. Wind speed and direction values
are hourly averages calculated from continuous monitoring data.
• I I