Environmental Monitoring Series
METHODOLOGY FOR MEASUREMENT OF
POLYCHLORINATED BIPHENYLS IN
AMBIENT AIR AND STATIONARY
SOURCES: A Review
Environmental Monitoring and Support Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL MONITORING series.
This series describes research conducted to develop new or improved methods
and instrumentation for the identification and quantification of environmental
•"Mutants at the lowest conceivably significant concentrations. It also includes
studies to determine the ambient concentrations of pollutants in the environment
and/or the variance of pollutants as a function of time or meteorological factors.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
METHODOLOGY FOR MEASUREMENT OF POLYCHLORINATED BIPHENYLS
IN AMBIENT AIR AND STATIONARY SOURCES' - A REVIEW
by
John H. Margeson
Quality Assurance Branch
Environmental Monitoring and Support Laboratory
Research Triangle Park, North Carolina 27711
ENVIRONMENTAL MONITORING AND SUPPORT LABORATORY
QUALITY ASSURANCE BRANCH
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NORTH CAROLINA 27711
AUGUST 1976
-------
ABSTRACT
The state of development of methodology for measurement of polychlorin-
ated biphenyls (PCBs) in ambient air and stationary sources was reviewed.
The most promising method for ambient air measurements involves
collection of PCBs on polyurethane foam, extraction with an organic solvent,
removal of interferences by column chromatography, and confirmation and
analysis by electron-capture gas chromatography. Quantitation by perchlorin-
ation of PCBs to decachlarobiphenyl (DCB) is the most promising quantitation
technique, but the procedure has not yet been perfected to the point where all
PCBs are quantitatively converted to DCB. Perfection of this technique should
allow for significant improvement in the quality of ambient PCB data being
generated.
Source and ambient methods differ mainly in sampling. Work on method-
ology for stationary sources is in the early stages of development and further
investigations are needed.
111
-------
CONTENTS
Abstract iii
List of Figures v
Acknowledgments vi
Background vii
1. Introduction 1
2. Physical and Chemical Properties 2
3. Ambient Air Methodology 4
Liquid Absorption Methods 4
Liquid Phases on Solid Supports 5
Solid Adsorbents 7
4. Sample Preparation 10
5. Qualitative and Quantitative Analyses 14
Comparison 15
Analyzed Aroclor Standards 17
Other Techniques 19
6. Stationary Sources 21
References 23
iv
-------
LIST OF FIGURES
Number Page
1 Assembled sampler and shelter with exploded view of
the filter holder 9
2 Aroclor 1254 (From Reference 23) 13
3 Comparison of electron capture chromatograms for Aroclors
1221, 1242, 1248, 1254, and 1260 17
4 Chromatogram of Perch!orinated PCB Mixture 20
-------
ACKNOWLEDGMENTS
The author would like to thank Mr. Merrill D. Jackson and
Dr. Robert G. Lewis of EPA's Health Effects Research Laboratory and
Dr. William J. Mitchell of EPA's Environmental Monitoring and Support
Laboratory for helpful discussions during the preparation of this
document. The author would also like to thank Mr. Don Lokey of EPA's
Strategies and Air Standards Division for making available the document
cited in reference 3 Environmental Assessment of PCB's in the Atmosphere,
prepared by the Mitre Corporation. The section on methods proved par-
ticularly useful and resulted in uncovering a number of helpful references,
Several passages of descriptive material on sampling procedure were used
directly as they appeared in the document.
vi
-------
SECTION 1
INTRODUCTION
Polychlorinated biphenyls (PCBs) were first introduced into the environ-
ment in 1929. ' Interest in PCBs accelerated in 1966 with the discovery of
their widespread occurrence in Sweden/ ' This discovery led to the recog-
nition that in many cases, environmental damage previously blamed on organ-
ochlorine pesticides (e.g., DDT) were actually due to PCBs. PCBs are
released to the environment by a number of pathways, including: incomplete
combustion in incinerators; vaporization from paints, plasticizers, and
coatings; and evaporation and distillation from bodies of water. In the
ambient air, PCBs exist in the vapor state as liquid aerosols and adsorbed
onto particulate matter.
The toxicity of PCBs is of particular concern because they are persistent
in the environment. ' PCBs, which are lipophilic, have been found in the re-
mains of wildlife^ and in human adipose tissue, ' and have been implicated
(3)
as carcinogens in laboratory studies in mammals. '
PCBs are manufactured in many countries. In the United States the sole
producer, the Monsanto Company, markets them under the trade name Aroclor.
Monsanto sold 73,100,000 pounds of Aroclor in the United States in 1970,
but in December of that year, the company voluntarily restricted sales to
closed-system applications.
Ambient concentrations are generally in the nanogram-per-cubic-meter
3
range; a concentration of 9.4 ng/m was measured in Providence, R. I.
-------
Measurement of PCBs emitted from stationary sources is just beginning; there-
fore, the state of the art in this area is changing rapidly.
Recognizing the potentially harmful effects of PCBs on humans, the
American Conference of Government Industrial Hygienists has set standards for
workplace exposure. Since PCBs released into the environment remain there
for a considerable length of time and have a high potential for being intro-
;
duced into the food chain, reliable methodology must be developed to identify
sources that release PCBs and to measure and control the amounts to which
humans are exposed. Methodology for measurement of ambient and stationary
source concentrations of PCBs must include the capability to quantitatively
sample particulate and vaporous forms at the nanogram- and microgram-per-cubic-
meter levels, respectively. Attention must be paid to removal of interferents
from samples taken from the ambient air and from stationary sources.
Analytical sensitivity obtained by use of electron-capture detection with
gas chromatography has been more than adequate. This technique has been used
for analysis of trace amounts of pesticides in soil and water/ ' Improvements
in the quantitation technique are needed, however.
This report reviews the state of the art of methodology for measuring
PCBs in ambient air and from stationary sources (with emphasis on the former).
Four main areas are covered: (1) physical and chemical properties of PCBs,
(2) sample preparation by extraction of PCBs from the collected sample and
removal of interferents, (3) analysis, and (4) quantitation.
-------
SECTION 2
PHYSICAL AND CHEMICAL PROPERTIES
Aroclors are prepared by catalytic chlorination of biphenyl with
(81
anhydrous chlorine.v ' This process produces a complex mixture of sub-
stitution products ranging in composition from one to ten chlorine atoms
fo\
per biphenyl molecule. 209 compounds are theoretically possible. ' The
compounds are named as derivatives of biphenyl:
3 2 2' 3'
\_K_V
6'
The Aroclors are characterized by a four-digit number. The first two digits
represent the type of molecule, 12 = chlorinated biphenyl (54 = chlorinated
terphenyl), and the last two digits represent the average weight percent of
lo\
chlorine in the mixture/ ' Thus, Aroclor 1242 is a mixture of chlorinated
biphenyls with an average of 42 % chlorine. (Aroclor 1016, a chlorinated
biphenyl with 42 % chlorine, deviates from this nomenclature.)
The environmentally important physical and chemical properties of the
Aroclors warrant discussion. All Aroclors are characterized by low water
solubility, the solubility decreasing with increasing chlorine content.
fo\
Aroclor 1242 has a reported water solubility of 200 ppb. ' Other workers
-------
( 52}
have found considerably lower values for Aroclor 1254. ' (Aroclors are
soluble in a number of organic solvents.) Vapor pressures of the more
volatile (those with lower degree of substitution) Aroclors are of the order
of 10"3 to 10~5 mm of Hg over the range 20-40°C (68-104°F) . The rate of
vaporization of PCBs from solid surfaces decreases as the extent of halo-
(53)
genation increases. ' The densities of Aroclors are much greater than
that of water.
Chemically the Aroclors are quite stable. They are resistant to
oxidation and change on exposure to acid or alkaline conditions at ambient
temperatures. The Aroclors do undergo reductive dechlori nation in the
atmosphere from exposure to ultraviolet radiation. This property is
important in analyzing environmental samples for PCBs and is discussed
below.
-------
SECTION 3
AMBIENT AIR METHODOLOGY
SAMPLING. The sampling procedures that have been used fall into three
general categories: (1) liquid absorption, (2) liquids on solid supports,
and (3) solid adsorbents.
Liquid Absorption
Liquid absorption methods usually involve the use of some type of
impinger or fritted-glass gas bubbler in conjunction with a vacuum pump.
One procedure utilizing a Greenburg-Smith Impinger, consists of drawing air
by vacuum pump through a trapping medium of ethylene glycol at a rate of
(0)
up to 30 liters per minute.v/ A collection efficiency of 75 to 83 % has
been reported/ ' Toluene/ ' hexane/ ' and hexylene glycor ' (for
chlorinated pesticides only), have also been used to collect PCBs.
EPA has used the ethylene glycol procedure to collect pesticides and
PCBs. ' The procedure consists of four Greenberg-Smith impingers each con-
taining 100 milliliters of ethylene glycol. A glass fiber filter is used
ahead of the impingers to trap particulate matter. Two impingers are operated
in series for 12 hours and the other two for an additional 12 hours. A column
of activated alumina has been used after the second impinger to trap pesti-
(14)
cides not retained by the ethylene glycol/ ' and it would presumably aid in
collecting PCBs.
Fritted glass bubblers with toluene as the solvent have been used
(3V
to collect PCBs.v ' This method is recommended by the American National
-------
Standards Institute.
Because of the ubiquitous nature of PCBs, glassware, solvents and
other equipment must be scrupuously cleaned.^ * '
Several limitations are inherent in liquid absorption systems. The
most important of these is the collection of a sufficient volume of air. The
highest airflow rate used is about 30 liters per minute; therefore, a 24-hour
sampling period results in the collection of PCBs from a total of only 43
(Q\
cubic meters of air.v ' The detection limit of electron capture detectors is
about 50 picograms per microliter.^ ' Assuming a sample volume (for analysis)
of about 10 ml and a 50 cubic meter air volume, this works out to a lower limit
of detection (LDL) of 10 ng/m3 - if collection efficiency for PCBs is 100%.
Presumably, the LDL could be lowered by extending the sampling time; however,
the more volatile and or less soluble PCBs may be lost from the impinger.
Liquid absorption methods appear suitable for sampling at sites where
relatively high concentrations are expected, such as near capacitor manu-
facturing plants and low temperature incinerator stacks.
Liquid Phases on Solid Supports.
These sampling procedures fall into two categories: (1) those that are
used in conjunction with a vacuum pump that include measurement of the air
volume sampled and therefore the concentration of PCB sampled and (2) those
employing static samplers where there is no measurement of air volume.
Static samplers, such as Nylon nets coated with a silicone oil, '
are useful for'qualitative determination of PCBs, but are not useful for
monitoring PCB concentrations.
In the other category of samplers, ceramic saddles coated with a
silicone oil'17' and Florisil - 5% glycerine^ ' have been used to sample
6
-------
PCBs. These systems allow higher airflow rates than do the liquid - impinger
samplers because of the porosity of the solid support. A lower limit of
detection of 2 ng/m , using electron capture gas chromatography, is claimed
(18)
for the latter procedure/ ' With the ceramic saddle - silicone oil system
a glass fiber filter was inserted in front of the coated saddles to trap
3
suspended particulate matter; a flow rate of 0.6 m /min and a collection
efficiency of 70% (for Aroclor 1254) were reported. It was expected that
PCBs in the vapor and liquid aerosol phase would easily pass through the
filter and be trapped in the silicone oil, while those PCBs adsorbed onto
solid particles would be retained by the filter. It was observed, however,
that less than 1% of the total PCB collected was on the filter.' ' In
maintaining their equilibrium vapor pressure with the surrounding air at the
high flow rates used, the particulate-bound PCBs were most likely entrained
in the airflow. Thus, no conclusion as to the relative concentration of PCBs
in the particulate and aerosol-vapor phase was possible. Other workers^ '
have suggested that PCBs adsorbed onto particulate matter are not retained on
glass fiber filters during sampling.
(14) (14)
Cottonseed-oil-coated glass beadsv ' and paraffin-coated Chromosorb Av '
have both been tested for collection of a variety of pesticides and may have
utility for collecting PCBs. While the former system appears promising, pro-
longed sampling of air can result in oxidation and polymerization of the
(29)
cottonseed oil making extraction of the oil difficult/ ' More work is needed
on liquid phase-solid support systems to define the collection efficiency for
the more volatile components of Aroclors, mono- and dichloro-biphenyls. Higher
collection efficiences than 70% are also desirable.
-------
Solid Adsorbents
Several solid adsorbents have been used to collect PCBs. Even though
low flow rates of 2 to 3 1/min were used, Florisil showed no breakthrough of
( 19}
PCBs from the first Florisil column after 5 days sampling/ ; Low analytical
blanks for the Florisil - 0.1 ng - using electron-capture gas chromatogrpahy
were obtained. This collection system warrants further investigation at
higher flow rates.
A highly promising sampling system is the polyurethane foam system of
Bidleman and Olney^ ' ' which has been tested for PCBs. In this system, an
8-by-l 0-inch glass-fiber filter for particulate collection is placed on top
of a container holding the solid trapping medium. The medium consists of
a porous polyurethane foam plug, pre-cleaned by washing with water and
extraction with acetone and petroleum ether. The entire apparatus is fitted
to a high volume sampler. Such an apparatus, developed under an EPA contract
with Environmental Sciences and Engineering, Inc./ ' is shown in Figure 1.
High airflow rates of 0.4 to 0.8 m/min have been obtained' ''' allowing
the collection of large air volumes.
The efficiency of this system for collection of tri-, tetra-, and
pentachloro- biphenyl vapors was found to be 99 to 96 %; 1 % or less of the
total PCB was formed on a backup plug. ' Similar quantitative recovery of
Aroclor 1221, which is 51 and 32 % mono- and dichloro- biphenyl, respectively,
has been obtained by other workers. ' Ambient PCB concentrations of 4 ng/m
have been measured in a 4-hour sampling time/ ' Refrigerated storage of
collected samples prior to analysis is recommended.
The ability to sample large volumes of air with quantitative retention
of PCBs indicates that the polyurethane foam sampling system is the most pro-
8
-------
Faceplate
30 on Stainless Steel Adapter
Polyurethane Foam
Plug Location
Filter
Position
Adapter
Wire
Retainer
Motor Unit
Adaoter
Exhaust Duct
(3 m minimum length)
Figure 1. Assembled sampler and shelter with exploded view of the filter holder.
-------
mising sampling system available at this time. In addition to quantitative
recovery, this system has the advantage, over liquid-phase solid support
sampling systems, of not requiring separation of a liquid sampling phase
prior to analysis. Additionally, the operating requirements of the sampler,
which are similar to the high volume sampler for measurement of suspended
(22)
particulate matter, . make it adaptable to routine sampling.
The polyurethane foam sampling system is being evaluated further under
the EPA contract with Environmental Science and Engineering cited above.
The effect of sampling time and temperature on retention of PCBs are being
studied.
The present extraction procedure for cleaning the polyurethane foam
plugs prior to sampling is lengthy: water wash, extraction for 12-hours with
acetone, followed by 2 hours with petroleum ether or hexane. ' Blank values
giving an electron-capture response of 500 ng/plug after extraction have
been obtained. ' It is not known whether this reponse is due to PCBs or
electron-accepting impurities from foam processing. Moderately high blank
values can be tolerated without seriously affecting the LDL, however. For
example, a blank value of 560 ng/plug means that to measure an ambient con-
centration of 10 ng/m of PCB (as decachlorobiphenyl) requires a sampling
3 (20)
time of 125 minutes at a sampling rate of 0.8 m /min. '
Since it is desirable to have a blank value that is reproducible from
plug to plug and obtainable with a shorter extraction time than above,
development of foam specifications were included in the above work. '
10
-------
SECTION 4
SAMPLE PREPARATION
Preparing a sample for analysis usually involves two procedures,
(1) removal of PCBs from the sampling medium by extraction with an organic
solvent and (2) treatment of the extract to remove interferences.
The solvents most frequently used to extract PCBs are petroleum
ether^ ' ' ' and hexane. ' ' Before use the solvent must be analyzed
to determine the PCB content; some pretreatment of the solvent is usually
required to reduce the PCB content to an analytically acceptable value. '
Pesticide grade solvents are commercially available. Sample extraction
times of 2 and 3 hours, and shaking the solvent-sample^ ' ' mixture in a
separatory funneP ' have been reported. Since removal of PCBs from the
polyurethane foam sampling mediunr ' ' appears to require considerably
less time than preparation of the foam for sampling, the blank response may
indeed be caused by electron-accepting compounds other than PCBs as was
observed earlier.
Since the sampling procedures for PCBs are not selective, they may
collect some interfering compounds along with PCBs; therefore, some treatment
of the extract prior to analysis may be required. Chlorinated and other
pesticides present a serious interference problem in analyzing residue
(23)
samples^ ' and are also the main interferents in air samples. Procedures
developed for removal of these interferents in residue analysis ' can be
11
-------
applied to air samples.
Figure 2 shows a chromatogram of an Aroclor 1254 - pesticide mixture;
the similarity of retention times for the Aroclor components and the pesti-
cides is striking and demonstrates the non-specificity of the electron capture
detector and the need for careful confirmation of the presence of PCBs by use
(23}
of at least two columns.v '
Two separate procedures are used to remove all of the interferring
pesticides prior to analysis: column chromatography using Florisil' '
(magnesia-silicate) —silicic acid and silica gel - Celite. *' The
Association of Analytical Chemists has a standard method for the Florisil
(21) (21)
procedure.v ' Both columns must be activated (heated)v ' to obtain optimum
separation; a water content of 3%^ ' in the silica gel - Celite procedure is
recommended for optimum separation. Florisil can be used to separate PCBs
from some pesticides (e.g., dieldrin and endrin), but the eluate retains DDT
and its metabolites. ' Silicic acid - Celite can be used to separate PCBs
(24 25)
from the latter compounds. *'
The stability of PCBs to alkali and acid permits removal of some
pesticides by chemical treatment. Treatment with alkali permits removal of
DDT.^ 3 ' Treatment with H2S04 destroys dieldrin and organophosphate
pesticides.(21'26)
Removal of interferences can require a significant amount of laboratory
work by the analyst. Sampling history and the identification of compounds in
_;
a sample chromatogram can be used to reduce this work. For example, sampling
in the area of a known source of PCBs should produce a high ratio of PCBs to
interferents. If this is confirmed by a sample chromatogram, interference
removal is probably not necessary. Removal of interferences is more likely
12
-------
1
8 12 16 20
Retention, minutes
28
32
Figure 2. Gas Chromatogram of:
Aroclor 1254 (solid line)
and pesticide mixture
(dotted line). Column
4% SE-30/6% QF-1, 200°C
carrier flow 70 ml/min.
(23)
1. Aldrin
2. Hept. Epoxide
3. p,p'-DDE
4. Dieldrin
5. p,p'-DDD
6. p,p'-DDT
7. Oil an I and Methoxychlor
8. Dilan II
13
-------
to be necessary when sampling in areas removed from PCB sources where inter-
ferents would be more predominate. Interference from DDT and its metabolites
cannot be discounted; nanogram quantities of DDT and p,p'-DDE resulting from
(27]
aerial fallout have been measured in southern California. '
14
-------
SECTION 5
QUALITATIVE AND QUANTITATIVE ANALYSES
PCBs are usually determined by gas-liquid chromatography using electron
capture detectors. ' Microcoulometric detectors have been used, but lack
(29)
the sensitivity required for analysis of PCB's in ambient air. ' Electron
(54)
capture detectors are specific only to electron-accepting molecules. '
Because of their high sensitivity, electron-capture detectors are preferred
for analysis of the low concentrations of PCBs present in ambient air.
A number of different liquid-solid support columns have been used to
separate PCBs. Liquid phases include: OV-17, SE-30 and QF-1;(7'20'30)OV-210
and OV-101; DC-200, and SF-96. ' Solid supports include: chromosorb W,
Gas Chrom P, and Gas Chrom Q.^ ' Selection of the proper column and optimi-
zation of chromatographic conditions is important to obtaining good resolution
of the Aroclor components and interferents.
During analysis it is very important to confirm the presence of PCBs,
because of the numerous interferents that can be present. Of the procedures -•
used for confirmation, perchlorination of the sample to convert PCBs to deca-
(31) (32)
chlorobiphenyr ' and mass spectrometry coupled with gas chromatographyv '
are the most absolute procedures. Qualitative identification of PCBs can be
made by comparing retention times of chromatographic peaks with those produced
by individual chlorinated biphenyls^ ' ' or Aroclors. Standard Aroclor
formulations are available from EPA.' ' All retention times should be relative
15
-------
to some arbitrary standard such as aldrin or p,p'-DDE. Gas - liquid chromato-
graphic retention times - relative to p,p'-DDE - for individual compounds in
the Aroclors are given in reference eight and twenty-eight and may be useful
to workers in the field of PCB methodology.
The various Aroclors exhibit different chromatograms, which can be
useful in distinguishing between them. The electron-capture chromatograms
for five different Aroclors are shown in Figure 3. The numbers on peaks
refer to individual compounds. Identification by the retention time pro-
cedure should be confirmed by use of two columns of different polarity. '
Confirmation of the existence of PCBs may also be obtained by thin layer
chromatography. '
Once the identification of PCBs has been confirmed, quantitation may be
achieved by (1) comparing properties of the sample chromatogram with those
produced by.Aroclor standards (2) using analyzed Aroclor standards, and
(3) perchlorinating the PCBs to decachlorobiphenyl. Accurate analysis of
PCBs is difficult at the present time. Each of the above procedures is
discussed below:
Comparison. Techniques based on comparison include subjectively
matching the chromatogram of the sample to that of the Aroclor it most
closely resembles.' ' Quantitation is obtained by integrating the area
under all of the peaks in the sample and comparing the total area Jo that
obtained, under the same chromatographic conditions, from a known weight
of the Aroclor standard. ' If the presence of more than one Aroclor is
indicated, the sample is quantitated using Aroclor standards judged appro-
priate for different portions of the chromatogram/ ' ' This "fingerprinting"
technique is the most widely used approach to quantitation because of its
(23)
practicality.
16
-------
AROCLOR 1221
AROCLOR 1242
AROCLOR 1248
AROCLOR 1254
0 24 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38
MINUTES
Figure 3. Comparison of electron capture chromatograms for Aroclor 1221, 1242,
1248, 1254, and 1260.
Source: Reference 11. Reprinted with permission from American National
Standards Institute, 1430 Broadway, New York, N. Y. 10018.
Publication C-107.1 - 1974, Copyrighted by ANSI.
17
-------
Other techniques are based on determining the ratio of peak heights
or areas between sample and Aroclor standard chromatograms for selected
peaks.<30>33>
The adequacy of all comparison procedures depends on the similarity of
sample and standard chromatograms. ' Because of photochemical breakdown
in the atmosphere and non-representative sampling of Aroclor components (due
to different vapor pressures of the components) the chromatogram obtained
fronrim-environmental sample is hardly ever identical to that of the original
Aroclor. Samples may also be composites of several Aroclors. Also the
electron-capture response of different chlorinated biphenyls can vary as much
(?^ 3R^
as 10,000 fold.v ' ' The weakness of comparative procedures, therefore,
lies i'n the impossibility^ ' of matching the sample and standard chromatograms.
Comparison techniques should be considered as semi-quantitative procedures.
Analyzed Aroclor Standards. By use of GC-mass spectrometry and a GC
with .an, electrolytic conductivity detector (to measure chlorine) the weight
(36)
percent of PCB in each peak of a given Aroclor has been determined. '
This information and the analyzed standards can be used to determine response
(36)
factors for corresponding peaks in an environmental PCB sample. ' The total
amount of PCB then becomes the amounts from all the individual peaks. The
authors in their paper have offered limited quantities of these standards to
interested parties.
Other workers' '' have used the above procedure to analyze environ-
mental samples for PCBs. This technique is an accurate means of analyzing
PCBs provided all peaks in the environmental samples are identified. The
technique, although accurate, is tedious. The obvious limitation of this
technique is the availability of the standards.
18
-------
Perchlon'nation. This technique of quantitation eliminates errors
inherent in the subjective nature of the comparison - "fingerprinting"
techniques, if the perchlorination itself is quantitative.
(39}
The reported procedurev ; involves reacting the sample, after removal
of interferences, with SbClr in chloroform solvent in a closed system at
165-175°C overnight. This chlorination converts the chlorinated biphenyls,
with different degrees of chlorination, to decachlorbiphenyl (DCB). Average
(39)
conversions of 93 to 100% were obtained with six different Aroclors. '
(The chloroform, of course, must be removed prior to analysis because it
responds to electron capture detectors.)
The main advantage of this technique over other techniques is that the
chromatogratn of the sample has only one peak, DCB. Quantitation is obtained
by comparing the peak height or area of the DCB peak with that of a known
weight of DCB of known purity. Comparison of the perchlorination technique
with the finger-printing technique gave good agreement when the DCB analysis
(39)
was reported in terms of the particular Aroclor. ' Since the response of
electron capture detectors depends on the number of chlorine atoms in the
biphenyl molecule, the perchlorination technique also increases the sensi-
tivity over that obtained with other quantitation techniques.
Some workers have been unsuccessful in obtaining quantitative con-
version of PCBs to DCB, ' typical results are shown in Figure 4. ' These
chromatograms were obtained using a starting mixture containing seven
different mono--di- and trichlorobiphenyls. One of the problems appears to
be trying to perch!orinate PCBs with varying degrees of chlorination - and
consequently varying susceptibilities to electophillic substitution - with
one set of reaction conditions. ' The use of milder perchlorination
conditions is indicated. ' In addition to optimizing the conditions of
19
-------
SbCl,-:200:lhr
and I02°:lhr
and 175°:15hr
63 ± 8
to DCB
conversion
SbCl5:175°:15hr
"Armour Conditions"
70 - 8% conversion
to DCB
Figure 4. Chromatogram of Perch!orinated PCB Mixture.
column: glass 6', CV-T7/QF-1 on chromosorb w at ~200°C
20
-------
perchlori nation, any significant amounts of biphenyl in samples will, of
i'
course, have to be removed prior to perchlori nation.
(41 )
Impurities, SbCl.Br and DCB, have been found in SbClr/ ' The former
compound produces bromononachlorobiphenyl and, of course, affects conversion
)
of PCBsv ' to DCB. Varying amounts of DCB have been found in different
commercial sources of SbCl^, but in general, the amount of DCB impurity is
not significant because of the small amount of SbClr required in the perchlor-
i nation reaction.
While further investigations are needed to optimize the perchlori nation
procedure, the technique has considerably more potential for accurately
quantitating PCBs than either comparison or analyzed standard techniques.
Further development of the perchlori nation procedure is in progress under an
EPA contract. ^
It should be noted that PCB values reported as ppm DCB will be higher
than values based on an individual Aroclor. The DCB analysis value can be
(39)
reliably converted to an individual Aroclor valuev ' only if a positive
identification of the Aroclor - in the collected sample - is achieved. As
discussed earlier, this identification is not an easy task.
(42)
Other Techniques. Computer controlled high resolution mass spectrometry. '
(44) (43 55)
polarography, ' and plasma chromatography, ' ' have been used to analyze
PCBs.
21
-------
SECTION 6
SUMMARY AND CONCLUSIONS
The most promising methodology for analyzing PCBs in ambient air at the
present time involves: (1) collecting the sample on polyurethane foam,
(2) extracting PCBs with an organic solvent, (3) removing interferents - as
necessary - by column chromatography, (4) confirming the presence of PCBs by
perch!orination in conjunction with electron capture - gas chromatography or
by mass spectrometry - gas chromatography, and (5) analysis with electron
capture - gas chromatography. A reliable and widely applicable quantisation
procedure has not yet been developed. However, perchlorination offers
considerable promise for fulfilling these requirements. Until such a pro-
cedure is developed, most workers will be forced to use subjective techniques
for quantitation.
Work on this method is still in the development stage and a number of
areas, as discussed in this review, need further investigation and improvement
before the method should be considered capable of producing reliable data.
Quantisation is the main area in need of improvement. A standard source of
high-purity decachlorobiphenyl would be useful if the perchlorination tech-
nique can be successfully developed. Quality control samples, e.g., PCBs on
polyurethane foam, are needed to aid the analyst in obtaining reliable data.
Methodology for measuring PCBs in ambient air is technically demanding
and requires considerable operator skill; however, the sampling procedure
22
-------
appears potentially suitable for routine use by technician-level personnel.
In operating a monitoring network, e.g., around a source of PCBs, collected
samples could be returned to a central laboratory that possessed the skills
required to carry out the analysis and quantisation.
23
-------
SECTION 7
STATIONARY SOURCES
As mentioned earlier, work on methodology for measurement of PCBs in
stationary sources appears to be in the early stages of development.
A common source of PCBs is incinerators which seldom maintain the
temperature of 2000°F required to obtain complete combustion.
The main difference between source and ambient methodology is in sampling.
In stationary sources, gases at varying elevated temperatures (e.g., 150 to
800°F, depending on the particular source)* ' need to be sampled isokineti-
cally. These gases contain varying amounts of water vapor and sometimes liquid
water. ' These properties of stationary source emissions impose some
additional requirements on sampling over those required for ambient sampling.
Namely, the absorbing medium, especially for liquids, must be cooled to retain
PCBs, and water must be removed before the gases reach the absorption medium.
Liquids, liquids on solid supports, and solids have been used to collect
PCBs. In laboratory studies, hexane, in impingers maintained at -78°C,
quantitatively recovered PCBs; continued bubbling of air through a hexane - PCB
solution containing 8 ppm PCBs showed no loss of PCBs.^ ' The system has been
used for sampling a waste treatment plant.^ ' Other workers^ ' have used
cooled hexane to sample incinerator stack emissions for PCBs. Aqueous 10 %
glycerine solutions contained in 2 absorption bottles showed 98 % recovery of
(48) (49)
PCBs; the system has been used to sample incinerator gases.v ' Toluenev
24
-------
has been recommended for collecting PCBs in source emissions.
(47)
Glycerine on Florisilv ' has been used to sample incinerator gases.
Two solid sampling materials, Tenax, poly(2,6-diphenyl-p-phenyleneoxide),
and Florisil have produced quantitative recovery of PCBs at elevated
(45)
temperatures.v ' Florisil is especially promising because of its low cost
and low blank values.
Interferences from stationary source emissions vary with the incineration
process and the materials being burned.
Electron capture - GC* ' and GC coupled with mass spectrometry^ '
have been used to analyze source samples. Quantisation by perchlorination
(47 51)
has been used in source analysis on at least two occasions, ' ' but neither
group of workers obtained quantitative conversion of PCBs to DCB. Bi- tri-,
and tetrachlorobiphenyls gave lower conversions than the more highly chlori-
(51)
nated components of PCBs.x '
Further investigations, especially in the area of sampling, are needed
to develop reliable methods for measuring PCBs in stationary sources. The
work being carried out to develop the perchlorination technique of quantisation
should be of mutual benefit to investigators interested in ambient and source
methodology.
25
-------
SECTION 7
REFERENCES
1. Soren Jensen, Polychlorinated Biphenyls as Contaminants of the
Environment-History. Proceedings of the National Swedish Environmental
Protection Board, Stockholm, Sweden, pp. 7-17 (September 29, 1970).
2. Price, H. A. and Welch, R. L. Occurrence of Polychlorinated Biphenyls
in Humans. Environmental Health Perspectives, pp. 73-78 (April 1972).
3. Fuller, B. et al. Environmental Assessment of Polychlorinated Biphenyls
in the Atmosphere. Mitre Corp. Technical Report MTR-7210 (April 1976)."
Prepared under EPA Contract 68-02-1495.
4. Hutzinger, 0. et al. The Chemistry of Polychlorinated Biphenyls.
Chemical Rubber Company Press, Chapter 1, Cleveland Ohio, 1974.
5. Bidleman, T. F. and Olney, C. E. Chlorinated Hydrocarbons in the
Sargasso Sea Atmosphere and Surface Water. Science, 183, 516-518,
(February 8, 1974).
6. Threshold Limit Values for Chemical Substances and Physical Agents.
American Conference of Governmental Industrial Hygiemsts, 1973.
Cincinnati, Ohio.
7. Manual of Analytical Methods for the Analysis of Pesticide Residues in
Human Environmental Samples. Environmental Toxicology Division,
Environmental Protection Agency, Research Triangle Park, N. C. 27711
(December, 1974).
8. Hutzinger, 0. et al. The Chemistry of Polychlorinated Biphenyls.
Chemical Rubber Company Press, Chapter 2, Cleveland Ohio, 1974.
9. Kutz, F. W. and Yang, H. S. C. A Note on Polychlorinated Biphenyls in
Air. National Conference on Polychlorinated Biphenyls, November 19-21,
1975. Chicago, Illinois, EPA Report No. 560/6-75-004.
10. Bidleman, T. F., and Olney, C. E. High Volume Collection of Atmospheric
Polychlorinated Biphenyls. Bull, of Environ. Cont. and Tox., 11,
442-450 (1974).
11. Guidelines for Handling and Disposal,of Capacitor - and Transformer -
Grade Askarels Containing Polychlorinated Biphenyls." ANSI Report C-107.1
(1974). American National Standards Institute, 1430 Broadway, New York,
New York 10018.
26
-------
12. Sampling Survey Related to Possible Emission of Polychlorinated Biphenyls.
EPA Region 5, Surveillance and Analysis Division, Chicago, Illinois.
October-November, 1975.
13. Stanley, C. W. et al. Measurement of Atmospheric Pesticides. Env.
Sci. and Tech., 5, 430-435 (1971).
14. Seiber, J. N. et al. Determination of Pesticides and Their Trans-
formation Products in Air. Environmental Dynamics of Pesticides,
pp. 17-43, Plenum Publishing Company, 227 West 17th Street, New York,
New York 10011.
15. Jackson, Merrill. Environmental Protection Agency, Research Triangle
Park, North Carolina 27711. Personal Communication.
16. Bengtson, S. A., and Sodergren, A. DDT and Polychlorinated Biphenyl
Residues in Airborne Fallout and Animals in Iceland. Ambio, 3(2), 84-87,
(1974).
17. Harvey, G. R., and Steinhaufr, W. G. Atmospheric Transport of Poly-
chlorinated Biphenyls to the North Atlantic. Atmospheric Environment, 8,
777-782 (1974).
18. Wakimoto, T. et. al. Method for the Quantisation of Organic Chlorine
Compounds in the Air by the Dry Sampling Method. Japan Analyst, 23(7),
790-793 (July 1974).
19. Giam, C. S. et al. Rapid and Inexpensive Method for Detection of Poly-
chlorinated Biphenyls and Phthalates in Air. Anal. Chem., 47^ 2319-2320
(1975).
20. Evaluation of a Method for the Analysis of Airborne Polychlorinated
Biphenyls, April 1976. Prepared under EPA Contract 68-01-2978.
21. Hotzinger, 0. et al. The Chemistry of Polychlorinated Biphenyls.
Chemical Rubber Company Press, Chapter 12, Cleveland, Ohio, 1974.
22. Code of Federal Regulations, Part 50, pp. 12-17, July 1, 1975. Reference
Method for the Determination of Suspended Particulates in the Atmosphere
(High Volume Method).
23. Manual of Analytical Quality Control for Pesticides in Human and Environ-
mental Media. Health Effects Research Laboratory, Environmental Toxi-
cology Division, Research Triangle Park, North Carolina 27711.
24. Armour, J. A. and Burke, J. A. Method for Separating PCBs from DDT and
Its Analogs. JAOAC. 53_, 761-768 (1970).
25. Snyder, D. and Reinert, R. Rapid Separation of Polychlorinated Biphenyls
. from,DDT and its Analogues on Silica Gel. Bull of Environ. Cont. and Tox.,
6, 385-390 (1971).
27
-------
26. Murphy, P. G. Sulfuric Acid for the Cleanup of Animal Tissues for
Analysis of Acid Stable Chlorinated Hydrocarbon Residues, JAOAC, 55,
1360-1362 (1972).
27. Young, D. R. Southern California Coastal Water Research Project, 1500
East Imperial Highway, El Segundo, California. Aerial Fallout of DDT
in Southern California. Submitted for publication in Bull, of Env. Cont.
and Tox.
28. Webb, A. G. and McCall, R. C. Identities of Polychlorinated Biphenyl
Isomers in Aroclors. JAOAC. 55. 748-752 (1972).
29. Lewis, Robert G. Environmental Protection Agency, Research Triangle
Park, North Carolina 27711. Personal communication.
30. Sherma, J. Gas Chromatography Analysis of Polychlorinated Biphenyls and
Other Nonpesticide Organic Pollutants. Advances in Chromatography, 12,
141-176 (1975), Marcel Dekker. ~~
31. Armour, J. A. Quantitative Perchlorination of Polychlorinated Biphenyls
as a Method for Confirmatory Residue Measurement and Identification.
JAOAC, 56_, 987-993 (1972).
32. Hutzinger, 0. et al. The Chemistry of Polychlorinated Biphenyls.
Chemical Rubber Company Press, Chapter 8, Cleveland, Ohio, 1974.
33. Polychlorinated Biphenyls - Environmental Impact; A Review by the Panel
on Hazardous Trace Substances. Environmental Research, 5, p. 345 (1972).
34. Analytical Reference Standards and Supplemental Data for Pesticides and
Other Selected Organic Compounds. EPA Report No. 600/9-76-012, May 1976.
Health Effects Research Laboratory, Research Triangle Park, N. C. 27711.
35. Zitko, V. et al. Retention Times and Electron-Capture Detector Responses
of Some Individual Chlorobiphenyls. Bull, of Environ. Cont. and Tox., 6.
160-163 (1971).
36. Webb, R. G. and McCall, A. C. Quantitative Polychlorinated Biphenyl
Standards for Electron Capture Gas Chromatography. J. Chrom. Sci., 11.
366-373 (1973).
37. Chau, A. S. Y. and Sampson, R. C. J. Electron Capture Gas Chromatographic
Methodology for the Quantisation of Polychlorinated Biphenyls: Survey and
Compromise. Environ. Letters, 8, 89-101 (1975).
38. Rote, J. W. and.Murphy, P. G. A Method for the Quantisation of Poly-
chlorinated Biphenyl Isomers. Bull, of Environ. Cont. and Tox., 6,
377-384 (1971).
39. Armour, J. A. Quantitative Perchlorination of Polychlorinated Biphenyls
as a Method for Confirmatory Residue Measurement and Identification.
JAOAC, 56. 987-993 (1973).
28
-------
40. Parris, 6. Notes on Polychlorinated Biphenyl Monitoring, Office of Toxic
Substances, Environmental Protection Agency, Washington, D.C. 20460.
41. Trotter, W. J., and Young, S. J. V. Limitation of the use of Antimony
Pentachloride for Perchlorination of Polychlorinated Biphenyls. JAOAC,
58_, 466-468 (1975).
42. Schuetzle, D. et al. Application of Computer Controlled High Resolution
Mass Spectrometry to the Analysis of Air Pollutants. Presented at an APCA
Meeting, Miami, Florida, June 18-22, 1972, as Paper No. 72-15.
43. Karasek, F. W. Plasma Chromatography of the Polychlorinated Biphenyls.
Anal. Chem.. 43_, 1982-1986 (1971).
44. Ishii, T. Polarographic Analysis of Air Pollutants. J. Pollution
Control - Japan, 8, 665-672 (1972).
45. Mitchell, W. J. Environmental Protection Agency, Research Triangle Park,
North Carolina 27711. Personal communication.
46. Abe, T., and Sone, M. Investigation of Measuring Method of Polychlorinated
Biphenyl in Gas Phase. Pollut. Contr. Tech. Center Rep., (Japan) No. 2,
pp. 24-27 (April 1974).
47. Ryota, S. et al. Determination of Polychlorinated Biphenyl in Dust, Ash
and Combustion Gas from City Waste Incinerators. Rep. Aichi Environmental
Res. Center (Japan), Vol. 2, pp. 43-49 (1974).
48. Kawase, Z. et al. Studies on Measurement of Polychlorinated Biphenyls in
Exhaust Gases. J. Japan Soc. Air Pollution, Vol. 8, No. 3, p. 598,
(October 1973).
49. Herman, T. S. Development of Sampling Procedures for Polycyclicorganic
Matter and Polychlorinated Biphenyls. EPA Report No. 650/2-75-007,
August 1974. Prepared under EPA Contract 68-02-1255.
50. Okuno, T., and Masahiko, T. Determination of Polychlorinated Biphenyl
in Stack Gas and in the Atmosphere. Japan Soc. Air Pollut. (Proc.),
p. 109 (November 7-9, 1972). Paper No. 64.
51. EPA Contract 68-02-1399. Unpublished information.
52. Schoor, W. P. Problems Associated with Low Solubility Compounds in
Aquatic Toxicity Tests: Theoretical Model and Solubility Characteristics
of Aroclor 1254 in Water. Water Research, 9_, 937-944 (1975).
53. Haque, R. and Kohmert, R. Studies on the Vapor Behavior of Selected
Polychlorinated Biphenyls. J. Env. Sci. Health B1.. 253-264 (1976).
54. E. D. Pellizzari. Electron Capture Detection in Gas Chromatography.
J. Chromatography,' Chromatographic Rev., 98_, 323,361 (1974).
29
-------
55. B. Bush and Fa-Chum Lo. Thin-Layer Chromatography for Quantitative
Polychlorinated Biphenyl Analysis. J. Chromatography, 77, 377-388
(1973).
56. Chromatograms prepared by George Parris, Office of Toxic Substances-EPA,
from work described in reference 20.
30
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing]
1. REPORT NO.
EPA-600/4-77-021
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
METHODOLOGY FOR MEASUREMENT OF POLYCHLORINATED BIPHENYLS
IN AMBIENT AIR AND STATIONARY SOURCES - A REVIEW
5. REPORT DATE
April 1977 iussinq date
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
John H. Margeson
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Quality Assurance Branch
Environmental Monitoring and Support Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
10. PROGRAM ELEMENT NO.
1HD621
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Monitoring and Support Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
EPA/600/08
15. SUPPLEMENTARY NOTES
16. ABSTRACT
The state of development of methodology for measurement of polychlorinated
biphenyls (PCBs) in ambient air and stationary sources was reviewed.
The most promising method for ambient air measurements involves collection of
PCBs on polyurethane foam, extraction with an organic solvent, removal of interferences
by column chromatography, and confirmation and analysis by electron-capture gas
chromatography. Quantisation by perchlorination of PCBs to decachlorobiphenyl (DCB)
is the most promising quantisation technique, but the procedure has not yet been
perfected to the point where all PCBs are quantitatively converted to DCB. Perfection
of this technique should allow for significant improvement in the quality of ambient
PCB data being generated.
Source and ambient methods differ mainly in sampling. Work on methodology for
stationary sources is in the early stages of development and further investigations
are needed.
The report contains 56 references.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COS AT I Field/Group
Air Pollution
Monitoring
Quality Assurance
Methods Evaluation
13B
13. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (TillsReport)
Unclassified
21. NO. OF PAGES
36
20. SECURITY fLfySS/This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
31
------- |